diff --git a/.clang-tidy b/.clang-tidy index de19059d09e..bb63bf2eea6 100644 --- a/.clang-tidy +++ b/.clang-tidy @@ -27,6 +27,8 @@ Checks: [ '-bugprone-not-null-terminated-result', '-bugprone-reserved-identifier', # useful but too slow, TODO retry when https://reviews.llvm.org/rG1c282052624f9d0bd273bde0b47b30c96699c6c7 is merged '-bugprone-unchecked-optional-access', + '-bugprone-crtp-constructor-accessibility', + '-bugprone-suspicious-stringview-data-usage', '-cert-dcl16-c', '-cert-dcl37-c', @@ -36,6 +38,7 @@ Checks: [ '-cert-msc51-cpp', '-cert-oop54-cpp', '-cert-oop57-cpp', + '-cert-err33-c', # Misreports on clang-19: it warns about all functions containing 'remove' in the name, not only about the standard library. '-clang-analyzer-optin.performance.Padding', @@ -99,6 +102,7 @@ Checks: [ '-modernize-use-emplace', '-modernize-use-nodiscard', '-modernize-use-trailing-return-type', + '-modernize-use-designated-initializers', '-performance-enum-size', '-performance-inefficient-string-concatenation', diff --git a/.git-blame-ignore-revs b/.git-blame-ignore-revs index 06e893fabb3..5088bf90ede 100644 --- a/.git-blame-ignore-revs +++ b/.git-blame-ignore-revs @@ -13,3 +13,6 @@ # dbms/ → src/ # (though it is unlikely that you will see it in blame) 06446b4f08a142d6f1bc30664c47ded88ab51782 + +# Applied Black formatter for Python code +e6f5a3f98b21ba99cf274a9833797889e020a2b3 diff --git a/.gitattributes b/.gitattributes index 56d6fecf4b8..dd94a48f8e7 100644 --- a/.gitattributes +++ b/.gitattributes @@ -2,3 +2,4 @@ contrib/* linguist-vendored *.h linguist-language=C++ tests/queries/0_stateless/data_json/* binary tests/queries/0_stateless/*.reference -crlf +src/Core/SettingsChangesHistory.cpp merge=union diff --git a/.github/actionlint.yml b/.github/actionlint.yml index 0f88f30d42c..4357bd3eb6b 100644 --- a/.github/actionlint.yml +++ b/.github/actionlint.yml @@ -7,3 +7,4 @@ self-hosted-runner: - stress-tester - style-checker - style-checker-aarch64 + - release-maker diff --git a/.github/actions/release/action.yml b/.github/actions/release/action.yml new file mode 100644 index 00000000000..c3897682a33 --- /dev/null +++ b/.github/actions/release/action.yml @@ -0,0 +1,168 @@ +name: Release + +description: Makes patch releases and creates new release branch + +inputs: + ref: + description: 'Git reference (branch or commit sha) from which to create the release' + required: true + type: string + type: + description: 'The type of release: "new" for a new release or "patch" for a patch release' + required: true + type: choice + options: + - patch + - new + dry-run: + description: 'Dry run' + required: false + default: true + type: boolean + token: + required: true + type: string + +runs: + using: "composite" + steps: + - name: Prepare Release Info + shell: bash + run: | + python3 ./tests/ci/create_release.py --prepare-release-info \ + --ref ${{ inputs.ref }} --release-type ${{ inputs.type }} \ + ${{ inputs.dry-run && '--dry-run' || '' }} + echo "::group::Release Info" + python3 -m json.tool /tmp/release_info.json + echo "::endgroup::" + release_tag=$(jq -r '.release_tag' /tmp/release_info.json) + commit_sha=$(jq -r '.commit_sha' /tmp/release_info.json) + echo "Release Tag: $release_tag" + echo "RELEASE_TAG=$release_tag" >> "$GITHUB_ENV" + echo "COMMIT_SHA=$commit_sha" >> "$GITHUB_ENV" + - name: Download All Release Artifacts + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + python3 ./tests/ci/create_release.py --download-packages ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Push Git Tag for the Release + shell: bash + run: | + python3 ./tests/ci/create_release.py --push-release-tag ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Push New Release Branch + if: ${{ inputs.type == 'new' }} + shell: bash + run: | + python3 ./tests/ci/create_release.py --push-new-release-branch ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Bump CH Version and Update Contributors' List + shell: bash + run: | + python3 ./tests/ci/create_release.py --create-bump-version-pr ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Bump Docker versions, Changelog, Security + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + git checkout master + python3 ./tests/ci/create_release.py --set-progress-started --progress "update changelog, docker version, security" + echo "List versions" + ./utils/list-versions/list-versions.sh > ./utils/list-versions/version_date.tsv + echo "Update docker version" + ./utils/list-versions/update-docker-version.sh + echo "Generate ChangeLog" + export CI=1 + docker run -u "${UID}:${GID}" -e PYTHONUNBUFFERED=1 -e CI=1 --network=host \ + --volume=".:/ClickHouse" clickhouse/style-test \ + /ClickHouse/tests/ci/changelog.py -v --debug-helpers \ + --gh-user-or-token=${{ inputs.token }} --jobs=5 \ + --output="/ClickHouse/docs/changelogs/${{ env.RELEASE_TAG }}.md" ${{ env.RELEASE_TAG }} + git add ./docs/changelogs/${{ env.RELEASE_TAG }}.md + echo "Generate Security" + python3 ./utils/security-generator/generate_security.py > SECURITY.md + git diff HEAD + - name: Create ChangeLog PR + if: ${{ inputs.type == 'patch' && ! inputs.dry-run }} + uses: peter-evans/create-pull-request@v6 + with: + author: "robot-clickhouse " + token: ${{ inputs.token }} + committer: "robot-clickhouse " + commit-message: Update version_date.tsv and changelogs after ${{ env.RELEASE_TAG }} + branch: auto/${{ env.RELEASE_TAG }} + assignees: ${{ github.event.sender.login }} # assign the PR to the tag pusher + delete-branch: true + title: Update version_date.tsv and changelog after ${{ env.RELEASE_TAG }} + labels: do not test + body: | + Update version_date.tsv and changelogs after ${{ env.RELEASE_TAG }} + ### Changelog category (leave one): + - Not for changelog (changelog entry is not required) + - name: Complete previous steps and Restore git state + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + python3 ./tests/ci/create_release.py --set-progress-completed + git reset --hard HEAD + git checkout "$GITHUB_REF_NAME" + - name: Create GH Release + shell: bash + if: ${{ inputs.type == 'patch' }} + run: | + python3 ./tests/ci/create_release.py --create-gh-release ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Export TGZ Packages + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + python3 ./tests/ci/artifactory.py --export-tgz ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Test TGZ Packages + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + python3 ./tests/ci/artifactory.py --test-tgz ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Export RPM Packages + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + python3 ./tests/ci/artifactory.py --export-rpm ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Test RPM Packages + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + python3 ./tests/ci/artifactory.py --test-rpm ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Export Debian Packages + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + python3 ./tests/ci/artifactory.py --export-debian ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Test Debian Packages + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + python3 ./tests/ci/artifactory.py --test-debian ${{ inputs.dry-run && '--dry-run' || '' }} + - name: Docker clickhouse/clickhouse-server building + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + cd "./tests/ci" + python3 ./create_release.py --set-progress-started --progress "docker server release" + export CHECK_NAME="Docker server image" + python3 docker_server.py --release-type auto --version ${{ env.RELEASE_TAG }} --check-name "$CHECK_NAME" --sha ${{ env.COMMIT_SHA }} ${{ ! inputs.dry-run && '--push' || '' }} + python3 ./create_release.py --set-progress-completed + - name: Docker clickhouse/clickhouse-keeper building + if: ${{ inputs.type == 'patch' }} + shell: bash + run: | + cd "./tests/ci" + python3 ./create_release.py --set-progress-started --progress "docker keeper release" + export CHECK_NAME="Docker keeper image" + python3 docker_server.py --release-type auto --version ${{ env.RELEASE_TAG }} --check-name "$CHECK_NAME" --sha ${{ env.COMMIT_SHA }} ${{ ! inputs.dry-run && '--push' || '' }} + python3 ./create_release.py --set-progress-completed + - name: Set current Release progress to Completed with OK + shell: bash + run: | + python3 ./tests/ci/create_release.py --set-progress-started --progress "completed" + python3 ./tests/ci/create_release.py --set-progress-completed + - name: Post Slack Message + if: ${{ !cancelled() }} + shell: bash + run: | + python3 ./tests/ci/create_release.py --post-status ${{ inputs.dry-run && '--dry-run' || '' }} diff --git a/.github/workflows/auto_release.yml b/.github/workflows/auto_release.yml index f1a6b307b40..457ffacc7a8 100644 --- a/.github/workflows/auto_release.yml +++ b/.github/workflows/auto_release.yml @@ -1,44 +1,110 @@ name: AutoRelease env: - # Force the stdout and stderr streams to be unbuffered PYTHONUNBUFFERED: 1 + DRY_RUN: true concurrency: - group: auto-release + group: release on: # yamllint disable-line rule:truthy - # schedule: - # - cron: '0 10-16 * * 1-5' + # Workflow uses a test bucket for packages and dry run mode (no real releases) + schedule: + - cron: '0 9 * * *' + - cron: '0 15 * * *' workflow_dispatch: + inputs: + dry-run: + description: 'Dry run' + required: false + default: true + type: boolean jobs: - CherryPick: - runs-on: [self-hosted, style-checker-aarch64] + AutoRelease: + runs-on: [self-hosted, release-maker] steps: + - name: DebugInfo + uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6 - name: Set envs - # https://docs.github.com/en/actions/learn-github-actions/workflow-commands-for-github-actions#multiline-strings run: | cat >> "$GITHUB_ENV" << 'EOF' - TEMP_PATH=${{runner.temp}}/cherry_pick ROBOT_CLICKHOUSE_SSH_KEY<> "$GITHUB_ENV" + - name: Set DRY_RUN for dispatch + if: ${{ github.event_name == 'workflow_dispatch' }} + run: echo "DRY_RUN=${{ github.event.inputs.dry-run }}" >> "$GITHUB_ENV" - name: Check out repository code uses: ClickHouse/checkout@v1 with: - clear-repository: true token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}} fetch-depth: 0 - - name: Auto-release + - name: Auto Release Prepare run: | cd "$GITHUB_WORKSPACE/tests/ci" - python3 auto_release.py --release-after-days=3 - - name: Cleanup - if: always() + python3 auto_release.py --prepare + echo "::group::Auto Release Info" + python3 -m json.tool /tmp/autorelease_info.json + echo "::endgroup::" + { + echo 'AUTO_RELEASE_PARAMS<> "$GITHUB_ENV" + - name: Post Release Branch statuses + run: | + cd "$GITHUB_WORKSPACE/tests/ci" + python3 auto_release.py --post-status + - name: Release ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[0].release_branch }} + if: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[0] && fromJson(env.AUTO_RELEASE_PARAMS).releases[0].ready }} + uses: ./.github/actions/release + with: + ref: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[0].commit_sha }} + type: patch + dry-run: ${{ env.DRY_RUN }} + token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}} + - name: Release ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[1].release_branch }} + if: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[0] && fromJson(env.AUTO_RELEASE_PARAMS).releases[1].ready }} + uses: ./.github/actions/release + with: + ref: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[1].commit_sha }} + type: patch + dry-run: ${{ env.DRY_RUN }} + token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}} + - name: Release ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[2].release_branch }} + if: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[2] && fromJson(env.AUTO_RELEASE_PARAMS).releases[2].ready }} + uses: ./.github/actions/release + with: + ref: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[2].commit_sha }} + type: patch + dry-run: ${{ env.DRY_RUN }} + token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}} + - name: Release ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[3].release_branch }} + if: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[3] && fromJson(env.AUTO_RELEASE_PARAMS).releases[3].ready }} + uses: ./.github/actions/release + with: + ref: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[3].commit_sha }} + type: patch + dry-run: ${{ env.DRY_RUN }} + token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}} + - name: Release ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[4].release_branch }} + if: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[4] && fromJson(env.AUTO_RELEASE_PARAMS).releases[4].ready }} + uses: ./.github/actions/release + with: + ref: ${{ fromJson(env.AUTO_RELEASE_PARAMS).releases[4].commit_sha }} + type: patch + dry-run: ${{ env.DRY_RUN }} + token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}} + - name: Post Slack Message + if: ${{ !cancelled() }} + run: | + cd "$GITHUB_WORKSPACE/tests/ci" + python3 auto_release.py --post-auto-release-complete --wf-status ${{ job.status }} + - name: Clean up run: | docker ps --quiet | xargs --no-run-if-empty docker kill ||: docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||: diff --git a/.github/workflows/backport_branches.yml b/.github/workflows/backport_branches.yml index 48e8fbbba05..322946ac77b 100644 --- a/.github/workflows/backport_branches.yml +++ b/.github/workflows/backport_branches.yml @@ -36,10 +36,6 @@ jobs: cd "$GITHUB_WORKSPACE/tests/ci" echo "Testing the main ci directory" python3 -m unittest discover -s . -p 'test_*.py' - for dir in *_lambda/; do - echo "Testing $dir" - python3 -m unittest discover -s "$dir" -p 'test_*.py' - done - name: PrepareRunConfig id: runconfig run: | @@ -62,7 +58,7 @@ jobs: BuildDockers: needs: [RunConfig] if: ${{ !failure() && !cancelled() }} - uses: ./.github/workflows/reusable_docker.yml + uses: ./.github/workflows/docker_test_images.yml with: data: ${{ needs.RunConfig.outputs.data }} CompatibilityCheckX86: @@ -167,10 +163,16 @@ jobs: steps: - name: Check out repository code uses: ClickHouse/checkout@v1 + - name: Download reports + run: | + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(needs.RunConfig.outputs.data) }} --pre --job-name Builds - name: Builds report run: | cd "$GITHUB_WORKSPACE/tests/ci" python3 ./build_report_check.py --reports package_release package_aarch64 package_asan package_tsan package_debug binary_darwin binary_darwin_aarch64 + - name: Set status + run: | + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(needs.RunConfig.outputs.data) }} --post --job-name Builds ############################################################################################ #################################### INSTALL PACKAGES ###################################### ############################################################################################ @@ -239,8 +241,9 @@ jobs: runner_type: stress-tester data: ${{ needs.RunConfig.outputs.data }} FinishCheck: - if: ${{ !failure() && !cancelled() }} + if: ${{ !cancelled() }} needs: + - RunConfig - Builds_Report - FunctionalStatelessTestAsan - FunctionalStatefulTestDebug @@ -255,6 +258,7 @@ jobs: with: clear-repository: true - name: Finish label + if: ${{ !failure() }} run: | cd "$GITHUB_WORKSPACE/tests/ci" # update mergeable check @@ -262,3 +266,10 @@ jobs: # update overall ci report python3 finish_check.py --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} python3 merge_pr.py + - name: Check Workflow results + run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF + python3 ./tests/ci/ci_buddy.py --check-wf-status diff --git a/.github/workflows/create_release.yml b/.github/workflows/create_release.yml new file mode 100644 index 00000000000..55644bdd503 --- /dev/null +++ b/.github/workflows/create_release.yml @@ -0,0 +1,44 @@ +name: CreateRelease + +concurrency: + group: release +'on': + workflow_dispatch: + inputs: + ref: + description: 'Git reference (branch or commit sha) from which to create the release' + required: true + type: string + type: + description: 'The type of release: "new" for a new release or "patch" for a patch release' + required: true + type: choice + options: + - patch + - new + dry-run: + description: 'Dry run' + required: false + default: true + type: boolean + +jobs: + CreateRelease: + env: + GH_TOKEN: ${{ secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN }} + runs-on: [self-hosted, release-maker] + steps: + - name: DebugInfo + uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6 + - name: Check out repository code + uses: ClickHouse/checkout@v1 + with: + token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}} + fetch-depth: 0 + - name: Call Release Action + uses: ./.github/actions/release + with: + ref: ${{ inputs.ref }} + type: ${{ inputs.type }} + dry-run: ${{ inputs.dry-run }} + token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}} diff --git a/.github/workflows/reusable_docker.yml b/.github/workflows/docker_test_images.yml similarity index 100% rename from .github/workflows/reusable_docker.yml rename to .github/workflows/docker_test_images.yml diff --git a/.github/workflows/jepsen.yml b/.github/workflows/jepsen.yml index db837ac1ec7..035ba2e5b98 100644 --- a/.github/workflows/jepsen.yml +++ b/.github/workflows/jepsen.yml @@ -9,19 +9,64 @@ on: # yamllint disable-line rule:truthy - cron: '0 */6 * * *' workflow_dispatch: jobs: + RunConfig: + runs-on: [self-hosted, style-checker-aarch64] + outputs: + data: ${{ steps.runconfig.outputs.CI_DATA }} + steps: + - name: DebugInfo + uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6 + - name: Check out repository code + uses: ClickHouse/checkout@v1 + with: + clear-repository: true # to ensure correct digests + fetch-depth: 0 # to get version + filter: tree:0 + - name: PrepareRunConfig + id: runconfig + run: | + echo "::group::configure CI run" + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --configure --workflow "$GITHUB_WORKFLOW" --outfile ${{ runner.temp }}/ci_run_data.json + echo "::endgroup::" + + echo "::group::CI run configure results" + python3 -m json.tool ${{ runner.temp }}/ci_run_data.json + echo "::endgroup::" + { + echo 'CI_DATA<> "$GITHUB_OUTPUT" KeeperJepsenRelease: - uses: ./.github/workflows/reusable_simple_job.yml + needs: [RunConfig] + uses: ./.github/workflows/reusable_test.yml with: - test_name: Jepsen keeper check - runner_type: style-checker - report_required: true + test_name: ClickHouse Keeper Jepsen + runner_type: style-checker-aarch64 + data: ${{ needs.RunConfig.outputs.data }} run_command: | python3 jepsen_check.py keeper - # ServerJepsenRelease: - # uses: ./.github/workflows/reusable_simple_job.yml - # with: - # test_name: Jepsen server check - # runner_type: style-checker - # run_command: | - # cd "$REPO_COPY/tests/ci" - # python3 jepsen_check.py server + ServerJepsenRelease: + if: false # skip for server + needs: [RunConfig] + uses: ./.github/workflows/reusable_test.yml + with: + test_name: ClickHouse Server Jepsen + runner_type: style-checker-aarch64 + data: ${{ needs.RunConfig.outputs.data }} + run_command: | + python3 jepsen_check.py server + CheckWorkflow: + if: ${{ !cancelled() }} + needs: [RunConfig, ServerJepsenRelease, KeeperJepsenRelease] + runs-on: [self-hosted, style-checker-aarch64] + steps: + - name: Check out repository code + uses: ClickHouse/checkout@v1 + - name: Check Workflow results + run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat >> "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF + python3 ./tests/ci/ci_buddy.py --check-wf-status diff --git a/.github/workflows/master.yml b/.github/workflows/master.yml index 2a7e6f737ab..acd7511d520 100644 --- a/.github/workflows/master.yml +++ b/.github/workflows/master.yml @@ -33,10 +33,6 @@ jobs: # cd "$GITHUB_WORKSPACE/tests/ci" # echo "Testing the main ci directory" # python3 -m unittest discover -s . -p 'test_*.py' -# for dir in *_lambda/; do -# echo "Testing $dir" -# python3 -m unittest discover -s "$dir" -p 'test_*.py' -# done - name: PrepareRunConfig id: runconfig run: | @@ -58,7 +54,7 @@ jobs: # BuildDockers: # needs: [RunConfig] # if: ${{ !failure() && !cancelled() }} -# uses: ./.github/workflows/reusable_docker.yml +# uses: ./.github/workflows/docker_test_images.yml # with: # data: ${{ needs.RunConfig.outputs.data }} # StyleCheck: @@ -125,34 +121,6 @@ jobs: runner_type: style-checker-aarch64 data: ${{ needs.RunConfig.outputs.data }} - MarkReleaseReady: - if: ${{ !failure() && !cancelled() }} - needs: [RunConfig, Builds_1, Builds_2] - runs-on: [self-hosted, style-checker-aarch64] - steps: - - name: Debug - run: | - echo need with different filters - cat << 'EOF' - ${{ toJSON(needs) }} - ${{ toJSON(needs.*.result) }} - no failures ${{ !contains(needs.*.result, 'failure') }} - no skips ${{ !contains(needs.*.result, 'skipped') }} - no both ${{ !(contains(needs.*.result, 'skipped') || contains(needs.*.result, 'failure')) }} - EOF - - name: Not ready - # fail the job to be able to restart it - if: ${{ contains(needs.*.result, 'skipped') || contains(needs.*.result, 'failure') }} - run: exit 1 - - name: Check out repository code - if: ${{ ! (contains(needs.*.result, 'skipped') || contains(needs.*.result, 'failure')) }} - uses: ClickHouse/checkout@v1 - - name: Mark Commit Release Ready - if: ${{ ! (contains(needs.*.result, 'skipped') || contains(needs.*.result, 'failure')) }} - run: | - cd "$GITHUB_WORKSPACE/tests/ci" - python3 mark_release_ready.py - FinishCheck: if: ${{ !cancelled() }} needs: [RunConfig, Builds_1, Builds_2, Builds_Report, Tests_1, Tests_2, Tests_3] @@ -164,3 +132,10 @@ jobs: run: | cd "$GITHUB_WORKSPACE/tests/ci" python3 finish_check.py --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} + - name: Check Workflow results + run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF + python3 ./tests/ci/ci_buddy.py --check-wf-status diff --git a/.github/workflows/merge_queue.yml b/.github/workflows/merge_queue.yml index 01685ee1f5a..64083668719 100644 --- a/.github/workflows/merge_queue.yml +++ b/.github/workflows/merge_queue.yml @@ -30,10 +30,6 @@ jobs: cd "$GITHUB_WORKSPACE/tests/ci" echo "Testing the main ci directory" python3 -m unittest discover -s . -p 'test_*.py' - for dir in *_lambda/; do - echo "Testing $dir" - python3 -m unittest discover -s "$dir" -p 'test_*.py' - done - name: PrepareRunConfig id: runconfig run: | @@ -51,7 +47,7 @@ jobs: BuildDockers: needs: [RunConfig] if: ${{ !failure() && !cancelled() && toJson(fromJson(needs.RunConfig.outputs.data).docker_data.missing_multi) != '[]' }} - uses: ./.github/workflows/reusable_docker.yml + uses: ./.github/workflows/docker_test_images.yml with: data: ${{ needs.RunConfig.outputs.data }} StyleCheck: @@ -97,7 +93,7 @@ jobs: data: ${{ needs.RunConfig.outputs.data }} CheckReadyForMerge: - if: ${{ !cancelled() && needs.StyleCheck.result == 'success' }} + if: ${{ !cancelled() }} # Test_2 or Test_3 must not have jobs required for Mergeable check needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Tests_1] runs-on: [self-hosted, style-checker-aarch64] @@ -105,6 +101,14 @@ jobs: - name: Check out repository code uses: ClickHouse/checkout@v1 - name: Check and set merge status + if: ${{ needs.StyleCheck.result == 'success' }} run: | cd "$GITHUB_WORKSPACE/tests/ci" python3 merge_pr.py --set-ci-status --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} + - name: Check Workflow results + run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF + python3 ./tests/ci/ci_buddy.py --check-wf-status diff --git a/.github/workflows/nightly.yml b/.github/workflows/nightly.yml index 3e1c5576e7d..ea9c125db70 100644 --- a/.github/workflows/nightly.yml +++ b/.github/workflows/nightly.yml @@ -40,7 +40,21 @@ jobs: } >> "$GITHUB_OUTPUT" BuildDockers: needs: [RunConfig] - uses: ./.github/workflows/reusable_docker.yml + uses: ./.github/workflows/docker_test_images.yml with: data: "${{ needs.RunConfig.outputs.data }}" set_latest: true + CheckWorkflow: + if: ${{ !cancelled() }} + needs: [RunConfig, BuildDockers] + runs-on: [self-hosted, style-checker-aarch64] + steps: + - name: Check out repository code + uses: ClickHouse/checkout@v1 + - name: Check Workflow results + run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF + python3 ./tests/ci/ci_buddy.py --check-wf-status diff --git a/.github/workflows/pull_request.yml b/.github/workflows/pull_request.yml index 4764e6d3c1a..34bf51871d2 100644 --- a/.github/workflows/pull_request.yml +++ b/.github/workflows/pull_request.yml @@ -48,10 +48,6 @@ jobs: cd "$GITHUB_WORKSPACE/tests/ci" echo "Testing the main ci directory" python3 -m unittest discover -s . -p 'test_*.py' - for dir in *_lambda/; do - echo "Testing $dir" - python3 -m unittest discover -s "$dir" -p 'test_*.py' - done - name: PrepareRunConfig id: runconfig run: | @@ -72,7 +68,7 @@ jobs: BuildDockers: needs: [RunConfig] if: ${{ !failure() && !cancelled() && toJson(fromJson(needs.RunConfig.outputs.data).docker_data.missing_multi) != '[]' }} - uses: ./.github/workflows/reusable_docker.yml + uses: ./.github/workflows/docker_test_images.yml with: data: ${{ needs.RunConfig.outputs.data }} StyleCheck: @@ -155,9 +151,10 @@ jobs: data: ${{ needs.RunConfig.outputs.data }} CheckReadyForMerge: - if: ${{ !cancelled() && needs.StyleCheck.result == 'success' }} - # Test_2 or Test_3 must not have jobs required for Mergeable check - needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Builds_2, Builds_Report, Tests_1] + if: ${{ !cancelled() }} + # Test_2 or Test_3 do not have the jobs required for Mergeable check, + # however, set them as "needs" to get all checks results before the automatic merge occurs. + needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Builds_2, Builds_Report, Tests_1, Tests_2, Tests_3] runs-on: [self-hosted, style-checker-aarch64] steps: - name: Check out repository code @@ -165,14 +162,22 @@ jobs: with: filter: tree:0 - name: Check and set merge status + if: ${{ needs.StyleCheck.result == 'success' }} run: | cd "$GITHUB_WORKSPACE/tests/ci" python3 merge_pr.py --set-ci-status --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} + - name: Check Workflow results + run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF + python3 ./tests/ci/ci_buddy.py --check-wf-status ################################# Stage Final ################################# # FinishCheck: - if: ${{ !cancelled() }} + if: ${{ !failure() && !cancelled() }} needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Builds_2, Builds_Report, Tests_1, Tests_2, Tests_3] runs-on: [self-hosted, style-checker-aarch64] steps: diff --git a/.github/workflows/release_branches.yml b/.github/workflows/release_branches.yml index 3e898c69ade..b79208b03a6 100644 --- a/.github/workflows/release_branches.yml +++ b/.github/workflows/release_branches.yml @@ -33,10 +33,6 @@ jobs: cd "$GITHUB_WORKSPACE/tests/ci" echo "Testing the main ci directory" python3 -m unittest discover -s . -p 'test_*.py' - for dir in *_lambda/; do - echo "Testing $dir" - python3 -m unittest discover -s "$dir" -p 'test_*.py' - done - name: PrepareRunConfig id: runconfig run: | @@ -57,7 +53,7 @@ jobs: BuildDockers: needs: [RunConfig] if: ${{ !failure() && !cancelled() }} - uses: ./.github/workflows/reusable_docker.yml + uses: ./.github/workflows/docker_test_images.yml with: data: ${{ needs.RunConfig.outputs.data }} CompatibilityCheckX86: @@ -184,10 +180,16 @@ jobs: steps: - name: Check out repository code uses: ClickHouse/checkout@v1 + - name: Download reports + run: | + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(needs.RunConfig.outputs.data) }} --pre --job-name Builds - name: Builds report run: | cd "$GITHUB_WORKSPACE/tests/ci" python3 ./build_report_check.py --reports package_release package_aarch64 package_asan package_msan package_ubsan package_tsan package_debug binary_darwin binary_darwin_aarch64 + - name: Set status + run: | + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(needs.RunConfig.outputs.data) }} --post --job-name Builds MarkReleaseReady: if: ${{ !failure() && !cancelled() }} needs: @@ -439,8 +441,9 @@ jobs: runner_type: stress-tester data: ${{ needs.RunConfig.outputs.data }} FinishCheck: - if: ${{ !failure() && !cancelled() }} + if: ${{ !cancelled() }} needs: + - RunConfig - DockerServerImage - DockerKeeperImage - Builds_Report @@ -476,9 +479,18 @@ jobs: with: clear-repository: true - name: Finish label + if: ${{ !failure() }} run: | cd "$GITHUB_WORKSPACE/tests/ci" # update mergeable check python3 merge_pr.py --set-ci-status --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} # update overall ci report python3 finish_check.py --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} + - name: Check Workflow results + run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF + + python3 ./tests/ci/ci_buddy.py --check-wf-status diff --git a/.github/workflows/reusable_test.yml b/.github/workflows/reusable_test.yml index c01dd8ca9d4..ca6df5df14e 100644 --- a/.github/workflows/reusable_test.yml +++ b/.github/workflows/reusable_test.yml @@ -102,6 +102,8 @@ jobs: --job-name '${{inputs.test_name}}' \ --run \ --run-command '''${{inputs.run_command}}''' + # shellcheck disable=SC2319 + echo "JOB_EXIT_CODE=$?" >> "$GITHUB_ENV" - name: Post run if: ${{ !cancelled() }} run: | diff --git a/.yamllint b/.yamllint index 9d6550ac960..7fb741ec9f4 100644 --- a/.yamllint +++ b/.yamllint @@ -13,5 +13,10 @@ rules: level: warning comments: min-spaces-from-content: 1 - document-start: - present: false + document-start: disable + colons: disable + indentation: disable + line-length: disable + trailing-spaces: disable + truthy: disable + new-line-at-end-of-file: disable diff --git a/CHANGELOG.md b/CHANGELOG.md index c4935f88245..07b37835dda 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,5 +1,6 @@ ### Table of Contents -**[ClickHouse release v24.6, 2024-06-27](#246)**
+**[ClickHouse release v24.7, 2024-07-30](#247)**
+**[ClickHouse release v24.6, 2024-07-01](#246)**
**[ClickHouse release v24.5, 2024-05-30](#245)**
**[ClickHouse release v24.4, 2024-04-30](#244)**
**[ClickHouse release v24.3 LTS, 2024-03-26](#243)**
@@ -9,107 +10,277 @@ # 2024 Changelog -### ClickHouse release 24.6, 2024-06-27 +### ClickHouse release 24.7, 2024-07-30 + +#### Backward Incompatible Change +* Forbid `CRATE MATERIALIZED VIEW ... ENGINE Replicated*MergeTree POPULATE AS SELECT ...` with Replicated databases. [#63963](https://github.com/ClickHouse/ClickHouse/pull/63963) ([vdimir](https://github.com/vdimir)). +* `clickhouse-keeper-client` will only accept paths in string literals, such as `ls '/hello/world'`, not bare strings such as `ls /hello/world`. [#65494](https://github.com/ClickHouse/ClickHouse/pull/65494) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Metric `KeeperOutstandingRequets` was renamed to `KeeperOutstandingRequests`. [#66206](https://github.com/ClickHouse/ClickHouse/pull/66206) ([Robert Schulze](https://github.com/rschu1ze)). +* Remove `is_deterministic` field from the `system.functions` table. [#66630](https://github.com/ClickHouse/ClickHouse/pull/66630) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Function `tuple` will now try to construct named tuples in query (controlled by `enable_named_columns_in_function_tuple`). Introduce function `tupleNames` to extract names from tuples. [#54881](https://github.com/ClickHouse/ClickHouse/pull/54881) ([Amos Bird](https://github.com/amosbird)). + +#### New Feature +* Add `ASOF JOIN` support for `full_sorting_join` algorithm. [#55051](https://github.com/ClickHouse/ClickHouse/pull/55051) ([vdimir](https://github.com/vdimir)). +* Add new window function `percent_rank`. [#62747](https://github.com/ClickHouse/ClickHouse/pull/62747) ([lgbo](https://github.com/lgbo-ustc)). +* Support JWT authentication in `clickhouse-client` (will be available only in ClickHouse Cloud). [#62829](https://github.com/ClickHouse/ClickHouse/pull/62829) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Add SQL functions `changeYear`, `changeMonth`, `changeDay`, `changeHour`, `changeMinute`, `changeSecond`. For example, `SELECT changeMonth(toDate('2024-06-14'), 7)` returns date `2024-07-14`. [#63186](https://github.com/ClickHouse/ClickHouse/pull/63186) ([cucumber95](https://github.com/cucumber95)). +* Introduce startup scripts, which allow the execution of preconfigured queries at the startup stage. [#64889](https://github.com/ClickHouse/ClickHouse/pull/64889) ([pufit](https://github.com/pufit)). +* Support accept_invalid_certificate in client's config in order to allow for client to connect over secure TCP to a server running with self-signed certificate - can be used as a shorthand for corresponding `openSSL` client settings `verificationMode=none` + `invalidCertificateHandler.name=AcceptCertificateHandler`. [#65238](https://github.com/ClickHouse/ClickHouse/pull/65238) ([peacewalker122](https://github.com/peacewalker122)). +* Add system.error_log which contains history of error values from table system.errors, periodically flushed to disk. [#65381](https://github.com/ClickHouse/ClickHouse/pull/65381) ([Pablo Marcos](https://github.com/pamarcos)). +* Add aggregate function `groupConcat`. About the same as `arrayStringConcat( groupArray(column), ',')` Can receive 2 parameters: a string delimiter and the number of elements to be processed. [#65451](https://github.com/ClickHouse/ClickHouse/pull/65451) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Add AzureQueue storage. [#65458](https://github.com/ClickHouse/ClickHouse/pull/65458) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Add a new setting to disable/enable writing page index into parquet files. [#65475](https://github.com/ClickHouse/ClickHouse/pull/65475) ([lgbo](https://github.com/lgbo-ustc)). +* Introduce `logger.console_log_level` server config to control the log level to the console (if enabled). [#65559](https://github.com/ClickHouse/ClickHouse/pull/65559) ([Azat Khuzhin](https://github.com/azat)). +* Automatically append a wildcard `*` to the end of a directory path with table function `file`. [#66019](https://github.com/ClickHouse/ClickHouse/pull/66019) ([Zhidong (David) Guo](https://github.com/Gun9niR)). +* Add `--memory-usage` option to client in non interactive mode. [#66393](https://github.com/ClickHouse/ClickHouse/pull/66393) ([vdimir](https://github.com/vdimir)). +* Make an interactive client for clickhouse-disks, add local disk from the local directory. [#64446](https://github.com/ClickHouse/ClickHouse/pull/64446) ([Daniil Ivanik](https://github.com/divanik)). +* When lightweight delete happens on a table with projection(s), users have choices either throw an exception (by default) or drop the projection [#65594](https://github.com/ClickHouse/ClickHouse/pull/65594) ([jsc0218](https://github.com/jsc0218)). + +#### Experimental Feature +* Change binary serialization of Variant data type: add `compact` mode to avoid writing the same discriminator multiple times for granules with single variant or with only NULL values. Add MergeTree setting `use_compact_variant_discriminators_serialization` that is enabled by default. Note that Variant type is still experimental and backward-incompatible change in serialization is ok. [#62774](https://github.com/ClickHouse/ClickHouse/pull/62774) ([Kruglov Pavel](https://github.com/Avogar)). +* Support rocksdb as backend storage of keeper. [#56626](https://github.com/ClickHouse/ClickHouse/pull/56626) ([Han Fei](https://github.com/hanfei1991)). +* Refactor JSONExtract functions, support more types including experimental Dynamic type. [#66046](https://github.com/ClickHouse/ClickHouse/pull/66046) ([Kruglov Pavel](https://github.com/Avogar)). +* Support null map subcolumn for Variant and Dynamic subcolumns. [#66178](https://github.com/ClickHouse/ClickHouse/pull/66178) ([Kruglov Pavel](https://github.com/Avogar)). +* Fix reading dynamic subcolumns from altered Memory table. Previously if `max_types` parameter of a Dynamic type was changed in Memory table via alter, further subcolumns reading can return wrong result. [#66066](https://github.com/ClickHouse/ClickHouse/pull/66066) ([Kruglov Pavel](https://github.com/Avogar)). +* Add support for `cluster_for_parallel_replicas` when using custom key parallel replicas. It allows you to use parallel replicas with custom key with MergeTree tables. [#65453](https://github.com/ClickHouse/ClickHouse/pull/65453) ([Antonio Andelic](https://github.com/antonio2368)). + +#### Performance Improvement +* Enable `optimize_functions_to_subcolumns` by default. [#58661](https://github.com/ClickHouse/ClickHouse/pull/58661) ([Anton Popov](https://github.com/CurtizJ)). +* Replace int to string algorithm with a faster one (from a modified amdn/itoa to a modified jeaiii/itoa). [#61661](https://github.com/ClickHouse/ClickHouse/pull/61661) ([Raúl Marín](https://github.com/Algunenano)). +* Sizes of hash tables created by join (`parallel_hash` algorithm) is collected and cached now. This information will be used to preallocate space in hash tables for subsequent query executions and save time on hash table resizes. [#64553](https://github.com/ClickHouse/ClickHouse/pull/64553) ([Nikita Taranov](https://github.com/nickitat)). +* Optimized queries with `ORDER BY` primary key and `WHERE` that have a condition with high selectivity by using of buffering. It is controlled by setting `read_in_order_use_buffering` (enabled by default) and can increase memory usage of query. [#64607](https://github.com/ClickHouse/ClickHouse/pull/64607) ([Anton Popov](https://github.com/CurtizJ)). +* Improve performance of loading `plain_rewritable` metadata. [#65634](https://github.com/ClickHouse/ClickHouse/pull/65634) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Attaching tables on read-only disks will use fewer resources by not loading outdated parts. [#65635](https://github.com/ClickHouse/ClickHouse/pull/65635) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Support minmax hyperrectangle for Set indices. [#65676](https://github.com/ClickHouse/ClickHouse/pull/65676) ([AntiTopQuark](https://github.com/AntiTopQuark)). +* Unload primary index of outdated parts to reduce total memory usage. [#65852](https://github.com/ClickHouse/ClickHouse/pull/65852) ([Anton Popov](https://github.com/CurtizJ)). +* Functions `replaceRegexpAll` and `replaceRegexpOne` are now significantly faster if the pattern is trivial, i.e. contains no metacharacters, pattern classes, flags, grouping characters etc. (Thanks to Taiyang Li). [#66185](https://github.com/ClickHouse/ClickHouse/pull/66185) ([Robert Schulze](https://github.com/rschu1ze)). +* s3 requests: Reduce retry time for queries, increase retries count for backups. 8.5 minutes and 100 retires for queries, 1.2 hours and 1000 retries for backup restore. [#65232](https://github.com/ClickHouse/ClickHouse/pull/65232) ([Sema Checherinda](https://github.com/CheSema)). +* Support query plan LIMIT optimization. Support LIMIT pushdown for PostgreSQL storage and table function. [#65454](https://github.com/ClickHouse/ClickHouse/pull/65454) ([Maksim Kita](https://github.com/kitaisreal)). +* Improved ZooKeeper load balancing. The current session doesn't expire until the optimal nodes become available despite `fallback_session_lifetime`. Added support for AZ-aware balancing. [#65570](https://github.com/ClickHouse/ClickHouse/pull/65570) ([Alexander Tokmakov](https://github.com/tavplubix)). +* DatabaseCatalog drops tables faster by using up to database_catalog_drop_table_concurrency threads. [#66065](https://github.com/ClickHouse/ClickHouse/pull/66065) ([Sema Checherinda](https://github.com/CheSema)). + +#### Improvement +* The setting `optimize_trivial_insert_select` is disabled by default. In most cases, it should be beneficial. Nevertheless, if you are seeing slower INSERT SELECT or increased memory usage, you can enable it back or `SET compatibility = '24.6'`. [#58970](https://github.com/ClickHouse/ClickHouse/pull/58970) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Print stacktrace and diagnostic info if `clickhouse-client` or `clickhouse-local` crashes. [#61109](https://github.com/ClickHouse/ClickHouse/pull/61109) ([Alexander Tokmakov](https://github.com/tavplubix)). +* The result of `SHOW INDEX | INDEXES | INDICES | KEYS` was previously sorted by the primary key column names. Since this was unintuitive, the result is now sorted by the position of the primary key columns within the primary key. [#61131](https://github.com/ClickHouse/ClickHouse/pull/61131) ([Robert Schulze](https://github.com/rschu1ze)). +* Change how deduplication for Materialized Views works. Fixed a lot of cases like: - on destination table: data is split for 2 or more blocks and that blocks is considered as duplicate when that block is inserted in parallel. - on MV destination table: the equal blocks are deduplicated, that happens when MV often produces equal data as a result for different input data due to performing aggregation. - on MV destination table: the equal blocks which comes from different MV are deduplicated. [#61601](https://github.com/ClickHouse/ClickHouse/pull/61601) ([Sema Checherinda](https://github.com/CheSema)). +* Allow matching column names in a case insensitive manner when reading json files (`input_format_json_case_insensitive_column_matching`). [#61750](https://github.com/ClickHouse/ClickHouse/pull/61750) ([kevinyhzou](https://github.com/KevinyhZou)). +* Support reading partitioned data DeltaLake data. Infer DeltaLake schema by reading metadata instead of data. [#63201](https://github.com/ClickHouse/ClickHouse/pull/63201) ([Kseniia Sumarokova](https://github.com/kssenii)). +* In composable protocols TLS layer accepted only `certificateFile` and `privateKeyFile` parameters. https://clickhouse.com/docs/en/operations/settings/composable-protocols. [#63985](https://github.com/ClickHouse/ClickHouse/pull/63985) ([Anton Ivashkin](https://github.com/ianton-ru)). +* Added profile event `SelectQueriesWithPrimaryKeyUsage` which indicates how many SELECT queries use the primary key to evaluate the WHERE clause. [#64492](https://github.com/ClickHouse/ClickHouse/pull/64492) ([0x01f](https://github.com/0xfei)). +* `StorageS3Queue` related fixes and improvements. Deduce a default value of `s3queue_processing_threads_num` according to the number of physical cpu cores on the server (instead of the previous default value as 1). Set default value of `s3queue_loading_retries` to 10. Fix possible vague "Uncaught exception" in exception column of `system.s3queue`. Do not increment retry count on `MEMORY_LIMIT_EXCEEDED` exception. Move files commit to a stage after insertion into table fully finished to avoid files being commited while not inserted. Add settings `s3queue_max_processed_files_before_commit`, `s3queue_max_processed_rows_before_commit`, `s3queue_max_processed_bytes_before_commit`, `s3queue_max_processing_time_sec_before_commit`, to better control commit and flush time. [#65046](https://github.com/ClickHouse/ClickHouse/pull/65046) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Support aliases in parametrized view function (only new analyzer). [#65190](https://github.com/ClickHouse/ClickHouse/pull/65190) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Updated to mask account key in logs in azureBlobStorage. [#65273](https://github.com/ClickHouse/ClickHouse/pull/65273) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Partition pruning for `IN` predicates when filter expression is a part of `PARTITION BY` expression. [#65335](https://github.com/ClickHouse/ClickHouse/pull/65335) ([Eduard Karacharov](https://github.com/korowa)). +* Add system tables with main information about all detached tables. [#65400](https://github.com/ClickHouse/ClickHouse/pull/65400) ([Konstantin Morozov](https://github.com/k-morozov)). +* `arrayMin`/`arrayMax` can be applicable to all data types that are comparable. [#65455](https://github.com/ClickHouse/ClickHouse/pull/65455) ([pn](https://github.com/chloro-pn)). +* Improved memory accounting for cgroups v2 to exclude the amount occupied by the page cache. [#65470](https://github.com/ClickHouse/ClickHouse/pull/65470) ([Nikita Taranov](https://github.com/nickitat)). +* Do not create format settings for each row when serializing chunks to insert to EmbeddedRocksDB table. [#65474](https://github.com/ClickHouse/ClickHouse/pull/65474) ([Duc Canh Le](https://github.com/canhld94)). +* Reduce `clickhouse-local` prompt to just `:)`. `getFQDNOrHostName()` takes too long on macOS, and we don't want a hostname in the prompt for `clickhouse-local` anyway. [#65510](https://github.com/ClickHouse/ClickHouse/pull/65510) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Avoid printing a message from jemalloc about per-CPU arenas on low-end virtual machines. [#65532](https://github.com/ClickHouse/ClickHouse/pull/65532) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Disable filesystem cache background download by default. It will be enabled back when we fix the issue with possible "Memory limit exceeded" because memory deallocation is done outside of query context (while buffer is allocated inside of query context) if we use background download threads. Plus we need to add a separate setting to define max size to download for background workers (currently it is limited by max_file_segment_size, which might be too big). [#65534](https://github.com/ClickHouse/ClickHouse/pull/65534) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Add new option to config `` which allow to specify how often clickhouse will reload config. [#65545](https://github.com/ClickHouse/ClickHouse/pull/65545) ([alesapin](https://github.com/alesapin)). +* Implement binary encoding for ClickHouse data types and add its specification in docs. Use it in Dynamic binary serialization, allow to use it in RowBinaryWithNamesAndTypes and Native formats under settings. [#65546](https://github.com/ClickHouse/ClickHouse/pull/65546) ([Kruglov Pavel](https://github.com/Avogar)). +* Improved ZooKeeper load balancing. The current session doesn't expire until the optimal nodes become available despite `fallback_session_lifetime`. Added support for AZ-aware balancing. [#65570](https://github.com/ClickHouse/ClickHouse/pull/65570) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Server settings `compiled_expression_cache_size` and `compiled_expression_cache_elements_size` are now shown in `system.server_settings`. [#65584](https://github.com/ClickHouse/ClickHouse/pull/65584) ([Robert Schulze](https://github.com/rschu1ze)). +* Add support for user identification based on x509 SubjectAltName extension. [#65626](https://github.com/ClickHouse/ClickHouse/pull/65626) ([Anton Kozlov](https://github.com/tonickkozlov)). +* `clickhouse-local` will respect the `max_server_memory_usage` and `max_server_memory_usage_to_ram_ratio` from the configuration file. It will also set the max memory usage to 90% of the system memory by default, like `clickhouse-server` does. [#65697](https://github.com/ClickHouse/ClickHouse/pull/65697) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add a script to backup your files to ClickHouse. [#65699](https://github.com/ClickHouse/ClickHouse/pull/65699) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* PostgreSQL source support cancel. [#65722](https://github.com/ClickHouse/ClickHouse/pull/65722) ([Maksim Kita](https://github.com/kitaisreal)). +* Make allow_experimental_analyzer be controlled by the initiator for distributed queries. This ensures compatibility and correctness during operations in mixed version clusters. [#65777](https://github.com/ClickHouse/ClickHouse/pull/65777) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Respect cgroup CPU limit in Keeper. [#65819](https://github.com/ClickHouse/ClickHouse/pull/65819) ([Antonio Andelic](https://github.com/antonio2368)). +* Allow to use `concat` function with empty arguments ``` sql :) select concat();. [#65887](https://github.com/ClickHouse/ClickHouse/pull/65887) ([李扬](https://github.com/taiyang-li)). +* Allow controlling named collections in clickhouse-local. [#65973](https://github.com/ClickHouse/ClickHouse/pull/65973) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Improve Azure profile events. [#65999](https://github.com/ClickHouse/ClickHouse/pull/65999) ([alesapin](https://github.com/alesapin)). +* Support ORC file read by writer time zone. [#66025](https://github.com/ClickHouse/ClickHouse/pull/66025) ([kevinyhzou](https://github.com/KevinyhZou)). +* Add settings to control connection to the PostgreSQL. * Setting `postgresql_connection_attempt_timeout` specifies the value passed to `connect_timeout` parameter of connection URL. * Setting `postgresql_connection_pool_retries` specifies the number of retries to establish a connection to the PostgreSQL end-point. [#66232](https://github.com/ClickHouse/ClickHouse/pull/66232) ([Dmitry Novik](https://github.com/novikd)). +* Reduce inaccuracy of input_wait_elapsed_us/input_wait_elapsed_us/elapsed_us. [#66239](https://github.com/ClickHouse/ClickHouse/pull/66239) ([Azat Khuzhin](https://github.com/azat)). +* Improve FilesystemCache ProfileEvents. [#66249](https://github.com/ClickHouse/ClickHouse/pull/66249) ([zhukai](https://github.com/nauu)). +* Add settings to ignore ON CLUSTER clause in queries for named collection management with replicated storage. [#66288](https://github.com/ClickHouse/ClickHouse/pull/66288) ([MikhailBurdukov](https://github.com/MikhailBurdukov)). +* Function `generateSnowflakeID` now allows to specify a machine ID as a parameter to prevent collisions in large clusters. [#66374](https://github.com/ClickHouse/ClickHouse/pull/66374) ([ZAWA_ll](https://github.com/Zawa-ll)). +* Disable suspending on Ctrl+Z in interactive mode. This is a common trap and is not expected behavior for almost all users. I imagine only a few extreme power users could appreciate suspending terminal applications to the background, but I don't know any. [#66511](https://github.com/ClickHouse/ClickHouse/pull/66511) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add option for validating the Primary key type in Dictionaries. Without this option for simple layouts any column type will be implicitly converted to UInt64. ### Documentation entry for user-facing changes. [#66595](https://github.com/ClickHouse/ClickHouse/pull/66595) ([MikhailBurdukov](https://github.com/MikhailBurdukov)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Fix unexpected size of low cardinality column in function calls. [#65298](https://github.com/ClickHouse/ClickHouse/pull/65298) ([Raúl Marín](https://github.com/Algunenano)). +* Check cyclic dependencies on CREATE/REPLACE/RENAME/EXCHANGE queries and throw an exception if there is a cyclic dependency. Previously such cyclic dependencies could lead to a deadlock during server startup. Also fix some bugs in dependencies creation. [#65405](https://github.com/ClickHouse/ClickHouse/pull/65405) ([Kruglov Pavel](https://github.com/Avogar)). +* Fix crash in maxIntersections. [#65689](https://github.com/ClickHouse/ClickHouse/pull/65689) ([Raúl Marín](https://github.com/Algunenano)). +* Fix the VALID UNTIL clause in the user definition resetting after a restart. [#66409](https://github.com/ClickHouse/ClickHouse/pull/66409) ([Nikolay Degterinsky](https://github.com/evillique)). +* Fix SHOW MERGES remaining time. [#66735](https://github.com/ClickHouse/ClickHouse/pull/66735) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* `Query was cancelled` might have been printed twice in clickhouse-client. This behaviour is fixed. [#66005](https://github.com/ClickHouse/ClickHouse/pull/66005) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Fixed crash while using MaterializedMySQL with TABLE OVERRIDE that maps MySQL NULL field into ClickHouse not NULL field. [#54649](https://github.com/ClickHouse/ClickHouse/pull/54649) ([Filipp Ozinov](https://github.com/bakwc)). +* Fix logical error when PREWHERE expression read no columns and table has no adaptive index granularity (very old table). [#59173](https://github.com/ClickHouse/ClickHouse/pull/59173) ([Alexander Gololobov](https://github.com/davenger)). +* Fix bug with cancellation buffer when canceling a query. [#64478](https://github.com/ClickHouse/ClickHouse/pull/64478) ([Sema Checherinda](https://github.com/CheSema)). +* Fix filling parts columns from metadata (when columns.txt does not exists). [#64757](https://github.com/ClickHouse/ClickHouse/pull/64757) ([Azat Khuzhin](https://github.com/azat)). +* Fix crash for `ALTER TABLE ... ON CLUSTER ... MODIFY SQL SECURITY`. [#64957](https://github.com/ClickHouse/ClickHouse/pull/64957) ([pufit](https://github.com/pufit)). +* Fix crash on destroying AccessControl: add explicit shutdown. [#64993](https://github.com/ClickHouse/ClickHouse/pull/64993) ([Vitaly Baranov](https://github.com/vitlibar)). +* Eliminate injective function in argument of functions `uniq*` recursively. This used to work correctly but was broken in the new analyzer. [#65140](https://github.com/ClickHouse/ClickHouse/pull/65140) ([Duc Canh Le](https://github.com/canhld94)). +* Fix unexpected projection name when query with CTE. [#65267](https://github.com/ClickHouse/ClickHouse/pull/65267) ([wudidapaopao](https://github.com/wudidapaopao)). +* Require `dictGet` privilege when accessing dictionaries via direct query or the `Dictionary` table engine. [#65359](https://github.com/ClickHouse/ClickHouse/pull/65359) ([Joe Lynch](https://github.com/joelynch)). +* Fix user-specific S3 auth with incremental backups. [#65481](https://github.com/ClickHouse/ClickHouse/pull/65481) ([Antonio Andelic](https://github.com/antonio2368)). +* Disable `non-intersecting-parts` optimization for queries with `FINAL` in case of `read-in-order` optimization was enabled. This could lead to an incorrect query result. As a workaround, disable `do_not_merge_across_partitions_select_final` and `split_parts_ranges_into_intersecting_and_non_intersecting_final` before this fix is merged. [#65505](https://github.com/ClickHouse/ClickHouse/pull/65505) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix getting exception `Index out of bound for blob metadata` in case all files from list batch were filtered out. [#65523](https://github.com/ClickHouse/ClickHouse/pull/65523) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix NOT_FOUND_COLUMN_IN_BLOCK for deduplicate merge of projection. [#65573](https://github.com/ClickHouse/ClickHouse/pull/65573) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)). +* Fixed bug in MergeJoin. Column in sparse serialisation might be treated as a column of its nested type though the required conversion wasn't performed. [#65632](https://github.com/ClickHouse/ClickHouse/pull/65632) ([Nikita Taranov](https://github.com/nickitat)). +* Fixed a bug that compatibility level '23.4' was not properly applied. [#65737](https://github.com/ClickHouse/ClickHouse/pull/65737) ([cw5121](https://github.com/cw5121)). +* Fix odbc table with nullable fields. [#65738](https://github.com/ClickHouse/ClickHouse/pull/65738) ([Rodolphe Dugé de Bernonville](https://github.com/RodolpheDuge)). +* Fix data race in `TCPHandler`, which could happen on fatal error. [#65744](https://github.com/ClickHouse/ClickHouse/pull/65744) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix invalid exceptions in function `parseDateTime` with `%F` and `%D` placeholders. [#65768](https://github.com/ClickHouse/ClickHouse/pull/65768) ([Antonio Andelic](https://github.com/antonio2368)). +* For queries that read from `PostgreSQL`, cancel the internal `PostgreSQL` query if the ClickHouse query is finished. Otherwise, `ClickHouse` query cannot be canceled until the internal `PostgreSQL` query is finished. [#65771](https://github.com/ClickHouse/ClickHouse/pull/65771) ([Maksim Kita](https://github.com/kitaisreal)). +* Fix a bug in short circuit logic when old analyzer and dictGetOrDefault is used. [#65802](https://github.com/ClickHouse/ClickHouse/pull/65802) ([jsc0218](https://github.com/jsc0218)). +* Fix a bug leads to EmbeddedRocksDB with TTL write corrupted SST files. [#65816](https://github.com/ClickHouse/ClickHouse/pull/65816) ([Duc Canh Le](https://github.com/canhld94)). +* Functions `bitTest`, `bitTestAll`, and `bitTestAny` now return an error if the specified bit index is out-of-bounds [#65818](https://github.com/ClickHouse/ClickHouse/pull/65818) ([Pablo Marcos](https://github.com/pamarcos)). +* Setting `join_any_take_last_row` is supported in any query with hash join. [#65820](https://github.com/ClickHouse/ClickHouse/pull/65820) ([vdimir](https://github.com/vdimir)). +* Better handling of join conditions involving `IS NULL` checks (for example `ON (a = b AND (a IS NOT NULL) AND (b IS NOT NULL) ) OR ( (a IS NULL) AND (b IS NULL) )` is rewritten to `ON a <=> b`), fix incorrect optimization when condition other then `IS NULL` are present. [#65835](https://github.com/ClickHouse/ClickHouse/pull/65835) ([vdimir](https://github.com/vdimir)). +* Functions `bitShiftLeft` and `bitShitfRight` return an error for out of bounds shift positions [#65838](https://github.com/ClickHouse/ClickHouse/pull/65838) ([Pablo Marcos](https://github.com/pamarcos)). +* Fix growing memory usage in S3Queue. [#65839](https://github.com/ClickHouse/ClickHouse/pull/65839) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix tie handling in `arrayAUC` to match sklearn. [#65840](https://github.com/ClickHouse/ClickHouse/pull/65840) ([gabrielmcg44](https://github.com/gabrielmcg44)). +* Fix possible issues with MySQL server protocol TLS connections. [#65917](https://github.com/ClickHouse/ClickHouse/pull/65917) ([Azat Khuzhin](https://github.com/azat)). +* Fix possible issues with MySQL client protocol TLS connections. [#65938](https://github.com/ClickHouse/ClickHouse/pull/65938) ([Azat Khuzhin](https://github.com/azat)). +* Fix handling of `SSL_ERROR_WANT_READ`/`SSL_ERROR_WANT_WRITE` with zero timeout. [#65941](https://github.com/ClickHouse/ClickHouse/pull/65941) ([Azat Khuzhin](https://github.com/azat)). +* Add missing settings `input_format_csv_skip_first_lines/input_format_tsv_skip_first_lines/input_format_csv_try_infer_numbers_from_strings/input_format_csv_try_infer_strings_from_quoted_tuples` in schema inference cache because they can change the resulting schema. It prevents from incorrect result of schema inference with these settings changed. [#65980](https://github.com/ClickHouse/ClickHouse/pull/65980) ([Kruglov Pavel](https://github.com/Avogar)). +* Column _size in s3 engine and s3 table function denotes the size of a file inside the archive, not a size of the archive itself. [#65993](https://github.com/ClickHouse/ClickHouse/pull/65993) ([Daniil Ivanik](https://github.com/divanik)). +* Fix resolving dynamic subcolumns in analyzer, avoid reading the whole column on dynamic subcolumn reading. [#66004](https://github.com/ClickHouse/ClickHouse/pull/66004) ([Kruglov Pavel](https://github.com/Avogar)). +* Fix config merging for from_env with replace overrides. [#66034](https://github.com/ClickHouse/ClickHouse/pull/66034) ([Azat Khuzhin](https://github.com/azat)). +* Fix a possible hanging in `GRPCServer` during shutdown. [#66061](https://github.com/ClickHouse/ClickHouse/pull/66061) ([Vitaly Baranov](https://github.com/vitlibar)). +* Fixed several cases in function `has` with non-constant `LowCardinality` arguments. [#66088](https://github.com/ClickHouse/ClickHouse/pull/66088) ([Anton Popov](https://github.com/CurtizJ)). +* Fix for `groupArrayIntersect`. It had incorrect behavior in the `merge()` function. Also, fixed behavior in `deserialise()` for numeric and general data. [#66103](https://github.com/ClickHouse/ClickHouse/pull/66103) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Fixed buffer overflow bug in `unbin`/`unhex` implementation. [#66106](https://github.com/ClickHouse/ClickHouse/pull/66106) ([Nikita Taranov](https://github.com/nickitat)). +* Disable the `merge-filters` optimization introduced in [#64760](https://github.com/ClickHouse/ClickHouse/issues/64760). It may cause an exception if optimization merges two filter expressions and does not apply a short-circuit evaluation. [#66126](https://github.com/ClickHouse/ClickHouse/pull/66126) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fixed the issue when the server failed to parse Avro files with negative block size arrays encoded, which is now allowed by the Avro specification. [#66130](https://github.com/ClickHouse/ClickHouse/pull/66130) ([Serge Klochkov](https://github.com/slvrtrn)). +* Fixed a bug in ZooKeeper client: a session could get stuck in unusable state after receiving a hardware error from ZooKeeper. For example, this might happen due to "soft memory limit" in ClickHouse Keeper. [#66140](https://github.com/ClickHouse/ClickHouse/pull/66140) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Fix issue in SumIfToCountIfVisitor and signed integers. [#66146](https://github.com/ClickHouse/ClickHouse/pull/66146) ([Raúl Marín](https://github.com/Algunenano)). +* Fix rare case with missing data in the result of distributed query. [#66174](https://github.com/ClickHouse/ClickHouse/pull/66174) ([vdimir](https://github.com/vdimir)). +* Fix order of parsing metadata fields in StorageDeltaLake. [#66211](https://github.com/ClickHouse/ClickHouse/pull/66211) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Don't throw `TIMEOUT_EXCEEDED` for `none_only_active` mode of `distributed_ddl_output_mode`. [#66218](https://github.com/ClickHouse/ClickHouse/pull/66218) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Fix handling limit for `system.numbers_mt` when no index can be used. [#66231](https://github.com/ClickHouse/ClickHouse/pull/66231) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Fixed how the ClickHouse server detects the maximum number of usable CPU cores as specified by cgroups v2 if the server runs in a container such as Docker. In more detail, containers often run their process in the root cgroup which has an empty name. In that case, ClickHouse ignored the CPU limits set by cgroups v2. [#66237](https://github.com/ClickHouse/ClickHouse/pull/66237) ([filimonov](https://github.com/filimonov)). +* Fix the `Not-ready set` error when a subquery with `IN` is used in the constraint. [#66261](https://github.com/ClickHouse/ClickHouse/pull/66261) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix error reporting while copying to S3 or AzureBlobStorage. [#66295](https://github.com/ClickHouse/ClickHouse/pull/66295) ([Vitaly Baranov](https://github.com/vitlibar)). +* Prevent watchdog from keeping descriptors of unlinked(rotated) log files. [#66334](https://github.com/ClickHouse/ClickHouse/pull/66334) ([Aleksei Filatov](https://github.com/aalexfvk)). +* Fix the bug that logicalexpressionoptimizerpass lost logical type of constant. [#66344](https://github.com/ClickHouse/ClickHouse/pull/66344) ([pn](https://github.com/chloro-pn)). +* Fix `Column identifier is already registered` error with `group_by_use_nulls=true` and new analyzer. [#66400](https://github.com/ClickHouse/ClickHouse/pull/66400) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix possible incorrect result for queries joining and filtering table external engine (like PostgreSQL), due to too aggressive filter pushdown. Since now, conditions from where section won't be send to external database in case of outer join with external table. [#66402](https://github.com/ClickHouse/ClickHouse/pull/66402) ([vdimir](https://github.com/vdimir)). +* Added missing column materialization for cross join. [#66413](https://github.com/ClickHouse/ClickHouse/pull/66413) ([lgbo](https://github.com/lgbo-ustc)). +* Fix `Cannot find column` error for queries with constant expression in `GROUP BY` key and new analyzer enabled. [#66433](https://github.com/ClickHouse/ClickHouse/pull/66433) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Avoid possible logical error during import from Npy format in case of bad array nesting level, fix testing of other kinds of errors. [#66461](https://github.com/ClickHouse/ClickHouse/pull/66461) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Fix wrong count() result when there is non-deterministic function in predicate. [#66510](https://github.com/ClickHouse/ClickHouse/pull/66510) ([Duc Canh Le](https://github.com/canhld94)). +* Correctly track memory for `Allocator::realloc`. [#66548](https://github.com/ClickHouse/ClickHouse/pull/66548) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix reading of uninitialized memory when hashing empty tuples. [#66562](https://github.com/ClickHouse/ClickHouse/pull/66562) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix an invalid result for queries with `WINDOW`. This could happen when `PARTITION` columns have sparse serialization and window functions are executed in parallel. [#66579](https://github.com/ClickHouse/ClickHouse/pull/66579) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix removing named collections in local storage. [#66599](https://github.com/ClickHouse/ClickHouse/pull/66599) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Fix `column_length` is not updated in `ColumnTuple::insertManyFrom`. [#66626](https://github.com/ClickHouse/ClickHouse/pull/66626) ([lgbo](https://github.com/lgbo-ustc)). +* Fix `Unknown identifier` and `Column is not under aggregate function` errors for queries with the expression `(column IS NULL).` The bug was triggered by [#65088](https://github.com/ClickHouse/ClickHouse/issues/65088), with the disabled analyzer only. [#66654](https://github.com/ClickHouse/ClickHouse/pull/66654) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix `Method getResultType is not supported for QUERY query node` error when scalar subquery was used as the first argument of IN (with new analyzer). [#66655](https://github.com/ClickHouse/ClickHouse/pull/66655) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix possible PARAMETER_OUT_OF_BOUND error during reading variant subcolumn. [#66659](https://github.com/ClickHouse/ClickHouse/pull/66659) ([Kruglov Pavel](https://github.com/Avogar)). +* Fix rare case of stuck merge after drop column. [#66707](https://github.com/ClickHouse/ClickHouse/pull/66707) ([Raúl Marín](https://github.com/Algunenano)). +* Fix assertion `isUniqTypes` when insert select from remote sources. [#66722](https://github.com/ClickHouse/ClickHouse/pull/66722) ([Sema Checherinda](https://github.com/CheSema)). +* Fix logical error in PrometheusRequestHandler. [#66621](https://github.com/ClickHouse/ClickHouse/pull/66621) ([Vitaly Baranov](https://github.com/vitlibar)). +* Fix `indexHint` function case found by fuzzer. [#66286](https://github.com/ClickHouse/ClickHouse/pull/66286) ([Anton Popov](https://github.com/CurtizJ)). +* Fix AST formatting of 'create table b empty as a'. [#64951](https://github.com/ClickHouse/ClickHouse/pull/64951) ([Michael Kolupaev](https://github.com/al13n321)). + +#### Build/Testing/Packaging Improvement +* Instantiate template methods ahead in different .cpp files, avoid too large translation units during compiling. [#64818](https://github.com/ClickHouse/ClickHouse/pull/64818) ([lgbo](https://github.com/lgbo-ustc)). + +### ClickHouse release 24.6, 2024-07-01 #### Backward Incompatible Change * Enable asynchronous load of databases and tables by default. See the `async_load_databases` in config.xml. While this change is fully compatible, it can introduce a difference in behavior. When `async_load_databases` is false, as in the previous versions, the server will not accept connections until all tables are loaded. When `async_load_databases` is true, as in the new version, the server can accept connections before all the tables are loaded. If a query is made to a table that is not yet loaded, it will wait for the table's loading, which can take considerable time. It can change the behavior of the server if it is part of a large distributed system under a load balancer. In the first case, the load balancer can get a connection refusal and quickly failover to another server. In the second case, the load balancer can connect to a server that is still loading the tables, and the query will have a higher latency. Moreover, if many queries accumulate in the waiting state, it can lead to a "thundering herd" problem when they start processing simultaneously. This can make a difference only for highly loaded distributed backends. You can set the value of `async_load_databases` to false to avoid this problem. [#57695](https://github.com/ClickHouse/ClickHouse/pull/57695) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Setting `replace_long_file_name_to_hash` is enabled by default for `MergeTree` tables. [#64457](https://github.com/ClickHouse/ClickHouse/pull/64457) ([Anton Popov](https://github.com/CurtizJ)). This setting is fully compatible, and no actions needed during upgrade. The new data format is supported from all versions starting from 23.9. After enabling this setting, you can no longer downgrade to a version 23.8 or older. * Some invalid queries will fail earlier during parsing. Note: disabled the support for inline KQL expressions (the experimental Kusto language) when they are put into a `kql` table function without a string literal, e.g. `kql(garbage | trash)` instead of `kql('garbage | trash')` or `kql($$garbage | trash$$)`. This feature was introduced unintentionally and should not exist. [#61500](https://github.com/ClickHouse/ClickHouse/pull/61500) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Rework parallel processing in `Ordered` mode of storage `S3Queue`. This PR is backward incompatible for Ordered mode if you used settings `s3queue_processing_threads_num` or `s3queue_total_shards_num`. Setting `s3queue_total_shards_num` is deleted, previously it was allowed to use only under `s3queue_allow_experimental_sharded_mode`, which is now deprecated. A new setting is added - `s3queue_buckets`. [#64349](https://github.com/ClickHouse/ClickHouse/pull/64349) ([Kseniia Sumarokova](https://github.com/kssenii)). * New functions `snowflakeIDToDateTime`, `snowflakeIDToDateTime64`, `dateTimeToSnowflakeID`, and `dateTime64ToSnowflakeID` were added. Unlike the existing functions `snowflakeToDateTime`, `snowflakeToDateTime64`, `dateTimeToSnowflake`, and `dateTime64ToSnowflake`, the new functions are compatible with function `generateSnowflakeID`, i.e. they accept the snowflake IDs generated by `generateSnowflakeID` and produce snowflake IDs of the same type as `generateSnowflakeID` (i.e. `UInt64`). Furthermore, the new functions default to the UNIX epoch (aka. 1970-01-01), just like `generateSnowflakeID`. If necessary, a different epoch, e.g. Twitter's/X's epoch 2010-11-04 aka. 1288834974657 msec since UNIX epoch, can be passed. The old conversion functions are deprecated and will be removed after a transition period: to use them regardless, enable setting `allow_deprecated_snowflake_conversion_functions`. [#64948](https://github.com/ClickHouse/ClickHouse/pull/64948) ([Robert Schulze](https://github.com/rschu1ze)). #### New Feature -* Introduce statistics of type "number of distinct values". [#59357](https://github.com/ClickHouse/ClickHouse/pull/59357) ([Han Fei](https://github.com/hanfei1991)). +* Allow to store named collections in ClickHouse Keeper. [#64574](https://github.com/ClickHouse/ClickHouse/pull/64574) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Support empty tuples. [#55061](https://github.com/ClickHouse/ClickHouse/pull/55061) ([Amos Bird](https://github.com/amosbird)). * Add Hilbert Curve encode and decode functions. [#60156](https://github.com/ClickHouse/ClickHouse/pull/60156) ([Artem Mustafin](https://github.com/Artemmm91)). -* Added support for reading LINESTRING geometry in WKT format using function `readWKTLineString`. [#62519](https://github.com/ClickHouse/ClickHouse/pull/62519) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). -* Allow to attach parts from a different disk. [#63087](https://github.com/ClickHouse/ClickHouse/pull/63087) ([Unalian](https://github.com/Unalian)). -* Allow proxy to be bypassed for hosts specified in `no_proxy` env variable and ClickHouse proxy configuration. [#63314](https://github.com/ClickHouse/ClickHouse/pull/63314) ([Arthur Passos](https://github.com/arthurpassos)). -* Added a new table function `loop` to support returning query results in an infinite loop. [#63452](https://github.com/ClickHouse/ClickHouse/pull/63452) ([Sariel](https://github.com/sarielwxm)). -* Added new SQL functions `generateSnowflakeID` for generating Twitter-style Snowflake IDs. [#63577](https://github.com/ClickHouse/ClickHouse/pull/63577) ([Danila Puzov](https://github.com/kazalika)). -* Add the ability to reshuffle rows during insert to optimize for size without violating the order set by `PRIMARY KEY`. It's controlled by the setting `optimize_row_order` (off by default). [#63578](https://github.com/ClickHouse/ClickHouse/pull/63578) ([Igor Markelov](https://github.com/ElderlyPassionFruit)). -* Added `merge_workload` and `mutation_workload` settings to regulate how resources are utilized and shared between merges, mutations and other workloads. [#64061](https://github.com/ClickHouse/ClickHouse/pull/64061) ([Sergei Trifonov](https://github.com/serxa)). -* Add support for comparing IPv4 and IPv6 types using the `=` operator. [#64292](https://github.com/ClickHouse/ClickHouse/pull/64292) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)). -* Allow to store named collections in zookeeper. [#64574](https://github.com/ClickHouse/ClickHouse/pull/64574) ([Kseniia Sumarokova](https://github.com/kssenii)). -* Support decimal arguments in binary math functions (pow, atan2, max2, min2, hypot). [#64582](https://github.com/ClickHouse/ClickHouse/pull/64582) ([Mikhail Gorshkov](https://github.com/mgorshkov)). * Add support for index analysis over `hilbertEncode`. [#64662](https://github.com/ClickHouse/ClickHouse/pull/64662) ([Artem Mustafin](https://github.com/Artemmm91)). +* Added support for reading `LINESTRING` geometry in the WKT format using function `readWKTLineString`. [#62519](https://github.com/ClickHouse/ClickHouse/pull/62519) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Allow to attach parts from a different disk. [#63087](https://github.com/ClickHouse/ClickHouse/pull/63087) ([Unalian](https://github.com/Unalian)). +* Added new SQL functions `generateSnowflakeID` for generating Twitter-style Snowflake IDs. [#63577](https://github.com/ClickHouse/ClickHouse/pull/63577) ([Danila Puzov](https://github.com/kazalika)). +* Added `merge_workload` and `mutation_workload` settings to regulate how resources are utilized and shared between merges, mutations and other workloads. [#64061](https://github.com/ClickHouse/ClickHouse/pull/64061) ([Sergei Trifonov](https://github.com/serxa)). +* Add support for comparing `IPv4` and `IPv6` types using the `=` operator. [#64292](https://github.com/ClickHouse/ClickHouse/pull/64292) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)). +* Support decimal arguments in binary math functions (pow, atan2, max2, min2, hypot). [#64582](https://github.com/ClickHouse/ClickHouse/pull/64582) ([Mikhail Gorshkov](https://github.com/mgorshkov)). * Added SQL functions `parseReadableSize` (along with `OrNull` and `OrZero` variants). [#64742](https://github.com/ClickHouse/ClickHouse/pull/64742) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)). * Add server settings `max_table_num_to_throw` and `max_database_num_to_throw` to limit the number of databases or tables on `CREATE` queries. [#64781](https://github.com/ClickHouse/ClickHouse/pull/64781) ([Xu Jia](https://github.com/XuJia0210)). -* Add _time virtual column to file alike storages (s3/file/hdfs/url/azureBlobStorage). [#64947](https://github.com/ClickHouse/ClickHouse/pull/64947) ([Ilya Golshtein](https://github.com/ilejn)). +* Add `_time` virtual column to file alike storages (s3/file/hdfs/url/azureBlobStorage). [#64947](https://github.com/ClickHouse/ClickHouse/pull/64947) ([Ilya Golshtein](https://github.com/ilejn)). * Introduced new functions `base64URLEncode`, `base64URLDecode` and `tryBase64URLDecode`. [#64991](https://github.com/ClickHouse/ClickHouse/pull/64991) ([Mikhail Gorshkov](https://github.com/mgorshkov)). * Add new function `editDistanceUTF8`, which calculates the [edit distance](https://en.wikipedia.org/wiki/Edit_distance) between two UTF8 strings. [#65269](https://github.com/ClickHouse/ClickHouse/pull/65269) ([LiuNeng](https://github.com/liuneng1994)). +* Add `http_response_headers` configuration to support custom response headers in custom HTTP handlers. [#63562](https://github.com/ClickHouse/ClickHouse/pull/63562) ([Grigorii](https://github.com/GSokol)). +* Added a new table function `loop` to support returning query results in an infinite loop. [#63452](https://github.com/ClickHouse/ClickHouse/pull/63452) ([Sariel](https://github.com/sarielwxm)). This is useful for testing. +* Introduced two additional columns in the `system.query_log`: `used_privileges` and `missing_privileges`. `used_privileges` is populated with the privileges that were checked during query execution, and `missing_privileges` contains required privileges that are missing. [#64597](https://github.com/ClickHouse/ClickHouse/pull/64597) ([Alexey Katsman](https://github.com/alexkats)). +* Added a setting `output_format_pretty_display_footer_column_names` which when enabled displays column names at the end of the table for long tables (50 rows by default), with the threshold value for minimum number of rows controlled by `output_format_pretty_display_footer_column_names_min_rows`. [#65144](https://github.com/ClickHouse/ClickHouse/pull/65144) ([Shaun Struwig](https://github.com/Blargian)). + +#### Experimental Feature +* Introduce statistics of type "number of distinct values". [#59357](https://github.com/ClickHouse/ClickHouse/pull/59357) ([Han Fei](https://github.com/hanfei1991)). +* Support statistics with ReplicatedMergeTree. [#64934](https://github.com/ClickHouse/ClickHouse/pull/64934) ([Han Fei](https://github.com/hanfei1991)). +* If "replica group" is configured for a `Replicated` database, automatically create a cluster that includes replicas from all groups. [#64312](https://github.com/ClickHouse/ClickHouse/pull/64312) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Add settings `parallel_replicas_custom_key_range_lower` and `parallel_replicas_custom_key_range_upper` to control how parallel replicas with dynamic shards parallelizes queries when using a range filter. [#64604](https://github.com/ClickHouse/ClickHouse/pull/64604) ([josh-hildred](https://github.com/josh-hildred)). #### Performance Improvement +* Add the ability to reshuffle rows during insert to optimize for size without violating the order set by `PRIMARY KEY`. It's controlled by the setting `optimize_row_order` (off by default). [#63578](https://github.com/ClickHouse/ClickHouse/pull/63578) ([Igor Markelov](https://github.com/ElderlyPassionFruit)). * Add a native parquet reader, which can read parquet binary to ClickHouse Columns directly. It's controlled by the setting `input_format_parquet_use_native_reader` (disabled by default). [#60361](https://github.com/ClickHouse/ClickHouse/pull/60361) ([ZhiHong Zhang](https://github.com/copperybean)). -* Reduce the number of virtual function calls in ColumnNullable::size. [#60556](https://github.com/ClickHouse/ClickHouse/pull/60556) ([HappenLee](https://github.com/HappenLee)). -* Speedup `splitByRegexp` when the regular expression argument is a single-character. [#62696](https://github.com/ClickHouse/ClickHouse/pull/62696) ([Robert Schulze](https://github.com/rschu1ze)). -* Speed up FixedHashTable by keeping track of the min and max keys used. This allows to reduce the number of cells that need to be verified. [#62746](https://github.com/ClickHouse/ClickHouse/pull/62746) ([Jiebin Sun](https://github.com/jiebinn)). -* Optimize the resolution of in(LowCardinality, ConstantSet). [#64060](https://github.com/ClickHouse/ClickHouse/pull/64060) ([Zhiguo Zhou](https://github.com/ZhiguoZh)). -* Use a thread pool to initialize and destroy hash tables inside `ConcurrentHashJoin`. [#64241](https://github.com/ClickHouse/ClickHouse/pull/64241) ([Nikita Taranov](https://github.com/nickitat)). -* Optimized vertical merges in tables with sparse columns. [#64311](https://github.com/ClickHouse/ClickHouse/pull/64311) ([Anton Popov](https://github.com/CurtizJ)). -* Enabled prefetches of data from remote filesystem during vertical merges. It improves latency of vertical merges in tables with data stored on remote filesystem. [#64314](https://github.com/ClickHouse/ClickHouse/pull/64314) ([Anton Popov](https://github.com/CurtizJ)). -* Reduce redundant calls to `isDefault()` of `ColumnSparse::filter` to improve performance. [#64426](https://github.com/ClickHouse/ClickHouse/pull/64426) ([Jiebin Sun](https://github.com/jiebinn)). -* Speedup `find_super_nodes` and `find_big_family` keeper-client commands by making multiple asynchronous getChildren requests. [#64628](https://github.com/ClickHouse/ClickHouse/pull/64628) ([Alexander Gololobov](https://github.com/davenger)). -* Improve function least/greatest for nullable numberic type arguments. [#64668](https://github.com/ClickHouse/ClickHouse/pull/64668) ([KevinyhZou](https://github.com/KevinyhZou)). -* Allow merging two consequent `FilterSteps` of a query plan. This improves filter-push-down optimization if the filter condition can be pushed down from the parent step. [#64760](https://github.com/ClickHouse/ClickHouse/pull/64760) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). -* Remove bad optimization in vertical final implementation and re-enable vertical final algorithm by default. [#64783](https://github.com/ClickHouse/ClickHouse/pull/64783) ([Duc Canh Le](https://github.com/canhld94)). -* Remove ALIAS nodes from the filter expression. This slightly improves performance for queries with `PREWHERE` (with the new analyzer). [#64793](https://github.com/ClickHouse/ClickHouse/pull/64793) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). -* Fix performance regression in cross join introduced in [#60459](https://github.com/ClickHouse/ClickHouse/issues/60459) (24.5). [#65243](https://github.com/ClickHouse/ClickHouse/pull/65243) ([Nikita Taranov](https://github.com/nickitat)). - -#### Improvement -* Support empty tuples. [#55061](https://github.com/ClickHouse/ClickHouse/pull/55061) ([Amos Bird](https://github.com/amosbird)). -* Hot reload storage policy for distributed tables when adding a new disk. [#58285](https://github.com/ClickHouse/ClickHouse/pull/58285) ([Duc Canh Le](https://github.com/canhld94)). -* Avoid possible deadlock during MergeTree index analysis when scheduling threads in a saturated service. [#59427](https://github.com/ClickHouse/ClickHouse/pull/59427) ([Sean Haynes](https://github.com/seandhaynes)). * Support partial trivial count optimization when the query filter is able to select exact ranges from merge tree tables. [#60463](https://github.com/ClickHouse/ClickHouse/pull/60463) ([Amos Bird](https://github.com/amosbird)). * Reduce max memory usage of multithreaded `INSERT`s by collecting chunks of multiple threads in a single transform. [#61047](https://github.com/ClickHouse/ClickHouse/pull/61047) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). * Reduce the memory usage when using Azure object storage by using fixed memory allocation, avoiding the allocation of an extra buffer. [#63160](https://github.com/ClickHouse/ClickHouse/pull/63160) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). -* Several minor corner case fixes to proxy support & tunneling. [#63427](https://github.com/ClickHouse/ClickHouse/pull/63427) ([Arthur Passos](https://github.com/arthurpassos)). -* Add `http_response_headers` setting to support custom response headers in custom HTTP handlers. [#63562](https://github.com/ClickHouse/ClickHouse/pull/63562) ([Grigorii](https://github.com/GSokol)). -* Improve io_uring resubmit visibility. Rename profile event `IOUringSQEsResubmits` -> `IOUringSQEsResubmitsAsync` and add a new one `IOUringSQEsResubmitsSync`. [#63699](https://github.com/ClickHouse/ClickHouse/pull/63699) ([Tomer Shafir](https://github.com/tomershafir)). -* Introduce assertions to verify all functions are called with columns of the right size. [#63723](https://github.com/ClickHouse/ClickHouse/pull/63723) ([Raúl Marín](https://github.com/Algunenano)). +* Reduce the number of virtual function calls in `ColumnNullable::size`. [#60556](https://github.com/ClickHouse/ClickHouse/pull/60556) ([HappenLee](https://github.com/HappenLee)). +* Speedup `splitByRegexp` when the regular expression argument is a single-character. [#62696](https://github.com/ClickHouse/ClickHouse/pull/62696) ([Robert Schulze](https://github.com/rschu1ze)). +* Speed up aggregation by 8-bit and 16-bit keys by keeping track of the min and max keys used. This allows to reduce the number of cells that need to be verified. [#62746](https://github.com/ClickHouse/ClickHouse/pull/62746) ([Jiebin Sun](https://github.com/jiebinn)). +* Optimize operator IN when the left hand side is `LowCardinality` and the right is a set of constants. [#64060](https://github.com/ClickHouse/ClickHouse/pull/64060) ([Zhiguo Zhou](https://github.com/ZhiguoZh)). +* Use a thread pool to initialize and destroy hash tables inside `ConcurrentHashJoin`. [#64241](https://github.com/ClickHouse/ClickHouse/pull/64241) ([Nikita Taranov](https://github.com/nickitat)). +* Optimized vertical merges in tables with sparse columns. [#64311](https://github.com/ClickHouse/ClickHouse/pull/64311) ([Anton Popov](https://github.com/CurtizJ)). +* Enabled prefetches of data from remote filesystem during vertical merges. It improves latency of vertical merges in tables with data stored on remote filesystem. [#64314](https://github.com/ClickHouse/ClickHouse/pull/64314) ([Anton Popov](https://github.com/CurtizJ)). +* Reduce redundant calls to `isDefault` of `ColumnSparse::filter` to improve performance. [#64426](https://github.com/ClickHouse/ClickHouse/pull/64426) ([Jiebin Sun](https://github.com/jiebinn)). +* Speedup `find_super_nodes` and `find_big_family` keeper-client commands by making multiple asynchronous getChildren requests. [#64628](https://github.com/ClickHouse/ClickHouse/pull/64628) ([Alexander Gololobov](https://github.com/davenger)). +* Improve function `least`/`greatest` for nullable numberic type arguments. [#64668](https://github.com/ClickHouse/ClickHouse/pull/64668) ([KevinyhZou](https://github.com/KevinyhZou)). +* Allow merging two consequent filtering steps of a query plan. This improves filter-push-down optimization if the filter condition can be pushed down from the parent step. [#64760](https://github.com/ClickHouse/ClickHouse/pull/64760) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Remove bad optimization in the vertical final implementation and re-enable vertical final algorithm by default. [#64783](https://github.com/ClickHouse/ClickHouse/pull/64783) ([Duc Canh Le](https://github.com/canhld94)). +* Remove ALIAS nodes from the filter expression. This slightly improves performance for queries with `PREWHERE` (with the new analyzer). [#64793](https://github.com/ClickHouse/ClickHouse/pull/64793) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Re-enable OpenSSL session caching. [#65111](https://github.com/ClickHouse/ClickHouse/pull/65111) ([Robert Schulze](https://github.com/rschu1ze)). +* Added settings to disable materialization of skip indexes and statistics on inserts (`materialize_skip_indexes_on_insert` and `materialize_statistics_on_insert`). [#64391](https://github.com/ClickHouse/ClickHouse/pull/64391) ([Anton Popov](https://github.com/CurtizJ)). +* Use the allocated memory size to calculate the row group size and reduce the peak memory of the parquet writer in the single-threaded mode. [#64424](https://github.com/ClickHouse/ClickHouse/pull/64424) ([LiuNeng](https://github.com/liuneng1994)). +* Improve the iterator of sparse column to reduce call of `size`. [#64497](https://github.com/ClickHouse/ClickHouse/pull/64497) ([Jiebin Sun](https://github.com/jiebinn)). +* Update condition to use server-side copy for backups to Azure blob storage. [#64518](https://github.com/ClickHouse/ClickHouse/pull/64518) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Optimized memory usage of vertical merges for tables with high number of skip indexes. [#64580](https://github.com/ClickHouse/ClickHouse/pull/64580) ([Anton Popov](https://github.com/CurtizJ)). + +#### Improvement * `SHOW CREATE TABLE` executed on top of system tables will now show the super handy comment unique for each table which will explain why this table is needed. [#63788](https://github.com/ClickHouse/ClickHouse/pull/63788) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). -* Added setting `metadata_storage_type` to keep free space on metadata storage disk. [#64128](https://github.com/ClickHouse/ClickHouse/pull/64128) ([MikhailBurdukov](https://github.com/MikhailBurdukov)). -* Add metrics to track the number of directories created and removed by the plain_rewritable metadata storage, and the number of entries in the local-to-remote in-memory map. [#64175](https://github.com/ClickHouse/ClickHouse/pull/64175) ([Julia Kartseva](https://github.com/jkartseva)). +* The second argument (scale) of functions `round()`, `roundBankers()`, `floor()`, `ceil()` and `trunc()` can now be non-const. [#64798](https://github.com/ClickHouse/ClickHouse/pull/64798) ([Mikhail Gorshkov](https://github.com/mgorshkov)). +* Hot reload storage policy for `Distributed` tables when adding a new disk. [#58285](https://github.com/ClickHouse/ClickHouse/pull/58285) ([Duc Canh Le](https://github.com/canhld94)). +* Avoid possible deadlock during MergeTree index analysis when scheduling threads in a saturated service. [#59427](https://github.com/ClickHouse/ClickHouse/pull/59427) ([Sean Haynes](https://github.com/seandhaynes)). +* Several minor corner case fixes to S3 proxy support & tunneling. [#63427](https://github.com/ClickHouse/ClickHouse/pull/63427) ([Arthur Passos](https://github.com/arthurpassos)). +* Improve io_uring resubmit visibility. Rename profile event `IOUringSQEsResubmits` -> `IOUringSQEsResubmitsAsync` and add a new one `IOUringSQEsResubmitsSync`. [#63699](https://github.com/ClickHouse/ClickHouse/pull/63699) ([Tomer Shafir](https://github.com/tomershafir)). +* Added a new setting, `metadata_keep_free_space_bytes` to keep free space on the metadata storage disk. [#64128](https://github.com/ClickHouse/ClickHouse/pull/64128) ([MikhailBurdukov](https://github.com/MikhailBurdukov)). +* Add metrics to track the number of directories created and removed by the `plain_rewritable` metadata storage, and the number of entries in the local-to-remote in-memory map. [#64175](https://github.com/ClickHouse/ClickHouse/pull/64175) ([Julia Kartseva](https://github.com/jkartseva)). * The query cache now considers identical queries with different settings as different. This increases robustness in cases where different settings (e.g. `limit` or `additional_table_filters`) would affect the query result. [#64205](https://github.com/ClickHouse/ClickHouse/pull/64205) ([Robert Schulze](https://github.com/rschu1ze)). -* Better Exception Message in Delete Table with Projection, users can understand the error and the steps should be taken. [#64212](https://github.com/ClickHouse/ClickHouse/pull/64212) ([jsc0218](https://github.com/jsc0218)). * Support the non standard error code `QpsLimitExceeded` in object storage as a retryable error. [#64225](https://github.com/ClickHouse/ClickHouse/pull/64225) ([Sema Checherinda](https://github.com/CheSema)). * Forbid converting a MergeTree table to replicated if the zookeeper path for this table already exists. [#64244](https://github.com/ClickHouse/ClickHouse/pull/64244) ([Kirill](https://github.com/kirillgarbar)). -* If "replica group" is configured for a `Replicated` database, automatically create a cluster that includes replicas from all groups. [#64312](https://github.com/ClickHouse/ClickHouse/pull/64312) ([Alexander Tokmakov](https://github.com/tavplubix)). -* Added settings to disable materialization of skip indexes and statistics on inserts (`materialize_skip_indexes_on_insert` and `materialize_statistics_on_insert`). [#64391](https://github.com/ClickHouse/ClickHouse/pull/64391) ([Anton Popov](https://github.com/CurtizJ)). -* Use the allocated memory size to calculate the row group size and reduce the peak memory of the parquet writer in single-threaded mode. [#64424](https://github.com/ClickHouse/ClickHouse/pull/64424) ([LiuNeng](https://github.com/liuneng1994)). -* Added new configuration input_format_parquet_prefer_block_bytes to control the average output block bytes, and modified the default value of input_format_parquet_max_block_size to 65409. [#64427](https://github.com/ClickHouse/ClickHouse/pull/64427) ([LiuNeng](https://github.com/liuneng1994)). +* Added a new setting `input_format_parquet_prefer_block_bytes` to control the average output block bytes, and modified the default value of `input_format_parquet_max_block_size` to 65409. [#64427](https://github.com/ClickHouse/ClickHouse/pull/64427) ([LiuNeng](https://github.com/liuneng1994)). +* Allow proxy to be bypassed for hosts specified in `no_proxy` env variable and ClickHouse proxy configuration. [#63314](https://github.com/ClickHouse/ClickHouse/pull/63314) ([Arthur Passos](https://github.com/arthurpassos)). * Always start Keeper with sufficient amount of threads in global thread pool. [#64444](https://github.com/ClickHouse/ClickHouse/pull/64444) ([Duc Canh Le](https://github.com/canhld94)). -* Settings from user config doesn't affect merges and mutations for MergeTree on top of object storage. [#64456](https://github.com/ClickHouse/ClickHouse/pull/64456) ([alesapin](https://github.com/alesapin)). -* Setting `replace_long_file_name_to_hash` is enabled by default for `MergeTree` tables. [#64457](https://github.com/ClickHouse/ClickHouse/pull/64457) ([Anton Popov](https://github.com/CurtizJ)). -* Improve the iterator of sparse column to reduce call of size(). [#64497](https://github.com/ClickHouse/ClickHouse/pull/64497) ([Jiebin Sun](https://github.com/jiebinn)). -* Update condition to use copy for azure blob storage. [#64518](https://github.com/ClickHouse/ClickHouse/pull/64518) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Settings from the user's config don't affect merges and mutations for `MergeTree` on top of object storage. [#64456](https://github.com/ClickHouse/ClickHouse/pull/64456) ([alesapin](https://github.com/alesapin)). * Support the non standard error code `TotalQpsLimitExceeded` in object storage as a retryable error. [#64520](https://github.com/ClickHouse/ClickHouse/pull/64520) ([Sema Checherinda](https://github.com/CheSema)). -* Optimized memory usage of vertical merges for tables with high number of skip indexes. [#64580](https://github.com/ClickHouse/ClickHouse/pull/64580) ([Anton Popov](https://github.com/CurtizJ)). -* Introduced two additional columns in the `system.query_log`: `used_privileges` and `missing_privileges`. `used_privileges` is populated with the privileges that were checked during query execution, and `missing_privileges` contains required privileges that are missing. [#64597](https://github.com/ClickHouse/ClickHouse/pull/64597) ([Alexey Katsman](https://github.com/alexkats)). -* Add settings `parallel_replicas_custom_key_range_lower` and `parallel_replicas_custom_key_range_upper` to control how parallel replicas with dynamic shards parallelizes queries when using a range filter. [#64604](https://github.com/ClickHouse/ClickHouse/pull/64604) ([josh-hildred](https://github.com/josh-hildred)). * Updated Advanced Dashboard for both open-source and ClickHouse Cloud versions to include a chart for 'Maximum concurrent network connections'. [#64610](https://github.com/ClickHouse/ClickHouse/pull/64610) ([Thom O'Connor](https://github.com/thomoco)). -* The second argument (scale) of functions `round()`, `roundBankers()`, `floor()`, `ceil()` and `trunc()` can now be non-const. [#64798](https://github.com/ClickHouse/ClickHouse/pull/64798) ([Mikhail Gorshkov](https://github.com/mgorshkov)). -* Improve progress report on zeros_mt and generateRandom. [#64804](https://github.com/ClickHouse/ClickHouse/pull/64804) ([Raúl Marín](https://github.com/Algunenano)). -* Add an asynchronous metric jemalloc.profile.active to show whether sampling is currently active. This is an activation mechanism in addition to prof.active; both must be active for the calling thread to sample. [#64842](https://github.com/ClickHouse/ClickHouse/pull/64842) ([Unalian](https://github.com/Unalian)). -* Support statistics with ReplicatedMergeTree. [#64934](https://github.com/ClickHouse/ClickHouse/pull/64934) ([Han Fei](https://github.com/hanfei1991)). +* Improve progress report on `zeros_mt` and `generateRandom`. [#64804](https://github.com/ClickHouse/ClickHouse/pull/64804) ([Raúl Marín](https://github.com/Algunenano)). +* Add an asynchronous metric `jemalloc.profile.active` to show whether sampling is currently active. This is an activation mechanism in addition to prof.active; both must be active for the calling thread to sample. [#64842](https://github.com/ClickHouse/ClickHouse/pull/64842) ([Unalian](https://github.com/Unalian)). * Remove mark of `allow_experimental_join_condition` as important. This mark may have prevented distributed queries in a mixed versions cluster from being executed successfully. [#65008](https://github.com/ClickHouse/ClickHouse/pull/65008) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). * Added server Asynchronous metrics `DiskGetObjectThrottler*` and `DiskGetObjectThrottler*` reflecting request per second rate limit defined with `s3_max_get_rps` and `s3_max_put_rps` disk settings and currently available number of requests that could be sent without hitting throttling limit on the disk. Metrics are defined for every disk that has a configured limit. [#65050](https://github.com/ClickHouse/ClickHouse/pull/65050) ([Sergei Trifonov](https://github.com/serxa)). -* Added a setting `output_format_pretty_display_footer_column_names` which when enabled displays column names at the end of the table for long tables (50 rows by default), with the threshold value for minimum number of rows controlled by `output_format_pretty_display_footer_column_names_min_rows`. [#65144](https://github.com/ClickHouse/ClickHouse/pull/65144) ([Shaun Struwig](https://github.com/Blargian)). -* Returned back the behaviour of how ClickHouse works and interprets Tuples in CSV format. This change effectively reverts https://github.com/ClickHouse/ClickHouse/pull/60994 and makes it available only under a few settings: `output_format_csv_serialize_tuple_into_separate_columns`, `input_format_csv_deserialize_separate_columns_into_tuple` and `input_format_csv_try_infer_strings_from_quoted_tuples`. [#65170](https://github.com/ClickHouse/ClickHouse/pull/65170) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). -* Initialize global trace collector for Poco::ThreadPool (needed for keeper, etc). [#65239](https://github.com/ClickHouse/ClickHouse/pull/65239) ([Kseniia Sumarokova](https://github.com/kssenii)). -* Add validation when creating a user with bcrypt_hash. [#65242](https://github.com/ClickHouse/ClickHouse/pull/65242) ([Raúl Marín](https://github.com/Algunenano)). -* Unite s3/hdfs/azure storage implementations into a single class working with IObjectStorage. Same for *Cluster, data lakes and Queue storages. [#59767](https://github.com/ClickHouse/ClickHouse/pull/59767) ([Kseniia Sumarokova](https://github.com/kssenii)). -* Refactor data part writer to remove dependencies on MergeTreeData and DataPart. [#63620](https://github.com/ClickHouse/ClickHouse/pull/63620) ([Alexander Gololobov](https://github.com/davenger)). -* Add profile events for number of rows read during/after prewhere. [#64198](https://github.com/ClickHouse/ClickHouse/pull/64198) ([Nikita Taranov](https://github.com/nickitat)). -* Print query in explain plan with parallel replicas. [#64298](https://github.com/ClickHouse/ClickHouse/pull/64298) ([vdimir](https://github.com/vdimir)). +* Initialize global trace collector for `Poco::ThreadPool` (needed for Keeper, etc). [#65239](https://github.com/ClickHouse/ClickHouse/pull/65239) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Add a validation when creating a user with `bcrypt_hash`. [#65242](https://github.com/ClickHouse/ClickHouse/pull/65242) ([Raúl Marín](https://github.com/Algunenano)). +* Add profile events for number of rows read during/after `PREWHERE`. [#64198](https://github.com/ClickHouse/ClickHouse/pull/64198) ([Nikita Taranov](https://github.com/nickitat)). +* Print query in `EXPLAIN PLAN` with parallel replicas. [#64298](https://github.com/ClickHouse/ClickHouse/pull/64298) ([vdimir](https://github.com/vdimir)). * Rename `allow_deprecated_functions` to `allow_deprecated_error_prone_window_functions`. [#64358](https://github.com/ClickHouse/ClickHouse/pull/64358) ([Raúl Marín](https://github.com/Algunenano)). -* Respect `max_read_buffer_size` setting for file descriptors as well in file() table function. [#64532](https://github.com/ClickHouse/ClickHouse/pull/64532) ([Azat Khuzhin](https://github.com/azat)). +* Respect `max_read_buffer_size` setting for file descriptors as well in the `file` table function. [#64532](https://github.com/ClickHouse/ClickHouse/pull/64532) ([Azat Khuzhin](https://github.com/azat)). * Disable transactions for unsupported storages even for materialized views. [#64918](https://github.com/ClickHouse/ClickHouse/pull/64918) ([alesapin](https://github.com/alesapin)). -* Refactor `KeyCondition` and key analysis to improve PartitionPruner and trivial count optimization. This is separated from [#60463](https://github.com/ClickHouse/ClickHouse/issues/60463) . [#61459](https://github.com/ClickHouse/ClickHouse/pull/61459) ([Amos Bird](https://github.com/amosbird)). +* Forbid `QUALIFY` clause in the old analyzer. The old analyzer ignored `QUALIFY`, so it could lead to unexpected data removal in mutations. [#65356](https://github.com/ClickHouse/ClickHouse/pull/65356) ([Dmitry Novik](https://github.com/novikd)). #### Bug Fix (user-visible misbehavior in an official stable release) +* A bug in Apache ORC library was fixed: Fixed ORC statistics calculation, when writing, for unsigned types on all platforms and Int8 on ARM. [#64563](https://github.com/ClickHouse/ClickHouse/pull/64563) ([Michael Kolupaev](https://github.com/al13n321)). +* Returned back the behaviour of how ClickHouse works and interprets Tuples in CSV format. This change effectively reverts https://github.com/ClickHouse/ClickHouse/pull/60994 and makes it available only under a few settings: `output_format_csv_serialize_tuple_into_separate_columns`, `input_format_csv_deserialize_separate_columns_into_tuple` and `input_format_csv_try_infer_strings_from_quoted_tuples`. [#65170](https://github.com/ClickHouse/ClickHouse/pull/65170) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). * Fix a permission error where a user in a specific situation can escalate their privileges on the default database without necessary grants. [#64769](https://github.com/ClickHouse/ClickHouse/pull/64769) ([pufit](https://github.com/pufit)). * Fix crash with UniqInjectiveFunctionsEliminationPass and uniqCombined. [#65188](https://github.com/ClickHouse/ClickHouse/pull/65188) ([Raúl Marín](https://github.com/Algunenano)). * Fix a bug in ClickHouse Keeper that causes digest mismatch during closing session. [#65198](https://github.com/ClickHouse/ClickHouse/pull/65198) ([Aleksei Filatov](https://github.com/aalexfvk)). -* Forbid `QUALIFY` clause in the old analyzer. The old analyzer ignored `QUALIFY`, so it could lead to unexpected data removal in mutations. [#65356](https://github.com/ClickHouse/ClickHouse/pull/65356) ([Dmitry Novik](https://github.com/novikd)). * Use correct memory alignment for Distinct combinator. Previously, crash could happen because of invalid memory allocation when the combinator was used. [#65379](https://github.com/ClickHouse/ClickHouse/pull/65379) ([Antonio Andelic](https://github.com/antonio2368)). * Fix crash with `DISTINCT` and window functions. [#64767](https://github.com/ClickHouse/ClickHouse/pull/64767) ([Igor Nikonov](https://github.com/devcrafter)). * Fixed 'set' skip index not working with IN and indexHint(). [#62083](https://github.com/ClickHouse/ClickHouse/pull/62083) ([Michael Kolupaev](https://github.com/al13n321)). @@ -132,7 +303,6 @@ * Fixed `optimize_read_in_order` behaviour for ORDER BY ... NULLS FIRST / LAST on tables with nullable keys. [#64483](https://github.com/ClickHouse/ClickHouse/pull/64483) ([Eduard Karacharov](https://github.com/korowa)). * Fix the `Expression nodes list expected 1 projection names` and `Unknown expression or identifier` errors for queries with aliases to `GLOBAL IN.`. [#64517](https://github.com/ClickHouse/ClickHouse/pull/64517) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). * Fix an error `Cannot find column` in distributed queries with constant CTE in the `GROUP BY` key. [#64519](https://github.com/ClickHouse/ClickHouse/pull/64519) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). -* Fixed ORC statistics calculation, when writing, for unsigned types on all platforms and Int8 on ARM. [#64563](https://github.com/ClickHouse/ClickHouse/pull/64563) ([Michael Kolupaev](https://github.com/al13n321)). * Fix the crash loop when restoring from backup is blocked by creating an MV with a definer that hasn't been restored yet. [#64595](https://github.com/ClickHouse/ClickHouse/pull/64595) ([pufit](https://github.com/pufit)). * Fix the output of function `formatDateTimeInJodaSyntax` when a formatter generates an uneven number of characters and the last character is `0`. For example, `SELECT formatDateTimeInJodaSyntax(toDate('2012-05-29'), 'D')` now correctly returns `150` instead of previously `15`. [#64614](https://github.com/ClickHouse/ClickHouse/pull/64614) ([LiuNeng](https://github.com/liuneng1994)). * Do not rewrite aggregation if `-If` combinator is already used. [#64638](https://github.com/ClickHouse/ClickHouse/pull/64638) ([Dmitry Novik](https://github.com/novikd)). @@ -166,21 +336,14 @@ * This PR ensures that the type of the constant(IN operator's second parameter) is always visible during the IN operator's type conversion process. Otherwise, losing type information may cause some conversions to fail, such as the conversion from DateTime to Date. This fixes ([#64487](https://github.com/ClickHouse/ClickHouse/issues/64487)). [#65315](https://github.com/ClickHouse/ClickHouse/pull/65315) ([pn](https://github.com/chloro-pn)). #### Build/Testing/Packaging Improvement -* Make `network` service be required when using the rc init script to start the ClickHouse server daemon. [#60650](https://github.com/ClickHouse/ClickHouse/pull/60650) ([Chun-Sheng, Li](https://github.com/peter279k)). -* Fix typo in test_hdfsCluster_unset_skip_unavailable_shards. The test writes data to unskip_unavailable_shards, but uses skip_unavailable_shards from the previous test. [#64243](https://github.com/ClickHouse/ClickHouse/pull/64243) ([Mikhail Artemenko](https://github.com/Michicosun)). -* Reduce the size of some slow tests. [#64387](https://github.com/ClickHouse/ClickHouse/pull/64387) ([Raúl Marín](https://github.com/Algunenano)). -* Reduce the size of some slow tests. [#64452](https://github.com/ClickHouse/ClickHouse/pull/64452) ([Raúl Marín](https://github.com/Algunenano)). -* Fix test_lost_part_other_replica. [#64512](https://github.com/ClickHouse/ClickHouse/pull/64512) ([Raúl Marín](https://github.com/Algunenano)). -* Add tests for experimental unequal joins and randomize new settings in clickhouse-test. [#64535](https://github.com/ClickHouse/ClickHouse/pull/64535) ([Nikita Fomichev](https://github.com/fm4v)). -* Upgrade tests: Update config and work with release candidates. [#64542](https://github.com/ClickHouse/ClickHouse/pull/64542) ([Raúl Marín](https://github.com/Algunenano)). -* Add support for LLVM XRay. [#64592](https://github.com/ClickHouse/ClickHouse/pull/64592) ([Tomer Shafir](https://github.com/tomershafir)). -* Speed up 02995_forget_partition. [#64761](https://github.com/ClickHouse/ClickHouse/pull/64761) ([Raúl Marín](https://github.com/Algunenano)). -* Fix 02790_async_queries_in_query_log. [#64764](https://github.com/ClickHouse/ClickHouse/pull/64764) ([Raúl Marín](https://github.com/Algunenano)). -* Support LLVM XRay on Linux amd64 only. [#64837](https://github.com/ClickHouse/ClickHouse/pull/64837) ([Tomer Shafir](https://github.com/tomershafir)). -* Get rid of custom code in `tests/ci/download_release_packages.py` and `tests/ci/get_previous_release_tag.py` to avoid issues after the https://github.com/ClickHouse/ClickHouse/pull/64759 is merged. [#64848](https://github.com/ClickHouse/ClickHouse/pull/64848) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). -* Decrease the `unit-test` image a few times. [#65102](https://github.com/ClickHouse/ClickHouse/pull/65102) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Add support for LLVM XRay. [#64592](https://github.com/ClickHouse/ClickHouse/pull/64592) [#64837](https://github.com/ClickHouse/ClickHouse/pull/64837) ([Tomer Shafir](https://github.com/tomershafir)). +* Unite s3/hdfs/azure storage implementations into a single class working with IObjectStorage. Same for *Cluster, data lakes and Queue storages. [#59767](https://github.com/ClickHouse/ClickHouse/pull/59767) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Refactor data part writer to remove dependencies on MergeTreeData and DataPart. [#63620](https://github.com/ClickHouse/ClickHouse/pull/63620) ([Alexander Gololobov](https://github.com/davenger)). +* Refactor `KeyCondition` and key analysis to improve PartitionPruner and trivial count optimization. This is separated from [#60463](https://github.com/ClickHouse/ClickHouse/issues/60463) . [#61459](https://github.com/ClickHouse/ClickHouse/pull/61459) ([Amos Bird](https://github.com/amosbird)). +* Introduce assertions to verify all functions are called with columns of the right size. [#63723](https://github.com/ClickHouse/ClickHouse/pull/63723) ([Raúl Marín](https://github.com/Algunenano)). +* Make `network` service be required when using the `rc` init script to start the ClickHouse server daemon. [#60650](https://github.com/ClickHouse/ClickHouse/pull/60650) ([Chun-Sheng, Li](https://github.com/peter279k)). +* Reduce the size of some slow tests. [#64387](https://github.com/ClickHouse/ClickHouse/pull/64387) [#64452](https://github.com/ClickHouse/ClickHouse/pull/64452) ([Raúl Marín](https://github.com/Algunenano)). * Replay ZooKeeper logs using keeper-bench. [#62481](https://github.com/ClickHouse/ClickHouse/pull/62481) ([Antonio Andelic](https://github.com/antonio2368)). -* Re-enable OpenSSL session caching. [#65111](https://github.com/ClickHouse/ClickHouse/pull/65111) ([Robert Schulze](https://github.com/rschu1ze)). ### ClickHouse release 24.5, 2024-05-30 diff --git a/CMakeLists.txt b/CMakeLists.txt index c4f093b1c99..f796e6c4616 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -319,7 +319,6 @@ endif() # Disable floating-point expression contraction in order to get consistent floating point calculation results across platforms set (COMPILER_FLAGS "${COMPILER_FLAGS} -ffp-contract=off") -# Our built-in unwinder only supports DWARF version up to 4. set (DEBUG_INFO_FLAGS "-g") # Disable omit frame pointer compiler optimization using -fno-omit-frame-pointer @@ -333,15 +332,15 @@ endif() set (CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${COMPILER_FLAGS} ${CMAKE_CXX_FLAGS_ADD}") set (CMAKE_CXX_FLAGS_RELWITHDEBINFO "${CMAKE_CXX_FLAGS_RELWITHDEBINFO} -O3 ${DEBUG_INFO_FLAGS} ${CMAKE_CXX_FLAGS_ADD}") -set (CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -O0 ${DEBUG_INFO_FLAGS} ${CMAKE_CXX_FLAGS_ADD}") +set (CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -Og ${DEBUG_INFO_FLAGS} ${CMAKE_CXX_FLAGS_ADD}") set (CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${COMPILER_FLAGS} ${CMAKE_C_FLAGS_ADD}") set (CMAKE_C_FLAGS_RELWITHDEBINFO "${CMAKE_C_FLAGS_RELWITHDEBINFO} -O3 ${DEBUG_INFO_FLAGS} ${CMAKE_C_FLAGS_ADD}") -set (CMAKE_C_FLAGS_DEBUG "${CMAKE_C_FLAGS_DEBUG} -O0 ${DEBUG_INFO_FLAGS} ${CMAKE_C_FLAGS_ADD}") +set (CMAKE_C_FLAGS_DEBUG "${CMAKE_C_FLAGS_DEBUG} -Og ${DEBUG_INFO_FLAGS} ${CMAKE_C_FLAGS_ADD}") set (CMAKE_ASM_FLAGS "${CMAKE_ASM_FLAGS} ${COMPILER_FLAGS} ${CMAKE_ASM_FLAGS_ADD}") set (CMAKE_ASM_FLAGS_RELWITHDEBINFO "${CMAKE_ASM_FLAGS_RELWITHDEBINFO} -O3 ${DEBUG_INFO_FLAGS} ${CMAKE_ASM_FLAGS_ADD}") -set (CMAKE_ASM_FLAGS_DEBUG "${CMAKE_ASM_FLAGS_DEBUG} -O0 ${DEBUG_INFO_FLAGS} ${CMAKE_ASM_FLAGS_ADD}") +set (CMAKE_ASM_FLAGS_DEBUG "${CMAKE_ASM_FLAGS_DEBUG} -Og ${DEBUG_INFO_FLAGS} ${CMAKE_ASM_FLAGS_ADD}") if (OS_DARWIN) set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -stdlib=libc++") diff --git a/README.md b/README.md index dc253d4db2d..98f9108f14c 100644 --- a/README.md +++ b/README.md @@ -34,14 +34,12 @@ curl https://clickhouse.com/ | sh Every month we get together with the community (users, contributors, customers, those interested in learning more about ClickHouse) to discuss what is coming in the latest release. If you are interested in sharing what you've built on ClickHouse, let us know. -* [v24.6 Community Call](https://clickhouse.com/company/events/v24-6-community-release-call) - Jul 2 +* [v24.7 Community Call](https://clickhouse.com/company/events/v24-7-community-release-call) - Jul 30 ## Upcoming Events Keep an eye out for upcoming meetups and events around the world. Somewhere else you want us to be? Please feel free to reach out to tyler `` clickhouse `` com. You can also peruse [ClickHouse Events](https://clickhouse.com/company/news-events) for a list of all upcoming trainings, meetups, speaking engagements, etc. -* [AWS Summit in DC](https://clickhouse.com/company/events/2024-06-aws-summit-dc) - Jun 26 -* [ClickHouse Meetup in Amsterdam](https://www.meetup.com/clickhouse-netherlands-user-group/events/300781068/) - Jun 27 * [ClickHouse Meetup in Paris](https://www.meetup.com/clickhouse-france-user-group/events/300783448/) - Jul 9 * [ClickHouse Cloud - Live Update Call](https://clickhouse.com/company/events/202407-cloud-update-live) - Jul 9 * [ClickHouse Meetup @ Ramp - New York City](https://www.meetup.com/clickhouse-new-york-user-group/events/300595845/) - Jul 9 diff --git a/SECURITY.md b/SECURITY.md index 8635951dc0e..53328b6e16b 100644 --- a/SECURITY.md +++ b/SECURITY.md @@ -14,6 +14,7 @@ The following versions of ClickHouse server are currently supported with securit | Version | Supported | |:-|:-| +| 24.6 | ✔️ | | 24.5 | ✔️ | | 24.4 | ✔️ | | 24.3 | ✔️ | diff --git a/base/base/cgroupsv2.cpp b/base/base/cgroupsv2.cpp index f20b9daf22e..87f62bf377d 100644 --- a/base/base/cgroupsv2.cpp +++ b/base/base/cgroupsv2.cpp @@ -3,8 +3,9 @@ #include #include -#include +#include +namespace fs = std::filesystem; bool cgroupsV2Enabled() { @@ -13,11 +14,11 @@ bool cgroupsV2Enabled() { /// This file exists iff the host has cgroups v2 enabled. auto controllers_file = default_cgroups_mount / "cgroup.controllers"; - if (!std::filesystem::exists(controllers_file)) + if (!fs::exists(controllers_file)) return false; return true; } - catch (const std::filesystem::filesystem_error &) /// all "underlying OS API errors", typically: permission denied + catch (const fs::filesystem_error &) /// all "underlying OS API errors", typically: permission denied { return false; /// not logging the exception as most callers fall back to cgroups v1 } @@ -33,8 +34,9 @@ bool cgroupsV2MemoryControllerEnabled() /// According to https://docs.kernel.org/admin-guide/cgroup-v2.html, file "cgroup.controllers" defines which controllers are available /// for the current + child cgroups. The set of available controllers can be restricted from level to level using file /// "cgroups.subtree_control". It is therefore sufficient to check the bottom-most nested "cgroup.controllers" file. - std::string cgroup = cgroupV2OfProcess(); - auto cgroup_dir = cgroup.empty() ? default_cgroups_mount : (default_cgroups_mount / cgroup); + fs::path cgroup_dir = cgroupV2PathOfProcess(); + if (cgroup_dir.empty()) + return false; std::ifstream controllers_file(cgroup_dir / "cgroup.controllers"); if (!controllers_file.is_open()) return false; @@ -46,7 +48,7 @@ bool cgroupsV2MemoryControllerEnabled() #endif } -std::string cgroupV2OfProcess() +fs::path cgroupV2PathOfProcess() { #if defined(OS_LINUX) chassert(cgroupsV2Enabled()); @@ -54,17 +56,18 @@ std::string cgroupV2OfProcess() /// A simpler way to get the membership is: std::ifstream cgroup_name_file("/proc/self/cgroup"); if (!cgroup_name_file.is_open()) - return ""; + return {}; /// With cgroups v2, there will be a *single* line with prefix "0::/" /// (see https://docs.kernel.org/admin-guide/cgroup-v2.html) std::string cgroup; std::getline(cgroup_name_file, cgroup); static const std::string v2_prefix = "0::/"; if (!cgroup.starts_with(v2_prefix)) - return ""; + return {}; cgroup = cgroup.substr(v2_prefix.length()); - return cgroup; + /// Note: The 'root' cgroup can have an empty cgroup name, this is valid + return default_cgroups_mount / cgroup; #else - return ""; + return {}; #endif } diff --git a/base/base/cgroupsv2.h b/base/base/cgroupsv2.h index 70219d87cd1..cfb916ff358 100644 --- a/base/base/cgroupsv2.h +++ b/base/base/cgroupsv2.h @@ -1,7 +1,6 @@ #pragma once #include -#include #if defined(OS_LINUX) /// I think it is possible to mount the cgroups hierarchy somewhere else (e.g. when in containers). @@ -16,7 +15,7 @@ bool cgroupsV2Enabled(); /// Assumes that cgroupsV2Enabled() is enabled. bool cgroupsV2MemoryControllerEnabled(); -/// Which cgroup does the process belong to? -/// Returns an empty string if the cgroup cannot be determined. +/// Detects which cgroup v2 the process belongs to and returns the filesystem path to the cgroup. +/// Returns an empty path the cgroup cannot be determined. /// Assumes that cgroupsV2Enabled() is enabled. -std::string cgroupV2OfProcess(); +std::filesystem::path cgroupV2PathOfProcess(); diff --git a/base/base/defines.h b/base/base/defines.h index 2fc54c37bde..5685a6d9833 100644 --- a/base/base/defines.h +++ b/base/base/defines.h @@ -87,10 +87,13 @@ # define ASAN_POISON_MEMORY_REGION(a, b) #endif -#if !defined(ABORT_ON_LOGICAL_ERROR) - #if !defined(NDEBUG) || defined(ADDRESS_SANITIZER) || defined(THREAD_SANITIZER) || defined(MEMORY_SANITIZER) || defined(UNDEFINED_BEHAVIOR_SANITIZER) - #define ABORT_ON_LOGICAL_ERROR - #endif +/// We used to have only ABORT_ON_LOGICAL_ERROR macro, but most of its uses were actually in places where we didn't care about logical errors +/// but wanted to check exactly if the current build type is debug or with sanitizer. This new macro is introduced to fix those places. +#if !defined(DEBUG_OR_SANITIZER_BUILD) +# if !defined(NDEBUG) || defined(ADDRESS_SANITIZER) || defined(THREAD_SANITIZER) || defined(MEMORY_SANITIZER) \ + || defined(UNDEFINED_BEHAVIOR_SANITIZER) +# define DEBUG_OR_SANITIZER_BUILD +# endif #endif /// chassert(x) is similar to assert(x), but: @@ -101,7 +104,7 @@ /// Also it makes sense to call abort() instead of __builtin_unreachable() in debug builds, /// because SIGABRT is easier to debug than SIGTRAP (the second one makes gdb crazy) #if !defined(chassert) - #if defined(ABORT_ON_LOGICAL_ERROR) +# if defined(DEBUG_OR_SANITIZER_BUILD) // clang-format off #include namespace DB diff --git a/base/base/demangle.h b/base/base/demangle.h index ddca264ecab..af9ccad16c1 100644 --- a/base/base/demangle.h +++ b/base/base/demangle.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include diff --git a/base/base/extended_types.h b/base/base/extended_types.h index 796167ab45d..3bf3f4ed31d 100644 --- a/base/base/extended_types.h +++ b/base/base/extended_types.h @@ -108,6 +108,14 @@ struct make_unsigned // NOLINT(readability-identifier-naming) using type = std::make_unsigned_t; }; +template <> struct make_unsigned { using type = UInt8; }; +template <> struct make_unsigned { using type = UInt8; }; +template <> struct make_unsigned { using type = UInt16; }; +template <> struct make_unsigned { using type = UInt16; }; +template <> struct make_unsigned { using type = UInt32; }; +template <> struct make_unsigned { using type = UInt32; }; +template <> struct make_unsigned { using type = UInt64; }; +template <> struct make_unsigned { using type = UInt64; }; template <> struct make_unsigned { using type = UInt128; }; template <> struct make_unsigned { using type = UInt128; }; template <> struct make_unsigned { using type = UInt256; }; @@ -121,6 +129,14 @@ struct make_signed // NOLINT(readability-identifier-naming) using type = std::make_signed_t; }; +template <> struct make_signed { using type = Int8; }; +template <> struct make_signed { using type = Int8; }; +template <> struct make_signed { using type = Int16; }; +template <> struct make_signed { using type = Int16; }; +template <> struct make_signed { using type = Int32; }; +template <> struct make_signed { using type = Int32; }; +template <> struct make_signed { using type = Int64; }; +template <> struct make_signed { using type = Int64; }; template <> struct make_signed { using type = Int128; }; template <> struct make_signed { using type = Int128; }; template <> struct make_signed { using type = Int256; }; diff --git a/base/base/getFQDNOrHostName.cpp b/base/base/getFQDNOrHostName.cpp index 2a4ba8e2e11..6b3da9699b9 100644 --- a/base/base/getFQDNOrHostName.cpp +++ b/base/base/getFQDNOrHostName.cpp @@ -6,6 +6,9 @@ namespace { std::string getFQDNOrHostNameImpl() { +#if defined(OS_DARWIN) + return Poco::Net::DNS::hostName(); +#else try { return Poco::Net::DNS::thisHost().name(); @@ -14,6 +17,7 @@ namespace { return Poco::Net::DNS::hostName(); } +#endif } } diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp index f47cba9833d..afdb6ba068a 100644 --- a/base/base/getMemoryAmount.cpp +++ b/base/base/getMemoryAmount.cpp @@ -23,8 +23,9 @@ std::optional getCgroupsV2MemoryLimit() if (!cgroupsV2MemoryControllerEnabled()) return {}; - std::string cgroup = cgroupV2OfProcess(); - auto current_cgroup = cgroup.empty() ? default_cgroups_mount : (default_cgroups_mount / cgroup); + std::filesystem::path current_cgroup = cgroupV2PathOfProcess(); + if (current_cgroup.empty()) + return {}; /// Open the bottom-most nested memory limit setting file. If there is no such file at the current /// level, try again at the parent level as memory settings are inherited. diff --git a/base/base/isSharedPtrUnique.h b/base/base/isSharedPtrUnique.h new file mode 100644 index 00000000000..c153605ecb1 --- /dev/null +++ b/base/base/isSharedPtrUnique.h @@ -0,0 +1,9 @@ +#pragma once + +#include + +template +bool isSharedPtrUnique(const std::shared_ptr & ptr) +{ + return ptr.use_count() == 1; +} diff --git a/base/base/itoa.cpp b/base/base/itoa.cpp index 9a2d02e3388..60231507c96 100644 --- a/base/base/itoa.cpp +++ b/base/base/itoa.cpp @@ -1,32 +1,3 @@ -// Based on https://github.com/amdn/itoa and combined with our optimizations -// -//=== itoa.cpp - Fast integer to ascii conversion --*- C++ -*-// -// -// The MIT License (MIT) -// Copyright (c) 2016 Arturo Martin-de-Nicolas -// -// Permission is hereby granted, free of charge, to any person obtaining a copy -// of this software and associated documentation files (the "Software"), to deal -// in the Software without restriction, including without limitation the rights -// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell -// copies of the Software, and to permit persons to whom the Software is -// furnished to do so, subject to the following conditions: -// -// The above copyright notice and this permission notice shall be included -// in all copies or substantial portions of the Software. -// -// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR -// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, -// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE -// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER -// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, -// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE -// SOFTWARE. -//===----------------------------------------------------------------------===// - -#include -#include -#include #include #include #include @@ -34,99 +5,15 @@ namespace { -template -ALWAYS_INLINE inline constexpr T pow10(size_t x) -{ - return x ? 10 * pow10(x - 1) : 1; -} - -// Division by a power of 10 is implemented using a multiplicative inverse. -// This strength reduction is also done by optimizing compilers, but -// presently the fastest results are produced by using the values -// for the multiplication and the shift as given by the algorithm -// described by Agner Fog in "Optimizing Subroutines in Assembly Language" -// -// http://www.agner.org/optimize/optimizing_assembly.pdf -// -// "Integer division by a constant (all processors) -// A floating point number can be divided by a constant by multiplying -// with the reciprocal. If we want to do the same with integers, we have -// to scale the reciprocal by 2n and then shift the product to the right -// by n. There are various algorithms for finding a suitable value of n -// and compensating for rounding errors. The algorithm described below -// was invented by Terje Mathisen, Norway, and not published elsewhere." - -/// Division by constant is performed by: -/// 1. Adding 1 if needed; -/// 2. Multiplying by another constant; -/// 3. Shifting right by another constant. -template -struct Division -{ - static constexpr bool add{add_}; - static constexpr UInt multiplier{multiplier_}; - static constexpr unsigned shift{shift_}; -}; - -/// Select a type with appropriate number of bytes from the list of types. -/// First parameter is the number of bytes requested. Then goes a list of types with 1, 2, 4, ... number of bytes. -/// Example: SelectType<4, uint8_t, uint16_t, uint32_t, uint64_t> will select uint32_t. -template -struct SelectType -{ - using Result = typename SelectType::Result; -}; - -template -struct SelectType<1, T, Ts...> -{ - using Result = T; -}; - - -/// Division by 10^N where N is the size of the type. -template -using DivisionBy10PowN = typename SelectType< - N, - Division, /// divide by 10 - Division, /// divide by 100 - Division, /// divide by 10000 - Division /// divide by 100000000 - >::Result; - -template -using UnsignedOfSize = typename SelectType::Result; - -/// Holds the result of dividing an unsigned N-byte variable by 10^N resulting in -template -struct QuotientAndRemainder -{ - UnsignedOfSize quotient; // quotient with fewer than 2*N decimal digits - UnsignedOfSize remainder; // remainder with at most N decimal digits -}; - -template -QuotientAndRemainder inline split(UnsignedOfSize value) -{ - constexpr DivisionBy10PowN division; - - UnsignedOfSize quotient = (division.multiplier * (UnsignedOfSize<2 * N>(value) + division.add)) >> division.shift; - UnsignedOfSize remainder = static_cast>(value - quotient * pow10>(N)); - - return {quotient, remainder}; -} - -ALWAYS_INLINE inline char * outDigit(char * p, uint8_t value) +ALWAYS_INLINE inline char * outOneDigit(char * p, uint8_t value) { *p = '0' + value; - ++p; - return p; + return p + 1; } // Using a lookup table to convert binary numbers from 0 to 99 // into ascii characters as described by Andrei Alexandrescu in // https://www.facebook.com/notes/facebook-engineering/three-optimization-tips-for-c/10151361643253920/ - const char digits[201] = "00010203040506070809" "10111213141516171819" "20212223242526272829" @@ -137,7 +24,6 @@ const char digits[201] = "00010203040506070809" "70717273747576777879" "80818283848586878889" "90919293949596979899"; - ALWAYS_INLINE inline char * outTwoDigits(char * p, uint8_t value) { memcpy(p, &digits[value * 2], 2); @@ -145,153 +31,260 @@ ALWAYS_INLINE inline char * outTwoDigits(char * p, uint8_t value) return p; } -namespace convert +namespace jeaiii { -template -char * head(char * p, UInt u); -template -char * tail(char * p, UInt u); +/* + MIT License -//===----------------------------------------------------------===// -// head: find most significant digit, skip leading zeros -//===----------------------------------------------------------===// + Copyright (c) 2022 James Edward Anhalt III - https://github.com/jeaiii/itoa -// "x" contains quotient and remainder after division by 10^N -// quotient is less than 10^N -template -ALWAYS_INLINE inline char * head(char * p, QuotientAndRemainder x) + Permission is hereby granted, free of charge, to any person obtaining a copy + of this software and associated documentation files (the "Software"), to deal + in the Software without restriction, including without limitation the rights + to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + copies of the Software, and to permit persons to whom the Software is + furnished to do so, subject to the following conditions: + + The above copyright notice and this permission notice shall be included in all + copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + SOFTWARE. +*/ +struct pair { - p = head(p, UnsignedOfSize(x.quotient)); - p = tail(p, x.remainder); - return p; -} + char dd[2]; + constexpr pair(char c) : dd{c, '\0'} { } /// NOLINT(google-explicit-constructor) + constexpr pair(int n) : dd{"0123456789"[n / 10], "0123456789"[n % 10]} { } /// NOLINT(google-explicit-constructor) +}; -// "u" is less than 10^2*N -template -ALWAYS_INLINE inline char * head(char * p, UInt u) +constexpr struct { - return u < pow10>(N) ? head(p, UnsignedOfSize(u)) : head(p, split(u)); -} + pair dd[100]{ + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, // + 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, // + 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, // + 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, // + 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, // + 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, // + 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, // + 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, // + 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, // + 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, // + }; + pair fd[100]{ + '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', // + 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, // + 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, // + 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, // + 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, // + 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, // + 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, // + 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, // + 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, // + 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, // + }; +} digits; -// recursion base case, selected when "u" is one byte -template <> -ALWAYS_INLINE inline char * head, 1>(char * p, UnsignedOfSize<1> u) +constexpr UInt64 mask24 = (UInt64(1) << 24) - 1; +constexpr UInt64 mask32 = (UInt64(1) << 32) - 1; +constexpr UInt64 mask57 = (UInt64(1) << 57) - 1; + +template +struct _cond { - return u < 10 ? outDigit(p, u) : outTwoDigits(p, u); -} - -//===----------------------------------------------------------===// -// tail: produce all digits including leading zeros -//===----------------------------------------------------------===// - -// recursive step, "u" is less than 10^2*N -template -ALWAYS_INLINE inline char * tail(char * p, UInt u) + using type = F; +}; +template +struct _cond { - QuotientAndRemainder x = split(u); - p = tail(p, UnsignedOfSize(x.quotient)); - p = tail(p, x.remainder); - return p; -} + using type = T; +}; +template +using cond = typename _cond::type; -// recursion base case, selected when "u" is one byte -template <> -ALWAYS_INLINE inline char * tail, 1>(char * p, UnsignedOfSize<1> u) +template +inline ALWAYS_INLINE char * to_text_from_integer(char * b, T i) { - return outTwoDigits(p, u); -} + constexpr auto q = sizeof(T); + using U = cond>>; -//===----------------------------------------------------------===// -// large values are >= 10^2*N -// where x contains quotient and remainder after division by 10^N -//===----------------------------------------------------------===// -template -ALWAYS_INLINE inline char * large(char * p, QuotientAndRemainder x) -{ - QuotientAndRemainder y = split(x.quotient); - p = head(p, UnsignedOfSize(y.quotient)); - p = tail(p, y.remainder); - p = tail(p, x.remainder); - return p; -} + // convert bool to int before test with unary + to silence warning if T happens to be bool + U const n = +i < 0 ? *b++ = '-', U(0) - U(i) : U(i); -//===----------------------------------------------------------===// -// handle values of "u" that might be >= 10^2*N -// where N is the size of "u" in bytes -//===----------------------------------------------------------===// -template -ALWAYS_INLINE inline char * uitoa(char * p, UInt u) -{ - if (u < pow10>(N)) - return head(p, UnsignedOfSize(u)); - QuotientAndRemainder x = split(u); + if (n < U(1e2)) + { + /// This is changed from the original jeaiii implementation + /// For small numbers the extra branch to call outOneDigit() is worth it as it saves some instructions + /// and a memory access (no need to read digits.fd[n]) + /// This is not true for pure random numbers, but that's not the common use case of a database + /// Original jeaii code + // *reinterpret_cast(b) = digits.fd[n]; + // return n < 10 ? b + 1 : b + 2; + return n < 10 ? outOneDigit(b, n) : outTwoDigits(b, n); + } + if (n < UInt32(1e6)) + { + if (sizeof(U) == 1 || n < U(1e4)) + { + auto f0 = UInt32(10 * (1 << 24) / 1e3 + 1) * n; + *reinterpret_cast(b) = digits.fd[f0 >> 24]; + if constexpr (sizeof(U) == 1) + b -= 1; + else + b -= n < U(1e3); + auto f2 = (f0 & mask24) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 24]; + return b + 4; + } + auto f0 = UInt64(10 * (1ull << 32ull) / 1e5 + 1) * n; + *reinterpret_cast(b) = digits.fd[f0 >> 32]; + if constexpr (sizeof(U) == 2) + b -= 1; + else + b -= n < U(1e5); + auto f2 = (f0 & mask32) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 32]; + auto f4 = (f2 & mask32) * 100; + *reinterpret_cast(b + 4) = digits.dd[f4 >> 32]; + return b + 6; + } + if (sizeof(U) == 4 || n < UInt64(1ull << 32ull)) + { + if (n < U(1e8)) + { + auto f0 = UInt64(10 * (1ull << 48ull) / 1e7 + 1) * n >> 16; + *reinterpret_cast(b) = digits.fd[f0 >> 32]; + b -= n < U(1e7); + auto f2 = (f0 & mask32) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 32]; + auto f4 = (f2 & mask32) * 100; + *reinterpret_cast(b + 4) = digits.dd[f4 >> 32]; + auto f6 = (f4 & mask32) * 100; + *reinterpret_cast(b + 6) = digits.dd[f6 >> 32]; + return b + 8; + } + auto f0 = UInt64(10 * (1ull << 57ull) / 1e9 + 1) * n; + *reinterpret_cast(b) = digits.fd[f0 >> 57]; + b -= n < UInt32(1e9); + auto f2 = (f0 & mask57) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 57]; + auto f4 = (f2 & mask57) * 100; + *reinterpret_cast(b + 4) = digits.dd[f4 >> 57]; + auto f6 = (f4 & mask57) * 100; + *reinterpret_cast(b + 6) = digits.dd[f6 >> 57]; + auto f8 = (f6 & mask57) * 100; + *reinterpret_cast(b + 8) = digits.dd[f8 >> 57]; + return b + 10; + } - return u < pow10>(2 * N) ? head(p, x) : large(p, x); -} + // if we get here U must be UInt64 but some compilers don't know that, so reassign n to a UInt64 to avoid warnings + UInt32 z = n % UInt32(1e8); + UInt64 u = n / UInt32(1e8); -// selected when "u" is one byte -template <> -ALWAYS_INLINE inline char * uitoa, 1>(char * p, UnsignedOfSize<1> u) -{ - if (u < 10) - return outDigit(p, u); - else if (u < 100) - return outTwoDigits(p, u); + if (u < UInt32(1e2)) + { + // u can't be 1 digit (if u < 10 it would have been handled above as a 9 digit 32bit number) + *reinterpret_cast(b) = digits.dd[u]; + b += 2; + } + else if (u < UInt32(1e6)) + { + if (u < UInt32(1e4)) + { + auto f0 = UInt32(10 * (1 << 24) / 1e3 + 1) * u; + *reinterpret_cast(b) = digits.fd[f0 >> 24]; + b -= u < UInt32(1e3); + auto f2 = (f0 & mask24) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 24]; + b += 4; + } + else + { + auto f0 = UInt64(10 * (1ull << 32ull) / 1e5 + 1) * u; + *reinterpret_cast(b) = digits.fd[f0 >> 32]; + b -= u < UInt32(1e5); + auto f2 = (f0 & mask32) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 32]; + auto f4 = (f2 & mask32) * 100; + *reinterpret_cast(b + 4) = digits.dd[f4 >> 32]; + b += 6; + } + } + else if (u < UInt32(1e8)) + { + auto f0 = UInt64(10 * (1ull << 48ull) / 1e7 + 1) * u >> 16; + *reinterpret_cast(b) = digits.fd[f0 >> 32]; + b -= u < UInt32(1e7); + auto f2 = (f0 & mask32) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 32]; + auto f4 = (f2 & mask32) * 100; + *reinterpret_cast(b + 4) = digits.dd[f4 >> 32]; + auto f6 = (f4 & mask32) * 100; + *reinterpret_cast(b + 6) = digits.dd[f6 >> 32]; + b += 8; + } + else if (u < UInt64(1ull << 32ull)) + { + auto f0 = UInt64(10 * (1ull << 57ull) / 1e9 + 1) * u; + *reinterpret_cast(b) = digits.fd[f0 >> 57]; + b -= u < UInt32(1e9); + auto f2 = (f0 & mask57) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 57]; + auto f4 = (f2 & mask57) * 100; + *reinterpret_cast(b + 4) = digits.dd[f4 >> 57]; + auto f6 = (f4 & mask57) * 100; + *reinterpret_cast(b + 6) = digits.dd[f6 >> 57]; + auto f8 = (f6 & mask57) * 100; + *reinterpret_cast(b + 8) = digits.dd[f8 >> 57]; + b += 10; + } else { - p = outDigit(p, u / 100); - p = outTwoDigits(p, u % 100); - return p; + UInt32 y = u % UInt32(1e8); + u /= UInt32(1e8); + + // u is 2, 3, or 4 digits (if u < 10 it would have been handled above) + if (u < UInt32(1e2)) + { + *reinterpret_cast(b) = digits.dd[u]; + b += 2; + } + else + { + auto f0 = UInt32(10 * (1 << 24) / 1e3 + 1) * u; + *reinterpret_cast(b) = digits.fd[f0 >> 24]; + b -= u < UInt32(1e3); + auto f2 = (f0 & mask24) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 24]; + b += 4; + } + // do 8 digits + auto f0 = (UInt64((1ull << 48ull) / 1e6 + 1) * y >> 16) + 1; + *reinterpret_cast(b) = digits.dd[f0 >> 32]; + auto f2 = (f0 & mask32) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 32]; + auto f4 = (f2 & mask32) * 100; + *reinterpret_cast(b + 4) = digits.dd[f4 >> 32]; + auto f6 = (f4 & mask32) * 100; + *reinterpret_cast(b + 6) = digits.dd[f6 >> 32]; + b += 8; } -} - -//===----------------------------------------------------------===// -// handle unsigned and signed integral operands -//===----------------------------------------------------------===// - -// itoa: handle unsigned integral operands (selected by SFINAE) -template -requires(!std::is_signed_v && std::is_integral_v) -ALWAYS_INLINE inline char * itoa(U u, char * p) -{ - return convert::uitoa(p, u); -} - -// itoa: handle signed integral operands (selected by SFINAE) -template -requires(std::is_signed_v && std::is_integral_v) -ALWAYS_INLINE inline char * itoa(I i, char * p) -{ - // Need "mask" to be filled with a copy of the sign bit. - // If "i" is a negative value, then the result of "operator >>" - // is implementation-defined, though usually it is an arithmetic - // right shift that replicates the sign bit. - // Use a conditional expression to be portable, - // a good optimizing compiler generates an arithmetic right shift - // and avoids the conditional branch. - UnsignedOfSize mask = i < 0 ? ~UnsignedOfSize(0) : 0; - // Now get the absolute value of "i" and cast to unsigned type UnsignedOfSize. - // Cannot use std::abs() because the result is undefined - // in 2's complement systems for the most-negative value. - // Want to avoid conditional branch for performance reasons since - // CPU branch prediction will be ineffective when negative values - // occur randomly. - // Let "u" be "i" cast to unsigned type UnsignedOfSize. - // Subtract "u" from 2*u if "i" is positive or 0 if "i" is negative. - // This yields the absolute value with the desired type without - // using a conditional branch and without invoking undefined or - // implementation defined behavior: - UnsignedOfSize u = ((2 * UnsignedOfSize(i)) & ~mask) - UnsignedOfSize(i); - // Unconditionally store a minus sign when producing digits - // in a forward direction and increment the pointer only if - // the value is in fact negative. - // This avoids a conditional branch and is safe because we will - // always produce at least one digit and it will overwrite the - // minus sign when the value is not negative. - *p = '-'; - p += (mask & 1); - p = convert::uitoa(p, u); - return p; + // do 8 digits + auto f0 = (UInt64((1ull << 48ull) / 1e6 + 1) * z >> 16) + 1; + *reinterpret_cast(b) = digits.dd[f0 >> 32]; + auto f2 = (f0 & mask32) * 100; + *reinterpret_cast(b + 2) = digits.dd[f2 >> 32]; + auto f4 = (f2 & mask32) * 100; + *reinterpret_cast(b + 4) = digits.dd[f4 >> 32]; + auto f6 = (f4 & mask32) * 100; + *reinterpret_cast(b + 6) = digits.dd[f6 >> 32]; + return b + 8; } } @@ -303,7 +296,7 @@ ALWAYS_INLINE inline char * writeUIntText(UInt128 _x, char * p) { /// If we the highest 64bit item is empty, we can print just the lowest item as u64 if (_x.items[UInt128::_impl::little(1)] == 0) - return convert::itoa(_x.items[UInt128::_impl::little(0)], p); + return jeaiii::to_text_from_integer(p, _x.items[UInt128::_impl::little(0)]); /// Doing operations using __int128 is faster and we already rely on this feature using T = unsigned __int128; @@ -334,7 +327,7 @@ ALWAYS_INLINE inline char * writeUIntText(UInt128 _x, char * p) current_block += max_multiple_of_hundred_blocks; } - char * highest_part_print = convert::itoa(uint64_t(x), p); + char * highest_part_print = jeaiii::to_text_from_integer(p, uint64_t(x)); for (int i = 0; i < current_block; i++) { outTwoDigits(highest_part_print, two_values[current_block - 1 - i]); @@ -450,12 +443,12 @@ ALWAYS_INLINE inline char * writeSIntText(T x, char * pos) char * itoa(UInt8 i, char * p) { - return convert::itoa(uint8_t(i), p); + return jeaiii::to_text_from_integer(p, uint8_t(i)); } char * itoa(Int8 i, char * p) { - return convert::itoa(int8_t(i), p); + return jeaiii::to_text_from_integer(p, int8_t(i)); } char * itoa(UInt128 i, char * p) @@ -481,7 +474,7 @@ char * itoa(Int256 i, char * p) #define DEFAULT_ITOA(T) \ char * itoa(T i, char * p) \ { \ - return convert::itoa(i, p); \ + return jeaiii::to_text_from_integer(p, i); \ } #define FOR_MISSING_INTEGER_TYPES(M) \ diff --git a/base/poco/Foundation/include/Poco/Format.h b/base/poco/Foundation/include/Poco/Format.h index f84be16d3ad..4f91dd44ca5 100644 --- a/base/poco/Foundation/include/Poco/Format.h +++ b/base/poco/Foundation/include/Poco/Format.h @@ -232,7 +232,7 @@ void Foundation_API format( const Any & value10); -void Foundation_API format(std::string & result, const std::string & fmt, const std::vector & values); +void Foundation_API formatVector(std::string & result, const std::string & fmt, const std::vector & values); /// Supports a variable number of arguments and is used by /// all other variants of format(). diff --git a/base/poco/Foundation/include/Poco/Logger.h b/base/poco/Foundation/include/Poco/Logger.h index 2a1cb33b407..74ddceea9dd 100644 --- a/base/poco/Foundation/include/Poco/Logger.h +++ b/base/poco/Foundation/include/Poco/Logger.h @@ -21,6 +21,8 @@ #include #include #include +#include +#include #include #include "Poco/Channel.h" diff --git a/base/poco/Foundation/include/Poco/Message.h b/base/poco/Foundation/include/Poco/Message.h index 9068e56a93c..756e427c5f5 100644 --- a/base/poco/Foundation/include/Poco/Message.h +++ b/base/poco/Foundation/include/Poco/Message.h @@ -19,6 +19,7 @@ #include +#include #include "Poco/Foundation.h" #include "Poco/Timestamp.h" diff --git a/base/poco/Foundation/include/Poco/RefCountedObject.h b/base/poco/Foundation/include/Poco/RefCountedObject.h index db966089e00..d0d964d8390 100644 --- a/base/poco/Foundation/include/Poco/RefCountedObject.h +++ b/base/poco/Foundation/include/Poco/RefCountedObject.h @@ -21,6 +21,8 @@ #include "Poco/AtomicCounter.h" #include "Poco/Foundation.h" +#include + namespace Poco { diff --git a/base/poco/Foundation/src/Format.cpp b/base/poco/Foundation/src/Format.cpp index 9872ddff042..94ab124510d 100644 --- a/base/poco/Foundation/src/Format.cpp +++ b/base/poco/Foundation/src/Format.cpp @@ -51,8 +51,8 @@ namespace } if (width != 0) str.width(width); } - - + + void parsePrec(std::ostream& str, std::string::const_iterator& itFmt, const std::string::const_iterator& endFmt) { if (itFmt != endFmt && *itFmt == '.') @@ -67,7 +67,7 @@ namespace if (prec >= 0) str.precision(prec); } } - + char parseMod(std::string::const_iterator& itFmt, const std::string::const_iterator& endFmt) { char mod = 0; @@ -77,13 +77,13 @@ namespace { case 'l': case 'h': - case 'L': + case 'L': case '?': mod = *itFmt++; break; } } return mod; } - + std::size_t parseIndex(std::string::const_iterator& itFmt, const std::string::const_iterator& endFmt) { int index = 0; @@ -110,8 +110,8 @@ namespace case 'f': str << std::fixed; break; } } - - + + void writeAnyInt(std::ostream& str, const Any& any) { if (any.type() == typeid(char)) @@ -201,7 +201,7 @@ namespace str << RefAnyCast(*itVal++); break; case 'z': - str << AnyCast(*itVal++); + str << AnyCast(*itVal++); break; case 'I': case 'D': @@ -303,7 +303,7 @@ void format(std::string& result, const std::string& fmt, const Any& value) { std::vector args; args.push_back(value); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -312,7 +312,7 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons std::vector args; args.push_back(value1); args.push_back(value2); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -322,7 +322,7 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons args.push_back(value1); args.push_back(value2); args.push_back(value3); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -333,7 +333,7 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons args.push_back(value2); args.push_back(value3); args.push_back(value4); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -345,7 +345,7 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons args.push_back(value3); args.push_back(value4); args.push_back(value5); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -358,7 +358,7 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons args.push_back(value4); args.push_back(value5); args.push_back(value6); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -372,7 +372,7 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons args.push_back(value5); args.push_back(value6); args.push_back(value7); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -387,7 +387,7 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons args.push_back(value6); args.push_back(value7); args.push_back(value8); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -403,7 +403,7 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons args.push_back(value7); args.push_back(value8); args.push_back(value9); - format(result, fmt, args); + formatVector(result, fmt, args); } @@ -420,16 +420,16 @@ void format(std::string& result, const std::string& fmt, const Any& value1, cons args.push_back(value8); args.push_back(value9); args.push_back(value10); - format(result, fmt, args); + formatVector(result, fmt, args); } -void format(std::string& result, const std::string& fmt, const std::vector& values) +void formatVector(std::string& result, const std::string& fmt, const std::vector& values) { std::string::const_iterator itFmt = fmt.begin(); std::string::const_iterator endFmt = fmt.end(); std::vector::const_iterator itVal = values.begin(); - std::vector::const_iterator endVal = values.end(); + std::vector::const_iterator endVal = values.end(); while (itFmt != endFmt) { switch (*itFmt) diff --git a/base/poco/MongoDB/src/ObjectId.cpp b/base/poco/MongoDB/src/ObjectId.cpp index 0125c246c2d..e360d129843 100644 --- a/base/poco/MongoDB/src/ObjectId.cpp +++ b/base/poco/MongoDB/src/ObjectId.cpp @@ -57,7 +57,7 @@ std::string ObjectId::toString(const std::string& fmt) const for (int i = 0; i < 12; ++i) { - s += format(fmt, (unsigned int) _id[i]); + s += Poco::format(fmt, (unsigned int) _id[i]); } return s; } diff --git a/base/poco/MongoDB/src/OpMsgCursor.cpp b/base/poco/MongoDB/src/OpMsgCursor.cpp index bc95851ae33..6abd45ecf76 100644 --- a/base/poco/MongoDB/src/OpMsgCursor.cpp +++ b/base/poco/MongoDB/src/OpMsgCursor.cpp @@ -43,9 +43,9 @@ namespace Poco { namespace MongoDB { -static const std::string keyCursor {"cursor"}; -static const std::string keyFirstBatch {"firstBatch"}; -static const std::string keyNextBatch {"nextBatch"}; +[[ maybe_unused ]] static const std::string keyCursor {"cursor"}; +[[ maybe_unused ]] static const std::string keyFirstBatch {"firstBatch"}; +[[ maybe_unused ]] static const std::string keyNextBatch {"nextBatch"}; static Poco::Int64 cursorIdFromResponse(const MongoDB::Document& doc); @@ -131,7 +131,7 @@ OpMsgMessage& OpMsgCursor::next(Connection& connection) connection.readResponse(_response); } else -#endif +#endif { _response.clear(); _query.setCursor(_cursorID, _batchSize); diff --git a/base/poco/Net/include/Poco/Net/NameValueCollection.h b/base/poco/Net/include/Poco/Net/NameValueCollection.h index be499838d0e..2337535bd11 100644 --- a/base/poco/Net/include/Poco/Net/NameValueCollection.h +++ b/base/poco/Net/include/Poco/Net/NameValueCollection.h @@ -79,7 +79,7 @@ namespace Net /// Returns the value of the first name-value pair with the given name. /// If no value with the given name has been found, the defaultValue is returned. - const std::vector> getAll(const std::string & name) const; + std::vector getAll(const std::string & name) const; /// Returns all values of all name-value pairs with the given name. /// /// Returns an empty vector if there are no name-value pairs with the given name. diff --git a/base/poco/Net/src/HTTPMessage.cpp b/base/poco/Net/src/HTTPMessage.cpp index c0083ec410c..b7ab5543a85 100644 --- a/base/poco/Net/src/HTTPMessage.cpp +++ b/base/poco/Net/src/HTTPMessage.cpp @@ -17,9 +17,9 @@ #include "Poco/NumberFormatter.h" #include "Poco/NumberParser.h" #include "Poco/String.h" +#include #include - using Poco::NumberFormatter; using Poco::NumberParser; using Poco::icompare; @@ -75,7 +75,7 @@ void HTTPMessage::setContentLength(std::streamsize length) erase(CONTENT_LENGTH); } - + std::streamsize HTTPMessage::getContentLength() const { const std::string& contentLength = get(CONTENT_LENGTH, EMPTY); @@ -98,7 +98,7 @@ void HTTPMessage::setContentLength64(Poco::Int64 length) erase(CONTENT_LENGTH); } - + Poco::Int64 HTTPMessage::getContentLength64() const { const std::string& contentLength = get(CONTENT_LENGTH, EMPTY); @@ -133,13 +133,13 @@ void HTTPMessage::setChunkedTransferEncoding(bool flag) setTransferEncoding(IDENTITY_TRANSFER_ENCODING); } - + bool HTTPMessage::getChunkedTransferEncoding() const { return icompare(getTransferEncoding(), CHUNKED_TRANSFER_ENCODING) == 0; } - + void HTTPMessage::setContentType(const std::string& mediaType) { if (mediaType.empty()) @@ -154,7 +154,7 @@ void HTTPMessage::setContentType(const MediaType& mediaType) setContentType(mediaType.toString()); } - + const std::string& HTTPMessage::getContentType() const { return get(CONTENT_TYPE, UNKNOWN_CONTENT_TYPE); diff --git a/base/poco/Net/src/NameValueCollection.cpp b/base/poco/Net/src/NameValueCollection.cpp index 783ed48cc30..e35d66d3bde 100644 --- a/base/poco/Net/src/NameValueCollection.cpp +++ b/base/poco/Net/src/NameValueCollection.cpp @@ -102,9 +102,9 @@ const std::string& NameValueCollection::get(const std::string& name, const std:: return defaultValue; } -const std::vector> NameValueCollection::getAll(const std::string& name) const +std::vector NameValueCollection::getAll(const std::string& name) const { - std::vector> values; + std::vector values; for (ConstIterator it = _map.find(name); it != _map.end(); it++) if (it->first == name) values.push_back(it->second); diff --git a/base/poco/NetSSL_OpenSSL/include/Poco/Net/SecureSocketImpl.h b/base/poco/NetSSL_OpenSSL/include/Poco/Net/SecureSocketImpl.h index 49c12b6b45f..890752c52da 100644 --- a/base/poco/NetSSL_OpenSSL/include/Poco/Net/SecureSocketImpl.h +++ b/base/poco/NetSSL_OpenSSL/include/Poco/Net/SecureSocketImpl.h @@ -235,8 +235,6 @@ namespace Net /// Note that simply closing a socket is not sufficient /// to be able to re-use it again. - Poco::Timespan getMaxTimeout(); - private: SecureSocketImpl(const SecureSocketImpl &); SecureSocketImpl & operator=(const SecureSocketImpl &); @@ -250,6 +248,9 @@ namespace Net Session::Ptr _pSession; friend class SecureStreamSocketImpl; + + Poco::Timespan getMaxTimeoutOrLimit(); + //// Return max(send, receive) if non zero, otherwise maximum timeout }; diff --git a/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp b/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp index efe25f65909..4873d259ae5 100644 --- a/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp +++ b/base/poco/NetSSL_OpenSSL/src/SecureSocketImpl.cpp @@ -199,7 +199,7 @@ void SecureSocketImpl::connectSSL(bool performHandshake) if (performHandshake && _pSocket->getBlocking()) { int ret; - Poco::Timespan remaining_time = getMaxTimeout(); + Poco::Timespan remaining_time = getMaxTimeoutOrLimit(); do { RemainingTimeCounter counter(remaining_time); @@ -302,7 +302,7 @@ int SecureSocketImpl::sendBytes(const void* buffer, int length, int flags) return rc; } - Poco::Timespan remaining_time = getMaxTimeout(); + Poco::Timespan remaining_time = getMaxTimeoutOrLimit(); do { RemainingTimeCounter counter(remaining_time); @@ -338,7 +338,7 @@ int SecureSocketImpl::receiveBytes(void* buffer, int length, int flags) return rc; } - Poco::Timespan remaining_time = getMaxTimeout(); + Poco::Timespan remaining_time = getMaxTimeoutOrLimit(); do { /// SSL record may consist of several TCP packets, @@ -372,7 +372,7 @@ int SecureSocketImpl::completeHandshake() poco_check_ptr (_pSSL); int rc; - Poco::Timespan remaining_time = getMaxTimeout(); + Poco::Timespan remaining_time = getMaxTimeoutOrLimit(); do { RemainingTimeCounter counter(remaining_time); @@ -453,18 +453,29 @@ X509* SecureSocketImpl::peerCertificate() const return 0; } -Poco::Timespan SecureSocketImpl::getMaxTimeout() +Poco::Timespan SecureSocketImpl::getMaxTimeoutOrLimit() { std::lock_guard lock(_mutex); Poco::Timespan remaining_time = _pSocket->getReceiveTimeout(); Poco::Timespan send_timeout = _pSocket->getSendTimeout(); if (remaining_time < send_timeout) remaining_time = send_timeout; + /// zero SO_SNDTIMEO/SO_RCVTIMEO works as no timeout, let's replicate this + /// + /// NOTE: we cannot use INT64_MAX (std::numeric_limits::max()), + /// since it will be later passed to poll() which accept int timeout, and + /// even though poll() accepts milliseconds and Timespan() accepts + /// microseconds, let's use smaller maximum value just to avoid some possible + /// issues, this should be enough anyway (it is ~24 days). + if (remaining_time == 0) + remaining_time = Poco::Timespan(std::numeric_limits::max()); return remaining_time; } bool SecureSocketImpl::mustRetry(int rc, Poco::Timespan& remaining_time) { + if (remaining_time == 0) + return false; std::lock_guard lock(_mutex); if (rc <= 0) { @@ -475,9 +486,7 @@ bool SecureSocketImpl::mustRetry(int rc, Poco::Timespan& remaining_time) case SSL_ERROR_WANT_READ: if (_pSocket->getBlocking()) { - /// Level-triggered mode of epoll_wait is used, so if SSL_read don't read all available data from socket, - /// epoll_wait returns true without waiting for new data even if remaining_time == 0 - if (_pSocket->pollImpl(remaining_time, Poco::Net::Socket::SELECT_READ) && remaining_time != 0) + if (_pSocket->pollImpl(remaining_time, Poco::Net::Socket::SELECT_READ)) return true; else throw Poco::TimeoutException(); @@ -486,13 +495,15 @@ bool SecureSocketImpl::mustRetry(int rc, Poco::Timespan& remaining_time) case SSL_ERROR_WANT_WRITE: if (_pSocket->getBlocking()) { - /// The same as for SSL_ERROR_WANT_READ - if (_pSocket->pollImpl(remaining_time, Poco::Net::Socket::SELECT_WRITE) && remaining_time != 0) + if (_pSocket->pollImpl(remaining_time, Poco::Net::Socket::SELECT_WRITE)) return true; else throw Poco::TimeoutException(); } break; + /// NOTE: POCO_EINTR is the same as SSL_ERROR_WANT_READ (at least in + /// OpenSSL), so this likely dead code, but let's leave it for + /// compatibility with other implementations case SSL_ERROR_SYSCALL: return socketError == POCO_EAGAIN || socketError == POCO_EINTR; default: diff --git a/cmake/autogenerated_versions.txt b/cmake/autogenerated_versions.txt index bb776fa9506..d69646d3694 100644 --- a/cmake/autogenerated_versions.txt +++ b/cmake/autogenerated_versions.txt @@ -2,11 +2,11 @@ # NOTE: VERSION_REVISION has nothing common with DBMS_TCP_PROTOCOL_VERSION, # only DBMS_TCP_PROTOCOL_VERSION should be incremented on protocol changes. -SET(VERSION_REVISION 54488) +SET(VERSION_REVISION 54489) SET(VERSION_MAJOR 24) -SET(VERSION_MINOR 7) +SET(VERSION_MINOR 8) SET(VERSION_PATCH 1) -SET(VERSION_GITHASH aa023477a9265e403982fca5ee29a714db5133d9) -SET(VERSION_DESCRIBE v24.7.1.1-testing) -SET(VERSION_STRING 24.7.1.1) +SET(VERSION_GITHASH 3f8b27d7accd2b5ec4afe7d0dd459115323304af) +SET(VERSION_DESCRIBE v24.8.1.1-testing) +SET(VERSION_STRING 24.8.1.1) # end of autochange diff --git a/cmake/limit_jobs.cmake b/cmake/limit_jobs.cmake index 8e48fc9b9d8..17d8dd42a2c 100644 --- a/cmake/limit_jobs.cmake +++ b/cmake/limit_jobs.cmake @@ -42,9 +42,19 @@ endif () # But use 2 parallel jobs, since: # - this is what llvm does # - and I've verfied that lld-11 does not use all available CPU time (in peak) while linking one binary -if (CMAKE_BUILD_TYPE_UC STREQUAL "RELWITHDEBINFO" AND ENABLE_THINLTO AND PARALLEL_LINK_JOBS GREATER 2) - message(STATUS "ThinLTO provides its own parallel linking - limiting parallel link jobs to 2.") - set (PARALLEL_LINK_JOBS 2) +if (CMAKE_BUILD_TYPE_UC STREQUAL "RELWITHDEBINFO" AND ENABLE_THINLTO) + if (ARCH_AARCH64) + # aarch64 builds start to often fail with OOMs (reason not yet clear), for now let's limit the concurrency + message(STATUS "ThinLTO provides its own parallel linking - limiting parallel link jobs to 1.") + set (PARALLEL_LINK_JOBS 1) + if (LINKER_NAME MATCHES "lld") + math(EXPR LTO_JOBS ${NUMBER_OF_LOGICAL_CORES}/4) + set (CMAKE_EXE_LINKER_FLAGS_RELWITHDEBINFO "${CMAKE_EXE_LINKER_FLAGS_RELWITHDEBINFO} -Wl,--thinlto-jobs=${LTO_JOBS}") + endif() + elseif (PARALLEL_LINK_JOBS GREATER 2) + message(STATUS "ThinLTO provides its own parallel linking - limiting parallel link jobs to 2.") + set (PARALLEL_LINK_JOBS 2) + endif () endif() message(STATUS "Building sub-tree with ${PARALLEL_COMPILE_JOBS} compile jobs and ${PARALLEL_LINK_JOBS} linker jobs (system: ${NUMBER_OF_LOGICAL_CORES} cores, ${TOTAL_PHYSICAL_MEMORY} MB RAM, 'OFF' means the native core count).") diff --git a/cmake/target.cmake b/cmake/target.cmake index d6c497955f6..3d0ecd032f9 100644 --- a/cmake/target.cmake +++ b/cmake/target.cmake @@ -84,5 +84,5 @@ if (CMAKE_CROSSCOMPILING) message (FATAL_ERROR "Trying to cross-compile to unsupported system: ${CMAKE_SYSTEM_NAME}!") endif () - message (STATUS "Cross-compiling for target: ${CMAKE_CXX_COMPILE_TARGET}") + message (STATUS "Cross-compiling for target: ${CMAKE_CXX_COMPILER_TARGET}") endif () diff --git a/contrib/CMakeLists.txt b/contrib/CMakeLists.txt index cf067530285..edea57a3c85 100644 --- a/contrib/CMakeLists.txt +++ b/contrib/CMakeLists.txt @@ -230,6 +230,8 @@ add_contrib (ulid-c-cmake ulid-c) add_contrib (libssh-cmake libssh) +add_contrib (prometheus-protobufs-cmake prometheus-protobufs prometheus-protobufs-gogo) + # Put all targets defined here and in subdirectories under "contrib/" folders in GUI-based IDEs. # Some of third-party projects may override CMAKE_FOLDER or FOLDER property of their targets, so they would not appear # in "contrib/..." as originally planned, so we workaround this by fixing FOLDER properties of all targets manually, diff --git a/contrib/avro b/contrib/avro index d43acc84d3d..545e7002683 160000 --- a/contrib/avro +++ b/contrib/avro @@ -1 +1 @@ -Subproject commit d43acc84d3d455b016f847d6666fbc3cd27f16a9 +Subproject commit 545e7002683cbc2198164d93088ac8e4955b4628 diff --git a/contrib/aws-cmake/CMakeLists.txt b/contrib/aws-cmake/CMakeLists.txt index abde20addaf..250b47b7c2c 100644 --- a/contrib/aws-cmake/CMakeLists.txt +++ b/contrib/aws-cmake/CMakeLists.txt @@ -125,7 +125,7 @@ configure_file("${AWS_SDK_CORE_DIR}/include/aws/core/SDKConfig.h.in" "${CMAKE_CURRENT_BINARY_DIR}/include/aws/core/SDKConfig.h" @ONLY) aws_get_version(AWS_CRT_CPP_VERSION_MAJOR AWS_CRT_CPP_VERSION_MINOR AWS_CRT_CPP_VERSION_PATCH FULL_VERSION GIT_HASH) -configure_file("${AWS_CRT_DIR}/include/aws/crt/Config.h.in" "${AWS_CRT_DIR}/include/aws/crt/Config.h" @ONLY) +configure_file("${AWS_CRT_DIR}/include/aws/crt/Config.h.in" "${CMAKE_CURRENT_BINARY_DIR}/include/aws/crt/Config.h" @ONLY) list(APPEND AWS_SOURCES ${AWS_SDK_CORE_SRC} ${AWS_SDK_CORE_NET_SRC} ${AWS_SDK_CORE_PLATFORM_SRC}) diff --git a/contrib/azure b/contrib/azure index 6262a76ef4c..ea3e19a7be0 160000 --- a/contrib/azure +++ b/contrib/azure @@ -1 +1 @@ -Subproject commit 6262a76ef4c4c330c84e58dd4f6f13f4e6230fcd +Subproject commit ea3e19a7be08519134c643177d56c7484dfec884 diff --git a/contrib/datasketches-cpp-cmake/CMakeLists.txt b/contrib/datasketches-cpp-cmake/CMakeLists.txt index b12a88ad57b..497d6956d0e 100644 --- a/contrib/datasketches-cpp-cmake/CMakeLists.txt +++ b/contrib/datasketches-cpp-cmake/CMakeLists.txt @@ -9,6 +9,7 @@ set(DATASKETCHES_LIBRARY theta) add_library(_datasketches INTERFACE) target_include_directories(_datasketches SYSTEM BEFORE INTERFACE "${ClickHouse_SOURCE_DIR}/contrib/datasketches-cpp/common/include" + "${ClickHouse_SOURCE_DIR}/contrib/datasketches-cpp/count/include" "${ClickHouse_SOURCE_DIR}/contrib/datasketches-cpp/theta/include") add_library(ch_contrib::datasketches ALIAS _datasketches) diff --git a/contrib/google-protobuf-cmake/protobuf_generate.cmake b/contrib/google-protobuf-cmake/protobuf_generate.cmake index 3e30b4e40fd..0731a81aeb8 100644 --- a/contrib/google-protobuf-cmake/protobuf_generate.cmake +++ b/contrib/google-protobuf-cmake/protobuf_generate.cmake @@ -157,15 +157,13 @@ function(protobuf_generate) set(_generated_srcs_all) foreach(_proto ${protobuf_generate_PROTOS}) - get_filename_component(_abs_file ${_proto} ABSOLUTE) - get_filename_component(_abs_dir ${_abs_file} DIRECTORY) - get_filename_component(_basename ${_proto} NAME_WE) - file(RELATIVE_PATH _rel_dir ${CMAKE_CURRENT_SOURCE_DIR} ${_abs_dir}) - - set(_possible_rel_dir) - if (NOT protobuf_generate_APPEND_PATH) - set(_possible_rel_dir ${_rel_dir}/) - endif() + # The protobuf compiler doesn't return paths to the files it generates so we have to calculate those paths here: + # _abs_file - absolute path to a .proto file, + # _possible_rel_dir - relative path to the .proto file from some import directory specified in Protobuf_IMPORT_DIRS, + # _basename - filename of the .proto file (without path and without extenstion). + get_proto_absolute_path(_abs_file "${_proto}" ${_protobuf_include_path}) + get_proto_relative_path(_possible_rel_dir "${_abs_file}" ${_protobuf_include_path}) + get_filename_component(_basename "${_abs_file}" NAME_WE) set(_generated_srcs) foreach(_ext ${protobuf_generate_GENERATE_EXTENSIONS}) @@ -173,7 +171,7 @@ function(protobuf_generate) endforeach() if(protobuf_generate_DESCRIPTORS AND protobuf_generate_LANGUAGE STREQUAL cpp) - set(_descriptor_file "${CMAKE_CURRENT_BINARY_DIR}/${_basename}.desc") + set(_descriptor_file "${protobuf_generate_PROTOC_OUT_DIR}/${_possible_rel_dir}${_basename}.desc") set(_dll_desc_out "--descriptor_set_out=${_descriptor_file}") list(APPEND _generated_srcs ${_descriptor_file}) endif() @@ -196,3 +194,36 @@ function(protobuf_generate) target_sources(${protobuf_generate_TARGET} PRIVATE ${_generated_srcs_all}) endif() endfunction() + +# Calculates the absolute path to a .proto file. +function(get_proto_absolute_path result proto) + cmake_path(IS_ABSOLUTE proto _is_abs_path) + if(_is_abs_path) + set(${result} "${proto}" PARENT_SCOPE) + return() + endif() + foreach(_include_dir ${ARGN}) + if(EXISTS "${_include_dir}/${proto}") + set(${result} "${_include_dir}/${proto}" PARENT_SCOPE) + return() + endif() + endforeach() + message(SEND_ERROR "Not found protobuf ${proto} in Protobuf_IMPORT_DIRS: ${ARGN}") +endfunction() + +# Calculates a relative path to a .proto file. The returned path is relative to one of include directories. +function(get_proto_relative_path result abs_path) + set(${result} "" PARENT_SCOPE) + get_filename_component(_abs_dir "${abs_path}" DIRECTORY) + foreach(_include_dir ${ARGN}) + cmake_path(IS_PREFIX _include_dir "${_abs_dir}" _is_prefix) + if(_is_prefix) + file(RELATIVE_PATH _rel_dir "${_include_dir}" "${_abs_dir}") + if(NOT _rel_dir STREQUAL "") + set(${result} "${_rel_dir}/" PARENT_SCOPE) + endif() + return() + endif() + endforeach() + message(WARNING "Not found protobuf ${abs_path} in Protobuf_IMPORT_DIRS: ${ARGN}") +endfunction() diff --git a/contrib/grpc b/contrib/grpc index 77b2737a709..1716359d2e2 160000 --- a/contrib/grpc +++ b/contrib/grpc @@ -1 +1 @@ -Subproject commit 77b2737a709d43d8c6895e3f03ca62b00bd9201c +Subproject commit 1716359d2e28d304a250f9df0e6c0ccad03de8db diff --git a/contrib/icu-cmake/CMakeLists.txt b/contrib/icu-cmake/CMakeLists.txt index a54bd8c1de2..0a650f2bcc0 100644 --- a/contrib/icu-cmake/CMakeLists.txt +++ b/contrib/icu-cmake/CMakeLists.txt @@ -5,7 +5,7 @@ else () endif () if (NOT ENABLE_ICU) - message(STATUS "Not using icu") + message(STATUS "Not using ICU") return() endif() diff --git a/contrib/jemalloc-cmake/CMakeLists.txt b/contrib/jemalloc-cmake/CMakeLists.txt index b633f0fda50..191886981c3 100644 --- a/contrib/jemalloc-cmake/CMakeLists.txt +++ b/contrib/jemalloc-cmake/CMakeLists.txt @@ -34,7 +34,7 @@ if (OS_LINUX) # avoid spurious latencies and additional work associated with # MADV_DONTNEED. See # https://github.com/ClickHouse/ClickHouse/issues/11121 for motivation. - set (JEMALLOC_CONFIG_MALLOC_CONF "percpu_arena:percpu,oversize_threshold:0,muzzy_decay_ms:0,dirty_decay_ms:5000") + set (JEMALLOC_CONFIG_MALLOC_CONF "percpu_arena:percpu,oversize_threshold:0,muzzy_decay_ms:0,dirty_decay_ms:5000,prof:true,prof_active:false,background_thread:true") else() set (JEMALLOC_CONFIG_MALLOC_CONF "oversize_threshold:0,muzzy_decay_ms:0,dirty_decay_ms:5000") endif() @@ -175,12 +175,19 @@ endif () target_compile_definitions(_jemalloc PRIVATE -DJEMALLOC_PROF=1) -# jemalloc provides support for two different libunwind flavors: the original HP libunwind and the one coming with gcc / g++ / libstdc++. -# The latter is identified by `JEMALLOC_PROF_LIBGCC` and uses `_Unwind_Backtrace` method instead of `unw_backtrace`. -# At the time ClickHouse uses LLVM libunwind which follows libgcc's way of backtracking. +# jemalloc provides support two unwind flavors: +# - JEMALLOC_PROF_LIBUNWIND - unw_backtrace() - gnu libunwind (compatible with llvm libunwind) +# - JEMALLOC_PROF_LIBGCC - _Unwind_Backtrace() - the original HP libunwind and the one coming with gcc / g++ / libstdc++. # -# ClickHouse has to provide `unw_backtrace` method by the means of [commit 8e2b31e](https://github.com/ClickHouse/libunwind/commit/8e2b31e766dd502f6df74909e04a7dbdf5182eb1). -target_compile_definitions (_jemalloc PRIVATE -DJEMALLOC_PROF_LIBGCC=1) +# But for JEMALLOC_PROF_LIBGCC it also calls _Unwind_Backtrace() during +# bootstraping of jemalloc, which may lead to deadlock, if the dlsym will do +# allocations somewhere (like glibc does prio 2.34, see [1]). +# +# [1]: https://sourceware.org/git/?p=glibc.git;a=commit;h=fada9018199c21c469ff0e731ef75c6020074ac9 +# +# And since ClickHouse unwind already supports unw_backtrace() we can safely +# switch to it to avoid this deadlock. +target_compile_definitions (_jemalloc PRIVATE -DJEMALLOC_PROF_LIBUNWIND=1) target_link_libraries (_jemalloc PRIVATE unwind) # for RTLD_NEXT diff --git a/contrib/jemalloc-cmake/README b/contrib/jemalloc-cmake/README index 91b58448c1f..89d7c818445 100644 --- a/contrib/jemalloc-cmake/README +++ b/contrib/jemalloc-cmake/README @@ -4,3 +4,14 @@ It allows to integrate JEMalloc into CMake project. - Added JEMALLOC_CONFIG_MALLOC_CONF substitution - Add musl support (USE_MUSL) - Also note, that darwin build requires JEMALLOC_PREFIX, while others do not +- JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE should be disabled + + CLOCK_MONOTONIC_COARSE can go backwards after clock_adjtime(ADJ_FREQUENCY) + Let's disable it for now, and this menas that CLOCK_MONOTONIC will be used, + and this, should not be a problem, since: + - jemalloc do not call clock_gettime() that frequently + - the difference is CLOCK_MONOTONIC 20ns and CLOCK_MONOTONIC_COARSE 4ns + + This can be done with the following command: + + gg JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE | cut -d: -f1 | xargs sed -i 's@#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE@/* #undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE */@' diff --git a/contrib/jemalloc-cmake/include_freebsd_ppc64le/jemalloc/internal/jemalloc_internal_defs.h.in b/contrib/jemalloc-cmake/include_freebsd_ppc64le/jemalloc/internal/jemalloc_internal_defs.h.in index e027ee97d1a..b58d5c1eb8e 100644 --- a/contrib/jemalloc-cmake/include_freebsd_ppc64le/jemalloc/internal/jemalloc_internal_defs.h.in +++ b/contrib/jemalloc-cmake/include_freebsd_ppc64le/jemalloc/internal/jemalloc_internal_defs.h.in @@ -96,7 +96,7 @@ /* * Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available. */ -#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE +/* #undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE */ /* * Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available. diff --git a/contrib/jemalloc-cmake/include_linux_aarch64/jemalloc/internal/jemalloc_internal_defs.h.in b/contrib/jemalloc-cmake/include_linux_aarch64/jemalloc/internal/jemalloc_internal_defs.h.in index 597f2d59933..8bfc4b60f3d 100644 --- a/contrib/jemalloc-cmake/include_linux_aarch64/jemalloc/internal/jemalloc_internal_defs.h.in +++ b/contrib/jemalloc-cmake/include_linux_aarch64/jemalloc/internal/jemalloc_internal_defs.h.in @@ -98,7 +98,7 @@ /* * Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available. */ -#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE +/* #undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE */ /* * Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available. diff --git a/contrib/jemalloc-cmake/include_linux_ppc64le/jemalloc/internal/jemalloc_internal_defs.h.in b/contrib/jemalloc-cmake/include_linux_ppc64le/jemalloc/internal/jemalloc_internal_defs.h.in index 0a72e71496b..d5197c6127f 100644 --- a/contrib/jemalloc-cmake/include_linux_ppc64le/jemalloc/internal/jemalloc_internal_defs.h.in +++ b/contrib/jemalloc-cmake/include_linux_ppc64le/jemalloc/internal/jemalloc_internal_defs.h.in @@ -98,7 +98,7 @@ /* * Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available. */ -#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE +/* #undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE */ /* * Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available. diff --git a/contrib/jemalloc-cmake/include_linux_riscv64/jemalloc/internal/jemalloc_internal_defs.h.in b/contrib/jemalloc-cmake/include_linux_riscv64/jemalloc/internal/jemalloc_internal_defs.h.in index 597f2d59933..8bfc4b60f3d 100644 --- a/contrib/jemalloc-cmake/include_linux_riscv64/jemalloc/internal/jemalloc_internal_defs.h.in +++ b/contrib/jemalloc-cmake/include_linux_riscv64/jemalloc/internal/jemalloc_internal_defs.h.in @@ -98,7 +98,7 @@ /* * Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available. */ -#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE +/* #undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE */ /* * Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available. diff --git a/contrib/jemalloc-cmake/include_linux_s390x/jemalloc/internal/jemalloc_internal_defs.h.in b/contrib/jemalloc-cmake/include_linux_s390x/jemalloc/internal/jemalloc_internal_defs.h.in index 531f2bca0c2..b3f3a489a83 100644 --- a/contrib/jemalloc-cmake/include_linux_s390x/jemalloc/internal/jemalloc_internal_defs.h.in +++ b/contrib/jemalloc-cmake/include_linux_s390x/jemalloc/internal/jemalloc_internal_defs.h.in @@ -96,7 +96,7 @@ /* * Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available. */ -#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE +/* #undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE */ /* * Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available. diff --git a/contrib/jemalloc-cmake/include_linux_x86_64/jemalloc/internal/jemalloc_internal_defs.h.in b/contrib/jemalloc-cmake/include_linux_x86_64/jemalloc/internal/jemalloc_internal_defs.h.in index d21098a4dcc..cf9acdb0b80 100644 --- a/contrib/jemalloc-cmake/include_linux_x86_64/jemalloc/internal/jemalloc_internal_defs.h.in +++ b/contrib/jemalloc-cmake/include_linux_x86_64/jemalloc/internal/jemalloc_internal_defs.h.in @@ -98,7 +98,7 @@ /* * Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available. */ -#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE +/* #undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE */ /* * Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available. diff --git a/contrib/jemalloc-cmake/include_linux_x86_64_musl/jemalloc/internal/jemalloc_internal_defs.h.in b/contrib/jemalloc-cmake/include_linux_x86_64_musl/jemalloc/internal/jemalloc_internal_defs.h.in index e08a2bed2ec..729bd82c2fa 100644 --- a/contrib/jemalloc-cmake/include_linux_x86_64_musl/jemalloc/internal/jemalloc_internal_defs.h.in +++ b/contrib/jemalloc-cmake/include_linux_x86_64_musl/jemalloc/internal/jemalloc_internal_defs.h.in @@ -99,7 +99,7 @@ /* * Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available. */ -#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE +/* #undef JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE */ /* * Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available. diff --git a/contrib/libpq-cmake/CMakeLists.txt b/contrib/libpq-cmake/CMakeLists.txt index 6a0012c01bf..246e19593f6 100644 --- a/contrib/libpq-cmake/CMakeLists.txt +++ b/contrib/libpq-cmake/CMakeLists.txt @@ -54,7 +54,6 @@ set(SRCS "${LIBPQ_SOURCE_DIR}/port/pgstrcasecmp.c" "${LIBPQ_SOURCE_DIR}/port/thread.c" "${LIBPQ_SOURCE_DIR}/port/path.c" - "${LIBPQ_SOURCE_DIR}/port/explicit_bzero.c" ) add_library(_libpq ${SRCS}) diff --git a/contrib/libunwind b/contrib/libunwind index d6a01c46327..a89d904befe 160000 --- a/contrib/libunwind +++ b/contrib/libunwind @@ -1 +1 @@ -Subproject commit d6a01c46327e56fd86beb8aaa31591fcd9a6b7df +Subproject commit a89d904befea07814628c6ce0b44083c4e149c62 diff --git a/contrib/libunwind-cmake/CMakeLists.txt b/contrib/libunwind-cmake/CMakeLists.txt index 37a2f29afcf..b566e8cb9b3 100644 --- a/contrib/libunwind-cmake/CMakeLists.txt +++ b/contrib/libunwind-cmake/CMakeLists.txt @@ -4,9 +4,6 @@ set(LIBUNWIND_CXX_SOURCES "${LIBUNWIND_SOURCE_DIR}/src/libunwind.cpp" "${LIBUNWIND_SOURCE_DIR}/src/Unwind-EHABI.cpp" "${LIBUNWIND_SOURCE_DIR}/src/Unwind-seh.cpp") -if (APPLE) - set(LIBUNWIND_CXX_SOURCES ${LIBUNWIND_CXX_SOURCES} "${LIBUNWIND_SOURCE_DIR}/src/Unwind_AppleExtras.cpp") -endif () set(LIBUNWIND_C_SOURCES "${LIBUNWIND_SOURCE_DIR}/src/UnwindLevel1.c" @@ -32,6 +29,7 @@ set_target_properties(unwind PROPERTIES FOLDER "contrib/libunwind-cmake") target_include_directories(unwind SYSTEM BEFORE PUBLIC $) target_compile_definitions(unwind PRIVATE -D_LIBUNWIND_NO_HEAP=1) +target_compile_definitions(unwind PRIVATE -D_LIBUNWIND_REMEMBER_STACK_ALLOC=1) # NOTE: from this macros sizeof(unw_context_t)/sizeof(unw_cursor_t) is depends, so it should be set always target_compile_definitions(unwind PUBLIC -D_LIBUNWIND_IS_NATIVE_ONLY) diff --git a/contrib/llvm-project b/contrib/llvm-project index d2142eed980..2a8967b60cb 160000 --- a/contrib/llvm-project +++ b/contrib/llvm-project @@ -1 +1 @@ -Subproject commit d2142eed98046a47ff7112e3cc1e197c8a5cd80f +Subproject commit 2a8967b60cbe5bc2df253712bac343cc5263c5fc diff --git a/contrib/mariadb-connector-c b/contrib/mariadb-connector-c index e39608998f5..d0a788c5b9f 160000 --- a/contrib/mariadb-connector-c +++ b/contrib/mariadb-connector-c @@ -1 +1 @@ -Subproject commit e39608998f5f6944ece9ec61f48e9172ec1de660 +Subproject commit d0a788c5b9fcaca2368d9233770d3ca91ea79f88 diff --git a/contrib/openssl b/contrib/openssl index 5d81fa7068f..66deddc1e53 160000 --- a/contrib/openssl +++ b/contrib/openssl @@ -1 +1 @@ -Subproject commit 5d81fa7068fc8c07f4d0997d5b703f3c541a637c +Subproject commit 66deddc1e53cda8706604a019777259372d1bd62 diff --git a/contrib/openssl-cmake/CMakeLists.txt b/contrib/openssl-cmake/CMakeLists.txt index 85de0340996..a9e4a3df698 100644 --- a/contrib/openssl-cmake/CMakeLists.txt +++ b/contrib/openssl-cmake/CMakeLists.txt @@ -1298,7 +1298,6 @@ elseif(ARCH_PPC64LE) ${OPENSSL_SOURCE_DIR}/crypto/camellia/camellia.c ${OPENSSL_SOURCE_DIR}/crypto/camellia/cmll_cbc.c ${OPENSSL_SOURCE_DIR}/crypto/chacha/chacha_enc.c - ${OPENSSL_SOURCE_DIR}/crypto/mem_clr.c ${OPENSSL_SOURCE_DIR}/crypto/rc4/rc4_enc.c ${OPENSSL_SOURCE_DIR}/crypto/rc4/rc4_skey.c ${OPENSSL_SOURCE_DIR}/crypto/sha/keccak1600.c diff --git a/contrib/orc b/contrib/orc index 947cebaf943..bcc025c0982 160000 --- a/contrib/orc +++ b/contrib/orc @@ -1 +1 @@ -Subproject commit 947cebaf9432d708253ac08dc3012daa6b4ede6f +Subproject commit bcc025c09828c556f54cfbdf83a66b9acae7d17f diff --git a/contrib/pocketfft b/contrib/pocketfft index 9efd4da52cf..f4c1aa8aa9c 160000 --- a/contrib/pocketfft +++ b/contrib/pocketfft @@ -1 +1 @@ -Subproject commit 9efd4da52cf8d28d14531d14e43ad9d913807546 +Subproject commit f4c1aa8aa9ce79ad39e80f2c9c41b92ead90fda3 diff --git a/contrib/prometheus-protobufs-cmake/CMakeLists.txt b/contrib/prometheus-protobufs-cmake/CMakeLists.txt new file mode 100644 index 00000000000..8c939902be7 --- /dev/null +++ b/contrib/prometheus-protobufs-cmake/CMakeLists.txt @@ -0,0 +1,34 @@ +option(ENABLE_PROMETHEUS_PROTOBUFS "Enable Prometheus Protobufs" ${ENABLE_PROTOBUF}) + +if(NOT ENABLE_PROMETHEUS_PROTOBUFS) + message(STATUS "Not using prometheus-protobufs") + return() +endif() + +set(Protobuf_INCLUDE_DIR "${ClickHouse_SOURCE_DIR}/contrib/google-protobuf/src") +set(Prometheus_INCLUDE_DIR "${ClickHouse_SOURCE_DIR}/contrib/prometheus-protobufs") +set(GogoProto_INCLUDE_DIR "${ClickHouse_SOURCE_DIR}/contrib/prometheus-protobufs-gogo") + +# Protobuf_IMPORT_DIRS specify where the protobuf compiler will look for .proto files. +set(Old_Protobuf_IMPORT_DIRS ${Protobuf_IMPORT_DIRS}) +list(APPEND Protobuf_IMPORT_DIRS "${Protobuf_INCLUDE_DIR}" "${Prometheus_INCLUDE_DIR}" "${GogoProto_INCLUDE_DIR}") + +PROTOBUF_GENERATE_CPP(prometheus_protobufs_sources prometheus_protobufs_headers + "prompb/remote.proto" + "prompb/types.proto" + "gogoproto/gogo.proto" +) + +set(Protobuf_IMPORT_DIRS ${Old_Protobuf_IMPORT_DIRS}) + +# Ignore warnings while compiling protobuf-generated *.pb.h and *.pb.cpp files. +set (CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -w") + +# Disable clang-tidy for protobuf-generated *.pb.h and *.pb.cpp files. +set (CMAKE_CXX_CLANG_TIDY "") + +add_library(_prometheus_protobufs ${prometheus_protobufs_sources} ${prometheus_protobufs_headers}) +target_include_directories(_prometheus_protobufs SYSTEM PUBLIC "${CMAKE_CURRENT_BINARY_DIR}") +target_link_libraries (_prometheus_protobufs PUBLIC ch_contrib::protobuf) + +add_library (ch_contrib::prometheus_protobufs ALIAS _prometheus_protobufs) diff --git a/contrib/prometheus-protobufs-gogo/LICENSE b/contrib/prometheus-protobufs-gogo/LICENSE new file mode 100644 index 00000000000..16be18e5c50 --- /dev/null +++ b/contrib/prometheus-protobufs-gogo/LICENSE @@ -0,0 +1,35 @@ +Copyright (c) 2022, The Cosmos SDK Authors. All rights reserved. +Copyright (c) 2013, The GoGo Authors. All rights reserved. + +Protocol Buffers for Go with Gadgets + +Go support for Protocol Buffers - Google's data interchange format + +Copyright 2010 The Go Authors. All rights reserved. +https://github.com/golang/protobuf + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are +met: + + * Redistributions of source code must retain the above copyright +notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above +copyright notice, this list of conditions and the following disclaimer +in the documentation and/or other materials provided with the +distribution. + * Neither the name of Google Inc. nor the names of its +contributors may be used to endorse or promote products derived from +this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/contrib/prometheus-protobufs-gogo/README b/contrib/prometheus-protobufs-gogo/README new file mode 100644 index 00000000000..c40bc42df66 --- /dev/null +++ b/contrib/prometheus-protobufs-gogo/README @@ -0,0 +1,4 @@ +File "gogoproto/gogo.proto" was downloaded from the "Protocol Buffers for Go with Gadgets" project: +https://github.com/cosmos/gogoproto/blob/main/gogoproto/gogo.proto + +File "gogoproto/gogo.proto" is used in ClickHouse to compile prometheus protobufs. diff --git a/contrib/prometheus-protobufs-gogo/gogoproto/gogo.proto b/contrib/prometheus-protobufs-gogo/gogoproto/gogo.proto new file mode 100644 index 00000000000..974b36a7ccd --- /dev/null +++ b/contrib/prometheus-protobufs-gogo/gogoproto/gogo.proto @@ -0,0 +1,145 @@ +// Protocol Buffers for Go with Gadgets +// +// Copyright (c) 2013, The GoGo Authors. All rights reserved. +// http://github.com/cosmos/gogoproto +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above +// copyright notice, this list of conditions and the following disclaimer +// in the documentation and/or other materials provided with the +// distribution. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +syntax = "proto2"; +package gogoproto; + +import "google/protobuf/descriptor.proto"; + +option java_package = "com.google.protobuf"; +option java_outer_classname = "GoGoProtos"; +option go_package = "github.com/cosmos/gogoproto/gogoproto"; + +extend google.protobuf.EnumOptions { + optional bool goproto_enum_prefix = 62001; + optional bool goproto_enum_stringer = 62021; + optional bool enum_stringer = 62022; + optional string enum_customname = 62023; + optional bool enumdecl = 62024; +} + +extend google.protobuf.EnumValueOptions { + optional string enumvalue_customname = 66001; +} + +extend google.protobuf.FileOptions { + optional bool goproto_getters_all = 63001; + optional bool goproto_enum_prefix_all = 63002; + optional bool goproto_stringer_all = 63003; + optional bool verbose_equal_all = 63004; + optional bool face_all = 63005; + optional bool gostring_all = 63006; + optional bool populate_all = 63007; + optional bool stringer_all = 63008; + optional bool onlyone_all = 63009; + + optional bool equal_all = 63013; + optional bool description_all = 63014; + optional bool testgen_all = 63015; + optional bool benchgen_all = 63016; + optional bool marshaler_all = 63017; + optional bool unmarshaler_all = 63018; + optional bool stable_marshaler_all = 63019; + + optional bool sizer_all = 63020; + + optional bool goproto_enum_stringer_all = 63021; + optional bool enum_stringer_all = 63022; + + optional bool unsafe_marshaler_all = 63023; + optional bool unsafe_unmarshaler_all = 63024; + + optional bool goproto_extensions_map_all = 63025; + optional bool goproto_unrecognized_all = 63026; + optional bool gogoproto_import = 63027; + optional bool protosizer_all = 63028; + optional bool compare_all = 63029; + optional bool typedecl_all = 63030; + optional bool enumdecl_all = 63031; + + optional bool goproto_registration = 63032; + optional bool messagename_all = 63033; + + optional bool goproto_sizecache_all = 63034; + optional bool goproto_unkeyed_all = 63035; +} + +extend google.protobuf.MessageOptions { + optional bool goproto_getters = 64001; + optional bool goproto_stringer = 64003; + optional bool verbose_equal = 64004; + optional bool face = 64005; + optional bool gostring = 64006; + optional bool populate = 64007; + optional bool stringer = 67008; + optional bool onlyone = 64009; + + optional bool equal = 64013; + optional bool description = 64014; + optional bool testgen = 64015; + optional bool benchgen = 64016; + optional bool marshaler = 64017; + optional bool unmarshaler = 64018; + optional bool stable_marshaler = 64019; + + optional bool sizer = 64020; + + optional bool unsafe_marshaler = 64023; + optional bool unsafe_unmarshaler = 64024; + + optional bool goproto_extensions_map = 64025; + optional bool goproto_unrecognized = 64026; + + optional bool protosizer = 64028; + optional bool compare = 64029; + + optional bool typedecl = 64030; + + optional bool messagename = 64033; + + optional bool goproto_sizecache = 64034; + optional bool goproto_unkeyed = 64035; +} + +extend google.protobuf.FieldOptions { + optional bool nullable = 65001; + optional bool embed = 65002; + optional string customtype = 65003; + optional string customname = 65004; + optional string jsontag = 65005; + optional string moretags = 65006; + optional string casttype = 65007; + optional string castkey = 65008; + optional string castvalue = 65009; + + optional bool stdtime = 65010; + optional bool stdduration = 65011; + optional bool wktpointer = 65012; + + optional string castrepeated = 65013; +} diff --git a/contrib/prometheus-protobufs/LICENSE b/contrib/prometheus-protobufs/LICENSE new file mode 100644 index 00000000000..261eeb9e9f8 --- /dev/null +++ b/contrib/prometheus-protobufs/LICENSE @@ -0,0 +1,201 @@ + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/contrib/prometheus-protobufs/README b/contrib/prometheus-protobufs/README new file mode 100644 index 00000000000..c557e59bb93 --- /dev/null +++ b/contrib/prometheus-protobufs/README @@ -0,0 +1,2 @@ +Files "prompb/remote.proto" and "prompb/types.proto" were downloaded from the Prometheus repository: +https://github.com/prometheus/prometheus/tree/main/prompb diff --git a/contrib/prometheus-protobufs/prompb/remote.proto b/contrib/prometheus-protobufs/prompb/remote.proto new file mode 100644 index 00000000000..50bb25e7fac --- /dev/null +++ b/contrib/prometheus-protobufs/prompb/remote.proto @@ -0,0 +1,88 @@ +// Copyright 2016 Prometheus Team +// Licensed under the Apache License, Version 2.0 (the "License"); +// you may not use this file except in compliance with the License. +// You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, software +// distributed under the License is distributed on an "AS IS" BASIS, +// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +// See the License for the specific language governing permissions and +// limitations under the License. + +syntax = "proto3"; +package prometheus; + +option go_package = "prompb"; + +import "prompb/types.proto"; +import "gogoproto/gogo.proto"; + +message WriteRequest { + repeated prometheus.TimeSeries timeseries = 1 [(gogoproto.nullable) = false]; + // Cortex uses this field to determine the source of the write request. + // We reserve it to avoid any compatibility issues. + reserved 2; + repeated prometheus.MetricMetadata metadata = 3 [(gogoproto.nullable) = false]; +} + +// ReadRequest represents a remote read request. +message ReadRequest { + repeated Query queries = 1; + + enum ResponseType { + // Server will return a single ReadResponse message with matched series that includes list of raw samples. + // It's recommended to use streamed response types instead. + // + // Response headers: + // Content-Type: "application/x-protobuf" + // Content-Encoding: "snappy" + SAMPLES = 0; + // Server will stream a delimited ChunkedReadResponse message that + // contains XOR or HISTOGRAM(!) encoded chunks for a single series. + // Each message is following varint size and fixed size bigendian + // uint32 for CRC32 Castagnoli checksum. + // + // Response headers: + // Content-Type: "application/x-streamed-protobuf; proto=prometheus.ChunkedReadResponse" + // Content-Encoding: "" + STREAMED_XOR_CHUNKS = 1; + } + + // accepted_response_types allows negotiating the content type of the response. + // + // Response types are taken from the list in the FIFO order. If no response type in `accepted_response_types` is + // implemented by server, error is returned. + // For request that do not contain `accepted_response_types` field the SAMPLES response type will be used. + repeated ResponseType accepted_response_types = 2; +} + +// ReadResponse is a response when response_type equals SAMPLES. +message ReadResponse { + // In same order as the request's queries. + repeated QueryResult results = 1; +} + +message Query { + int64 start_timestamp_ms = 1; + int64 end_timestamp_ms = 2; + repeated prometheus.LabelMatcher matchers = 3; + prometheus.ReadHints hints = 4; +} + +message QueryResult { + // Samples within a time series must be ordered by time. + repeated prometheus.TimeSeries timeseries = 1; +} + +// ChunkedReadResponse is a response when response_type equals STREAMED_XOR_CHUNKS. +// We strictly stream full series after series, optionally split by time. This means that a single frame can contain +// partition of the single series, but once a new series is started to be streamed it means that no more chunks will +// be sent for previous one. Series are returned sorted in the same way TSDB block are internally. +message ChunkedReadResponse { + repeated prometheus.ChunkedSeries chunked_series = 1; + + // query_index represents an index of the query from ReadRequest.queries these chunks relates to. + int64 query_index = 2; +} diff --git a/contrib/prometheus-protobufs/prompb/types.proto b/contrib/prometheus-protobufs/prompb/types.proto new file mode 100644 index 00000000000..61fc1e0143e --- /dev/null +++ b/contrib/prometheus-protobufs/prompb/types.proto @@ -0,0 +1,187 @@ +// Copyright 2017 Prometheus Team +// Licensed under the Apache License, Version 2.0 (the "License"); +// you may not use this file except in compliance with the License. +// You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, software +// distributed under the License is distributed on an "AS IS" BASIS, +// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +// See the License for the specific language governing permissions and +// limitations under the License. + +syntax = "proto3"; +package prometheus; + +option go_package = "prompb"; + +import "gogoproto/gogo.proto"; + +message MetricMetadata { + enum MetricType { + UNKNOWN = 0; + COUNTER = 1; + GAUGE = 2; + HISTOGRAM = 3; + GAUGEHISTOGRAM = 4; + SUMMARY = 5; + INFO = 6; + STATESET = 7; + } + + // Represents the metric type, these match the set from Prometheus. + // Refer to github.com/prometheus/common/model/metadata.go for details. + MetricType type = 1; + string metric_family_name = 2; + string help = 4; + string unit = 5; +} + +message Sample { + double value = 1; + // timestamp is in ms format, see model/timestamp/timestamp.go for + // conversion from time.Time to Prometheus timestamp. + int64 timestamp = 2; +} + +message Exemplar { + // Optional, can be empty. + repeated Label labels = 1 [(gogoproto.nullable) = false]; + double value = 2; + // timestamp is in ms format, see model/timestamp/timestamp.go for + // conversion from time.Time to Prometheus timestamp. + int64 timestamp = 3; +} + +// A native histogram, also known as a sparse histogram. +// Original design doc: +// https://docs.google.com/document/d/1cLNv3aufPZb3fNfaJgdaRBZsInZKKIHo9E6HinJVbpM/edit +// The appendix of this design doc also explains the concept of float +// histograms. This Histogram message can represent both, the usual +// integer histogram as well as a float histogram. +message Histogram { + enum ResetHint { + UNKNOWN = 0; // Need to test for a counter reset explicitly. + YES = 1; // This is the 1st histogram after a counter reset. + NO = 2; // There was no counter reset between this and the previous Histogram. + GAUGE = 3; // This is a gauge histogram where counter resets don't happen. + } + + oneof count { // Count of observations in the histogram. + uint64 count_int = 1; + double count_float = 2; + } + double sum = 3; // Sum of observations in the histogram. + // The schema defines the bucket schema. Currently, valid numbers + // are -4 <= n <= 8. They are all for base-2 bucket schemas, where 1 + // is a bucket boundary in each case, and then each power of two is + // divided into 2^n logarithmic buckets. Or in other words, each + // bucket boundary is the previous boundary times 2^(2^-n). In the + // future, more bucket schemas may be added using numbers < -4 or > + // 8. + sint32 schema = 4; + double zero_threshold = 5; // Breadth of the zero bucket. + oneof zero_count { // Count in zero bucket. + uint64 zero_count_int = 6; + double zero_count_float = 7; + } + + // Negative Buckets. + repeated BucketSpan negative_spans = 8 [(gogoproto.nullable) = false]; + // Use either "negative_deltas" or "negative_counts", the former for + // regular histograms with integer counts, the latter for float + // histograms. + repeated sint64 negative_deltas = 9; // Count delta of each bucket compared to previous one (or to zero for 1st bucket). + repeated double negative_counts = 10; // Absolute count of each bucket. + + // Positive Buckets. + repeated BucketSpan positive_spans = 11 [(gogoproto.nullable) = false]; + // Use either "positive_deltas" or "positive_counts", the former for + // regular histograms with integer counts, the latter for float + // histograms. + repeated sint64 positive_deltas = 12; // Count delta of each bucket compared to previous one (or to zero for 1st bucket). + repeated double positive_counts = 13; // Absolute count of each bucket. + + ResetHint reset_hint = 14; + // timestamp is in ms format, see model/timestamp/timestamp.go for + // conversion from time.Time to Prometheus timestamp. + int64 timestamp = 15; +} + +// A BucketSpan defines a number of consecutive buckets with their +// offset. Logically, it would be more straightforward to include the +// bucket counts in the Span. However, the protobuf representation is +// more compact in the way the data is structured here (with all the +// buckets in a single array separate from the Spans). +message BucketSpan { + sint32 offset = 1; // Gap to previous span, or starting point for 1st span (which can be negative). + uint32 length = 2; // Length of consecutive buckets. +} + +// TimeSeries represents samples and labels for a single time series. +message TimeSeries { + // For a timeseries to be valid, and for the samples and exemplars + // to be ingested by the remote system properly, the labels field is required. + repeated Label labels = 1 [(gogoproto.nullable) = false]; + repeated Sample samples = 2 [(gogoproto.nullable) = false]; + repeated Exemplar exemplars = 3 [(gogoproto.nullable) = false]; + repeated Histogram histograms = 4 [(gogoproto.nullable) = false]; +} + +message Label { + string name = 1; + string value = 2; +} + +message Labels { + repeated Label labels = 1 [(gogoproto.nullable) = false]; +} + +// Matcher specifies a rule, which can match or set of labels or not. +message LabelMatcher { + enum Type { + EQ = 0; + NEQ = 1; + RE = 2; + NRE = 3; + } + Type type = 1; + string name = 2; + string value = 3; +} + +message ReadHints { + int64 step_ms = 1; // Query step size in milliseconds. + string func = 2; // String representation of surrounding function or aggregation. + int64 start_ms = 3; // Start time in milliseconds. + int64 end_ms = 4; // End time in milliseconds. + repeated string grouping = 5; // List of label names used in aggregation. + bool by = 6; // Indicate whether it is without or by. + int64 range_ms = 7; // Range vector selector range in milliseconds. +} + +// Chunk represents a TSDB chunk. +// Time range [min, max] is inclusive. +message Chunk { + int64 min_time_ms = 1; + int64 max_time_ms = 2; + + // We require this to match chunkenc.Encoding. + enum Encoding { + UNKNOWN = 0; + XOR = 1; + HISTOGRAM = 2; + FLOAT_HISTOGRAM = 3; + } + Encoding type = 3; + bytes data = 4; +} + +// ChunkedSeries represents single, encoded time series. +message ChunkedSeries { + // Labels should be sorted. + repeated Label labels = 1 [(gogoproto.nullable) = false]; + // Chunks will be in start time order and may overlap. + repeated Chunk chunks = 2 [(gogoproto.nullable) = false]; +} diff --git a/contrib/rocksdb b/contrib/rocksdb index 3a0b80ca9d6..be366233921 160000 --- a/contrib/rocksdb +++ b/contrib/rocksdb @@ -1 +1 @@ -Subproject commit 3a0b80ca9d6eebb38fad7ea3f41dfc9db4f6a984 +Subproject commit be366233921293bd07a84dc4ea6991858665f202 diff --git a/contrib/rocksdb-cmake/CMakeLists.txt b/contrib/rocksdb-cmake/CMakeLists.txt index c4220ba90ac..3a14407166c 100644 --- a/contrib/rocksdb-cmake/CMakeLists.txt +++ b/contrib/rocksdb-cmake/CMakeLists.txt @@ -1,24 +1,17 @@ -option (ENABLE_ROCKSDB "Enable rocksdb library" ${ENABLE_LIBRARIES}) +option (ENABLE_ROCKSDB "Enable RocksDB" ${ENABLE_LIBRARIES}) if (NOT ENABLE_ROCKSDB) - message (STATUS "Not using rocksdb") + message (STATUS "Not using RocksDB") return() endif() -## this file is extracted from `contrib/rocksdb/CMakeLists.txt` -set(ROCKSDB_SOURCE_DIR "${ClickHouse_SOURCE_DIR}/contrib/rocksdb") -list(APPEND CMAKE_MODULE_PATH "${ROCKSDB_SOURCE_DIR}/cmake/modules/") - -set(PORTABLE ON) -## always disable jemalloc for rocksdb by default -## because it introduces non-standard jemalloc APIs +# Always disable jemalloc for rocksdb by default because it introduces non-standard jemalloc APIs option(WITH_JEMALLOC "build with JeMalloc" OFF) -set(USE_SNAPPY OFF) -if (TARGET ch_contrib::snappy) - set(USE_SNAPPY ON) -endif() -option(WITH_SNAPPY "build with SNAPPY" ${USE_SNAPPY}) -## lz4, zlib, zstd is enabled in ClickHouse by default + +option(WITH_LIBURING "build with liburing" OFF) # TODO could try to enable this conditionally, depending on ClickHouse's ENABLE_LIBURING + +# ClickHouse cannot be compiled without snappy, lz4, zlib, zstd +option(WITH_SNAPPY "build with SNAPPY" ON) option(WITH_LZ4 "build with lz4" ON) option(WITH_ZLIB "build with zlib" ON) option(WITH_ZSTD "build with zstd" ON) @@ -26,78 +19,46 @@ option(WITH_ZSTD "build with zstd" ON) # third-party/folly is only validated to work on Linux and Windows for now. # So only turn it on there by default. if(CMAKE_SYSTEM_NAME MATCHES "Linux|Windows") - if(MSVC AND MSVC_VERSION LESS 1910) - # Folly does not compile with MSVC older than VS2017 - option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" OFF) - else() - option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" ON) - endif() + option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" ON) else() option(WITH_FOLLY_DISTRIBUTED_MUTEX "build with folly::DistributedMutex" OFF) endif() -if( NOT DEFINED CMAKE_CXX_STANDARD ) - set(CMAKE_CXX_STANDARD 11) +if(WITH_SNAPPY) + add_definitions(-DSNAPPY) + list(APPEND THIRDPARTY_LIBS ch_contrib::snappy) endif() -if(MSVC) - option(WITH_XPRESS "build with windows built in compression" OFF) - include("${ROCKSDB_SOURCE_DIR}/thirdparty.inc") -else() - if(CMAKE_SYSTEM_NAME MATCHES "FreeBSD" AND NOT CMAKE_SYSTEM_NAME MATCHES "kFreeBSD") - # FreeBSD has jemalloc as default malloc - # but it does not have all the jemalloc files in include/... - set(WITH_JEMALLOC ON) - else() - if(WITH_JEMALLOC AND TARGET ch_contrib::jemalloc) - add_definitions(-DROCKSDB_JEMALLOC -DJEMALLOC_NO_DEMANGLE) - list(APPEND THIRDPARTY_LIBS ch_contrib::jemalloc) - endif() - endif() - - if(WITH_SNAPPY) - add_definitions(-DSNAPPY) - list(APPEND THIRDPARTY_LIBS ch_contrib::snappy) - endif() - - if(WITH_ZLIB) - add_definitions(-DZLIB) - list(APPEND THIRDPARTY_LIBS ch_contrib::zlib) - endif() - - if(WITH_LZ4) - add_definitions(-DLZ4) - list(APPEND THIRDPARTY_LIBS ch_contrib::lz4) - endif() - - if(WITH_ZSTD) - add_definitions(-DZSTD) - list(APPEND THIRDPARTY_LIBS ch_contrib::zstd) - endif() +if(WITH_ZLIB) + add_definitions(-DZLIB) + list(APPEND THIRDPARTY_LIBS ch_contrib::zlib) endif() -if(CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64") - if(POWER9) - set(HAS_POWER9 1) - set(HAS_ALTIVEC 1) - else() - set(HAS_POWER8 1) - set(HAS_ALTIVEC 1) - endif(POWER9) -endif(CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64") +if(WITH_LZ4) + add_definitions(-DLZ4) + list(APPEND THIRDPARTY_LIBS ch_contrib::lz4) +endif() -if(CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64|AARCH64|arm64|ARM64") - set(HAS_ARMV8_CRC 1) - set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -march=armv8-a+crc+crypto -Wno-unused-function") - set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=armv8-a+crc+crypto -Wno-unused-function") -endif(CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64|AARCH64|arm64|ARM64") +if(WITH_ZSTD) + add_definitions(-DZSTD) + list(APPEND THIRDPARTY_LIBS ch_contrib::zstd) +endif() +option(PORTABLE "build a portable binary" ON) -if(ENABLE_AVX2 AND ENABLE_PCLMULQDQ) +if(ENABLE_SSE42 AND ENABLE_PCLMULQDQ) add_definitions(-DHAVE_SSE42) add_definitions(-DHAVE_PCLMUL) endif() +if(CMAKE_SYSTEM_PROCESSOR MATCHES "arm64|aarch64|AARCH64") + set (HAS_ARMV8_CRC 1) + # the original build descriptions set specific flags for ARM. These flags are already subsumed by ClickHouse's general + # ARM flags, see cmake/cpu_features.cmake + # set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -march=armv8-a+crc+crypto -Wno-unused-function") + # set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=armv8-a+crc+crypto -Wno-unused-function") +endif() + set (HAVE_THREAD_LOCAL 1) if(HAVE_THREAD_LOCAL) add_definitions(-DROCKSDB_SUPPORT_THREAD_LOCAL) @@ -107,8 +68,6 @@ if(CMAKE_SYSTEM_NAME MATCHES "Darwin") add_definitions(-DOS_MACOSX) elseif(CMAKE_SYSTEM_NAME MATCHES "Linux") add_definitions(-DOS_LINUX) -elseif(CMAKE_SYSTEM_NAME MATCHES "SunOS") - add_definitions(-DOS_SOLARIS) elseif(CMAKE_SYSTEM_NAME MATCHES "FreeBSD") add_definitions(-DOS_FREEBSD) elseif(CMAKE_SYSTEM_NAME MATCHES "Android") @@ -123,12 +82,10 @@ endif() if (OS_LINUX) add_definitions(-DROCKSDB_SCHED_GETCPU_PRESENT) - add_definitions(-DROCKSDB_AUXV_SYSAUXV_PRESENT) add_definitions(-DROCKSDB_AUXV_GETAUXVAL_PRESENT) -elseif (OS_FREEBSD) - add_definitions(-DROCKSDB_AUXV_SYSAUXV_PRESENT) endif() +set(ROCKSDB_SOURCE_DIR "${ClickHouse_SOURCE_DIR}/contrib/rocksdb") include_directories(${ROCKSDB_SOURCE_DIR}) include_directories("${ROCKSDB_SOURCE_DIR}/include") @@ -136,11 +93,11 @@ if(WITH_FOLLY_DISTRIBUTED_MUTEX) include_directories("${ROCKSDB_SOURCE_DIR}/third-party/folly") endif() -# Main library source code - set(SOURCES ${ROCKSDB_SOURCE_DIR}/cache/cache.cc ${ROCKSDB_SOURCE_DIR}/cache/cache_entry_roles.cc + ${ROCKSDB_SOURCE_DIR}/cache/cache_key.cc + ${ROCKSDB_SOURCE_DIR}/cache/cache_reservation_manager.cc ${ROCKSDB_SOURCE_DIR}/cache/clock_cache.cc ${ROCKSDB_SOURCE_DIR}/cache/lru_cache.cc ${ROCKSDB_SOURCE_DIR}/cache/sharded_cache.cc @@ -156,6 +113,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/db/blob/blob_log_format.cc ${ROCKSDB_SOURCE_DIR}/db/blob/blob_log_sequential_reader.cc ${ROCKSDB_SOURCE_DIR}/db/blob/blob_log_writer.cc + ${ROCKSDB_SOURCE_DIR}/db/blob/prefetch_buffer_collection.cc ${ROCKSDB_SOURCE_DIR}/db/builder.cc ${ROCKSDB_SOURCE_DIR}/db/c.cc ${ROCKSDB_SOURCE_DIR}/db/column_family.cc @@ -229,6 +187,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/env/file_system_tracer.cc ${ROCKSDB_SOURCE_DIR}/env/fs_remap.cc ${ROCKSDB_SOURCE_DIR}/env/mock_env.cc + ${ROCKSDB_SOURCE_DIR}/env/unique_id_gen.cc ${ROCKSDB_SOURCE_DIR}/file/delete_scheduler.cc ${ROCKSDB_SOURCE_DIR}/file/file_prefetch_buffer.cc ${ROCKSDB_SOURCE_DIR}/file/file_util.cc @@ -247,6 +206,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/memory/concurrent_arena.cc ${ROCKSDB_SOURCE_DIR}/memory/jemalloc_nodump_allocator.cc ${ROCKSDB_SOURCE_DIR}/memory/memkind_kmem_allocator.cc + ${ROCKSDB_SOURCE_DIR}/memory/memory_allocator.cc ${ROCKSDB_SOURCE_DIR}/memtable/alloc_tracker.cc ${ROCKSDB_SOURCE_DIR}/memtable/hash_linklist_rep.cc ${ROCKSDB_SOURCE_DIR}/memtable/hash_skiplist_rep.cc @@ -322,6 +282,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/table/table_factory.cc ${ROCKSDB_SOURCE_DIR}/table/table_properties.cc ${ROCKSDB_SOURCE_DIR}/table/two_level_iterator.cc + ${ROCKSDB_SOURCE_DIR}/table/unique_id.cc ${ROCKSDB_SOURCE_DIR}/test_util/sync_point.cc ${ROCKSDB_SOURCE_DIR}/test_util/sync_point_impl.cc ${ROCKSDB_SOURCE_DIR}/test_util/testutil.cc @@ -333,9 +294,12 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/tools/ldb_tool.cc ${ROCKSDB_SOURCE_DIR}/tools/sst_dump_tool.cc ${ROCKSDB_SOURCE_DIR}/tools/trace_analyzer_tool.cc - ${ROCKSDB_SOURCE_DIR}/trace_replay/trace_replay.cc ${ROCKSDB_SOURCE_DIR}/trace_replay/block_cache_tracer.cc ${ROCKSDB_SOURCE_DIR}/trace_replay/io_tracer.cc + ${ROCKSDB_SOURCE_DIR}/trace_replay/trace_record_handler.cc + ${ROCKSDB_SOURCE_DIR}/trace_replay/trace_record_result.cc + ${ROCKSDB_SOURCE_DIR}/trace_replay/trace_record.cc + ${ROCKSDB_SOURCE_DIR}/trace_replay/trace_replay.cc ${ROCKSDB_SOURCE_DIR}/util/coding.cc ${ROCKSDB_SOURCE_DIR}/util/compaction_job_stats_impl.cc ${ROCKSDB_SOURCE_DIR}/util/comparator.cc @@ -347,6 +311,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/util/murmurhash.cc ${ROCKSDB_SOURCE_DIR}/util/random.cc ${ROCKSDB_SOURCE_DIR}/util/rate_limiter.cc + ${ROCKSDB_SOURCE_DIR}/util/regex.cc ${ROCKSDB_SOURCE_DIR}/util/ribbon_config.cc ${ROCKSDB_SOURCE_DIR}/util/slice.cc ${ROCKSDB_SOURCE_DIR}/util/file_checksum_helper.cc @@ -362,18 +327,23 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/utilities/blob_db/blob_db_impl_filesnapshot.cc ${ROCKSDB_SOURCE_DIR}/utilities/blob_db/blob_dump_tool.cc ${ROCKSDB_SOURCE_DIR}/utilities/blob_db/blob_file.cc + ${ROCKSDB_SOURCE_DIR}/utilities/cache_dump_load.cc + ${ROCKSDB_SOURCE_DIR}/utilities/cache_dump_load_impl.cc ${ROCKSDB_SOURCE_DIR}/utilities/cassandra/cassandra_compaction_filter.cc ${ROCKSDB_SOURCE_DIR}/utilities/cassandra/format.cc ${ROCKSDB_SOURCE_DIR}/utilities/cassandra/merge_operator.cc ${ROCKSDB_SOURCE_DIR}/utilities/checkpoint/checkpoint_impl.cc + ${ROCKSDB_SOURCE_DIR}/utilities/compaction_filters.cc ${ROCKSDB_SOURCE_DIR}/utilities/compaction_filters/remove_emptyvalue_compactionfilter.cc ${ROCKSDB_SOURCE_DIR}/utilities/debug.cc ${ROCKSDB_SOURCE_DIR}/utilities/env_mirror.cc ${ROCKSDB_SOURCE_DIR}/utilities/env_timed.cc ${ROCKSDB_SOURCE_DIR}/utilities/fault_injection_env.cc ${ROCKSDB_SOURCE_DIR}/utilities/fault_injection_fs.cc + ${ROCKSDB_SOURCE_DIR}/utilities/fault_injection_secondary_cache.cc ${ROCKSDB_SOURCE_DIR}/utilities/leveldb_options/leveldb_options.cc ${ROCKSDB_SOURCE_DIR}/utilities/memory/memory_util.cc + ${ROCKSDB_SOURCE_DIR}/utilities/merge_operators.cc ${ROCKSDB_SOURCE_DIR}/utilities/merge_operators/bytesxor.cc ${ROCKSDB_SOURCE_DIR}/utilities/merge_operators/max.cc ${ROCKSDB_SOURCE_DIR}/utilities/merge_operators/put.cc @@ -393,6 +363,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/utilities/simulator_cache/sim_cache.cc ${ROCKSDB_SOURCE_DIR}/utilities/table_properties_collectors/compact_on_deletion_collector.cc ${ROCKSDB_SOURCE_DIR}/utilities/trace/file_trace_reader_writer.cc + ${ROCKSDB_SOURCE_DIR}/utilities/trace/replayer_impl.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/lock/lock_manager.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/lock/point/point_lock_tracker.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/lock/point/point_lock_manager.cc @@ -411,6 +382,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/utilities/transactions/write_unprepared_txn.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/write_unprepared_txn_db.cc ${ROCKSDB_SOURCE_DIR}/utilities/ttl/db_ttl_impl.cc + ${ROCKSDB_SOURCE_DIR}/utilities/wal_filter.cc ${ROCKSDB_SOURCE_DIR}/utilities/write_batch_with_index/write_batch_with_index.cc ${ROCKSDB_SOURCE_DIR}/utilities/write_batch_with_index/write_batch_with_index_internal.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/lock/range/range_tree/lib/locktree/concurrent_tree.cc @@ -425,7 +397,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/utilities/transactions/lock/range/range_tree/lib/standalone_port.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/lock/range/range_tree/lib/util/dbt.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/lock/range/range_tree/lib/util/memarena.cc - rocksdb_build_version.cc) + build_version.cc) # generated by hand if(ENABLE_SSE42 AND ENABLE_PCLMULQDQ) set_source_files_properties( @@ -462,5 +434,6 @@ endif() add_library(_rocksdb ${SOURCES}) add_library(ch_contrib::rocksdb ALIAS _rocksdb) target_link_libraries(_rocksdb PRIVATE ${THIRDPARTY_LIBS} ${SYSTEM_LIBS}) + # SYSTEM is required to overcome some issues target_include_directories(_rocksdb SYSTEM BEFORE INTERFACE "${ROCKSDB_SOURCE_DIR}/include") diff --git a/contrib/rocksdb-cmake/rocksdb_build_version.cc b/contrib/rocksdb-cmake/build_version.cc similarity index 100% rename from contrib/rocksdb-cmake/rocksdb_build_version.cc rename to contrib/rocksdb-cmake/build_version.cc diff --git a/contrib/s2geometry b/contrib/s2geometry index 0547c383717..6522a40338d 160000 --- a/contrib/s2geometry +++ b/contrib/s2geometry @@ -1 +1 @@ -Subproject commit 0547c38371777a1c1c8be263a6f05c3bf71bb05b +Subproject commit 6522a40338d58752c2a4227a3fc2bc4107c73e43 diff --git a/contrib/s2geometry-cmake/CMakeLists.txt b/contrib/s2geometry-cmake/CMakeLists.txt index 6632f9c27d5..48562b8cead 100644 --- a/contrib/s2geometry-cmake/CMakeLists.txt +++ b/contrib/s2geometry-cmake/CMakeLists.txt @@ -1,7 +1,7 @@ -option(ENABLE_S2_GEOMETRY "Enable S2 geometry library" ${ENABLE_LIBRARIES}) +option(ENABLE_S2_GEOMETRY "Enable S2 Geometry" ${ENABLE_LIBRARIES}) if (NOT ENABLE_S2_GEOMETRY) - message(STATUS "Not using S2 geometry") + message(STATUS "Not using S2 Geometry") return() endif() @@ -38,6 +38,7 @@ set(S2_SRCS "${S2_SOURCE_DIR}/s2/s2cell_index.cc" "${S2_SOURCE_DIR}/s2/s2cell_union.cc" "${S2_SOURCE_DIR}/s2/s2centroids.cc" + "${S2_SOURCE_DIR}/s2/s2chain_interpolation_query.cc" "${S2_SOURCE_DIR}/s2/s2closest_cell_query.cc" "${S2_SOURCE_DIR}/s2/s2closest_edge_query.cc" "${S2_SOURCE_DIR}/s2/s2closest_point_query.cc" @@ -46,6 +47,7 @@ set(S2_SRCS "${S2_SOURCE_DIR}/s2/s2coords.cc" "${S2_SOURCE_DIR}/s2/s2crossing_edge_query.cc" "${S2_SOURCE_DIR}/s2/s2debug.cc" + "${S2_SOURCE_DIR}/s2/s2density_tree.cc" "${S2_SOURCE_DIR}/s2/s2earth.cc" "${S2_SOURCE_DIR}/s2/s2edge_clipping.cc" "${S2_SOURCE_DIR}/s2/s2edge_crosser.cc" @@ -53,8 +55,10 @@ set(S2_SRCS "${S2_SOURCE_DIR}/s2/s2edge_distances.cc" "${S2_SOURCE_DIR}/s2/s2edge_tessellator.cc" "${S2_SOURCE_DIR}/s2/s2error.cc" + "${S2_SOURCE_DIR}/s2/s2fractal.cc" "${S2_SOURCE_DIR}/s2/s2furthest_edge_query.cc" "${S2_SOURCE_DIR}/s2/s2hausdorff_distance_query.cc" + "${S2_SOURCE_DIR}/s2/s2index_cell_data.cc" "${S2_SOURCE_DIR}/s2/s2latlng.cc" "${S2_SOURCE_DIR}/s2/s2latlng_rect.cc" "${S2_SOURCE_DIR}/s2/s2latlng_rect_bounder.cc" @@ -63,10 +67,10 @@ set(S2_SRCS "${S2_SOURCE_DIR}/s2/s2lax_polyline_shape.cc" "${S2_SOURCE_DIR}/s2/s2loop.cc" "${S2_SOURCE_DIR}/s2/s2loop_measures.cc" + "${S2_SOURCE_DIR}/s2/s2max_distance_targets.cc" "${S2_SOURCE_DIR}/s2/s2measures.cc" "${S2_SOURCE_DIR}/s2/s2memory_tracker.cc" "${S2_SOURCE_DIR}/s2/s2metrics.cc" - "${S2_SOURCE_DIR}/s2/s2max_distance_targets.cc" "${S2_SOURCE_DIR}/s2/s2min_distance_targets.cc" "${S2_SOURCE_DIR}/s2/s2padded_cell.cc" "${S2_SOURCE_DIR}/s2/s2point_compression.cc" @@ -80,10 +84,11 @@ set(S2_SRCS "${S2_SOURCE_DIR}/s2/s2predicates.cc" "${S2_SOURCE_DIR}/s2/s2projections.cc" "${S2_SOURCE_DIR}/s2/s2r2rect.cc" - "${S2_SOURCE_DIR}/s2/s2region.cc" - "${S2_SOURCE_DIR}/s2/s2region_term_indexer.cc" + "${S2_SOURCE_DIR}/s2/s2random.cc" "${S2_SOURCE_DIR}/s2/s2region_coverer.cc" "${S2_SOURCE_DIR}/s2/s2region_intersection.cc" + "${S2_SOURCE_DIR}/s2/s2region_sharder.cc" + "${S2_SOURCE_DIR}/s2/s2region_term_indexer.cc" "${S2_SOURCE_DIR}/s2/s2region_union.cc" "${S2_SOURCE_DIR}/s2/s2shape_index.cc" "${S2_SOURCE_DIR}/s2/s2shape_index_buffered_region.cc" @@ -94,9 +99,12 @@ set(S2_SRCS "${S2_SOURCE_DIR}/s2/s2shapeutil_coding.cc" "${S2_SOURCE_DIR}/s2/s2shapeutil_contains_brute_force.cc" "${S2_SOURCE_DIR}/s2/s2shapeutil_conversion.cc" + "${S2_SOURCE_DIR}/s2/s2shapeutil_count_vertices.cc" "${S2_SOURCE_DIR}/s2/s2shapeutil_edge_iterator.cc" + "${S2_SOURCE_DIR}/s2/s2shapeutil_edge_wrap.cc" "${S2_SOURCE_DIR}/s2/s2shapeutil_get_reference_point.cc" "${S2_SOURCE_DIR}/s2/s2shapeutil_visit_crossing_edge_pairs.cc" + "${S2_SOURCE_DIR}/s2/s2testing.cc" "${S2_SOURCE_DIR}/s2/s2text_format.cc" "${S2_SOURCE_DIR}/s2/s2wedge_relations.cc" "${S2_SOURCE_DIR}/s2/s2winding_operation.cc" @@ -140,6 +148,7 @@ target_link_libraries(_s2 PRIVATE absl::strings absl::type_traits absl::utility + absl::vlog_is_on ) target_include_directories(_s2 SYSTEM BEFORE PUBLIC "${S2_SOURCE_DIR}/") diff --git a/contrib/sysroot b/contrib/sysroot index 39c4713334f..cc385041b22 160000 --- a/contrib/sysroot +++ b/contrib/sysroot @@ -1 +1 @@ -Subproject commit 39c4713334f9f156dbf508f548d510d9129a657c +Subproject commit cc385041b226d1fc28ead14dbab5d40a5f821dd8 diff --git a/contrib/vectorscan b/contrib/vectorscan index 38431d11178..d29730e1cb9 160000 --- a/contrib/vectorscan +++ b/contrib/vectorscan @@ -1 +1 @@ -Subproject commit 38431d111781843741a781a57a6381a527d900a4 +Subproject commit d29730e1cb9daaa66bda63426cdce83505d2c809 diff --git a/contrib/vectorscan-cmake/CMakeLists.txt b/contrib/vectorscan-cmake/CMakeLists.txt index d6c626c1612..35d5fd3dc82 100644 --- a/contrib/vectorscan-cmake/CMakeLists.txt +++ b/contrib/vectorscan-cmake/CMakeLists.txt @@ -1,11 +1,8 @@ -# We use vectorscan, a portable and API/ABI-compatible drop-in replacement for hyperscan. - +# Vectorscan is drop-in replacement for Hyperscan. if ((ARCH_AMD64 AND NOT NO_SSE3_OR_HIGHER) OR ARCH_AARCH64) - option (ENABLE_VECTORSCAN "Enable vectorscan library" ${ENABLE_LIBRARIES}) + option (ENABLE_VECTORSCAN "Enable vectorscan" ${ENABLE_LIBRARIES}) endif() -# TODO PPC should generally work but needs manual generation of ppc/config.h file on a PPC machine - if (NOT ENABLE_VECTORSCAN) message (STATUS "Not using vectorscan") return() @@ -272,34 +269,24 @@ if (ARCH_AARCH64) ) endif() -# TODO -# if (ARCH_PPC64LE) -# list(APPEND SRCS -# "${LIBRARY_DIR}/src/util/supervector/arch/ppc64el/impl.cpp" -# ) -# endif() - add_library (_vectorscan ${SRCS}) -target_compile_options (_vectorscan PRIVATE - -fno-sanitize=undefined # assume the library takes care of itself - -O2 -fno-strict-aliasing -fno-omit-frame-pointer -fvisibility=hidden # options from original build system -) # library has too much debug information if (OMIT_HEAVY_DEBUG_SYMBOLS) target_compile_options (_vectorscan PRIVATE -g0) endif() -# Include version header manually generated by running the original build system -target_include_directories (_vectorscan SYSTEM PRIVATE common) +target_include_directories (_vectorscan SYSTEM PUBLIC "${LIBRARY_DIR}/src") + +# Makes the version header visible. It was generated by running the native build system manually. +# Please update whenever you update vectorscan. +target_include_directories (_vectorscan SYSTEM PUBLIC common) # vectorscan inherited some patched in-source versions of boost headers to fix a bug in # boost 1.69. This bug has been solved long ago but vectorscan's source code still # points to the patched versions, so include it here. target_include_directories (_vectorscan SYSTEM PRIVATE "${LIBRARY_DIR}/include") -target_include_directories (_vectorscan SYSTEM PUBLIC "${LIBRARY_DIR}/src") - # Include platform-specific config header generated by manually running the original build system # Please regenerate these files if you update vectorscan. diff --git a/contrib/vectorscan-cmake/common/hs_version.h b/contrib/vectorscan-cmake/common/hs_version.h index 8315b44fb2a..3d266484095 100644 --- a/contrib/vectorscan-cmake/common/hs_version.h +++ b/contrib/vectorscan-cmake/common/hs_version.h @@ -32,8 +32,12 @@ /** * A version string to identify this release of Hyperscan. */ -#define HS_VERSION_STRING "5.4.7 2022-06-20" +#define HS_VERSION_STRING "5.4.11 2024-07-04" #define HS_VERSION_32BIT ((5 << 24) | (1 << 16) | (7 << 8) | 0) +#define HS_MAJOR 5 +#define HS_MINOR 4 +#define HS_PATCH 11 + #endif /* HS_VERSION_H_C6428FAF8E3713 */ diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile index 24f38740ff5..c59ef1b919a 100644 --- a/docker/keeper/Dockerfile +++ b/docker/keeper/Dockerfile @@ -34,7 +34,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.5.3.5" +ARG VERSION="24.6.2.17" ARG PACKAGES="clickhouse-keeper" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/packager/binary-builder/build.sh b/docker/packager/binary-builder/build.sh index 032aceb0af3..bd5f2fe8466 100755 --- a/docker/packager/binary-builder/build.sh +++ b/docker/packager/binary-builder/build.sh @@ -111,6 +111,7 @@ fi mv ./programs/clickhouse* /output || mv ./programs/*_fuzzer /output [ -x ./programs/self-extracting/clickhouse ] && mv ./programs/self-extracting/clickhouse /output [ -x ./programs/self-extracting/clickhouse-stripped ] && mv ./programs/self-extracting/clickhouse-stripped /output +[ -x ./programs/self-extracting/clickhouse-keeper ] && mv ./programs/self-extracting/clickhouse-keeper /output mv ./src/unit_tests_dbms /output ||: # may not exist for some binary builds mv ./programs/*.dict ./programs/*.options ./programs/*_seed_corpus.zip /output ||: # libFuzzer oss-fuzz compatible infrastructure diff --git a/docker/packager/packager b/docker/packager/packager index 2dcbd8d695e..da4af7fc1be 100755 --- a/docker/packager/packager +++ b/docker/packager/packager @@ -276,10 +276,7 @@ def parse_env_variables( if is_release_build(debug_build, package_type, sanitizer, coverage): cmake_flags.append("-DSPLIT_DEBUG_SYMBOLS=ON") result.append("WITH_PERFORMANCE=1") - if is_cross_arm: - cmake_flags.append("-DBUILD_STANDALONE_KEEPER=1") - else: - result.append("BUILD_MUSL_KEEPER=1") + cmake_flags.append("-DBUILD_STANDALONE_KEEPER=1") elif package_type == "fuzzers": cmake_flags.append("-DENABLE_FUZZING=1") cmake_flags.append("-DENABLE_PROTOBUF=1") diff --git a/docker/reqgenerator.py b/docker/reqgenerator.py new file mode 100644 index 00000000000..6c1d89ac0ac --- /dev/null +++ b/docker/reqgenerator.py @@ -0,0 +1,47 @@ +#!/usr/bin/env python3 +# To run this script you must install docker and piddeptree python package +# + +import subprocess +import os +import sys + + +def build_docker_deps(image_name, imagedir): + cmd = f"""docker run --entrypoint "/bin/bash" {image_name} -c "pip install pipdeptree 2>/dev/null 1>/dev/null && pipdeptree --freeze --warn silence | sed 's/ \+//g' | sort | uniq" > {imagedir}/requirements.txt""" + subprocess.check_call(cmd, shell=True) + + +def check_docker_file_install_with_pip(filepath): + image_name = None + with open(filepath, "r") as f: + for line in f: + if "docker build" in line: + arr = line.split(" ") + if len(arr) > 4: + image_name = arr[4] + if "pip3 install" in line or "pip install" in line: + return image_name, True + return image_name, False + + +def process_affected_images(images_dir): + for root, _dirs, files in os.walk(images_dir): + for f in files: + if f == "Dockerfile": + docker_file_path = os.path.join(root, f) + print("Checking image on path", docker_file_path) + image_name, has_pip = check_docker_file_install_with_pip( + docker_file_path + ) + if has_pip: + print("Find pip in", image_name) + try: + build_docker_deps(image_name, root) + except Exception as ex: + print(ex) + else: + print("Pip not found in", docker_file_path) + + +process_affected_images(sys.argv[1]) diff --git a/docker/server/Dockerfile.alpine b/docker/server/Dockerfile.alpine index c71319a2a7e..240df79aeb1 100644 --- a/docker/server/Dockerfile.alpine +++ b/docker/server/Dockerfile.alpine @@ -32,7 +32,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.5.3.5" +ARG VERSION="24.6.2.17" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu index ed8cf3d657d..ac64655991a 100644 --- a/docker/server/Dockerfile.ubuntu +++ b/docker/server/Dockerfile.ubuntu @@ -28,7 +28,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list ARG REPO_CHANNEL="stable" ARG REPOSITORY="deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb ${REPO_CHANNEL} main" -ARG VERSION="24.5.3.5" +ARG VERSION="24.6.2.17" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" #docker-official-library:off diff --git a/docker/test/base/Dockerfile b/docker/test/base/Dockerfile index 2317f84e0cb..a81826ed6b5 100644 --- a/docker/test/base/Dockerfile +++ b/docker/test/base/Dockerfile @@ -23,15 +23,17 @@ RUN apt-get update \ # and MEMORY_LIMIT_EXCEEDED exceptions in Functional tests (total memory limit in Functional tests is ~55.24 GiB). # TSAN will flush shadow memory when reaching this limit. # It may cause false-negatives, but it's better than OOM. -RUN echo "TSAN_OPTIONS='verbosity=1000 halt_on_error=1 abort_on_error=1 history_size=7 memory_limit_mb=46080 second_deadlock_stack=1'" >> /etc/environment -RUN echo "UBSAN_OPTIONS='print_stacktrace=1'" >> /etc/environment -RUN echo "MSAN_OPTIONS='abort_on_error=1 poison_in_dtor=1'" >> /etc/environment -RUN echo "LSAN_OPTIONS='suppressions=/usr/share/clickhouse-test/config/lsan_suppressions.txt'" >> /etc/environment +# max_allocation_size_mb is set to 32GB, so we have much bigger chance to run into memory limit than the limitation of the sanitizers +RUN echo "TSAN_OPTIONS='verbosity=1000 halt_on_error=1 abort_on_error=1 history_size=7 memory_limit_mb=46080 second_deadlock_stack=1 max_allocation_size_mb=32768'" >> /etc/environment +RUN echo "UBSAN_OPTIONS='print_stacktrace=1 max_allocation_size_mb=32768'" >> /etc/environment +RUN echo "MSAN_OPTIONS='abort_on_error=1 poison_in_dtor=1 max_allocation_size_mb=32768'" >> /etc/environment +RUN echo "LSAN_OPTIONS='suppressions=/usr/share/clickhouse-test/config/lsan_suppressions.txt max_allocation_size_mb=32768'" >> /etc/environment # Sanitizer options for current shell (not current, but the one that will be spawned on "docker run") # (but w/o verbosity for TSAN, otherwise test.reference will not match) -ENV TSAN_OPTIONS='halt_on_error=1 abort_on_error=1 history_size=7 memory_limit_mb=46080 second_deadlock_stack=1' -ENV UBSAN_OPTIONS='print_stacktrace=1' -ENV MSAN_OPTIONS='abort_on_error=1 poison_in_dtor=1' +ENV TSAN_OPTIONS='halt_on_error=1 abort_on_error=1 history_size=7 memory_limit_mb=46080 second_deadlock_stack=1 max_allocation_size_mb=32768' +ENV UBSAN_OPTIONS='print_stacktrace=1 max_allocation_size_mb=32768' +ENV MSAN_OPTIONS='abort_on_error=1 poison_in_dtor=1 max_allocation_size_mb=32768' +ENV LSAN_OPTIONS='max_allocation_size_mb=32768' # for external_symbolizer_path RUN ln -s /usr/bin/llvm-symbolizer-${LLVM_VERSION} /usr/bin/llvm-symbolizer diff --git a/docker/test/fasttest/Dockerfile b/docker/test/fasttest/Dockerfile index e0be261d5e8..4cac2ee6135 100644 --- a/docker/test/fasttest/Dockerfile +++ b/docker/test/fasttest/Dockerfile @@ -19,19 +19,20 @@ RUN apt-get update \ odbcinst \ psmisc \ python3 \ - python3-lxml \ python3-pip \ - python3-requests \ - python3-termcolor \ unixodbc \ pv \ jq \ zstd \ --yes --no-install-recommends \ && apt-get clean \ - && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* + && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* \ + && groupadd --system --gid 1000 clickhouse \ + && useradd --system --gid 1000 --uid 1000 -m clickhouse +# ^ For some reason, groupadd and useradd are needed for tests with 'expect', but I don't know, why. -RUN pip3 install numpy==1.26.3 scipy==1.12.0 pandas==1.5.3 Jinja2==3.1.3 +COPY requirements.txt / +RUN pip3 install --no-cache-dir -r /requirements.txt # This symlink is required by gcc to find the lld linker RUN ln -s /usr/bin/lld-${LLVM_VERSION} /usr/bin/ld.lld @@ -39,6 +40,10 @@ RUN ln -s /usr/bin/lld-${LLVM_VERSION} /usr/bin/ld.lld # https://salsa.debian.org/pkg-llvm-team/llvm-toolchain/-/commit/992e52c0b156a5ba9c6a8a54f8c4857ddd3d371d RUN sed -i '/_IMPORT_CHECK_FILES_FOR_\(mlir-\|llvm-bolt\|merge-fdata\|MLIR\)/ {s|^|#|}' /usr/lib/llvm-${LLVM_VERSION}/lib/cmake/llvm/LLVMExports-*.cmake +# LLVM changes paths for compiler-rt libraries. For some reason clang-18.1.8 cannot catch up libraries from default install path. +# It's very dirty workaround, better to build compiler and LLVM ourself and use it. Details: https://github.com/llvm/llvm-project/issues/95792 +RUN test ! -d /usr/lib/llvm-18/lib/clang/18/lib/x86_64-pc-linux-gnu || ln -s /usr/lib/llvm-18/lib/clang/18/lib/x86_64-pc-linux-gnu /usr/lib/llvm-18/lib/clang/18/lib/x86_64-unknown-linux-gnu + ARG CCACHE_VERSION=4.6.1 RUN mkdir /tmp/ccache \ && cd /tmp/ccache \ diff --git a/docker/test/fasttest/requirements.txt b/docker/test/fasttest/requirements.txt new file mode 100644 index 00000000000..993ea22e5ae --- /dev/null +++ b/docker/test/fasttest/requirements.txt @@ -0,0 +1,41 @@ +Jinja2==3.1.3 +MarkupSafe==2.1.5 +PyJWT==2.3.0 +PyYAML==6.0.1 +Pygments==2.11.2 +SecretStorage==3.3.1 +blinker==1.4 +certifi==2020.6.20 +chardet==4.0.0 +cryptography==3.4.8 +dbus-python==1.2.18 +distro==1.7.0 +httplib2==0.20.2 +idna==3.3 +importlib-metadata==4.6.4 +jeepney==0.7.1 +keyring==23.5.0 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +lxml==4.8.0 +more-itertools==8.10.0 +numpy==1.26.3 +oauthlib==3.2.0 +packaging==24.1 +pandas==1.5.3 +pip==24.1.1 +pipdeptree==2.23.0 +pyparsing==2.4.7 +python-apt==2.4.0+ubuntu3 +python-dateutil==2.9.0.post0 +pytz==2024.1 +requests==2.32.3 +scipy==1.12.0 +setuptools==59.6.0 +six==1.16.0 +termcolor==1.1.0 +urllib3==1.26.5 +wadllib==1.3.6 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/fasttest/run.sh b/docker/test/fasttest/run.sh index 4d5159cfa9e..26283afc86a 100755 --- a/docker/test/fasttest/run.sh +++ b/docker/test/fasttest/run.sh @@ -9,7 +9,7 @@ trap 'kill $(jobs -pr) ||:' EXIT stage=${stage:-} # Compiler version, normally set by Dockerfile -export LLVM_VERSION=${LLVM_VERSION:-17} +export LLVM_VERSION=${LLVM_VERSION:-18} # A variable to pass additional flags to CMake. # Here we explicitly default it to nothing so that bash doesn't complain about @@ -84,6 +84,8 @@ function start_server echo "ClickHouse server pid '$server_pid' started and responded" } +export -f start_server + function clone_root { [ "$UID" -eq 0 ] && git config --global --add safe.directory "$FASTTEST_SOURCE" @@ -254,6 +256,22 @@ function configure rm -f "$FASTTEST_DATA/config.d/secure_ports.xml" } +function timeout_with_logging() { + local exit_code=0 + + timeout -s TERM --preserve-status "${@}" || exit_code="${?}" + + echo "Checking if it is a timeout. The code 124 will indicate a timeout." + if [[ "${exit_code}" -eq "124" ]] + then + echo "The command 'timeout ${*}' has been killed by timeout." + else + echo "No, it isn't a timeout." + fi + + return $exit_code +} + function run_tests { clickhouse-server --version @@ -269,6 +287,11 @@ function run_tests NPROC=1 fi + export CLICKHOUSE_CONFIG_DIR=$FASTTEST_DATA + export CLICKHOUSE_CONFIG="$FASTTEST_DATA/config.xml" + export CLICKHOUSE_USER_FILES="$FASTTEST_DATA/user_files" + export CLICKHOUSE_SCHEMA_FILES="$FASTTEST_DATA/format_schemas" + local test_opts=( --hung-check --fast-tests-only @@ -292,6 +315,8 @@ function run_tests clickhouse stop --pid-path "$FASTTEST_DATA" } +export -f run_tests + case "$stage" in "") ls -la @@ -315,7 +340,7 @@ case "$stage" in configure 2>&1 | ts '%Y-%m-%d %H:%M:%S' | tee "$FASTTEST_OUTPUT/install_log.txt" ;& "run_tests") - run_tests + timeout_with_logging 35m bash -c run_tests ||: /process_functional_tests_result.py --in-results-dir "$FASTTEST_OUTPUT/" \ --out-results-file "$FASTTEST_OUTPUT/test_results.tsv" \ --out-status-file "$FASTTEST_OUTPUT/check_status.tsv" || echo -e "failure\tCannot parse results" > "$FASTTEST_OUTPUT/check_status.tsv" diff --git a/docker/test/fuzzer/Dockerfile b/docker/test/fuzzer/Dockerfile index d3f78ac1d95..e1fb09b8ed5 100644 --- a/docker/test/fuzzer/Dockerfile +++ b/docker/test/fuzzer/Dockerfile @@ -31,7 +31,8 @@ RUN apt-get update \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* -RUN pip3 install Jinja2 +COPY requirements.txt / +RUN pip3 install --no-cache-dir -r /requirements.txt COPY * / diff --git a/docker/test/fuzzer/query-fuzzer-tweaks-users.xml b/docker/test/fuzzer/query-fuzzer-tweaks-users.xml index e2a4976b385..d5b876a4c85 100644 --- a/docker/test/fuzzer/query-fuzzer-tweaks-users.xml +++ b/docker/test/fuzzer/query-fuzzer-tweaks-users.xml @@ -20,7 +20,7 @@ - 10G + 5G diff --git a/docker/test/fuzzer/requirements.txt b/docker/test/fuzzer/requirements.txt new file mode 100644 index 00000000000..3dce93e023b --- /dev/null +++ b/docker/test/fuzzer/requirements.txt @@ -0,0 +1,27 @@ +blinker==1.4 +cryptography==3.4.8 +dbus-python==1.2.18 +distro==1.7.0 +httplib2==0.20.2 +importlib-metadata==4.6.4 +jeepney==0.7.1 +Jinja2==3.1.4 +keyring==23.5.0 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +MarkupSafe==2.1.5 +more-itertools==8.10.0 +oauthlib==3.2.0 +packaging==24.1 +pip==24.1.1 +pipdeptree==2.23.0 +PyJWT==2.3.0 +pyparsing==2.4.7 +python-apt==2.4.0+ubuntu3 +SecretStorage==3.3.1 +setuptools==59.6.0 +six==1.16.0 +wadllib==1.3.6 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/fuzzer/run-fuzzer.sh b/docker/test/fuzzer/run-fuzzer.sh index 6191aeaf304..b8f967ed9c2 100755 --- a/docker/test/fuzzer/run-fuzzer.sh +++ b/docker/test/fuzzer/run-fuzzer.sh @@ -208,7 +208,6 @@ handle SIGPIPE nostop noprint pass handle SIGTERM nostop noprint pass handle SIGUSR1 nostop noprint pass handle SIGUSR2 nostop noprint pass -handle SIGSEGV nostop pass handle SIG$RTMIN nostop noprint pass info signals continue diff --git a/docker/test/integration/base/Dockerfile b/docker/test/integration/base/Dockerfile index 270b40e23a6..469251f648c 100644 --- a/docker/test/integration/base/Dockerfile +++ b/docker/test/integration/base/Dockerfile @@ -33,7 +33,8 @@ RUN apt-get update \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* -RUN pip3 install pycurl +COPY requirements.txt / +RUN pip3 install --no-cache-dir -r requirements.txt && rm -rf /root/.cache/pip # Architecture of the image when BuildKit/buildx is used ARG TARGETARCH diff --git a/docker/test/integration/base/requirements.txt b/docker/test/integration/base/requirements.txt new file mode 100644 index 00000000000..d195d8deaf6 --- /dev/null +++ b/docker/test/integration/base/requirements.txt @@ -0,0 +1,26 @@ +blinker==1.4 +cryptography==3.4.8 +dbus-python==1.2.18 +distro==1.7.0 +httplib2==0.20.2 +importlib-metadata==4.6.4 +jeepney==0.7.1 +keyring==23.5.0 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +more-itertools==8.10.0 +oauthlib==3.2.0 +packaging==24.1 +pip==24.1.1 +pipdeptree==2.23.0 +pycurl==7.45.3 +PyJWT==2.3.0 +pyparsing==2.4.7 +python-apt==2.4.0+ubuntu3 +SecretStorage==3.3.1 +setuptools==59.6.0 +six==1.16.0 +wadllib==1.3.6 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/integration/resolver/Dockerfile b/docker/test/integration/resolver/Dockerfile index 01b9b777614..b35a7262651 100644 --- a/docker/test/integration/resolver/Dockerfile +++ b/docker/test/integration/resolver/Dockerfile @@ -2,4 +2,5 @@ # Helper docker container to run python bottle apps FROM python:3 -RUN python -m pip install bottle +COPY requirements.txt / +RUN python -m pip install --no-cache-dir -r requirements.txt diff --git a/docker/test/integration/resolver/requirements.txt b/docker/test/integration/resolver/requirements.txt new file mode 100644 index 00000000000..fbf85295329 --- /dev/null +++ b/docker/test/integration/resolver/requirements.txt @@ -0,0 +1,6 @@ +bottle==0.12.25 +packaging==24.1 +pip==23.2.1 +pipdeptree==2.23.0 +setuptools==69.0.3 +wheel==0.42.0 diff --git a/docker/test/integration/runner/Dockerfile b/docker/test/integration/runner/Dockerfile index 23d8a37d822..d250b746e7d 100644 --- a/docker/test/integration/runner/Dockerfile +++ b/docker/test/integration/runner/Dockerfile @@ -26,7 +26,6 @@ RUN apt-get update \ libicu-dev \ bsdutils \ curl \ - python3-pika \ liblua5.1-dev \ luajit \ libssl-dev \ @@ -61,49 +60,8 @@ RUN curl -fsSL https://download.docker.com/linux/ubuntu/gpg | apt-key add - \ # kazoo 2.10.0 is broken # https://s3.amazonaws.com/clickhouse-test-reports/59337/524625a1d2f4cc608a3f1059e3df2c30f353a649/integration_tests__asan__analyzer__[5_6].html -RUN python3 -m pip install --no-cache-dir \ - PyMySQL==1.1.0 \ - asyncio==3.4.3 \ - avro==1.10.2 \ - azure-storage-blob==12.19.0 \ - boto3==1.34.24 \ - cassandra-driver==3.29.0 \ - confluent-kafka==2.3.0 \ - delta-spark==2.3.0 \ - dict2xml==1.7.4 \ - dicttoxml==1.7.16 \ - docker==6.1.3 \ - docker-compose==1.29.2 \ - grpcio==1.60.0 \ - grpcio-tools==1.60.0 \ - kafka-python==2.0.2 \ - lz4==4.3.3 \ - minio==7.2.3 \ - nats-py==2.6.0 \ - protobuf==4.25.2 \ - kazoo==2.9.0 \ - psycopg2-binary==2.9.6 \ - pyhdfs==0.3.1 \ - pymongo==3.11.0 \ - pyspark==3.3.2 \ - pytest==7.4.4 \ - pytest-order==1.0.0 \ - pytest-random==0.2 \ - pytest-repeat==0.9.3 \ - pytest-timeout==2.2.0 \ - pytest-xdist==3.5.0 \ - pytest-reportlog==0.4.0 \ - pytz==2023.3.post1 \ - pyyaml==5.3.1 \ - redis==5.0.1 \ - requests-kerberos==0.14.0 \ - tzlocal==2.1 \ - retry==0.9.2 \ - bs4==0.0.2 \ - lxml==5.1.0 \ - urllib3==2.0.7 \ - jwcrypto==1.5.6 -# bs4, lxml are for cloud tests, do not delete +COPY requirements.txt / +RUN python3 -m pip install --no-cache-dir -r requirements.txt # Hudi supports only spark 3.3.*, not 3.4 RUN curl -fsSL -O https://archive.apache.org/dist/spark/spark-3.3.2/spark-3.3.2-bin-hadoop3.tgz \ diff --git a/docker/test/integration/runner/requirements.txt b/docker/test/integration/runner/requirements.txt new file mode 100644 index 00000000000..8a77d8abf77 --- /dev/null +++ b/docker/test/integration/runner/requirements.txt @@ -0,0 +1,113 @@ +PyHDFS==0.3.1 +PyJWT==2.3.0 +PyMySQL==1.1.0 +PyNaCl==1.5.0 +PyYAML==5.3.1 +SecretStorage==3.3.1 +argon2-cffi-bindings==21.2.0 +argon2-cffi==23.1.0 +async-timeout==4.0.3 +asyncio==3.4.3 +attrs==23.2.0 +avro==1.10.2 +azure-core==1.30.1 +azure-storage-blob==12.19.0 +bcrypt==4.1.3 +beautifulsoup4==4.12.3 +blinker==1.4 +boto3==1.34.24 +botocore==1.34.101 +bs4==0.0.2 +cassandra-driver==3.29.0 +certifi==2024.2.2 +cffi==1.16.0 +charset-normalizer==3.3.2 +click==8.1.7 +confluent-kafka==2.3.0 +cryptography==3.4.8 +dbus-python==1.2.18 +decorator==5.1.1 +delta-spark==2.3.0 +dict2xml==1.7.4 +dicttoxml==1.7.16 +distro-info==1.1+ubuntu0.2 +distro==1.7.0 +docker-compose==1.29.2 +docker==6.1.3 +dockerpty==0.4.1 +docopt==0.6.2 +exceptiongroup==1.2.1 +execnet==2.1.1 +geomet==0.2.1.post1 +grpcio-tools==1.60.0 +grpcio==1.60.0 +gssapi==1.8.3 +httplib2==0.20.2 +idna==3.7 +importlib-metadata==4.6.4 +iniconfig==2.0.0 +isodate==0.6.1 +jeepney==0.7.1 +jmespath==1.0.1 +jsonschema==3.2.0 +jwcrypto==1.5.6 +kafka-python==2.0.2 +kazoo==2.9.0 +keyring==23.5.0 +krb5==0.5.1 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +lxml==5.1.0 +lz4==4.3.3 +minio==7.2.3 +more-itertools==8.10.0 +nats-py==2.6.0 +oauthlib==3.2.0 +packaging==24.0 +paramiko==3.4.0 +pika==1.2.0 +pip==24.1.1 +pipdeptree==2.23.0 +pluggy==1.5.0 +protobuf==4.25.2 +psycopg2-binary==2.9.6 +py4j==0.10.9.5 +py==1.11.0 +pycparser==2.22 +pycryptodome==3.20.0 +pymongo==3.11.0 +pyparsing==2.4.7 +pyrsistent==0.20.0 +pyspark==3.3.2 +pyspnego==0.10.2 +pytest-order==1.0.0 +pytest-random==0.2 +pytest-repeat==0.9.3 +pytest-reportlog==0.4.0 +pytest-timeout==2.2.0 +pytest-xdist==3.5.0 +pytest==7.4.4 +python-apt==2.4.0+ubuntu3 +python-dateutil==2.9.0.post0 +python-dotenv==0.21.1 +pytz==2023.3.post1 +redis==5.0.1 +requests-kerberos==0.14.0 +requests==2.31.0 +retry==0.9.2 +s3transfer==0.10.1 +setuptools==59.6.0 +simplejson==3.19.2 +six==1.16.0 +soupsieve==2.5 +texttable==1.7.0 +tomli==2.0.1 +typing_extensions==4.11.0 +tzlocal==2.1 +unattended-upgrades==0.1 +urllib3==2.0.7 +wadllib==1.3.6 +websocket-client==0.59.0 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/libfuzzer/Dockerfile b/docker/test/libfuzzer/Dockerfile index c9802a0e44e..3ffae0cd921 100644 --- a/docker/test/libfuzzer/Dockerfile +++ b/docker/test/libfuzzer/Dockerfile @@ -1,3 +1,4 @@ +# docker build -t clickhouse/libfuzzer . ARG FROM_TAG=latest FROM clickhouse/test-base:$FROM_TAG @@ -29,15 +30,12 @@ RUN apt-get update \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* -RUN pip3 install Jinja2 - -COPY * / +COPY requirements.txt / +RUN pip3 install --no-cache-dir -r /requirements.txt ENV FUZZER_ARGS="-max_total_time=60" SHELL ["/bin/bash", "-c"] -CMD set -o pipefail \ - && timeout -s 9 1h /run_libfuzzer.py 2>&1 | ts "$(printf '%%Y-%%m-%%d %%H:%%M:%%S\t')" | tee main.log # docker run --network=host --volume :/workspace -e PR_TO_TEST=<> -e SHA_TO_TEST=<> clickhouse/libfuzzer diff --git a/docker/test/libfuzzer/requirements.txt b/docker/test/libfuzzer/requirements.txt new file mode 100644 index 00000000000..3dce93e023b --- /dev/null +++ b/docker/test/libfuzzer/requirements.txt @@ -0,0 +1,27 @@ +blinker==1.4 +cryptography==3.4.8 +dbus-python==1.2.18 +distro==1.7.0 +httplib2==0.20.2 +importlib-metadata==4.6.4 +jeepney==0.7.1 +Jinja2==3.1.4 +keyring==23.5.0 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +MarkupSafe==2.1.5 +more-itertools==8.10.0 +oauthlib==3.2.0 +packaging==24.1 +pip==24.1.1 +pipdeptree==2.23.0 +PyJWT==2.3.0 +pyparsing==2.4.7 +python-apt==2.4.0+ubuntu3 +SecretStorage==3.3.1 +setuptools==59.6.0 +six==1.16.0 +wadllib==1.3.6 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/performance-comparison/Dockerfile b/docker/test/performance-comparison/Dockerfile index 1835900b316..c68a39f6f70 100644 --- a/docker/test/performance-comparison/Dockerfile +++ b/docker/test/performance-comparison/Dockerfile @@ -23,7 +23,6 @@ RUN apt-get update \ python3 \ python3-dev \ python3-pip \ - python3-setuptools \ rsync \ tree \ tzdata \ @@ -33,12 +32,14 @@ RUN apt-get update \ cargo \ ripgrep \ zstd \ - && pip3 --no-cache-dir install 'clickhouse-driver==0.2.1' scipy \ && apt-get purge --yes python3-dev g++ \ && apt-get autoremove --yes \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* +COPY requirements.txt / +RUN pip3 --no-cache-dir install -r requirements.txt + COPY run.sh / CMD ["bash", "/run.sh"] diff --git a/docker/test/performance-comparison/requirements.txt b/docker/test/performance-comparison/requirements.txt new file mode 100644 index 00000000000..932527cc022 --- /dev/null +++ b/docker/test/performance-comparison/requirements.txt @@ -0,0 +1,32 @@ +blinker==1.4 +clickhouse-driver==0.2.7 +cryptography==3.4.8 +dbus-python==1.2.18 +distro==1.7.0 +httplib2==0.20.2 +importlib-metadata==4.6.4 +jeepney==0.7.1 +keyring==23.5.0 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +more-itertools==8.10.0 +numpy==1.26.3 +oauthlib==3.2.0 +packaging==24.1 +pip==24.1.1 +pipdeptree==2.23.0 +Pygments==2.11.2 +PyJWT==2.3.0 +pyparsing==2.4.7 +python-apt==2.4.0+ubuntu3 +pytz==2023.4 +PyYAML==6.0.1 +scipy==1.12.0 +SecretStorage==3.3.1 +setuptools==59.6.0 +six==1.16.0 +tzlocal==2.1 +wadllib==1.3.6 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/sqlancer/Dockerfile b/docker/test/sqlancer/Dockerfile index 82fc2598397..3b919ffb3e3 100644 --- a/docker/test/sqlancer/Dockerfile +++ b/docker/test/sqlancer/Dockerfile @@ -6,7 +6,7 @@ ARG apt_archive="http://archive.ubuntu.com" RUN sed -i "s|http://archive.ubuntu.com|$apt_archive|g" /etc/apt/sources.list RUN apt-get update --yes \ - && env DEBIAN_FRONTEND=noninteractive apt-get install wget git default-jdk maven python3 --yes --no-install-recommends \ + && env DEBIAN_FRONTEND=noninteractive apt-get install wget git python3 default-jdk maven --yes --no-install-recommends \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* diff --git a/docker/test/sqllogic/Dockerfile b/docker/test/sqllogic/Dockerfile index 1ea1e52e6fa..1425e12cd84 100644 --- a/docker/test/sqllogic/Dockerfile +++ b/docker/test/sqllogic/Dockerfile @@ -18,11 +18,8 @@ RUN apt-get update --yes \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* -RUN pip3 install \ - numpy \ - pyodbc \ - deepdiff \ - sqlglot +COPY requirements.txt / +RUN pip3 install --no-cache-dir -r /requirements.txt ARG odbc_driver_url="https://github.com/ClickHouse/clickhouse-odbc/releases/download/v1.1.6.20200320/clickhouse-odbc-1.1.6-Linux.tar.gz" diff --git a/docker/test/sqllogic/requirements.txt b/docker/test/sqllogic/requirements.txt new file mode 100644 index 00000000000..abc0a368659 --- /dev/null +++ b/docker/test/sqllogic/requirements.txt @@ -0,0 +1,30 @@ +blinker==1.4 +cryptography==3.4.8 +dbus-python==1.2.18 +deepdiff==7.0.1 +distro==1.7.0 +httplib2==0.20.2 +importlib-metadata==4.6.4 +jeepney==0.7.1 +keyring==23.5.0 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +more-itertools==8.10.0 +numpy==1.26.4 +oauthlib==3.2.0 +ordered-set==4.1.0 +packaging==24.1 +pip==24.1.1 +pipdeptree==2.23.0 +PyJWT==2.3.0 +pyodbc==5.1.0 +pyparsing==2.4.7 +python-apt==2.4.0+ubuntu3 +SecretStorage==3.3.1 +setuptools==59.6.0 +six==1.16.0 +sqlglot==23.16.0 +wadllib==1.3.6 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/sqltest/Dockerfile b/docker/test/sqltest/Dockerfile index 7f59f65761f..71d915b0c7a 100644 --- a/docker/test/sqltest/Dockerfile +++ b/docker/test/sqltest/Dockerfile @@ -14,9 +14,8 @@ RUN apt-get update --yes \ && apt-get clean \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* -RUN pip3 install \ - pyyaml \ - clickhouse-driver +COPY requirements.txt / +RUN pip3 install --no-cache-dir -r /requirements.txt ARG sqltest_repo="https://github.com/elliotchance/sqltest/" diff --git a/docker/test/sqltest/requirements.txt b/docker/test/sqltest/requirements.txt new file mode 100644 index 00000000000..4a0ae3edbac --- /dev/null +++ b/docker/test/sqltest/requirements.txt @@ -0,0 +1,29 @@ +blinker==1.4 +clickhouse-driver==0.2.7 +cryptography==3.4.8 +dbus-python==1.2.18 +distro==1.7.0 +httplib2==0.20.2 +importlib-metadata==4.6.4 +jeepney==0.7.1 +keyring==23.5.0 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +more-itertools==8.10.0 +oauthlib==3.2.0 +packaging==24.1 +pip==24.1.1 +pipdeptree==2.23.0 +PyJWT==2.3.0 +pyparsing==2.4.7 +python-apt==2.4.0+ubuntu3 +pytz==2024.1 +PyYAML==6.0.1 +SecretStorage==3.3.1 +setuptools==59.6.0 +six==1.16.0 +tzlocal==5.2 +wadllib==1.3.6 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/stateful/Dockerfile b/docker/test/stateful/Dockerfile index 355e70f180e..0daf88cad7e 100644 --- a/docker/test/stateful/Dockerfile +++ b/docker/test/stateful/Dockerfile @@ -6,7 +6,6 @@ FROM clickhouse/stateless-test:$FROM_TAG RUN apt-get update -y \ && env DEBIAN_FRONTEND=noninteractive \ apt-get install --yes --no-install-recommends \ - python3-requests \ nodejs \ npm \ && apt-get clean \ diff --git a/docker/test/stateful/run.sh b/docker/test/stateful/run.sh index 09a9f51084b..8e2f1890f89 100755 --- a/docker/test/stateful/run.sh +++ b/docker/test/stateful/run.sh @@ -4,6 +4,9 @@ source /setup_export_logs.sh set -e -x +MAX_RUN_TIME=${MAX_RUN_TIME:-3600} +MAX_RUN_TIME=$((MAX_RUN_TIME == 0 ? 3600 : MAX_RUN_TIME)) + # Choose random timezone for this test run TZ="$(rg -v '#' /usr/share/zoneinfo/zone.tab | awk '{print $3}' | shuf | head -n1)" echo "Choosen random timezone $TZ" @@ -16,11 +19,17 @@ dpkg -i package_folder/clickhouse-client_*.deb ln -s /usr/share/clickhouse-test/clickhouse-test /usr/bin/clickhouse-test +# shellcheck disable=SC1091 +source /utils.lib + # install test configs /usr/share/clickhouse-test/config/install.sh azurite-blob --blobHost 0.0.0.0 --blobPort 10000 --silent --inMemoryPersistence & + ./setup_minio.sh stateful +./mc admin trace clickminio > /test_output/minio.log & +MC_ADMIN_PID=$! config_logs_export_cluster /etc/clickhouse-server/config.d/system_logs_export.yaml @@ -182,8 +191,8 @@ else ENGINE = CollapsingMergeTree(Sign) PARTITION BY toYYYYMM(StartDate) ORDER BY (CounterID, StartDate, intHash32(UserID), VisitID) SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='s3_cache'" - clickhouse-client --query "INSERT INTO test.hits SELECT * FROM datasets.hits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0" - clickhouse-client --query "INSERT INTO test.visits SELECT * FROM datasets.visits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0" + clickhouse-client --query "INSERT INTO test.hits SELECT * FROM datasets.hits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16" + clickhouse-client --query "INSERT INTO test.visits SELECT * FROM datasets.visits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16" clickhouse-client --query "DROP TABLE datasets.visits_v1 SYNC" clickhouse-client --query "DROP TABLE datasets.hits_v1 SYNC" else @@ -191,7 +200,7 @@ else clickhouse-client --query "RENAME TABLE datasets.visits_v1 TO test.visits" fi clickhouse-client --query "CREATE TABLE test.hits_s3 (WatchID UInt64, JavaEnable UInt8, Title String, GoodEvent Int16, EventTime DateTime, EventDate Date, CounterID UInt32, ClientIP UInt32, ClientIP6 FixedString(16), RegionID UInt32, UserID UInt64, CounterClass Int8, OS UInt8, UserAgent UInt8, URL String, Referer String, URLDomain String, RefererDomain String, Refresh UInt8, IsRobot UInt8, RefererCategories Array(UInt16), URLCategories Array(UInt16), URLRegions Array(UInt32), RefererRegions Array(UInt32), ResolutionWidth UInt16, ResolutionHeight UInt16, ResolutionDepth UInt8, FlashMajor UInt8, FlashMinor UInt8, FlashMinor2 String, NetMajor UInt8, NetMinor UInt8, UserAgentMajor UInt16, UserAgentMinor FixedString(2), CookieEnable UInt8, JavascriptEnable UInt8, IsMobile UInt8, MobilePhone UInt8, MobilePhoneModel String, Params String, IPNetworkID UInt32, TraficSourceID Int8, SearchEngineID UInt16, SearchPhrase String, AdvEngineID UInt8, IsArtifical UInt8, WindowClientWidth UInt16, WindowClientHeight UInt16, ClientTimeZone Int16, ClientEventTime DateTime, SilverlightVersion1 UInt8, SilverlightVersion2 UInt8, SilverlightVersion3 UInt32, SilverlightVersion4 UInt16, PageCharset String, CodeVersion UInt32, IsLink UInt8, IsDownload UInt8, IsNotBounce UInt8, FUniqID UInt64, HID UInt32, IsOldCounter UInt8, IsEvent UInt8, IsParameter UInt8, DontCountHits UInt8, WithHash UInt8, HitColor FixedString(1), UTCEventTime DateTime, Age UInt8, Sex UInt8, Income UInt8, Interests UInt16, Robotness UInt8, GeneralInterests Array(UInt16), RemoteIP UInt32, RemoteIP6 FixedString(16), WindowName Int32, OpenerName Int32, HistoryLength Int16, BrowserLanguage FixedString(2), BrowserCountry FixedString(2), SocialNetwork String, SocialAction String, HTTPError UInt16, SendTiming Int32, DNSTiming Int32, ConnectTiming Int32, ResponseStartTiming Int32, ResponseEndTiming Int32, FetchTiming Int32, RedirectTiming Int32, DOMInteractiveTiming Int32, DOMContentLoadedTiming Int32, DOMCompleteTiming Int32, LoadEventStartTiming Int32, LoadEventEndTiming Int32, NSToDOMContentLoadedTiming Int32, FirstPaintTiming Int32, RedirectCount Int8, SocialSourceNetworkID UInt8, SocialSourcePage String, ParamPrice Int64, ParamOrderID String, ParamCurrency FixedString(3), ParamCurrencyID UInt16, GoalsReached Array(UInt32), OpenstatServiceName String, OpenstatCampaignID String, OpenstatAdID String, OpenstatSourceID String, UTMSource String, UTMMedium String, UTMCampaign String, UTMContent String, UTMTerm String, FromTag String, HasGCLID UInt8, RefererHash UInt64, URLHash UInt64, CLID UInt32, YCLID UInt64, ShareService String, ShareURL String, ShareTitle String, ParsedParams Nested(Key1 String, Key2 String, Key3 String, Key4 String, Key5 String, ValueDouble Float64), IslandID FixedString(16), RequestNum UInt32, RequestTry UInt8) ENGINE = MergeTree() PARTITION BY toYYYYMM(EventDate) ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='s3_cache'" - clickhouse-client --query "INSERT INTO test.hits_s3 SELECT * FROM test.hits SETTINGS enable_filesystem_cache_on_write_operations=0" + clickhouse-client --query "INSERT INTO test.hits_s3 SELECT * FROM test.hits SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16" fi clickhouse-client --query "SHOW TABLES FROM test" @@ -213,6 +222,10 @@ function run_tests() ADDITIONAL_OPTIONS+=('--s3-storage') fi + if [[ -n "$USE_AZURE_STORAGE_FOR_MERGE_TREE" ]] && [[ "$USE_AZURE_STORAGE_FOR_MERGE_TREE" -eq 1 ]]; then + ADDITIONAL_OPTIONS+=('--azure-blob-storage') + fi + if [[ -n "$USE_DATABASE_ORDINARY" ]] && [[ "$USE_DATABASE_ORDINARY" -eq 1 ]]; then ADDITIONAL_OPTIONS+=('--db-engine=Ordinary') fi @@ -232,7 +245,25 @@ function run_tests() } export -f run_tests -timeout "$MAX_RUN_TIME" bash -c run_tests ||: + +function timeout_with_logging() { + local exit_code=0 + + timeout -s TERM --preserve-status "${@}" || exit_code="${?}" + + echo "Checking if it is a timeout. The code 124 will indicate a timeout." + if [[ "${exit_code}" -eq "124" ]] + then + echo "The command 'timeout ${*}' has been killed by timeout." + else + echo "No, it isn't a timeout." + fi + + return $exit_code +} + +TIMEOUT=$((MAX_RUN_TIME - 700)) +timeout_with_logging "$TIMEOUT" bash -c run_tests ||: echo "Files in current directory" ls -la ./ @@ -247,6 +278,8 @@ if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]] sudo clickhouse stop --pid-path /var/run/clickhouse-server2 ||: fi +# Kill minio admin client to stop collecting logs +kill $MC_ADMIN_PID rg -Fa "" /var/log/clickhouse-server/clickhouse-server.log ||: zstd --threads=0 < /var/log/clickhouse-server/clickhouse-server.log > /test_output/clickhouse-server.log.zst ||: @@ -268,3 +301,5 @@ if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]] mv /var/log/clickhouse-server/stderr1.log /test_output/ ||: mv /var/log/clickhouse-server/stderr2.log /test_output/ ||: fi + +collect_core_dumps diff --git a/docker/test/stateless/Dockerfile b/docker/test/stateless/Dockerfile index c3d80a7334b..a0e5513a3a2 100644 --- a/docker/test/stateless/Dockerfile +++ b/docker/test/stateless/Dockerfile @@ -25,10 +25,7 @@ RUN apt-get update -y \ openssl \ postgresql-client \ python3 \ - python3-lxml \ python3-pip \ - python3-requests \ - python3-termcolor \ qemu-user-static \ sqlite3 \ sudo \ @@ -51,7 +48,8 @@ RUN curl -OL https://github.com/protocolbuffers/protobuf/releases/download/v${PR && unzip protoc-${PROTOC_VERSION}-linux-x86_64.zip -d /usr/local \ && rm protoc-${PROTOC_VERSION}-linux-x86_64.zip -RUN pip3 install numpy==1.26.3 scipy==1.12.0 pandas==1.5.3 Jinja2==3.1.3 pyarrow==15.0.0 +COPY requirements.txt / +RUN pip3 install --no-cache-dir -r /requirements.txt RUN mkdir -p /tmp/clickhouse-odbc-tmp \ && cd /tmp/clickhouse-odbc-tmp \ @@ -88,6 +86,7 @@ RUN curl -L --no-verbose -O 'https://archive.apache.org/dist/hadoop/common/hadoo ENV MINIO_ROOT_USER="clickhouse" ENV MINIO_ROOT_PASSWORD="clickhouse" ENV EXPORT_S3_STORAGE_POLICIES=1 +ENV CLICKHOUSE_GRPC_CLIENT="/usr/share/clickhouse-utils/grpc-client/clickhouse-grpc-client.py" RUN npm install -g azurite@3.30.0 \ && npm install -g tslib && npm install -g node diff --git a/docker/test/stateless/attach_gdb.lib b/docker/test/stateless/attach_gdb.lib index eb54f920b98..d288288bb17 100644 --- a/docker/test/stateless/attach_gdb.lib +++ b/docker/test/stateless/attach_gdb.lib @@ -20,7 +20,6 @@ handle SIGPIPE nostop noprint pass handle SIGTERM nostop noprint pass handle SIGUSR1 nostop noprint pass handle SIGUSR2 nostop noprint pass -handle SIGSEGV nostop pass handle SIG$RTMIN nostop noprint pass info signals continue diff --git a/docker/test/stateless/requirements.txt b/docker/test/stateless/requirements.txt new file mode 100644 index 00000000000..74860d5fec3 --- /dev/null +++ b/docker/test/stateless/requirements.txt @@ -0,0 +1,53 @@ +awscli==1.22.34 +blinker==1.4 +botocore==1.23.34 +certifi==2020.6.20 +chardet==4.0.0 +colorama==0.4.4 +cryptography==3.4.8 +dbus-python==1.2.18 +distro==1.7.0 +docutils==0.17.1 +grpcio==1.47.0 +gyp==0.1 +httplib2==0.20.2 +idna==3.3 +importlib-metadata==4.6.4 +jeepney==0.7.1 +Jinja2==3.1.3 +jmespath==0.10.0 +keyring==23.5.0 +launchpadlib==1.10.16 +lazr.restfulclient==0.14.4 +lazr.uri==1.0.6 +lxml==4.8.0 +MarkupSafe==2.1.5 +more-itertools==8.10.0 +numpy==1.26.3 +oauthlib==3.2.0 +packaging==24.1 +pandas==1.5.3 +pip==24.1.1 +pipdeptree==2.23.0 +protobuf==4.25.3 +pyarrow==15.0.0 +pyasn1==0.4.8 +PyJWT==2.3.0 +pyparsing==2.4.7 +python-apt==2.4.0+ubuntu3 +python-dateutil==2.8.1 +pytz==2024.1 +PyYAML==6.0.1 +requests==2.32.3 +roman==3.3 +rsa==4.8 +s3transfer==0.5.0 +scipy==1.12.0 +SecretStorage==3.3.1 +setuptools==59.6.0 +six==1.16.0 +termcolor==1.1.0 +urllib3==1.26.5 +wadllib==1.3.6 +wheel==0.37.1 +zipp==1.0.0 diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index 3ce489b9e0e..f9f96c76d59 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -6,19 +6,24 @@ source /setup_export_logs.sh # fail on errors, verbose and export all env variables set -e -x -a +MAX_RUN_TIME=${MAX_RUN_TIME:-9000} +MAX_RUN_TIME=$((MAX_RUN_TIME == 0 ? 9000 : MAX_RUN_TIME)) + +USE_DATABASE_REPLICATED=${USE_DATABASE_REPLICATED:=0} +USE_SHARED_CATALOG=${USE_SHARED_CATALOG:=0} + # Choose random timezone for this test run. # # NOTE: that clickhouse-test will randomize session_timezone by itself as well # (it will choose between default server timezone and something specific). TZ="$(rg -v '#' /usr/share/zoneinfo/zone.tab | awk '{print $3}' | shuf | head -n1)" -echo "Choosen random timezone $TZ" +echo "Chosen random timezone $TZ" ln -snf "/usr/share/zoneinfo/$TZ" /etc/localtime && echo "$TZ" > /etc/timezone dpkg -i package_folder/clickhouse-common-static_*.deb dpkg -i package_folder/clickhouse-common-static-dbg_*.deb -# Accept failure in the next two commands until 24.4 is released (for compatibility and Bugfix validation run) -dpkg -i package_folder/clickhouse-odbc-bridge_*.deb || true -dpkg -i package_folder/clickhouse-library-bridge_*.deb || true +dpkg -i package_folder/clickhouse-odbc-bridge_*.deb +dpkg -i package_folder/clickhouse-library-bridge_*.deb dpkg -i package_folder/clickhouse-server_*.deb dpkg -i package_folder/clickhouse-client_*.deb @@ -43,6 +48,9 @@ source /utils.lib /usr/share/clickhouse-test/config/install.sh ./setup_minio.sh stateless +./mc admin trace clickminio > /test_output/minio.log & +MC_ADMIN_PID=$! + ./setup_hdfs_minicluster.sh config_logs_export_cluster /etc/clickhouse-server/config.d/system_logs_export.yaml @@ -55,12 +63,6 @@ if [[ -n "$BUGFIX_VALIDATE_CHECK" ]] && [[ "$BUGFIX_VALIDATE_CHECK" -eq 1 ]]; th rm /etc/clickhouse-server/users.d/s3_cache_new.xml rm /etc/clickhouse-server/config.d/zero_copy_destructive_operations.xml - #todo: remove these after 24.3 released. - sudo sed -i "s|azure<|azure_blob_storage<|" /etc/clickhouse-server/config.d/azure_storage_conf.xml - - #todo: remove these after 24.3 released. - sudo sed -i "s|local<|local_blob_storage<|" /etc/clickhouse-server/config.d/storage_conf.xml - function remove_keeper_config() { sudo sed -i "/<$1>$2<\/$1>/d" /etc/clickhouse-server/config.d/keeper_port.xml @@ -96,7 +98,7 @@ fi # simplest way to forward env variables to server sudo -E -u clickhouse /usr/bin/clickhouse-server --config /etc/clickhouse-server/config.xml --daemon --pid-file /var/run/clickhouse-server/clickhouse-server.pid -if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then +if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then sudo sed -i "s|/var/lib/clickhouse/filesystem_caches/|/var/lib/clickhouse/filesystem_caches_1/|" /etc/clickhouse-server1/config.d/filesystem_caches_path.xml sudo sed -i "s|/var/lib/clickhouse/filesystem_caches/|/var/lib/clickhouse/filesystem_caches_2/|" /etc/clickhouse-server2/config.d/filesystem_caches_path.xml @@ -128,12 +130,9 @@ if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]] --keeper_server.tcp_port 29181 --keeper_server.server_id 3 \ --prometheus.port 29988 \ --macros.shard s2 # It doesn't work :( - - MAX_RUN_TIME=$((MAX_RUN_TIME < 9000 ? MAX_RUN_TIME : 9000)) # min(MAX_RUN_TIME, 2.5 hours) - MAX_RUN_TIME=$((MAX_RUN_TIME != 0 ? MAX_RUN_TIME : 9000)) # set to 2.5 hours if 0 (unlimited) fi -if [[ -n "$USE_SHARED_CATALOG" ]] && [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then +if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then sudo cat /etc/clickhouse-server1/config.d/filesystem_caches_path.xml \ | sed "s|/var/lib/clickhouse/filesystem_caches/|/var/lib/clickhouse/filesystem_caches_1/|" \ > /etc/clickhouse-server1/config.d/filesystem_caches_path.xml.tmp @@ -155,9 +154,6 @@ if [[ -n "$USE_SHARED_CATALOG" ]] && [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then --keeper_server.tcp_port 19181 --keeper_server.server_id 2 \ --prometheus.port 19988 \ --macros.replica r2 # It doesn't work :( - - MAX_RUN_TIME=$((MAX_RUN_TIME < 9000 ? MAX_RUN_TIME : 9000)) # min(MAX_RUN_TIME, 2.5 hours) - MAX_RUN_TIME=$((MAX_RUN_TIME != 0 ? MAX_RUN_TIME : 9000)) # set to 2.5 hours if 0 (unlimited) fi # Wait for the server to start, but not for too long. @@ -168,7 +164,6 @@ do done setup_logs_replication - attach_gdb_to_clickhouse || true # FIXME: to not break old builds, clean on 2023-09-01 function fn_exists() { @@ -207,21 +202,21 @@ function run_tests() if [[ -n "$USE_AZURE_STORAGE_FOR_MERGE_TREE" ]] && [[ "$USE_AZURE_STORAGE_FOR_MERGE_TREE" -eq 1 ]]; then # to disable the same tests - ADDITIONAL_OPTIONS+=('--s3-storage') + ADDITIONAL_OPTIONS+=('--azure-blob-storage') # azurite is slow, but with these two settings it can be super slow ADDITIONAL_OPTIONS+=('--no-random-settings') ADDITIONAL_OPTIONS+=('--no-random-merge-tree-settings') fi - if [[ -n "$USE_SHARED_CATALOG" ]] && [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then + if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then ADDITIONAL_OPTIONS+=('--shared-catalog') fi - if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then + if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then ADDITIONAL_OPTIONS+=('--replicated-database') # Too many tests fail for DatabaseReplicated in parallel. ADDITIONAL_OPTIONS+=('--jobs') - ADDITIONAL_OPTIONS+=('2') + ADDITIONAL_OPTIONS+=('3') elif [[ 1 == $(clickhouse-client --query "SELECT value LIKE '%SANITIZE_COVERAGE%' FROM system.build_options WHERE name = 'CXX_FLAGS'") ]]; then # Coverage on a per-test basis could only be collected sequentially. # Do not set the --jobs parameter. @@ -252,24 +247,37 @@ function run_tests() try_run_with_retry 10 clickhouse-client -q "insert into system.zookeeper (name, path, value) values ('auxiliary_zookeeper2', '/test/chroot/', '')" + TIMEOUT=$((MAX_RUN_TIME - 800 > 8400 ? 8400 : MAX_RUN_TIME - 800)) + START_TIME=${SECONDS} set +e - clickhouse-test --testname --shard --zookeeper --check-zookeeper-session --hung-check --print-time \ - --no-drop-if-fail --test-runs "$NUM_TRIES" "${ADDITIONAL_OPTIONS[@]}" 2>&1 \ + timeout --preserve-status --signal TERM --kill-after 60m ${TIMEOUT}s \ + clickhouse-test --testname --shard --zookeeper --check-zookeeper-session --hung-check --print-time \ + --no-drop-if-fail --test-runs "$NUM_TRIES" "${ADDITIONAL_OPTIONS[@]}" 2>&1 \ | ts '%Y-%m-%d %H:%M:%S' \ | tee -a test_output/test_result.txt set -e + DURATION=$((START_TIME - SECONDS)) + + echo "Elapsed ${DURATION} seconds." + if [[ $DURATION -ge $TIMEOUT ]] + then + echo "It looks like the command is terminated by the timeout, which is ${TIMEOUT} seconds." + fi } export -f run_tests + +# This should be enough to setup job and collect artifacts +TIMEOUT=$((MAX_RUN_TIME - 700)) if [ "$NUM_TRIES" -gt "1" ]; then # We don't run tests with Ordinary database in PRs, only in master. # So run new/changed tests with Ordinary at least once in flaky check. - timeout_with_logging "$MAX_RUN_TIME" bash -c 'NUM_TRIES=1; USE_DATABASE_ORDINARY=1; run_tests' \ - | sed 's/All tests have finished//' | sed 's/No tests were run//' ||: + timeout_with_logging "$TIMEOUT" bash -c 'NUM_TRIES=1; USE_DATABASE_ORDINARY=1; run_tests' \ + | sed 's/All tests have finished/Redacted: a message about tests finish is deleted/' | sed 's/No tests were run/Redacted: a message about no tests run is deleted/' ||: fi -timeout_with_logging "$MAX_RUN_TIME" bash -c run_tests ||: +timeout_with_logging "$TIMEOUT" bash -c run_tests ||: echo "Files in current directory" ls -la ./ @@ -290,7 +298,7 @@ do err=$(clickhouse-client -q "select * from system.$table into outfile '/test_output/$table.tsv.gz' format TSVWithNamesAndTypes") echo "$err" [[ "0" != "${#err}" ]] && failed_to_save_logs=1 - if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then + if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then err=$( { clickhouse-client --port 19000 -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst; } 2>&1 ) echo "$err" [[ "0" != "${#err}" ]] && failed_to_save_logs=1 @@ -299,7 +307,7 @@ do [[ "0" != "${#err}" ]] && failed_to_save_logs=1 fi - if [[ -n "$USE_SHARED_CATALOG" ]] && [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then + if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then err=$( { clickhouse-client --port 19000 -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst; } 2>&1 ) echo "$err" [[ "0" != "${#err}" ]] && failed_to_save_logs=1 @@ -310,15 +318,20 @@ done # Why do we read data with clickhouse-local? # Because it's the simplest way to read it when server has crashed. sudo clickhouse stop ||: -if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then + + +if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then sudo clickhouse stop --pid-path /var/run/clickhouse-server1 ||: sudo clickhouse stop --pid-path /var/run/clickhouse-server2 ||: fi -if [[ -n "$USE_SHARED_CATALOG" ]] && [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then +if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then sudo clickhouse stop --pid-path /var/run/clickhouse-server1 ||: fi +# Kill minio admin client to stop collecting logs +kill $MC_ADMIN_PID + rg -Fa "" /var/log/clickhouse-server/clickhouse-server.log ||: rg -A50 -Fa "============" /var/log/clickhouse-server/stderr.log ||: zstd --threads=0 < /var/log/clickhouse-server/clickhouse-server.log > /test_output/clickhouse-server.log.zst & @@ -342,12 +355,13 @@ if [ $failed_to_save_logs -ne 0 ]; then for table in query_log zookeeper_log trace_log transactions_info_log metric_log blob_storage_log error_log do clickhouse-local "$data_path_config" --only-system-tables --stacktrace -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.tsv.zst ||: - if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then + + if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then clickhouse-local --path /var/lib/clickhouse1/ --only-system-tables --stacktrace -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst ||: clickhouse-local --path /var/lib/clickhouse2/ --only-system-tables --stacktrace -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.2.tsv.zst ||: fi - if [[ -n "$USE_SHARED_CATALOG" ]] && [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then + if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then clickhouse-local --path /var/lib/clickhouse1/ --only-system-tables --stacktrace -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst ||: fi done @@ -383,7 +397,8 @@ rm -rf /var/lib/clickhouse/data/system/*/ tar -chf /test_output/store.tar /var/lib/clickhouse/store ||: tar -chf /test_output/metadata.tar /var/lib/clickhouse/metadata/*.sql ||: -if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then + +if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then rg -Fa "" /var/log/clickhouse-server/clickhouse-server1.log ||: rg -Fa "" /var/log/clickhouse-server/clickhouse-server2.log ||: zstd --threads=0 < /var/log/clickhouse-server/clickhouse-server1.log > /test_output/clickhouse-server1.log.zst ||: @@ -394,9 +409,11 @@ if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]] tar -chf /test_output/coordination2.tar /var/lib/clickhouse2/coordination ||: fi -if [[ -n "$USE_SHARED_CATALOG" ]] && [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then +if [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then rg -Fa "" /var/log/clickhouse-server/clickhouse-server1.log ||: zstd --threads=0 < /var/log/clickhouse-server/clickhouse-server1.log > /test_output/clickhouse-server1.log.zst ||: mv /var/log/clickhouse-server/stderr1.log /test_output/ ||: tar -chf /test_output/coordination1.tar /var/lib/clickhouse1/coordination ||: fi + +collect_core_dumps diff --git a/docker/test/stateless/setup_hdfs_minicluster.sh b/docker/test/stateless/setup_hdfs_minicluster.sh index 6671e73562a..15a54f59096 100755 --- a/docker/test/stateless/setup_hdfs_minicluster.sh +++ b/docker/test/stateless/setup_hdfs_minicluster.sh @@ -10,7 +10,7 @@ cd hadoop-3.3.1 export JAVA_HOME=/usr mkdir -p target/test/data chown clickhouse ./target/test/data -sudo -E -u clickhouse bin/mapred minicluster -format -nomr -nnport 12222 >> /test_output/garbage.log 2>&1 & +sudo -E -u clickhouse bin/mapred minicluster -format -nomr -nnport 12222 >> /test_output/hdfs_minicluster.log 2>&1 & while ! nc -z localhost 12222; do sleep 1 diff --git a/docker/test/stateless/stress_tests.lib b/docker/test/stateless/stress_tests.lib index c069ccbdd8d..682da1df837 100644 --- a/docker/test/stateless/stress_tests.lib +++ b/docker/test/stateless/stress_tests.lib @@ -1,8 +1,5 @@ #!/bin/bash -# core.COMM.PID-TID -sysctl kernel.core_pattern='core.%e.%p-%P' - OK="\tOK\t\\N\t" FAIL="\tFAIL\t\\N\t" @@ -315,12 +312,4 @@ function collect_query_and_trace_logs() done } -function collect_core_dumps() -{ - find . -type f -maxdepth 1 -name 'core.*' | while read -r core; do - zstd --threads=0 "$core" - mv "$core.zst" /test_output/ - done -} - # vi: ft=bash diff --git a/docker/test/stateless/utils.lib b/docker/test/stateless/utils.lib index 9b6ab535a90..cb257536c36 100644 --- a/docker/test/stateless/utils.lib +++ b/docker/test/stateless/utils.lib @@ -1,5 +1,10 @@ #!/bin/bash +# core.COMM.PID-TID +sysctl kernel.core_pattern='core.%e.%p-%P' +# ASAN doesn't work with suid_dumpable=2 +sysctl fs.suid_dumpable=1 + function run_with_retry() { if [[ $- =~ e ]]; then @@ -38,14 +43,25 @@ function fn_exists() { function timeout_with_logging() { local exit_code=0 - timeout "${@}" || exit_code="${?}" + timeout -s TERM --preserve-status "${@}" || exit_code="${?}" + echo "Checking if it is a timeout. The code 124 will indicate a timeout." if [[ "${exit_code}" -eq "124" ]] then - echo "The command 'timeout ${*}' has been killed by timeout" + echo "The command 'timeout ${*}' has been killed by timeout." + else + echo "No, it isn't a timeout." fi return $exit_code } +function collect_core_dumps() +{ + find . -type f -maxdepth 1 -name 'core.*' | while read -r core; do + zstd --threads=0 "$core" + mv "$core.zst" /test_output/ + done +} + # vi: ft=bash diff --git a/docker/test/stress/run.sh b/docker/test/stress/run.sh index 6d121ba4142..b21114e456f 100644 --- a/docker/test/stress/run.sh +++ b/docker/test/stress/run.sh @@ -21,6 +21,9 @@ source /attach_gdb.lib # shellcheck source=../stateless/stress_tests.lib source /stress_tests.lib +# shellcheck disable=SC1091 +source /utils.lib + install_packages package_folder # Thread Fuzzer allows to check more permutations of possible thread scheduling @@ -110,6 +113,15 @@ start_server clickhouse-client --query "SHOW TABLES FROM datasets" clickhouse-client --query "SHOW TABLES FROM test" +if [[ "$USE_S3_STORAGE_FOR_MERGE_TREE" == "1" ]]; then + TEMP_POLICY="s3_cache" +elif [[ "$USE_AZURE_STORAGE_FOR_MERGE_TREE" == "1" ]]; then + TEMP_POLICY="azure_cache" +else + TEMP_POLICY="default" +fi + + clickhouse-client --query "CREATE TABLE test.hits_s3 (WatchID UInt64, JavaEnable UInt8, Title String, GoodEvent Int16, EventTime DateTime, EventDate Date, CounterID UInt32, ClientIP UInt32, ClientIP6 FixedString(16), RegionID UInt32, UserID UInt64, CounterClass Int8, OS UInt8, UserAgent UInt8, URL String, Referer String, URLDomain String, RefererDomain String, @@ -135,7 +147,7 @@ clickhouse-client --query "CREATE TABLE test.hits_s3 (WatchID UInt64, JavaEnabl URLHash UInt64, CLID UInt32, YCLID UInt64, ShareService String, ShareURL String, ShareTitle String, ParsedParams Nested(Key1 String, Key2 String, Key3 String, Key4 String, Key5 String, ValueDouble Float64), IslandID FixedString(16), RequestNum UInt32, RequestTry UInt8) ENGINE = MergeTree() PARTITION BY toYYYYMM(EventDate) - ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='s3_cache'" + ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='$TEMP_POLICY'" clickhouse-client --query "CREATE TABLE test.hits (WatchID UInt64, JavaEnable UInt8, Title String, GoodEvent Int16, EventTime DateTime, EventDate Date, CounterID UInt32, ClientIP UInt32, ClientIP6 FixedString(16), RegionID UInt32, UserID UInt64, CounterClass Int8, OS UInt8, UserAgent UInt8, URL String, Referer String, URLDomain String, @@ -161,7 +173,7 @@ clickhouse-client --query "CREATE TABLE test.hits (WatchID UInt64, JavaEnable U URLHash UInt64, CLID UInt32, YCLID UInt64, ShareService String, ShareURL String, ShareTitle String, ParsedParams Nested(Key1 String, Key2 String, Key3 String, Key4 String, Key5 String, ValueDouble Float64), IslandID FixedString(16), RequestNum UInt32, RequestTry UInt8) ENGINE = MergeTree() PARTITION BY toYYYYMM(EventDate) - ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='s3_cache'" + ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='$TEMP_POLICY'" clickhouse-client --query "CREATE TABLE test.visits (CounterID UInt32, StartDate Date, Sign Int8, IsNew UInt8, VisitID UInt64, UserID UInt64, StartTime DateTime, Duration UInt32, UTCStartTime DateTime, PageViews Int32, Hits Int32, IsBounce UInt8, Referer String, StartURL String, RefererDomain String, StartURLDomain String, @@ -195,11 +207,11 @@ clickhouse-client --query "CREATE TABLE test.visits (CounterID UInt32, StartDat Market Nested(Type UInt8, GoalID UInt32, OrderID String, OrderPrice Int64, PP UInt32, DirectPlaceID UInt32, DirectOrderID UInt32, DirectBannerID UInt32, GoodID String, GoodName String, GoodQuantity Int32, GoodPrice Int64), IslandID FixedString(16)) ENGINE = CollapsingMergeTree(Sign) PARTITION BY toYYYYMM(StartDate) ORDER BY (CounterID, StartDate, intHash32(UserID), VisitID) - SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='s3_cache'" + SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='$TEMP_POLICY'" -clickhouse-client --query "INSERT INTO test.hits_s3 SELECT * FROM datasets.hits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0" -clickhouse-client --query "INSERT INTO test.hits SELECT * FROM datasets.hits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0" -clickhouse-client --query "INSERT INTO test.visits SELECT * FROM datasets.visits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0" +clickhouse-client --query "INSERT INTO test.hits_s3 SELECT * FROM datasets.hits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16" +clickhouse-client --query "INSERT INTO test.hits SELECT * FROM datasets.hits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16" +clickhouse-client --query "INSERT INTO test.visits SELECT * FROM datasets.visits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16" clickhouse-client --query "DROP TABLE datasets.visits_v1 SYNC" clickhouse-client --query "DROP TABLE datasets.hits_v1 SYNC" @@ -211,19 +223,29 @@ clickhouse-client --query "SYSTEM STOP THREAD FUZZER" stop_server # Let's enable S3 storage by default -export USE_S3_STORAGE_FOR_MERGE_TREE=1 export RANDOMIZE_OBJECT_KEY_TYPE=1 export ZOOKEEPER_FAULT_INJECTION=1 export THREAD_POOL_FAULT_INJECTION=1 configure -# But we still need default disk because some tables loaded only into it -sudo cat /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml \ - | sed "s|
s3
|
s3
default|" \ - > /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml.tmp -mv /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml.tmp /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml -sudo chown clickhouse /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml -sudo chgrp clickhouse /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml +if [[ "$USE_S3_STORAGE_FOR_MERGE_TREE" == "1" ]]; then + # But we still need default disk because some tables loaded only into it + sudo cat /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml \ + | sed "s|
s3
|
s3
default|" \ + > /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml.tmp + mv /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml.tmp /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml + sudo chown clickhouse /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml + sudo chgrp clickhouse /etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml +elif [[ "$USE_AZURE_STORAGE_FOR_MERGE_TREE" == "1" ]]; then + # But we still need default disk because some tables loaded only into it + sudo cat /etc/clickhouse-server/config.d/azure_storage_policy_by_default.xml \ + | sed "s|
azure
|
azure
default|" \ + > /etc/clickhouse-server/config.d/azure_storage_policy_by_default.xml.tmp + mv /etc/clickhouse-server/config.d/azure_storage_policy_by_default.xml.tmp /etc/clickhouse-server/config.d/azure_storage_policy_by_default.xml + sudo chown clickhouse /etc/clickhouse-server/config.d/azure_storage_policy_by_default.xml + sudo chgrp clickhouse /etc/clickhouse-server/config.d/azure_storage_policy_by_default.xml +fi + sudo cat /etc/clickhouse-server/config.d/logger_trace.xml \ | sed "s|trace|test|" \ diff --git a/docker/test/style/Dockerfile b/docker/test/style/Dockerfile index 7cd712b73f6..cdc1d1fa095 100644 --- a/docker/test/style/Dockerfile +++ b/docker/test/style/Dockerfile @@ -23,22 +23,8 @@ RUN apt-get update && env DEBIAN_FRONTEND=noninteractive apt-get install --yes \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* # python-magic is the same version as in Ubuntu 22.04 -RUN pip3 install \ - PyGithub \ - black==23.12.0 \ - boto3 \ - codespell==2.2.1 \ - mypy==1.8.0 \ - pylint==3.1.0 \ - python-magic==0.4.24 \ - flake8==4.0.1 \ - requests \ - thefuzz \ - tqdm==4.66.4 \ - types-requests \ - unidiff \ - jwt \ - && rm -rf /root/.cache/pip +COPY requirements.txt / +RUN pip3 install --no-cache-dir -r requirements.txt RUN echo "en_US.UTF-8 UTF-8" > /etc/locale.gen && locale-gen en_US.UTF-8 ENV LC_ALL en_US.UTF-8 diff --git a/docker/test/style/requirements.txt b/docker/test/style/requirements.txt new file mode 100644 index 00000000000..ed73d0d3636 --- /dev/null +++ b/docker/test/style/requirements.txt @@ -0,0 +1,58 @@ +aiohttp==3.9.5 +aiosignal==1.3.1 +astroid==3.1.0 +async-timeout==4.0.3 +attrs==23.2.0 +black==24.4.2 +boto3==1.34.131 +botocore==1.34.131 +certifi==2024.6.2 +cffi==1.16.0 +charset-normalizer==3.3.2 +click==8.1.7 +codespell==2.2.1 +cryptography==42.0.8 +Deprecated==1.2.14 +dill==0.3.8 +flake8==4.0.1 +frozenlist==1.4.1 +idna==3.7 +isort==5.13.2 +jmespath==1.0.1 +jwt==1.3.1 +mccabe==0.6.1 +multidict==6.0.5 +mypy==1.8.0 +mypy-extensions==1.0.0 +packaging==24.1 +pathspec==0.9.0 +pip==24.1.1 +pipdeptree==2.23.0 +platformdirs==4.2.2 +pycodestyle==2.8.0 +pycparser==2.22 +pyflakes==2.4.0 +PyGithub==2.3.0 +PyJWT==2.8.0 +pylint==3.1.0 +PyNaCl==1.5.0 +python-dateutil==2.9.0.post0 +python-magic==0.4.24 +PyYAML==6.0.1 +rapidfuzz==3.9.3 +requests==2.32.3 +s3transfer==0.10.1 +setuptools==59.6.0 +six==1.16.0 +thefuzz==0.22.1 +tomli==2.0.1 +tomlkit==0.12.5 +tqdm==4.66.4 +types-requests==2.32.0.20240622 +typing_extensions==4.12.2 +unidiff==0.7.5 +urllib3==2.2.2 +wheel==0.37.1 +wrapt==1.16.0 +yamllint==1.26.3 +yarl==1.9.4 diff --git a/docker/test/util/Dockerfile b/docker/test/util/Dockerfile index f13bb576f79..babddbd228c 100644 --- a/docker/test/util/Dockerfile +++ b/docker/test/util/Dockerfile @@ -17,6 +17,7 @@ ENV DEBIAN_FRONTEND=noninteractive LLVM_VERSION=18 RUN apt-get update \ && apt-get install \ + sudo \ apt-transport-https \ apt-utils \ ca-certificates \ diff --git a/docker/test/util/process_functional_tests_result.py b/docker/test/util/process_functional_tests_result.py index fd4cc9f4bf7..3da1a8f3674 100755 --- a/docker/test/util/process_functional_tests_result.py +++ b/docker/test/util/process_functional_tests_result.py @@ -11,6 +11,8 @@ TIMEOUT_SIGN = "[ Timeout! " UNKNOWN_SIGN = "[ UNKNOWN " SKIPPED_SIGN = "[ SKIPPED " HUNG_SIGN = "Found hung queries in processlist" +SERVER_DIED_SIGN = "Server died, terminating all processes" +SERVER_DIED_SIGN2 = "Server does not respond to health check" DATABASE_SIGN = "Database: " SUCCESS_FINISH_SIGNS = ["All tests have finished", "No tests were run"] @@ -25,6 +27,7 @@ def process_test_log(log_path, broken_tests): failed = 0 success = 0 hung = False + server_died = False retries = False success_finish = False test_results = [] @@ -41,6 +44,8 @@ def process_test_log(log_path, broken_tests): if HUNG_SIGN in line: hung = True break + if SERVER_DIED_SIGN in line or SERVER_DIED_SIGN2 in line: + server_died = True if RETRIES_SIGN in line: retries = True if any( @@ -107,12 +112,12 @@ def process_test_log(log_path, broken_tests): # Python does not support TSV, so we have to escape '\t' and '\n' manually # and hope that complex escape sequences will not break anything test_results = [ - ( + [ test[0], test[1], test[2], "".join(test[3])[:4096].replace("\t", "\\t").replace("\n", "\\n"), - ) + ] for test in test_results ] @@ -123,6 +128,7 @@ def process_test_log(log_path, broken_tests): failed, success, hung, + server_died, success_finish, retries, test_results, @@ -150,6 +156,7 @@ def process_result(result_path, broken_tests): failed, success, hung, + server_died, success_finish, retries, test_results, @@ -164,14 +171,23 @@ def process_result(result_path, broken_tests): if hung: description = "Some queries hung, " state = "failure" - test_results.append(("Some queries hung", "FAIL", "0", "")) + test_results.append(["Some queries hung", "FAIL", "0", ""]) + elif server_died: + description = "Server died, " + state = "failure" + # When ClickHouse server crashes, some tests are still running + # and fail because they cannot connect to server + for result in test_results: + if result[1] == "FAIL": + result[1] = "SERVER_DIED" + test_results.append(["Server died", "FAIL", "0", ""]) elif not success_finish: description = "Tests are not finished, " state = "failure" - test_results.append(("Tests are not finished", "FAIL", "0", "")) + test_results.append(["Tests are not finished", "FAIL", "0", ""]) elif retries: description = "Some tests restarted, " - test_results.append(("Some tests restarted", "SKIPPED", "0", "")) + test_results.append(["Some tests restarted", "SKIPPED", "0", ""]) else: description = "" @@ -218,5 +234,21 @@ if __name__ == "__main__": state, description, test_results = process_result(args.in_results_dir, broken_tests) logging.info("Result parsed") status = (state, description) + + def test_result_comparator(item): + # sort by status then by check name + order = { + "FAIL": 0, + "SERVER_DIED": 1, + "Timeout": 2, + "NOT_FAILED": 3, + "BROKEN": 4, + "OK": 5, + "SKIPPED": 6, + } + return order.get(item[1], 10), str(item[0]), item[1] + + test_results.sort(key=test_result_comparator) + write_results(args.out_results_file, args.out_status_file, test_results, status) logging.info("Result written") diff --git a/docs/changelogs/v24.5.4.49-stable.md b/docs/changelogs/v24.5.4.49-stable.md new file mode 100644 index 00000000000..56d497d5f59 --- /dev/null +++ b/docs/changelogs/v24.5.4.49-stable.md @@ -0,0 +1,41 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.5.4.49-stable (63b760955a0) FIXME as compared to v24.5.3.5-stable (e0eb66f8e17) + +#### Improvement +* Backported in [#65886](https://github.com/ClickHouse/ClickHouse/issues/65886): Always start Keeper with sufficient amount of threads in global thread pool. [#64444](https://github.com/ClickHouse/ClickHouse/pull/64444) ([Duc Canh Le](https://github.com/canhld94)). +* Backported in [#65304](https://github.com/ClickHouse/ClickHouse/issues/65304): Returned back the behaviour of how ClickHouse works and interprets Tuples in CSV format. This change effectively reverts https://github.com/ClickHouse/ClickHouse/pull/60994 and makes it available only under a few settings: `output_format_csv_serialize_tuple_into_separate_columns`, `input_format_csv_deserialize_separate_columns_into_tuple` and `input_format_csv_try_infer_strings_from_quoted_tuples`. [#65170](https://github.com/ClickHouse/ClickHouse/pull/65170) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Backported in [#65896](https://github.com/ClickHouse/ClickHouse/issues/65896): Respect cgroup CPU limit in Keeper. [#65819](https://github.com/ClickHouse/ClickHouse/pull/65819) ([Antonio Andelic](https://github.com/antonio2368)). + +#### Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC) +* Backported in [#65287](https://github.com/ClickHouse/ClickHouse/issues/65287): Fix crash with UniqInjectiveFunctionsEliminationPass and uniqCombined. [#65188](https://github.com/ClickHouse/ClickHouse/pull/65188) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#65374](https://github.com/ClickHouse/ClickHouse/issues/65374): Fix a bug in ClickHouse Keeper that causes digest mismatch during closing session. [#65198](https://github.com/ClickHouse/ClickHouse/pull/65198) ([Aleksei Filatov](https://github.com/aalexfvk)). +* Backported in [#65437](https://github.com/ClickHouse/ClickHouse/issues/65437): Forbid `QUALIFY` clause in the old analyzer. The old analyzer ignored `QUALIFY`, so it could lead to unexpected data removal in mutations. [#65356](https://github.com/ClickHouse/ClickHouse/pull/65356) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#65450](https://github.com/ClickHouse/ClickHouse/issues/65450): Use correct memory alignment for Distinct combinator. Previously, crash could happen because of invalid memory allocation when the combinator was used. [#65379](https://github.com/ClickHouse/ClickHouse/pull/65379) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#65712](https://github.com/ClickHouse/ClickHouse/issues/65712): Fix crash in maxIntersections. [#65689](https://github.com/ClickHouse/ClickHouse/pull/65689) ([Raúl Marín](https://github.com/Algunenano)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#65681](https://github.com/ClickHouse/ClickHouse/issues/65681): Fix `duplicate alias` error for distributed queries with `ARRAY JOIN`. [#64226](https://github.com/ClickHouse/ClickHouse/pull/64226) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#65331](https://github.com/ClickHouse/ClickHouse/issues/65331): Fix the crash loop when restoring from backup is blocked by creating an MV with a definer that hasn't been restored yet. [#64595](https://github.com/ClickHouse/ClickHouse/pull/64595) ([pufit](https://github.com/pufit)). +* Backported in [#64835](https://github.com/ClickHouse/ClickHouse/issues/64835): Fix bug which could lead to non-working TTLs with expressions. [#64694](https://github.com/ClickHouse/ClickHouse/pull/64694) ([alesapin](https://github.com/alesapin)). +* Backported in [#65542](https://github.com/ClickHouse/ClickHouse/issues/65542): Fix crash for `ALTER TABLE ... ON CLUSTER ... MODIFY SQL SECURITY`. [#64957](https://github.com/ClickHouse/ClickHouse/pull/64957) ([pufit](https://github.com/pufit)). +* Backported in [#65580](https://github.com/ClickHouse/ClickHouse/issues/65580): Fix crash on destroying AccessControl: add explicit shutdown. [#64993](https://github.com/ClickHouse/ClickHouse/pull/64993) ([Vitaly Baranov](https://github.com/vitlibar)). +* Backported in [#65618](https://github.com/ClickHouse/ClickHouse/issues/65618): Fix possible infinite query duration in case of cyclic aliases. Fixes [#64849](https://github.com/ClickHouse/ClickHouse/issues/64849). [#65081](https://github.com/ClickHouse/ClickHouse/pull/65081) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#65617](https://github.com/ClickHouse/ClickHouse/issues/65617): Fix aggregate function name rewriting in the new analyzer. [#65110](https://github.com/ClickHouse/ClickHouse/pull/65110) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#65732](https://github.com/ClickHouse/ClickHouse/issues/65732): Eliminate injective function in argument of functions `uniq*` recursively. This used to work correctly but was broken in the new analyzer. [#65140](https://github.com/ClickHouse/ClickHouse/pull/65140) ([Duc Canh Le](https://github.com/canhld94)). +* Backported in [#65265](https://github.com/ClickHouse/ClickHouse/issues/65265): Fix the bug in Hashed and Hashed_Array dictionary short circuit evaluation, which may read uninitialized number, leading to various errors. [#65256](https://github.com/ClickHouse/ClickHouse/pull/65256) ([jsc0218](https://github.com/jsc0218)). +* Backported in [#65663](https://github.com/ClickHouse/ClickHouse/issues/65663): Disable `non-intersecting-parts` optimization for queries with `FINAL` in case of `read-in-order` optimization was enabled. This could lead to an incorrect query result. As a workaround, disable `do_not_merge_across_partitions_select_final` and `split_parts_ranges_into_intersecting_and_non_intersecting_final` before this fix is merged. [#65505](https://github.com/ClickHouse/ClickHouse/pull/65505) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#65788](https://github.com/ClickHouse/ClickHouse/issues/65788): Fixed bug in MergeJoin. Column in sparse serialisation might be treated as a column of its nested type though the required conversion wasn't performed. [#65632](https://github.com/ClickHouse/ClickHouse/pull/65632) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#65812](https://github.com/ClickHouse/ClickHouse/issues/65812): Fix invalid exceptions in function `parseDateTime` with `%F` and `%D` placeholders. [#65768](https://github.com/ClickHouse/ClickHouse/pull/65768) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#65828](https://github.com/ClickHouse/ClickHouse/issues/65828): Fix a bug in short circuit logic when old analyzer and dictGetOrDefault is used. [#65802](https://github.com/ClickHouse/ClickHouse/pull/65802) ([jsc0218](https://github.com/jsc0218)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#65412](https://github.com/ClickHouse/ClickHouse/issues/65412): Re-enable OpenSSL session caching. [#65111](https://github.com/ClickHouse/ClickHouse/pull/65111) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#65905](https://github.com/ClickHouse/ClickHouse/issues/65905): Fix bug with session closing in Keeper. [#65735](https://github.com/ClickHouse/ClickHouse/pull/65735) ([Antonio Andelic](https://github.com/antonio2368)). + diff --git a/docs/changelogs/v24.6.1.4423-stable.md b/docs/changelogs/v24.6.1.4423-stable.md new file mode 100644 index 00000000000..f7af9cbaf8d --- /dev/null +++ b/docs/changelogs/v24.6.1.4423-stable.md @@ -0,0 +1,735 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.6.1.4423-stable (dcced7c8478) FIXME as compared to v24.4.1.2088-stable (6d4b31322d1) + +#### Backward Incompatible Change +* Enable asynchronous load of databases and tables by default. See the `async_load_databases` in config.xml. While this change is fully compatible, it can introduce a difference in behavior. When `async_load_databases` is false, as in the previous versions, the server will not accept connections until all tables are loaded. When `async_load_databases` is true, as in the new version, the server can accept connections before all the tables are loaded. If a query is made to a table that is not yet loaded, it will wait for the table's loading, which can take considerable time. It can change the behavior of the server if it is part of a large distributed system under a load balancer. In the first case, the load balancer can get a connection refusal and quickly failover to another server. In the second case, the load balancer can connect to a server that is still loading the tables, and the query will have a higher latency. Moreover, if many queries accumulate in the waiting state, it can lead to a "thundering herd" problem when they start processing simultaneously. This can make a difference only for highly loaded distributed backends. You can set the value of `async_load_databases` to false to avoid this problem. [#57695](https://github.com/ClickHouse/ClickHouse/pull/57695) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Some invalid queries will fail earlier during parsing. Note: disabled the support for inline KQL expressions (the experimental Kusto language) when they are put into a `kql` table function without a string literal, e.g. `kql(garbage | trash)` instead of `kql('garbage | trash')` or `kql($$garbage | trash$$)`. This feature was introduced unintentionally and should not exist. [#61500](https://github.com/ClickHouse/ClickHouse/pull/61500) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Renamed "inverted indexes" to "full-text indexes" which is a less technical / more user-friendly name. This also changes internal table metadata and breaks tables with existing (experimental) inverted indexes. Please make to drop such indexes before upgrade and re-create them after upgrade. [#62884](https://github.com/ClickHouse/ClickHouse/pull/62884) ([Robert Schulze](https://github.com/rschu1ze)). +* Usage of functions `neighbor`, `runningAccumulate`, `runningDifferenceStartingWithFirstValue`, `runningDifference` deprecated (because it is error-prone). Proper window functions should be used instead. To enable them back, set `allow_deprecated_functions=1`. [#63132](https://github.com/ClickHouse/ClickHouse/pull/63132) ([Nikita Taranov](https://github.com/nickitat)). +* Queries from `system.columns` will work faster if there is a large number of columns, but many databases or tables are not granted for `SHOW TABLES`. Note that in previous versions, if you grant `SHOW COLUMNS` to individual columns without granting `SHOW TABLES` to the corresponding tables, the `system.columns` table will show these columns, but in a new version, it will skip the table entirely. Remove trace log messages "Access granted" and "Access denied" that slowed down queries. [#63439](https://github.com/ClickHouse/ClickHouse/pull/63439) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Rework parallel processing in `Ordered` mode of storage `S3Queue`. This PR is backward incompatible for Ordered mode if you used settings `s3queue_processing_threads_num` or `s3queue_total_shards_num`. Setting `s3queue_total_shards_num` is deleted, previously it was allowed to use only under `s3queue_allow_experimental_sharded_mode`, which is now deprecated. A new setting is added - `s3queue_buckets`. [#64349](https://github.com/ClickHouse/ClickHouse/pull/64349) ([Kseniia Sumarokova](https://github.com/kssenii)). +* New functions `snowflakeIDToDateTime`, `snowflakeIDToDateTime64`, `dateTimeToSnowflakeID`, and `dateTime64ToSnowflakeID` were added. Unlike the existing functions `snowflakeToDateTime`, `snowflakeToDateTime64`, `dateTimeToSnowflake`, and `dateTime64ToSnowflake`, the new functions are compatible with function `generateSnowflakeID`, i.e. they accept the snowflake IDs generated by `generateSnowflakeID` and produce snowflake IDs of the same type as `generateSnowflakeID` (i.e. `UInt64`). Furthermore, the new functions default to the UNIX epoch (aka. 1970-01-01), just like `generateSnowflakeID`. If necessary, a different epoch, e.g. Twitter's/X's epoch 2010-11-04 aka. 1288834974657 msec since UNIX epoch, can be passed. The old conversion functions are deprecated and will be removed after a transition period: to use them regardless, enable setting `allow_deprecated_snowflake_conversion_functions`. [#64948](https://github.com/ClickHouse/ClickHouse/pull/64948) ([Robert Schulze](https://github.com/rschu1ze)). + +#### New Feature +* Provide support for AzureBlobStorage function in ClickHouse server to use Azure Workload identity to authenticate against Azure blob storage. If `use_workload_identity` parameter is set in config, [workload identity](https://github.com/Azure/azure-sdk-for-cpp/tree/main/sdk/identity/azure-identity#authenticate-azure-hosted-applications) is used for authentication. [#57881](https://github.com/ClickHouse/ClickHouse/pull/57881) ([Vinay Suryadevara](https://github.com/vinay92-ch)). +* Introduce bulk loading to StorageEmbeddedRocksDB by creating and ingesting SST file instead of relying on rocksdb build-in memtable. This help to increase importing speed, especially for long-running insert query to StorageEmbeddedRocksDB tables. Also, introduce `StorageEmbeddedRocksDB` table settings. [#59163](https://github.com/ClickHouse/ClickHouse/pull/59163) ([Duc Canh Le](https://github.com/canhld94)). +* Introduce statistics of type "number of distinct values". [#59357](https://github.com/ClickHouse/ClickHouse/pull/59357) ([Han Fei](https://github.com/hanfei1991)). +* User can now parse CRLF with TSV format using a setting `input_format_tsv_crlf_end_of_line`. Closes [#56257](https://github.com/ClickHouse/ClickHouse/issues/56257). [#59747](https://github.com/ClickHouse/ClickHouse/pull/59747) ([Shaun Struwig](https://github.com/Blargian)). +* Add Hilbert Curve encode and decode functions. [#60156](https://github.com/ClickHouse/ClickHouse/pull/60156) ([Artem Mustafin](https://github.com/Artemmm91)). +* Adds the Form Format to read/write a single record in the application/x-www-form-urlencoded format. [#60199](https://github.com/ClickHouse/ClickHouse/pull/60199) ([Shaun Struwig](https://github.com/Blargian)). +* Added possibility to compress in CROSS JOIN. [#60459](https://github.com/ClickHouse/ClickHouse/pull/60459) ([p1rattttt](https://github.com/p1rattttt)). +* New setting `input_format_force_null_for_omitted_fields` that forces NULL values for omitted fields. [#60887](https://github.com/ClickHouse/ClickHouse/pull/60887) ([Constantine Peresypkin](https://github.com/pkit)). +* Support join with inequal conditions which involve columns from both left and right table. e.g. `t1.y < t2.y`. To enable, `SET allow_experimental_join_condition = 1`. [#60920](https://github.com/ClickHouse/ClickHouse/pull/60920) ([lgbo](https://github.com/lgbo-ustc)). +* Earlier our s3 storage and s3 table function didn't support selecting from archive files. I created a solution that allows to iterate over files inside archives in S3. [#62259](https://github.com/ClickHouse/ClickHouse/pull/62259) ([Daniil Ivanik](https://github.com/divanik)). +* Support for conditional function `clamp`. [#62377](https://github.com/ClickHouse/ClickHouse/pull/62377) ([skyoct](https://github.com/skyoct)). +* Add npy output format. [#62430](https://github.com/ClickHouse/ClickHouse/pull/62430) ([豪肥肥](https://github.com/HowePa)). +* Added support for reading LINESTRING geometry in WKT format using function `readWKTLineString`. [#62519](https://github.com/ClickHouse/ClickHouse/pull/62519) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Added SQL functions `generateUUIDv7`, `generateUUIDv7ThreadMonotonic`, `generateUUIDv7NonMonotonic` (with different monotonicity/performance trade-offs) to generate version 7 UUIDs aka. timestamp-based UUIDs with random component. Also added a new function `UUIDToNum` to extract bytes from a UUID and a new function `UUIDv7ToDateTime` to extract timestamp component from a UUID version 7. [#62852](https://github.com/ClickHouse/ClickHouse/pull/62852) ([Alexey Petrunyaka](https://github.com/pet74alex)). +* Implement Dynamic data type that allows to store values of any type inside it without knowing all of them in advance. Dynamic type is available under a setting `allow_experimental_dynamic_type`. [#63058](https://github.com/ClickHouse/ClickHouse/pull/63058) ([Kruglov Pavel](https://github.com/Avogar)). +* Allow to attach parts from a different disk. [#63087](https://github.com/ClickHouse/ClickHouse/pull/63087) ([Unalian](https://github.com/Unalian)). +* Allow proxy to be bypassed for hosts specified in `no_proxy` env variable and ClickHouse proxy configuration. [#63314](https://github.com/ClickHouse/ClickHouse/pull/63314) ([Arthur Passos](https://github.com/arthurpassos)). +* Introduce bulk loading to StorageEmbeddedRocksDB by creating and ingesting SST file instead of relying on rocksdb build-in memtable. This help to increase importing speed, especially for long-running insert query to StorageEmbeddedRocksDB tables. Also, introduce StorageEmbeddedRocksDB table settings. [#63324](https://github.com/ClickHouse/ClickHouse/pull/63324) ([Duc Canh Le](https://github.com/canhld94)). +* Raw as a synonym for TSVRaw. [#63394](https://github.com/ClickHouse/ClickHouse/pull/63394) ([Unalian](https://github.com/Unalian)). +* Added possibility to do cross join in temporary file if size exceeds limits. [#63432](https://github.com/ClickHouse/ClickHouse/pull/63432) ([p1rattttt](https://github.com/p1rattttt)). +* Added a new table function `loop` to support returning query results in an infinite loop. [#63452](https://github.com/ClickHouse/ClickHouse/pull/63452) ([Sariel](https://github.com/sarielwxm)). +* Added new SQL functions `generateSnowflakeID` for generating Twitter-style Snowflake IDs. [#63577](https://github.com/ClickHouse/ClickHouse/pull/63577) ([Danila Puzov](https://github.com/kazalika)). +* Add the ability to reshuffle rows during insert to optimize for size without violating the order set by `PRIMARY KEY`. It's controlled by the setting `optimize_row_order` (off by default). [#63578](https://github.com/ClickHouse/ClickHouse/pull/63578) ([Igor Markelov](https://github.com/ElderlyPassionFruit)). +* On Linux and MacOS, if the program has STDOUT redirected to a file with a compression extension, use the corresponding compression method instead of nothing (making it behave similarly to `INTO OUTFILE` ). [#63662](https://github.com/ClickHouse/ClickHouse/pull/63662) ([v01dXYZ](https://github.com/v01dXYZ)). +* Added `merge_workload` and `mutation_workload` settings to regulate how resources are utilized and shared between merges, mutations and other workloads. [#64061](https://github.com/ClickHouse/ClickHouse/pull/64061) ([Sergei Trifonov](https://github.com/serxa)). +* Change warning on high number of attached tables to differentiate tables, views and dictionaries. [#64180](https://github.com/ClickHouse/ClickHouse/pull/64180) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)). +* Add support for comparing IPv4 and IPv6 types using the `=` operator. [#64292](https://github.com/ClickHouse/ClickHouse/pull/64292) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)). +* Allow to store named collections in zookeeper. [#64574](https://github.com/ClickHouse/ClickHouse/pull/64574) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Support decimal arguments in binary math functions (pow(), atan2(), max2, min2(), hypot(). [#64582](https://github.com/ClickHouse/ClickHouse/pull/64582) ([Mikhail Gorshkov](https://github.com/mgorshkov)). +* Add support for index analysis over `hilbertEncode`. [#64662](https://github.com/ClickHouse/ClickHouse/pull/64662) ([Artem Mustafin](https://github.com/Artemmm91)). +* Added SQL functions `parseReadableSize` (along with `OrNull` and `OrZero` variants). [#64742](https://github.com/ClickHouse/ClickHouse/pull/64742) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)). +* Add server settings `max_table_num_to_throw` and `max_database_num_to_throw` to limit the number of databases or tables on `CREATE` queries. [#64781](https://github.com/ClickHouse/ClickHouse/pull/64781) ([Xu Jia](https://github.com/XuJia0210)). +* Add _time virtual column to file alike storages (s3/file/hdfs/url/azureBlobStorage). [#64947](https://github.com/ClickHouse/ClickHouse/pull/64947) ([Ilya Golshtein](https://github.com/ilejn)). +* Introduced new functions `base64URLEncode`, `base64URLDecode` and `tryBase64URLDecode`. [#64991](https://github.com/ClickHouse/ClickHouse/pull/64991) ([Mikhail Gorshkov](https://github.com/mgorshkov)). +* Add new function `editDistanceUTF8`, which calculates the [edit distance](https://en.wikipedia.org/wiki/Edit_distance) between two UTF8 strings. [#65269](https://github.com/ClickHouse/ClickHouse/pull/65269) ([LiuNeng](https://github.com/liuneng1994)). + +#### Performance Improvement +* Skip merging of newly created projection blocks during `INSERT`-s. [#59405](https://github.com/ClickHouse/ClickHouse/pull/59405) ([Nikita Taranov](https://github.com/nickitat)). +* Add a native parquet reader, which can read parquet binary to ClickHouse Columns directly. It's controlled by the setting `input_format_parquet_use_native_reader` (disabled by default). [#60361](https://github.com/ClickHouse/ClickHouse/pull/60361) ([ZhiHong Zhang](https://github.com/copperybean)). +* Reduce the number of virtual function calls in ColumnNullable::size(). [#60556](https://github.com/ClickHouse/ClickHouse/pull/60556) ([HappenLee](https://github.com/HappenLee)). +* Process string functions XXXUTF8 'asciily' if input strings are all ascii chars. Inspired by https://github.com/apache/doris/pull/29799. Overall speed up by 1.07x~1.62x. Notice that peak memory usage had been decreased in some cases. [#61632](https://github.com/ClickHouse/ClickHouse/pull/61632) ([李扬](https://github.com/taiyang-li)). +* Improved performance of selection (`{}`) globs in StorageS3. [#62120](https://github.com/ClickHouse/ClickHouse/pull/62120) ([Andrey Zvonov](https://github.com/zvonand)). +* HostResolver has each IP address several times. If remote host has several IPs and by some reason (firewall rules for example) access on some IPs allowed and on others forbidden, than only first record of forbidden IPs marked as failed, and in each try these IPs have a chance to be chosen (and failed again). Even if fix this, every 120 seconds DNS cache dropped, and IPs can be chosen again. [#62652](https://github.com/ClickHouse/ClickHouse/pull/62652) ([Anton Ivashkin](https://github.com/ianton-ru)). +* Speedup `splitByRegexp` when the regular expression argument is a single-character. [#62696](https://github.com/ClickHouse/ClickHouse/pull/62696) ([Robert Schulze](https://github.com/rschu1ze)). +* Speed up FixedHashTable by keeping track of the min and max keys used. This allows to reduce the number of cells that need to be verified. [#62746](https://github.com/ClickHouse/ClickHouse/pull/62746) ([Jiebin Sun](https://github.com/jiebinn)). +* Add a new configuration`prefer_merge_sort_block_bytes` to control the memory usage and speed up sorting 2 times when merging when there are many columns. [#62904](https://github.com/ClickHouse/ClickHouse/pull/62904) ([LiuNeng](https://github.com/liuneng1994)). +* `clickhouse-local` will start faster. In previous versions, it was not deleting temporary directories by mistake. Now it will. This closes [#62941](https://github.com/ClickHouse/ClickHouse/issues/62941). [#63074](https://github.com/ClickHouse/ClickHouse/pull/63074) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Micro-optimizations for the new analyzer. [#63429](https://github.com/ClickHouse/ClickHouse/pull/63429) ([Raúl Marín](https://github.com/Algunenano)). +* Index analysis will work if `DateTime` is compared to `DateTime64`. This closes [#63441](https://github.com/ClickHouse/ClickHouse/issues/63441). [#63443](https://github.com/ClickHouse/ClickHouse/pull/63443) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Index analysis will work if `DateTime` is compared to `DateTime64`. This closes [#63441](https://github.com/ClickHouse/ClickHouse/issues/63441). [#63532](https://github.com/ClickHouse/ClickHouse/pull/63532) ([Raúl Marín](https://github.com/Algunenano)). +* Optimize the resolution of in(LowCardinality, ConstantSet). [#64060](https://github.com/ClickHouse/ClickHouse/pull/64060) ([Zhiguo Zhou](https://github.com/ZhiguoZh)). +* Speed up indices of type `set` a little (around 1.5 times) by removing garbage. [#64098](https://github.com/ClickHouse/ClickHouse/pull/64098) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Use a thread pool to initialize and destroy hash tables inside `ConcurrentHashJoin`. [#64241](https://github.com/ClickHouse/ClickHouse/pull/64241) ([Nikita Taranov](https://github.com/nickitat)). +* Optimized vertical merges in tables with sparse columns. [#64311](https://github.com/ClickHouse/ClickHouse/pull/64311) ([Anton Popov](https://github.com/CurtizJ)). +* Enabled prefetches of data from remote filesystem during vertical merges. It improves latency of vertical merges in tables with data stored on remote filesystem. [#64314](https://github.com/ClickHouse/ClickHouse/pull/64314) ([Anton Popov](https://github.com/CurtizJ)). +* Reduce redundant calls to `isDefault()` of `ColumnSparse::filter` to improve performance. [#64426](https://github.com/ClickHouse/ClickHouse/pull/64426) ([Jiebin Sun](https://github.com/jiebinn)). +* Speedup `find_super_nodes` and `find_big_family` keeper-client commands by making multiple asynchronous getChildren requests. [#64628](https://github.com/ClickHouse/ClickHouse/pull/64628) ([Alexander Gololobov](https://github.com/davenger)). +* Improve function least/greatest for nullable numberic type arguments. [#64668](https://github.com/ClickHouse/ClickHouse/pull/64668) ([KevinyhZou](https://github.com/KevinyhZou)). +* Allow merging two consequent `FilterSteps` of a query plan. This improves filter-push-down optimization if the filter condition can be pushed down from the parent step. [#64760](https://github.com/ClickHouse/ClickHouse/pull/64760) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Remove bad optimization in vertical final implementation and re-enable vertical final algorithm by default. [#64783](https://github.com/ClickHouse/ClickHouse/pull/64783) ([Duc Canh Le](https://github.com/canhld94)). +* Remove ALIAS nodes from the filter expression. This slightly improves performance for queries with `PREWHERE` (with new analyzer). [#64793](https://github.com/ClickHouse/ClickHouse/pull/64793) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix performance regression in cross join introduced in [#60459](https://github.com/ClickHouse/ClickHouse/issues/60459) (24.5). [#65243](https://github.com/ClickHouse/ClickHouse/pull/65243) ([Nikita Taranov](https://github.com/nickitat)). + +#### Improvement +* Support empty tuples. [#55061](https://github.com/ClickHouse/ClickHouse/pull/55061) ([Amos Bird](https://github.com/amosbird)). +* Hot reload storage policy for distributed tables when adding a new disk. [#58285](https://github.com/ClickHouse/ClickHouse/pull/58285) ([Duc Canh Le](https://github.com/canhld94)). +* Maps can now have `Float32`, `Float64`, `Array(T)`, `Map(K,V)` and `Tuple(T1, T2, ...)` as keys. Closes [#54537](https://github.com/ClickHouse/ClickHouse/issues/54537). [#59318](https://github.com/ClickHouse/ClickHouse/pull/59318) ([李扬](https://github.com/taiyang-li)). +* Avoid possible deadlock during MergeTree index analysis when scheduling threads in a saturated service. [#59427](https://github.com/ClickHouse/ClickHouse/pull/59427) ([Sean Haynes](https://github.com/seandhaynes)). +* Multiline strings with border preservation and column width change. [#59940](https://github.com/ClickHouse/ClickHouse/pull/59940) ([Volodyachan](https://github.com/Volodyachan)). +* Make rabbitmq nack broken messages. Closes [#45350](https://github.com/ClickHouse/ClickHouse/issues/45350). [#60312](https://github.com/ClickHouse/ClickHouse/pull/60312) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Support partial trivial count optimization when the query filter is able to select exact ranges from merge tree tables. [#60463](https://github.com/ClickHouse/ClickHouse/pull/60463) ([Amos Bird](https://github.com/amosbird)). +* Fix a crash in asynchronous stack unwinding (such as when using the sampling query profiler) while interpreting debug info. This closes [#60460](https://github.com/ClickHouse/ClickHouse/issues/60460). [#60468](https://github.com/ClickHouse/ClickHouse/pull/60468) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Reduce max memory usage of multithreaded `INSERT`s by collecting chunks of multiple threads in a single transform. [#61047](https://github.com/ClickHouse/ClickHouse/pull/61047) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Distinct messages for s3 error 'no key' for cases disk and storage. [#61108](https://github.com/ClickHouse/ClickHouse/pull/61108) ([Sema Checherinda](https://github.com/CheSema)). +* Less contention in filesystem cache (part 4). Allow to keep filesystem cache not filled to the limit by doing additional eviction in the background (controlled by `keep_free_space_size(elements)_ratio`). This allows to release pressure from space reservation for queries (on `tryReserve` method). Also this is done in a lock free way as much as possible, e.g. should not block normal cache usage. [#61250](https://github.com/ClickHouse/ClickHouse/pull/61250) ([Kseniia Sumarokova](https://github.com/kssenii)). +* The progress bar will work for trivial queries with LIMIT from `system.zeros`, `system.zeros_mt` (it already works for `system.numbers` and `system.numbers_mt`), and the `generateRandom` table function. As a bonus, if the total number of records is greater than the `max_rows_to_read` limit, it will throw an exception earlier. This closes [#58183](https://github.com/ClickHouse/ClickHouse/issues/58183). [#61823](https://github.com/ClickHouse/ClickHouse/pull/61823) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* YAML Merge Key support. [#62685](https://github.com/ClickHouse/ClickHouse/pull/62685) ([Azat Khuzhin](https://github.com/azat)). +* Enhance error message when non-deterministic function is used with Replicated source. [#62896](https://github.com/ClickHouse/ClickHouse/pull/62896) ([Grégoire Pineau](https://github.com/lyrixx)). +* Fix interserver secret for Distributed over Distributed from `remote`. [#63013](https://github.com/ClickHouse/ClickHouse/pull/63013) ([Azat Khuzhin](https://github.com/azat)). +* Allow using `clickhouse-local` and its shortcuts `clickhouse` and `ch` with a query or queries file as a positional argument. Examples: `ch "SELECT 1"`, `ch --param_test Hello "SELECT {test:String}"`, `ch query.sql`. This closes [#62361](https://github.com/ClickHouse/ClickHouse/issues/62361). [#63081](https://github.com/ClickHouse/ClickHouse/pull/63081) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Support configuration substitutions from YAML files. [#63106](https://github.com/ClickHouse/ClickHouse/pull/63106) ([Eduard Karacharov](https://github.com/korowa)). +* Reduce the memory usage when using Azure object storage by using fixed memory allocation, avoiding the allocation of an extra buffer. [#63160](https://github.com/ClickHouse/ClickHouse/pull/63160) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Add TTL information in system parts_columns table. [#63200](https://github.com/ClickHouse/ClickHouse/pull/63200) ([litlig](https://github.com/litlig)). +* Keep previous data in terminal after picking from skim suggestions. [#63261](https://github.com/ClickHouse/ClickHouse/pull/63261) ([FlameFactory](https://github.com/FlameFactory)). +* Width of fields now correctly calculate, ignoring ANSI escape sequences. [#63270](https://github.com/ClickHouse/ClickHouse/pull/63270) ([Shaun Struwig](https://github.com/Blargian)). +* Enable plain_rewritable metadata for local and Azure (azure_blob_storage) object storages. [#63365](https://github.com/ClickHouse/ClickHouse/pull/63365) ([Julia Kartseva](https://github.com/jkartseva)). +* Support English-style Unicode quotes, e.g. “Hello”, ‘world’. This is questionable in general but helpful when you type your query in a word processor, such as Google Docs. This closes [#58634](https://github.com/ClickHouse/ClickHouse/issues/58634). [#63381](https://github.com/ClickHouse/ClickHouse/pull/63381) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Allowed to create MaterializedMySQL database without connection to MySQL. [#63397](https://github.com/ClickHouse/ClickHouse/pull/63397) ([Kirill](https://github.com/kirillgarbar)). +* Remove copying data when writing to filesystem cache. [#63401](https://github.com/ClickHouse/ClickHouse/pull/63401) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Update the usage of error code `NUMBER_OF_ARGUMENTS_DOESNT_MATCH` by more accurate error codes when appropriate. [#63406](https://github.com/ClickHouse/ClickHouse/pull/63406) ([Yohann Jardin](https://github.com/yohannj)). +* Several minor corner case fixes to proxy support & tunneling. [#63427](https://github.com/ClickHouse/ClickHouse/pull/63427) ([Arthur Passos](https://github.com/arthurpassos)). +* `os_user` and `client_hostname` are now correctly set up for queries for command line suggestions in clickhouse-client. This closes [#63430](https://github.com/ClickHouse/ClickHouse/issues/63430). [#63433](https://github.com/ClickHouse/ClickHouse/pull/63433) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fixed tabulation from line numbering, correct handling of length when moving a line if the value has a tab, added tests. [#63493](https://github.com/ClickHouse/ClickHouse/pull/63493) ([Volodyachan](https://github.com/Volodyachan)). +* Add this `aggregate_function_group_array_has_limit_size`setting to support discarding data in some scenarios. [#63516](https://github.com/ClickHouse/ClickHouse/pull/63516) ([zhongyuankai](https://github.com/zhongyuankai)). +* Automatically mark a replica of Replicated database as lost and start recovery if some DDL task fails more than `max_retries_before_automatic_recovery` (100 by default) times in a row with the same error. Also, fixed a bug that could cause skipping DDL entries when an exception is thrown during an early stage of entry execution. [#63549](https://github.com/ClickHouse/ClickHouse/pull/63549) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Add `http_response_headers` setting to support custom response headers in custom HTTP handlers. [#63562](https://github.com/ClickHouse/ClickHouse/pull/63562) ([Grigorii](https://github.com/GSokol)). +* Automatically correct `max_block_size=0` to default value. [#63587](https://github.com/ClickHouse/ClickHouse/pull/63587) ([Antonio Andelic](https://github.com/antonio2368)). +* Account failed files in `s3queue_tracked_file_ttl_sec` and `s3queue_traked_files_limit` for `StorageS3Queue`. [#63638](https://github.com/ClickHouse/ClickHouse/pull/63638) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Add a build_id ALIAS column to trace_log to facilitate auto renaming upon detecting binary changes. This is to address [#52086](https://github.com/ClickHouse/ClickHouse/issues/52086). [#63656](https://github.com/ClickHouse/ClickHouse/pull/63656) ([Zimu Li](https://github.com/woodlzm)). +* Enable truncate operation for object storage disks. [#63693](https://github.com/ClickHouse/ClickHouse/pull/63693) ([MikhailBurdukov](https://github.com/MikhailBurdukov)). +* Improve io_uring resubmits visibility. Rename profile event `IOUringSQEsResubmits` -> `IOUringSQEsResubmitsAsync` and add a new one `IOUringSQEsResubmitsSync`. [#63699](https://github.com/ClickHouse/ClickHouse/pull/63699) ([Tomer Shafir](https://github.com/tomershafir)). +* Introduce assertions to verify all functions are called with columns of the right size. [#63723](https://github.com/ClickHouse/ClickHouse/pull/63723) ([Raúl Marín](https://github.com/Algunenano)). +* The loading of the keywords list is now dependent on the server revision and will be disabled for the old versions of ClickHouse server. CC @azat. [#63786](https://github.com/ClickHouse/ClickHouse/pull/63786) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* `SHOW CREATE TABLE` executed on top of system tables will now show the super handy comment unique for each table which will explain why this table is needed. [#63788](https://github.com/ClickHouse/ClickHouse/pull/63788) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Allow trailing commas in the columns list in the INSERT query. For example, `INSERT INTO test (a, b, c, ) VALUES ...`. [#63803](https://github.com/ClickHouse/ClickHouse/pull/63803) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Better exception messages for the `Regexp` format. [#63804](https://github.com/ClickHouse/ClickHouse/pull/63804) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Allow trailing commas in the `Values` format. For example, this query is allowed: `INSERT INTO test (a, b, c) VALUES (4, 5, 6,);`. [#63810](https://github.com/ClickHouse/ClickHouse/pull/63810) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Clickhouse disks have to read server setting to obtain actual metadata format version. [#63831](https://github.com/ClickHouse/ClickHouse/pull/63831) ([Sema Checherinda](https://github.com/CheSema)). +* Disable pretty format restrictions (`output_format_pretty_max_rows`/`output_format_pretty_max_value_width`) when stdout is not TTY. [#63942](https://github.com/ClickHouse/ClickHouse/pull/63942) ([Azat Khuzhin](https://github.com/azat)). +* Exception handling now works when ClickHouse is used inside AWS Lambda. Author: [Alexey Coolnev](https://github.com/acoolnev). [#64014](https://github.com/ClickHouse/ClickHouse/pull/64014) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Throw `CANNOT_DECOMPRESS` instread of `CORRUPTED_DATA` on invalid compressed data passed via HTTP. [#64036](https://github.com/ClickHouse/ClickHouse/pull/64036) ([vdimir](https://github.com/vdimir)). +* A tip for a single large number in Pretty formats now works for Nullable and LowCardinality. This closes [#61993](https://github.com/ClickHouse/ClickHouse/issues/61993). [#64084](https://github.com/ClickHouse/ClickHouse/pull/64084) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Now backups with azure blob storage will use multicopy. [#64116](https://github.com/ClickHouse/ClickHouse/pull/64116) ([alesapin](https://github.com/alesapin)). +* Added a new setting, `metadata_keep_free_space_bytes` to keep free space on the metadata storage disk. [#64128](https://github.com/ClickHouse/ClickHouse/pull/64128) ([MikhailBurdukov](https://github.com/MikhailBurdukov)). +* Add metrics, logs, and thread names around parts filtering with indices. [#64130](https://github.com/ClickHouse/ClickHouse/pull/64130) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Allow to use native copy for azure even with different containers. [#64154](https://github.com/ClickHouse/ClickHouse/pull/64154) ([alesapin](https://github.com/alesapin)). +* Add metrics to track the number of directories created and removed by the plain_rewritable metadata storage, and the number of entries in the local-to-remote in-memory map. [#64175](https://github.com/ClickHouse/ClickHouse/pull/64175) ([Julia Kartseva](https://github.com/jkartseva)). +* Finally enable native copy for azure. [#64182](https://github.com/ClickHouse/ClickHouse/pull/64182) ([alesapin](https://github.com/alesapin)). +* Ignore `allow_suspicious_primary_key` on `ATTACH` and verify on `ALTER`. [#64202](https://github.com/ClickHouse/ClickHouse/pull/64202) ([Azat Khuzhin](https://github.com/azat)). +* The query cache now considers identical queries with different settings as different. This increases robustness in cases where different settings (e.g. `limit` or `additional_table_filters`) would affect the query result. [#64205](https://github.com/ClickHouse/ClickHouse/pull/64205) ([Robert Schulze](https://github.com/rschu1ze)). +* Better Exception Message in Delete Table with Projection, users can understand the error and the steps should be taken. [#64212](https://github.com/ClickHouse/ClickHouse/pull/64212) ([jsc0218](https://github.com/jsc0218)). +* Support the non standard error code `QpsLimitExceeded` in object storage as a retryable error. [#64225](https://github.com/ClickHouse/ClickHouse/pull/64225) ([Sema Checherinda](https://github.com/CheSema)). +* Forbid converting a MergeTree table to replicated if the zookeeper path for this table already exists. [#64244](https://github.com/ClickHouse/ClickHouse/pull/64244) ([Kirill](https://github.com/kirillgarbar)). +* If "replica group" is configured for a `Replicated` database, automatically create a cluster that includes replicas from all groups. [#64312](https://github.com/ClickHouse/ClickHouse/pull/64312) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Added settings to disable materialization of skip indexes and statistics on inserts (`materialize_skip_indexes_on_insert` and `materialize_statistics_on_insert`). [#64391](https://github.com/ClickHouse/ClickHouse/pull/64391) ([Anton Popov](https://github.com/CurtizJ)). +* Use the allocated memory size to calculate the row group size and reduce the peak memory of the parquet writer in single-threaded mode. [#64424](https://github.com/ClickHouse/ClickHouse/pull/64424) ([LiuNeng](https://github.com/liuneng1994)). +* Added new configuration input_format_parquet_prefer_block_bytes to control the average output block bytes, and modified the default value of input_format_parquet_max_block_size to 65409. [#64427](https://github.com/ClickHouse/ClickHouse/pull/64427) ([LiuNeng](https://github.com/liuneng1994)). +* Always start Keeper with sufficient amount of threads in global thread pool. [#64444](https://github.com/ClickHouse/ClickHouse/pull/64444) ([Duc Canh Le](https://github.com/canhld94)). +* Settings from user config doesn't affect merges and mutations for MergeTree on top of object storage. [#64456](https://github.com/ClickHouse/ClickHouse/pull/64456) ([alesapin](https://github.com/alesapin)). +* Setting `replace_long_file_name_to_hash` is enabled by default for `MergeTree` tables. [#64457](https://github.com/ClickHouse/ClickHouse/pull/64457) ([Anton Popov](https://github.com/CurtizJ)). +* Improve the iterator of sparse column to reduce call of size(). [#64497](https://github.com/ClickHouse/ClickHouse/pull/64497) ([Jiebin Sun](https://github.com/jiebinn)). +* Update condition to use copy for azure blob storage. [#64518](https://github.com/ClickHouse/ClickHouse/pull/64518) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Support the non standard error code `TotalQpsLimitExceeded` in object storage as a retryable error. [#64520](https://github.com/ClickHouse/ClickHouse/pull/64520) ([Sema Checherinda](https://github.com/CheSema)). +* Optimized memory usage of vertical merges for tables with high number of skip indexes. [#64580](https://github.com/ClickHouse/ClickHouse/pull/64580) ([Anton Popov](https://github.com/CurtizJ)). +* Introduced two additional columns in the `system.query_log`: `used_privileges` and `missing_privileges`. `used_privileges` is populated with the privileges that were checked during query execution, and `missing_privileges` contains required privileges that are missing. [#64597](https://github.com/ClickHouse/ClickHouse/pull/64597) ([Alexey Katsman](https://github.com/alexkats)). +* Add settings `parallel_replicas_custom_key_range_lower` and `parallel_replicas_custom_key_range_upper` to control how parallel replicas with dynamic shards parallelizes queries when using a range filter. [#64604](https://github.com/ClickHouse/ClickHouse/pull/64604) ([josh-hildred](https://github.com/josh-hildred)). +* Updated Advanced Dashboard for both open-source and ClickHouse Cloud versions to include a chart for 'Maximum concurrent network connections'. [#64610](https://github.com/ClickHouse/ClickHouse/pull/64610) ([Thom O'Connor](https://github.com/thomoco)). +* The second argument (scale) of functions `round()`, `roundBankers()`, `floor()`, `ceil()` and `trunc()` can now be non-const. [#64798](https://github.com/ClickHouse/ClickHouse/pull/64798) ([Mikhail Gorshkov](https://github.com/mgorshkov)). +* Improve progress report on zeros_mt and generateRandom. [#64804](https://github.com/ClickHouse/ClickHouse/pull/64804) ([Raúl Marín](https://github.com/Algunenano)). +* Add an asynchronous metric jemalloc.profile.active to show whether sampling is currently active. This is an activation mechanism in addition to prof.active; both must be active for the calling thread to sample. [#64842](https://github.com/ClickHouse/ClickHouse/pull/64842) ([Unalian](https://github.com/Unalian)). +* Support statistics with ReplicatedMergeTree. [#64934](https://github.com/ClickHouse/ClickHouse/pull/64934) ([Han Fei](https://github.com/hanfei1991)). +* Don't mark of `allow_experimental_join_condition` as IMPORTANT. This may have prevented distributed queries in a mixed versions cluster from being executed successfully. [#65008](https://github.com/ClickHouse/ClickHouse/pull/65008) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Backported in [#65716](https://github.com/ClickHouse/ClickHouse/issues/65716): `StorageS3Queue` related fixes and improvements. Deduce a default value of `s3queue_processing_threads_num` according to the number of physical cpu cores on the server (instead of the previous default value as 1). Set default value of `s3queue_loading_retries` to 10. Fix possible vague "Uncaught exception" in exception column of `system.s3queue`. Do not increment retry count on `MEMORY_LIMIT_EXCEEDED` exception. Move files commit to a stage after insertion into table fully finished to avoid files being commited while not inserted. Add settings `s3queue_max_processed_files_before_commit`, `s3queue_max_processed_rows_before_commit`, `s3queue_max_processed_bytes_before_commit`, `s3queue_max_processing_time_sec_before_commit`, to better control commit and flush time. [#65046](https://github.com/ClickHouse/ClickHouse/pull/65046) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Added server Asynchronous metrics `DiskGetObjectThrottler*` and `DiskGetObjectThrottler*` reflecting request per second rate limit defined with `s3_max_get_rps` and `s3_max_put_rps` disk settings and currently available number of requests that could be sent without hitting throttling limit on the disk. Metrics are defined for every disk that has a configured limit. [#65050](https://github.com/ClickHouse/ClickHouse/pull/65050) ([Sergei Trifonov](https://github.com/serxa)). +* Added a setting `output_format_pretty_display_footer_column_names` which when enabled displays column names at the end of the table for long tables (50 rows by default), with the threshold value for minimum number of rows controlled by `output_format_pretty_display_footer_column_names_min_rows`. [#65144](https://github.com/ClickHouse/ClickHouse/pull/65144) ([Shaun Struwig](https://github.com/Blargian)). +* Returned back the behaviour of how ClickHouse works and interprets Tuples in CSV format. This change effectively reverts https://github.com/ClickHouse/ClickHouse/pull/60994 and makes it available only under a few settings: `output_format_csv_serialize_tuple_into_separate_columns`, `input_format_csv_deserialize_separate_columns_into_tuple` and `input_format_csv_try_infer_strings_from_quoted_tuples`. [#65170](https://github.com/ClickHouse/ClickHouse/pull/65170) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Initialize global trace collector for Poco::ThreadPool (needed for keeper, etc). [#65239](https://github.com/ClickHouse/ClickHouse/pull/65239) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Add validation when creating a user with bcrypt_hash. [#65242](https://github.com/ClickHouse/ClickHouse/pull/65242) ([Raúl Marín](https://github.com/Algunenano)). + +#### Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC) +* Fix a permission error where a user in a specific situation can escalate their privileges on the default database without necessary grants. [#64769](https://github.com/ClickHouse/ClickHouse/pull/64769) ([pufit](https://github.com/pufit)). +* Fix crash with UniqInjectiveFunctionsEliminationPass and uniqCombined. [#65188](https://github.com/ClickHouse/ClickHouse/pull/65188) ([Raúl Marín](https://github.com/Algunenano)). +* Fix a bug in ClickHouse Keeper that causes digest mismatch during closing session. [#65198](https://github.com/ClickHouse/ClickHouse/pull/65198) ([Aleksei Filatov](https://github.com/aalexfvk)). +* Forbid `QUALIFY` clause in the old analyzer. The old analyzer ignored `QUALIFY`, so it could lead to unexpected data removal in mutations. [#65356](https://github.com/ClickHouse/ClickHouse/pull/65356) ([Dmitry Novik](https://github.com/novikd)). +* Use correct memory alignment for Distinct combinator. Previously, crash could happen because of invalid memory allocation when the combinator was used. [#65379](https://github.com/ClickHouse/ClickHouse/pull/65379) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#65846](https://github.com/ClickHouse/ClickHouse/issues/65846): Check cyclic dependencies on CREATE/REPLACE/RENAME/EXCHANGE queries and throw an exception if there is a cyclic dependency. Previously such cyclic dependencies could lead to a deadlock during server startup. Closes [#65355](https://github.com/ClickHouse/ClickHouse/issues/65355). Also fix some bugs in dependencies creation. [#65405](https://github.com/ClickHouse/ClickHouse/pull/65405) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#65714](https://github.com/ClickHouse/ClickHouse/issues/65714): Fix crash in maxIntersections. [#65689](https://github.com/ClickHouse/ClickHouse/pull/65689) ([Raúl Marín](https://github.com/Algunenano)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Fix making backup when multiple shards are used. This PR fixes [#56566](https://github.com/ClickHouse/ClickHouse/issues/56566). [#57684](https://github.com/ClickHouse/ClickHouse/pull/57684) ([Vitaly Baranov](https://github.com/vitlibar)). +* Fix passing projections/indexes from CREATE query into inner table of MV. [#59183](https://github.com/ClickHouse/ClickHouse/pull/59183) ([Azat Khuzhin](https://github.com/azat)). +* Fix boundRatio incorrect merge. [#60532](https://github.com/ClickHouse/ClickHouse/pull/60532) ([Tao Wang](https://github.com/wangtZJU)). +* Fix crash when using some functions with low-cardinality columns. [#61966](https://github.com/ClickHouse/ClickHouse/pull/61966) ([Michael Kolupaev](https://github.com/al13n321)). +* Fixed 'set' skip index not working with IN and indexHint(). [#62083](https://github.com/ClickHouse/ClickHouse/pull/62083) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix queries with FINAL give wrong result when table does not use adaptive granularity. [#62432](https://github.com/ClickHouse/ClickHouse/pull/62432) ([Duc Canh Le](https://github.com/canhld94)). +* Improve the detection of cgroups v2 memory controller in unusual locations. This fixes a warning that the cgroup memory observer was disabled because no cgroups v1 or v2 current memory file could be found. [#62903](https://github.com/ClickHouse/ClickHouse/pull/62903) ([Robert Schulze](https://github.com/rschu1ze)). +* Fix subsequent use of external tables in client. [#62964](https://github.com/ClickHouse/ClickHouse/pull/62964) ([Azat Khuzhin](https://github.com/azat)). +* Fix crash with untuple and unresolved lambda. [#63131](https://github.com/ClickHouse/ClickHouse/pull/63131) ([Raúl Marín](https://github.com/Algunenano)). +* Fix bug which could lead to server to accept connections before server is actually loaded. [#63181](https://github.com/ClickHouse/ClickHouse/pull/63181) ([alesapin](https://github.com/alesapin)). +* Fix intersect parts when restart after drop range. [#63202](https://github.com/ClickHouse/ClickHouse/pull/63202) ([Han Fei](https://github.com/hanfei1991)). +* Fix a misbehavior when SQL security defaults don't load for old tables during server startup. [#63209](https://github.com/ClickHouse/ClickHouse/pull/63209) ([pufit](https://github.com/pufit)). +* JOIN filter push down filled join fix. Closes [#63228](https://github.com/ClickHouse/ClickHouse/issues/63228). [#63234](https://github.com/ClickHouse/ClickHouse/pull/63234) ([Maksim Kita](https://github.com/kitaisreal)). +* Fix infinite loop while listing objects in Azure blob storage. [#63257](https://github.com/ClickHouse/ClickHouse/pull/63257) ([Julia Kartseva](https://github.com/jkartseva)). +* CROSS join can be executed with any value `join_algorithm` setting, close [#62431](https://github.com/ClickHouse/ClickHouse/issues/62431). [#63273](https://github.com/ClickHouse/ClickHouse/pull/63273) ([vdimir](https://github.com/vdimir)). +* Fixed a potential crash caused by a `no space left` error when temporary data in the cache is used. [#63346](https://github.com/ClickHouse/ClickHouse/pull/63346) ([vdimir](https://github.com/vdimir)). +* Fix bug which could potentially lead to rare LOGICAL_ERROR during SELECT query with message: `Unexpected return type from materialize. Expected type_XXX. Got type_YYY.` Introduced in [#59379](https://github.com/ClickHouse/ClickHouse/issues/59379). [#63353](https://github.com/ClickHouse/ClickHouse/pull/63353) ([alesapin](https://github.com/alesapin)). +* Fix `X-ClickHouse-Timezone` header returning wrong timezone when using `session_timezone` as query level setting. [#63377](https://github.com/ClickHouse/ClickHouse/pull/63377) ([Andrey Zvonov](https://github.com/zvonand)). +* Fix debug assert when using grouping WITH ROLLUP and LowCardinality types. [#63398](https://github.com/ClickHouse/ClickHouse/pull/63398) ([Raúl Marín](https://github.com/Algunenano)). +* Fix logical errors in queries with `GROUPING SETS` and `WHERE` and `group_by_use_nulls = true`, close [#60538](https://github.com/ClickHouse/ClickHouse/issues/60538). [#63405](https://github.com/ClickHouse/ClickHouse/pull/63405) ([vdimir](https://github.com/vdimir)). +* Fix backup of projection part in case projection was removed from table metadata, but part still has projection. [#63426](https://github.com/ClickHouse/ClickHouse/pull/63426) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix 'Every derived table must have its own alias' error for MYSQL dictionary source, close [#63341](https://github.com/ClickHouse/ClickHouse/issues/63341). [#63481](https://github.com/ClickHouse/ClickHouse/pull/63481) ([vdimir](https://github.com/vdimir)). +* Insert QueryFinish on AsyncInsertFlush with no data. [#63483](https://github.com/ClickHouse/ClickHouse/pull/63483) ([Raúl Marín](https://github.com/Algunenano)). +* Fix `system.query_log.used_dictionaries` logging. [#63487](https://github.com/ClickHouse/ClickHouse/pull/63487) ([Eduard Karacharov](https://github.com/korowa)). +* Support executing function during assignment of parameterized view value. [#63502](https://github.com/ClickHouse/ClickHouse/pull/63502) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Avoid segafult in `MergeTreePrefetchedReadPool` while fetching projection parts. [#63513](https://github.com/ClickHouse/ClickHouse/pull/63513) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix rabbitmq heap-use-after-free found by clang-18, which can happen if an error is thrown from RabbitMQ during initialization of exchange and queues. [#63515](https://github.com/ClickHouse/ClickHouse/pull/63515) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix crash on exit with sentry enabled (due to openssl destroyed before sentry). [#63548](https://github.com/ClickHouse/ClickHouse/pull/63548) ([Azat Khuzhin](https://github.com/azat)). +* Fixed parquet memory tracking. [#63584](https://github.com/ClickHouse/ClickHouse/pull/63584) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix support for Array and Map with Keyed hashing functions and materialized keys. [#63628](https://github.com/ClickHouse/ClickHouse/pull/63628) ([Salvatore Mesoraca](https://github.com/aiven-sal)). +* Fixed Parquet filter pushdown not working with Analyzer. [#63642](https://github.com/ClickHouse/ClickHouse/pull/63642) ([Michael Kolupaev](https://github.com/al13n321)). +* It is forbidden to convert MergeTree to replicated if the zookeeper path for this table already exists. [#63670](https://github.com/ClickHouse/ClickHouse/pull/63670) ([Kirill](https://github.com/kirillgarbar)). +* Read only the necessary columns from VIEW (new analyzer). Closes [#62594](https://github.com/ClickHouse/ClickHouse/issues/62594). [#63688](https://github.com/ClickHouse/ClickHouse/pull/63688) ([Maksim Kita](https://github.com/kitaisreal)). +* Fix rare case with missing data in the result of distributed query. [#63691](https://github.com/ClickHouse/ClickHouse/pull/63691) ([vdimir](https://github.com/vdimir)). +* Fix [#63539](https://github.com/ClickHouse/ClickHouse/issues/63539). Forbid WINDOW redefinition in new analyzer. [#63694](https://github.com/ClickHouse/ClickHouse/pull/63694) ([Dmitry Novik](https://github.com/novikd)). +* Flatten_nested is broken with replicated database. [#63695](https://github.com/ClickHouse/ClickHouse/pull/63695) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix `SIZES_OF_COLUMNS_DOESNT_MATCH` error for queries with `arrayJoin` function in `WHERE`. Fixes [#63653](https://github.com/ClickHouse/ClickHouse/issues/63653). [#63722](https://github.com/ClickHouse/ClickHouse/pull/63722) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix `Not found column` and `CAST AS Map from array requires nested tuple of 2 elements` exceptions for distributed queries which use `Map(Nothing, Nothing)` type. Fixes [#63637](https://github.com/ClickHouse/ClickHouse/issues/63637). [#63753](https://github.com/ClickHouse/ClickHouse/pull/63753) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix possible `ILLEGAL_COLUMN` error in `partial_merge` join, close [#37928](https://github.com/ClickHouse/ClickHouse/issues/37928). [#63755](https://github.com/ClickHouse/ClickHouse/pull/63755) ([vdimir](https://github.com/vdimir)). +* `query_plan_remove_redundant_distinct` can break queries with WINDOW FUNCTIONS (with `allow_experimental_analyzer` is on). Fixes [#62820](https://github.com/ClickHouse/ClickHouse/issues/62820). [#63776](https://github.com/ClickHouse/ClickHouse/pull/63776) ([Igor Nikonov](https://github.com/devcrafter)). +* Fix possible crash with SYSTEM UNLOAD PRIMARY KEY. [#63778](https://github.com/ClickHouse/ClickHouse/pull/63778) ([Raúl Marín](https://github.com/Algunenano)). +* Fix a query with a duplicating cycling alias. Fixes [#63320](https://github.com/ClickHouse/ClickHouse/issues/63320). [#63791](https://github.com/ClickHouse/ClickHouse/pull/63791) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fixed performance degradation of parsing data formats in INSERT query. This closes [#62918](https://github.com/ClickHouse/ClickHouse/issues/62918). This partially reverts [#42284](https://github.com/ClickHouse/ClickHouse/issues/42284), which breaks the original design and introduces more problems. [#63801](https://github.com/ClickHouse/ClickHouse/pull/63801) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add 'endpoint_subpath' S3 URI setting to allow plain_rewritable disks to share the same endpoint. [#63806](https://github.com/ClickHouse/ClickHouse/pull/63806) ([Julia Kartseva](https://github.com/jkartseva)). +* Fix queries using parallel read buffer (e.g. with max_download_thread > 0) getting stuck when threads cannot be allocated. [#63814](https://github.com/ClickHouse/ClickHouse/pull/63814) ([Antonio Andelic](https://github.com/antonio2368)). +* Allow JOIN filter push down to both streams if only single equivalent column is used in query. Closes [#63799](https://github.com/ClickHouse/ClickHouse/issues/63799). [#63819](https://github.com/ClickHouse/ClickHouse/pull/63819) ([Maksim Kita](https://github.com/kitaisreal)). +* Remove the data from all disks after DROP with the Lazy database engines. Without these changes, orhpaned will remain on the disks. [#63848](https://github.com/ClickHouse/ClickHouse/pull/63848) ([MikhailBurdukov](https://github.com/MikhailBurdukov)). +* Fix incorrect select query result when parallel replicas were used to read from a Materialized View. [#63861](https://github.com/ClickHouse/ClickHouse/pull/63861) ([Nikita Taranov](https://github.com/nickitat)). +* Fixes in `find_super_nodes` and `find_big_family` command of keeper-client: - do not fail on ZNONODE errors - find super nodes inside super nodes - properly calculate subtree node count. [#63862](https://github.com/ClickHouse/ClickHouse/pull/63862) ([Alexander Gololobov](https://github.com/davenger)). +* Fix a error `Database name is empty` for remote queries with lambdas over the cluster with modified default database. Fixes [#63471](https://github.com/ClickHouse/ClickHouse/issues/63471). [#63864](https://github.com/ClickHouse/ClickHouse/pull/63864) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix SIGSEGV due to CPU/Real (`query_profiler_real_time_period_ns`/`query_profiler_cpu_time_period_ns`) profiler (has been an issue since 2022, that leads to periodic server crashes, especially if you were using distributed engine). [#63865](https://github.com/ClickHouse/ClickHouse/pull/63865) ([Azat Khuzhin](https://github.com/azat)). +* Fixed `EXPLAIN CURRENT TRANSACTION` query. [#63926](https://github.com/ClickHouse/ClickHouse/pull/63926) ([Anton Popov](https://github.com/CurtizJ)). +* Fix analyzer - IN function with arbitrary deep sub-selects in materialized view to use insertion block. [#63930](https://github.com/ClickHouse/ClickHouse/pull/63930) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)). +* Allow `ALTER TABLE .. MODIFY|RESET SETTING` and `ALTER TABLE .. MODIFY COMMENT` for plain_rewritable disk. [#63933](https://github.com/ClickHouse/ClickHouse/pull/63933) ([Julia Kartseva](https://github.com/jkartseva)). +* Fix Recursive CTE with distributed queries. Closes [#63790](https://github.com/ClickHouse/ClickHouse/issues/63790). [#63939](https://github.com/ClickHouse/ClickHouse/pull/63939) ([Maksim Kita](https://github.com/kitaisreal)). +* Fixed reading of columns of type `Tuple(Map(LowCardinality(String), String), ...)`. [#63956](https://github.com/ClickHouse/ClickHouse/pull/63956) ([Anton Popov](https://github.com/CurtizJ)). +* Fix resolve of unqualified COLUMNS matcher. Preserve the input columns order and forbid usage of unknown identifiers. [#63962](https://github.com/ClickHouse/ClickHouse/pull/63962) ([Dmitry Novik](https://github.com/novikd)). +* Fix the `Not found column` error for queries with `skip_unused_shards = 1`, `LIMIT BY`, and the new analyzer. Fixes [#63943](https://github.com/ClickHouse/ClickHouse/issues/63943). [#63983](https://github.com/ClickHouse/ClickHouse/pull/63983) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* (Low-quality third-party Kusto Query Language). Resolve Client Abortion Issue When Using KQL Table Function in Interactive Mode. [#63992](https://github.com/ClickHouse/ClickHouse/pull/63992) ([Yong Wang](https://github.com/kashwy)). +* Fix an `Cyclic aliases` error for cyclic aliases of different type (expression and function). [#63993](https://github.com/ClickHouse/ClickHouse/pull/63993) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Deserialize untrusted binary inputs in a safer way. [#64024](https://github.com/ClickHouse/ClickHouse/pull/64024) ([Robert Schulze](https://github.com/rschu1ze)). +* Do not throw `Storage doesn't support FINAL` error for remote queries over non-MergeTree tables with `final = true` and new analyzer. Fixes [#63960](https://github.com/ClickHouse/ClickHouse/issues/63960). [#64037](https://github.com/ClickHouse/ClickHouse/pull/64037) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Add missing settings to recoverLostReplica. [#64040](https://github.com/ClickHouse/ClickHouse/pull/64040) ([Raúl Marín](https://github.com/Algunenano)). +* Fix unwind on SIGSEGV on aarch64 (due to small stack for signal). [#64058](https://github.com/ClickHouse/ClickHouse/pull/64058) ([Azat Khuzhin](https://github.com/azat)). +* This fix will use a proper redefined context with the correct definer for each individual view in the query pipeline. [#64079](https://github.com/ClickHouse/ClickHouse/pull/64079) ([pufit](https://github.com/pufit)). +* Fix analyzer: "Not found column" error is fixed when using INTERPOLATE. [#64096](https://github.com/ClickHouse/ClickHouse/pull/64096) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)). +* Fix azure backup writing multipart blocks as 1mb (read buffer size) instead of max_upload_part_size. [#64117](https://github.com/ClickHouse/ClickHouse/pull/64117) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix creating backups to S3 buckets with different credentials from the disk containing the file. [#64153](https://github.com/ClickHouse/ClickHouse/pull/64153) ([Antonio Andelic](https://github.com/antonio2368)). +* Prevent LOGICAL_ERROR on CREATE TABLE as MaterializedView. [#64174](https://github.com/ClickHouse/ClickHouse/pull/64174) ([Raúl Marín](https://github.com/Algunenano)). +* The query cache now considers two identical queries against different databases as different. The previous behavior could be used to bypass missing privileges to read from a table. [#64199](https://github.com/ClickHouse/ClickHouse/pull/64199) ([Robert Schulze](https://github.com/rschu1ze)). +* Fix possible abort on uncaught exception in ~WriteBufferFromFileDescriptor in StatusFile. [#64206](https://github.com/ClickHouse/ClickHouse/pull/64206) ([Kruglov Pavel](https://github.com/Avogar)). +* Ignore `text_log` config when using Keeper. [#64218](https://github.com/ClickHouse/ClickHouse/pull/64218) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix `duplicate alias` error for distributed queries with `ARRAY JOIN`. [#64226](https://github.com/ClickHouse/ClickHouse/pull/64226) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix unexpected accurateCast from string to integer. [#64255](https://github.com/ClickHouse/ClickHouse/pull/64255) ([wudidapaopao](https://github.com/wudidapaopao)). +* Fixed CNF simplification, in case any OR group contains mutually exclusive atoms. [#64256](https://github.com/ClickHouse/ClickHouse/pull/64256) ([Eduard Karacharov](https://github.com/korowa)). +* Fix Query Tree size validation. [#64377](https://github.com/ClickHouse/ClickHouse/pull/64377) ([Dmitry Novik](https://github.com/novikd)). +* Fix `Logical error: Bad cast` for `Buffer` table with `PREWHERE`. [#64388](https://github.com/ClickHouse/ClickHouse/pull/64388) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Prevent recursive logging in `blob_storage_log` when it's stored on object storage. [#64393](https://github.com/ClickHouse/ClickHouse/pull/64393) ([vdimir](https://github.com/vdimir)). +* Fixed `CREATE TABLE AS` queries for tables with default expressions. [#64455](https://github.com/ClickHouse/ClickHouse/pull/64455) ([Anton Popov](https://github.com/CurtizJ)). +* Fixed `optimize_read_in_order` behaviour for ORDER BY ... NULLS FIRST / LAST on tables with nullable keys. [#64483](https://github.com/ClickHouse/ClickHouse/pull/64483) ([Eduard Karacharov](https://github.com/korowa)). +* Fix the `Expression nodes list expected 1 projection names` and `Unknown expression or identifier` errors for queries with aliases to `GLOBAL IN.`. [#64517](https://github.com/ClickHouse/ClickHouse/pull/64517) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix an error `Cannot find column` in distributed queries with constant CTE in the `GROUP BY` key. [#64519](https://github.com/ClickHouse/ClickHouse/pull/64519) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fixed ORC statistics calculation, when writing, for unsigned types on all platforms and Int8 on ARM. [#64563](https://github.com/ClickHouse/ClickHouse/pull/64563) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix the crash loop when restoring from backup is blocked by creating an MV with a definer that hasn't been restored yet. [#64595](https://github.com/ClickHouse/ClickHouse/pull/64595) ([pufit](https://github.com/pufit)). +* Fix the output of function `formatDateTimeInJodaSyntax` when a formatter generates an uneven number of characters and the last character is `0`. For example, `SELECT formatDateTimeInJodaSyntax(toDate('2012-05-29'), 'D')` now correctly returns `150` instead of previously `15`. [#64614](https://github.com/ClickHouse/ClickHouse/pull/64614) ([LiuNeng](https://github.com/liuneng1994)). +* Do not rewrite aggregation if `-If` combinator is already used. [#64638](https://github.com/ClickHouse/ClickHouse/pull/64638) ([Dmitry Novik](https://github.com/novikd)). +* Fix type inference for float (in case of small buffer, i.e. `--max_read_buffer_size 1`). [#64641](https://github.com/ClickHouse/ClickHouse/pull/64641) ([Azat Khuzhin](https://github.com/azat)). +* Fix bug which could lead to non-working TTLs with expressions. [#64694](https://github.com/ClickHouse/ClickHouse/pull/64694) ([alesapin](https://github.com/alesapin)). +* Fix removing the `WHERE` and `PREWHERE` expressions, which are always true (for the new analyzer). [#64695](https://github.com/ClickHouse/ClickHouse/pull/64695) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fixed excessive part elimination by token-based text indexes (`ngrambf` , `full_text`) when filtering by result of `startsWith`, `endsWith`, `match`, `multiSearchAny`. [#64720](https://github.com/ClickHouse/ClickHouse/pull/64720) ([Eduard Karacharov](https://github.com/korowa)). +* Fixes incorrect behaviour of ANSI CSI escaping in the `UTF8::computeWidth` function. [#64756](https://github.com/ClickHouse/ClickHouse/pull/64756) ([Shaun Struwig](https://github.com/Blargian)). +* Fix a case of incorrect removal of `ORDER BY` / `LIMIT BY` across subqueries. [#64766](https://github.com/ClickHouse/ClickHouse/pull/64766) ([Raúl Marín](https://github.com/Algunenano)). +* Fix (experimental) unequal join with subqueries for sets which are in the mixed join conditions. [#64775](https://github.com/ClickHouse/ClickHouse/pull/64775) ([lgbo](https://github.com/lgbo-ustc)). +* Fix crash in a local cache over `plain_rewritable` disk. [#64778](https://github.com/ClickHouse/ClickHouse/pull/64778) ([Julia Kartseva](https://github.com/jkartseva)). +* Keeper fix: return correct value for `zk_latest_snapshot_size` in `mntr` command. [#64784](https://github.com/ClickHouse/ClickHouse/pull/64784) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix `Cannot find column` in distributed query with `ARRAY JOIN` by `Nested` column. Fixes [#64755](https://github.com/ClickHouse/ClickHouse/issues/64755). [#64801](https://github.com/ClickHouse/ClickHouse/pull/64801) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix memory leak in slru cache policy. [#64803](https://github.com/ClickHouse/ClickHouse/pull/64803) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fixed possible incorrect memory tracking in several kinds of queries: queries that read any data from S3, queries via http protocol, asynchronous inserts. [#64844](https://github.com/ClickHouse/ClickHouse/pull/64844) ([Anton Popov](https://github.com/CurtizJ)). +* Fix the `Block structure mismatch` error for queries reading with `PREWHERE` from the materialized view when the materialized view has columns of different types than the source table. Fixes [#64611](https://github.com/ClickHouse/ClickHouse/issues/64611). [#64855](https://github.com/ClickHouse/ClickHouse/pull/64855) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix rare crash when table has TTL with subquery + database replicated + parallel replicas + analyzer. It's really rare, but please don't use TTLs with subqueries. [#64858](https://github.com/ClickHouse/ClickHouse/pull/64858) ([alesapin](https://github.com/alesapin)). +* Fix duplicating `Delete` events in `blob_storage_log` in case of large batch to delete. [#64924](https://github.com/ClickHouse/ClickHouse/pull/64924) ([vdimir](https://github.com/vdimir)). +* Backported in [#65544](https://github.com/ClickHouse/ClickHouse/issues/65544): Fix crash for `ALTER TABLE ... ON CLUSTER ... MODIFY SQL SECURITY`. [#64957](https://github.com/ClickHouse/ClickHouse/pull/64957) ([pufit](https://github.com/pufit)). +* Fixed `Session moved to another server` error from [Zoo]Keeper that might happen after server startup when the config has includes from [Zoo]Keeper. [#64986](https://github.com/ClickHouse/ClickHouse/pull/64986) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#65582](https://github.com/ClickHouse/ClickHouse/issues/65582): Fix crash on destroying AccessControl: add explicit shutdown. [#64993](https://github.com/ClickHouse/ClickHouse/pull/64993) ([Vitaly Baranov](https://github.com/vitlibar)). +* Fix `ALTER MODIFY COMMENT` query that was broken for parameterized VIEWs in https://github.com/ClickHouse/ClickHouse/pull/54211. [#65031](https://github.com/ClickHouse/ClickHouse/pull/65031) ([Nikolay Degterinsky](https://github.com/evillique)). +* Fix `host_id` in DatabaseReplicated when `cluster_secure_connection` parameter is enabled. Previously all the connections within the cluster created by DatabaseReplicated were not secure, even if the parameter was enabled. [#65054](https://github.com/ClickHouse/ClickHouse/pull/65054) ([Nikolay Degterinsky](https://github.com/evillique)). +* Fixing the `Not-ready Set` error after the `PREWHERE` optimization for StorageMerge. [#65057](https://github.com/ClickHouse/ClickHouse/pull/65057) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Avoid writing to finalized buffer in File-like storages. [#65063](https://github.com/ClickHouse/ClickHouse/pull/65063) ([Kruglov Pavel](https://github.com/Avogar)). +* Fix possible infinite query duration in case of cyclic aliases. Fixes [#64849](https://github.com/ClickHouse/ClickHouse/issues/64849). [#65081](https://github.com/ClickHouse/ClickHouse/pull/65081) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix the `Unknown expression identifier` error for remote queries with `INTERPOLATE (alias)` (new analyzer). Fixes [#64636](https://github.com/ClickHouse/ClickHouse/issues/64636). [#65090](https://github.com/ClickHouse/ClickHouse/pull/65090) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix pushing arithmetic operations out of aggregation. In the new analyzer, optimization was applied only once. [#65104](https://github.com/ClickHouse/ClickHouse/pull/65104) ([Dmitry Novik](https://github.com/novikd)). +* Fix aggregate function name rewriting in the new analyzer. [#65110](https://github.com/ClickHouse/ClickHouse/pull/65110) ([Dmitry Novik](https://github.com/novikd)). +* Respond with 5xx instead of 200 OK in case of receive timeout while reading (parts of) the request body from the client socket. [#65118](https://github.com/ClickHouse/ClickHouse/pull/65118) ([Julian Maicher](https://github.com/jmaicher)). +* Backported in [#65734](https://github.com/ClickHouse/ClickHouse/issues/65734): Eliminate injective function in argument of functions `uniq*` recursively. This used to work correctly but was broken in the new analyzer. [#65140](https://github.com/ClickHouse/ClickHouse/pull/65140) ([Duc Canh Le](https://github.com/canhld94)). +* Fix possible crash for hedged requests. [#65206](https://github.com/ClickHouse/ClickHouse/pull/65206) ([Azat Khuzhin](https://github.com/azat)). +* Fix the bug in Hashed and Hashed_Array dictionary short circuit evaluation, which may read uninitialized number, leading to various errors. [#65256](https://github.com/ClickHouse/ClickHouse/pull/65256) ([jsc0218](https://github.com/jsc0218)). +* This PR ensures that the type of the constant(IN operator's second parameter) is always visible during the IN operator's type conversion process. Otherwise, losing type information may cause some conversions to fail, such as the conversion from DateTime to Date. fix ([#64487](https://github.com/ClickHouse/ClickHouse/issues/64487)). [#65315](https://github.com/ClickHouse/ClickHouse/pull/65315) ([pn](https://github.com/chloro-pn)). +* Backported in [#65665](https://github.com/ClickHouse/ClickHouse/issues/65665): Disable `non-intersecting-parts` optimization for queries with `FINAL` in case of `read-in-order` optimization was enabled. This could lead to an incorrect query result. As a workaround, disable `do_not_merge_across_partitions_select_final` and `split_parts_ranges_into_intersecting_and_non_intersecting_final` before this fix is merged. [#65505](https://github.com/ClickHouse/ClickHouse/pull/65505) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#65606](https://github.com/ClickHouse/ClickHouse/issues/65606): Fix getting exception `Index out of bound for blob metadata` in case all files from list batch were filtered out. [#65523](https://github.com/ClickHouse/ClickHouse/pull/65523) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#65790](https://github.com/ClickHouse/ClickHouse/issues/65790): Fixed bug in MergeJoin. Column in sparse serialisation might be treated as a column of its nested type though the required conversion wasn't performed. [#65632](https://github.com/ClickHouse/ClickHouse/pull/65632) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#65814](https://github.com/ClickHouse/ClickHouse/issues/65814): Fix invalid exceptions in function `parseDateTime` with `%F` and `%D` placeholders. [#65768](https://github.com/ClickHouse/ClickHouse/pull/65768) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#65830](https://github.com/ClickHouse/ClickHouse/issues/65830): Fix a bug in short circuit logic when old analyzer and dictGetOrDefault is used. [#65802](https://github.com/ClickHouse/ClickHouse/pull/65802) ([jsc0218](https://github.com/jsc0218)). + +#### Build/Testing/Packaging Improvement +* ClickHouse is built with clang-18. A lot of new checks from clang-tidy-18 have been enabled. [#60469](https://github.com/ClickHouse/ClickHouse/pull/60469) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Make `network` service be required when using the rc init script to start the ClickHouse server daemon. [#60650](https://github.com/ClickHouse/ClickHouse/pull/60650) ([Chun-Sheng, Li](https://github.com/peter279k)). +* Re-enable broken s390x build in CI. [#63135](https://github.com/ClickHouse/ClickHouse/pull/63135) ([Harry Lee](https://github.com/HarryLeeIBM)). +* The Dockerfile is reviewed by the docker official library in https://github.com/docker-library/official-images/pull/15846. [#63400](https://github.com/ClickHouse/ClickHouse/pull/63400) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Information about every symbol in every translation unit will be collected in the CI database for every build in the CI. This closes [#63494](https://github.com/ClickHouse/ClickHouse/issues/63494). [#63495](https://github.com/ClickHouse/ClickHouse/pull/63495) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Experimentally support loongarch64 as a new platform for ClickHouse. [#63733](https://github.com/ClickHouse/ClickHouse/pull/63733) ([qiangxuhui](https://github.com/qiangxuhui)). +* Update Apache Datasketches library. It resolves [#63858](https://github.com/ClickHouse/ClickHouse/issues/63858). [#63923](https://github.com/ClickHouse/ClickHouse/pull/63923) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Enable GRPC support for aarch64 linux while cross-compiling binary. [#64072](https://github.com/ClickHouse/ClickHouse/pull/64072) ([alesapin](https://github.com/alesapin)). +* Fix typo in test_hdfsCluster_unset_skip_unavailable_shards. The test writes data to unskip_unavailable_shards, but uses skip_unavailable_shards from the previous test. [#64243](https://github.com/ClickHouse/ClickHouse/pull/64243) ([Mikhail Artemenko](https://github.com/Michicosun)). +* Reduce the size of some slow tests. [#64387](https://github.com/ClickHouse/ClickHouse/pull/64387) ([Raúl Marín](https://github.com/Algunenano)). +* Reduce the size of some slow tests. [#64452](https://github.com/ClickHouse/ClickHouse/pull/64452) ([Raúl Marín](https://github.com/Algunenano)). +* Fix test_lost_part_other_replica. [#64512](https://github.com/ClickHouse/ClickHouse/pull/64512) ([Raúl Marín](https://github.com/Algunenano)). +* Add tests for experimental unequal joins and randomize new settings in clickhouse-test. [#64535](https://github.com/ClickHouse/ClickHouse/pull/64535) ([Nikita Fomichev](https://github.com/fm4v)). +* Upgrade tests: Update config and work with release candidates. [#64542](https://github.com/ClickHouse/ClickHouse/pull/64542) ([Raúl Marín](https://github.com/Algunenano)). +* Add support for LLVM XRay. [#64592](https://github.com/ClickHouse/ClickHouse/pull/64592) ([Tomer Shafir](https://github.com/tomershafir)). +* Speed up 02995_forget_partition. [#64761](https://github.com/ClickHouse/ClickHouse/pull/64761) ([Raúl Marín](https://github.com/Algunenano)). +* Fix 02790_async_queries_in_query_log. [#64764](https://github.com/ClickHouse/ClickHouse/pull/64764) ([Raúl Marín](https://github.com/Algunenano)). +* Support LLVM XRay on Linux amd64 only. [#64837](https://github.com/ClickHouse/ClickHouse/pull/64837) ([Tomer Shafir](https://github.com/tomershafir)). +* Get rid of custom code in `tests/ci/download_release_packages.py` and `tests/ci/get_previous_release_tag.py` to avoid issues after the https://github.com/ClickHouse/ClickHouse/pull/64759 is merged. [#64848](https://github.com/ClickHouse/ClickHouse/pull/64848) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Decrease the `unit-test` image a few times. [#65102](https://github.com/ClickHouse/ClickHouse/pull/65102) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). + +#### NO CL CATEGORY + +* Backported in [#65568](https://github.com/ClickHouse/ClickHouse/issues/65568):. [#65498](https://github.com/ClickHouse/ClickHouse/pull/65498) ([Sergei Trifonov](https://github.com/serxa)). +* Backported in [#65693](https://github.com/ClickHouse/ClickHouse/issues/65693):. [#65686](https://github.com/ClickHouse/ClickHouse/pull/65686) ([Raúl Marín](https://github.com/Algunenano)). + +#### NO CL ENTRY + +* NO CL ENTRY: 'Revert "Do not remove server constants from GROUP BY key for secondary query."'. [#63297](https://github.com/ClickHouse/ClickHouse/pull/63297) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* NO CL ENTRY: 'Revert "Introduce bulk loading to StorageEmbeddedRocksDB"'. [#63316](https://github.com/ClickHouse/ClickHouse/pull/63316) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* NO CL ENTRY: 'Revert "Revert "Do not remove server constants from GROUP BY key for secondary query.""'. [#63415](https://github.com/ClickHouse/ClickHouse/pull/63415) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* NO CL ENTRY: 'Revert "Fix index analysis for `DateTime64`"'. [#63525](https://github.com/ClickHouse/ClickHouse/pull/63525) ([Raúl Marín](https://github.com/Algunenano)). +* NO CL ENTRY: 'Revert "Update gui.md - Add ch-ui to open-source available tools."'. [#64064](https://github.com/ClickHouse/ClickHouse/pull/64064) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* NO CL ENTRY: 'Revert "Prevent conversion to Replicated if zookeeper path already exists"'. [#64214](https://github.com/ClickHouse/ClickHouse/pull/64214) ([Sergei Trifonov](https://github.com/serxa)). +* NO CL ENTRY: 'Revert "Refactoring of Server.h: Isolate server management from other logic"'. [#64425](https://github.com/ClickHouse/ClickHouse/pull/64425) ([Alexander Tokmakov](https://github.com/tavplubix)). +* NO CL ENTRY: 'Revert "Remove some unnecessary `UNREACHABLE`s"'. [#64430](https://github.com/ClickHouse/ClickHouse/pull/64430) ([Alexander Tokmakov](https://github.com/tavplubix)). +* NO CL ENTRY: 'Revert "CI: fix build_report selection in case of job reuse"'. [#64516](https://github.com/ClickHouse/ClickHouse/pull/64516) ([Max K.](https://github.com/maxknv)). +* NO CL ENTRY: 'Revert "Revert "CI: fix build_report selection in case of job reuse""'. [#64531](https://github.com/ClickHouse/ClickHouse/pull/64531) ([Max K.](https://github.com/maxknv)). +* NO CL ENTRY: 'Revert "Add `fromReadableSize` function"'. [#64616](https://github.com/ClickHouse/ClickHouse/pull/64616) ([Robert Schulze](https://github.com/rschu1ze)). +* NO CL ENTRY: 'Update CHANGELOG.md'. [#64816](https://github.com/ClickHouse/ClickHouse/pull/64816) ([Paweł Kudzia](https://github.com/pakud)). +* NO CL ENTRY: 'Revert "Reduce lock contention for MergeTree tables (by renaming parts without holding lock)"'. [#64899](https://github.com/ClickHouse/ClickHouse/pull/64899) ([alesapin](https://github.com/alesapin)). +* NO CL ENTRY: 'Revert "Add dynamic untracked memory limits for more precise memory tracking"'. [#64969](https://github.com/ClickHouse/ClickHouse/pull/64969) ([Sergei Trifonov](https://github.com/serxa)). +* NO CL ENTRY: 'Revert "Fix duplicating Delete events in blob_storage_log"'. [#65049](https://github.com/ClickHouse/ClickHouse/pull/65049) ([Alexander Tokmakov](https://github.com/tavplubix)). +* NO CL ENTRY: 'Revert "Revert "Fix duplicating Delete events in blob_storage_log""'. [#65053](https://github.com/ClickHouse/ClickHouse/pull/65053) ([vdimir](https://github.com/vdimir)). +* NO CL ENTRY: 'Revert "S3: reduce retires time for queries, increase retries count for backups"'. [#65148](https://github.com/ClickHouse/ClickHouse/pull/65148) ([Raúl Marín](https://github.com/Algunenano)). +* NO CL ENTRY: 'Revert "Small fix for 02340_parts_refcnt_mergetree"'. [#65149](https://github.com/ClickHouse/ClickHouse/pull/65149) ([Raúl Marín](https://github.com/Algunenano)). +* NO CL ENTRY: 'Revert "Change default s3_throw_on_zero_files_match to true, document that presigned S3 URLs are not supported"'. [#65250](https://github.com/ClickHouse/ClickHouse/pull/65250) ([Max K.](https://github.com/maxknv)). +* NO CL ENTRY: 'Revert "Fix AWS ECS"'. [#65361](https://github.com/ClickHouse/ClickHouse/pull/65361) ([Alexander Tokmakov](https://github.com/tavplubix)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Try abort on current thread join. [#42544](https://github.com/ClickHouse/ClickHouse/pull/42544) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* This change was reverted. [#51008](https://github.com/ClickHouse/ClickHouse/pull/51008) ([Michael Kolupaev](https://github.com/al13n321)). +* Analyzer fuzzer 2. [#57098](https://github.com/ClickHouse/ClickHouse/pull/57098) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Analyzer fuzzer 4. [#57101](https://github.com/ClickHouse/ClickHouse/pull/57101) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Check python code with flake8. [#58349](https://github.com/ClickHouse/ClickHouse/pull/58349) ([Azat Khuzhin](https://github.com/azat)). +* Unite s3/hdfs/azure storage implementations into a single class working with IObjectStorage. Same for *Cluster, data lakes and Queue storages. [#59767](https://github.com/ClickHouse/ClickHouse/pull/59767) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Remove http_max_chunk_size setting (too internal). [#60852](https://github.com/ClickHouse/ClickHouse/pull/60852) ([Azat Khuzhin](https://github.com/azat)). +* Fix race in refreshable materialized views causing SELECT to fail sometimes. [#60883](https://github.com/ClickHouse/ClickHouse/pull/60883) ([Michael Kolupaev](https://github.com/al13n321)). +* Refactor KeyCondition and key analysis to improve PartitionPruner and trivial count optimization. This is separated from [#60463](https://github.com/ClickHouse/ClickHouse/issues/60463) . [#61459](https://github.com/ClickHouse/ClickHouse/pull/61459) ([Amos Bird](https://github.com/amosbird)). +* Implement cumulative A Sync status. [#61464](https://github.com/ClickHouse/ClickHouse/pull/61464) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Parallel replicas: table check failover. [#61935](https://github.com/ClickHouse/ClickHouse/pull/61935) ([Igor Nikonov](https://github.com/devcrafter)). +* This change was reverted. [#61973](https://github.com/ClickHouse/ClickHouse/pull/61973) ([Azat Khuzhin](https://github.com/azat)). +* Avoid crashing on column type mismatch in a few dozen places. [#62087](https://github.com/ClickHouse/ClickHouse/pull/62087) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix optimize_if_chain_to_multiif const NULL handling. [#62104](https://github.com/ClickHouse/ClickHouse/pull/62104) ([Michael Kolupaev](https://github.com/al13n321)). +* Use intrusive lists for `ResourceRequest` instead of deque. [#62165](https://github.com/ClickHouse/ClickHouse/pull/62165) ([Sergei Trifonov](https://github.com/serxa)). +* Analyzer: Fix validateAggregates for tables with different aliases. [#62346](https://github.com/ClickHouse/ClickHouse/pull/62346) ([vdimir](https://github.com/vdimir)). +* Improve code and tests of `DROP` of multiple tables. [#62359](https://github.com/ClickHouse/ClickHouse/pull/62359) ([zhongyuankai](https://github.com/zhongyuankai)). +* Fix exception message during writing to partitioned s3/hdfs/azure path with globs. [#62423](https://github.com/ClickHouse/ClickHouse/pull/62423) ([Kruglov Pavel](https://github.com/Avogar)). +* Support UBSan on Clang-19 (master). [#62466](https://github.com/ClickHouse/ClickHouse/pull/62466) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Replay ZK logs using keeper-bench. [#62481](https://github.com/ClickHouse/ClickHouse/pull/62481) ([Antonio Andelic](https://github.com/antonio2368)). +* Save the stacktrace of thread waiting on failing AsyncLoader job. [#62719](https://github.com/ClickHouse/ClickHouse/pull/62719) ([Sergei Trifonov](https://github.com/serxa)). +* group_by_use_nulls strikes back. [#62922](https://github.com/ClickHouse/ClickHouse/pull/62922) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Analyzer: prefer column name to alias from array join. [#62995](https://github.com/ClickHouse/ClickHouse/pull/62995) ([vdimir](https://github.com/vdimir)). +* CI: try separate the workflows file for GitHub's Merge Queue. [#63123](https://github.com/ClickHouse/ClickHouse/pull/63123) ([Max K.](https://github.com/maxknv)). +* Try to fix coverage tests. [#63130](https://github.com/ClickHouse/ClickHouse/pull/63130) ([Raúl Marín](https://github.com/Algunenano)). +* Fix azure backup flaky test. [#63158](https://github.com/ClickHouse/ClickHouse/pull/63158) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Merging [#60920](https://github.com/ClickHouse/ClickHouse/issues/60920). [#63159](https://github.com/ClickHouse/ClickHouse/pull/63159) ([vdimir](https://github.com/vdimir)). +* QueryAnalysisPass improve QUALIFY validation. [#63162](https://github.com/ClickHouse/ClickHouse/pull/63162) ([Maksim Kita](https://github.com/kitaisreal)). +* Add numpy tests for different endianness. [#63189](https://github.com/ClickHouse/ClickHouse/pull/63189) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Clean the `_work` directory between runner's launches. Fallback to auto-update actions runner if it fails to start. Make the `init-network.sh` sourceable and executable. [#63195](https://github.com/ClickHouse/ClickHouse/pull/63195) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Add ability to run Azure tests in PR with label. [#63196](https://github.com/ClickHouse/ClickHouse/pull/63196) ([alesapin](https://github.com/alesapin)). +* Fix possible endless loop while reading from azure. [#63197](https://github.com/ClickHouse/ClickHouse/pull/63197) ([Anton Popov](https://github.com/CurtizJ)). +* Add information about materialized view security bug fix into the changelog. [#63204](https://github.com/ClickHouse/ClickHouse/pull/63204) ([pufit](https://github.com/pufit)). +* Disable one test from 02994_sanity_check_settings. [#63208](https://github.com/ClickHouse/ClickHouse/pull/63208) ([Raúl Marín](https://github.com/Algunenano)). +* Enable custom parquet encoder by default, attempt 2. [#63210](https://github.com/ClickHouse/ClickHouse/pull/63210) ([Michael Kolupaev](https://github.com/al13n321)). +* Update version after release. [#63215](https://github.com/ClickHouse/ClickHouse/pull/63215) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update version_date.tsv and changelogs after v24.4.1.2088-stable. [#63217](https://github.com/ClickHouse/ClickHouse/pull/63217) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Update version_date.tsv and changelogs after v24.3.3.102-lts. [#63226](https://github.com/ClickHouse/ClickHouse/pull/63226) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Update version_date.tsv and changelogs after v24.2.3.70-stable. [#63227](https://github.com/ClickHouse/ClickHouse/pull/63227) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Return back [#61551](https://github.com/ClickHouse/ClickHouse/issues/61551) (More optimal loading of marks). [#63233](https://github.com/ClickHouse/ClickHouse/pull/63233) ([Anton Popov](https://github.com/CurtizJ)). +* Hide CI options under a spoiler. [#63237](https://github.com/ClickHouse/ClickHouse/pull/63237) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Add azure run with msan. [#63238](https://github.com/ClickHouse/ClickHouse/pull/63238) ([alesapin](https://github.com/alesapin)). +* Now syntax for this command is following: `TRUNCATE ALL TABLES FROM [IF EXISTS] `. [#63241](https://github.com/ClickHouse/ClickHouse/pull/63241) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Minor follow-up to a renaming PR. [#63260](https://github.com/ClickHouse/ClickHouse/pull/63260) ([Robert Schulze](https://github.com/rschu1ze)). +* Followup for [#62613](https://github.com/ClickHouse/ClickHouse/issues/62613) Adding back checks similar to these: https://github.com/ClickHouse/ClickHouse/pull/62613/files#diff-70859078da57ecdfc66d26f732c0d7718d269e82bdc80e62b39f5ffeab36c05bL99 https://github.com/ClickHouse/ClickHouse/pull/62613/files#diff-70859078da57ecdfc66d26f732c0d7718d269e82bdc80e62b39f5ffeab36c05bL144-L149. [#63274](https://github.com/ClickHouse/ClickHouse/pull/63274) ([Alexander Gololobov](https://github.com/davenger)). +* This setting was added in 24.5, not 24.4. [#63278](https://github.com/ClickHouse/ClickHouse/pull/63278) ([Raúl Marín](https://github.com/Algunenano)). +* Improve cloud backport script. [#63282](https://github.com/ClickHouse/ClickHouse/pull/63282) ([Raúl Marín](https://github.com/Algunenano)). +* Update version_date.tsv and changelogs after v23.8.14.6-lts. [#63285](https://github.com/ClickHouse/ClickHouse/pull/63285) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Fix azure flaky test. [#63286](https://github.com/ClickHouse/ClickHouse/pull/63286) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Fix deadlock in `CacheDictionaryUpdateQueue` in case of exception in constructor. [#63287](https://github.com/ClickHouse/ClickHouse/pull/63287) ([Nikita Taranov](https://github.com/nickitat)). +* DiskApp: fix 'list --recursive /' and crash on invalid arguments. [#63296](https://github.com/ClickHouse/ClickHouse/pull/63296) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix terminate because of unhandled exception in `MergeTreeDeduplicationLog::shutdown`. [#63298](https://github.com/ClickHouse/ClickHouse/pull/63298) ([Nikita Taranov](https://github.com/nickitat)). +* Move s3_plain_rewritable unit test to shell. [#63317](https://github.com/ClickHouse/ClickHouse/pull/63317) ([Julia Kartseva](https://github.com/jkartseva)). +* Add tests for [#63264](https://github.com/ClickHouse/ClickHouse/issues/63264). [#63321](https://github.com/ClickHouse/ClickHouse/pull/63321) ([Raúl Marín](https://github.com/Algunenano)). +* Try fix segfault in `MergeTreeReadPoolBase::createTask`. [#63323](https://github.com/ClickHouse/ClickHouse/pull/63323) ([Antonio Andelic](https://github.com/antonio2368)). +* Reduce time-to-insert profiling data in case of logs cluster issues. [#63325](https://github.com/ClickHouse/ClickHouse/pull/63325) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Update README.md. [#63326](https://github.com/ClickHouse/ClickHouse/pull/63326) ([Tyler Hannan](https://github.com/tylerhannan)). +* This should fix failures with error like `Permission denied ["/var/lib/clickhouse/disks/s3/store/364/3643ff83-0996-4a4a-a90b-a96e66a10c74"]` when table dir was chmod-ed by DatabaseCatalog. [#63330](https://github.com/ClickHouse/ClickHouse/pull/63330) ([Alexander Gololobov](https://github.com/davenger)). +* Use `/commit/` to have the URLs in [reports](https://play.clickhouse.com/play?user=play#c2VsZWN0IGRpc3RpbmN0IGNvbW1pdF91cmwgZnJvbSBjaGVja3Mgd2hlcmUgY2hlY2tfc3RhcnRfdGltZSA+PSBub3coKSAtIGludGVydmFsIDEgbW9udGggYW5kIHB1bGxfcmVxdWVzdF9udW1iZXI9NjA1MzI=) like https://github.com/ClickHouse/ClickHouse/commit/44f8bc5308b53797bec8cccc3bd29fab8a00235d and not like https://github.com/ClickHouse/ClickHouse/commits/44f8bc5308b53797bec8cccc3bd29fab8a00235d. [#63331](https://github.com/ClickHouse/ClickHouse/pull/63331) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Add test for [#56287](https://github.com/ClickHouse/ClickHouse/issues/56287). [#63340](https://github.com/ClickHouse/ClickHouse/pull/63340) ([Raúl Marín](https://github.com/Algunenano)). +* Update README.md. [#63350](https://github.com/ClickHouse/ClickHouse/pull/63350) ([Tyler Hannan](https://github.com/tylerhannan)). +* Add test for [#48049](https://github.com/ClickHouse/ClickHouse/issues/48049). [#63351](https://github.com/ClickHouse/ClickHouse/pull/63351) ([Raúl Marín](https://github.com/Algunenano)). +* Add option `query_id_prefix` to `clickhouse-benchmark`. [#63352](https://github.com/ClickHouse/ClickHouse/pull/63352) ([Anton Popov](https://github.com/CurtizJ)). +* New version is fantatish (at least with Ubuntu 22.04.4 LTS): ``` azurite --version /usr/local/lib/node_modules/azurite/dist/src/common/persistence/MemoryExtentStore.js:53 return this._chunks.get(categoryName)?.chunks.get(id); ^. [#63354](https://github.com/ClickHouse/ClickHouse/pull/63354) ([alesapin](https://github.com/alesapin)). +* Randomize setting `enable_block_offset_column` in stress tests. [#63355](https://github.com/ClickHouse/ClickHouse/pull/63355) ([Anton Popov](https://github.com/CurtizJ)). +* Fix AST parsing of invalid type names. [#63357](https://github.com/ClickHouse/ClickHouse/pull/63357) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix some 00002_log_and_exception_messages_formatting flakiness. [#63358](https://github.com/ClickHouse/ClickHouse/pull/63358) ([Michael Kolupaev](https://github.com/al13n321)). +* Add tags for the test 03000_traverse_shadow_system_data_paths.sql to make it stable. [#63366](https://github.com/ClickHouse/ClickHouse/pull/63366) ([Aleksei Filatov](https://github.com/aalexfvk)). +* Add a test for [#55655](https://github.com/ClickHouse/ClickHouse/issues/55655). [#63380](https://github.com/ClickHouse/ClickHouse/pull/63380) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix data race in `reportBrokenPart`. [#63396](https://github.com/ClickHouse/ClickHouse/pull/63396) ([Antonio Andelic](https://github.com/antonio2368)). +* Workaround for `oklch()` inside canvas bug for firefox. [#63404](https://github.com/ClickHouse/ClickHouse/pull/63404) ([Sergei Trifonov](https://github.com/serxa)). +* Add test for issue [#47862](https://github.com/ClickHouse/ClickHouse/issues/47862). [#63424](https://github.com/ClickHouse/ClickHouse/pull/63424) ([Robert Schulze](https://github.com/rschu1ze)). +* Fix parsing of `CREATE INDEX` query. [#63425](https://github.com/ClickHouse/ClickHouse/pull/63425) ([Anton Popov](https://github.com/CurtizJ)). +* We are using Shared Catalog in the CI Logs cluster. [#63442](https://github.com/ClickHouse/ClickHouse/pull/63442) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix collection of coverage data in the CI Logs cluster. [#63453](https://github.com/ClickHouse/ClickHouse/pull/63453) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix flaky test for rocksdb bulk sink. [#63457](https://github.com/ClickHouse/ClickHouse/pull/63457) ([Duc Canh Le](https://github.com/canhld94)). +* Extra constraints for stress and fuzzer tests. [#63470](https://github.com/ClickHouse/ClickHouse/pull/63470) ([Raúl Marín](https://github.com/Algunenano)). +* io_uring: refactor get reader from context. [#63475](https://github.com/ClickHouse/ClickHouse/pull/63475) ([Tomer Shafir](https://github.com/tomershafir)). +* Analyzer setting max_streams_to_max_threads_ratio overflow fix. [#63478](https://github.com/ClickHouse/ClickHouse/pull/63478) ([Maksim Kita](https://github.com/kitaisreal)). +* Provides setting `output_format_pretty_preserve_border_for_multiline_string` which allows to render multiline strings in pretty format better. The default value for this setting is true. [#63479](https://github.com/ClickHouse/ClickHouse/pull/63479) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Fix logical error when reloading config with customly created web disk broken after [#56367](https://github.com/ClickHouse/ClickHouse/issues/56367). [#63484](https://github.com/ClickHouse/ClickHouse/pull/63484) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Add test for [#49307](https://github.com/ClickHouse/ClickHouse/issues/49307). [#63486](https://github.com/ClickHouse/ClickHouse/pull/63486) ([Anton Popov](https://github.com/CurtizJ)). +* Remove leftovers of GCC support in cmake rules. [#63488](https://github.com/ClickHouse/ClickHouse/pull/63488) ([Azat Khuzhin](https://github.com/azat)). +* Fix ProfileEventTimeIncrement code. [#63489](https://github.com/ClickHouse/ClickHouse/pull/63489) ([Azat Khuzhin](https://github.com/azat)). +* MergeTreePrefetchedReadPool: Print parent name when logging projection parts. [#63522](https://github.com/ClickHouse/ClickHouse/pull/63522) ([Raúl Marín](https://github.com/Algunenano)). +* Correctly stop `asyncCopy` tasks in all cases. [#63523](https://github.com/ClickHouse/ClickHouse/pull/63523) ([Antonio Andelic](https://github.com/antonio2368)). +* Almost everything should work on AArch64 (Part of [#58061](https://github.com/ClickHouse/ClickHouse/issues/58061)). [#63527](https://github.com/ClickHouse/ClickHouse/pull/63527) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update randomization of `old_parts_lifetime`. [#63530](https://github.com/ClickHouse/ClickHouse/pull/63530) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Update 02240_system_filesystem_cache_table.sh. [#63531](https://github.com/ClickHouse/ClickHouse/pull/63531) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix data race in `DistributedSink`. [#63538](https://github.com/ClickHouse/ClickHouse/pull/63538) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix azure tests run on master. [#63540](https://github.com/ClickHouse/ClickHouse/pull/63540) ([alesapin](https://github.com/alesapin)). +* The commit 2b8254f987a65d5c21d74fe67b4ee9757970466e was not synced into the cloud because it was falsely marked as a success by `upstream_pr.head.sha`. Here we'll try our best to find a proper commit, and won't make anything if we can't. [#63543](https://github.com/ClickHouse/ClickHouse/pull/63543) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Add `no-s3-storage` tag to local_plain_rewritable ut. [#63546](https://github.com/ClickHouse/ClickHouse/pull/63546) ([Julia Kartseva](https://github.com/jkartseva)). +* Add `jwcrypto` to integration tests runner. [#63551](https://github.com/ClickHouse/ClickHouse/pull/63551) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Go back to upstream lz4. [#63574](https://github.com/ClickHouse/ClickHouse/pull/63574) ([Raúl Marín](https://github.com/Algunenano)). +* Fix logical error in ColumnTuple::tryInsert(). [#63583](https://github.com/ClickHouse/ClickHouse/pull/63583) ([Michael Kolupaev](https://github.com/al13n321)). +* harmonize sumMap error messages on ILLEGAL_TYPE_OF_ARGUMENT. [#63619](https://github.com/ClickHouse/ClickHouse/pull/63619) ([Yohann Jardin](https://github.com/yohannj)). +* Refactor data part writer to remove dependencies on MergeTreeData and DataPart. [#63620](https://github.com/ClickHouse/ClickHouse/pull/63620) ([Alexander Gololobov](https://github.com/davenger)). +* Update README.md. [#63631](https://github.com/ClickHouse/ClickHouse/pull/63631) ([Tyler Hannan](https://github.com/tylerhannan)). +* Ignore global profiler if system.trace_log is not enabled and fix really disable it for keeper standalone build. [#63632](https://github.com/ClickHouse/ClickHouse/pull/63632) ([Azat Khuzhin](https://github.com/azat)). +* Fixes for 00002_log_and_exception_messages_formatting. [#63634](https://github.com/ClickHouse/ClickHouse/pull/63634) ([Azat Khuzhin](https://github.com/azat)). +* Fix 02362_part_log_merge_algorithm flaky test. [#63635](https://github.com/ClickHouse/ClickHouse/pull/63635) ([Miсhael Stetsyuk](https://github.com/mstetsyuk)). +* Fix tests flakiness due to long SYSTEM FLUSH LOGS (explicitly specify old_parts_lifetime). [#63639](https://github.com/ClickHouse/ClickHouse/pull/63639) ([Azat Khuzhin](https://github.com/azat)). +* Update clickhouse-test help section. [#63663](https://github.com/ClickHouse/ClickHouse/pull/63663) ([Ali](https://github.com/xogoodnow)). +* Fix bad test `02950_part_log_bytes_uncompressed`. [#63672](https://github.com/ClickHouse/ClickHouse/pull/63672) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Remove leftovers of `optimize_monotonous_functions_in_order_by`. [#63674](https://github.com/ClickHouse/ClickHouse/pull/63674) ([Nikita Taranov](https://github.com/nickitat)). +* tests: attempt to fix 02340_parts_refcnt_mergetree flakiness. [#63684](https://github.com/ClickHouse/ClickHouse/pull/63684) ([Azat Khuzhin](https://github.com/azat)). +* Parallel replicas: simple cleanup. [#63685](https://github.com/ClickHouse/ClickHouse/pull/63685) ([Igor Nikonov](https://github.com/devcrafter)). +* Cancel S3 reads properly when parallel reads are used. [#63687](https://github.com/ClickHouse/ClickHouse/pull/63687) ([Antonio Andelic](https://github.com/antonio2368)). +* Explaining insertion order of the Map datatype. [#63690](https://github.com/ClickHouse/ClickHouse/pull/63690) ([Mark Needham](https://github.com/mneedham)). +* selectRangesToRead() simple cleanup. [#63692](https://github.com/ClickHouse/ClickHouse/pull/63692) ([Igor Nikonov](https://github.com/devcrafter)). +* Fix fuzzed analyzer_join_with_constant query. [#63702](https://github.com/ClickHouse/ClickHouse/pull/63702) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Add missing explicit instantiations of ColumnUnique. [#63718](https://github.com/ClickHouse/ClickHouse/pull/63718) ([Raúl Marín](https://github.com/Algunenano)). +* Better asserts in ColumnString.h. [#63719](https://github.com/ClickHouse/ClickHouse/pull/63719) ([Raúl Marín](https://github.com/Algunenano)). +* Try to fix flaky s3 tests test_seekable_formats and test_seekable_formats_url. [#63720](https://github.com/ClickHouse/ClickHouse/pull/63720) ([Kruglov Pavel](https://github.com/Avogar)). +* Don't randomize some settings in 02941_variant_type_* tests to avoid timeouts. [#63721](https://github.com/ClickHouse/ClickHouse/pull/63721) ([Kruglov Pavel](https://github.com/Avogar)). +* Fix flaky 03145_non_loaded_projection_backup.sh. [#63728](https://github.com/ClickHouse/ClickHouse/pull/63728) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Userspace page cache: don't collect stats if cache is unused. [#63730](https://github.com/ClickHouse/ClickHouse/pull/63730) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix insignificant UBSAN error in QueryAnalyzer::replaceNodesWithPositionalArguments(). [#63734](https://github.com/ClickHouse/ClickHouse/pull/63734) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix a bug in resolving matcher inside lambda inside ARRAY JOIN. [#63744](https://github.com/ClickHouse/ClickHouse/pull/63744) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Self explanatory. [#63754](https://github.com/ClickHouse/ClickHouse/pull/63754) ([Arthur Passos](https://github.com/arthurpassos)). +* Do not hide disk name. [#63756](https://github.com/ClickHouse/ClickHouse/pull/63756) ([Kseniia Sumarokova](https://github.com/kssenii)). +* CI: remove Cancel and Debug workflows as redundant. [#63757](https://github.com/ClickHouse/ClickHouse/pull/63757) ([Max K.](https://github.com/maxknv)). +* Security Policy: Add notification process. [#63773](https://github.com/ClickHouse/ClickHouse/pull/63773) ([Leticia Webb](https://github.com/leticiawebb)). +* Fix typo. [#63774](https://github.com/ClickHouse/ClickHouse/pull/63774) ([Anton Popov](https://github.com/CurtizJ)). +* Fix fuzzer when only explicit faults are used. [#63775](https://github.com/ClickHouse/ClickHouse/pull/63775) ([Raúl Marín](https://github.com/Algunenano)). +* Settings typo. [#63782](https://github.com/ClickHouse/ClickHouse/pull/63782) ([Rory Crispin](https://github.com/RoryCrispin)). +* Ref. [#63479](https://github.com/ClickHouse/ClickHouse/issues/63479). [#63783](https://github.com/ClickHouse/ClickHouse/pull/63783) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Fix test_odbc_interaction from aarch64 [#61457](https://github.com/ClickHouse/ClickHouse/issues/61457). [#63787](https://github.com/ClickHouse/ClickHouse/pull/63787) ([alesapin](https://github.com/alesapin)). +* Fix test `test_catboost_evaluate` for aarch64. [#61457](https://github.com/ClickHouse/ClickHouse/issues/61457). [#63789](https://github.com/ClickHouse/ClickHouse/pull/63789) ([alesapin](https://github.com/alesapin)). +* Rewrite plan for parallel replicas in Planner. [#63796](https://github.com/ClickHouse/ClickHouse/pull/63796) ([Igor Nikonov](https://github.com/devcrafter)). +* Follow-up for the `binary_symbols` table in CI. [#63802](https://github.com/ClickHouse/ClickHouse/pull/63802) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Support INSERT with VALUES in the ANTLR syntax file. [#63811](https://github.com/ClickHouse/ClickHouse/pull/63811) ([GG Bond](https://github.com/zzyReal666)). +* Fix race in `ReplicatedMergeTreeLogEntryData`. [#63816](https://github.com/ClickHouse/ClickHouse/pull/63816) ([Antonio Andelic](https://github.com/antonio2368)). +* Allow allocation during job destructor in `ThreadPool`. [#63829](https://github.com/ClickHouse/ClickHouse/pull/63829) ([Antonio Andelic](https://github.com/antonio2368)). +* Remove HDFS from disks config for one integration test for arm. [#61457](https://github.com/ClickHouse/ClickHouse/issues/61457). [#63832](https://github.com/ClickHouse/ClickHouse/pull/63832) ([alesapin](https://github.com/alesapin)). +* io_uring: add basic io_uring clickhouse perf test. [#63835](https://github.com/ClickHouse/ClickHouse/pull/63835) ([Tomer Shafir](https://github.com/tomershafir)). +* Bump version for old image in test_short_strings_aggregation to make it work on arm. [#61457](https://github.com/ClickHouse/ClickHouse/issues/61457). [#63836](https://github.com/ClickHouse/ClickHouse/pull/63836) ([alesapin](https://github.com/alesapin)). +* fix typo. [#63838](https://github.com/ClickHouse/ClickHouse/pull/63838) ([Alexander Gololobov](https://github.com/davenger)). +* Disable test `test_non_default_compression/test.py::test_preconfigured_deflateqpl_codec` on arm. [#61457](https://github.com/ClickHouse/ClickHouse/issues/61457). [#63839](https://github.com/ClickHouse/ClickHouse/pull/63839) ([alesapin](https://github.com/alesapin)). +* This PR was reverted. [#63857](https://github.com/ClickHouse/ClickHouse/pull/63857) ([Sema Checherinda](https://github.com/CheSema)). +* Remove unnecessary logging statements in MergeJoinTransform.cpp. [#63860](https://github.com/ClickHouse/ClickHouse/pull/63860) ([vdimir](https://github.com/vdimir)). +* Temporary disables 3 integration tcs on arm until https://github.com/clickhouse/clickhouse/issues/63855 is resolved. [#63867](https://github.com/ClickHouse/ClickHouse/pull/63867) ([Max K.](https://github.com/maxknv)). +* Fix some settings values in 02455_one_row_from_csv_memory_usage test to make it less flaky. [#63874](https://github.com/ClickHouse/ClickHouse/pull/63874) ([Kruglov Pavel](https://github.com/Avogar)). +* Randomise `allow_experimental_parallel_reading_from_replicas` in stress tests. [#63899](https://github.com/ClickHouse/ClickHouse/pull/63899) ([Nikita Taranov](https://github.com/nickitat)). +* Fix logs test for binary data by converting it to a valid UTF8 string. [#63909](https://github.com/ClickHouse/ClickHouse/pull/63909) ([Alexey Katsman](https://github.com/alexkats)). +* More sanity checks for parallel replicas. [#63910](https://github.com/ClickHouse/ClickHouse/pull/63910) ([Nikita Taranov](https://github.com/nickitat)). +* Include checks like `Stateless tests (asan, distributed cache, meta storage in keeper, s3 storage) [2/3]` in `Mergeable Check` and `A Sync`. [#63945](https://github.com/ClickHouse/ClickHouse/pull/63945) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Insignificant libunwind build fixes. [#63946](https://github.com/ClickHouse/ClickHouse/pull/63946) ([Azat Khuzhin](https://github.com/azat)). +* Revert multiline pretty changes due to performance problems. [#63947](https://github.com/ClickHouse/ClickHouse/pull/63947) ([Raúl Marín](https://github.com/Algunenano)). +* Some usability improvements for c++expr script. [#63948](https://github.com/ClickHouse/ClickHouse/pull/63948) ([Azat Khuzhin](https://github.com/azat)). +* Fix 02124_insert_deduplication_token_multiple_blocks. [#63950](https://github.com/ClickHouse/ClickHouse/pull/63950) ([Han Fei](https://github.com/hanfei1991)). +* CI: aarch64: disable arm integration tests with kerberaized kafka. [#63961](https://github.com/ClickHouse/ClickHouse/pull/63961) ([Max K.](https://github.com/maxknv)). +* Make events like [timeouts](https://play.clickhouse.com/play?user=play#U0VMRUNUICogRlJPTSBjaGVja3MgV0hFUkUgdGVzdF9uYW1lID09ICdDaGVjayB0aW1lb3V0IGV4cGlyZWQnIEFORCBjaGVja19zdGFydF90aW1lIEJFVFdFRU4gdG9EYXRlKCcyMDI0LTA1LTEwJykgQU5EIHRvRGF0ZSgnMjAyNC0wNS0xNScp) visible in CI DB. [#63982](https://github.com/ClickHouse/ClickHouse/pull/63982) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Slightly better setting `force_optimize_projection_name`. [#63997](https://github.com/ClickHouse/ClickHouse/pull/63997) ([Anton Popov](https://github.com/CurtizJ)). +* chore(ci-workers): remove reusable from tailscale key. [#63999](https://github.com/ClickHouse/ClickHouse/pull/63999) ([Gabriel Martinez](https://github.com/GMartinez-Sisti)). +* Better script to collect symbols statistics. [#64013](https://github.com/ClickHouse/ClickHouse/pull/64013) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix a typo in Analyzer. [#64022](https://github.com/ClickHouse/ClickHouse/pull/64022) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix libbcrypt for FreeBSD build. [#64023](https://github.com/ClickHouse/ClickHouse/pull/64023) ([Azat Khuzhin](https://github.com/azat)). +* Remove some unnecessary `UNREACHABLE`s. [#64035](https://github.com/ClickHouse/ClickHouse/pull/64035) ([Robert Schulze](https://github.com/rschu1ze)). +* Add `ClickHouseVersion.copy` method. Create a branch release in advance without spinning out the release to increase the stability. [#64039](https://github.com/ClickHouse/ClickHouse/pull/64039) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Fix searching for libclang_rt.builtins.*.a on FreeBSD. [#64051](https://github.com/ClickHouse/ClickHouse/pull/64051) ([Azat Khuzhin](https://github.com/azat)). +* The mime type is not 100% reliable for Python and shell scripts without shebangs; add a check for file extension. [#64062](https://github.com/ClickHouse/ClickHouse/pull/64062) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Fix waiting for mutations with retriable errors. [#64063](https://github.com/ClickHouse/ClickHouse/pull/64063) ([Alexander Tokmakov](https://github.com/tavplubix)). +* harmonize h3PointDist* error messages. [#64080](https://github.com/ClickHouse/ClickHouse/pull/64080) ([Yohann Jardin](https://github.com/yohannj)). +* This log message is better in Trace. [#64081](https://github.com/ClickHouse/ClickHouse/pull/64081) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Prevent stack overflow in Fuzzer and Stress test. [#64082](https://github.com/ClickHouse/ClickHouse/pull/64082) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* tests: fix expected error for 03036_reading_s3_archives (fixes CI). [#64089](https://github.com/ClickHouse/ClickHouse/pull/64089) ([Azat Khuzhin](https://github.com/azat)). +* Fix sanitizers. [#64090](https://github.com/ClickHouse/ClickHouse/pull/64090) ([Azat Khuzhin](https://github.com/azat)). +* Update llvm/clang to 18.1.6. [#64091](https://github.com/ClickHouse/ClickHouse/pull/64091) ([Azat Khuzhin](https://github.com/azat)). +* Set green Mergeable Check status only after all required checks are passed with success - All non-required checks are started at stage Test_3 when all required checks are passed in Test_1/2. [#64093](https://github.com/ClickHouse/ClickHouse/pull/64093) ([Max K.](https://github.com/maxknv)). +* Move `isAllASCII` from UTFHelper to StringUtils. [#64108](https://github.com/ClickHouse/ClickHouse/pull/64108) ([Robert Schulze](https://github.com/rschu1ze)). +* Throw out some `inline`s. [#64110](https://github.com/ClickHouse/ClickHouse/pull/64110) ([Robert Schulze](https://github.com/rschu1ze)). +* Clean up .clang-tidy after transition to Clang 18. [#64111](https://github.com/ClickHouse/ClickHouse/pull/64111) ([Robert Schulze](https://github.com/rschu1ze)). +* Ignore exception when checking for cgroupsv2. [#64118](https://github.com/ClickHouse/ClickHouse/pull/64118) ([Robert Schulze](https://github.com/rschu1ze)). +* Add retries in git submodule update. [#64125](https://github.com/ClickHouse/ClickHouse/pull/64125) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* See https://s3.amazonaws.com/clickhouse-test-reports/63946/86cf1e13d866333b8a511badd7f2fe186d810646/ast_fuzzer__ubsan_.html. [#64127](https://github.com/ClickHouse/ClickHouse/pull/64127) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Refactoring of Server.h: Isolate server management from other logic. [#64132](https://github.com/ClickHouse/ClickHouse/pull/64132) ([TTPO100AJIEX](https://github.com/TTPO100AJIEX)). +* Syncing code. [#64135](https://github.com/ClickHouse/ClickHouse/pull/64135) ([Antonio Andelic](https://github.com/antonio2368)). +* Losen build resource limits for unusual architectures. [#64152](https://github.com/ClickHouse/ClickHouse/pull/64152) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* fix clang tidy. [#64179](https://github.com/ClickHouse/ClickHouse/pull/64179) ([Han Fei](https://github.com/hanfei1991)). +* Fix: 02124_insert_deduplication_token_multiple_blocks_replica. [#64181](https://github.com/ClickHouse/ClickHouse/pull/64181) ([Igor Nikonov](https://github.com/devcrafter)). +* Fix global query profiler. [#64187](https://github.com/ClickHouse/ClickHouse/pull/64187) ([Azat Khuzhin](https://github.com/azat)). +* CI: cancel running PR wf after adding to MQ. [#64188](https://github.com/ClickHouse/ClickHouse/pull/64188) ([Max K.](https://github.com/maxknv)). +* Add profile events for number of rows read during/after prewhere. [#64198](https://github.com/ClickHouse/ClickHouse/pull/64198) ([Nikita Taranov](https://github.com/nickitat)). +* Add debug logging to EmbeddedRocksDBBulkSink. [#64203](https://github.com/ClickHouse/ClickHouse/pull/64203) ([vdimir](https://github.com/vdimir)). +* Fix special builds (due to excessive resource usage - memory/CPU). [#64204](https://github.com/ClickHouse/ClickHouse/pull/64204) ([Azat Khuzhin](https://github.com/azat)). +* Update InterpreterCreateQuery.cpp. [#64207](https://github.com/ClickHouse/ClickHouse/pull/64207) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Remove unused storage_snapshot field from MergeTreeSelectProcessor. [#64217](https://github.com/ClickHouse/ClickHouse/pull/64217) ([Alexander Gololobov](https://github.com/davenger)). +* Add test for [#37090](https://github.com/ClickHouse/ClickHouse/issues/37090). [#64220](https://github.com/ClickHouse/ClickHouse/pull/64220) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Small cli tool. [#64227](https://github.com/ClickHouse/ClickHouse/pull/64227) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Make `settings_changes_history` const. [#64230](https://github.com/ClickHouse/ClickHouse/pull/64230) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* test for [#45804](https://github.com/ClickHouse/ClickHouse/issues/45804). [#64245](https://github.com/ClickHouse/ClickHouse/pull/64245) ([Denny Crane](https://github.com/den-crane)). +* Update version after release. [#64283](https://github.com/ClickHouse/ClickHouse/pull/64283) ([Raúl Marín](https://github.com/Algunenano)). +* Followup for [#63691](https://github.com/ClickHouse/ClickHouse/issues/63691). [#64285](https://github.com/ClickHouse/ClickHouse/pull/64285) ([vdimir](https://github.com/vdimir)). +* CI: dependency fix for changelog.py. [#64293](https://github.com/ClickHouse/ClickHouse/pull/64293) ([Max K.](https://github.com/maxknv)). +* Print query in explain plan with parallel replicas. [#64298](https://github.com/ClickHouse/ClickHouse/pull/64298) ([vdimir](https://github.com/vdimir)). +* CI: Cancel sync wf on new push. [#64299](https://github.com/ClickHouse/ClickHouse/pull/64299) ([Max K.](https://github.com/maxknv)). +* CI: master workflow with folded jobs. [#64340](https://github.com/ClickHouse/ClickHouse/pull/64340) ([Max K.](https://github.com/maxknv)). +* CI: Sync, Merge check, CI gh's statuses fixes. [#64348](https://github.com/ClickHouse/ClickHouse/pull/64348) ([Max K.](https://github.com/maxknv)). +* Enable 02494_query_cache_nested_query_bug for Analyzer. [#64357](https://github.com/ClickHouse/ClickHouse/pull/64357) ([Robert Schulze](https://github.com/rschu1ze)). +* Rename allow_deprecated_functions to allow_deprecated_error_prone_window_functions. [#64358](https://github.com/ClickHouse/ClickHouse/pull/64358) ([Raúl Marín](https://github.com/Algunenano)). +* Change input_format_parquet_use_native_reader to 24.6. [#64359](https://github.com/ClickHouse/ClickHouse/pull/64359) ([Raúl Marín](https://github.com/Algunenano)). +* Update description for settings `cross_join_min_rows_to_compress` and `cross_join_min_bytes_to_compress`. [#64360](https://github.com/ClickHouse/ClickHouse/pull/64360) ([Nikita Fomichev](https://github.com/fm4v)). +* Changed the unreleased setting `aggregate_function_group_array_has_limit_size` to `aggregate_function_group_array_action_when_limit_is_reached`. [#64362](https://github.com/ClickHouse/ClickHouse/pull/64362) ([Raúl Marín](https://github.com/Algunenano)). +* Split tests 03039_dynamic_all_merge_algorithms to avoid timeouts. [#64363](https://github.com/ClickHouse/ClickHouse/pull/64363) ([Kruglov Pavel](https://github.com/Avogar)). +* Try to fix GWPAsan. [#64365](https://github.com/ClickHouse/ClickHouse/pull/64365) ([Antonio Andelic](https://github.com/antonio2368)). +* CI: add secrets to reusable stage wf yml. [#64366](https://github.com/ClickHouse/ClickHouse/pull/64366) ([Max K.](https://github.com/maxknv)). +* Do not run tests tagged 'no-s3-storage-with-slow-build' with ASan. [#64367](https://github.com/ClickHouse/ClickHouse/pull/64367) ([vdimir](https://github.com/vdimir)). +* This change was reverted. [#64386](https://github.com/ClickHouse/ClickHouse/pull/64386) ([Francisco J. Jurado Moreno](https://github.com/Beetelbrox)). +* Update s3queue.md. [#64389](https://github.com/ClickHouse/ClickHouse/pull/64389) ([Kseniia Sumarokova](https://github.com/kssenii)). +* test for [#64211](https://github.com/ClickHouse/ClickHouse/issues/64211). [#64390](https://github.com/ClickHouse/ClickHouse/pull/64390) ([Denny Crane](https://github.com/den-crane)). +* Follow-up to [#59767](https://github.com/ClickHouse/ClickHouse/issues/59767). [#64398](https://github.com/ClickHouse/ClickHouse/pull/64398) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Remove wrong comment. [#64403](https://github.com/ClickHouse/ClickHouse/pull/64403) ([Sergei Trifonov](https://github.com/serxa)). +* Follow up to [#59767](https://github.com/ClickHouse/ClickHouse/issues/59767). [#64404](https://github.com/ClickHouse/ClickHouse/pull/64404) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Refactor s3 settings (move settings parsing into single place). [#64412](https://github.com/ClickHouse/ClickHouse/pull/64412) ([Kseniia Sumarokova](https://github.com/kssenii)). +* This PR was reverted. [#64423](https://github.com/ClickHouse/ClickHouse/pull/64423) ([Sergei Trifonov](https://github.com/serxa)). +* Fix test after [#64404](https://github.com/ClickHouse/ClickHouse/issues/64404). [#64432](https://github.com/ClickHouse/ClickHouse/pull/64432) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Faster TestKeeper shutdown. [#64433](https://github.com/ClickHouse/ClickHouse/pull/64433) ([Alexander Gololobov](https://github.com/davenger)). +* Remove some logging. [#64434](https://github.com/ClickHouse/ClickHouse/pull/64434) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Revert "Revert "Remove some unnecessary UNREACHABLEs"". [#64435](https://github.com/ClickHouse/ClickHouse/pull/64435) ([Robert Schulze](https://github.com/rschu1ze)). +* Clean settings in 02943_variant_read_subcolumns test. [#64437](https://github.com/ClickHouse/ClickHouse/pull/64437) ([Kruglov Pavel](https://github.com/Avogar)). +* Add a comment after [#64226](https://github.com/ClickHouse/ClickHouse/issues/64226). [#64449](https://github.com/ClickHouse/ClickHouse/pull/64449) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* CI: fix build_report selection in case of job reuse. [#64459](https://github.com/ClickHouse/ClickHouse/pull/64459) ([Max K.](https://github.com/maxknv)). +* Add Critical bugfix category in PR template. [#64480](https://github.com/ClickHouse/ClickHouse/pull/64480) ([Max K.](https://github.com/maxknv)). +* Remove `generateSnowflakeIDThreadMonotonic`. [#64499](https://github.com/ClickHouse/ClickHouse/pull/64499) ([Robert Schulze](https://github.com/rschu1ze)). +* Move analyzer attempt 2. [#64500](https://github.com/ClickHouse/ClickHouse/pull/64500) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Sync some code back from internal to public repository. [#64502](https://github.com/ClickHouse/ClickHouse/pull/64502) ([Robert Schulze](https://github.com/rschu1ze)). +* Remove `generateUUIDv7(NonMonotonic|ThreadMonotonic)` functions. [#64506](https://github.com/ClickHouse/ClickHouse/pull/64506) ([Robert Schulze](https://github.com/rschu1ze)). +* Fix bash completion for settings. [#64521](https://github.com/ClickHouse/ClickHouse/pull/64521) ([Azat Khuzhin](https://github.com/azat)). +* Use max_read_buffer_size for file descriptors as well in file(). [#64532](https://github.com/ClickHouse/ClickHouse/pull/64532) ([Azat Khuzhin](https://github.com/azat)). +* Temporarily disable `enable_vertical_final` setting by default. This feature should not be used in older releases because it [might crash](https://github.com/ClickHouse/ClickHouse/issues/64543), but it's already fixed in 24.6 where this setting change has been reverted and `enable_vertical_final` is again enabled by default. [#64544](https://github.com/ClickHouse/ClickHouse/pull/64544) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Removed excessive calls to `flush logs` and disabled under sanitizers. [#64550](https://github.com/ClickHouse/ClickHouse/pull/64550) ([Nikita Taranov](https://github.com/nickitat)). +* Sync code moved in private repo back back to public repo. [#64551](https://github.com/ClickHouse/ClickHouse/pull/64551) ([Robert Schulze](https://github.com/rschu1ze)). +* Add support for custom type to ASTLiteral, or else the type may be lost when parse the ast. E.g. set a ASTLiteral to DataTime32 with value 19870, then it will be parsed to Int16. [#64562](https://github.com/ClickHouse/ClickHouse/pull/64562) ([shuai.xu](https://github.com/shuai-xu)). +* Add a temporary known host for git over ssh. [#64569](https://github.com/ClickHouse/ClickHouse/pull/64569) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Cache first analysis result in ReadFromMergeTree. [#64579](https://github.com/ClickHouse/ClickHouse/pull/64579) ([Igor Nikonov](https://github.com/devcrafter)). +* Derive script parameters (labels) from the --repo/--from-repo - fix to not create backports for all release branches if backport for specific branch only. [#64603](https://github.com/ClickHouse/ClickHouse/pull/64603) ([Max K.](https://github.com/maxknv)). +* CI fixes. [#64605](https://github.com/ClickHouse/ClickHouse/pull/64605) ([Max K.](https://github.com/maxknv)). +* Double-checking [#59318](https://github.com/ClickHouse/ClickHouse/issues/59318) and docs for `Map`. [#64606](https://github.com/ClickHouse/ClickHouse/pull/64606) ([Robert Schulze](https://github.com/rschu1ze)). +* Update CHANGELOG.md. [#64609](https://github.com/ClickHouse/ClickHouse/pull/64609) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Tests: Convert numeric to symbolic error codes. [#64635](https://github.com/ClickHouse/ClickHouse/pull/64635) ([Robert Schulze](https://github.com/rschu1ze)). +* Move NamedCollectionsFactory into a separate file. [#64642](https://github.com/ClickHouse/ClickHouse/pull/64642) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Shuffle tests for parallel execution. [#64646](https://github.com/ClickHouse/ClickHouse/pull/64646) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* CI: Do not upload binaries for special builds in PRs. [#64653](https://github.com/ClickHouse/ClickHouse/pull/64653) ([Max K.](https://github.com/maxknv)). +* Update changelog. [#64654](https://github.com/ClickHouse/ClickHouse/pull/64654) ([Robert Schulze](https://github.com/rschu1ze)). +* Parallel replicas: simple cleanup. [#64655](https://github.com/ClickHouse/ClickHouse/pull/64655) ([Igor Nikonov](https://github.com/devcrafter)). +* Be more graceful with existing tables with `inverted` indexes. [#64656](https://github.com/ClickHouse/ClickHouse/pull/64656) ([Robert Schulze](https://github.com/rschu1ze)). +* CI: Build Report Check to verify only enabled builds. [#64669](https://github.com/ClickHouse/ClickHouse/pull/64669) ([Max K.](https://github.com/maxknv)). +* Tests: Convert error numbers to symbolic error codes, pt. II. [#64670](https://github.com/ClickHouse/ClickHouse/pull/64670) ([Robert Schulze](https://github.com/rschu1ze)). +* Split query analyzer. [#64672](https://github.com/ClickHouse/ClickHouse/pull/64672) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* By the end of CI, CI_Running status must be SUCCESS or FAILURE never PENDING. [#64693](https://github.com/ClickHouse/ClickHouse/pull/64693) ([Max K.](https://github.com/maxknv)). +* The following list of merged PRs is not present in the release branch and was added to the changelog by mistake:. [#64704](https://github.com/ClickHouse/ClickHouse/pull/64704) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* CI: MergeQueue: add binary_release and unit tests. [#64705](https://github.com/ClickHouse/ClickHouse/pull/64705) ([Max K.](https://github.com/maxknv)). +* Fix to get first good enough GH token instead of getting and comparing all of them. [#64709](https://github.com/ClickHouse/ClickHouse/pull/64709) ([Max K.](https://github.com/maxknv)). +* Check for missing Upload ID in CreateMultipartUpload reply. [#64714](https://github.com/ClickHouse/ClickHouse/pull/64714) ([Michael Kolupaev](https://github.com/al13n321)). +* Update version_date.tsv and changelogs after v24.5.1.1763-stable. [#64715](https://github.com/ClickHouse/ClickHouse/pull/64715) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Fix (unreleased) `loop()` table function crashing on empty table name. [#64716](https://github.com/ClickHouse/ClickHouse/pull/64716) ([Michael Kolupaev](https://github.com/al13n321)). +* Update CHANGELOG.md. [#64730](https://github.com/ClickHouse/ClickHouse/pull/64730) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* CI: ci.py refactoring. [#64734](https://github.com/ClickHouse/ClickHouse/pull/64734) ([Max K.](https://github.com/maxknv)). +* Return the explanation for session moved error. [#64747](https://github.com/ClickHouse/ClickHouse/pull/64747) ([Antonio Andelic](https://github.com/antonio2368)). +* Adjust the version_helper and script to a new release scheme. [#64759](https://github.com/ClickHouse/ClickHouse/pull/64759) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Do not try to write columns.txt if it does not exist for write-once storages. [#64762](https://github.com/ClickHouse/ClickHouse/pull/64762) ([Azat Khuzhin](https://github.com/azat)). +* Update 02482_load_parts_refcounts.sh. [#64765](https://github.com/ClickHouse/ClickHouse/pull/64765) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Fix crash with DISTINCT and window functions. [#64767](https://github.com/ClickHouse/ClickHouse/pull/64767) ([Igor Nikonov](https://github.com/devcrafter)). +* Fix assert in IObjectStorageIteratorAsync. [#64770](https://github.com/ClickHouse/ClickHouse/pull/64770) ([Michael Kolupaev](https://github.com/al13n321)). +* Make table functions always report engine 'StorageProxy' in system.tables. [#64771](https://github.com/ClickHouse/ClickHouse/pull/64771) ([Michael Kolupaev](https://github.com/al13n321)). +* Ask about company name on GitHub. [#64774](https://github.com/ClickHouse/ClickHouse/pull/64774) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix flaky tests about SQLite. [#64776](https://github.com/ClickHouse/ClickHouse/pull/64776) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Remove iostream debug helpers. [#64777](https://github.com/ClickHouse/ClickHouse/pull/64777) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Remove unnecessary comment. [#64785](https://github.com/ClickHouse/ClickHouse/pull/64785) ([Raúl Marín](https://github.com/Algunenano)). +* Follow-ups to some PRs. [#64787](https://github.com/ClickHouse/ClickHouse/pull/64787) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Attempt to fix 02228_merge_tree_insert_memory_usage.sql flakiness for s3. [#64800](https://github.com/ClickHouse/ClickHouse/pull/64800) ([Raúl Marín](https://github.com/Algunenano)). +* Add regression test for filter propagation through `Merge` engine. [#64806](https://github.com/ClickHouse/ClickHouse/pull/64806) ([Nikita Taranov](https://github.com/nickitat)). +* Migrate changelog.py to a descendant of fuzzywuzzy. [#64807](https://github.com/ClickHouse/ClickHouse/pull/64807) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* A follow-up for https://github.com/ClickHouse/ClickHouse/pull/64039 and [#64759](https://github.com/ClickHouse/ClickHouse/issues/64759). [#64813](https://github.com/ClickHouse/ClickHouse/pull/64813) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Make row order optimization non-experimental. [#64814](https://github.com/ClickHouse/ClickHouse/pull/64814) ([Robert Schulze](https://github.com/rschu1ze)). +* Didn't catch it at the time when all versions belonged to the current year. [#64817](https://github.com/ClickHouse/ClickHouse/pull/64817) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Fix clang-tidy build. [#64823](https://github.com/ClickHouse/ClickHouse/pull/64823) ([Robert Schulze](https://github.com/rschu1ze)). +* Sets all builds that we run tests on to normal build list. [#64824](https://github.com/ClickHouse/ClickHouse/pull/64824) ([Max K.](https://github.com/maxknv)). +* CI: fix CI await feature. [#64825](https://github.com/ClickHouse/ClickHouse/pull/64825) ([Max K.](https://github.com/maxknv)). +* Fix clang-tidy. [#64827](https://github.com/ClickHouse/ClickHouse/pull/64827) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Upload blob_storage_log from stateless tests. [#64843](https://github.com/ClickHouse/ClickHouse/pull/64843) ([alesapin](https://github.com/alesapin)). +* Follow-up to [#64349](https://github.com/ClickHouse/ClickHouse/issues/64349). [#64845](https://github.com/ClickHouse/ClickHouse/pull/64845) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Simplify handling of old 'inverted' indexes. [#64846](https://github.com/ClickHouse/ClickHouse/pull/64846) ([Robert Schulze](https://github.com/rschu1ze)). +* Use issue templates defined in YAML provide more user-friendly experience. [#64850](https://github.com/ClickHouse/ClickHouse/pull/64850) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Handle logs from rocksdb by ClickHouse internal logging. [#64856](https://github.com/ClickHouse/ClickHouse/pull/64856) ([Azat Khuzhin](https://github.com/azat)). +* Follow-up for https://github.com/ClickHouse/ClickHouse/pull/59357. [#64860](https://github.com/ClickHouse/ClickHouse/pull/64860) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* added mlock and mlockall to aspell-dict to be ignored. [#64863](https://github.com/ClickHouse/ClickHouse/pull/64863) ([Ali](https://github.com/xogoodnow)). +* A tiny fix for fancy quotes. [#64883](https://github.com/ClickHouse/ClickHouse/pull/64883) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix possible loss of "Query was cancelled" message in client. [#64888](https://github.com/ClickHouse/ClickHouse/pull/64888) ([Azat Khuzhin](https://github.com/azat)). +* We accidentally lost the way to set `PR Check` failure at some point. [#64890](https://github.com/ClickHouse/ClickHouse/pull/64890) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Fix global trace collector. [#64896](https://github.com/ClickHouse/ClickHouse/pull/64896) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix test_mask_sensitive_info/test.py::test_create_table. [#64901](https://github.com/ClickHouse/ClickHouse/pull/64901) ([Azat Khuzhin](https://github.com/azat)). +* Update 03165_string_functions_with_token_text_indexes.sql. [#64903](https://github.com/ClickHouse/ClickHouse/pull/64903) ([Alexander Tokmakov](https://github.com/tavplubix)). +* When the branch is removed, it's impossible to get the diff by the labels. `print` in imported files spoils the `ipython` output. [#64904](https://github.com/ClickHouse/ClickHouse/pull/64904) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Disable transactions for unsupported storages even for materialized v…. [#64918](https://github.com/ClickHouse/ClickHouse/pull/64918) ([alesapin](https://github.com/alesapin)). +* additional log for cleanupDetachedTables. [#64919](https://github.com/ClickHouse/ClickHouse/pull/64919) ([Konstantin Morozov](https://github.com/k-morozov)). +* Fix tupleConcat of two empty tuples. This fixes [#64885](https://github.com/ClickHouse/ClickHouse/issues/64885). [#64923](https://github.com/ClickHouse/ClickHouse/pull/64923) ([Amos Bird](https://github.com/amosbird)). +* CI: Minor fixes in ci scripts. [#64950](https://github.com/ClickHouse/ClickHouse/pull/64950) ([Max K.](https://github.com/maxknv)). +* Fix error message (it was strange). [#64952](https://github.com/ClickHouse/ClickHouse/pull/64952) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update fmtlib version to 9.1.0. [#64959](https://github.com/ClickHouse/ClickHouse/pull/64959) ([Duc Canh Le](https://github.com/canhld94)). +* Test 02908_many_requests_to_system_replicas makes a lot of heavy requests and it overloads server if it's an ASAN build. [#64966](https://github.com/ClickHouse/ClickHouse/pull/64966) ([Alexander Gololobov](https://github.com/davenger)). +* Fix (unreleased) bug in short circuit evaluation. [#64967](https://github.com/ClickHouse/ClickHouse/pull/64967) ([Raúl Marín](https://github.com/Algunenano)). +* Update version_date.tsv and changelogs after v24.4.2.141-stable. [#64968](https://github.com/ClickHouse/ClickHouse/pull/64968) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Fix `test_attach_partition_using_copy`. [#64977](https://github.com/ClickHouse/ClickHouse/pull/64977) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Faster processing of scheduler queue activations. [#64985](https://github.com/ClickHouse/ClickHouse/pull/64985) ([Sergei Trifonov](https://github.com/serxa)). +* CI: Fix nightly workflow. [#64987](https://github.com/ClickHouse/ClickHouse/pull/64987) ([Max K.](https://github.com/maxknv)). +* Fix innocuous data race in detectLanguage. [#64988](https://github.com/ClickHouse/ClickHouse/pull/64988) ([Raúl Marín](https://github.com/Algunenano)). +* CI: Builds in CI settings. [#64994](https://github.com/ClickHouse/ClickHouse/pull/64994) ([Max K.](https://github.com/maxknv)). +* REVERTED. [#65009](https://github.com/ClickHouse/ClickHouse/pull/65009) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* CI: Fix backports. [#65010](https://github.com/ClickHouse/ClickHouse/pull/65010) ([Max K.](https://github.com/maxknv)). +* Try fix 03143_prewhere_profile_events. [#65014](https://github.com/ClickHouse/ClickHouse/pull/65014) ([Nikita Taranov](https://github.com/nickitat)). +* Fix 03165_string_functions_with_token_text_indexes. [#65018](https://github.com/ClickHouse/ClickHouse/pull/65018) ([Julia Kartseva](https://github.com/jkartseva)). +* This change was reverted. [#65028](https://github.com/ClickHouse/ClickHouse/pull/65028) ([Sergei Trifonov](https://github.com/serxa)). +* Bump googletest to latest HEAD. [#65038](https://github.com/ClickHouse/ClickHouse/pull/65038) ([Robert Schulze](https://github.com/rschu1ze)). +* Improve comment about AsynchronousMetrics. [#65040](https://github.com/ClickHouse/ClickHouse/pull/65040) ([Antonio Andelic](https://github.com/antonio2368)). +* CI: Remove fuzzer build from normal CI run (bugfix). [#65041](https://github.com/ClickHouse/ClickHouse/pull/65041) ([Max K.](https://github.com/maxknv)). +* CI config refactoring. [#65045](https://github.com/ClickHouse/ClickHouse/pull/65045) ([Max K.](https://github.com/maxknv)). +* Bump abseil to latest HEAD. [#65048](https://github.com/ClickHouse/ClickHouse/pull/65048) ([Robert Schulze](https://github.com/rschu1ze)). +* Capture weak_ptr of ContextAccess for safety. [#65051](https://github.com/ClickHouse/ClickHouse/pull/65051) ([Alexander Gololobov](https://github.com/davenger)). +* Stateless tests: add test for SIZES_OF_NESTED_COLUMNS_ARE_INCONSISTENT. [#65056](https://github.com/ClickHouse/ClickHouse/pull/65056) ([Nikita Fomichev](https://github.com/fm4v)). +* Increase timeout in wait_for_all_mutations. [#65058](https://github.com/ClickHouse/ClickHouse/pull/65058) ([Alexander Gololobov](https://github.com/davenger)). +* Tests for _time virtual column in file alike storages. [#65064](https://github.com/ClickHouse/ClickHouse/pull/65064) ([Ilya Golshtein](https://github.com/ilejn)). +* Update odbc-bridge.md. [#65099](https://github.com/ClickHouse/ClickHouse/pull/65099) ([Alexander Gololobov](https://github.com/davenger)). +* Small fix for 02340_parts_refcnt_mergetree. [#65105](https://github.com/ClickHouse/ClickHouse/pull/65105) ([Nikita Taranov](https://github.com/nickitat)). +* Re-enable OpenSSL session caching. [#65111](https://github.com/ClickHouse/ClickHouse/pull/65111) ([Robert Schulze](https://github.com/rschu1ze)). +* Update test_replicated_database/test.py. [#65112](https://github.com/ClickHouse/ClickHouse/pull/65112) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Fix false positives leaky memory warnings in OpenSSL. [#65125](https://github.com/ClickHouse/ClickHouse/pull/65125) ([Robert Schulze](https://github.com/rschu1ze)). +* Fix `Initiator received more initial requests than there are replicas` with `loop` engine. [#65133](https://github.com/ClickHouse/ClickHouse/pull/65133) ([Nikita Taranov](https://github.com/nickitat)). +* Fix 'Tasks in BackgroundSchedulePool cannot throw' caused by MergeTreeData::loadUnexpectedDataParts(). [#65135](https://github.com/ClickHouse/ClickHouse/pull/65135) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix bad error message. [#65137](https://github.com/ClickHouse/ClickHouse/pull/65137) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Just fixing flaky unit tests. [#65152](https://github.com/ClickHouse/ClickHouse/pull/65152) ([Sema Checherinda](https://github.com/CheSema)). +* This change was reverted. [#65164](https://github.com/ClickHouse/ClickHouse/pull/65164) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Ensure submodules are named consistently. [#65167](https://github.com/ClickHouse/ClickHouse/pull/65167) ([Robert Schulze](https://github.com/rschu1ze)). +* Remove obsolete fix from aws submodule. [#65168](https://github.com/ClickHouse/ClickHouse/pull/65168) ([Robert Schulze](https://github.com/rschu1ze)). +* CI: Fix not-merged cherry-picks for backports. [#65181](https://github.com/ClickHouse/ClickHouse/pull/65181) ([Max K.](https://github.com/maxknv)). +* Add an assertion in ReplicatedMergeTreeQueue. [#65184](https://github.com/ClickHouse/ClickHouse/pull/65184) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Fix bug in unreleased code. [#65185](https://github.com/ClickHouse/ClickHouse/pull/65185) ([Raúl Marín](https://github.com/Algunenano)). +* Fix docs for skipping-indexes.md. [#65194](https://github.com/ClickHouse/ClickHouse/pull/65194) ([morning-color](https://github.com/morning-color)). +* Fix the descriptions of some server settings. [#65200](https://github.com/ClickHouse/ClickHouse/pull/65200) ([Raúl Marín](https://github.com/Algunenano)). +* Fix issue after [#64813](https://github.com/ClickHouse/ClickHouse/issues/64813) with broken search in the changelog, and missing zstd in a style-check image. [#65202](https://github.com/ClickHouse/ClickHouse/pull/65202) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Fix bug in unreleased code. [#65203](https://github.com/ClickHouse/ClickHouse/pull/65203) ([Raúl Marín](https://github.com/Algunenano)). +* Add test prewhere merge. [#65207](https://github.com/ClickHouse/ClickHouse/pull/65207) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Sync ProfileEvents.h. [#65208](https://github.com/ClickHouse/ClickHouse/pull/65208) ([Kseniia Sumarokova](https://github.com/kssenii)). +* FinishCheck to set failure if workflow failed. [#65228](https://github.com/ClickHouse/ClickHouse/pull/65228) ([Max K.](https://github.com/maxknv)). +* Update version_date.tsv and changelogs after v24.3.4.147-lts. [#65235](https://github.com/ClickHouse/ClickHouse/pull/65235) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Update version_date.tsv and changelogs after v24.5.3.5-stable. [#65240](https://github.com/ClickHouse/ClickHouse/pull/65240) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Fails sometimes for debug build https://s3.amazonaws.com/clickhouse-test-reports/0/af6afd904316bfb771737faa147ce8aea72dd705/stateless_tests__debug__[4_5].html. [#65245](https://github.com/ClickHouse/ClickHouse/pull/65245) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix libunwind in CI. [#65247](https://github.com/ClickHouse/ClickHouse/pull/65247) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* CI: Do not skip FinishCheck in Merge Queue. [#65249](https://github.com/ClickHouse/ClickHouse/pull/65249) ([Max K.](https://github.com/maxknv)). +* Add a test just in case. [#65271](https://github.com/ClickHouse/ClickHouse/pull/65271) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Disable 02581_share_big_sets_between_multiple_mutations_tasks_long in coverage run. [#65295](https://github.com/ClickHouse/ClickHouse/pull/65295) ([Alexander Gololobov](https://github.com/davenger)). +* Update version_date.tsv and changelogs after v23.8.15.35-lts. [#65300](https://github.com/ClickHouse/ClickHouse/pull/65300) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* mute test test_query_is_canceled_with_inf_retries. [#65301](https://github.com/ClickHouse/ClickHouse/pull/65301) ([Sema Checherinda](https://github.com/CheSema)). +* Fix silly typo that caused wrong tags messages. [#65307](https://github.com/ClickHouse/ClickHouse/pull/65307) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Save server data for failed stateless tests. [#65309](https://github.com/ClickHouse/ClickHouse/pull/65309) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Fix 01246_buffer_flush flakiness (by tuning timeouts). [#65310](https://github.com/ClickHouse/ClickHouse/pull/65310) ([Azat Khuzhin](https://github.com/azat)). +* Remove outdated override in stress tests. [#65323](https://github.com/ClickHouse/ClickHouse/pull/65323) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix bad code in `system.session_log`. [#65332](https://github.com/ClickHouse/ClickHouse/pull/65332) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* add tests for 'boom filter index with map'. [#65333](https://github.com/ClickHouse/ClickHouse/pull/65333) ([iceFireser](https://github.com/iceFireser)). +* Fix crash in 03036_dynamic_read_subcolumns. [#65341](https://github.com/ClickHouse/ClickHouse/pull/65341) ([Kruglov Pavel](https://github.com/Avogar)). +* Move tests 02942_variant_cast and 02944_variant_as_common_type to analyzer_tech_debt.txt. [#65342](https://github.com/ClickHouse/ClickHouse/pull/65342) ([Kruglov Pavel](https://github.com/Avogar)). +* REVERTED. [#65384](https://github.com/ClickHouse/ClickHouse/pull/65384) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* CI: Add Non-blocking (Woolen wolfdog) CI mode. [#65385](https://github.com/ClickHouse/ClickHouse/pull/65385) ([Max K.](https://github.com/maxknv)). +* Fix compatibility release check. [#65394](https://github.com/ClickHouse/ClickHouse/pull/65394) ([Alexey Katsman](https://github.com/alexkats)). +* Move a leaksan suppression from Poco into OpenSSL. [#65396](https://github.com/ClickHouse/ClickHouse/pull/65396) ([Robert Schulze](https://github.com/rschu1ze)). +* Fix tidy build. [#65415](https://github.com/ClickHouse/ClickHouse/pull/65415) ([Sergei Trifonov](https://github.com/serxa)). +* Remove Tests dependency on Builds_2. No tests depend on Builds_2. [#65416](https://github.com/ClickHouse/ClickHouse/pull/65416) ([Max K.](https://github.com/maxknv)). +* CI: PR workflow dependencies fix. [#65442](https://github.com/ClickHouse/ClickHouse/pull/65442) ([Max K.](https://github.com/maxknv)). +* Fix test_storage_s3_queue/test.py::test_max_set_age. [#65452](https://github.com/ClickHouse/ClickHouse/pull/65452) ([Kseniia Sumarokova](https://github.com/kssenii)). +* CI: Rename A Sync status. [#65456](https://github.com/ClickHouse/ClickHouse/pull/65456) ([Max K.](https://github.com/maxknv)). +* CI: Rename sync status. [#65464](https://github.com/ClickHouse/ClickHouse/pull/65464) ([Max K.](https://github.com/maxknv)). +* This change was reverted. [#65466](https://github.com/ClickHouse/ClickHouse/pull/65466) ([Sergei Trifonov](https://github.com/serxa)). +* Remove a feature wasn't part of any release yet. [#65480](https://github.com/ClickHouse/ClickHouse/pull/65480) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#65657](https://github.com/ClickHouse/ClickHouse/issues/65657): Fix of `PlanSquashingTransform`: pipeline stuck. [#65487](https://github.com/ClickHouse/ClickHouse/pull/65487) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Backported in [#65504](https://github.com/ClickHouse/ClickHouse/issues/65504): Fix bad test `02922_deduplication_with_zero_copy`. [#65492](https://github.com/ClickHouse/ClickHouse/pull/65492) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#65591](https://github.com/ClickHouse/ClickHouse/issues/65591): Setting `uniform_snowflake_conversion_functions` (not in any release yet) was replaced by setting `allow_deprecated_snowflake_conversion_functions`. The latter controls if the legacy snowflake conversion functions are available (by default, they are not). [#65522](https://github.com/ClickHouse/ClickHouse/pull/65522) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#65759](https://github.com/ClickHouse/ClickHouse/issues/65759): Renames Build report jobs. [#65554](https://github.com/ClickHouse/ClickHouse/pull/65554) ([Max K.](https://github.com/maxknv)). +* Backported in [#65773](https://github.com/ClickHouse/ClickHouse/issues/65773): `base64En/Decode64Url` --> `base64En/Decode64URL`. [#65760](https://github.com/ClickHouse/ClickHouse/pull/65760) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#65805](https://github.com/ClickHouse/ClickHouse/issues/65805): CI: Fix for Builds report job in backports and releases. [#65774](https://github.com/ClickHouse/ClickHouse/pull/65774) ([Max K.](https://github.com/maxknv)). + diff --git a/docs/changelogs/v24.6.2.17-stable.md b/docs/changelogs/v24.6.2.17-stable.md new file mode 100644 index 00000000000..820937f6291 --- /dev/null +++ b/docs/changelogs/v24.6.2.17-stable.md @@ -0,0 +1,26 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.6.2.17-stable (5710a8b5c0c) FIXME as compared to v24.6.1.4423-stable (dcced7c8478) + +#### New Feature +* Backported in [#66002](https://github.com/ClickHouse/ClickHouse/issues/66002): Add AzureQueue storage. [#65458](https://github.com/ClickHouse/ClickHouse/pull/65458) ([Kseniia Sumarokova](https://github.com/kssenii)). + +#### Improvement +* Backported in [#65898](https://github.com/ClickHouse/ClickHouse/issues/65898): Respect cgroup CPU limit in Keeper. [#65819](https://github.com/ClickHouse/ClickHouse/pull/65819) ([Antonio Andelic](https://github.com/antonio2368)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#65935](https://github.com/ClickHouse/ClickHouse/issues/65935): For queries that read from `PostgreSQL`, cancel the internal `PostgreSQL` query if the ClickHouse query is finished. Otherwise, `ClickHouse` query cannot be canceled until the internal `PostgreSQL` query is finished. [#65771](https://github.com/ClickHouse/ClickHouse/pull/65771) ([Maksim Kita](https://github.com/kitaisreal)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#65907](https://github.com/ClickHouse/ClickHouse/issues/65907): Fix bug with session closing in Keeper. [#65735](https://github.com/ClickHouse/ClickHouse/pull/65735) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#65962](https://github.com/ClickHouse/ClickHouse/issues/65962): Add missing workload identity changes. [#65848](https://github.com/ClickHouse/ClickHouse/pull/65848) ([SmitaRKulkarni](https://github.com/SmitaRKulkarni)). +* Backported in [#66033](https://github.com/ClickHouse/ClickHouse/issues/66033): Follow up to [#65046](https://github.com/ClickHouse/ClickHouse/issues/65046). [#65928](https://github.com/ClickHouse/ClickHouse/pull/65928) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#66076](https://github.com/ClickHouse/ClickHouse/issues/66076): Fix support of non-const scale arguments in rounding functions. [#65983](https://github.com/ClickHouse/ClickHouse/pull/65983) ([Mikhail Gorshkov](https://github.com/mgorshkov)). +* Backported in [#66017](https://github.com/ClickHouse/ClickHouse/issues/66017): Fix race in s3queue. [#65986](https://github.com/ClickHouse/ClickHouse/pull/65986) ([Kseniia Sumarokova](https://github.com/kssenii)). + diff --git a/docs/en/development/build-cross-riscv.md b/docs/en/development/build-cross-riscv.md index 759d97823e2..dd97b6081e8 100644 --- a/docs/en/development/build-cross-riscv.md +++ b/docs/en/development/build-cross-riscv.md @@ -11,7 +11,7 @@ This is for the case when you have Linux machine and want to use it to build `cl The cross-build for RISC-V 64 is based on the [Build instructions](../development/build.md), follow them first. -## Install Clang-16 +## Install Clang-18 Follow the instructions from https://apt.llvm.org/ for your Ubuntu or Debian setup or do ``` diff --git a/docs/en/development/developer-instruction.md b/docs/en/development/developer-instruction.md index 0a1fe58b16f..4b11fb71230 100644 --- a/docs/en/development/developer-instruction.md +++ b/docs/en/development/developer-instruction.md @@ -226,15 +226,59 @@ Other IDEs you can use are [Sublime Text](https://www.sublimetext.com/), [Visual ## Writing Code {#writing-code} -The description of ClickHouse architecture can be found here: https://clickhouse.com/docs/en/development/architecture/ +Below you can find some quick links which may be useful when writing code for ClickHouse: -The Code Style Guide: https://clickhouse.com/docs/en/development/style/ +- [ClickHouse architecture description](https://clickhouse.com/docs/en/development/architecture/). +- [The code style guide](https://clickhouse.com/docs/en/development/style/). +- [Adding third-party libraries](https://clickhouse.com/docs/en/development/contrib/#adding-third-party-libraries) +- [Writing tests](https://clickhouse.com/docs/en/development/tests/) +- [List of open issues](https://github.com/ClickHouse/ClickHouse/issues?q=is%3Aopen+is%3Aissue+label%3Ahacktoberfest) -Adding third-party libraries: https://clickhouse.com/docs/en/development/contrib/#adding-third-party-libraries +## Writing Documentation {#writing-documentation} -Writing tests: https://clickhouse.com/docs/en/development/tests/ +As part of every pull request which adds a new feature, it is necessary to write documentation for it. If you'd like to preview your documentation changes the instructions for how to build the documentation page locally are available in the README.md file [here](https://github.com/ClickHouse/clickhouse-docs). When adding a new function to ClickHouse you can use the template below as a guide: -List of tasks: https://github.com/ClickHouse/ClickHouse/issues?q=is%3Aopen+is%3Aissue+label%3Ahacktoberfest +```markdown +# newFunctionName + +A short description of the function goes here. It should describe briefly what it does and a typical usage case. + +**Syntax** + +\```sql +newFunctionName(arg1, arg2[, arg3]) +\``` + +**Arguments** + +- `arg1` — Description of the argument. [DataType](../data-types/float.md) +- `arg2` — Description of the argument. [DataType](../data-types/float.md) +- `arg3` — Description of optional argument (optional). [DataType](../data-types/float.md) + +**Implementation Details** + +A description of implementation details if relevant. + +**Returned value** + +- Returns {insert what the function returns here}. [DataType](../data-types/float.md) + +**Example** + +Query: + +\```sql +SELECT 'write your example query here'; +\``` + +Response: + +\```response +┌───────────────────────────────────┐ +│ the result of the query │ +└───────────────────────────────────┘ +\``` +``` ## Test Data {#test-data} diff --git a/docs/en/development/tests.md b/docs/en/development/tests.md index 8dff6f0ed1d..269995a1a96 100644 --- a/docs/en/development/tests.md +++ b/docs/en/development/tests.md @@ -28,7 +28,7 @@ run, for example, the test `01428_hash_set_nan_key`, change to the repository folder and run the following command: ``` -PATH=$PATH: tests/clickhouse-test 01428_hash_set_nan_key +PATH=:$PATH tests/clickhouse-test 01428_hash_set_nan_key ``` Test results (`stderr` and `stdout`) are written to files `01428_hash_set_nan_key.[stderr|stdout]` which diff --git a/docs/en/engines/table-engines/integrations/azureBlobStorage.md b/docs/en/engines/table-engines/integrations/azureBlobStorage.md index dfc27d6b8cf..bdf96832e9d 100644 --- a/docs/en/engines/table-engines/integrations/azureBlobStorage.md +++ b/docs/en/engines/table-engines/integrations/azureBlobStorage.md @@ -56,6 +56,15 @@ SELECT * FROM test_table; - `_size` — Size of the file in bytes. Type: `Nullable(UInt64)`. If the size is unknown, the value is `NULL`. - `_time` — Last modified time of the file. Type: `Nullable(DateTime)`. If the time is unknown, the value is `NULL`. +## Authentication + +Currently there are 3 ways to authenticate: +- `Managed Identity` - Can be used by providing an `endpoint`, `connection_string` or `storage_account_url`. +- `SAS Token` - Can be used by providing an `endpoint`, `connection_string` or `storage_account_url`. It is identified by presence of '?' in the url. +- `Workload Identity` - Can be used by providing an `endpoint` or `storage_account_url`. If `use_workload_identity` parameter is set in config, ([workload identity](https://github.com/Azure/azure-sdk-for-cpp/tree/main/sdk/identity/azure-identity#authenticate-azure-hosted-applications)) is used for authentication. + + + ## See also [Azure Blob Storage Table Function](/docs/en/sql-reference/table-functions/azureBlobStorage) diff --git a/docs/en/engines/table-engines/integrations/nats.md b/docs/en/engines/table-engines/integrations/nats.md index 9f7409a6893..78ce537224c 100644 --- a/docs/en/engines/table-engines/integrations/nats.md +++ b/docs/en/engines/table-engines/integrations/nats.md @@ -58,7 +58,7 @@ Optional parameters: - `nats_max_reconnect` – Maximum amount of reconnection attempts per try to connect to NATS. Default: `5`. - `nats_reconnect_wait` – Amount of time in milliseconds to sleep between each reconnect attempt. Default: `5000`. - `nats_server_list` - Server list for connection. Can be specified to connect to NATS cluster. -- `nats_skip_broken_messages` - NATS message parser tolerance to schema-incompatible messages per block. Default: `0`. If `nats_skip_broken_messages = N` then the engine skips *N* RabbitMQ messages that cannot be parsed (a message equals a row of data). +- `nats_skip_broken_messages` - NATS message parser tolerance to schema-incompatible messages per block. Default: `0`. If `nats_skip_broken_messages = N` then the engine skips *N* NATS messages that cannot be parsed (a message equals a row of data). - `nats_max_block_size` - Number of row collected by poll(s) for flushing data from NATS. Default: [max_insert_block_size](../../../operations/settings/settings.md#max_insert_block_size). - `nats_flush_interval_ms` - Timeout for flushing data read from NATS. Default: [stream_flush_interval_ms](../../../operations/settings/settings.md#stream-flush-interval-ms). - `nats_username` - NATS username. @@ -67,7 +67,7 @@ Optional parameters: - `nats_credential_file` - Path to a NATS credentials file. - `nats_startup_connect_tries` - Number of connect tries at startup. Default: `5`. - `nats_max_rows_per_message` — The maximum number of rows written in one NATS message for row-based formats. (default : `1`). -- `nats_handle_error_mode` — How to handle errors for RabbitMQ engine. Possible values: default (the exception will be thrown if we fail to parse a message), stream (the exception message and raw message will be saved in virtual columns `_error` and `_raw_message`). +- `nats_handle_error_mode` — How to handle errors for NATS engine. Possible values: default (the exception will be thrown if we fail to parse a message), stream (the exception message and raw message will be saved in virtual columns `_error` and `_raw_message`). SSL connection: diff --git a/docs/en/engines/table-engines/integrations/s3queue.md b/docs/en/engines/table-engines/integrations/s3queue.md index 97ca79501a7..06325fa15fb 100644 --- a/docs/en/engines/table-engines/integrations/s3queue.md +++ b/docs/en/engines/table-engines/integrations/s3queue.md @@ -28,6 +28,8 @@ CREATE TABLE s3_queue_engine_table (name String, value UInt32) [s3queue_cleanup_interval_max_ms = 30000,] ``` +Starting with `24.7` settings without `s3queue_` prefix are also supported. + **Engine parameters** - `path` — Bucket url with path to file. Supports following wildcards in readonly mode: `*`, `**`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc'`, `'def'` — strings. For more information see [below](#wildcards-in-path). @@ -73,7 +75,7 @@ SETTINGS Possible values: - unordered — With unordered mode, the set of all already processed files is tracked with persistent nodes in ZooKeeper. -- ordered — With ordered mode, only the max name of the successfully consumed file, and the names of files that will be retried after unsuccessful loading attempt are being stored in ZooKeeper. +- ordered — With ordered mode, the files are processed in lexicographic order. It means that if file named 'BBB' was processed at some point and later on a file named 'AA' is added to the bucket, it will be ignored. Only the max name (in lexicographic sense) of the successfully consumed file, and the names of files that will be retried after unsuccessful loading attempt are being stored in ZooKeeper. Default value: `ordered` in versions before 24.6. Starting with 24.6 there is no default value, the setting becomes required to be specified manually. For tables created on earlier versions the default value will remain `Ordered` for compatibility. diff --git a/docs/en/engines/table-engines/mergetree-family/mergetree.md b/docs/en/engines/table-engines/mergetree-family/mergetree.md index f0c4e1b0e34..7ffbd9a5bae 100644 --- a/docs/en/engines/table-engines/mergetree-family/mergetree.md +++ b/docs/en/engines/table-engines/mergetree-family/mergetree.md @@ -993,11 +993,15 @@ They can be used for prewhere optimization only if we enable `set allow_statisti - `TDigest` - Stores distribution of values from numeric columns in [TDigest](https://github.com/tdunning/t-digest) sketch. + [TDigest](https://github.com/tdunning/t-digest) sketches which allow to compute approximate percentiles (e.g. the 90th percentile) for numeric columns. - `Uniq` - - Estimate the number of distinct values of a column by HyperLogLog. + + [HyperLogLog](https://en.wikipedia.org/wiki/HyperLogLog) sketches which provide an estimation how many distinct values a column contains. + +- `count_min` + + [Count-min](https://en.wikipedia.org/wiki/Count%E2%80%93min_sketch) sketches which provide an approximate count of the frequency of each value in a column. ## Column-level Settings {#column-level-settings} diff --git a/docs/en/getting-started/example-datasets/images/stackoverflow.png b/docs/en/getting-started/example-datasets/images/stackoverflow.png new file mode 100644 index 00000000000..f31acdc8cc3 Binary files /dev/null and b/docs/en/getting-started/example-datasets/images/stackoverflow.png differ diff --git a/docs/en/getting-started/example-datasets/nypd_complaint_data.md b/docs/en/getting-started/example-datasets/nypd_complaint_data.md index a178fe456a6..1caa1ab6a9a 100644 --- a/docs/en/getting-started/example-datasets/nypd_complaint_data.md +++ b/docs/en/getting-started/example-datasets/nypd_complaint_data.md @@ -55,7 +55,7 @@ CMPLNT_FR_TM Nullable(String) ``` :::tip -Most of the time the above command will let you know which fields in the input data are numeric, and which are strings, and which are tuples. This is not always the case. Because ClickHouse is routineley used with datasets containing billions of records there is a default number (100) of rows examined to [infer the schema](/docs/en/integrations/data-ingestion/data-formats/json.md#relying-on-schema-inference) in order to avoid parsing billions of rows to infer the schema. The response below may not match what you see, as the dataset is updated several times each year. Looking at the Data Dictionary you can see that CMPLNT_NUM is specified as text, and not numeric. By overriding the default of 100 rows for inference with the setting `SETTINGS input_format_max_rows_to_read_for_schema_inference=2000` +Most of the time the above command will let you know which fields in the input data are numeric, and which are strings, and which are tuples. This is not always the case. Because ClickHouse is routineley used with datasets containing billions of records there is a default number (100) of rows examined to [infer the schema](/en/integrations/data-formats/json/inference) in order to avoid parsing billions of rows to infer the schema. The response below may not match what you see, as the dataset is updated several times each year. Looking at the Data Dictionary you can see that CMPLNT_NUM is specified as text, and not numeric. By overriding the default of 100 rows for inference with the setting `SETTINGS input_format_max_rows_to_read_for_schema_inference=2000` you can get a better idea of the content. Note: as of version 22.5 the default is now 25,000 rows for inferring the schema, so only change the setting if you are on an older version or if you need more than 25,000 rows to be sampled. diff --git a/docs/en/getting-started/example-datasets/stackoverflow.md b/docs/en/getting-started/example-datasets/stackoverflow.md new file mode 100644 index 00000000000..e982a3c3dfc --- /dev/null +++ b/docs/en/getting-started/example-datasets/stackoverflow.md @@ -0,0 +1,394 @@ +--- +slug: /en/getting-started/example-datasets/stackoverflow +sidebar_label: Stack Overflow +sidebar_position: 1 +description: Analyzing Stack Overflow data with ClickHouse +--- + +# Analyzing Stack Overflow data with ClickHouse + +This dataset contains every `Post`, `User`, `Vote`, `Comment`, `Badge, `PostHistory`, and `PostLink` that has occurred on Stack Overflow. + +Users can either download pre-prepared Parquet versions of the data, containing every post up to April 2024, or download the latest data in XML format and load this. Stack Overflow provide updates to this data periodically - historically every 3 months. + +The following diagram shows the schema for the available tables assuming Parquet format. + +![Stack Overflow schema](./images/stackoverflow.png) + +A description of the schema of this data can be found [here](https://meta.stackexchange.com/questions/2677/database-schema-documentation-for-the-public-data-dump-and-sede). + +## Pre-prepared data + +We provide a copy of this data in Parquet format, up to date as of April 2024. While small for ClickHouse with respect to the number of rows (60 million posts), this dataset contains significant volumes of text and large String columns. + +```sql +CREATE DATABASE stackoverflow +``` + +The following timings are for a 96 GiB, 24 vCPU ClickHouse Cloud cluster located in `eu-west-2`. The dataset is located in `eu-west-3`. + +### Posts + +```sql +CREATE TABLE stackoverflow.posts +( + `Id` Int32 CODEC(Delta(4), ZSTD(1)), + `PostTypeId` Enum8('Question' = 1, 'Answer' = 2, 'Wiki' = 3, 'TagWikiExcerpt' = 4, 'TagWiki' = 5, 'ModeratorNomination' = 6, 'WikiPlaceholder' = 7, 'PrivilegeWiki' = 8), + `AcceptedAnswerId` UInt32, + `CreationDate` DateTime64(3, 'UTC'), + `Score` Int32, + `ViewCount` UInt32 CODEC(Delta(4), ZSTD(1)), + `Body` String, + `OwnerUserId` Int32, + `OwnerDisplayName` String, + `LastEditorUserId` Int32, + `LastEditorDisplayName` String, + `LastEditDate` DateTime64(3, 'UTC') CODEC(Delta(8), ZSTD(1)), + `LastActivityDate` DateTime64(3, 'UTC'), + `Title` String, + `Tags` String, + `AnswerCount` UInt16 CODEC(Delta(2), ZSTD(1)), + `CommentCount` UInt8, + `FavoriteCount` UInt8, + `ContentLicense` LowCardinality(String), + `ParentId` String, + `CommunityOwnedDate` DateTime64(3, 'UTC'), + `ClosedDate` DateTime64(3, 'UTC') +) +ENGINE = MergeTree +PARTITION BY toYear(CreationDate) +ORDER BY (PostTypeId, toDate(CreationDate), CreationDate) + +INSERT INTO stackoverflow.posts SELECT * FROM s3('https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/posts/*.parquet') + +0 rows in set. Elapsed: 265.466 sec. Processed 59.82 million rows, 38.07 GB (225.34 thousand rows/s., 143.42 MB/s.) +``` + +Posts are also available by year e.g. [https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/posts/2020.parquet](https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/posts/2020.parquet) + + +### Votes + +```sql +CREATE TABLE stackoverflow.votes +( + `Id` UInt32, + `PostId` Int32, + `VoteTypeId` UInt8, + `CreationDate` DateTime64(3, 'UTC'), + `UserId` Int32, + `BountyAmount` UInt8 +) +ENGINE = MergeTree +ORDER BY (VoteTypeId, CreationDate, PostId, UserId) + +INSERT INTO stackoverflow.votes SELECT * FROM s3('https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/votes/*.parquet') + +0 rows in set. Elapsed: 21.605 sec. Processed 238.98 million rows, 2.13 GB (11.06 million rows/s., 98.46 MB/s.) +``` + +Votes are also available by year e.g. [https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/posts/2020.parquet](https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/votes/2020.parquet) + + +### Comments + +```sql +CREATE TABLE stackoverflow.comments +( + `Id` UInt32, + `PostId` UInt32, + `Score` UInt16, + `Text` String, + `CreationDate` DateTime64(3, 'UTC'), + `UserId` Int32, + `UserDisplayName` LowCardinality(String) +) +ENGINE = MergeTree +ORDER BY CreationDate + +INSERT INTO stackoverflow.comments SELECT * FROM s3('https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/comments/*.parquet') + +0 rows in set. Elapsed: 56.593 sec. Processed 90.38 million rows, 11.14 GB (1.60 million rows/s., 196.78 MB/s.) +``` + +Comments are also available by year e.g. [https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/posts/2020.parquet](https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/comments/2020.parquet) + +### Users + +```sql +CREATE TABLE stackoverflow.users +( + `Id` Int32, + `Reputation` LowCardinality(String), + `CreationDate` DateTime64(3, 'UTC') CODEC(Delta(8), ZSTD(1)), + `DisplayName` String, + `LastAccessDate` DateTime64(3, 'UTC'), + `AboutMe` String, + `Views` UInt32, + `UpVotes` UInt32, + `DownVotes` UInt32, + `WebsiteUrl` String, + `Location` LowCardinality(String), + `AccountId` Int32 +) +ENGINE = MergeTree +ORDER BY (Id, CreationDate) + +INSERT INTO stackoverflow.users SELECT * FROM s3('https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/users.parquet') + +0 rows in set. Elapsed: 10.988 sec. Processed 22.48 million rows, 1.36 GB (2.05 million rows/s., 124.10 MB/s.) +``` + +### Badges + +```sql +CREATE TABLE stackoverflow.badges +( + `Id` UInt32, + `UserId` Int32, + `Name` LowCardinality(String), + `Date` DateTime64(3, 'UTC'), + `Class` Enum8('Gold' = 1, 'Silver' = 2, 'Bronze' = 3), + `TagBased` Bool +) +ENGINE = MergeTree +ORDER BY UserId + +INSERT INTO stackoverflow.badges SELECT * FROM s3('https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/badges.parquet') + +0 rows in set. Elapsed: 6.635 sec. Processed 51.29 million rows, 797.05 MB (7.73 million rows/s., 120.13 MB/s.) +``` + +### `PostLinks` + +```sql +CREATE TABLE stackoverflow.postlinks +( + `Id` UInt64, + `CreationDate` DateTime64(3, 'UTC'), + `PostId` Int32, + `RelatedPostId` Int32, + `LinkTypeId` Enum8('Linked' = 1, 'Duplicate' = 3) +) +ENGINE = MergeTree +ORDER BY (PostId, RelatedPostId) + +INSERT INTO stackoverflow.postlinks SELECT * FROM s3('https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/postlinks.parquet') + +0 rows in set. Elapsed: 1.534 sec. Processed 6.55 million rows, 129.70 MB (4.27 million rows/s., 84.57 MB/s.) +``` + +### `PostHistory` + +```sql +CREATE TABLE stackoverflow.posthistory +( + `Id` UInt64, + `PostHistoryTypeId` UInt8, + `PostId` Int32, + `RevisionGUID` String, + `CreationDate` DateTime64(3, 'UTC'), + `UserId` Int32, + `Text` String, + `ContentLicense` LowCardinality(String), + `Comment` String, + `UserDisplayName` String +) +ENGINE = MergeTree +ORDER BY (CreationDate, PostId) + +INSERT INTO stackoverflow.posthistory SELECT * FROM s3('https://datasets-documentation.s3.eu-west-3.amazonaws.com/stackoverflow/parquet/posthistory/*.parquet') + +0 rows in set. Elapsed: 422.795 sec. Processed 160.79 million rows, 67.08 GB (380.30 thousand rows/s., 158.67 MB/s.) +``` + +## Original dataset + +The original dataset is available in compressed (7zip) XML format at [https://archive.org/download/stackexchange](https://archive.org/download/stackexchange) - files with prefix `stackoverflow.com*`. + +### Download + +```bash +wget https://archive.org/download/stackexchange/stackoverflow.com-Badges.7z +wget https://archive.org/download/stackexchange/stackoverflow.com-Comments.7z +wget https://archive.org/download/stackexchange/stackoverflow.com-PostHistory.7z +wget https://archive.org/download/stackexchange/stackoverflow.com-PostLinks.7z +wget https://archive.org/download/stackexchange/stackoverflow.com-Posts.7z +wget https://archive.org/download/stackexchange/stackoverflow.com-Users.7z +wget https://archive.org/download/stackexchange/stackoverflow.com-Votes.7z +``` + +These files are up to 35GB and can take around 30 mins to download depending on internet connection - the download server throttles at around 20MB/sec. + +### Convert to JSON + +At the time of writing, ClickHouse does not have native support for XML as an input format. To load the data into ClickHouse we first convert to NDJSON. + +To convert XML to JSON we recommend the [`xq`](https://github.com/kislyuk/yq) linux tool, a simple `jq` wrapper for XML documents. + +Install xq and jq: + +```bash +sudo apt install jq +pip install yq +``` + +The following steps apply to any of the above files. We use the `stackoverflow.com-Posts.7z` file as an example. Modify as required. + +Extract the file using [p7zip](https://p7zip.sourceforge.net/). This will produce a single xml file - in this case `Posts.xml`. + +> Files are compressed approximately 4.5x. At 22GB compressed, the posts file requires around 97G uncompressed. + +```bash +p7zip -d stackoverflow.com-Posts.7z +``` + +The following splits the xml file into files, each containing 10000 rows. + +```bash +mkdir posts +cd posts +# the following splits the input xml file into sub files of 10000 rows +tail +3 ../Posts.xml | head -n -1 | split -l 10000 --filter='{ printf "\n"; cat - ; printf "\n"; } > $FILE' - +``` + +After running the above users will have a set of files, each with 10000 lines. This ensures the memory overhead of the next command is not excessive (xml to JSON conversion is done in memory). + +```bash +find . -maxdepth 1 -type f -exec xq -c '.rows.row[]' {} \; | sed -e 's:"@:":g' > posts_v2.json +``` + +The above command will produce a single `posts.json` file. + +Load into ClickHouse with the following command. Note the schema is specified for the `posts.json` file. This will need to be adjusted per data type to align with the target table. + +```bash +clickhouse local --query "SELECT * FROM file('posts.json', JSONEachRow, 'Id Int32, PostTypeId UInt8, AcceptedAnswerId UInt32, CreationDate DateTime64(3, \'UTC\'), Score Int32, ViewCount UInt32, Body String, OwnerUserId Int32, OwnerDisplayName String, LastEditorUserId Int32, LastEditorDisplayName String, LastEditDate DateTime64(3, \'UTC\'), LastActivityDate DateTime64(3, \'UTC\'), Title String, Tags String, AnswerCount UInt16, CommentCount UInt8, FavoriteCount UInt8, ContentLicense String, ParentId String, CommunityOwnedDate DateTime64(3, \'UTC\'), ClosedDate DateTime64(3, \'UTC\')') FORMAT Native" | clickhouse client --host --secure --password --query "INSERT INTO stackoverflow.posts_v2 FORMAT Native" +``` + +## Example queries + +A few simple questions to you get started. + +### Most popular tags on Stack Overflow + +```sql + +SELECT + arrayJoin(arrayFilter(t -> (t != ''), splitByChar('|', Tags))) AS Tags, + count() AS c +FROM stackoverflow.posts +GROUP BY Tags +ORDER BY c DESC +LIMIT 10 + +┌─Tags───────┬───────c─┐ +│ javascript │ 2527130 │ +│ python │ 2189638 │ +│ java │ 1916156 │ +│ c# │ 1614236 │ +│ php │ 1463901 │ +│ android │ 1416442 │ +│ html │ 1186567 │ +│ jquery │ 1034621 │ +│ c++ │ 806202 │ +│ css │ 803755 │ +└────────────┴─────────┘ + +10 rows in set. Elapsed: 1.013 sec. Processed 59.82 million rows, 1.21 GB (59.07 million rows/s., 1.19 GB/s.) +Peak memory usage: 224.03 MiB. +``` + +### User with the most answers (active accounts) + +Account requires a `UserId`. + +```sql +SELECT + any(OwnerUserId) UserId, + OwnerDisplayName, + count() AS c +FROM stackoverflow.posts WHERE OwnerDisplayName != '' AND PostTypeId='Answer' AND OwnerUserId != 0 +GROUP BY OwnerDisplayName +ORDER BY c DESC +LIMIT 5 + +┌─UserId─┬─OwnerDisplayName─┬────c─┐ +│ 22656 │ Jon Skeet │ 2727 │ +│ 23354 │ Marc Gravell │ 2150 │ +│ 12950 │ tvanfosson │ 1530 │ +│ 3043 │ Joel Coehoorn │ 1438 │ +│ 10661 │ S.Lott │ 1087 │ +└────────┴──────────────────┴──────┘ + +5 rows in set. Elapsed: 0.154 sec. Processed 35.83 million rows, 193.39 MB (232.33 million rows/s., 1.25 GB/s.) +Peak memory usage: 206.45 MiB. +``` + +### ClickHouse related posts with the most views + +```sql +SELECT + Id, + Title, + ViewCount, + AnswerCount +FROM stackoverflow.posts +WHERE Title ILIKE '%ClickHouse%' +ORDER BY ViewCount DESC +LIMIT 10 + +┌───────Id─┬─Title────────────────────────────────────────────────────────────────────────────┬─ViewCount─┬─AnswerCount─┐ +│ 52355143 │ Is it possible to delete old records from clickhouse table? │ 41462 │ 3 │ +│ 37954203 │ Clickhouse Data Import │ 38735 │ 3 │ +│ 37901642 │ Updating data in Clickhouse │ 36236 │ 6 │ +│ 58422110 │ Pandas: How to insert dataframe into Clickhouse │ 29731 │ 4 │ +│ 63621318 │ DBeaver - Clickhouse - SQL Error [159] .. Read timed out │ 27350 │ 1 │ +│ 47591813 │ How to filter clickhouse table by array column contents? │ 27078 │ 2 │ +│ 58728436 │ How to search the string in query with case insensitive on Clickhouse database? │ 26567 │ 3 │ +│ 65316905 │ Clickhouse: DB::Exception: Memory limit (for query) exceeded │ 24899 │ 2 │ +│ 49944865 │ How to add a column in clickhouse │ 24424 │ 1 │ +│ 59712399 │ How to cast date Strings to DateTime format with extended parsing in ClickHouse? │ 22620 │ 1 │ +└──────────┴──────────────────────────────────────────────────────────────────────────────────┴───────────┴─────────────┘ + +10 rows in set. Elapsed: 0.472 sec. Processed 59.82 million rows, 1.91 GB (126.63 million rows/s., 4.03 GB/s.) +Peak memory usage: 240.01 MiB. +``` + +### Most controversial posts + +```sql +SELECT + Id, + Title, + UpVotes, + DownVotes, + abs(UpVotes - DownVotes) AS Controversial_ratio +FROM stackoverflow.posts +INNER JOIN +( + SELECT + PostId, + countIf(VoteTypeId = 2) AS UpVotes, + countIf(VoteTypeId = 3) AS DownVotes + FROM stackoverflow.votes + GROUP BY PostId + HAVING (UpVotes > 10) AND (DownVotes > 10) +) AS votes ON posts.Id = votes.PostId +WHERE Title != '' +ORDER BY Controversial_ratio ASC +LIMIT 3 + +┌───────Id─┬─Title─────────────────────────────────────────────┬─UpVotes─┬─DownVotes─┬─Controversial_ratio─┐ +│ 583177 │ VB.NET Infinite For Loop │ 12 │ 12 │ 0 │ +│ 9756797 │ Read console input as enumerable - one statement? │ 16 │ 16 │ 0 │ +│ 13329132 │ What's the point of ARGV in Ruby? │ 22 │ 22 │ 0 │ +└──────────┴───────────────────────────────────────────────────┴─────────┴───────────┴─────────────────────┘ + +3 rows in set. Elapsed: 4.779 sec. Processed 298.80 million rows, 3.16 GB (62.52 million rows/s., 661.05 MB/s.) +Peak memory usage: 6.05 GiB. +``` + +## Attribution + +We thank Stack Overflow for providing this data under the `cc-by-sa 4.0` license, acknowledging their efforts and the original source of the data at [https://archive.org/details/stackexchange](https://archive.org/details/stackexchange). diff --git a/docs/en/getting-started/install.md b/docs/en/getting-started/install.md index 98e73dec451..6209ef3c8ee 100644 --- a/docs/en/getting-started/install.md +++ b/docs/en/getting-started/install.md @@ -15,7 +15,7 @@ You have four options for getting up and running with ClickHouse: - **[ClickHouse Cloud](https://clickhouse.com/cloud/):** The official ClickHouse as a service, - built by, maintained and supported by the creators of ClickHouse - **[Quick Install](#quick-install):** an easy-to-download binary for testing and developing with ClickHouse -- **[Production Deployments](#available-installation-options):** ClickHouse can run on any Linux, FreeBSD, or macOS with x86-64, ARM, or PowerPC64LE CPU architecture +- **[Production Deployments](#available-installation-options):** ClickHouse can run on any Linux, FreeBSD, or macOS with x86-64, modern ARM (ARMv8.2-A up), or PowerPC64LE CPU architecture - **[Docker Image](https://hub.docker.com/r/clickhouse/clickhouse-server/):** use the official Docker image in Docker Hub ## ClickHouse Cloud diff --git a/docs/en/interfaces/cli.md b/docs/en/interfaces/cli.md index e18ff6f1a3f..849b3d4b486 100644 --- a/docs/en/interfaces/cli.md +++ b/docs/en/interfaces/cli.md @@ -185,6 +185,7 @@ You can pass parameters to `clickhouse-client` (all parameters have a default va - `--format, -f` – Use the specified default format to output the result. - `--vertical, -E` – If specified, use the [Vertical format](../interfaces/formats.md#vertical) by default to output the result. This is the same as `–format=Vertical`. In this format, each value is printed on a separate line, which is helpful when displaying wide tables. - `--time, -t` – If specified, print the query execution time to ‘stderr’ in non-interactive mode. +- `--memory-usage` – If specified, print memory usage to ‘stderr’ in non-interactive mode]. Possible values: 'none' - do not print memory usage, 'default' - print number of bytes, 'readable' - print memory usage in human-readable format. - `--stacktrace` – If specified, also print the stack trace if an exception occurs. - `--config-file` – The name of the configuration file. - `--secure` – If specified, will connect to server over secure connection (TLS). You might need to configure your CA certificates in the [configuration file](#configuration_files). The available configuration settings are the same as for [server-side TLS configuration](../operations/server-configuration-parameters/settings.md#server_configuration_parameters-openssl). @@ -339,7 +340,7 @@ clickhouse-client clickhouse://some_user%40some_mail.com@localhost:9000 Connect to one of provides hosts: `192.168.1.15`, `192.168.1.25`. ``` bash -clickhouse-client clickhouse://192.168.1.15,192.168.1.25 +clickhouse-client clickhouse://192.168.1.15,192.168.1.25 ``` ### Configuration Files {#configuration_files} @@ -367,7 +368,7 @@ Example of a config file: ``` Or the same config in a YAML format: - + ```yaml user: username password: 'password' diff --git a/docs/en/interfaces/formats.md b/docs/en/interfaces/formats.md index a81a17e65d6..9c4c082bc3a 100644 --- a/docs/en/interfaces/formats.md +++ b/docs/en/interfaces/formats.md @@ -67,6 +67,7 @@ The supported formats are: | [Prometheus](#prometheus) | ✗ | ✔ | | [Protobuf](#protobuf) | ✔ | ✔ | | [ProtobufSingle](#protobufsingle) | ✔ | ✔ | +| [ProtobufList](#protobuflist) | ✔ | ✔ | | [Avro](#data-format-avro) | ✔ | ✔ | | [AvroConfluent](#data-format-avro-confluent) | ✔ | ✗ | | [Parquet](#data-format-parquet) | ✔ | ✔ | @@ -1535,6 +1536,10 @@ the columns from input data will be mapped to the columns from the table by thei Otherwise, the first row will be skipped. If setting [input_format_with_types_use_header](/docs/en/operations/settings/settings-formats.md/#input_format_with_types_use_header) is set to 1, the types from input data will be compared with the types of the corresponding columns from the table. Otherwise, the second row will be skipped. +If setting [output_format_binary_encode_types_in_binary_format](/docs/en/operations/settings/settings-formats.md/#output_format_binary_encode_types_in_binary_format) is set to 1, +the types in header will be written using [binary encoding](/docs/en/sql-reference/data-types/data-types-binary-encoding.md) instead of strings with type names in RowBinaryWithNamesAndTypes output format. +If setting [input_format_binary_encode_types_in_binary_format](/docs/en/operations/settings/settings-formats.md/#input_format_binary_encode_types_in_binary_format) is set to 1, +the types in header will be read using [binary encoding](/docs/en/sql-reference/data-types/data-types-binary-encoding.md) instead of strings with type names in RowBinaryWithNamesAndTypes input format. ::: ## RowBinaryWithDefaults {#rowbinarywithdefaults} @@ -1948,6 +1953,35 @@ SYSTEM DROP FORMAT SCHEMA CACHE FOR Protobuf Same as [Protobuf](#protobuf) but for storing/parsing single Protobuf message without length delimiters. +## ProtobufList {#protobuflist} + +Similar to Protobuf but rows are represented as a sequence of sub-messages contained in a message with fixed name "Envelope". + +Usage example: + +``` sql +SELECT * FROM test.table FORMAT ProtobufList SETTINGS format_schema = 'schemafile:MessageType' +``` + +``` bash +cat protobuflist_messages.bin | clickhouse-client --query "INSERT INTO test.table FORMAT ProtobufList SETTINGS format_schema='schemafile:MessageType'" +``` + +where the file `schemafile.proto` looks like this: + +``` capnp +syntax = "proto3"; +message Envelope { + message MessageType { + string name = 1; + string surname = 2; + uint32 birthDate = 3; + repeated string phoneNumbers = 4; + }; + MessageType row = 1; +}; +``` + ## Avro {#data-format-avro} [Apache Avro](https://avro.apache.org/) is a row-oriented data serialization framework developed within Apache’s Hadoop project. diff --git a/docs/en/interfaces/mysql.md b/docs/en/interfaces/mysql.md index 42820505406..30a8d9569a9 100644 --- a/docs/en/interfaces/mysql.md +++ b/docs/en/interfaces/mysql.md @@ -6,7 +6,20 @@ sidebar_label: MySQL Interface # MySQL Interface -ClickHouse supports the MySQL wire protocol. This allow tools that are MySQL-compatible to interact with ClickHouse seamlessly (e.g. [Looker Studio](../integrations/data-visualization/looker-studio-and-clickhouse.md)). +ClickHouse supports the MySQL wire protocol. This allows certain clients that do not have native ClickHouse connectors leverage the MySQL protocol instead, and it has been validated with the following BI tools: + +- [Looker Studio](../integrations/data-visualization/looker-studio-and-clickhouse.md) +- [Tableau Online](../integrations/tableau-online) +- [QuickSight](../integrations/quicksight) + +If you are trying other untested clients or integrations, keep in mind that there could be the following limitations: + +- SSL implementation might not be fully compatible; there could be potential [TLS SNI](https://www.cloudflare.com/learning/ssl/what-is-sni/) issues. +- A particular tool might require dialect features (e.g., MySQL-specific functions or settings) that are not implemented yet. + +If there is a native driver available (e.g., [DBeaver](../integrations/dbeaver)), it is always preferred to use it instead of the MySQL interface. Additionally, while most of the MySQL language clients should work fine, MySQL interface is not guaranteed to be a drop-in replacement for a codebase with existing MySQL queries. + +If your use case involves a particular tool that does not have a native ClickHouse driver, and you would like to use it via the MySQL interface and you found certain incompatibilities - please [create an issue](https://github.com/ClickHouse/ClickHouse/issues) in the ClickHouse repository. ## Enabling the MySQL Interface On ClickHouse Cloud diff --git a/docs/en/operations/backup.md b/docs/en/operations/backup.md index 46c24ad8491..fc861e25e9f 100644 --- a/docs/en/operations/backup.md +++ b/docs/en/operations/backup.md @@ -84,6 +84,7 @@ The BACKUP and RESTORE statements take a list of DATABASE and TABLE names, a des - [`compression_method`](/docs/en/sql-reference/statements/create/table.md/#column-compression-codecs) and compression_level - `password` for the file on disk - `base_backup`: the destination of the previous backup of this source. For example, `Disk('backups', '1.zip')` + - `use_same_s3_credentials_for_base_backup`: whether base backup to S3 should inherit credentials from the query. Only works with `S3`. - `structure_only`: if enabled, allows to only backup or restore the CREATE statements without the data of tables - `storage_policy`: storage policy for the tables being restored. See [Using Multiple Block Devices for Data Storage](../engines/table-engines/mergetree-family/mergetree.md#table_engine-mergetree-multiple-volumes). This setting is only applicable to the `RESTORE` command. The specified storage policy applies only to tables with an engine from the `MergeTree` family. - `s3_storage_class`: the storage class used for S3 backup. For example, `STANDARD` diff --git a/docs/en/operations/configuration-files.md b/docs/en/operations/configuration-files.md index 57fea3cca3a..4a818410ebe 100644 --- a/docs/en/operations/configuration-files.md +++ b/docs/en/operations/configuration-files.md @@ -124,7 +124,7 @@ which is equal to #### Default values for from_env and from_zk attributes -It's possible to set the default value and substitute it only if the environment variable or zookeeper node is set using `replace="1"`. +It's possible to set the default value and substitute it only if the environment variable or zookeeper node is set using `replace="1"` (must be declared before from_env). With previous example, but `MAX_QUERY_SIZE` is unset: @@ -132,7 +132,7 @@ With previous example, but `MAX_QUERY_SIZE` is unset: - 150000 + 150000 diff --git a/docs/en/operations/external-authenticators/ssl-x509.md b/docs/en/operations/external-authenticators/ssl-x509.md index 109913c2b18..09fac45d7ae 100644 --- a/docs/en/operations/external-authenticators/ssl-x509.md +++ b/docs/en/operations/external-authenticators/ssl-x509.md @@ -6,23 +6,30 @@ import SelfManaged from '@site/docs/en/_snippets/_self_managed_only_no_roadmap.m -[SSL 'strict' option](../server-configuration-parameters/settings.md#server_configuration_parameters-openssl) enables mandatory certificate validation for the incoming connections. In this case, only connections with trusted certificates can be established. Connections with untrusted certificates will be rejected. Thus, certificate validation allows to uniquely authenticate an incoming connection. `Common Name` field of the certificate is used to identify connected user. This allows to associate multiple certificates with the same user. Additionally, reissuing and revoking of the certificates does not affect the ClickHouse configuration. +[SSL 'strict' option](../server-configuration-parameters/settings.md#server_configuration_parameters-openssl) enables mandatory certificate validation for the incoming connections. In this case, only connections with trusted certificates can be established. Connections with untrusted certificates will be rejected. Thus, certificate validation allows to uniquely authenticate an incoming connection. `Common Name` or `subjectAltName extension` field of the certificate is used to identify the connected user. This allows to associate multiple certificates with the same user. Additionally, reissuing and revoking of the certificates does not affect the ClickHouse configuration. -To enable SSL certificate authentication, a list of `Common Name`'s for each ClickHouse user must be specified in the settings file `users.xml `: +To enable SSL certificate authentication, a list of `Common Name`'s or `Subject Alt Name`'s for each ClickHouse user must be specified in the settings file `users.xml `: **Example** ```xml - + host.domain.com:example_user host.domain.com:example_user_dev - + + + + DNS:host.domain.com + + + + ``` diff --git a/docs/en/operations/named-collections.md b/docs/en/operations/named-collections.md index 91438cfb675..59ee05d1f9e 100644 --- a/docs/en/operations/named-collections.md +++ b/docs/en/operations/named-collections.md @@ -5,6 +5,10 @@ sidebar_label: "Named collections" title: "Named collections" --- +import CloudNotSupportedBadge from '@theme/badges/CloudNotSupportedBadge'; + + + Named collections provide a way to store collections of key-value pairs to be used to configure integrations with external sources. You can use named collections with dictionaries, tables, table functions, and object storage. diff --git a/docs/en/operations/opentelemetry.md b/docs/en/operations/opentelemetry.md index 70f64d08ba3..fe60ceedc0b 100644 --- a/docs/en/operations/opentelemetry.md +++ b/docs/en/operations/opentelemetry.md @@ -2,15 +2,11 @@ slug: /en/operations/opentelemetry sidebar_position: 62 sidebar_label: Tracing ClickHouse with OpenTelemetry -title: "[experimental] Tracing ClickHouse with OpenTelemetry" +title: "Tracing ClickHouse with OpenTelemetry" --- [OpenTelemetry](https://opentelemetry.io/) is an open standard for collecting traces and metrics from the distributed application. ClickHouse has some support for OpenTelemetry. -:::note -This is an experimental feature that will change in backwards-incompatible ways in future releases. -::: - ## Supplying Trace Context to ClickHouse ClickHouse accepts trace context HTTP headers, as described by the [W3C recommendation](https://www.w3.org/TR/trace-context/). It also accepts trace context over a native protocol that is used for communication between ClickHouse servers or between the client and server. For manual testing, trace context headers conforming to the Trace Context recommendation can be supplied to `clickhouse-client` using `--opentelemetry-traceparent` and `--opentelemetry-tracestate` flags. diff --git a/docs/en/operations/server-configuration-parameters/settings.md b/docs/en/operations/server-configuration-parameters/settings.md index 27661bc075b..55e8f6fdf6b 100644 --- a/docs/en/operations/server-configuration-parameters/settings.md +++ b/docs/en/operations/server-configuration-parameters/settings.md @@ -498,6 +498,8 @@ Default: 0.9 Interval in seconds during which the server's maximum allowed memory consumption is adjusted by the corresponding threshold in cgroups. (see settings `cgroup_memory_watcher_hard_limit_ratio` and `cgroup_memory_watcher_soft_limit_ratio`). +To disable the cgroup observer, set this value to `0`. + Type: UInt64 Default: 15 diff --git a/docs/en/operations/settings/merge-tree-settings.md b/docs/en/operations/settings/merge-tree-settings.md index 9879ee35612..7278b91f90d 100644 --- a/docs/en/operations/settings/merge-tree-settings.md +++ b/docs/en/operations/settings/merge-tree-settings.md @@ -974,6 +974,13 @@ Default value: false - [exclude_deleted_rows_for_part_size_in_merge](#exclude_deleted_rows_for_part_size_in_merge) setting +## use_compact_variant_discriminators_serialization {#use_compact_variant_discriminators_serialization} + +Enables compact mode for binary serialization of discriminators in Variant data type. +This mode allows to use significantly less memory for storing discriminators in parts when there is mostly one variant or a lot of NULL values. + +Default value: true + ## merge_workload Used to regulate how resources are utilized and shared between merges and other workloads. Specified value is used as `workload` setting value for background merges of this table. If not specified (empty string), then server setting `merge_workload` is used instead. @@ -1023,7 +1030,7 @@ A table with no primary key represents the extreme case of a single equivalence The fewer and the larger the equivalence classes are, the higher the degree of freedom when re-shuffling rows. -The heuristics applied to find the best row order within each equivalence class is suggested by D. Lemir, O. Kaser in [Reordering columns for smaller indexes](https://doi.org/10.1016/j.ins.2011.02.002) and based on sorting the rows within each equivalence class by ascending cardinality of the non-primary key columns. +The heuristics applied to find the best row order within each equivalence class is suggested by D. Lemire, O. Kaser in [Reordering columns for smaller indexes](https://doi.org/10.1016/j.ins.2011.02.002) and based on sorting the rows within each equivalence class by ascending cardinality of the non-primary key columns. It performs three steps: 1. Find all equivalence classes based on the row values in primary key columns. 2. For each equivalence class, calculate (usually estimate) the cardinalities of the non-primary-key columns. diff --git a/docs/en/operations/settings/settings-formats.md b/docs/en/operations/settings/settings-formats.md index 530023df5b7..f8b40cd81ac 100644 --- a/docs/en/operations/settings/settings-formats.md +++ b/docs/en/operations/settings/settings-formats.md @@ -1951,6 +1951,18 @@ The maximum allowed size for String in RowBinary format. It prevents allocating Default value: `1GiB`. +### output_format_binary_encode_types_in_binary_format {#output_format_binary_encode_types_in_binary_format} + +Write data types in [binary format](../../sql-reference/data-types/data-types-binary-encoding.md) instead of type names in RowBinaryWithNamesAndTypes output format. + +Disabled by default. + +### input_format_binary_decode_types_in_binary_format {#input_format_binary_decode_types_in_binary_format} + +Read data types in [binary format](../../sql-reference/data-types/data-types-binary-encoding.md) instead of type names in RowBinaryWithNamesAndTypes input format. + +Disabled by default. + ## Native format settings {#native-format-settings} ### input_format_native_allow_types_conversion {#input_format_native_allow_types_conversion} @@ -1958,3 +1970,15 @@ Default value: `1GiB`. Allow types conversion in Native input format between columns from input data and requested columns. Enabled by default. + +### output_format_native_encode_types_in_binary_format {#output_format_native_encode_types_in_binary_format} + +Write data types in [binary format](../../sql-reference/data-types/data-types-binary-encoding.md) instead of type names in Native output format. + +Disabled by default. + +### input_format_native_decode_types_in_binary_format {#input_format_native_decode_types_in_binary_format} + +Read data types in [binary format](../../sql-reference/data-types/data-types-binary-encoding.md) instead of type names in Native input format. + +Disabled by default. \ No newline at end of file diff --git a/docs/en/operations/settings/settings-users.md b/docs/en/operations/settings/settings-users.md index 96477f777a9..ef1e58fd18e 100644 --- a/docs/en/operations/settings/settings-users.md +++ b/docs/en/operations/settings/settings-users.md @@ -22,6 +22,21 @@ Structure of the `users` section: + + + ssh-ed25519 + AAAAC3NzaC1lZDI1NTE5AAAAIDNf0r6vRl24Ix3tv2IgPmNPO2ATa2krvt80DdcTatLj + + + ecdsa-sha2-nistp256 + AAAAE2VjZHNhLXNoYTItbmlzdHAyNTYAAAAIbmlzdHAyNTYAAABBBNxeV2uN5UY6CUbCzTA1rXfYimKQA5ivNIqxdax4bcMXz4D0nSk2l5E1TkR5mG8EBWtmExSPbcEPJ8V7lyWWbA8= + + + ssh-rsa + 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 + + + 0|1 @@ -79,6 +94,24 @@ Password can be specified in plaintext or in SHA256 (hex format). The first line of the result is the password. The second line is the corresponding double SHA1 hash. +### username/ssh-key {#user-sshkey} + +This setting allows authenticating with SSH keys. + +Given a SSH key (as generated by `ssh-keygen`) like +``` +ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIDNf0r6vRl24Ix3tv2IgPmNPO2ATa2krvt80DdcTatLj john@example.com +``` +The `ssh_key` element is expected to be +``` + + ssh-ed25519 + AAAAC3NzaC1lZDI1NTE5AAAAIDNf0r6vRl24Ix3tv2IgPmNPO2ATa2krvt80DdcTatLj + +``` + +Substitute `ssh-ed25519` with `ssh-rsa` or `ecdsa-sha2-nistp256` for the other supported algorithms. + ### access_management {#access_management-user-setting} This setting enables or disables using of SQL-driven [access control and account management](../../guides/sre/user-management/index.md#access-control) for the user. diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index 62f59559034..68fc2af0843 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -1170,6 +1170,10 @@ Data in the VALUES clause of INSERT queries is processed by a separate stream pa Default value: 262144 (= 256 KiB). +:::note +`max_query_size` cannot be set within an SQL query (e.g., `SELECT now() SETTINGS max_query_size=10000`) because ClickHouse needs to allocate a buffer to parse the query, and this buffer size is determined by the `max_query_size` setting, which must be configured before the query is executed. +::: + ## max_parser_depth {#max_parser_depth} Limits maximum recursion depth in the recursive descent parser. Allows controlling the stack size. @@ -1354,12 +1358,25 @@ Connection pool size for PostgreSQL table engine and database engine. Default value: 16 +## postgresql_connection_attempt_timeout {#postgresql-connection-attempt-timeout} + +Connection timeout in seconds of a single attempt to connect PostgreSQL end-point. +The value is passed as a `connect_timeout` parameter of the connection URL. + +Default value: `2`. + ## postgresql_connection_pool_wait_timeout {#postgresql-connection-pool-wait-timeout} Connection pool push/pop timeout on empty pool for PostgreSQL table engine and database engine. By default it will block on empty pool. Default value: 5000 +## postgresql_connection_pool_retries {#postgresql-connection-pool-retries} + +The maximum number of retries to establish a connection with the PostgreSQL end-point. + +Default value: `2`. + ## postgresql_connection_pool_auto_close_connection {#postgresql-connection-pool-auto-close-connection} Close connection before returning connection to the pool. @@ -2536,7 +2553,7 @@ Possible values: - 0 — Optimization disabled. - 1 — Optimization enabled. -Default value: `0`. +Default value: `1`. ## optimize_trivial_count_query {#optimize-trivial-count-query} diff --git a/docs/en/operations/startup-scripts.md b/docs/en/operations/startup-scripts.md new file mode 100644 index 00000000000..91aa4772bcf --- /dev/null +++ b/docs/en/operations/startup-scripts.md @@ -0,0 +1,30 @@ +--- +slug: /en/operations/startup-scripts +sidebar_label: Startup Scripts +--- + +# Startup Scripts + +ClickHouse can run arbitrary SQL queries from the server configuration during startup. This can be useful for migrations or automatic schema creation. + +```xml + + + + CREATE ROLE OR REPLACE test_role + + + CREATE TABLE TestTable (id UInt64) ENGINE=TinyLog + SELECT 1; + + + +``` + +ClickHouse executes all queries from the `startup_scripts` sequentially in the specified order. If any of the queries fail, the execution of the following queries won't be interrupted. + +You can specify a conditional query in the config. In that case, the corresponding query executes only when the condition query returns the value `1` or `true`. + +:::note +If the condition query returns any other value than `1` or `true`, the result will be interpreted as `false`, and the corresponding won't be executed. +::: diff --git a/docs/en/operations/system-tables/detached_tables.md b/docs/en/operations/system-tables/detached_tables.md new file mode 100644 index 00000000000..f669c8fd251 --- /dev/null +++ b/docs/en/operations/system-tables/detached_tables.md @@ -0,0 +1,35 @@ +--- +slug: /en/operations/system-tables/detached_tables +--- +# detached_tables + +Contains information of each detached table. + +Columns: + +- `database` ([String](../../sql-reference/data-types/string.md)) — The name of the database the table is in. + +- `table` ([String](../../sql-reference/data-types/string.md)) — Table name. + +- `uuid` ([UUID](../../sql-reference/data-types/uuid.md)) — Table uuid (Atomic database). + +- `metadata_path` ([String](../../sql-reference/data-types/string.md)) - Path to the table metadata in the file system. + +- `is_permanently` ([UInt8](../../sql-reference/data-types/int-uint.md)) - Flag indicates that the table was detached PERMANENTLY. + + +**Example** + +```sql +SELECT * FROM system.detached_tables FORMAT Vertical; +``` + +```text +Row 1: +────── +database: base +table: t1 +uuid: 81b1c20a-b7c6-4116-a2ce-7583fb6b6736 +metadata_path: /var/lib/clickhouse/store/461/461cf698-fd0b-406d-8c01-5d8fd5748a91/t1.sql +is_permanently: 1 +``` diff --git a/docs/en/operations/system-tables/functions.md b/docs/en/operations/system-tables/functions.md index d52bf24f289..da0ef04bd27 100644 --- a/docs/en/operations/system-tables/functions.md +++ b/docs/en/operations/system-tables/functions.md @@ -9,7 +9,6 @@ Columns: - `name` ([String](../../sql-reference/data-types/string.md)) – The name of the function. - `is_aggregate` ([UInt8](../../sql-reference/data-types/int-uint.md)) — Whether the function is an aggregate function. -- `is_deterministic` ([Nullable](../../sql-reference/data-types/nullable.md)([UInt8](../../sql-reference/data-types/int-uint.md))) - Whether the function is deterministic. - `case_insensitive`, ([UInt8](../../sql-reference/data-types/int-uint.md)) - Whether the function name can be used case-insensitively. - `alias_to`, ([String](../../sql-reference/data-types/string.md)) - The original function name, if the function name is an alias. - `create_query`, ([String](../../sql-reference/data-types/enum.md)) - Unused. diff --git a/docs/en/operations/system-tables/metrics.md b/docs/en/operations/system-tables/metrics.md index 83ce817b7db..f253b164e2a 100644 --- a/docs/en/operations/system-tables/metrics.md +++ b/docs/en/operations/system-tables/metrics.md @@ -357,7 +357,7 @@ Number of currently running inserts to Kafka Number of alive connections -### KeeperOutstandingRequets +### KeeperOutstandingRequests Number of outstanding requests diff --git a/docs/en/operations/utilities/clickhouse-disks.md b/docs/en/operations/utilities/clickhouse-disks.md index 76db9e41836..e22bc06b641 100644 --- a/docs/en/operations/utilities/clickhouse-disks.md +++ b/docs/en/operations/utilities/clickhouse-disks.md @@ -4,35 +4,56 @@ sidebar_position: 59 sidebar_label: clickhouse-disks --- -# clickhouse-disks +# Clickhouse-disks -A utility providing filesystem-like operations for ClickHouse disks. +A utility providing filesystem-like operations for ClickHouse disks. It can work in both interactive and not interactive modes. -Program-wide options: +## Program-wide options * `--config-file, -C` -- path to ClickHouse config, defaults to `/etc/clickhouse-server/config.xml`. * `--save-logs` -- Log progress of invoked commands to `/var/log/clickhouse-server/clickhouse-disks.log`. * `--log-level` -- What [type](../server-configuration-parameters/settings#server_configuration_parameters-logger) of events to log, defaults to `none`. * `--disk` -- what disk to use for `mkdir, move, read, write, remove` commands. Defaults to `default`. +* `--query, -q` -- single query that can be executed without launching interactive mode +* `--help, -h` -- print all the options and commands with description + +## Default Disks +After the launch two disks are initialized. The first one is a disk `local` that is supposed to imitate local file system from which clickhouse-disks utility was launched. The second one is a disk `default` that is mounted to the local filesystem in the directory that can be found in config as a parameter `clickhouse/path` (default value is `/var/lib/clickhouse`). + +## Clickhouse-disks state +For each disk that was added the utility stores current directory (as in a usual filesystem). User can change current directory and switch between disks. + +State is reflected in a prompt "`disk_name`:`path_name`" ## Commands -* `copy [--disk-from d1] [--disk-to d2] `. - Recursively copy data from `FROM_PATH` at disk `d1` (defaults to `disk` value if not provided) - to `TO_PATH` at disk `d2` (defaults to `disk` value if not provided). -* `move `. - Move file or directory from `FROM_PATH` to `TO_PATH`. -* `remove `. - Remove `PATH` recursively. -* `link `. - Create a hardlink from `FROM_PATH` to `TO_PATH`. -* `list [--recursive] ...` - List files at `PATH`s. Non-recursive by default. -* `list-disks`. +In these documentation file all mandatory positional arguments are referred as ``, named arguments are referred as `[--parameter value]`. All positional parameters could be mentioned as a named parameter with a corresponding name. + +* `cd (change-dir, change_dir) [--disk disk] ` + Change directory to path `path` on disk `disk` (default value is a current disk). No disk switching happens. +* `copy (cp) [--disk-from disk_1] [--disk-to disk_2] `. + Recursively copy data from `path-from` at disk `disk_1` (default value is a current disk (parameter `disk` in a non-interactive mode)) + to `path-to` at disk `disk_2` (default value is a current disk (parameter `disk` in a non-interactive mode)). +* `current_disk_with_path (current, current_disk, current_path)` + Print current state in format: + `Disk: "current_disk" Path: "current path on current disk"` +* `help []` + Print help message about command `command`. If `command` is not specified print information about all commands. +* `move (mv) `. + Move file or directory from `path-from` to `path-to` within current disk. +* `remove (rm, delete) `. + Remove `path` recursively on a current disk. +* `link (ln) `. + Create a hardlink from `path-from` to `path-to` on a current disk. +* `list (ls) [--recursive] ` + List files at `path`s on a current disk. Non-recursive by default. +* `list-disks (list_disks, ls-disks, ls_disks)`. List disks names. -* `mkdir [--recursive] `. +* `mkdir [--recursive] ` on a current disk. Create a directory. Non-recursive by default. -* `read: []` - Read a file from `FROM_PATH` to `TO_PATH` (`stdout` if not supplied). -* `write [FROM_PATH] `. - Write a file from `FROM_PATH` (`stdin` if not supplied) to `TO_PATH`. +* `read (r) [--path-to path]` + Read a file from `path-from` to `path` (`stdout` if not supplied). +* `switch-disk [--path path] ` + Switch to disk `disk` on path `path` (if `path` is not specified default value is a previous path on disk `disk`). +* `write (w) [--path-from path] `. + Write a file from `path` (`stdin` if `path` is not supplied, input must finish by Ctrl+D) to `path-to`. diff --git a/docs/en/operations/utilities/clickhouse-local.md b/docs/en/operations/utilities/clickhouse-local.md index 93a3fecf3c6..c20e4fc3b09 100644 --- a/docs/en/operations/utilities/clickhouse-local.md +++ b/docs/en/operations/utilities/clickhouse-local.md @@ -16,7 +16,7 @@ sidebar_label: clickhouse-local While `clickhouse-local` is a great tool for development and testing purposes, and for processing files, it is not suitable for serving end users or applications. In these scenarios, it is recommended to use the open-source [ClickHouse](https://clickhouse.com/docs/en/install). ClickHouse is a powerful OLAP database that is designed to handle large-scale analytical workloads. It provides fast and efficient processing of complex queries on large datasets, making it ideal for use in production environments where high-performance is critical. Additionally, ClickHouse offers a wide range of features such as replication, sharding, and high availability, which are essential for scaling up to handle large datasets and serving applications. If you need to handle larger datasets or serve end users or applications, we recommend using open-source ClickHouse instead of `clickhouse-local`. -Please read the docs below that show example use cases for `clickhouse-local`, such as [querying local CSVs](#query-data-in-a-csv-file-using-sql) or [reading a parquet file in S3](#query-data-in-a-parquet-file-in-aws-s3). +Please read the docs below that show example use cases for `clickhouse-local`, such as [querying local file](#query_data_in_file) or [reading a parquet file in S3](#query-data-in-a-parquet-file-in-aws-s3). ## Download clickhouse-local @@ -236,10 +236,10 @@ Read 2 rows, 32.00 B in 0.000 sec., 5182 rows/sec., 80.97 KiB/sec. Previous example is the same as: ``` bash -$ echo -e "1,2\n3,4" | clickhouse-local --query " +$ echo -e "1,2\n3,4" | clickhouse-local -n --query " CREATE TABLE table (a Int64, b Int64) ENGINE = File(CSV, stdin); SELECT a, b FROM table; - DROP TABLE table" + DROP TABLE table;" Read 2 rows, 32.00 B in 0.000 sec., 4987 rows/sec., 77.93 KiB/sec. 1 2 3 4 diff --git a/docs/en/sql-reference/aggregate-functions/index.md b/docs/en/sql-reference/aggregate-functions/index.md index 96bf0c5d93b..5056ef2c7aa 100644 --- a/docs/en/sql-reference/aggregate-functions/index.md +++ b/docs/en/sql-reference/aggregate-functions/index.md @@ -18,7 +18,7 @@ ClickHouse also supports: During aggregation, all `NULL` arguments are skipped. If the aggregation has several arguments it will ignore any row in which one or more of them are NULL. -There is an exception to this rule, which are the functions [`first_value`](../../sql-reference/aggregate-functions/reference/first_value.md), [`last_value`](../../sql-reference/aggregate-functions/reference/last_value.md) and their aliases when followed by the modifier `RESPECT NULLS`: `FIRST_VALUE(b) RESPECT NULLS`. +There is an exception to this rule, which are the functions [`first_value`](../../sql-reference/aggregate-functions/reference/first_value.md), [`last_value`](../../sql-reference/aggregate-functions/reference/last_value.md) and their aliases (`any` and `anyLast` respectively) when followed by the modifier `RESPECT NULLS`. For example, `FIRST_VALUE(b) RESPECT NULLS`. **Examples:** diff --git a/docs/en/sql-reference/aggregate-functions/reference/aggthrow.md b/docs/en/sql-reference/aggregate-functions/reference/aggthrow.md new file mode 100644 index 00000000000..fdbfd5b9e41 --- /dev/null +++ b/docs/en/sql-reference/aggregate-functions/reference/aggthrow.md @@ -0,0 +1,37 @@ +--- +slug: /en/sql-reference/aggregate-functions/reference/aggthrow +sidebar_position: 101 +--- + +# aggThrow + +This function can be used for the purpose of testing exception safety. It will throw an exception on creation with the specified probability. + +**Syntax** + +```sql +aggThrow(throw_prob) +``` + +**Arguments** + +- `throw_prob` — Probability to throw on creation. [Float64](../../data-types/float.md). + +**Returned value** + +- An exception: `Code: 503. DB::Exception: Aggregate function aggThrow has thrown exception successfully`. + +**Example** + +Query: + +```sql +SELECT number % 2 AS even, aggThrow(number) FROM numbers(10) GROUP BY even; +``` + +Result: + +```response +Received exception: +Code: 503. DB::Exception: Aggregate function aggThrow has thrown exception successfully: While executing AggregatingTransform. (AGGREGATE_FUNCTION_THROW) +``` diff --git a/docs/en/sql-reference/aggregate-functions/reference/any.md b/docs/en/sql-reference/aggregate-functions/reference/any.md index cdff7dde4a9..972263585f2 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/any.md +++ b/docs/en/sql-reference/aggregate-functions/reference/any.md @@ -5,12 +5,12 @@ sidebar_position: 102 # any -Selects the first encountered value of a column. +Selects the first encountered value of a column, ignoring any `NULL` values. **Syntax** ```sql -any(column) +any(column) [RESPECT NULLS] ``` Aliases: `any_value`, [`first_value`](../reference/first_value.md). @@ -20,7 +20,9 @@ Aliases: `any_value`, [`first_value`](../reference/first_value.md). **Returned value** -By default, it ignores NULL values and returns the first NOT NULL value found in the column. Like [`first_value`](../../../sql-reference/aggregate-functions/reference/first_value.md) it supports `RESPECT NULLS`, in which case it will select the first value passed, independently on whether it's NULL or not. +:::note +Supports the `RESPECT NULLS` modifier after the function name. Using this modifier will ensure the function selects the first value passed, regardless of whether it is `NULL` or not. +::: :::note The return type of the function is the same as the input, except for LowCardinality which is discarded. This means that given no rows as input it will return the default value of that type (0 for integers, or Null for a Nullable() column). You might use the `-OrNull` [combinator](../../../sql-reference/aggregate-functions/combinators.md) ) to modify this behaviour. diff --git a/docs/en/sql-reference/aggregate-functions/reference/any_respect_nulls.md b/docs/en/sql-reference/aggregate-functions/reference/any_respect_nulls.md deleted file mode 100644 index 99104a9b8c7..00000000000 --- a/docs/en/sql-reference/aggregate-functions/reference/any_respect_nulls.md +++ /dev/null @@ -1,44 +0,0 @@ ---- -slug: /en/sql-reference/aggregate-functions/reference/any_respect_nulls -sidebar_position: 103 ---- - -# any_respect_nulls - -Selects the first encountered value of a column, irregardless of whether it is a `NULL` value or not. - -Alias: `any_value_respect_nulls`, `first_value_repect_nulls`. - -**Syntax** - -```sql -any_respect_nulls(column) -``` - -**Parameters** -- `column`: The column name. - -**Returned value** - -- The last value encountered, irregardless of whether it is a `NULL` value or not. - -**Example** - -Query: - -```sql -CREATE TABLE any_nulls (city Nullable(String)) ENGINE=Log; - -INSERT INTO any_nulls (city) VALUES (NULL), ('Amsterdam'), ('New York'), ('Tokyo'), ('Valencia'), (NULL); - -SELECT any(city), any_respect_nulls(city) FROM any_nulls; -``` - -```response -┌─any(city)─┬─any_respect_nulls(city)─┐ -│ Amsterdam │ ᴺᵁᴸᴸ │ -└───────────┴─────────────────────────┘ -``` - -**See Also** -- [any](../reference/any.md) diff --git a/docs/en/sql-reference/aggregate-functions/reference/anylast.md b/docs/en/sql-reference/aggregate-functions/reference/anylast.md index e43bc07fbdc..202d2e9fb10 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/anylast.md +++ b/docs/en/sql-reference/aggregate-functions/reference/anylast.md @@ -5,17 +5,21 @@ sidebar_position: 105 # anyLast -Selects the last value encountered. The result is just as indeterminate as for the [any](../../../sql-reference/aggregate-functions/reference/any.md) function. +Selects the last value encountered, ignoring any `NULL` values by default. The result is just as indeterminate as for the [any](../../../sql-reference/aggregate-functions/reference/any.md) function. **Syntax** ```sql -anyLast(column) +anyLast(column) [RESPECT NULLS] ``` **Parameters** - `column`: The column name. +:::note +Supports the `RESPECT NULLS` modifier after the function name. Using this modifier will ensure the function selects the first value passed, regardless of whether it is `NULL` or not. +::: + **Returned value** - The last value encountered. diff --git a/docs/en/sql-reference/aggregate-functions/reference/anylast_respect_nulls.md b/docs/en/sql-reference/aggregate-functions/reference/anylast_respect_nulls.md deleted file mode 100644 index 8f093cfdb61..00000000000 --- a/docs/en/sql-reference/aggregate-functions/reference/anylast_respect_nulls.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -slug: /en/sql-reference/aggregate-functions/reference/anylast_respect_nulls -sidebar_position: 106 ---- - -# anyLast_respect_nulls - -Selects the last value encountered, irregardless of whether it is `NULL` or not. - -**Syntax** - -```sql -anyLast_respect_nulls(column) -``` - -**Parameters** -- `column`: The column name. - -**Returned value** - -- The last value encountered, irregardless of whether it is `NULL` or not. - -**Example** - -Query: - -```sql -CREATE TABLE any_last_nulls (city Nullable(String)) ENGINE=Log; - -INSERT INTO any_last_nulls (city) VALUES ('Amsterdam'),(NULL),('New York'),('Tokyo'),('Valencia'),(NULL); - -SELECT anyLast(city), anyLast_respect_nulls(city) FROM any_last_nulls; -``` - -```response -┌─anyLast(city)─┬─anyLast_respect_nulls(city)─┐ -│ Valencia │ ᴺᵁᴸᴸ │ -└───────────────┴─────────────────────────────┘ -``` \ No newline at end of file diff --git a/docs/en/sql-reference/aggregate-functions/reference/index.md b/docs/en/sql-reference/aggregate-functions/reference/index.md index e3725b6a430..2dce0afe2e1 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/index.md +++ b/docs/en/sql-reference/aggregate-functions/reference/index.md @@ -43,11 +43,11 @@ Standard aggregate functions: ClickHouse-specific aggregate functions: +- [aggThrow](../reference/aggthrow.md) - [analysisOfVariance](../reference/analysis_of_variance.md) -- [any](../reference/any_respect_nulls.md) +- [any](../reference/any.md) - [anyHeavy](../reference/anyheavy.md) - [anyLast](../reference/anylast.md) -- [anyLast](../reference/anylast_respect_nulls.md) - [boundingRatio](../reference/boundrat.md) - [first_value](../reference/first_value.md) - [last_value](../reference/last_value.md) diff --git a/docs/en/sql-reference/aggregate-functions/reference/maxmap.md b/docs/en/sql-reference/aggregate-functions/reference/maxmap.md index c9c6913249c..73075c0823d 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/maxmap.md +++ b/docs/en/sql-reference/aggregate-functions/reference/maxmap.md @@ -5,23 +5,45 @@ sidebar_position: 165 # maxMap -Syntax: `maxMap(key, value)` or `maxMap(Tuple(key, value))` - Calculates the maximum from `value` array according to the keys specified in the `key` array. -Passing a tuple of keys and value arrays is identical to passing two arrays of keys and values. +**Syntax** -The number of elements in `key` and `value` must be the same for each row that is totaled. +```sql +maxMap(key, value) +``` +or +```sql +maxMap(Tuple(key, value)) +``` -Returns a tuple of two arrays: keys and values calculated for the corresponding keys. +Alias: `maxMappedArrays` -Example: +:::note +- Passing a tuple of keys and value arrays is identical to passing two arrays of keys and values. +- The number of elements in `key` and `value` must be the same for each row that is totaled. +::: + +**Parameters** + +- `key` — Array of keys. [Array](../../data-types/array.md). +- `value` — Array of values. [Array](../../data-types/array.md). + +**Returned value** + +- Returns a tuple of two arrays: keys in sorted order, and values calculated for the corresponding keys. [Tuple](../../data-types/tuple.md)([Array](../../data-types/array.md), [Array](../../data-types/array.md)). + +**Example** + +Query: ``` sql SELECT maxMap(a, b) FROM values('a Array(Char), b Array(Int64)', (['x', 'y'], [2, 2]), (['y', 'z'], [3, 1])) ``` +Result: + ``` text ┌─maxMap(a, b)───────────┐ │ [['x','y','z'],[2,3,1]]│ diff --git a/docs/en/sql-reference/aggregate-functions/reference/minmap.md b/docs/en/sql-reference/aggregate-functions/reference/minmap.md index b1fbb9e49f3..c0f340b3f3f 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/minmap.md +++ b/docs/en/sql-reference/aggregate-functions/reference/minmap.md @@ -5,23 +5,45 @@ sidebar_position: 169 # minMap -Syntax: `minMap(key, value)` or `minMap(Tuple(key, value))` - Calculates the minimum from `value` array according to the keys specified in the `key` array. -Passing a tuple of keys and value ​​arrays is identical to passing two arrays of keys and values. +**Syntax** -The number of elements in `key` and `value` must be the same for each row that is totaled. +```sql +`minMap(key, value)` +``` +or +```sql +minMap(Tuple(key, value)) +``` -Returns a tuple of two arrays: keys in sorted order, and values calculated for the corresponding keys. +Alias: `minMappedArrays` -Example: +:::note +- Passing a tuple of keys and value arrays is identical to passing an array of keys and an array of values. +- The number of elements in `key` and `value` must be the same for each row that is totaled. +::: + +**Parameters** + +- `key` — Array of keys. [Array](../../data-types/array.md). +- `value` — Array of values. [Array](../../data-types/array.md). + +**Returned value** + +- Returns a tuple of two arrays: keys in sorted order, and values calculated for the corresponding keys. [Tuple](../../data-types/tuple.md)([Array](../../data-types/array.md), [Array](../../data-types/array.md)). + +**Example** + +Query: ``` sql SELECT minMap(a, b) FROM values('a Array(Int32), b Array(Int64)', ([1, 2], [2, 2]), ([2, 3], [1, 1])) ``` +Result: + ``` text ┌─minMap(a, b)──────┐ │ ([1,2,3],[2,1,1]) │ diff --git a/docs/en/sql-reference/aggregate-functions/reference/singlevalueornull.md b/docs/en/sql-reference/aggregate-functions/reference/singlevalueornull.md index 21344b58ba6..19154c488d9 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/singlevalueornull.md +++ b/docs/en/sql-reference/aggregate-functions/reference/singlevalueornull.md @@ -16,7 +16,7 @@ singleValueOrNull(x) **Parameters** -- `x` — Column of any [data type](../../data-types/index.md). +- `x` — Column of any [data type](../../data-types/index.md) (except [Map](../../data-types/map.md), [Array](../../data-types/array.md) or [Tuple](../../data-types/tuple) which cannot be of type [Nullable](../../data-types/nullable.md)). **Returned values** diff --git a/docs/en/sql-reference/data-types/array.md b/docs/en/sql-reference/data-types/array.md index e5a8ce5d18b..1737fdd88b2 100644 --- a/docs/en/sql-reference/data-types/array.md +++ b/docs/en/sql-reference/data-types/array.md @@ -96,3 +96,22 @@ Result: │ 1 │ [2] │ [[4,1]] │ └───────────┴───────────┴───────────┘ ``` + +## Reading nested subcolumns from Array + +If nested type `T` inside `Array` has subcolumns (for example, if it's a [named tuple](./tuple.md)), you can read its subcolumns from an `Array(T)` type with the same subcolumn names. The type of a subcolumn will be `Array` of the type of original subcolumn. + +**Example** + +```sql +CREATE TABLE t_arr (arr Array(Tuple(field1 UInt32, field2 String))) ENGINE = MergeTree ORDER BY tuple(); +INSERT INTO t_arr VALUES ([(1, 'Hello'), (2, 'World')]), ([(3, 'This'), (4, 'is'), (5, 'subcolumn')]); +SELECT arr.field1, toTypeName(arr.field1), arr.field2, toTypeName(arr.field2) from t_arr; +``` + +```test +┌─arr.field1─┬─toTypeName(arr.field1)─┬─arr.field2────────────────┬─toTypeName(arr.field2)─┐ +│ [1,2] │ Array(UInt32) │ ['Hello','World'] │ Array(String) │ +│ [3,4,5] │ Array(UInt32) │ ['This','is','subcolumn'] │ Array(String) │ +└────────────┴────────────────────────┴───────────────────────────┴────────────────────────┘ +``` diff --git a/docs/en/sql-reference/data-types/data-types-binary-encoding.md b/docs/en/sql-reference/data-types/data-types-binary-encoding.md new file mode 100644 index 00000000000..812e946e43e --- /dev/null +++ b/docs/en/sql-reference/data-types/data-types-binary-encoding.md @@ -0,0 +1,115 @@ +--- +slug: /en/sql-reference/data-types/data-types-binary-encoding +sidebar_position: 56 +sidebar_label: Data types binary encoding specification. +--- + + +# Data types binary encoding specification + +This specification describes the binary format that can be used for binary encoding and decoding of ClickHouse data types. This format is used in `Dynamic` column [binary serialization](dynamic.md#binary-output-format) and can be used in input/output formats [RowBinaryWithNamesAndTypes](../../interfaces/formats.md#rowbinarywithnamesandtypes) and [Native](../../interfaces/formats.md#native) under corresponding settings. + +The table below describes how each data type is represented in binary format. Each data type encoding consist of 1 byte that indicates the type and some optional additional information. +`var_uint` in the binary encoding means that the size is encoded using Variable-Length Quantity compression. + +| ClickHouse data type | Binary encoding | +|--------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| `Nothing` | `0x00` | +| `UInt8` | `0x01` | +| `UInt16` | `0x02` | +| `UInt32` | `0x03` | +| `UInt64` | `0x04` | +| `UInt128` | `0x05` | +| `UInt256` | `0x06` | +| `Int8` | `0x07` | +| `Int16` | `0x08` | +| `Int32` | `0x09` | +| `Int64` | `0x0A` | +| `Int128` | `0x0B` | +| `Int256` | `0x0C` | +| `Float32` | `0x0D` | +| `Float64` | `0x0E` | +| `Date` | `0x0F` | +| `Date32` | `0x10` | +| `DateTime` | `0x11` | +| `DateTime(time_zone)` | `0x12` | +| `DateTime64(P)` | `0x13` | +| `DateTime64(P, time_zone)` | `0x14` | +| `String` | `0x15` | +| `FixedString(N)` | `0x16` | +| `Enum8` | `0x17...` | +| `Enum16` | `0x18...>` | +| `Decimal32(P, S)` | `0x19` | +| `Decimal64(P, S)` | `0x1A` | +| `Decimal128(P, S)` | `0x1B` | +| `Decimal256(P, S)` | `0x1C` | +| `UUID` | `0x1D` | +| `Array(T)` | `0x1E` | +| `Tuple(T1, ..., TN)` | `0x1F...` | +| `Tuple(name1 T1, ..., nameN TN)` | `0x20...` | +| `Set` | `0x21` | +| `Interval` | `0x22` (see [interval kind binary encoding](#interval-kind-binary-encoding)) | +| `Nullable(T)` | `0x23` | +| `Function` | `0x24...` | +| `AggregateFunction(function_name(param_1, ..., param_N), arg_T1, ..., arg_TN)` | `0x25......` (see [aggregate function parameter binary encoding](#aggregate-function-parameter-binary-encoding)) | +| `LowCardinality(T)` | `0x26` | +| `Map(K, V)` | `0x27` | +| `IPv4` | `0x28` | +| `IPv6` | `0x29` | +| `Variant(T1, ..., TN)` | `0x2A...` | +| `Dynamic(max_types=N)` | `0x2B` | +| `Custom type` (`Ring`, `Polygon`, etc) | `0x2C` | +| `Bool` | `0x2D` | +| `SimpleAggregateFunction(function_name(param_1, ..., param_N), arg_T1, ..., arg_TN)` | `0x2E......` (see [aggregate function parameter binary encoding](#aggregate-function-parameter-binary-encoding)) | +| `Nested(name1 T1, ..., nameN TN)` | `0x2F...` | + + +### Interval kind binary encoding + +The table below describes how different interval kinds of `Interval` data type are encoded. + +| Interval kind | Binary encoding | +|---------------|-----------------| +| `Nanosecond` | `0x00` | +| `Microsecond` | `0x01` | +| `Millisecond` | `0x02` | +| `Second` | `0x03` | +| `Minute` | `0x04` | +| `Hour` | `0x05` | +| `Day` | `0x06` | +| `Week` | `0x07` | +| `Month` | `0x08` | +| `Quarter` | `0x09` | +| `Year` | `0x1A` | + +### Aggregate function parameter binary encoding + +The table below describes how parameters of `AggragateFunction` and `SimpleAggregateFunction` are encoded. +The encoding of a parameter consists of 1 byte indicating the type of the parameter and the value itself. + +| Parameter type | Binary encoding | +|--------------------------|--------------------------------------------------------------------------------------------------------------------------------| +| `Null` | `0x00` | +| `UInt64` | `0x01` | +| `Int64` | `0x02` | +| `UInt128` | `0x03` | +| `Int128` | `0x04` | +| `UInt128` | `0x05` | +| `Int128` | `0x06` | +| `Float64` | `0x07` | +| `Decimal32` | `0x08` | +| `Decimal64` | `0x09` | +| `Decimal128` | `0x0A` | +| `Decimal256` | `0x0B` | +| `String` | `0x0C` | +| `Array` | `0x0D...` | +| `Tuple` | `0x0E...` | +| `Map` | `0x0F...` | +| `IPv4` | `0x10` | +| `IPv6` | `0x11` | +| `UUID` | `0x12` | +| `Bool` | `0x13` | +| `Object` | `0x14...` | +| `AggregateFunctionState` | `0x15` | +| `Negative infinity` | `0xFE` | +| `Positive infinity` | `0xFF` | diff --git a/docs/en/sql-reference/data-types/dynamic.md b/docs/en/sql-reference/data-types/dynamic.md index 955fd54e641..8be81471377 100644 --- a/docs/en/sql-reference/data-types/dynamic.md +++ b/docs/en/sql-reference/data-types/dynamic.md @@ -1,6 +1,6 @@ --- slug: /en/sql-reference/data-types/dynamic -sidebar_position: 56 +sidebar_position: 62 sidebar_label: Dynamic --- @@ -493,3 +493,44 @@ SELECT count(), dynamicType(d), _part FROM test GROUP BY _part, dynamicType(d) O ``` As we can see, ClickHouse kept the most frequent types `UInt64` and `Array(UInt64)` and casted all other types to `String`. + +## JSONExtract functions with Dynamic + +All `JSONExtract*` functions support `Dynamic` type: + +```sql +SELECT JSONExtract('{"a" : [1, 2, 3]}', 'a', 'Dynamic') AS dynamic, dynamicType(dynamic) AS dynamic_type; +``` + +```text +┌─dynamic─┬─dynamic_type───────────┐ +│ [1,2,3] │ Array(Nullable(Int64)) │ +└─────────┴────────────────────────┘ +``` + +```sql +SELECT JSONExtract('{"obj" : {"a" : 42, "b" : "Hello", "c" : [1,2,3]}}', 'obj', 'Map(String, Variant(UInt32, String, Array(UInt32)))') AS map_of_dynamics, mapApply((k, v) -> (k, variantType(v)), map_of_dynamics) AS map_of_dynamic_types``` + +```text +┌─map_of_dynamics──────────────────┬─map_of_dynamic_types────────────────────────────┐ +│ {'a':42,'b':'Hello','c':[1,2,3]} │ {'a':'UInt32','b':'String','c':'Array(UInt32)'} │ +└──────────────────────────────────┴─────────────────────────────────────────────────┘ +``` + +```sql +SELECT JSONExtractKeysAndValues('{"a" : 42, "b" : "Hello", "c" : [1,2,3]}', 'Variant(UInt32, String, Array(UInt32))') AS dynamics, arrayMap(x -> (x.1, variantType(x.2)), dynamics) AS dynamic_types``` +``` + +```text +┌─dynamics───────────────────────────────┬─dynamic_types─────────────────────────────────────────┐ +│ [('a',42),('b','Hello'),('c',[1,2,3])] │ [('a','UInt32'),('b','String'),('c','Array(UInt32)')] │ +└────────────────────────────────────────┴───────────────────────────────────────────────────────┘ +``` + +### Binary output format + +In RowBinary format values of `Dynamic` type are serialized in the following format: + +```text + +``` diff --git a/docs/en/sql-reference/data-types/json.md b/docs/en/sql-reference/data-types/json.md index 39e37abad82..f218c8d0339 100644 --- a/docs/en/sql-reference/data-types/json.md +++ b/docs/en/sql-reference/data-types/json.md @@ -1,24 +1,20 @@ --- -slug: /en/sql-reference/data-types/json +slug: /en/sql-reference/data-types/object-data-type sidebar_position: 26 -sidebar_label: JSON +sidebar_label: Object Data Type +keywords: [object, data type] --- -# JSON +# Object Data Type (deprecated) -:::note -This feature is experimental and is not production-ready. If you need to work with JSON documents, consider using [this guide](/docs/en/integrations/data-ingestion/data-formats/json.md) instead. -::: +**This feature is not production-ready and is now deprecated.** If you need to work with JSON documents, consider using [this guide](/docs/en/integrations/data-formats/json/overview) instead. A new implementation to support JSON object is in progress and can be tracked [here](https://github.com/ClickHouse/ClickHouse/issues/54864). + +
Stores JavaScript Object Notation (JSON) documents in a single column. `JSON` is an alias for `Object('json')`. -:::note -The JSON data type is an obsolete feature. Do not use it. -If you want to use it, set `allow_experimental_object_type = 1`. -::: - ## Example **Example 1** @@ -49,7 +45,7 @@ SELECT o.a, o.b.c, o.b.d[3] FROM json **Example 2** -To be able to create an ordered `MergeTree` family table the sorting key has to be extracted into its column. For example, to insert a file of compressed HTTP access logs in JSON format: +To be able to create an ordered `MergeTree` family table, the sorting key has to be extracted into its column. For example, to insert a file of compressed HTTP access logs in JSON format: ```sql CREATE TABLE logs @@ -69,7 +65,7 @@ FROM file('access.json.gz', JSONAsString) ## Displaying JSON columns -When displaying a `JSON` column ClickHouse only shows the field values by default (because internally, it is represented as a tuple). You can display the field names as well by setting `output_format_json_named_tuples_as_objects = 1`: +When displaying a `JSON` column, ClickHouse only shows the field values by default (because internally, it is represented as a tuple). You can also display the field names by setting `output_format_json_named_tuples_as_objects = 1`: ```sql SET output_format_json_named_tuples_as_objects = 1 @@ -83,4 +79,5 @@ SELECT * FROM json FORMAT JSONEachRow ## Related Content +- [Using JSON in ClickHouse](/docs/en/integrations/data-formats/json) - [Getting Data Into ClickHouse - Part 2 - A JSON detour](https://clickhouse.com/blog/getting-data-into-clickhouse-part-2-json) diff --git a/docs/en/sql-reference/data-types/lowcardinality.md b/docs/en/sql-reference/data-types/lowcardinality.md index 133ac2bd72e..a92f6f4dce6 100644 --- a/docs/en/sql-reference/data-types/lowcardinality.md +++ b/docs/en/sql-reference/data-types/lowcardinality.md @@ -56,7 +56,6 @@ Functions: ## Related content -- [Reducing ClickHouse Storage Cost with the Low Cardinality Type – Lessons from an Instana Engineer](https://altinity.com/blog/2020-5-20-reducing-clickhouse-storage-cost-with-the-low-cardinality-type-lessons-from-an-instana-engineer) -- [String Optimization (video presentation in Russian)](https://youtu.be/rqf-ILRgBdY?list=PL0Z2YDlm0b3iwXCpEFiOOYmwXzVmjJfEt). [Slides in English](https://github.com/ClickHouse/clickhouse-presentations/raw/master/meetup19/string_optimization.pdf) - Blog: [Optimizing ClickHouse with Schemas and Codecs](https://clickhouse.com/blog/optimize-clickhouse-codecs-compression-schema) - Blog: [Working with time series data in ClickHouse](https://clickhouse.com/blog/working-with-time-series-data-and-functions-ClickHouse) +- [String Optimization (video presentation in Russian)](https://youtu.be/rqf-ILRgBdY?list=PL0Z2YDlm0b3iwXCpEFiOOYmwXzVmjJfEt). [Slides in English](https://github.com/ClickHouse/clickhouse-presentations/raw/master/meetup19/string_optimization.pdf) diff --git a/docs/en/sql-reference/functions/array-functions.md b/docs/en/sql-reference/functions/array-functions.md index d87ca4a0fe7..1b52440903d 100644 --- a/docs/en/sql-reference/functions/array-functions.md +++ b/docs/en/sql-reference/functions/array-functions.md @@ -3080,4 +3080,4 @@ Result: ## Distance functions -All supported functions are described in [distance functions documentation](../../sql-reference/functions/distance-functions.md). +All supported functions are described in [distance functions documentation](../../sql-reference/functions/distance-functions.md). \ No newline at end of file diff --git a/docs/en/sql-reference/functions/bit-functions.md b/docs/en/sql-reference/functions/bit-functions.md index a48893b93bf..5ab7e07fcad 100644 --- a/docs/en/sql-reference/functions/bit-functions.md +++ b/docs/en/sql-reference/functions/bit-functions.md @@ -173,7 +173,7 @@ See function [substring](string-functions.md#substring). ## bitTest -Takes any integer and converts it into [binary form](https://en.wikipedia.org/wiki/Binary_number), returns the value of a bit at specified position. The countdown starts from 0 from the right to the left. +Takes any integer and converts it into [binary form](https://en.wikipedia.org/wiki/Binary_number), returns the value of a bit at specified position. Counting is right-to-left, starting at 0. **Syntax** @@ -226,7 +226,7 @@ Result: ## bitTestAll -Returns result of [logical conjuction](https://en.wikipedia.org/wiki/Logical_conjunction) (AND operator) of all bits at given positions. The countdown starts from 0 from the right to the left. +Returns result of [logical conjuction](https://en.wikipedia.org/wiki/Logical_conjunction) (AND operator) of all bits at given positions. Counting is right-to-left, starting at 0. The conjuction for bit-wise operations: @@ -289,7 +289,7 @@ Result: ## bitTestAny -Returns result of [logical disjunction](https://en.wikipedia.org/wiki/Logical_disjunction) (OR operator) of all bits at given positions. The countdown starts from 0 from the right to the left. +Returns result of [logical disjunction](https://en.wikipedia.org/wiki/Logical_disjunction) (OR operator) of all bits at given positions. Counting is right-to-left, starting at 0. The disjunction for bit-wise operations: diff --git a/docs/en/sql-reference/functions/date-time-functions.md b/docs/en/sql-reference/functions/date-time-functions.md index b532e0de8f0..4f5e5a5d716 100644 --- a/docs/en/sql-reference/functions/date-time-functions.md +++ b/docs/en/sql-reference/functions/date-time-functions.md @@ -83,7 +83,57 @@ Result: ``` ## makeDate32 -Like [makeDate](#makedate) but produces a [Date32](../data-types/date32.md). +Creates a date of type [Date32](../../sql-reference/data-types/date32.md) from a year, month, day (or optionally a year and a day). + +**Syntax** + +```sql +makeDate32(year, [month,] day) +``` + +**Arguments** + +- `year` — Year. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). +- `month` — Month (optional). [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). +- `day` — Day. [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). + +:::note +If `month` is omitted then `day` should take a value between `1` and `365`, otherwise it should take a value between `1` and `31`. +::: + +**Returned values** + +- A date created from the arguments. [Date32](../../sql-reference/data-types/date32.md). + +**Examples** + +Create a date from a year, month, and day: + +Query: + +```sql +SELECT makeDate32(2024, 1, 1); +``` + +Result: + +```response +2024-01-01 +``` + +Create a Date from a year and day of year: + +Query: + +``` sql +SELECT makeDate32(2024, 100); +``` + +Result: + +```response +2024-04-09 +``` ## makeDateTime @@ -125,12 +175,38 @@ Result: ## makeDateTime64 -Like [makeDateTime](#makedatetime) but produces a [DateTime64](../data-types/datetime64.md). +Creates a [DateTime64](../../sql-reference/data-types/datetime64.md) data type value from its components: year, month, day, hour, minute, second. With optional sub-second precision. **Syntax** +```sql +makeDateTime64(year, month, day, hour, minute, second[, precision]) +``` + +**Arguments** + +- `year` — Year (0-9999). [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). +- `month` — Month (1-12). [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). +- `day` — Day (1-31). [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). +- `hour` — Hour (0-23). [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). +- `minute` — Minute (0-59). [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). +- `second` — Second (0-59). [Integer](../../sql-reference/data-types/int-uint.md), [Float](../../sql-reference/data-types/float.md) or [Decimal](../../sql-reference/data-types/decimal.md). +- `precision` — Optional precision of the sub-second component (0-9). [Integer](../../sql-reference/data-types/int-uint.md). + +**Returned value** + +- A date and time created from the supplied arguments. [DateTime64](../../sql-reference/data-types/datetime64.md). + +**Example** + ``` sql -makeDateTime64(year, month, day, hour, minute, second[, fraction[, precision[, timezone]]]) +SELECT makeDateTime64(2023, 5, 15, 10, 30, 45, 779, 5); +``` + +```response +┌─makeDateTime64(2023, 5, 15, 10, 30, 45, 779, 5)─┐ +│ 2023-05-15 10:30:45.00779 │ +└─────────────────────────────────────────────────┘ ``` ## timestamp @@ -2622,6 +2698,204 @@ Like function `YYYYMMDDhhmmssToDate()` but produces a [DateTime64](../data-types Accepts an additional, optional `precision` parameter after the `timezone` parameter. +## changeYear + +Changes the year component of a date or date time. + +**Syntax** +``` sql + +changeYear(date_or_datetime, value) +``` + +**Arguments** + +- `date_or_datetime` - a [Date](../data-types/date.md), [Date32](../data-types/date32.md), [DateTime](../data-types/datetime.md) or [DateTime64](../data-types/datetime64.md) +- `value` - a new value of the year. [Integer](../../sql-reference/data-types/int-uint.md). + +**Returned value** + +- The same type as `date_or_datetime`. + +**Example** + +``` sql +SELECT changeYear(toDate('1999-01-01'), 2000), changeYear(toDateTime64('1999-01-01 00:00:00.000', 3), 2000); +``` + +Result: + +``` +┌─changeYear(toDate('1999-01-01'), 2000)─┬─changeYear(toDateTime64('1999-01-01 00:00:00.000', 3), 2000)─┐ +│ 2000-01-01 │ 2000-01-01 00:00:00.000 │ +└────────────────────────────────────────┴──────────────────────────────────────────────────────────────┘ +``` + +## changeMonth + +Changes the month component of a date or date time. + +**Syntax** + +``` sql +changeMonth(date_or_datetime, value) +``` + +**Arguments** + +- `date_or_datetime` - a [Date](../data-types/date.md), [Date32](../data-types/date32.md), [DateTime](../data-types/datetime.md) or [DateTime64](../data-types/datetime64.md) +- `value` - a new value of the month. [Integer](../../sql-reference/data-types/int-uint.md). + +**Returned value** + +- Returns a value of same type as `date_or_datetime`. + +**Example** + +``` sql +SELECT changeMonth(toDate('1999-01-01'), 2), changeMonth(toDateTime64('1999-01-01 00:00:00.000', 3), 2); +``` + +Result: + +``` +┌─changeMonth(toDate('1999-01-01'), 2)─┬─changeMonth(toDateTime64('1999-01-01 00:00:00.000', 3), 2)─┐ +│ 1999-02-01 │ 1999-02-01 00:00:00.000 │ +└──────────────────────────────────────┴────────────────────────────────────────────────────────────┘ +``` + +## changeDay + +Changes the day component of a date or date time. + +**Syntax** + +``` sql +changeDay(date_or_datetime, value) +``` + +**Arguments** + +- `date_or_datetime` - a [Date](../data-types/date.md), [Date32](../data-types/date32.md), [DateTime](../data-types/datetime.md) or [DateTime64](../data-types/datetime64.md) +- `value` - a new value of the day. [Integer](../../sql-reference/data-types/int-uint.md). + +**Returned value** + +- Returns a value of same type as `date_or_datetime`. + +**Example** + +``` sql +SELECT changeDay(toDate('1999-01-01'), 5), changeDay(toDateTime64('1999-01-01 00:00:00.000', 3), 5); +``` + +Result: + +``` +┌─changeDay(toDate('1999-01-01'), 5)─┬─changeDay(toDateTime64('1999-01-01 00:00:00.000', 3), 5)─┐ +│ 1999-01-05 │ 1999-01-05 00:00:00.000 │ +└────────────────────────────────────┴──────────────────────────────────────────────────────────┘ +``` + +## changeHour + +Changes the hour component of a date or date time. + +**Syntax** + +``` sql +changeHour(date_or_datetime, value) +``` + +**Arguments** + +- `date_or_datetime` - a [Date](../data-types/date.md), [Date32](../data-types/date32.md), [DateTime](../data-types/datetime.md) or [DateTime64](../data-types/datetime64.md) +- `value` - a new value of the hour. [Integer](../../sql-reference/data-types/int-uint.md). + +**Returned value** + +- Returns a value of same type as `date_or_datetime`. If the input is a [Date](../data-types/date.md), return [DateTime](../data-types/datetime.md). If the input is a [Date32](../data-types/date32.md), return [DateTime64](../data-types/datetime64.md). + +**Example** + +``` sql +SELECT changeHour(toDate('1999-01-01'), 14), changeHour(toDateTime64('1999-01-01 00:00:00.000', 3), 14); +``` + +Result: + +``` +┌─changeHour(toDate('1999-01-01'), 14)─┬─changeHour(toDateTime64('1999-01-01 00:00:00.000', 3), 14)─┐ +│ 1999-01-01 14:00:00 │ 1999-01-01 14:00:00.000 │ +└──────────────────────────────────────┴────────────────────────────────────────────────────────────┘ +``` + +## changeMinute + +Changes the minute component of a date or date time. + +**Syntax** + +``` sql +changeMinute(date_or_datetime, value) +``` + +**Arguments** + +- `date_or_datetime` - a [Date](../data-types/date.md), [Date32](../data-types/date32.md), [DateTime](../data-types/datetime.md) or [DateTime64](../data-types/datetime64.md) +- `value` - a new value of the minute. [Integer](../../sql-reference/data-types/int-uint.md). + +**Returned value** + +- Returns a value of same type as `date_or_datetime`. If the input is a [Date](../data-types/date.md), return [DateTime](../data-types/datetime.md). If the input is a [Date32](../data-types/date32.md), return [DateTime64](../data-types/datetime64.md). + +**Example** + +``` sql + SELECT changeMinute(toDate('1999-01-01'), 15), changeMinute(toDateTime64('1999-01-01 00:00:00.000', 3), 15); +``` + +Result: + +``` +┌─changeMinute(toDate('1999-01-01'), 15)─┬─changeMinute(toDateTime64('1999-01-01 00:00:00.000', 3), 15)─┐ +│ 1999-01-01 00:15:00 │ 1999-01-01 00:15:00.000 │ +└────────────────────────────────────────┴──────────────────────────────────────────────────────────────┘ +``` + +## changeSecond + +Changes the second component of a date or date time. + +**Syntax** + +``` sql +changeSecond(date_or_datetime, value) +``` + +**Arguments** + +- `date_or_datetime` - a [Date](../data-types/date.md), [Date32](../data-types/date32.md), [DateTime](../data-types/datetime.md) or [DateTime64](../data-types/datetime64.md) +- `value` - a new value of the second. [Integer](../../sql-reference/data-types/int-uint.md). + +**Returned value** + +- Returns a value of same type as `date_or_datetime`. If the input is a [Date](../data-types/date.md), return [DateTime](../data-types/datetime.md). If the input is a [Date32](../data-types/date32.md), return [DateTime64](../data-types/datetime64.md). + +**Example** + +``` sql +SELECT changeSecond(toDate('1999-01-01'), 15), changeSecond(toDateTime64('1999-01-01 00:00:00.000', 3), 15); +``` + +Result: + +``` +┌─changeSecond(toDate('1999-01-01'), 15)─┬─changeSecond(toDateTime64('1999-01-01 00:00:00.000', 3), 15)─┐ +│ 1999-01-01 00:00:15 │ 1999-01-01 00:00:15.000 │ +└────────────────────────────────────────┴──────────────────────────────────────────────────────────────┘ +``` + ## addYears Adds a specified number of years to a date, a date with time or a string-encoded date / date with time. @@ -2638,6 +2912,7 @@ addYears(date, num) - `num`: Number of years to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` plus `num` years. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -2675,6 +2950,7 @@ addQuarters(date, num) - `num`: Number of quarters to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` plus `num` quarters. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -2712,6 +2988,7 @@ addMonths(date, num) - `num`: Number of months to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` plus `num` months. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -2749,6 +3026,7 @@ addWeeks(date, num) - `num`: Number of weeks to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` plus `num` weeks. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -2786,6 +3064,7 @@ addDays(date, num) - `num`: Number of days to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` plus `num` days. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -2823,6 +3102,7 @@ addHours(date, num) - `num`: Number of hours to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** +o - Returns `date` plus `num` hours. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -2860,6 +3140,7 @@ addMinutes(date, num) - `num`: Number of minutes to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` plus `num` minutes. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -2897,6 +3178,7 @@ addSeconds(date, num) - `num`: Number of seconds to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` plus `num` seconds. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -2934,6 +3216,7 @@ addMilliseconds(date_time, num) - `num`: Number of milliseconds to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date_time` plus `num` milliseconds. [DateTime64](../data-types/datetime64.md). **Example** @@ -2969,6 +3252,7 @@ addMicroseconds(date_time, num) - `num`: Number of microseconds to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date_time` plus `num` microseconds. [DateTime64](../data-types/datetime64.md). **Example** @@ -3004,6 +3288,7 @@ addNanoseconds(date_time, num) - `num`: Number of nanoseconds to add. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date_time` plus `num` nanoseconds. [DateTime64](../data-types/datetime64.md). **Example** @@ -3039,6 +3324,7 @@ addInterval(interval_1, interval_2) - `interval_2`: Second interval to be added. [interval](../data-types/special-data-types/interval.md). **Returned value** + - Returns a tuple of intervals. [tuple](../data-types/tuple.md)([interval](../data-types/special-data-types/interval.md)). :::note @@ -3085,6 +3371,7 @@ addTupleOfIntervals(interval_1, interval_2) - `intervals`: Tuple of intervals to add to `date`. [tuple](../data-types/tuple.md)([interval](../data-types/special-data-types/interval.md)). **Returned value** + - Returns `date` with added `intervals`. [date](../data-types/date.md)/[date32](../data-types/date32.md)/[datetime](../data-types/datetime.md)/[datetime64](../data-types/datetime64.md). **Example** @@ -3119,6 +3406,7 @@ subtractYears(date, num) - `num`: Number of years to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` minus `num` years. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -3156,6 +3444,7 @@ subtractQuarters(date, num) - `num`: Number of quarters to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` minus `num` quarters. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -3193,6 +3482,7 @@ subtractMonths(date, num) - `num`: Number of months to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` minus `num` months. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -3230,6 +3520,7 @@ subtractWeeks(date, num) - `num`: Number of weeks to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` minus `num` weeks. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -3267,6 +3558,7 @@ subtractDays(date, num) - `num`: Number of days to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` minus `num` days. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -3304,6 +3596,7 @@ subtractHours(date, num) - `num`: Number of hours to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` minus `num` hours. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[Datetime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -3341,6 +3634,7 @@ subtractMinutes(date, num) - `num`: Number of minutes to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` minus `num` minutes. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -3378,6 +3672,7 @@ subtractSeconds(date, num) - `num`: Number of seconds to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date` minus `num` seconds. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** @@ -3415,6 +3710,7 @@ subtractMilliseconds(date_time, num) - `num`: Number of milliseconds to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date_time` minus `num` milliseconds. [DateTime64](../data-types/datetime64.md). **Example** @@ -3450,6 +3746,7 @@ subtractMicroseconds(date_time, num) - `num`: Number of microseconds to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date_time` minus `num` microseconds. [DateTime64](../data-types/datetime64.md). **Example** @@ -3485,6 +3782,7 @@ subtractNanoseconds(date_time, num) - `num`: Number of nanoseconds to subtract. [(U)Int*](../data-types/int-uint.md), [Float*](../data-types/float.md). **Returned value** + - Returns `date_time` minus `num` nanoseconds. [DateTime64](../data-types/datetime64.md). **Example** @@ -3520,6 +3818,7 @@ subtractInterval(interval_1, interval_2) - `interval_2`: Second interval to be negated. [interval](../data-types/special-data-types/interval.md). **Returned value** + - Returns a tuple of intervals. [tuple](../data-types/tuple.md)([interval](../data-types/special-data-types/interval.md)). :::note @@ -3566,6 +3865,7 @@ subtractTupleOfIntervals(interval_1, interval_2) - `intervals`: Tuple of intervals to subtract from `date`. [tuple](../data-types/tuple.md)([interval](../data-types/special-data-types/interval.md)). **Returned value** + - Returns `date` with subtracted `intervals`. [Date](../data-types/date.md)/[Date32](../data-types/date32.md)/[DateTime](../data-types/datetime.md)/[DateTime64](../data-types/datetime64.md). **Example** diff --git a/docs/en/sql-reference/functions/hash-functions.md b/docs/en/sql-reference/functions/hash-functions.md index e431ed75465..7c977e7d6dc 100644 --- a/docs/en/sql-reference/functions/hash-functions.md +++ b/docs/en/sql-reference/functions/hash-functions.md @@ -314,10 +314,71 @@ SELECT groupBitXor(cityHash64(*)) FROM table Calculates a 32-bit hash code from any type of integer. This is a relatively fast non-cryptographic hash function of average quality for numbers. +**Syntax** + +```sql +intHash32(int) +``` + +**Arguments** + +- `int` — Integer to hash. [(U)Int*](../data-types/int-uint.md). + +**Returned value** + +- 32-bit hash code. [UInt32](../data-types/int-uint.md). + +**Example** + +Query: + +```sql +SELECT intHash32(42); +``` + +Result: + +```response +┌─intHash32(42)─┐ +│ 1228623923 │ +└───────────────┘ +``` + ## intHash64 Calculates a 64-bit hash code from any type of integer. -It works faster than intHash32. Average quality. +This is a relatively fast non-cryptographic hash function of average quality for numbers. +It works faster than [intHash32](#inthash32). + +**Syntax** + +```sql +intHash64(int) +``` + +**Arguments** + +- `int` — Integer to hash. [(U)Int*](../data-types/int-uint.md). + +**Returned value** + +- 64-bit hash code. [UInt64](../data-types/int-uint.md). + +**Example** + +Query: + +```sql +SELECT intHash64(42); +``` + +Result: + +```response +┌────────intHash64(42)─┐ +│ 11490350930367293593 │ +└──────────────────────┘ +``` ## SHA1, SHA224, SHA256, SHA512, SHA512_256 diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md index 58fc1eba02e..012b0db96b5 100644 --- a/docs/en/sql-reference/functions/other-functions.md +++ b/docs/en/sql-reference/functions/other-functions.md @@ -76,7 +76,7 @@ WHERE macro = 'test'; └───────┴──────────────┘ ``` -## FQDN +## fqdn Returns the fully qualified domain name of the ClickHouse server. @@ -86,7 +86,7 @@ Returns the fully qualified domain name of the ClickHouse server. fqdn(); ``` -This function is case-insensitive. +Aliases: `fullHostName`, `FQDN`. **Returned value** @@ -2102,14 +2102,14 @@ Result: └─────────────────┘ ``` -## filesystemFree +## filesystemUnreserved -Returns the total amount of the free space on the filesystem hosting the database persistence. See also `filesystemAvailable` +Returns the total amount of the free space on the filesystem hosting the database persistence. (previously `filesystemFree`). See also [`filesystemAvailable`](#filesystemavailable). **Syntax** ```sql -filesystemFree() +filesystemUnreserved() ``` **Returned value** @@ -2121,7 +2121,7 @@ filesystemFree() Query: ```sql -SELECT formatReadableSize(filesystemFree()) AS "Free space"; +SELECT formatReadableSize(filesystemUnreserved()) AS "Free space"; ``` Result: @@ -2449,11 +2449,11 @@ As you can see, `runningAccumulate` merges states for each group of rows separat ## joinGet -The function lets you extract data from the table the same way as from a [dictionary](../../sql-reference/dictionaries/index.md). - -Gets the data from [Join](../../engines/table-engines/special/join.md#creating-a-table) tables using the specified join key. +The function lets you extract data from the table the same way as from a [dictionary](../../sql-reference/dictionaries/index.md). Gets the data from [Join](../../engines/table-engines/special/join.md#creating-a-table) tables using the specified join key. +:::note Only supports tables created with the `ENGINE = Join(ANY, LEFT, )` statement. +::: **Syntax** @@ -2463,26 +2463,32 @@ joinGet(join_storage_table_name, `value_column`, join_keys) **Arguments** -- `join_storage_table_name` — an [identifier](../../sql-reference/syntax.md#syntax-identifiers) indicating where the search is performed. The identifier is searched in the default database (see setting `default_database` in the config file). To override the default database, use `USE db_name` or specify the database and the table through the separator `db_name.db_table` as in the example. +- `join_storage_table_name` — an [identifier](../../sql-reference/syntax.md#syntax-identifiers) indicating where the search is performed. - `value_column` — name of the column of the table that contains required data. - `join_keys` — list of keys. +:::note +The identifier is searched for in the default database (see setting `default_database` in the config file). To override the default database, use `USE db_name` or specify the database and the table through the separator `db_name.db_table` as in the example. +::: + **Returned value** -Returns a list of values corresponded to list of keys. - -If certain does not exist in source table then `0` or `null` will be returned based on [join_use_nulls](../../operations/settings/settings.md#join_use_nulls) setting. +- Returns a list of values corresponded to the list of keys. +:::note +If a certain key does not exist in source table then `0` or `null` will be returned based on [join_use_nulls](../../operations/settings/settings.md#join_use_nulls) setting during table creation. More info about `join_use_nulls` in [Join operation](../../engines/table-engines/special/join.md). +::: **Example** Input table: ```sql -CREATE DATABASE db_test -CREATE TABLE db_test.id_val(`id` UInt32, `val` UInt32) ENGINE = Join(ANY, LEFT, id) SETTINGS join_use_nulls = 1 -INSERT INTO db_test.id_val VALUES (1,11)(2,12)(4,13) +CREATE DATABASE db_test; +CREATE TABLE db_test.id_val(`id` UInt32, `val` UInt32) ENGINE = Join(ANY, LEFT, id); +INSERT INTO db_test.id_val VALUES (1, 11)(2, 12)(4, 13); +SELECT * FROM db_test.id_val; ``` ```text @@ -2496,18 +2502,116 @@ INSERT INTO db_test.id_val VALUES (1,11)(2,12)(4,13) Query: ```sql -SELECT joinGet(db_test.id_val, 'val', toUInt32(number)) from numbers(4) SETTINGS join_use_nulls = 1 +SELECT number, joinGet(db_test.id_val, 'val', toUInt32(number)) from numbers(4); ``` Result: ```text -┌─joinGet(db_test.id_val, 'val', toUInt32(number))─┐ -│ 0 │ -│ 11 │ -│ 12 │ -│ 0 │ -└──────────────────────────────────────────────────┘ + ┌─number─┬─joinGet('db_test.id_val', 'val', toUInt32(number))─┐ +1. │ 0 │ 0 │ +2. │ 1 │ 11 │ +3. │ 2 │ 12 │ +4. │ 3 │ 0 │ + └────────┴────────────────────────────────────────────────────┘ +``` + +Setting `join_use_nulls` can be used during table creation to change the behaviour of what gets returned if no key exists in the source table. + +```sql +CREATE DATABASE db_test; +CREATE TABLE db_test.id_val_nulls(`id` UInt32, `val` UInt32) ENGINE = Join(ANY, LEFT, id) SETTINGS join_use_nulls=1; +INSERT INTO db_test.id_val_nulls VALUES (1, 11)(2, 12)(4, 13); +SELECT * FROM db_test.id_val_nulls; +``` + +```text +┌─id─┬─val─┐ +│ 4 │ 13 │ +│ 2 │ 12 │ +│ 1 │ 11 │ +└────┴─────┘ +``` + +Query: + +```sql +SELECT number, joinGet(db_test.id_val_nulls, 'val', toUInt32(number)) from numbers(4); +``` + +Result: + +```text + ┌─number─┬─joinGet('db_test.id_val_nulls', 'val', toUInt32(number))─┐ +1. │ 0 │ ᴺᵁᴸᴸ │ +2. │ 1 │ 11 │ +3. │ 2 │ 12 │ +4. │ 3 │ ᴺᵁᴸᴸ │ + └────────┴──────────────────────────────────────────────────────────┘ +``` + +## joinGetOrNull + +Like [joinGet](#joinget) but returns `NULL` when the key is missing instead of returning the default value. + +**Syntax** + +```sql +joinGetOrNull(join_storage_table_name, `value_column`, join_keys) +``` + +**Arguments** + +- `join_storage_table_name` — an [identifier](../../sql-reference/syntax.md#syntax-identifiers) indicating where the search is performed. +- `value_column` — name of the column of the table that contains required data. +- `join_keys` — list of keys. + +:::note +The identifier is searched for in the default database (see setting `default_database` in the config file). To override the default database, use `USE db_name` or specify the database and the table through the separator `db_name.db_table` as in the example. +::: + +**Returned value** + +- Returns a list of values corresponded to the list of keys. + +:::note +If a certain key does not exist in source table then `NULL` is returned for that key. +::: + +**Example** + +Input table: + +```sql +CREATE DATABASE db_test; +CREATE TABLE db_test.id_val(`id` UInt32, `val` UInt32) ENGINE = Join(ANY, LEFT, id); +INSERT INTO db_test.id_val VALUES (1, 11)(2, 12)(4, 13); +SELECT * FROM db_test.id_val; +``` + +```text +┌─id─┬─val─┐ +│ 4 │ 13 │ +│ 2 │ 12 │ +│ 1 │ 11 │ +└────┴─────┘ +``` + +Query: + +```sql +SELECT number, joinGetOrNull(db_test.id_val, 'val', toUInt32(number)) from numbers(4); +``` + +Result: + +```text + ┌─number─┬─joinGetOrNull('db_test.id_val', 'val', toUInt32(number))─┐ +1. │ 0 │ ᴺᵁᴸᴸ │ +2. │ 1 │ 11 │ +3. │ 2 │ 12 │ +4. │ 3 │ ᴺᵁᴸᴸ │ + └────────┴──────────────────────────────────────────────────────────┘ ``` ## catboostEvaluate @@ -2984,6 +3088,66 @@ Result: └─────────┘ ``` +## partitionID + +Computes the [partition ID](../../engines/table-engines/mergetree-family/custom-partitioning-key.md). + +:::note +This function is slow and should not be called for large amount of rows. +::: + +**Syntax** + +```sql +partitionID(x[, y, ...]); +``` + +**Arguments** + +- `x` — Column for which to return the partition ID. +- `y, ...` — Remaining N columns for which to return the partition ID (optional). + +**Returned Value** + +- Partition ID that the row would belong to. [String](../data-types/string.md). + +**Example** + +Query: + +```sql +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab +( + i int, + j int +) +ENGINE = MergeTree +PARTITION BY i +ORDER BY tuple(); + +INSERT INTO tab VALUES (1, 1), (1, 2), (1, 3), (2, 4), (2, 5), (2, 6); + +SELECT i, j, partitionID(i), _partition_id FROM tab ORDER BY i, j; +``` + +Result: + +```response +┌─i─┬─j─┬─partitionID(i)─┬─_partition_id─┐ +│ 1 │ 1 │ 1 │ 1 │ +│ 1 │ 2 │ 1 │ 1 │ +│ 1 │ 3 │ 1 │ 1 │ +└───┴───┴────────────────┴───────────────┘ +┌─i─┬─j─┬─partitionID(i)─┬─_partition_id─┐ +│ 2 │ 4 │ 2 │ 2 │ +│ 2 │ 5 │ 2 │ 2 │ +│ 2 │ 6 │ 2 │ 2 │ +└───┴───┴────────────────┴───────────────┘ +``` + + ## shardNum Returns the index of a shard which processes a part of data in a distributed query. Indices are started from `1`. @@ -3860,3 +4024,138 @@ Result: └───────────────┘ ``` +## transactionID + +Returns the ID of a [transaction](https://clickhouse.com/docs/en/guides/developer/transactional#transactions-commit-and-rollback). + +:::note +This function is part of an experimental feature set. Enable experimental transaction support by adding this setting to your configuration: + +``` + + 1 + +``` + +For more information see the page [Transactional (ACID) support](https://clickhouse.com/docs/en/guides/developer/transactional#transactions-commit-and-rollback). +::: + +**Syntax** + +```sql +transactionID() +``` + +**Returned value** + +- Returns a tuple consisting of `start_csn`, `local_tid` and `host_id`. [Tuple](../data-types/tuple.md). + +- `start_csn`: Global sequential number, the newest commit timestamp that was seen when this transaction began. [UInt64](../data-types/int-uint.md). +- `local_tid`: Local sequential number that is unique for each transaction started by this host within a specific start_csn. [UInt64](../data-types/int-uint.md). +- `host_id`: UUID of the host that has started this transaction. [UUID](../data-types/uuid.md). + +**Example** + +Query: + +```sql +BEGIN TRANSACTION; +SELECT transactionID(); +ROLLBACK; +``` + +Result: + +```response +┌─transactionID()────────────────────────────────┐ +│ (32,34,'0ee8b069-f2bb-4748-9eae-069c85b5252b') │ +└────────────────────────────────────────────────┘ +``` + +## transactionLatestSnapshot + +Returns the newest snapshot (Commit Sequence Number) of a [transaction](https://clickhouse.com/docs/en/guides/developer/transactional#transactions-commit-and-rollback) that is available for reading. + +:::note +This function is part of an experimental feature set. Enable experimental transaction support by adding this setting to your configuration: + +``` + + 1 + +``` + +For more information see the page [Transactional (ACID) support](https://clickhouse.com/docs/en/guides/developer/transactional#transactions-commit-and-rollback). +::: + +**Syntax** + +```sql +transactionLatestSnapshot() +``` + +**Returned value** + +- Returns the latest snapshot (CSN) of a transaction. [UInt64](../data-types/int-uint.md) + +**Example** + +Query: + +```sql +BEGIN TRANSACTION; +SELECT transactionLatestSnapshot(); +ROLLBACK; +``` + +Result: + +```response +┌─transactionLatestSnapshot()─┐ +│ 32 │ +└─────────────────────────────┘ +``` + +## transactionOldestSnapshot + +Returns the oldest snapshot (Commit Sequence Number) that is visible for some running [transaction](https://clickhouse.com/docs/en/guides/developer/transactional#transactions-commit-and-rollback). + +:::note +This function is part of an experimental feature set. Enable experimental transaction support by adding this setting to your configuration: + +``` + + 1 + +``` + +For more information see the page [Transactional (ACID) support](https://clickhouse.com/docs/en/guides/developer/transactional#transactions-commit-and-rollback). +::: + +**Syntax** + +```sql +transactionOldestSnapshot() +``` + +**Returned value** + +- Returns the oldest snapshot (CSN) of a transaction. [UInt64](../data-types/int-uint.md) + +**Example** + +Query: + +```sql +BEGIN TRANSACTION; +SELECT transactionLatestSnapshot(); +ROLLBACK; +``` + +Result: + +```response +┌─transactionOldestSnapshot()─┐ +│ 32 │ +└─────────────────────────────┘ +``` diff --git a/docs/en/sql-reference/functions/string-functions.md b/docs/en/sql-reference/functions/string-functions.md index a258456345e..e2f1b8c7f14 100644 --- a/docs/en/sql-reference/functions/string-functions.md +++ b/docs/en/sql-reference/functions/string-functions.md @@ -12,9 +12,7 @@ Functions for [searching](string-search-functions.md) in strings and for [replac ## empty -Checks whether the input string is empty. - -A string is considered non-empty if it contains at least one byte, even if this byte is a space or the null byte. +Checks whether the input string is empty. A string is considered non-empty if it contains at least one byte, even if this byte is a space or the null byte. The function is also available for [arrays](array-functions.md#function-empty) and [UUIDs](uuid-functions.md#empty). @@ -48,9 +46,7 @@ Result: ## notEmpty -Checks whether the input string is non-empty. - -A string is considered non-empty if it contains at least one byte, even if this byte is a space or the null byte. +Checks whether the input string is non-empty. A string is considered non-empty if it contains at least one byte, even if this byte is a space or the null byte. The function is also available for [arrays](array-functions.md#function-notempty) and [UUIDs](uuid-functions.md#notempty). @@ -96,7 +92,7 @@ length(s) **Parameters** -- `s`: An input string or array. [String](../data-types/string)/[Array](../data-types/array). +- `s` — An input string or array. [String](../data-types/string)/[Array](../data-types/array). **Returned value** @@ -149,7 +145,7 @@ lengthUTF8(s) **Parameters** -- `s`: String containing valid UTF-8 encoded text. [String](../data-types/string). +- `s` — String containing valid UTF-8 encoded text. [String](../data-types/string). **Returned value** @@ -183,8 +179,8 @@ left(s, offset) **Parameters** -- `s`: The string to calculate a substring from. [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md). -- `offset`: The number of bytes of the offset. [UInt*](../data-types/int-uint). +- `s` — The string to calculate a substring from. [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md). +- `offset` — The number of bytes of the offset. [UInt*](../data-types/int-uint). **Returned value** @@ -230,8 +226,8 @@ leftUTF8(s, offset) **Parameters** -- `s`: The UTF-8 encoded string to calculate a substring from. [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md). -- `offset`: The number of bytes of the offset. [UInt*](../data-types/int-uint). +- `s` — The UTF-8 encoded string to calculate a substring from. [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md). +- `offset` — The number of bytes of the offset. [UInt*](../data-types/int-uint). **Returned value** @@ -347,8 +343,8 @@ right(s, offset) **Parameters** -- `s`: The string to calculate a substring from. [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md). -- `offset`: The number of bytes of the offset. [UInt*](../data-types/int-uint). +- `s` — The string to calculate a substring from. [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md). +- `offset` — The number of bytes of the offset. [UInt*](../data-types/int-uint). **Returned value** @@ -394,8 +390,8 @@ rightUTF8(s, offset) **Parameters** -- `s`: The UTF-8 encoded string to calculate a substring from. [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md). -- `offset`: The number of bytes of the offset. [UInt*](../data-types/int-uint). +- `s` — The UTF-8 encoded string to calculate a substring from. [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md). +- `offset` — The number of bytes of the offset. [UInt*](../data-types/int-uint). **Returned value** @@ -547,7 +543,7 @@ Alias: `ucase` **Parameters** -- `input`: A string type [String](../data-types/string.md). +- `input` — A string type [String](../data-types/string.md). **Returned value** @@ -571,17 +567,47 @@ SELECT upper('clickhouse'); Converts a string to lowercase, assuming that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. -Does not detect the language, e.g. for Turkish the result might not be exactly correct (i/İ vs. i/I). +:::note +Does not detect the language, e.g. for Turkish the result might not be exactly correct (i/İ vs. i/I). If the length of the UTF-8 byte sequence is different for upper and lower case of a code point (such as `ẞ` and `ß`), the result may be incorrect for this code point. +::: -If the length of the UTF-8 byte sequence is different for upper and lower case of a code point, the result may be incorrect for this code point. +**Syntax** + +```sql +lowerUTF8(input) +``` + +**Parameters** + +- `input` — A string type [String](../data-types/string.md). + +**Returned value** + +- A [String](../data-types/string.md) data type value. + +**Example** + +Query: + +``` sql +SELECT lowerUTF8('MÜNCHEN') as Lowerutf8; +``` + +Result: + +``` response +┌─Lowerutf8─┐ +│ münchen │ +└───────────┘ +``` ## upperUTF8 Converts a string to uppercase, assuming that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. -Does not detect the language, e.g. for Turkish the result might not be exactly correct (i/İ vs. i/I). - -If the length of the UTF-8 byte sequence is different for upper and lower case of a code point, the result may be incorrect for this code point. +:::note +Does not detect the language, e.g. for Turkish the result might not be exactly correct (i/İ vs. i/I). If the length of the UTF-8 byte sequence is different for upper and lower case of a code point (such as `ẞ` and `ß`), the result may be incorrect for this code point. +::: **Syntax** @@ -591,7 +617,7 @@ upperUTF8(input) **Parameters** -- `input`: A string type [String](../data-types/string.md). +- `input` — A string type [String](../data-types/string.md). **Returned value** @@ -605,6 +631,8 @@ Query: SELECT upperUTF8('München') as Upperutf8; ``` +Result: + ``` response ┌─Upperutf8─┐ │ MÜNCHEN │ @@ -615,6 +643,34 @@ SELECT upperUTF8('München') as Upperutf8; Returns 1, if the set of bytes constitutes valid UTF-8-encoded text, otherwise 0. +**Syntax** + +``` sql +isValidUTF8(input) +``` + +**Parameters** + +- `input` — A string type [String](../data-types/string.md). + +**Returned value** + +- Returns `1`, if the set of bytes constitutes valid UTF-8-encoded text, otherwise `0`. + +Query: + +``` sql +SELECT isValidUTF8('\xc3\xb1') AS valid, isValidUTF8('\xc3\x28') AS invalid; +``` + +Result: + +``` response +┌─valid─┬─invalid─┐ +│ 1 │ 0 │ +└───────┴─────────┘ +``` + ## toValidUTF8 Replaces invalid UTF-8 characters by the `�` (U+FFFD) character. All running in a row invalid characters are collapsed into the one replacement character. @@ -736,7 +792,7 @@ concat(s1, s2, ...) **Arguments** -At least one value of arbitrary type. +Values of arbitrary type. Arguments which are not of types [String](../data-types/string.md) or [FixedString](../data-types/fixedstring.md) are converted to strings using their default serialization. As this decreases performance, it is not recommended to use non-String/FixedString arguments. @@ -884,7 +940,7 @@ Returns the substring of a string `s` which starts at the specified byte index ` substring(s, offset[, length]) ``` -Alias: +Aliases: - `substr` - `mid` - `byteSlice` @@ -927,9 +983,9 @@ substringUTF8(s, offset[, length]) **Arguments** -- `s`: The string to calculate a substring from. [String](../data-types/string.md), [FixedString](../data-types/fixedstring.md) or [Enum](../data-types/enum.md) -- `offset`: The starting position of the substring in `s` . [(U)Int*](../data-types/int-uint.md). -- `length`: The maximum length of the substring. [(U)Int*](../data-types/int-uint.md). Optional. +- `s` — The string to calculate a substring from. [String](../data-types/string.md), [FixedString](../data-types/fixedstring.md) or [Enum](../data-types/enum.md) +- `offset` — The starting position of the substring in `s` . [(U)Int*](../data-types/int-uint.md). +- `length` — The maximum length of the substring. [(U)Int*](../data-types/int-uint.md). Optional. **Returned value** @@ -965,9 +1021,9 @@ Alias: `SUBSTRING_INDEX` **Arguments** -- s: The string to extract substring from. [String](../data-types/string.md). -- delim: The character to split. [String](../data-types/string.md). -- count: The number of occurrences of the delimiter to count before extracting the substring. If count is positive, everything to the left of the final delimiter (counting from the left) is returned. If count is negative, everything to the right of the final delimiter (counting from the right) is returned. [UInt or Int](../data-types/int-uint.md) +- s — The string to extract substring from. [String](../data-types/string.md). +- delim — The character to split. [String](../data-types/string.md). +- count — The number of occurrences of the delimiter to count before extracting the substring. If count is positive, everything to the left of the final delimiter (counting from the left) is returned. If count is negative, everything to the right of the final delimiter (counting from the right) is returned. [UInt or Int](../data-types/int-uint.md) **Example** @@ -996,9 +1052,9 @@ substringIndexUTF8(s, delim, count) **Arguments** -- `s`: The string to extract substring from. [String](../data-types/string.md). -- `delim`: The character to split. [String](../data-types/string.md). -- `count`: The number of occurrences of the delimiter to count before extracting the substring. If count is positive, everything to the left of the final delimiter (counting from the left) is returned. If count is negative, everything to the right of the final delimiter (counting from the right) is returned. [UInt or Int](../data-types/int-uint.md) +- `s` — The string to extract substring from. [String](../data-types/string.md). +- `delim` — The character to split. [String](../data-types/string.md). +- `count` — The number of occurrences of the delimiter to count before extracting the substring. If count is positive, everything to the left of the final delimiter (counting from the left) is returned. If count is negative, everything to the right of the final delimiter (counting from the right) is returned. [UInt or Int](../data-types/int-uint.md) **Returned value** @@ -1168,14 +1224,14 @@ Result: └────────────────────────────┘ ``` -## base64UrlEncode +## base64URLEncode Encodes an URL (String or FixedString) as base64 with URL-specific modifications, according to [RFC 4648](https://datatracker.ietf.org/doc/html/rfc4648#section-5). **Syntax** ```sql -base64UrlEncode(url) +base64URLEncode(url) ``` **Arguments** @@ -1189,13 +1245,13 @@ base64UrlEncode(url) **Example** ``` sql -SELECT base64UrlEncode('https://clickhouse.com'); +SELECT base64URLEncode('https://clickhouse.com'); ``` Result: ```result -┌─base64UrlEncode('https://clickhouse.com')─┐ +┌─base64URLEncode('https://clickhouse.com')─┐ │ aHR0cDovL2NsaWNraG91c2UuY29t │ └───────────────────────────────────────────┘ ``` @@ -1234,19 +1290,19 @@ Result: └──────────────────────────────────┘ ``` -## base64UrlDecode +## base64URLDecode Accepts a base64-encoded URL and decodes it from base64 with URL-specific modifications, according to [RFC 4648](https://datatracker.ietf.org/doc/html/rfc4648#section-5). Throws an exception in case of an error. **Syntax** ```sql -base64UrlDecode(encodedUrl) +base64URLDecode(encodedUrl) ``` **Arguments** -- `encodedUrl` — [String](../data-types/string.md) column or constant. If the string is not a valid Base64-encoded value with URL-specific modifications, an exception is thrown. +- `encodedURL` — [String](../data-types/string.md) column or constant. If the string is not a valid Base64-encoded value with URL-specific modifications, an exception is thrown. **Returned value** @@ -1255,13 +1311,13 @@ base64UrlDecode(encodedUrl) **Example** ``` sql -SELECT base64UrlDecode('aHR0cDovL2NsaWNraG91c2UuY29t'); +SELECT base64URLDecode('aHR0cDovL2NsaWNraG91c2UuY29t'); ``` Result: ```result -┌─base64UrlDecode('aHR0cDovL2NsaWNraG91c2UuY29t')─┐ +┌─base64URLDecode('aHR0cDovL2NsaWNraG91c2UuY29t')─┐ │ https://clickhouse.com │ └─────────────────────────────────────────────────┘ ``` @@ -1278,7 +1334,7 @@ tryBase64Decode(encoded) **Arguments** -- `encoded`: [String](../data-types/string.md) column or constant. If the string is not a valid Base64-encoded value, returns an empty string. +- `encoded` — [String](../data-types/string.md) column or constant. If the string is not a valid Base64-encoded value, returns an empty string. **Returned value** @@ -1298,19 +1354,19 @@ SELECT tryBase64Decode('RW5jb2RlZA==') as res, tryBase64Decode('invalid') as res └────────────┴─────────────┘ ``` -## tryBase64UrlDecode +## tryBase64URLDecode -Like `base64UrlDecode` but returns an empty string in case of error. +Like `base64URLDecode` but returns an empty string in case of error. **Syntax** ```sql -tryBase64UrlDecode(encodedUrl) +tryBase64URLDecode(encodedUrl) ``` **Parameters** -- `encodedUrl`: [String](../data-types/string.md) column or constant. If the string is not a valid Base64-encoded value with URL-specific modifications, returns an empty string. +- `encodedURL` — [String](../data-types/string.md) column or constant. If the string is not a valid Base64-encoded value with URL-specific modifications, returns an empty string. **Returned value** @@ -1321,7 +1377,7 @@ tryBase64UrlDecode(encodedUrl) Query: ```sql -SELECT tryBase64UrlDecode('aHR0cDovL2NsaWNraG91c2UuY29t') as res, tryBase64Decode('aHR0cHM6Ly9jbGlja') as res_invalid; +SELECT tryBase64URLDecode('aHR0cDovL2NsaWNraG91c2UuY29t') as res, tryBase64Decode('aHR0cHM6Ly9jbGlja') as res_invalid; ``` ```response @@ -1556,7 +1612,7 @@ The result type is UInt64. ## normalizeQuery -Replaces literals, sequences of literals and complex aliases with placeholders. +Replaces literals, sequences of literals and complex aliases (containing whitespace, more than two digits or at least 36 bytes long such as UUIDs) with placeholder `?`. **Syntax** @@ -1574,6 +1630,8 @@ normalizeQuery(x) **Example** +Query: + ``` sql SELECT normalizeQuery('[1, 2, 3, x]') AS query; ``` @@ -1586,9 +1644,44 @@ Result: └──────────┘ ``` +## normalizeQueryKeepNames + +Replaces literals, sequences of literals with placeholder `?` but does not replace complex aliases (containing whitespace, more than two digits +or at least 36 bytes long such as UUIDs). This helps better analyze complex query logs. + +**Syntax** + +``` sql +normalizeQueryKeepNames(x) +``` + +**Arguments** + +- `x` — Sequence of characters. [String](../data-types/string.md). + +**Returned value** + +- Sequence of characters with placeholders. [String](../data-types/string.md). + +**Example** + +Query: + +``` sql +SELECT normalizeQuery('SELECT 1 AS aComplexName123'), normalizeQueryKeepNames('SELECT 1 AS aComplexName123'); +``` + +Result: + +```result +┌─normalizeQuery('SELECT 1 AS aComplexName123')─┬─normalizeQueryKeepNames('SELECT 1 AS aComplexName123')─┐ +│ SELECT ? AS `?` │ SELECT ? AS aComplexName123 │ +└───────────────────────────────────────────────┴────────────────────────────────────────────────────────┘ +``` + ## normalizedQueryHash -Returns identical 64bit hash values without the values of literals for similar queries. Can be helpful to analyze query log. +Returns identical 64bit hash values without the values of literals for similar queries. Can be helpful to analyze query logs. **Syntax** @@ -1606,6 +1699,8 @@ normalizedQueryHash(x) **Example** +Query: + ``` sql SELECT normalizedQueryHash('SELECT 1 AS `xyz`') != normalizedQueryHash('SELECT 1 AS `abc`') AS res; ``` @@ -1618,6 +1713,43 @@ Result: └─────┘ ``` +## normalizedQueryHashKeepNames + +Like [normalizedQueryHash](#normalizedqueryhash) it returns identical 64bit hash values without the values of literals for similar queries but it does not replace complex aliases (containing whitespace, more than two digits +or at least 36 bytes long such as UUIDs) with a placeholder before hashing. Can be helpful to analyze query logs. + +**Syntax** + +``` sql +normalizedQueryHashKeepNames(x) +``` + +**Arguments** + +- `x` — Sequence of characters. [String](../data-types/string.md). + +**Returned value** + +- Hash value. [UInt64](../data-types/int-uint.md#uint-ranges). + +**Example** + +``` sql +SELECT normalizedQueryHash('SELECT 1 AS `xyz123`') != normalizedQueryHash('SELECT 1 AS `abc123`') AS normalizedQueryHash; +SELECT normalizedQueryHashKeepNames('SELECT 1 AS `xyz123`') != normalizedQueryHashKeepNames('SELECT 1 AS `abc123`') AS normalizedQueryHashKeepNames; +``` + +Result: + +```result +┌─normalizedQueryHash─┐ +│ 0 │ +└─────────────────────┘ +┌─normalizedQueryHashKeepNames─┐ +│ 1 │ +└──────────────────────────────┘ +``` + ## normalizeUTF8NFC Converts a string to [NFC normalized form](https://en.wikipedia.org/wiki/Unicode_equivalence#Normal_forms), assuming the string is valid UTF8-encoded text. @@ -1936,7 +2068,7 @@ soundex(val) **Arguments** -- `val` - Input value. [String](../data-types/string.md) +- `val` — Input value. [String](../data-types/string.md) **Returned value** @@ -1969,7 +2101,7 @@ punycodeEncode(val) **Arguments** -- `val` - Input value. [String](../data-types/string.md) +- `val` — Input value. [String](../data-types/string.md) **Returned value** @@ -2002,7 +2134,7 @@ punycodeEncode(val) **Arguments** -- `val` - Punycode-encoded string. [String](../data-types/string.md) +- `val` — Punycode-encoded string. [String](../data-types/string.md) **Returned value** @@ -2028,7 +2160,7 @@ Like `punycodeDecode` but returns an empty string if no valid Punycode-encoded s ## idnaEncode -Returns the the ASCII representation (ToASCII algorithm) of a domain name according to the [Internationalized Domain Names in Applications](https://en.wikipedia.org/wiki/Internationalized_domain_name#Internationalizing_Domain_Names_in_Applications) (IDNA) mechanism. +Returns the ASCII representation (ToASCII algorithm) of a domain name according to the [Internationalized Domain Names in Applications](https://en.wikipedia.org/wiki/Internationalized_domain_name#Internationalizing_Domain_Names_in_Applications) (IDNA) mechanism. The input string must be UTF-encoded and translatable to an ASCII string, otherwise an exception is thrown. Note: No percent decoding or trimming of tabs, spaces or control characters is performed. @@ -2040,7 +2172,7 @@ idnaEncode(val) **Arguments** -- `val` - Input value. [String](../data-types/string.md) +- `val` — Input value. [String](../data-types/string.md) **Returned value** @@ -2066,7 +2198,7 @@ Like `idnaEncode` but returns an empty string in case of an error instead of thr ## idnaDecode -Returns the the Unicode (UTF-8) representation (ToUnicode algorithm) of a domain name according to the [Internationalized Domain Names in Applications](https://en.wikipedia.org/wiki/Internationalized_domain_name#Internationalizing_Domain_Names_in_Applications) (IDNA) mechanism. +Returns the Unicode (UTF-8) representation (ToUnicode algorithm) of a domain name according to the [Internationalized Domain Names in Applications](https://en.wikipedia.org/wiki/Internationalized_domain_name#Internationalizing_Domain_Names_in_Applications) (IDNA) mechanism. In case of an error (e.g. because the input is invalid), the input string is returned. Note that repeated application of `idnaEncode()` and `idnaDecode()` does not necessarily return the original string due to case normalization. @@ -2078,7 +2210,7 @@ idnaDecode(val) **Arguments** -- `val` - Input value. [String](../data-types/string.md) +- `val` — Input value. [String](../data-types/string.md) **Returned value** @@ -2122,7 +2254,7 @@ Result: └───────────────────────────────────────────┘ ``` -Alias: mismatches +Alias: `mismatches` ## stringJaccardIndex @@ -2176,7 +2308,7 @@ Result: └─────────────────────────────────────┘ ``` -Alias: levenshteinDistance +Alias: `levenshteinDistance` ## editDistanceUTF8 @@ -2202,7 +2334,7 @@ Result: └─────────────────────────────────────┘ ``` -Alias: levenshteinDistanceUTF8 +Alias: `levenshteinDistanceUTF8` ## damerauLevenshteinDistance @@ -2280,13 +2412,93 @@ Result: Convert the first letter of each word to upper case and the rest to lower case. Words are sequences of alphanumeric characters separated by non-alphanumeric characters. +:::note +Because `initCap` converts only the first letter of each word to upper case you may observe unexpected behaviour for words containing apostrophes or capital letters. For example: + +```sql +SELECT initCap('mother''s daughter'), initCap('joe McAdam'); +``` + +will return + +```response +┌─initCap('mother\'s daughter')─┬─initCap('joe McAdam')─┐ +│ Mother'S Daughter │ Joe Mcadam │ +└───────────────────────────────┴───────────────────────┘ +``` + +This is a known behaviour, with no plans currently to fix it. +::: + +**Syntax** + +```sql +initcap(val) +``` + +**Arguments** + +- `val` — Input value. [String](../data-types/string.md). + +**Returned value** + +- `val` with the first letter of each word converted to upper case. [String](../data-types/string.md). + +**Example** + +Query: + +```sql +SELECT initcap('building for fast'); +``` + +Result: + +```text +┌─initcap('building for fast')─┐ +│ Building For Fast │ +└──────────────────────────────┘ +``` + ## initcapUTF8 -Like [initcap](#initcap), assuming that the string contains valid UTF-8 encoded text. If this assumption is violated, no exception is thrown and the result is undefined. - -Does not detect the language, e.g. for Turkish the result might not be exactly correct (i/İ vs. i/I). +Like [initcap](#initcap), `initcapUTF8` converts the first letter of each word to upper case and the rest to lower case. Assumes that the string contains valid UTF-8 encoded text. +If this assumption is violated, no exception is thrown and the result is undefined. +:::note +This function does not detect the language, e.g. for Turkish the result might not be exactly correct (i/İ vs. i/I). If the length of the UTF-8 byte sequence is different for upper and lower case of a code point, the result may be incorrect for this code point. +::: + +**Syntax** + +```sql +initcapUTF8(val) +``` + +**Arguments** + +- `val` — Input value. [String](../data-types/string.md). + +**Returned value** + +- `val` with the first letter of each word converted to upper case. [String](../data-types/string.md). + +**Example** + +Query: + +```sql +SELECT initcapUTF8('не тормозит'); +``` + +Result: + +```text +┌─initcapUTF8('не тормозит')─┐ +│ Не Тормозит │ +└────────────────────────────┘ +``` ## firstLine @@ -2300,7 +2512,7 @@ firstLine(val) **Arguments** -- `val` - Input value. [String](../data-types/string.md) +- `val` — Input value. [String](../data-types/string.md) **Returned value** diff --git a/docs/en/sql-reference/functions/string-replace-functions.md b/docs/en/sql-reference/functions/string-replace-functions.md index 7aeb1f5b2a7..8793ebdd1a3 100644 --- a/docs/en/sql-reference/functions/string-replace-functions.md +++ b/docs/en/sql-reference/functions/string-replace-functions.md @@ -34,7 +34,7 @@ Alias: `replace`. Replaces the first occurrence of the substring matching the regular expression `pattern` (in [re2 syntax](https://github.com/google/re2/wiki/Syntax)) in `haystack` by the `replacement` string. -`replacement` can containing substitutions `\0-\9`. +`replacement` can contain substitutions `\0-\9`. Substitutions `\1-\9` correspond to the 1st to 9th capturing group (submatch), substitution `\0` corresponds to the entire match. To use a verbatim `\` character in the `pattern` or `replacement` strings, escape it using `\`. diff --git a/docs/en/sql-reference/functions/time-window-functions.md b/docs/en/sql-reference/functions/time-window-functions.md index 2cec1987c20..5169d4487ec 100644 --- a/docs/en/sql-reference/functions/time-window-functions.md +++ b/docs/en/sql-reference/functions/time-window-functions.md @@ -6,44 +6,122 @@ sidebar_label: Time Window # Time Window Functions -Time window functions return the inclusive lower and exclusive upper bound of the corresponding window. The functions for working with WindowView are listed below: +Time window functions return the inclusive lower and exclusive upper bound of the corresponding window. The functions for working with [WindowView](../statements/create/view.md/#window-view-experimental) are listed below: ## tumble A tumbling time window assigns records to non-overlapping, continuous windows with a fixed duration (`interval`). +**Syntax** + ``` sql tumble(time_attr, interval [, timezone]) ``` **Arguments** -- `time_attr` - Date and time. [DateTime](../data-types/datetime.md) data type. -- `interval` - Window interval in [Interval](../data-types/special-data-types/interval.md) data type. +- `time_attr` — Date and time. [DateTime](../data-types/datetime.md). +- `interval` — Window interval in [Interval](../data-types/special-data-types/interval.md). - `timezone` — [Timezone name](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone) (optional). **Returned values** -- The inclusive lower and exclusive upper bound of the corresponding tumbling window. [Tuple](../data-types/tuple.md)([DateTime](../data-types/datetime.md), [DateTime](../data-types/datetime.md))`. +- The inclusive lower and exclusive upper bound of the corresponding tumbling window. [Tuple](../data-types/tuple.md)([DateTime](../data-types/datetime.md), [DateTime](../data-types/datetime.md)). **Example** Query: ``` sql -SELECT tumble(now(), toIntervalDay('1')) +SELECT tumble(now(), toIntervalDay('1')); ``` Result: ``` text ┌─tumble(now(), toIntervalDay('1'))─────────────┐ -│ ['2020-01-01 00:00:00','2020-01-02 00:00:00'] │ +│ ('2024-07-04 00:00:00','2024-07-05 00:00:00') │ └───────────────────────────────────────────────┘ ``` +## tumbleStart + +Returns the inclusive lower bound of the corresponding [tumbling window](#tumble). + +**Syntax** + +``` sql +tumbleStart(time_attr, interval [, timezone]); +``` + +**Arguments** + +- `time_attr` — Date and time. [DateTime](../data-types/datetime.md). +- `interval` — Window interval in [Interval](../data-types/special-data-types/interval.md). +- `timezone` — [Timezone name](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone) (optional). + +The parameters above can also be passed to the function as a [tuple](../data-types/tuple.md). + +**Returned values** + +- The inclusive lower bound of the corresponding tumbling window. [DateTime](../data-types/datetime.md), [Tuple](../data-types/tuple.md) or [UInt32](../data-types/int-uint.md). + +**Example** + +Query: + +```sql +SELECT tumbleStart(now(), toIntervalDay('1')); +``` + +Result: + +```response +┌─tumbleStart(now(), toIntervalDay('1'))─┐ +│ 2024-07-04 00:00:00 │ +└────────────────────────────────────────┘ +``` + +## tumbleEnd + +Returns the exclusive upper bound of the corresponding [tumbling window](#tumble). + +**Syntax** + +``` sql +tumbleEnd(time_attr, interval [, timezone]); +``` + +**Arguments** + +- `time_attr` — Date and time. [DateTime](../data-types/datetime.md). +- `interval` — Window interval in [Interval](../data-types/special-data-types/interval.md). +- `timezone` — [Timezone name](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone) (optional). + +The parameters above can also be passed to the function as a [tuple](../data-types/tuple.md). + +**Returned values** + +- The inclusive lower bound of the corresponding tumbling window. [DateTime](../data-types/datetime.md), [Tuple](../data-types/tuple.md) or [UInt32](../data-types/int-uint.md). + +**Example** + +Query: + +```sql +SELECT tumbleEnd(now(), toIntervalDay('1')); +``` + +Result: + +```response +┌─tumbleEnd(now(), toIntervalDay('1'))─┐ +│ 2024-07-05 00:00:00 │ +└──────────────────────────────────────┘ +``` + ## hop -A hopping time window has a fixed duration (`window_interval`) and hops by a specified hop interval (`hop_interval`). If the `hop_interval` is smaller than the `window_interval`, hopping windows are overlapping. Thus, records can be assigned to multiple windows. +A hopping time window has a fixed duration (`window_interval`) and hops by a specified hop interval (`hop_interval`). If the `hop_interval` is smaller than the `window_interval`, hopping windows are overlapping. Thus, records can be assigned to multiple windows. ``` sql hop(time_attr, hop_interval, window_interval [, timezone]) @@ -51,65 +129,118 @@ hop(time_attr, hop_interval, window_interval [, timezone]) **Arguments** -- `time_attr` - Date and time. [DateTime](../data-types/datetime.md) data type. -- `hop_interval` - Hop interval in [Interval](../data-types/special-data-types/interval.md) data type. Should be a positive number. -- `window_interval` - Window interval in [Interval](../data-types/special-data-types/interval.md) data type. Should be a positive number. -- `timezone` — [Timezone name](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone) (optional). +- `time_attr` — Date and time. [DateTime](../data-types/datetime.md). +- `hop_interval` — Positive Hop interval. [Interval](../data-types/special-data-types/interval.md). +- `window_interval` — Positive Window interval. [Interval](../data-types/special-data-types/interval.md). +- `timezone` — [Timezone name](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone) (optional). **Returned values** -- The inclusive lower and exclusive upper bound of the corresponding hopping window. Since one record can be assigned to multiple hop windows, the function only returns the bound of the **first** window when hop function is used **without** `WINDOW VIEW`. [Tuple](../data-types/tuple.md)([DateTime](../data-types/datetime.md), [DateTime](../data-types/datetime.md))`. +- The inclusive lower and exclusive upper bound of the corresponding hopping window. [Tuple](../data-types/tuple.md)([DateTime](../data-types/datetime.md), [DateTime](../data-types/datetime.md))`. + +:::note +Since one record can be assigned to multiple hop windows, the function only returns the bound of the **first** window when hop function is used **without** `WINDOW VIEW`. +::: **Example** Query: ``` sql -SELECT hop(now(), INTERVAL '1' SECOND, INTERVAL '2' SECOND) +SELECT hop(now(), INTERVAL '1' DAY, INTERVAL '2' DAY); ``` Result: ``` text -┌─hop(now(), toIntervalSecond('1'), toIntervalSecond('2'))──┐ -│ ('2020-01-14 16:58:22','2020-01-14 16:58:24') │ -└───────────────────────────────────────────────────────────┘ -``` - -## tumbleStart - -Returns the inclusive lower bound of the corresponding tumbling window. - -``` sql -tumbleStart(bounds_tuple); -tumbleStart(time_attr, interval [, timezone]); -``` - -## tumbleEnd - -Returns the exclusive upper bound of the corresponding tumbling window. - -``` sql -tumbleEnd(bounds_tuple); -tumbleEnd(time_attr, interval [, timezone]); +┌─hop(now(), toIntervalDay('1'), toIntervalDay('2'))─┐ +│ ('2024-07-03 00:00:00','2024-07-05 00:00:00') │ +└────────────────────────────────────────────────────┘ ``` ## hopStart -Returns the inclusive lower bound of the corresponding hopping window. +Returns the inclusive lower bound of the corresponding [hopping window](#hop). + +**Syntax** ``` sql -hopStart(bounds_tuple); hopStart(time_attr, hop_interval, window_interval [, timezone]); ``` +**Arguments** + +- `time_attr` — Date and time. [DateTime](../data-types/datetime.md). +- `hop_interval` — Positive Hop interval. [Interval](../data-types/special-data-types/interval.md). +- `window_interval` — Positive Window interval. [Interval](../data-types/special-data-types/interval.md). +- `timezone` — [Timezone name](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone) (optional). + +The parameters above can also be passed to the function as a [tuple](../data-types/tuple.md). + +**Returned values** + +- The inclusive lower bound of the corresponding hopping window. [DateTime](../data-types/datetime.md), [Tuple](../data-types/tuple.md) or [UInt32](../data-types/int-uint.md). + +:::note +Since one record can be assigned to multiple hop windows, the function only returns the bound of the **first** window when hop function is used **without** `WINDOW VIEW`. +::: + +**Example** + +Query: + +``` sql +SELECT hopStart(now(), INTERVAL '1' DAY, INTERVAL '2' DAY); +``` + +Result: + +``` text +┌─hopStart(now(), toIntervalDay('1'), toIntervalDay('2'))─┐ +│ 2024-07-03 00:00:00 │ +└─────────────────────────────────────────────────────────┘ +``` ## hopEnd -Returns the exclusive upper bound of the corresponding hopping window. +Returns the exclusive upper bound of the corresponding [hopping window](#hop). + +**Syntax** ``` sql -hopEnd(bounds_tuple); hopEnd(time_attr, hop_interval, window_interval [, timezone]); +``` +**Arguments** + +- `time_attr` — Date and time. [DateTime](../data-types/datetime.md). +- `hop_interval` — Positive Hop interval. [Interval](../data-types/special-data-types/interval.md). +- `window_interval` — Positive Window interval. [Interval](../data-types/special-data-types/interval.md). +- `timezone` — [Timezone name](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-timezone) (optional). + +The parameters above can also be passed to the function as a [tuple](../data-types/tuple.md). + +**Returned values** + +- The exclusive upper bound of the corresponding hopping window. [DateTime](../data-types/datetime.md), [Tuple](../data-types/tuple.md) or [UInt32](../data-types/int-uint.md). + +:::note +Since one record can be assigned to multiple hop windows, the function only returns the bound of the **first** window when hop function is used **without** `WINDOW VIEW`. +::: + +**Example** + +Query: + +``` sql +SELECT hopEnd(now(), INTERVAL '1' DAY, INTERVAL '2' DAY); +``` + +Result: + +``` text +┌─hopEnd(now(), toIntervalDay('1'), toIntervalDay('2'))─┐ +│ 2024-07-05 00:00:00 │ +└───────────────────────────────────────────────────────┘ + ``` ## Related content diff --git a/docs/en/sql-reference/functions/tuple-functions.md b/docs/en/sql-reference/functions/tuple-functions.md index 0663be08240..3b4d68e44b2 100644 --- a/docs/en/sql-reference/functions/tuple-functions.md +++ b/docs/en/sql-reference/functions/tuple-functions.md @@ -7,7 +7,7 @@ sidebar_label: Tuples ## tuple A function that allows grouping multiple columns. -For columns with the types T1, T2, ..., it returns a Tuple(T1, T2, ...) type tuple containing these columns. There is no cost to execute the function. +For columns C1, C2, ... with the types T1, T2, ..., it returns a named Tuple(C1 T1, C2 T2, ...) type tuple containing these columns if their names are unique and can be treated as unquoted identifiers, otherwise a Tuple(T1, T2, ...) is returned. There is no cost to execute the function. Tuples are normally used as intermediate values for an argument of IN operators, or for creating a list of formal parameters of lambda functions. Tuples can’t be written to a table. The function implements the operator `(x, y, ...)`. @@ -259,6 +259,60 @@ Result: └───────────────────────────────────────┘ ``` +## tupleNames + +Converts a tuple into an array of column names. For a tuple in the form `Tuple(a T, b T, ...)`, it returns an array of strings representing the named columns of the tuple. If the tuple elements do not have explicit names, their indices will be used as the column names instead. + +**Syntax** + +``` sql +tupleNames(tuple) +``` + +**Arguments** + +- `tuple` — Named tuple. [Tuple](../../sql-reference/data-types/tuple.md) with any types of values. + +**Returned value** + +- An array with strings. + +Type: [Array](../../sql-reference/data-types/array.md)([Tuple](../../sql-reference/data-types/tuple.md)([String](../../sql-reference/data-types/string.md), ...)). + +**Example** + +Query: + +``` sql +CREATE TABLE tupletest (col Tuple(user_ID UInt64, session_ID UInt64)) ENGINE = Memory; + +INSERT INTO tupletest VALUES (tuple(1, 2)); + +SELECT tupleNames(col) FROM tupletest; +``` + +Result: + +``` text +┌─tupleNames(col)──────────┐ +│ ['user_ID','session_ID'] │ +└──────────────────────────┘ +``` + +If you pass a simple tuple to the function, ClickHouse uses the indexes of the columns as their names: + +``` sql +SELECT tupleNames(tuple(3, 2, 1)); +``` + +Result: + +``` text +┌─tupleNames((3, 2, 1))─┐ +│ ['1','2','3'] │ +└───────────────────────┘ +``` + ## tuplePlus Calculates the sum of corresponding values of two tuples of the same size. diff --git a/docs/en/sql-reference/functions/tuple-map-functions.md b/docs/en/sql-reference/functions/tuple-map-functions.md index ad40725d680..24b356eca87 100644 --- a/docs/en/sql-reference/functions/tuple-map-functions.md +++ b/docs/en/sql-reference/functions/tuple-map-functions.md @@ -600,7 +600,7 @@ mapApply(func, map) **Arguments** -- `func` - [Lambda function](../../sql-reference/functions/index.md#higher-order-functions---operator-and-lambdaparams-expr-function). +- `func` — [Lambda function](../../sql-reference/functions/index.md#higher-order-functions---operator-and-lambdaparams-expr-function). - `map` — [Map](../data-types/map.md). **Returned value** @@ -831,7 +831,39 @@ SELECT mapSort((k, v) -> v, map('key2', 2, 'key3', 1, 'key1', 3)) AS map; └──────────────────────────────┘ ``` -For more details see the [reference](../../sql-reference/functions/array-functions.md#array_functions-sort) for `arraySort` function. +For more details see the [reference](../../sql-reference/functions/array-functions.md#array_functions-sort) for `arraySort` function. + +## mapPartialSort + +Sorts the elements of a map in ascending order with additional `limit` argument allowing partial sorting. +If the `func` function is specified, the sorting order is determined by the result of the `func` function applied to the keys and values of the map. + +**Syntax** + +```sql +mapPartialSort([func,] limit, map) +``` +**Arguments** + +- `func` – Optional function to apply to the keys and values of the map. [Lambda function](../../sql-reference/functions/index.md#higher-order-functions---operator-and-lambdaparams-expr-function). +- `limit` – Elements in range [1..limit] are sorted. [(U)Int](../data-types/int-uint.md). +- `map` – Map to sort. [Map](../data-types/map.md). + +**Returned value** + +- Partially sorted map. [Map](../data-types/map.md). + +**Example** + +``` sql +SELECT mapPartialSort((k, v) -> v, 2, map('k1', 3, 'k2', 1, 'k3', 2)); +``` + +``` text +┌─mapPartialSort(lambda(tuple(k, v), v), 2, map('k1', 3, 'k2', 1, 'k3', 2))─┐ +│ {'k2':1,'k3':2,'k1':3} │ +└───────────────────────────────────────────────────────────────────────────┘ +``` ## mapReverseSort(\[func,\], map) @@ -861,3 +893,35 @@ SELECT mapReverseSort((k, v) -> v, map('key2', 2, 'key3', 1, 'key1', 3)) AS map; ``` For more details see function [arrayReverseSort](../../sql-reference/functions/array-functions.md#array_functions-reverse-sort). + +## mapPartialReverseSort + +Sorts the elements of a map in descending order with additional `limit` argument allowing partial sorting. +If the `func` function is specified, the sorting order is determined by the result of the `func` function applied to the keys and values of the map. + +**Syntax** + +```sql +mapPartialReverseSort([func,] limit, map) +``` +**Arguments** + +- `func` – Optional function to apply to the keys and values of the map. [Lambda function](../../sql-reference/functions/index.md#higher-order-functions---operator-and-lambdaparams-expr-function). +- `limit` – Elements in range [1..limit] are sorted. [(U)Int](../data-types/int-uint.md). +- `map` – Map to sort. [Map](../data-types/map.md). + +**Returned value** + +- Partially sorted map. [Map](../data-types/map.md). + +**Example** + +``` sql +SELECT mapPartialReverseSort((k, v) -> v, 2, map('k1', 3, 'k2', 1, 'k3', 2)); +``` + +``` text +┌─mapPartialReverseSort(lambda(tuple(k, v), v), 2, map('k1', 3, 'k2', 1, 'k3', 2))─┐ +│ {'k1':3,'k3':2,'k2':1} │ +└──────────────────────────────────────────────────────────────────────────────────┘ +``` \ No newline at end of file diff --git a/docs/en/sql-reference/functions/uuid-functions.md b/docs/en/sql-reference/functions/uuid-functions.md index e990023efbc..0bde207dcc9 100644 --- a/docs/en/sql-reference/functions/uuid-functions.md +++ b/docs/en/sql-reference/functions/uuid-functions.md @@ -567,12 +567,13 @@ While no standard or recommendation exists for the epoch of Snowflake IDs, imple **Syntax** ``` sql -generateSnowflakeID([expr]) +generateSnowflakeID([expr, [machine_id]]) ``` **Arguments** - `expr` — An arbitrary [expression](../../sql-reference/syntax.md#syntax-expressions) used to bypass [common subexpression elimination](../../sql-reference/functions/index.md#common-subexpression-elimination) if the function is called multiple times in a query. The value of the expression has no effect on the returned Snowflake ID. Optional. +- `machine_id` — A machine ID, the lowest 10 bits are used. [Int64](../data-types/int-uint.md). Optional. **Returned value** @@ -608,6 +609,16 @@ SELECT generateSnowflakeID(1), generateSnowflakeID(2); └────────────────────────┴────────────────────────┘ ``` +**Example with expression and a machine ID** + +``` +SELECT generateSnowflakeID('expr', 1); + +┌─generateSnowflakeID('expr', 1)─┐ +│ 7201148511606784002 │ +└────────────────────────────────┘ +``` + ## snowflakeToDateTime :::warning diff --git a/docs/en/sql-reference/statements/alter/named-collection.md b/docs/en/sql-reference/statements/alter/named-collection.md index 71d4bfadd9c..ab772fe4dcf 100644 --- a/docs/en/sql-reference/statements/alter/named-collection.md +++ b/docs/en/sql-reference/statements/alter/named-collection.md @@ -3,6 +3,10 @@ slug: /en/sql-reference/statements/alter/named-collection sidebar_label: NAMED COLLECTION --- +import CloudNotSupportedBadge from '@theme/badges/CloudNotSupportedBadge'; + + + # ALTER NAMED COLLECTION This query intends to modify already existing named collections. diff --git a/docs/en/sql-reference/statements/alter/statistics.md b/docs/en/sql-reference/statements/alter/statistics.md index 80024781f88..6880cef0e5c 100644 --- a/docs/en/sql-reference/statements/alter/statistics.md +++ b/docs/en/sql-reference/statements/alter/statistics.md @@ -28,6 +28,6 @@ There is an example adding two statistics types to two columns: ALTER TABLE t1 MODIFY STATISTICS c, d TYPE TDigest, Uniq; ``` -:::note +:::note Statistic manipulation is supported only for tables with [`*MergeTree`](../../../engines/table-engines/mergetree-family/mergetree.md) engine (including [replicated](../../../engines/table-engines/mergetree-family/replication.md) variants). ::: diff --git a/docs/en/sql-reference/statements/alter/user.md b/docs/en/sql-reference/statements/alter/user.md index b5c156f56a9..6216b83c2ef 100644 --- a/docs/en/sql-reference/statements/alter/user.md +++ b/docs/en/sql-reference/statements/alter/user.md @@ -12,7 +12,7 @@ Syntax: ``` sql ALTER USER [IF EXISTS] name1 [ON CLUSTER cluster_name1] [RENAME TO new_name1] [, name2 [ON CLUSTER cluster_name2] [RENAME TO new_name2] ...] - [NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name'}] + [NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name' | SAN 'TYPE:subject_alt_name'}] [[ADD | DROP] HOST {LOCAL | NAME 'name' | REGEXP 'name_regexp' | IP 'address' | LIKE 'pattern'} [,...] | ANY | NONE] [VALID UNTIL datetime] [DEFAULT ROLE role [,...] | ALL | ALL EXCEPT role [,...] ] diff --git a/docs/en/sql-reference/statements/alter/view.md b/docs/en/sql-reference/statements/alter/view.md index fb7a5bd7c03..5f3dae0a9c0 100644 --- a/docs/en/sql-reference/statements/alter/view.md +++ b/docs/en/sql-reference/statements/alter/view.md @@ -134,8 +134,8 @@ PRIMARY KEY (event_type, ts) ORDER BY (event_type, ts, browser) SETTINGS index_granularity = 8192 --- !!! The columns' definition is unchanged but it does not matter, we are not quering --- MATERIALIZED VIEW, we are quering TO (storage) table. +-- !!! The columns' definition is unchanged but it does not matter, we are not querying +-- MATERIALIZED VIEW, we are querying TO (storage) table. -- SELECT section is updated. SHOW CREATE TABLE mv FORMAT TSVRaw; diff --git a/docs/en/sql-reference/statements/create/named-collection.md b/docs/en/sql-reference/statements/create/named-collection.md index f69fa2e3678..a4e146c814c 100644 --- a/docs/en/sql-reference/statements/create/named-collection.md +++ b/docs/en/sql-reference/statements/create/named-collection.md @@ -3,6 +3,10 @@ slug: /en/sql-reference/statements/create/named-collection sidebar_label: NAMED COLLECTION --- +import CloudNotSupportedBadge from '@theme/badges/CloudNotSupportedBadge'; + + + # CREATE NAMED COLLECTION Creates a new named collection. diff --git a/docs/en/sql-reference/statements/create/table.md b/docs/en/sql-reference/statements/create/table.md index 0253bc647e6..9c8984d698f 100644 --- a/docs/en/sql-reference/statements/create/table.md +++ b/docs/en/sql-reference/statements/create/table.md @@ -21,7 +21,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] name2 [type2] [NULL|NOT NULL] [DEFAULT|MATERIALIZED|EPHEMERAL|ALIAS expr2] [COMMENT 'comment for column'] [compression_codec] [TTL expr2], ... ) ENGINE = engine - COMMENT 'comment for table' + [COMMENT 'comment for table'] ``` Creates a table named `table_name` in the `db` database or the current database if `db` is not set, with the structure specified in brackets and the `engine` engine. @@ -152,7 +152,7 @@ SELECT * FROM test; `MATERIALIZED expr` -Materialized expression. Values of such columns are always calculated, they cannot be specified in INSERT queries. +Materialized expression. Values of such columns are automatically calculated according to the specified materialized expression when rows are inserted. Values cannot be explicitly specified during `INSERT`s. Also, default value columns of this type are not included in the result of `SELECT *`. This is to preserve the invariant that the result of a `SELECT *` can always be inserted back into the table using `INSERT`. This behavior can be disabled with setting `asterisk_include_materialized_columns`. @@ -626,11 +626,6 @@ SELECT * FROM base.t1; You can add a comment to the table when you creating it. -:::note -The comment clause is supported by all table engines except [Kafka](../../../engines/table-engines/integrations/kafka.md), [RabbitMQ](../../../engines/table-engines/integrations/rabbitmq.md) and [EmbeddedRocksDB](../../../engines/table-engines/integrations/embedded-rocksdb.md). -::: - - **Syntax** ``` sql diff --git a/docs/en/sql-reference/statements/create/user.md b/docs/en/sql-reference/statements/create/user.md index aee98cfcd10..8c9143ee086 100644 --- a/docs/en/sql-reference/statements/create/user.md +++ b/docs/en/sql-reference/statements/create/user.md @@ -12,7 +12,7 @@ Syntax: ``` sql CREATE USER [IF NOT EXISTS | OR REPLACE] name1 [ON CLUSTER cluster_name1] [, name2 [ON CLUSTER cluster_name2] ...] - [NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name'} | {WITH ssh_key BY KEY 'public_key' TYPE 'ssh-rsa|...'} | {WITH http SERVER 'server_name' [SCHEME 'Basic']}] + [NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name' | SAN 'TYPE:subject_alt_name'} | {WITH ssh_key BY KEY 'public_key' TYPE 'ssh-rsa|...'} | {WITH http SERVER 'server_name' [SCHEME 'Basic']}] [HOST {LOCAL | NAME 'name' | REGEXP 'name_regexp' | IP 'address' | LIKE 'pattern'} [,...] | ANY | NONE] [VALID UNTIL datetime] [IN access_storage_type] diff --git a/docs/en/sql-reference/statements/create/view.md b/docs/en/sql-reference/statements/create/view.md index 1fabb6d8cc7..2931f7020fb 100644 --- a/docs/en/sql-reference/statements/create/view.md +++ b/docs/en/sql-reference/statements/create/view.md @@ -16,6 +16,7 @@ Syntax: CREATE [OR REPLACE] VIEW [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster_name] [DEFINER = { user | CURRENT_USER }] [SQL SECURITY { DEFINER | INVOKER | NONE }] AS SELECT ... +[COMMENT 'comment'] ``` Normal views do not store any data. They just perform a read from another table on each access. In other words, a normal view is nothing more than a saved query. When reading from a view, this saved query is used as a subquery in the [FROM](../../../sql-reference/statements/select/from.md) clause. @@ -57,6 +58,7 @@ SELECT * FROM view(column1=value1, column2=value2 ...) CREATE MATERIALIZED VIEW [IF NOT EXISTS] [db.]table_name [ON CLUSTER] [TO[db.]name] [ENGINE = engine] [POPULATE] [DEFINER = { user | CURRENT_USER }] [SQL SECURITY { DEFINER | INVOKER | NONE }] AS SELECT ... +[COMMENT 'comment'] ``` :::tip @@ -161,6 +163,7 @@ RANDOMIZE FOR interval DEPENDS ON [db.]name [, [db.]name [, ...]] [TO[db.]name] [(columns)] [ENGINE = engine] [EMPTY] AS SELECT ... +[COMMENT 'comment'] ``` where `interval` is a sequence of simple intervals: ```sql @@ -267,7 +270,10 @@ This is an experimental feature that may change in backwards-incompatible ways i ::: ``` sql -CREATE WINDOW VIEW [IF NOT EXISTS] [db.]table_name [TO [db.]table_name] [INNER ENGINE engine] [ENGINE engine] [WATERMARK strategy] [ALLOWED_LATENESS interval_function] [POPULATE] AS SELECT ... GROUP BY time_window_function +CREATE WINDOW VIEW [IF NOT EXISTS] [db.]table_name [TO [db.]table_name] [INNER ENGINE engine] [ENGINE engine] [WATERMARK strategy] [ALLOWED_LATENESS interval_function] [POPULATE] +AS SELECT ... +GROUP BY time_window_function +[COMMENT 'comment'] ``` Window view can aggregate data by time window and output the results when the window is ready to fire. It stores the partial aggregation results in an inner(or specified) table to reduce latency and can push the processing result to a specified table or push notifications using the WATCH query. diff --git a/docs/en/sql-reference/statements/kill.md b/docs/en/sql-reference/statements/kill.md index b665ad85a09..6e18ace10c7 100644 --- a/docs/en/sql-reference/statements/kill.md +++ b/docs/en/sql-reference/statements/kill.md @@ -58,6 +58,8 @@ KILL QUERY WHERE query_id='2-857d-4a57-9ee0-327da5d60a90' KILL QUERY WHERE user='username' SYNC ``` +:::tip If you are killing a query in ClickHouse Cloud or in a self-managed cluster, then be sure to use the ```ON CLUSTER [cluster-name]``` option, in order to ensure the query is killed on all replicas::: + Read-only users can only stop their own queries. By default, the asynchronous version of queries is used (`ASYNC`), which does not wait for confirmation that queries have stopped. @@ -131,6 +133,7 @@ KILL MUTATION WHERE database = 'default' AND table = 'table' -- Cancel the specific mutation: KILL MUTATION WHERE database = 'default' AND table = 'table' AND mutation_id = 'mutation_3.txt' ``` +:::tip If you are killing a mutation in ClickHouse Cloud or in a self-managed cluster, then be sure to use the ```ON CLUSTER [cluster-name]``` option, in order to ensure the mutation is killed on all replicas::: The query is useful when a mutation is stuck and cannot finish (e.g. if some function in the mutation query throws an exception when applied to the data contained in the table). diff --git a/docs/en/sql-reference/statements/select/join.md b/docs/en/sql-reference/statements/select/join.md index 34c6016235a..96d9d26977d 100644 --- a/docs/en/sql-reference/statements/select/join.md +++ b/docs/en/sql-reference/statements/select/join.md @@ -297,7 +297,7 @@ Algorithm requires the special column in tables. This column: - Must contain an ordered sequence. - Can be one of the following types: [Int, UInt](../../../sql-reference/data-types/int-uint.md), [Float](../../../sql-reference/data-types/float.md), [Date](../../../sql-reference/data-types/date.md), [DateTime](../../../sql-reference/data-types/datetime.md), [Decimal](../../../sql-reference/data-types/decimal.md). -- Can’t be the only column in the `JOIN` clause. +- For `hash` join algorithm it can’t be the only column in the `JOIN` clause. Syntax `ASOF JOIN ... ON`: @@ -337,7 +337,8 @@ For example, consider the following tables: `ASOF JOIN` can take the timestamp of a user event from `table_1` and find an event in `table_2` where the timestamp is closest to the timestamp of the event from `table_1` corresponding to the closest match condition. Equal timestamp values are the closest if available. Here, the `user_id` column can be used for joining on equality and the `ev_time` column can be used for joining on the closest match. In our example, `event_1_1` can be joined with `event_2_1` and `event_1_2` can be joined with `event_2_3`, but `event_2_2` can’t be joined. :::note -`ASOF` join is **not** supported in the [Join](../../../engines/table-engines/special/join.md) table engine. +`ASOF JOIN` is supported only by `hash` and `full_sorting_merge` join algorithms. +It's **not** supported in the [Join](../../../engines/table-engines/special/join.md) table engine. ::: ## PASTE JOIN Usage diff --git a/docs/en/sql-reference/statements/system.md b/docs/en/sql-reference/statements/system.md index e6d3439d2b9..35f2f15dd80 100644 --- a/docs/en/sql-reference/statements/system.md +++ b/docs/en/sql-reference/statements/system.md @@ -18,10 +18,21 @@ Reloads all dictionaries that have been successfully loaded before. By default, dictionaries are loaded lazily (see [dictionaries_lazy_load](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-dictionaries_lazy_load)), so instead of being loaded automatically at startup, they are initialized on first access through dictGet function or SELECT from tables with ENGINE = Dictionary. The `SYSTEM RELOAD DICTIONARIES` query reloads such dictionaries (LOADED). Always returns `Ok.` regardless of the result of the dictionary update. +**Syntax** + +```sql +SYSTEM RELOAD DICTIONARIES [ON CLUSTER cluster_name] +``` + ## RELOAD DICTIONARY Completely reloads a dictionary `dictionary_name`, regardless of the state of the dictionary (LOADED / NOT_LOADED / FAILED). Always returns `Ok.` regardless of the result of updating the dictionary. + +``` sql +SYSTEM RELOAD DICTIONARY [ON CLUSTER cluster_name] dictionary_name +``` + The status of the dictionary can be checked by querying the `system.dictionaries` table. ``` sql diff --git a/docs/en/sql-reference/table-functions/azureBlobStorageCluster.md b/docs/en/sql-reference/table-functions/azureBlobStorageCluster.md index 20dfd35d5db..ca3959d6ed1 100644 --- a/docs/en/sql-reference/table-functions/azureBlobStorageCluster.md +++ b/docs/en/sql-reference/table-functions/azureBlobStorageCluster.md @@ -1,6 +1,6 @@ --- slug: /en/sql-reference/table-functions/azureBlobStorageCluster -sidebar_position: 55 +sidebar_position: 15 sidebar_label: azureBlobStorageCluster title: "azureBlobStorageCluster Table Function" --- diff --git a/docs/en/sql-reference/table-functions/file.md b/docs/en/sql-reference/table-functions/file.md index 3a3162dad9a..44b1b50620a 100644 --- a/docs/en/sql-reference/table-functions/file.md +++ b/docs/en/sql-reference/table-functions/file.md @@ -130,7 +130,9 @@ SELECT * FROM file('user_files/archives/archive{1..2}.zip :: table.csv'); ## Globs in path -Paths may use globbing. Files must match the whole path pattern, not only the suffix or prefix. +Paths may use globbing. Files must match the whole path pattern, not only the suffix or prefix. There is one exception that if the path refers to an existing +directory and does not use globs, a `*` will be implicitly added to the path so +all the files in the directory are selected. - `*` — Represents arbitrarily many characters except `/` but including the empty string. - `?` — Represents an arbitrary single character. @@ -163,6 +165,12 @@ An alternative path expression which achieves the same: SELECT count(*) FROM file('{some,another}_dir/*', 'TSV', 'name String, value UInt32'); ``` +Query the total number of rows in `some_dir` using the implicit `*`: + +```sql +SELECT count(*) FROM file('some_dir', 'TSV', 'name String, value UInt32'); +``` + :::note If your listing of files contains number ranges with leading zeros, use the construction with braces for each digit separately or use `?`. ::: diff --git a/docs/en/sql-reference/table-functions/s3.md b/docs/en/sql-reference/table-functions/s3.md index 1a7e2b8d66a..35e5d86034c 100644 --- a/docs/en/sql-reference/table-functions/s3.md +++ b/docs/en/sql-reference/table-functions/s3.md @@ -269,9 +269,9 @@ FROM s3( ## Virtual Columns {#virtual-columns} -- `_path` — Path to the file. Type: `LowCardinalty(String)`. -- `_file` — Name of the file. Type: `LowCardinalty(String)`. -- `_size` — Size of the file in bytes. Type: `Nullable(UInt64)`. If the file size is unknown, the value is `NULL`. +- `_path` — Path to the file. Type: `LowCardinalty(String)`. In case of archive, shows path in a format: "{path_to_archive}::{path_to_file_inside_archive}" +- `_file` — Name of the file. Type: `LowCardinalty(String)`. In case of archive shows name of the file inside the archive. +- `_size` — Size of the file in bytes. Type: `Nullable(UInt64)`. If the file size is unknown, the value is `NULL`. In case of archive shows uncompressed file size of the file inside the archive. - `_time` — Last modified time of the file. Type: `Nullable(DateTime)`. If the time is unknown, the value is `NULL`. ## Storage Settings {#storage-settings} diff --git a/docs/en/sql-reference/window-functions/dense_rank.md b/docs/en/sql-reference/window-functions/dense_rank.md new file mode 100644 index 00000000000..d6445b68c55 --- /dev/null +++ b/docs/en/sql-reference/window-functions/dense_rank.md @@ -0,0 +1,73 @@ +--- +slug: /en/sql-reference/window-functions/dense_rank +sidebar_label: dense_rank +sidebar_position: 7 +--- + +# dense_rank + +Ranks the current row within its partition without gaps. In other words, if the value of any new row encountered is equal to the value of one of the previous rows then it will receive the next successive rank without any gaps in ranking. + +The [rank](./rank.md) function provides the same behaviour, but with gaps in ranking. + +**Syntax** + +```sql +dense_rank (column_name) + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [ROWS or RANGE expression_to_bound_rows_withing_the_group]] | [window_name]) +FROM table_name +WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column]) +``` + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Returned value** + +- A number for the current row within its partition, without gaps in ranking. [UInt64](../data-types/int-uint.md). + +**Example** + +The following example is based on the example provided in the video instructional [Ranking window functions in ClickHouse](https://youtu.be/Yku9mmBYm_4?si=XIMu1jpYucCQEoXA). + +Query: + +```sql +CREATE TABLE salaries +( + `team` String, + `player` String, + `salary` UInt32, + `position` String +) +Engine = Memory; + +INSERT INTO salaries FORMAT Values + ('Port Elizabeth Barbarians', 'Gary Chen', 195000, 'F'), + ('New Coreystad Archdukes', 'Charles Juarez', 190000, 'F'), + ('Port Elizabeth Barbarians', 'Michael Stanley', 150000, 'D'), + ('New Coreystad Archdukes', 'Scott Harrison', 150000, 'D'), + ('Port Elizabeth Barbarians', 'Robert George', 195000, 'M'), + ('South Hampton Seagulls', 'Douglas Benson', 150000, 'M'), + ('South Hampton Seagulls', 'James Henderson', 140000, 'M'); +``` + +```sql +SELECT player, salary, + dense_rank() OVER (ORDER BY salary DESC) AS dense_rank +FROM salaries; +``` + +Result: + +```response + ┌─player──────────┬─salary─┬─dense_rank─┐ +1. │ Gary Chen │ 195000 │ 1 │ +2. │ Robert George │ 195000 │ 1 │ +3. │ Charles Juarez │ 190000 │ 2 │ +4. │ Michael Stanley │ 150000 │ 3 │ +5. │ Douglas Benson │ 150000 │ 3 │ +6. │ Scott Harrison │ 150000 │ 3 │ +7. │ James Henderson │ 140000 │ 4 │ + └─────────────────┴────────┴────────────┘ +``` \ No newline at end of file diff --git a/docs/en/sql-reference/window-functions/first_value.md b/docs/en/sql-reference/window-functions/first_value.md new file mode 100644 index 00000000000..30c3b1f99dc --- /dev/null +++ b/docs/en/sql-reference/window-functions/first_value.md @@ -0,0 +1,79 @@ +--- +slug: /en/sql-reference/window-functions/first_value +sidebar_label: first_value +sidebar_position: 3 +--- + +# first_value + +Returns the first value evaluated within its ordered frame. By default, NULL arguments are skipped, however the `RESPECT NULLS` modifier can be used to override this behaviour. + +**Syntax** + +```sql +first_value (column_name) [[RESPECT NULLS] | [IGNORE NULLS]] + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [ROWS or RANGE expression_to_bound_rows_withing_the_group]] | [window_name]) +FROM table_name +WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column]) +``` + +Alias: `any`. + +:::note +Using the optional modifier `RESPECT NULLS` after `first_value(column_name)` will ensure that `NULL` arguments are not skipped. +See [NULL processing](../aggregate-functions/index.md/#null-processing) for more information. +::: + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Returned value** + +- The first value evaluated within its ordered frame. + +**Example** + +In this example the `first_value` function is used to find the highest paid footballer from a fictional dataset of salaries of Premier League football players. + +Query: + +```sql +DROP TABLE IF EXISTS salaries; +CREATE TABLE salaries +( + `team` String, + `player` String, + `salary` UInt32, + `position` String +) +Engine = Memory; + +INSERT INTO salaries FORMAT Values + ('Port Elizabeth Barbarians', 'Gary Chen', 196000, 'F'), + ('New Coreystad Archdukes', 'Charles Juarez', 190000, 'F'), + ('Port Elizabeth Barbarians', 'Michael Stanley', 100000, 'D'), + ('New Coreystad Archdukes', 'Scott Harrison', 180000, 'D'), + ('Port Elizabeth Barbarians', 'Robert George', 195000, 'M'), + ('South Hampton Seagulls', 'Douglas Benson', 150000, 'M'), + ('South Hampton Seagulls', 'James Henderson', 140000, 'M'); +``` + +```sql +SELECT player, salary, + first_value(player) OVER (ORDER BY salary DESC) AS highest_paid_player +FROM salaries; +``` + +Result: + +```response + ┌─player──────────┬─salary─┬─highest_paid_player─┐ +1. │ Gary Chen │ 196000 │ Gary Chen │ +2. │ Robert George │ 195000 │ Gary Chen │ +3. │ Charles Juarez │ 190000 │ Gary Chen │ +4. │ Scott Harrison │ 180000 │ Gary Chen │ +5. │ Douglas Benson │ 150000 │ Gary Chen │ +6. │ James Henderson │ 140000 │ Gary Chen │ +7. │ Michael Stanley │ 100000 │ Gary Chen │ + └─────────────────┴────────┴─────────────────────┘ +``` \ No newline at end of file diff --git a/docs/en/sql-reference/window-functions/index.md b/docs/en/sql-reference/window-functions/index.md index 3a8afd10359..0c3e2ea1cb6 100644 --- a/docs/en/sql-reference/window-functions/index.md +++ b/docs/en/sql-reference/window-functions/index.md @@ -1,10 +1,11 @@ --- slug: /en/sql-reference/window-functions/ -sidebar_position: 62 sidebar_label: Window Functions -title: Window Functions +sidebar_position: 1 --- +# Window Functions + Windows functions let you perform calculations across a set of rows that are related to the current row. Some of the calculations that you can do are similar to those that can be done with an aggregate function, but a window function doesn't cause rows to be grouped into a single output - the individual rows are still returned. @@ -12,8 +13,8 @@ Some of the calculations that you can do are similar to those that can be done w ClickHouse supports the standard grammar for defining windows and window functions. The table below indicates whether a feature is currently supported. -| Feature | Supported? | -|------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Feature | Supported? | +|--------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | ad hoc window specification (`count(*) over (partition by id order by time desc)`) | ✅ | | expressions involving window functions, e.g. `(count(*) over ()) / 2)` | ✅ | | `WINDOW` clause (`select ... from table window w as (partition by id)`) | ✅ | @@ -23,6 +24,7 @@ ClickHouse supports the standard grammar for defining windows and window functio | `GROUPS` frame | ❌ | | Calculating aggregate functions over a frame (`sum(value) over (order by time)`) | ✅ (All aggregate functions are supported) | | `rank()`, `dense_rank()`, `row_number()` | ✅ | +| `percent_rank()` | ✅ Efficiently computes the relative standing of a value within a partition in a dataset. This function effectively replaces the more verbose and computationally intensive manual SQL calculation expressed as `ifNull((rank() OVER(PARTITION BY x ORDER BY y) - 1) / nullif(count(1) OVER(PARTITION BY x) - 1, 0), 0)`| | `lag/lead(value, offset)` | ❌
You can use one of the following workarounds:
1) `any(value) over (.... rows between preceding and preceding)`, or `following` for `lead`
2) `lagInFrame/leadInFrame`, which are analogous, but respect the window frame. To get behavior identical to `lag/lead`, use `rows between unbounded preceding and unbounded following` | | ntile(buckets) | ✅
Specify window like, (partition by x order by y rows between unbounded preceding and unrounded following). | @@ -74,14 +76,14 @@ WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column] These functions can be used only as a window function. -- `row_number()` - Number the current row within its partition starting from 1. -- `first_value(x)` - Return the first non-NULL value evaluated within its ordered frame. -- `last_value(x)` - Return the last non-NULL value evaluated within its ordered frame. -- `nth_value(x, offset)` - Return the first non-NULL value evaluated against the nth row (offset) in its ordered frame. -- `rank()` - Rank the current row within its partition with gaps. -- `dense_rank()` - Rank the current row within its partition without gaps. -- `lagInFrame(x)` - Return a value evaluated at the row that is at a specified physical offset row before the current row within the ordered frame. -- `leadInFrame(x)` - Return a value evaluated at the row that is offset rows after the current row within the ordered frame. +- [`row_number()`](./row_number.md) - Number the current row within its partition starting from 1. +- [`first_value(x)`](./first_value.md) - Return the first value evaluated within its ordered frame. +- [`last_value(x)`](./last_value.md) - Return the last value evaluated within its ordered frame. +- [`nth_value(x, offset)`](./nth_value.md) - Return the first non-NULL value evaluated against the nth row (offset) in its ordered frame. +- [`rank()`](./rank.md) - Rank the current row within its partition with gaps. +- [`dense_rank()`](./dense_rank.md) - Rank the current row within its partition without gaps. +- [`lagInFrame(x)`](./lagInFrame.md) - Return a value evaluated at the row that is at a specified physical offset row before the current row within the ordered frame. +- [`leadInFrame(x)`](./leadInFrame.md) - Return a value evaluated at the row that is offset rows after the current row within the ordered frame. ## Examples diff --git a/docs/en/sql-reference/window-functions/lagInFrame.md b/docs/en/sql-reference/window-functions/lagInFrame.md new file mode 100644 index 00000000000..049e095c10f --- /dev/null +++ b/docs/en/sql-reference/window-functions/lagInFrame.md @@ -0,0 +1,79 @@ +--- +slug: /en/sql-reference/window-functions/lagInFrame +sidebar_label: lagInFrame +sidebar_position: 8 +--- + +# lagInFrame + +Returns a value evaluated at the row that is at a specified physical offset row before the current row within the ordered frame. + +**Syntax** + +```sql +lagInFrame(x[, offset[, default]]) + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [ROWS or RANGE expression_to_bound_rows_withing_the_group]] | [window_name]) +FROM table_name +WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column]) +``` + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Parameters** +- `x` — Column name. +- `offset` — Offset to apply. [(U)Int*](../data-types/int-uint.md). (Optional - `1` by default). +- `default` — Value to return if calculated row exceeds the boundaries of the window frame. (Optional - `null` by default). + +**Returned value** + +- Value evaluated at the row that is at a specified physical offset before the current row within the ordered frame. + +**Example** + +This example looks at historical data for a specific stock and uses the `lagInFrame` function to calculate a day-to-day delta and percentage change in the closing price of the stock. + +Query: + +```sql +CREATE TABLE stock_prices +( + `date` Date, + `open` Float32, -- opening price + `high` Float32, -- daily high + `low` Float32, -- daily low + `close` Float32, -- closing price + `volume` UInt32 -- trade volume +) +Engine = Memory; + +INSERT INTO stock_prices FORMAT Values + ('2024-06-03', 113.62, 115.00, 112.00, 115.00, 438392000), + ('2024-06-04', 115.72, 116.60, 114.04, 116.44, 403324000), + ('2024-06-05', 118.37, 122.45, 117.47, 122.44, 528402000), + ('2024-06-06', 124.05, 125.59, 118.32, 121.00, 664696000), + ('2024-06-07', 119.77, 121.69, 118.02, 120.89, 412386000); +``` + +```sql +SELECT + date, + close, + lagInFrame(close, 1, close) OVER (ORDER BY date ASC) AS previous_day_close, + COALESCE(ROUND(close - previous_day_close, 2)) AS delta, + COALESCE(ROUND((delta / previous_day_close) * 100, 2)) AS percent_change +FROM stock_prices +ORDER BY date DESC; +``` + +Result: + +```response + ┌───────date─┬──close─┬─previous_day_close─┬─delta─┬─percent_change─┐ +1. │ 2024-06-07 │ 120.89 │ 121 │ -0.11 │ -0.09 │ +2. │ 2024-06-06 │ 121 │ 122.44 │ -1.44 │ -1.18 │ +3. │ 2024-06-05 │ 122.44 │ 116.44 │ 6 │ 5.15 │ +4. │ 2024-06-04 │ 116.44 │ 115 │ 1.44 │ 1.25 │ +5. │ 2024-06-03 │ 115 │ 115 │ 0 │ 0 │ + └────────────┴────────┴────────────────────┴───────┴────────────────┘ +``` \ No newline at end of file diff --git a/docs/en/sql-reference/window-functions/last_value.md b/docs/en/sql-reference/window-functions/last_value.md new file mode 100644 index 00000000000..dd7f5fa078a --- /dev/null +++ b/docs/en/sql-reference/window-functions/last_value.md @@ -0,0 +1,79 @@ +--- +slug: /en/sql-reference/window-functions/last_value +sidebar_label: last_value +sidebar_position: 4 +--- + +# last_value + +Returns the last value evaluated within its ordered frame. By default, NULL arguments are skipped, however the `RESPECT NULLS` modifier can be used to override this behaviour. + +**Syntax** + +```sql +last_value (column_name) [[RESPECT NULLS] | [IGNORE NULLS]] + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [ROWS or RANGE expression_to_bound_rows_withing_the_group]] | [window_name]) +FROM table_name +WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column]) +``` + +Alias: `anyLast`. + +:::note +Using the optional modifier `RESPECT NULLS` after `first_value(column_name)` will ensure that `NULL` arguments are not skipped. +See [NULL processing](../aggregate-functions/index.md/#null-processing) for more information. +::: + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Returned value** + +- The last value evaluated within its ordered frame. + +**Example** + +In this example the `last_value` function is used to find the highest paid footballer from a fictional dataset of salaries of Premier League football players. + +Query: + +```sql +DROP TABLE IF EXISTS salaries; +CREATE TABLE salaries +( + `team` String, + `player` String, + `salary` UInt32, + `position` String +) +Engine = Memory; + +INSERT INTO salaries FORMAT Values + ('Port Elizabeth Barbarians', 'Gary Chen', 196000, 'F'), + ('New Coreystad Archdukes', 'Charles Juarez', 190000, 'F'), + ('Port Elizabeth Barbarians', 'Michael Stanley', 100000, 'D'), + ('New Coreystad Archdukes', 'Scott Harrison', 180000, 'D'), + ('Port Elizabeth Barbarians', 'Robert George', 195000, 'M'), + ('South Hampton Seagulls', 'Douglas Benson', 150000, 'M'), + ('South Hampton Seagulls', 'James Henderson', 140000, 'M'); +``` + +```sql +SELECT player, salary, + last_value(player) OVER (ORDER BY salary DESC RANGE BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS lowest_paid_player +FROM salaries; +``` + +Result: + +```response + ┌─player──────────┬─salary─┬─lowest_paid_player─┐ +1. │ Gary Chen │ 196000 │ Michael Stanley │ +2. │ Robert George │ 195000 │ Michael Stanley │ +3. │ Charles Juarez │ 190000 │ Michael Stanley │ +4. │ Scott Harrison │ 180000 │ Michael Stanley │ +5. │ Douglas Benson │ 150000 │ Michael Stanley │ +6. │ James Henderson │ 140000 │ Michael Stanley │ +7. │ Michael Stanley │ 100000 │ Michael Stanley │ + └─────────────────┴────────┴────────────────────┘ +``` \ No newline at end of file diff --git a/docs/en/sql-reference/window-functions/leadInFrame.md b/docs/en/sql-reference/window-functions/leadInFrame.md new file mode 100644 index 00000000000..fc1b92cc266 --- /dev/null +++ b/docs/en/sql-reference/window-functions/leadInFrame.md @@ -0,0 +1,60 @@ +--- +slug: /en/sql-reference/window-functions/leadInFrame +sidebar_label: leadInFrame +sidebar_position: 9 +--- + +# leadInFrame + +Returns a value evaluated at the row that is offset rows after the current row within the ordered frame. + +**Syntax** + +```sql +leadInFrame(x[, offset[, default]]) + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [ROWS or RANGE expression_to_bound_rows_withing_the_group]] | [window_name]) +FROM table_name +WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column]) +``` + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Parameters** +- `x` — Column name. +- `offset` — Offset to apply. [(U)Int*](../data-types/int-uint.md). (Optional - `1` by default). +- `default` — Value to return if calculated row exceeds the boundaries of the window frame. (Optional - `null` by default). + +**Returned value** + +- value evaluated at the row that is offset rows after the current row within the ordered frame. + +**Example** + +This example looks at [historical data](https://www.kaggle.com/datasets/sazidthe1/nobel-prize-data) for Nobel Prize winners and uses the `leadInFrame` function to return a list of successive winners in the physics category. + +Query: + +```sql +CREATE OR REPLACE VIEW nobel_prize_laureates AS FROM file('nobel_laureates_data.csv') SELECT *; +``` + +```sql +FROM nobel_prize_laureates SELECT fullName, leadInFrame(year, 1, year) OVER (PARTITION BY category ORDER BY year) AS year, category, motivation WHERE category == 'physics' ORDER BY year DESC LIMIT 9; +``` + +Result: + +```response + ┌─fullName─────────┬─year─┬─category─┬─motivation─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┐ +1. │ Pierre Agostini │ 2023 │ physics │ for experimental methods that generate attosecond pulses of light for the study of electron dynamics in matter │ +2. │ Ferenc Krausz │ 2023 │ physics │ for experimental methods that generate attosecond pulses of light for the study of electron dynamics in matter │ +3. │ Anne L Huillier │ 2023 │ physics │ for experimental methods that generate attosecond pulses of light for the study of electron dynamics in matter │ +4. │ Alain Aspect │ 2022 │ physics │ for experiments with entangled photons establishing the violation of Bell inequalities and pioneering quantum information science │ +5. │ Anton Zeilinger │ 2022 │ physics │ for experiments with entangled photons establishing the violation of Bell inequalities and pioneering quantum information science │ +6. │ John Clauser │ 2022 │ physics │ for experiments with entangled photons establishing the violation of Bell inequalities and pioneering quantum information science │ +7. │ Syukuro Manabe │ 2021 │ physics │ for the physical modelling of Earths climate quantifying variability and reliably predicting global warming │ +8. │ Klaus Hasselmann │ 2021 │ physics │ for the physical modelling of Earths climate quantifying variability and reliably predicting global warming │ +9. │ Giorgio Parisi │ 2021 │ physics │ for the discovery of the interplay of disorder and fluctuations in physical systems from atomic to planetary scales │ + └──────────────────┴──────┴──────────┴────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┘ +``` \ No newline at end of file diff --git a/docs/en/sql-reference/window-functions/nth_value.md b/docs/en/sql-reference/window-functions/nth_value.md new file mode 100644 index 00000000000..aa5baf651a8 --- /dev/null +++ b/docs/en/sql-reference/window-functions/nth_value.md @@ -0,0 +1,75 @@ +--- +slug: /en/sql-reference/window-functions/nth_value +sidebar_label: nth_value +sidebar_position: 5 +--- + +# nth_value + +Returns the first non-NULL value evaluated against the nth row (offset) in its ordered frame. + +**Syntax** + +```sql +nth_value (x, offset) + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [ROWS or RANGE expression_to_bound_rows_withing_the_group]] | [window_name]) +FROM table_name +WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column]) +``` + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Parameters** + +- `x` — Column name. +- `offset` — nth row to evaluate current row against. + +**Returned value** + +- The first non-NULL value evaluated against the nth row (offset) in its ordered frame. + +**Example** + +In this example the `nth-value` function is used to find the third-highest salary from a fictional dataset of salaries of Premier League football players. + +Query: + +```sql +DROP TABLE IF EXISTS salaries; +CREATE TABLE salaries +( + `team` String, + `player` String, + `salary` UInt32, + `position` String +) +Engine = Memory; + +INSERT INTO salaries FORMAT Values + ('Port Elizabeth Barbarians', 'Gary Chen', 195000, 'F'), + ('New Coreystad Archdukes', 'Charles Juarez', 190000, 'F'), + ('Port Elizabeth Barbarians', 'Michael Stanley', 100000, 'D'), + ('New Coreystad Archdukes', 'Scott Harrison', 180000, 'D'), + ('Port Elizabeth Barbarians', 'Robert George', 195000, 'M'), + ('South Hampton Seagulls', 'Douglas Benson', 150000, 'M'), + ('South Hampton Seagulls', 'James Henderson', 140000, 'M'); +``` + +```sql +SELECT player, salary, nth_value(player,3) OVER(ORDER BY salary DESC) AS third_highest_salary FROM salaries; +``` + +Result: + +```response + ┌─player──────────┬─salary─┬─third_highest_salary─┐ +1. │ Gary Chen │ 195000 │ │ +2. │ Robert George │ 195000 │ │ +3. │ Charles Juarez │ 190000 │ Charles Juarez │ +4. │ Scott Harrison │ 180000 │ Charles Juarez │ +5. │ Douglas Benson │ 150000 │ Charles Juarez │ +6. │ James Henderson │ 140000 │ Charles Juarez │ +7. │ Michael Stanley │ 100000 │ Charles Juarez │ + └─────────────────┴────────┴──────────────────────┘ +``` \ No newline at end of file diff --git a/docs/en/sql-reference/window-functions/rank.md b/docs/en/sql-reference/window-functions/rank.md new file mode 100644 index 00000000000..dff5e154151 --- /dev/null +++ b/docs/en/sql-reference/window-functions/rank.md @@ -0,0 +1,74 @@ +--- +slug: /en/sql-reference/window-functions/rank +sidebar_label: rank +sidebar_position: 6 +--- + +# rank + +Ranks the current row within its partition with gaps. In other words, if the value of any row it encounters is equal to the value of a previous row then it will receive the same rank as that previous row. +The rank of the next row is then equal to the rank of the previous row plus a gap equal to the number of times the previous rank was given. + +The [dense_rank](./dense_rank.md) function provides the same behaviour but without gaps in ranking. + +**Syntax** + +```sql +rank (column_name) + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [ROWS or RANGE expression_to_bound_rows_withing_the_group]] | [window_name]) +FROM table_name +WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column]) +``` + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Returned value** + +- A number for the current row within its partition, including gaps. [UInt64](../data-types/int-uint.md). + +**Example** + +The following example is based on the example provided in the video instructional [Ranking window functions in ClickHouse](https://youtu.be/Yku9mmBYm_4?si=XIMu1jpYucCQEoXA). + +Query: + +```sql +CREATE TABLE salaries +( + `team` String, + `player` String, + `salary` UInt32, + `position` String +) +Engine = Memory; + +INSERT INTO salaries FORMAT Values + ('Port Elizabeth Barbarians', 'Gary Chen', 195000, 'F'), + ('New Coreystad Archdukes', 'Charles Juarez', 190000, 'F'), + ('Port Elizabeth Barbarians', 'Michael Stanley', 150000, 'D'), + ('New Coreystad Archdukes', 'Scott Harrison', 150000, 'D'), + ('Port Elizabeth Barbarians', 'Robert George', 195000, 'M'), + ('South Hampton Seagulls', 'Douglas Benson', 150000, 'M'), + ('South Hampton Seagulls', 'James Henderson', 140000, 'M'); +``` + +```sql +SELECT player, salary, + rank() OVER (ORDER BY salary DESC) AS rank +FROM salaries; +``` + +Result: + +```response + ┌─player──────────┬─salary─┬─rank─┐ +1. │ Gary Chen │ 195000 │ 1 │ +2. │ Robert George │ 195000 │ 1 │ +3. │ Charles Juarez │ 190000 │ 3 │ +4. │ Douglas Benson │ 150000 │ 4 │ +5. │ Michael Stanley │ 150000 │ 4 │ +6. │ Scott Harrison │ 150000 │ 4 │ +7. │ James Henderson │ 140000 │ 7 │ + └─────────────────┴────────┴──────┘ +``` \ No newline at end of file diff --git a/docs/en/sql-reference/window-functions/row_number.md b/docs/en/sql-reference/window-functions/row_number.md new file mode 100644 index 00000000000..f1c331f89a3 --- /dev/null +++ b/docs/en/sql-reference/window-functions/row_number.md @@ -0,0 +1,67 @@ +--- +slug: /en/sql-reference/window-functions/row_number +sidebar_label: row_number +sidebar_position: 2 +--- + +# row_number + +Numbers the current row within its partition starting from 1. + +**Syntax** + +```sql +row_number (column_name) + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [ROWS or RANGE expression_to_bound_rows_withing_the_group]] | [window_name]) +FROM table_name +WINDOW window_name as ([[PARTITION BY grouping_column] [ORDER BY sorting_column]) +``` + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Returned value** + +- A number for the current row within its partition. [UInt64](../data-types/int-uint.md). + +**Example** + +The following example is based on the example provided in the video instructional [Ranking window functions in ClickHouse](https://youtu.be/Yku9mmBYm_4?si=XIMu1jpYucCQEoXA). + +Query: + +```sql +CREATE TABLE salaries +( + `team` String, + `player` String, + `salary` UInt32, + `position` String +) +Engine = Memory; + +INSERT INTO salaries FORMAT Values + ('Port Elizabeth Barbarians', 'Gary Chen', 195000, 'F'), + ('New Coreystad Archdukes', 'Charles Juarez', 190000, 'F'), + ('Port Elizabeth Barbarians', 'Michael Stanley', 150000, 'D'), + ('New Coreystad Archdukes', 'Scott Harrison', 150000, 'D'), + ('Port Elizabeth Barbarians', 'Robert George', 195000, 'M'); +``` + +```sql +SELECT player, salary, + row_number() OVER (ORDER BY salary DESC) AS row_number +FROM salaries; +``` + +Result: + +```response + ┌─player──────────┬─salary─┬─row_number─┐ +1. │ Gary Chen │ 195000 │ 1 │ +2. │ Robert George │ 195000 │ 2 │ +3. │ Charles Juarez │ 190000 │ 3 │ +4. │ Scott Harrison │ 150000 │ 4 │ +5. │ Michael Stanley │ 150000 │ 5 │ + └─────────────────┴────────┴────────────┘ +``` \ No newline at end of file diff --git a/docs/ru/getting-started/playground.md b/docs/ru/getting-started/playground.md index c8c987eec9e..a2d5498fb9a 100644 --- a/docs/ru/getting-started/playground.md +++ b/docs/ru/getting-started/playground.md @@ -6,38 +6,38 @@ sidebar_label: Playground # ClickHouse Playground {#clickhouse-playground} -[ClickHouse Playground](https://play.clickhouse.com/play?user=play) allows people to experiment with ClickHouse by running queries instantly, without setting up their server or cluster. -Several example datasets are available in Playground. +[ClickHouse Playground](https://play.clickhouse.com/play?user=play) позволяет пользователям экспериментировать с ClickHouse, выполняя запросы мгновенно, без необходимости настройки сервера или кластера. +В Playground доступны несколько примеров наборов данных. -You can make queries to Playground using any HTTP client, for example [curl](https://curl.haxx.se) or [wget](https://www.gnu.org/software/wget/), or set up a connection using [JDBC](../interfaces/jdbc.md) or [ODBC](../interfaces/odbc.md) drivers. More information about software products that support ClickHouse is available [here](../interfaces/index.md). +Вы можете выполнять запросы к Playground, используя любой HTTP-клиент, например [curl](https://curl.haxx.se) или [wget](https://www.gnu.org/software/wget/), или настроить соединение, используя драйверы [JDBC](../interfaces/jdbc.md) или [ODBC](../interfaces/odbc.md). Дополнительную информацию о программных продуктах, поддерживающих ClickHouse, можно найти [здесь](../interfaces/index.md). -## Credentials {#credentials} +## Учетные данные {#credentials} -| Parameter | Value | +| Параметр | Значение | |:--------------------|:-----------------------------------| -| HTTPS endpoint | `https://play.clickhouse.com:443/` | -| Native TCP endpoint | `play.clickhouse.com:9440` | -| User | `explorer` or `play` | -| Password | (empty) | +| HTTPS-адрес | `https://play.clickhouse.com:443/` | +| TCP-адрес | `play.clickhouse.com:9440` | +| Пользователь | `explorer` или `play` | +| Пароль | (пусто) | -## Limitations {#limitations} +## Ограничения {#limitations} -The queries are executed as a read-only user. It implies some limitations: +Запросы выполняются от имени пользователя с правами только на чтение. Это предполагает некоторые ограничения: -- DDL queries are not allowed -- INSERT queries are not allowed +- DDL-запросы не разрешены +- INSERT-запросы не разрешены -The service also have quotas on its usage. +Сервис также имеет квоты на использование. -## Examples {#examples} +## Примеры {#examples} -HTTPS endpoint example with `curl`: +Пример использования HTTPS-адреса с `curl`: -``` bash +```bash curl "https://play.clickhouse.com/?user=explorer" --data-binary "SELECT 'Play ClickHouse'" ``` -TCP endpoint example with [CLI](../interfaces/cli.md): +Пример использования TCP-адреса с [CLI](../interfaces/cli.md): ``` bash clickhouse client --secure --host play.clickhouse.com --user explorer diff --git a/docs/ru/introduction/distinctive-features.md b/docs/ru/introduction/distinctive-features.md index dafaf055980..da820c90a1e 100644 --- a/docs/ru/introduction/distinctive-features.md +++ b/docs/ru/introduction/distinctive-features.md @@ -12,7 +12,7 @@ sidebar_label: "Отличительные возможности ClickHouse" Этот пункт пришлось выделить, так как существуют системы, которые могут хранить значения отдельных столбцов по отдельности, но не могут эффективно выполнять аналитические запросы в силу оптимизации под другой сценарий работы. Примеры: HBase, BigTable, Cassandra, HyperTable. В этих системах вы получите пропускную способность в районе сотен тысяч строк в секунду, но не сотен миллионов строк в секунду. -Также стоит заметить, что ClickHouse является системой управления базами данных, а не системой для одной базой данных. То есть, ClickHouse позволяет создавать таблицы и базы данных во время выполнения (runtime), загружать данные и выполнять запросы без переконфигурирования и перезапуска сервера. +Также стоит заметить, что ClickHouse является системой управления базами данных, а не системой для одной базы данных. То есть, ClickHouse позволяет создавать таблицы и базы данных во время выполнения (runtime), загружать данные и выполнять запросы без переконфигурирования и перезапуска сервера. ## Сжатие данных {#szhatie-dannykh} diff --git a/docs/ru/operations/external-authenticators/ssl-x509.md b/docs/ru/operations/external-authenticators/ssl-x509.md index affdf87b199..7f1fb03962c 100644 --- a/docs/ru/operations/external-authenticators/ssl-x509.md +++ b/docs/ru/operations/external-authenticators/ssl-x509.md @@ -3,23 +3,30 @@ slug: /ru/operations/external-authenticators/ssl-x509 --- # Аутентификация по сертификату SSL X.509 {#ssl-external-authentication} -[Опция 'strict'](../server-configuration-parameters/settings.md#server_configuration_parameters-openssl) включает обязательную проверку сертификатов входящих соединений в библиотеке `SSL`. В этом случае могут быть установлены только соединения, представившие действительный сертификат. Соединения с недоверенными сертификатами будут отвергнуты. Таким образом, проверка сертификата позволяет однозначно аутентифицировать входящее соединение. Идентификация пользователя осуществляется по полю `Common Name` сертификата. Это позволяет ассоциировать несколько сертификатов с одним и тем же пользователем. Дополнительно, перевыпуск и отзыв сертификата не требуют изменения конфигурации ClickHouse. +[Опция 'strict'](../server-configuration-parameters/settings.md#server_configuration_parameters-openssl) включает обязательную проверку сертификатов входящих соединений в библиотеке `SSL`. В этом случае могут быть установлены только соединения, представившие действительный сертификат. Соединения с недоверенными сертификатами будут отвергнуты. Таким образом, проверка сертификата позволяет однозначно аутентифицировать входящее соединение. Идентификация пользователя осуществляется по полю `Common Name` или `subjectAltName` сертификата. Это позволяет ассоциировать несколько сертификатов с одним и тем же пользователем. Дополнительно, перевыпуск и отзыв сертификата не требуют изменения конфигурации ClickHouse. -Для включения аутентификации по SSL сертификату, необходимо указать список `Common Name` для каждого пользователя ClickHouse в файле настройки `config.xml`: +Для включения аутентификации по SSL сертификату, необходимо указать список `Common Name` или `subjectAltName` для каждого пользователя ClickHouse в файле настройки `config.xml`: **Example** ```xml - - + + host.domain.com:example_user host.domain.com:example_user_dev - + - + + + + DNS:host.domain.com + + + + ``` diff --git a/docs/ru/operations/settings/settings.md b/docs/ru/operations/settings/settings.md index 3a70a0bac12..84bbf6c83d3 100644 --- a/docs/ru/operations/settings/settings.md +++ b/docs/ru/operations/settings/settings.md @@ -2077,7 +2077,7 @@ SELECT * FROM test_table - 0 — оптимизация отключена. - 1 — оптимизация включена. -Значение по умолчанию: `0`. +Значение по умолчанию: `1`. ## optimize_trivial_count_query {#optimize-trivial-count-query} diff --git a/docs/ru/sql-reference/functions/string-functions.md b/docs/ru/sql-reference/functions/string-functions.md index fa76e84f130..2436581fc7f 100644 --- a/docs/ru/sql-reference/functions/string-functions.md +++ b/docs/ru/sql-reference/functions/string-functions.md @@ -538,7 +538,7 @@ SELECT base58Decode('3dc8KtHrwM'); Синоним: `TO_BASE64`. -## base64UrlEncode(s) +## base64URLEncode(s) Производит кодирование URL (String или FixedString) в base64-представление в соответствии с [RFC 4648](https://tools.ietf.org/html/rfc4648). @@ -548,7 +548,7 @@ SELECT base58Decode('3dc8KtHrwM'); Синоним: `FROM_BASE64`. -## base64UrlDecode(s) +## base64URLDecode(s) Декодирует base64-представление URL в исходную строку в соответствии с [RFC 4648](https://tools.ietf.org/html/rfc4648). При невозможности декодирования выбрасывает исключение @@ -556,9 +556,9 @@ SELECT base58Decode('3dc8KtHrwM'); Функционал аналогичен base64Decode, но при невозможности декодирования возвращает пустую строку. -## tryBase64UrlDecode(s) +## tryBase64URLDecode(s) -Функционал аналогичен base64UrlDecode, но при невозможности декодирования возвращает пустую строку. +Функционал аналогичен base64URLDecode, но при невозможности декодирования возвращает пустую строку. ## endsWith(s, suffix) {#endswith} diff --git a/docs/ru/sql-reference/statements/create/user.md b/docs/ru/sql-reference/statements/create/user.md index 76cfdb251dc..fac2cacf8cc 100644 --- a/docs/ru/sql-reference/statements/create/user.md +++ b/docs/ru/sql-reference/statements/create/user.md @@ -13,7 +13,7 @@ sidebar_label: "Пользователь" ``` sql CREATE USER [IF NOT EXISTS | OR REPLACE] name1 [ON CLUSTER cluster_name1] [, name2 [ON CLUSTER cluster_name2] ...] - [NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name'} | {WITH ssh_key BY KEY 'public_key' TYPE 'ssh-rsa|...'}] + [NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name' | SAN 'TYPE:subject_alt_name'} | {WITH ssh_key BY KEY 'public_key' TYPE 'ssh-rsa|...'}] [HOST {LOCAL | NAME 'name' | REGEXP 'name_regexp' | IP 'address' | LIKE 'pattern'} [,...] | ANY | NONE] [DEFAULT ROLE role [,...]] [DEFAULT DATABASE database | NONE] diff --git a/docs/zh/changelog/index.md b/docs/zh/changelog/index.md index c91d8bcf4d1..cd77a8c03cf 100644 --- a/docs/zh/changelog/index.md +++ b/docs/zh/changelog/index.md @@ -356,7 +356,7 @@ sidebar_label: "\u53D8\u66F4\u65E5\u5FD7" #### 新功能 {#new-feature-1} -- 添加 `deduplicate_blocks_in_dependent_materialized_views` 用于控制具有实例化视图的表中幂等插入的行为的选项。 这个新功能是由Altinity的特殊要求添加到错误修正版本中的。 +- 添加 `deduplicate_blocks_in_dependent_materialized_views` 用于控制具有实例化视图的表中幂等插入的行为的选项。 [#9070](https://github.com/ClickHouse/ClickHouse/pull/9070) [(urykhy)](https://github.com/urykhy) ### ClickHouse版本v20.1.2.4,2020-01-22 {#clickhouse-release-v20-1-2-4-2020-01-22} diff --git a/docs/zh/engines/table-engines/mergetree-family/mergetree.md b/docs/zh/engines/table-engines/mergetree-family/mergetree.md index 67bd681269b..d5ece5b23a9 100644 --- a/docs/zh/engines/table-engines/mergetree-family/mergetree.md +++ b/docs/zh/engines/table-engines/mergetree-family/mergetree.md @@ -201,18 +201,18 @@ ClickHouse 不要求主键唯一,所以您可以插入多条具有相同主键 主键中列的数量并没有明确的限制。依据数据结构,您可以在主键包含多些或少些列。这样可以: - - 改善索引的性能。 +- 改善索引的性能。 - - 如果当前主键是 `(a, b)` ,在下列情况下添加另一个 `c` 列会提升性能: + 如果当前主键是 `(a, b)` ,在下列情况下添加另一个 `c` 列会提升性能: - - 查询会使用 `c` 列作为条件 - - 很长的数据范围( `index_granularity` 的数倍)里 `(a, b)` 都是相同的值,并且这样的情况很普遍。换言之,就是加入另一列后,可以让您的查询略过很长的数据范围。 + - 查询会使用 `c` 列作为条件 + - 很长的数据范围( `index_granularity` 的数倍)里 `(a, b)` 都是相同的值,并且这样的情况很普遍。换言之,就是加入另一列后,可以让您的查询略过很长的数据范围。 - - 改善数据压缩。 +- 改善数据压缩。 - ClickHouse 以主键排序片段数据,所以,数据的一致性越高,压缩越好。 + ClickHouse 以主键排序片段数据,所以,数据的一致性越高,压缩越好。 - - 在[CollapsingMergeTree](collapsingmergetree.md#table_engine-collapsingmergetree) 和 [SummingMergeTree](summingmergetree.md) 引擎里进行数据合并时会提供额外的处理逻辑。 +- 在[CollapsingMergeTree](collapsingmergetree.md#table_engine-collapsingmergetree) 和 [SummingMergeTree](summingmergetree.md) 引擎里进行数据合并时会提供额外的处理逻辑。 在这种情况下,指定与主键不同的 *排序键* 也是有意义的。 diff --git a/docs/zh/sql-reference/functions/conditional-functions.md b/docs/zh/sql-reference/functions/conditional-functions.md index e0f8f7a4777..378e4d897c0 100644 --- a/docs/zh/sql-reference/functions/conditional-functions.md +++ b/docs/zh/sql-reference/functions/conditional-functions.md @@ -82,14 +82,14 @@ FROM LEFT_RIGHT SELECT left, right, - if(left < right, 'left is smaller than right', 'right is greater or equal than left') AS is_smaller + if(left < right, 'left is smaller than right', 'right is smaller or equal than left') AS is_smaller FROM LEFT_RIGHT WHERE isNotNull(left) AND isNotNull(right) ┌─left─┬─right─┬─is_smaller──────────────────────────┐ │ 1 │ 3 │ left is smaller than right │ -│ 2 │ 2 │ right is greater or equal than left │ -│ 3 │ 1 │ right is greater or equal than left │ +│ 2 │ 2 │ right is smaller or equal than left │ +│ 3 │ 1 │ right is smaller or equal than left │ └──────┴───────┴─────────────────────────────────────┘ ``` diff --git a/programs/CMakeLists.txt b/programs/CMakeLists.txt index 4640882f2be..ce3a4659e0e 100644 --- a/programs/CMakeLists.txt +++ b/programs/CMakeLists.txt @@ -4,6 +4,9 @@ if (USE_CLANG_TIDY) set (CMAKE_CXX_CLANG_TIDY "${CLANG_TIDY_PATH}") endif () +set(MAX_LINKER_MEMORY 3500) +include(../cmake/limit_jobs.cmake) + include(${ClickHouse_SOURCE_DIR}/cmake/split_debug_symbols.cmake) # The `clickhouse` binary is a multi purpose tool that contains multiple execution modes (client, server, etc.), @@ -66,18 +69,18 @@ else() message(STATUS "Library bridge mode: OFF") endif() -if (ENABLE_CLICKHOUSE_KEEPER) - message(STATUS "ClickHouse keeper mode: ON") -else() - message(STATUS "ClickHouse keeper mode: OFF") -endif() - if (ENABLE_CLICKHOUSE_KEEPER_CONVERTER) message(STATUS "ClickHouse keeper-converter mode: ON") else() message(STATUS "ClickHouse keeper-converter mode: OFF") endif() +if (ENABLE_CLICKHOUSE_KEEPER) + message(STATUS "ClickHouse Keeper: ON") +else() + message(STATUS "ClickHouse Keeper: OFF") +endif() + if (ENABLE_CLICKHOUSE_KEEPER_CLIENT) message(STATUS "ClickHouse keeper-client mode: ON") else() @@ -131,10 +134,6 @@ add_subdirectory (static-files-disk-uploader) add_subdirectory (su) add_subdirectory (disks) -if (ENABLE_CLICKHOUSE_KEEPER) - add_subdirectory (keeper) -endif() - if (ENABLE_CLICKHOUSE_KEEPER_CONVERTER) add_subdirectory (keeper-converter) endif() @@ -143,6 +142,10 @@ if (ENABLE_CLICKHOUSE_KEEPER_CLIENT) add_subdirectory (keeper-client) endif() +if (ENABLE_CLICKHOUSE_KEEPER) + add_subdirectory (keeper) +endif() + if (ENABLE_CLICKHOUSE_ODBC_BRIDGE) add_subdirectory (odbc-bridge) endif () diff --git a/programs/client/Client.cpp b/programs/client/Client.cpp index c4878b18f00..887c5cb86bc 100644 --- a/programs/client/Client.cpp +++ b/programs/client/Client.cpp @@ -24,9 +24,8 @@ #include #include #include - +#include #include -#include #include #include @@ -49,6 +48,8 @@ #include #include +#include + namespace fs = std::filesystem; using namespace std::literals; @@ -185,6 +186,8 @@ void Client::parseConnectionsCredentials(Poco::Util::AbstractConfiguration & con history_file = home_path + "/" + history_file.substr(1); config.setString("history_file", history_file); } + if (config.has(prefix + ".accept-invalid-certificate")) + config.setBool("accept-invalid-certificate", config.getBool(prefix + ".accept-invalid-certificate")); } if (!connection_name.empty() && !connection_found) @@ -248,6 +251,10 @@ std::vector Client::loadWarningMessages() } } +Poco::Util::LayeredConfiguration & Client::getClientConfiguration() +{ + return config(); +} void Client::initialize(Poco::Util::Application & self) { @@ -272,6 +279,12 @@ void Client::initialize(Poco::Util::Application & self) else if (config().has("connection")) throw Exception(ErrorCodes::BAD_ARGUMENTS, "--connection was specified, but config does not exist"); + if (config().has("accept-invalid-certificate")) + { + config().setString("openSSL.client.invalidCertificateHandler.name", "AcceptCertificateHandler"); + config().setString("openSSL.client.verificationMode", "none"); + } + /** getenv is thread-safe in Linux glibc and in all sane libc implementations. * But the standard does not guarantee that subsequent calls will not rewrite the value by returned pointer. * @@ -697,9 +710,7 @@ bool Client::processWithFuzzing(const String & full_query) const char * begin = full_query.data(); orig_ast = parseQuery(begin, begin + full_query.size(), global_context->getSettingsRef(), - /*allow_multi_statements=*/ true, - /*is_interactive=*/ is_interactive, - /*ignore_error=*/ ignore_error); + /*allow_multi_statements=*/ true); } catch (const Exception & e) { @@ -728,7 +739,7 @@ bool Client::processWithFuzzing(const String & full_query) } if (auto *q = orig_ast->as()) { - if (auto *setDialect = q->changes.tryGet("dialect"); setDialect && setDialect->safeGet() == "kusto") + if (auto *set_dialect = q->changes.tryGet("dialect"); set_dialect && set_dialect->safeGet() == "kusto") return true; } @@ -1115,6 +1126,7 @@ void Client::processOptions(const OptionsDescription & options_description, if (!options["user"].defaulted()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "User and JWT flags can't be specified together"); config().setString("jwt", options["jwt"].as()); + config().setString("user", ""); } if (options.count("accept-invalid-certificate")) { diff --git a/programs/client/Client.h b/programs/client/Client.h index bef948b3c1e..229608f787d 100644 --- a/programs/client/Client.h +++ b/programs/client/Client.h @@ -16,6 +16,9 @@ public: int main(const std::vector & /*args*/) override; protected: + + Poco::Util::LayeredConfiguration & getClientConfiguration() override; + bool processWithFuzzing(const String & full_query) override; std::optional processFuzzingStep(const String & query_to_execute, const ASTPtr & parsed_query); diff --git a/programs/client/clickhouse-client.xml b/programs/client/clickhouse-client.xml index d0deb818c1e..c32b63413e9 100644 --- a/programs/client/clickhouse-client.xml +++ b/programs/client/clickhouse-client.xml @@ -1,5 +1,6 @@ + true @@ -72,6 +73,7 @@ Default: "hostname" will be used. --> default + 127.0.0.1 9000 diff --git a/programs/disks/CMakeLists.txt b/programs/disks/CMakeLists.txt index f0949fcfceb..7e8afe084fb 100644 --- a/programs/disks/CMakeLists.txt +++ b/programs/disks/CMakeLists.txt @@ -1,6 +1,8 @@ set (CLICKHOUSE_DISKS_SOURCES DisksApp.cpp + DisksClient.cpp ICommand.cpp + CommandChangeDirectory.cpp CommandCopy.cpp CommandLink.cpp CommandList.cpp @@ -9,10 +11,14 @@ set (CLICKHOUSE_DISKS_SOURCES CommandMove.cpp CommandRead.cpp CommandRemove.cpp - CommandWrite.cpp) + CommandSwitchDisk.cpp + CommandWrite.cpp + CommandHelp.cpp + CommandTouch.cpp + CommandGetCurrentDiskAndPath.cpp) if (CLICKHOUSE_CLOUD) - set (CLICKHOUSE_DISKS_SOURCES ${CLICKHOUSE_DISKS_SOURCES} CommandPackedIO.cpp) + set (CLICKHOUSE_DISKS_SOURCES ${CLICKHOUSE_DISKS_SOURCES} CommandPackedIO.cpp) endif () set (CLICKHOUSE_DISKS_LINK diff --git a/programs/disks/CommandChangeDirectory.cpp b/programs/disks/CommandChangeDirectory.cpp new file mode 100644 index 00000000000..b545f37de72 --- /dev/null +++ b/programs/disks/CommandChangeDirectory.cpp @@ -0,0 +1,35 @@ +#include +#include +#include "DisksApp.h" +#include "DisksClient.h" +#include "ICommand.h" + +namespace DB +{ + +class CommandChangeDirectory final : public ICommand +{ +public: + explicit CommandChangeDirectory() : ICommand() + { + command_name = "cd"; + description = "Change directory (makes sense only in interactive mode)"; + options_description.add_options()("path", po::value(), "the path to which we want to change (mandatory, positional)")( + "disk", po::value(), "A disk where the path is changed (without disk switching)"); + positional_options_description.add("path", 1); + } + + void executeImpl(const CommandLineOptions & options, DisksClient & client) override + { + DiskWithPath & disk = getDiskWithPath(client, options, "disk"); + String path = getValueFromCommandLineOptionsThrow(options, "path"); + disk.setPath(path); + } +}; + +CommandPtr makeCommandChangeDirectory() +{ + return std::make_shared(); +} + +} diff --git a/programs/disks/CommandCopy.cpp b/programs/disks/CommandCopy.cpp index f176fa277d7..e3051f2702c 100644 --- a/programs/disks/CommandCopy.cpp +++ b/programs/disks/CommandCopy.cpp @@ -1,6 +1,8 @@ -#include "ICommand.h" #include +#include "Common/Exception.h" #include +#include "DisksClient.h" +#include "ICommand.h" namespace DB { @@ -10,59 +12,89 @@ namespace ErrorCodes extern const int BAD_ARGUMENTS; } + class CommandCopy final : public ICommand { public: - CommandCopy() + explicit CommandCopy() : ICommand() { command_name = "copy"; - command_option_description.emplace(createOptionsDescription("Allowed options", getTerminalWidth())); - description = "Recursively copy data from `FROM_PATH` to `TO_PATH`"; - usage = "copy [OPTION]... "; - command_option_description->add_options() - ("disk-from", po::value(), "disk from which we copy") - ("disk-to", po::value(), "disk to which we copy"); + description = "Recursively copy data from `path-from` to `path-to`"; + options_description.add_options()( + "disk-from", po::value(), "disk from which we copy is executed (default value is a current disk)")( + "disk-to", po::value(), "disk to which copy is executed (default value is a current disk)")( + "path-from", po::value(), "path from which copy is executed (mandatory, positional)")( + "path-to", po::value(), "path to which copy is executed (mandatory, positional)")( + "recursive,r", "recursively copy the directory (required to remove a directory)"); + positional_options_description.add("path-from", 1); + positional_options_description.add("path-to", 1); } - void processOptions( - Poco::Util::LayeredConfiguration & config, - po::variables_map & options) const override + void executeImpl(const CommandLineOptions & options, DisksClient & client) override { - if (options.count("disk-from")) - config.setString("disk-from", options["disk-from"].as()); - if (options.count("disk-to")) - config.setString("disk-to", options["disk-to"].as()); - } + auto disk_from = getDiskWithPath(client, options, "disk-from"); + auto disk_to = getDiskWithPath(client, options, "disk-to"); + String path_from = disk_from.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path-from")); + String path_to = disk_to.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path-to")); + bool recursive = options.count("recursive"); - void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) override - { - if (command_arguments.size() != 2) + if (!disk_from.getDisk()->exists(path_from)) { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "cannot stat '{}' on disk '{}': No such file or directory", + path_from, + disk_from.getDisk()->getName()); } + else if (disk_from.getDisk()->isFile(path_from)) + { + auto target_location = getTargetLocation(path_from, disk_to, path_to); + if (!disk_to.getDisk()->exists(target_location) || disk_to.getDisk()->isFile(target_location)) + { + disk_from.getDisk()->copyFile( + path_from, + *disk_to.getDisk(), + target_location, + /* read_settings= */ {}, + /* write_settings= */ {}, + /* cancellation_hook= */ {}); + } + else + { + throw Exception( + ErrorCodes::BAD_ARGUMENTS, "cannot overwrite directory {} with non-directory {}", target_location, path_from); + } + } + else if (disk_from.getDisk()->isDirectory(path_from)) + { + if (!recursive) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "--recursive not specified; omitting directory {}", path_from); + } + auto target_location = getTargetLocation(path_from, disk_to, path_to); - String disk_name_from = config.getString("disk-from", config.getString("disk", "default")); - String disk_name_to = config.getString("disk-to", config.getString("disk", "default")); - - const String & path_from = command_arguments[0]; - const String & path_to = command_arguments[1]; - - DiskPtr disk_from = disk_selector->get(disk_name_from); - DiskPtr disk_to = disk_selector->get(disk_name_to); - - String relative_path_from = validatePathAndGetAsRelative(path_from); - String relative_path_to = validatePathAndGetAsRelative(path_to); - - disk_from->copyDirectoryContent(relative_path_from, disk_to, relative_path_to, /* read_settings= */ {}, /* write_settings= */ {}, /* cancellation_hook= */ {}); + if (disk_to.getDisk()->isFile(target_location)) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "cannot overwrite non-directory {} with directory {}", path_to, target_location); + } + else if (!disk_to.getDisk()->exists(target_location)) + { + disk_to.getDisk()->createDirectory(target_location); + } + disk_from.getDisk()->copyDirectoryContent( + path_from, + disk_to.getDisk(), + target_location, + /* read_settings= */ {}, + /* write_settings= */ {}, + /* cancellation_hook= */ {}); + } } }; + +CommandPtr makeCommandCopy() +{ + return std::make_shared(); } -std::unique_ptr makeCommandCopy() -{ - return std::make_unique(); } diff --git a/programs/disks/CommandGetCurrentDiskAndPath.cpp b/programs/disks/CommandGetCurrentDiskAndPath.cpp new file mode 100644 index 00000000000..15f8ef5aae8 --- /dev/null +++ b/programs/disks/CommandGetCurrentDiskAndPath.cpp @@ -0,0 +1,30 @@ +#include +#include +#include "DisksApp.h" +#include "DisksClient.h" +#include "ICommand.h" + +namespace DB +{ + +class CommandGetCurrentDiskAndPath final : public ICommand +{ +public: + explicit CommandGetCurrentDiskAndPath() : ICommand() + { + command_name = "current_disk_with_path"; + description = "Prints current disk and path (which coincide with the prompt)"; + } + + void executeImpl(const CommandLineOptions &, DisksClient & client) override + { + auto disk = client.getCurrentDiskWithPath(); + std::cout << "Disk: " << disk.getDisk()->getName() << "\nPath: " << disk.getCurrentPath() << std::endl; + } +}; + +CommandPtr makeCommandGetCurrentDiskAndPath() +{ + return std::make_shared(); +} +} diff --git a/programs/disks/CommandHelp.cpp b/programs/disks/CommandHelp.cpp new file mode 100644 index 00000000000..a3aee9498d3 --- /dev/null +++ b/programs/disks/CommandHelp.cpp @@ -0,0 +1,43 @@ +#include "DisksApp.h" +#include "ICommand.h" + +#include +#include + +namespace DB +{ + +class CommandHelp final : public ICommand +{ +public: + explicit CommandHelp(const DisksApp & disks_app_) : disks_app(disks_app_) + { + command_name = "help"; + description = "Print help message about available commands"; + options_description.add_options()( + "command", po::value(), "A command to help with (optional, positional), if not specified, help lists all the commands"); + positional_options_description.add("command", 1); + } + + void executeImpl(const CommandLineOptions & options, DisksClient & /*client*/) override + { + std::optional command = getValueFromCommandLineOptionsWithOptional(options, "command"); + if (command.has_value()) + { + disks_app.printCommandHelpMessage(command.value()); + } + else + { + disks_app.printAvailableCommandsHelpMessage(); + } + } + + const DisksApp & disks_app; +}; + +CommandPtr makeCommandHelp(const DisksApp & disks_app) +{ + return std::make_shared(disks_app); +} + +} diff --git a/programs/disks/CommandLink.cpp b/programs/disks/CommandLink.cpp index dbaa3162f82..11c196cafc5 100644 --- a/programs/disks/CommandLink.cpp +++ b/programs/disks/CommandLink.cpp @@ -1,14 +1,9 @@ -#include "ICommand.h" #include +#include "ICommand.h" namespace DB { -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - class CommandLink final : public ICommand { public: @@ -16,42 +11,27 @@ public: { command_name = "link"; description = "Create hardlink from `from_path` to `to_path`"; - usage = "link [OPTION]... "; + options_description.add_options()( + "path-from", po::value(), "the path from which a hard link will be created (mandatory, positional)")( + "path-to", po::value(), "the path where a hard link will be created (mandatory, positional)"); + positional_options_description.add("path-from", 1); + positional_options_description.add("path-to", 1); } - void processOptions( - Poco::Util::LayeredConfiguration &, - po::variables_map &) const override + void executeImpl(const CommandLineOptions & options, DisksClient & client) override { - } + auto disk = client.getCurrentDiskWithPath(); - void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) override - { - if (command_arguments.size() != 2) - { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); - } + const String & path_from = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path-from")); + const String & path_to = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path-to")); - String disk_name = config.getString("disk", "default"); - - const String & path_from = command_arguments[0]; - const String & path_to = command_arguments[1]; - - DiskPtr disk = disk_selector->get(disk_name); - - String relative_path_from = validatePathAndGetAsRelative(path_from); - String relative_path_to = validatePathAndGetAsRelative(path_to); - - disk->createHardLink(relative_path_from, relative_path_to); + disk.getDisk()->createHardLink(path_from, path_to); } }; + +CommandPtr makeCommandLink() +{ + return std::make_shared(); } -std::unique_ptr makeCommandLink() -{ - return std::make_unique(); } diff --git a/programs/disks/CommandList.cpp b/programs/disks/CommandList.cpp index 7213802ea86..77479b1d217 100644 --- a/programs/disks/CommandList.cpp +++ b/programs/disks/CommandList.cpp @@ -1,98 +1,95 @@ -#include "ICommand.h" #include #include +#include "DisksApp.h" +#include "DisksClient.h" +#include "ICommand.h" namespace DB { -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - class CommandList final : public ICommand { public: - CommandList() + explicit CommandList() : ICommand() { command_name = "list"; - command_option_description.emplace(createOptionsDescription("Allowed options", getTerminalWidth())); description = "List files at path[s]"; - usage = "list [OPTION]... ..."; - command_option_description->add_options() - ("recursive", "recursively list all directories"); + options_description.add_options()("recursive", "recursively list the directory")("all", "show hidden files")( + "path", po::value(), "the path of listing (mandatory, positional)"); + positional_options_description.add("path", 1); } - void processOptions( - Poco::Util::LayeredConfiguration & config, - po::variables_map & options) const override + void executeImpl(const CommandLineOptions & options, DisksClient & client) override { - if (options.count("recursive")) - config.setBool("recursive", true); - } - - void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) override - { - if (command_arguments.size() != 1) - { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); - } - - String disk_name = config.getString("disk", "default"); - - const String & path = command_arguments[0]; - - DiskPtr disk = disk_selector->get(disk_name); - - String relative_path = validatePathAndGetAsRelative(path); - - bool recursive = config.getBool("recursive", false); + bool recursive = options.count("recursive"); + bool show_hidden = options.count("all"); + auto disk = client.getCurrentDiskWithPath(); + String path = getValueFromCommandLineOptionsWithDefault(options, "path", "."); if (recursive) - listRecursive(disk, relative_path); + listRecursive(disk, path, show_hidden); else - list(disk, relative_path); + list(disk, path, show_hidden); } private: - static void list(const DiskPtr & disk, const std::string & relative_path) + static void list(const DiskWithPath & disk, const std::string & path, bool show_hidden) { - std::vector file_names; - disk->listFiles(relative_path, file_names); + std::vector file_names = disk.listAllFilesByPath(path); + std::vector selected_and_sorted_file_names{}; for (const auto & file_name : file_names) - std::cout << file_name << '\n'; + if (show_hidden || (!file_name.starts_with('.'))) + selected_and_sorted_file_names.push_back(file_name); + + std::sort(selected_and_sorted_file_names.begin(), selected_and_sorted_file_names.end()); + for (const auto & file_name : selected_and_sorted_file_names) + { + std::cout << file_name << "\n"; + } } - static void listRecursive(const DiskPtr & disk, const std::string & relative_path) + static void listRecursive(const DiskWithPath & disk, const std::string & relative_path, bool show_hidden) { - std::vector file_names; - disk->listFiles(relative_path, file_names); + std::vector file_names = disk.listAllFilesByPath(relative_path); + std::vector selected_and_sorted_file_names{}; std::cout << relative_path << ":\n"; - if (!file_names.empty()) - { - for (const auto & file_name : file_names) - std::cout << file_name << '\n'; - std::cout << "\n"; - } - for (const auto & file_name : file_names) + if (show_hidden || (!file_name.starts_with('.'))) + selected_and_sorted_file_names.push_back(file_name); + + std::sort(selected_and_sorted_file_names.begin(), selected_and_sorted_file_names.end()); + for (const auto & file_name : selected_and_sorted_file_names) { - auto path = relative_path.empty() ? file_name : (relative_path + "/" + file_name); - if (disk->isDirectory(path)) - listRecursive(disk, path); + std::cout << file_name << "\n"; + } + std::cout << "\n"; + + for (const auto & file_name : selected_and_sorted_file_names) + { + auto path = [&]() -> String + { + if (relative_path.ends_with("/")) + { + return relative_path + file_name; + } + else + { + return relative_path + "/" + file_name; + } + }(); + if (disk.isDirectory(path)) + { + listRecursive(disk, path, show_hidden); + } } } }; -} -std::unique_ptr makeCommandList() +CommandPtr makeCommandList() { - return std::make_unique(); + return std::make_shared(); +} } diff --git a/programs/disks/CommandListDisks.cpp b/programs/disks/CommandListDisks.cpp index 79da021fd00..9fb67fed5e0 100644 --- a/programs/disks/CommandListDisks.cpp +++ b/programs/disks/CommandListDisks.cpp @@ -1,68 +1,40 @@ -#include "ICommand.h" +#include #include +#include +#include "DisksClient.h" +#include "ICommand.h" namespace DB { -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - class CommandListDisks final : public ICommand { public: - CommandListDisks() + explicit CommandListDisks() : ICommand() { command_name = "list-disks"; - description = "List disks names"; - usage = "list-disks [OPTION]"; + description = "Lists all available disks"; } - void processOptions( - Poco::Util::LayeredConfiguration &, - po::variables_map &) const override - {} - - void execute( - const std::vector & command_arguments, - std::shared_ptr &, - Poco::Util::LayeredConfiguration & config) override + void executeImpl(const CommandLineOptions &, DisksClient & client) override { - if (!command_arguments.empty()) + std::vector sorted_and_selected{}; + for (const auto & disk_name : client.getAllDiskNames()) { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); + sorted_and_selected.push_back(disk_name + ":" + client.getDiskWithPath(disk_name).getAbsolutePath("")); } - - constexpr auto config_prefix = "storage_configuration.disks"; - constexpr auto default_disk_name = "default"; - - Poco::Util::AbstractConfiguration::Keys keys; - config.keys(config_prefix, keys); - - bool has_default_disk = false; - - /// For the output to be ordered - std::set disks; - - for (const auto & disk_name : keys) + std::sort(sorted_and_selected.begin(), sorted_and_selected.end()); + for (const auto & disk_name : sorted_and_selected) { - if (disk_name == default_disk_name) - has_default_disk = true; - disks.insert(disk_name); + std::cout << disk_name << "\n"; } - - if (!has_default_disk) - disks.insert(default_disk_name); - - for (const auto & disk : disks) - std::cout << disk << '\n'; } -}; -} -std::unique_ptr makeCommandListDisks() +private: +}; + +CommandPtr makeCommandListDisks() { - return std::make_unique(); + return std::make_shared(); +} } diff --git a/programs/disks/CommandMkDir.cpp b/programs/disks/CommandMkDir.cpp index 6d33bdec498..c6222f326d4 100644 --- a/programs/disks/CommandMkDir.cpp +++ b/programs/disks/CommandMkDir.cpp @@ -6,61 +6,35 @@ namespace DB { -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - class CommandMkDir final : public ICommand { public: CommandMkDir() { command_name = "mkdir"; - command_option_description.emplace(createOptionsDescription("Allowed options", getTerminalWidth())); - description = "Create a directory"; - usage = "mkdir [OPTION]... "; - command_option_description->add_options() - ("recursive", "recursively create directories"); + description = "Creates a directory"; + options_description.add_options()("parents", "recursively create directories")( + "path", po::value(), "the path on which directory should be created (mandatory, positional)"); + positional_options_description.add("path", 1); } - void processOptions( - Poco::Util::LayeredConfiguration & config, - po::variables_map & options) const override + void executeImpl(const CommandLineOptions & options, DisksClient & client) override { - if (options.count("recursive")) - config.setBool("recursive", true); - } + bool recursive = options.count("parents"); + auto disk = client.getCurrentDiskWithPath(); - void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) override - { - if (command_arguments.size() != 1) - { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); - } - - String disk_name = config.getString("disk", "default"); - - const String & path = command_arguments[0]; - - DiskPtr disk = disk_selector->get(disk_name); - - String relative_path = validatePathAndGetAsRelative(path); - bool recursive = config.getBool("recursive", false); + String path = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path")); if (recursive) - disk->createDirectories(relative_path); + disk.getDisk()->createDirectories(path); else - disk->createDirectory(relative_path); + disk.getDisk()->createDirectory(path); } }; + +CommandPtr makeCommandMkDir() +{ + return std::make_shared(); } -std::unique_ptr makeCommandMkDir() -{ - return std::make_unique(); } diff --git a/programs/disks/CommandMove.cpp b/programs/disks/CommandMove.cpp index 75cf96252ed..e3d485032e0 100644 --- a/programs/disks/CommandMove.cpp +++ b/programs/disks/CommandMove.cpp @@ -1,5 +1,5 @@ -#include "ICommand.h" #include +#include "ICommand.h" namespace DB { @@ -9,6 +9,7 @@ namespace ErrorCodes extern const int BAD_ARGUMENTS; } + class CommandMove final : public ICommand { public: @@ -16,44 +17,62 @@ public: { command_name = "move"; description = "Move file or directory from `from_path` to `to_path`"; - usage = "move [OPTION]... "; + options_description.add_options()("path-from", po::value(), "path from which we copy (mandatory, positional)")( + "path-to", po::value(), "path to which we copy (mandatory, positional)"); + positional_options_description.add("path-from", 1); + positional_options_description.add("path-to", 1); } - void processOptions( - Poco::Util::LayeredConfiguration &, - po::variables_map &) const override - {} - - void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) override + void executeImpl(const CommandLineOptions & options, DisksClient & client) override { - if (command_arguments.size() != 2) + auto disk = client.getCurrentDiskWithPath(); + + String path_from = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path-from")); + String path_to = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path-to")); + + if (disk.getDisk()->isFile(path_from)) { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); + disk.getDisk()->moveFile(path_from, path_to); + } + else if (disk.getDisk()->isDirectory(path_from)) + { + auto target_location = getTargetLocation(path_from, disk, path_to); + if (!disk.getDisk()->exists(target_location)) + { + disk.getDisk()->createDirectory(target_location); + disk.getDisk()->moveDirectory(path_from, target_location); + } + else + { + if (disk.getDisk()->isFile(target_location)) + { + throw Exception( + ErrorCodes::BAD_ARGUMENTS, "cannot overwrite non-directory '{}' with directory '{}'", target_location, path_from); + } + if (!disk.getDisk()->isDirectoryEmpty(target_location)) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "cannot move '{}' to '{}': Directory not empty", path_from, target_location); + } + else + { + disk.getDisk()->moveDirectory(path_from, target_location); + } + } + } + else if (!disk.getDisk()->exists(path_from)) + { + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "cannot stat '{}' on disk: '{}': No such file or directory", + path_from, + disk.getDisk()->getName()); } - - String disk_name = config.getString("disk", "default"); - - const String & path_from = command_arguments[0]; - const String & path_to = command_arguments[1]; - - DiskPtr disk = disk_selector->get(disk_name); - - String relative_path_from = validatePathAndGetAsRelative(path_from); - String relative_path_to = validatePathAndGetAsRelative(path_to); - - if (disk->isFile(relative_path_from)) - disk->moveFile(relative_path_from, relative_path_to); - else - disk->moveDirectory(relative_path_from, relative_path_to); } }; + +CommandPtr makeCommandMove() +{ + return std::make_shared(); } -std::unique_ptr makeCommandMove() -{ - return std::make_unique(); } diff --git a/programs/disks/CommandRead.cpp b/programs/disks/CommandRead.cpp index 0f3ac7ab98c..277e735f507 100644 --- a/programs/disks/CommandRead.cpp +++ b/programs/disks/CommandRead.cpp @@ -1,78 +1,52 @@ -#include "ICommand.h" -#include #include #include #include +#include #include +#include "ICommand.h" namespace DB { -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - class CommandRead final : public ICommand { public: CommandRead() { command_name = "read"; - command_option_description.emplace(createOptionsDescription("Allowed options", getTerminalWidth())); - description = "Read a file from `FROM_PATH` to `TO_PATH`"; - usage = "read [OPTION]... []"; - command_option_description->add_options() - ("output", po::value(), "file to which we are reading, defaults to `stdout`"); + description = "Read a file from `path-from` to `path-to`"; + options_description.add_options()("path-from", po::value(), "file from which we are reading (mandatory, positional)")( + "path-to", po::value(), "file to which we are writing, defaults to `stdout`"); + positional_options_description.add("path-from", 1); } - void processOptions( - Poco::Util::LayeredConfiguration & config, - po::variables_map & options) const override + void executeImpl(const CommandLineOptions & options, DisksClient & client) override { - if (options.count("output")) - config.setString("output", options["output"].as()); - } + auto disk = client.getCurrentDiskWithPath(); + String path_from = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path-from")); + std::optional path_to = getValueFromCommandLineOptionsWithOptional(options, "path-to"); - void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) override - { - if (command_arguments.size() != 1) + auto in = disk.getDisk()->readFile(path_from); + std::unique_ptr out = {}; + if (path_to.has_value()) { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); - } - - String disk_name = config.getString("disk", "default"); - - DiskPtr disk = disk_selector->get(disk_name); - - String relative_path = validatePathAndGetAsRelative(command_arguments[0]); - - String path_output = config.getString("output", ""); - - if (!path_output.empty()) - { - String relative_path_output = validatePathAndGetAsRelative(path_output); - - auto in = disk->readFile(relative_path); - auto out = disk->writeFile(relative_path_output); + String relative_path_to = disk.getRelativeFromRoot(path_to.value()); + out = disk.getDisk()->writeFile(relative_path_to); copyData(*in, *out); - out->finalize(); } else { - auto in = disk->readFile(relative_path); - std::unique_ptr out = std::make_unique(STDOUT_FILENO); + out = std::make_unique(STDOUT_FILENO); copyData(*in, *out); + out->write('\n'); } + out->finalize(); } }; + +CommandPtr makeCommandRead() +{ + return std::make_shared(); } -std::unique_ptr makeCommandRead() -{ - return std::make_unique(); } diff --git a/programs/disks/CommandRemove.cpp b/programs/disks/CommandRemove.cpp index 0c631eacff3..4388f6c0b14 100644 --- a/programs/disks/CommandRemove.cpp +++ b/programs/disks/CommandRemove.cpp @@ -1,5 +1,6 @@ -#include "ICommand.h" #include +#include "Common/Exception.h" +#include "ICommand.h" namespace DB { @@ -9,46 +10,49 @@ namespace ErrorCodes extern const int BAD_ARGUMENTS; } + class CommandRemove final : public ICommand { public: CommandRemove() { command_name = "remove"; - description = "Remove file or directory with all children. Throws exception if file doesn't exists.\nPath should be in format './' or './path' or 'path'"; - usage = "remove [OPTION]... "; + description = "Remove file or directory. Throws exception if file doesn't exists"; + options_description.add_options()("path", po::value(), "path that is going to be deleted (mandatory, positional)")( + "recursive,r", "recursively removes the directory (required to remove a directory)"); + positional_options_description.add("path", 1); } - void processOptions( - Poco::Util::LayeredConfiguration &, - po::variables_map &) const override - {} - - void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) override + void executeImpl(const CommandLineOptions & options, DisksClient & client) override { - if (command_arguments.size() != 1) + auto disk = client.getCurrentDiskWithPath(); + const String & path = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path")); + bool recursive = options.count("recursive"); + if (!disk.getDisk()->exists(path)) { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Path {} on disk {} doesn't exist", path, disk.getDisk()->getName()); + } + else if (disk.getDisk()->isDirectory(path)) + { + if (!recursive) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "cannot remove '{}': Is a directory", path); + } + else + { + disk.getDisk()->removeRecursive(path); + } + } + else + { + disk.getDisk()->removeFileIfExists(path); } - - String disk_name = config.getString("disk", "default"); - - const String & path = command_arguments[0]; - - DiskPtr disk = disk_selector->get(disk_name); - - String relative_path = validatePathAndGetAsRelative(path); - - disk->removeRecursive(relative_path); } }; + +CommandPtr makeCommandRemove() +{ + return std::make_shared(); } -std::unique_ptr makeCommandRemove() -{ - return std::make_unique(); } diff --git a/programs/disks/CommandSwitchDisk.cpp b/programs/disks/CommandSwitchDisk.cpp new file mode 100644 index 00000000000..fa02d991365 --- /dev/null +++ b/programs/disks/CommandSwitchDisk.cpp @@ -0,0 +1,35 @@ +#include +#include +#include +#include "DisksApp.h" +#include "ICommand.h" + +namespace DB +{ + +class CommandSwitchDisk final : public ICommand +{ +public: + explicit CommandSwitchDisk() : ICommand() + { + command_name = "switch-disk"; + description = "Switch disk (makes sense only in interactive mode)"; + options_description.add_options()("disk", po::value(), "the disk to switch to (mandatory, positional)")( + "path", po::value(), "the path to switch on the disk"); + positional_options_description.add("disk", 1); + } + + void executeImpl(const CommandLineOptions & options, DisksClient & client) override + { + String disk = getValueFromCommandLineOptions(options, "disk"); + std::optional path = getValueFromCommandLineOptionsWithOptional(options, "path"); + + client.switchToDisk(disk, path); + } +}; + +CommandPtr makeCommandSwitchDisk() +{ + return std::make_shared(); +} +} diff --git a/programs/disks/CommandTouch.cpp b/programs/disks/CommandTouch.cpp new file mode 100644 index 00000000000..c0bdb64cf9e --- /dev/null +++ b/programs/disks/CommandTouch.cpp @@ -0,0 +1,34 @@ +#include +#include +#include "DisksApp.h" +#include "DisksClient.h" +#include "ICommand.h" + +namespace DB +{ + +class CommandTouch final : public ICommand +{ +public: + explicit CommandTouch() : ICommand() + { + command_name = "touch"; + description = "Create a file by path"; + options_description.add_options()("path", po::value(), "the path of listing (mandatory, positional)"); + positional_options_description.add("path", 1); + } + + void executeImpl(const CommandLineOptions & options, DisksClient & client) override + { + auto disk = client.getCurrentDiskWithPath(); + String path = getValueFromCommandLineOptionsThrow(options, "path"); + + disk.getDisk()->createFile(disk.getRelativeFromRoot(path)); + } +}; + +CommandPtr makeCommandTouch() +{ + return std::make_shared(); +} +} diff --git a/programs/disks/CommandWrite.cpp b/programs/disks/CommandWrite.cpp index 7ded37e067a..9c82132e284 100644 --- a/programs/disks/CommandWrite.cpp +++ b/programs/disks/CommandWrite.cpp @@ -1,79 +1,57 @@ -#include "ICommand.h" #include +#include "ICommand.h" -#include #include #include #include +#include namespace DB { -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - class CommandWrite final : public ICommand { public: CommandWrite() { command_name = "write"; - command_option_description.emplace(createOptionsDescription("Allowed options", getTerminalWidth())); - description = "Write a file from `FROM_PATH` to `TO_PATH`"; - usage = "write [OPTION]... [] "; - command_option_description->add_options() - ("input", po::value(), "file from which we are reading, defaults to `stdin`"); + description = "Write a file from `path-from` to `path-to`"; + options_description.add_options()("path-from", po::value(), "file from which we are reading, defaults to `stdin` (input from `stdin` is finished by Ctrl+D)")( + "path-to", po::value(), "file to which we are writing (mandatory, positional)"); + positional_options_description.add("path-to", 1); } - void processOptions( - Poco::Util::LayeredConfiguration & config, - po::variables_map & options) const override + + void executeImpl(const CommandLineOptions & options, DisksClient & client) override { - if (options.count("input")) - config.setString("input", options["input"].as()); - } + auto disk = client.getCurrentDiskWithPath(); - void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) override - { - if (command_arguments.size() != 1) + std::optional path_from = getValueFromCommandLineOptionsWithOptional(options, "path-from"); + + String path_to = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow(options, "path-to")); + + auto in = [&]() -> std::unique_ptr { - printHelpMessage(); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); - } + if (!path_from.has_value()) + { + return std::make_unique(STDIN_FILENO); + } + else + { + String relative_path_from = disk.getRelativeFromRoot(path_from.value()); + return disk.getDisk()->readFile(relative_path_from); + } + }(); - String disk_name = config.getString("disk", "default"); - - const String & path = command_arguments[0]; - - DiskPtr disk = disk_selector->get(disk_name); - - String relative_path = validatePathAndGetAsRelative(path); - - String path_input = config.getString("input", ""); - std::unique_ptr in; - if (path_input.empty()) - { - in = std::make_unique(STDIN_FILENO); - } - else - { - String relative_path_input = validatePathAndGetAsRelative(path_input); - in = disk->readFile(relative_path_input); - } - - auto out = disk->writeFile(relative_path); + auto out = disk.getDisk()->writeFile(path_to); copyData(*in, *out); out->finalize(); } }; + +CommandPtr makeCommandWrite() +{ + return std::make_shared(); } -std::unique_ptr makeCommandWrite() -{ - return std::make_unique(); } diff --git a/programs/disks/DisksApp.cpp b/programs/disks/DisksApp.cpp index 5da5ab4bae9..59ba45b9451 100644 --- a/programs/disks/DisksApp.cpp +++ b/programs/disks/DisksApp.cpp @@ -1,11 +1,22 @@ #include "DisksApp.h" +#include +#include +#include "Common/Exception.h" +#include "Common/filesystemHelpers.h" +#include +#include "DisksClient.h" #include "ICommand.h" +#include "ICommand_fwd.h" + +#include +#include +#include +#include #include -#include #include - +#include namespace DB { @@ -13,74 +24,289 @@ namespace DB namespace ErrorCodes { extern const int BAD_ARGUMENTS; + extern const int LOGICAL_ERROR; +}; + +LineReader::Patterns DisksApp::query_extenders = {"\\"}; +LineReader::Patterns DisksApp::query_delimiters = {""}; +String DisksApp::word_break_characters = " \t\v\f\a\b\r\n"; + +CommandPtr DisksApp::getCommandByName(const String & command) const +{ + try + { + if (auto it = aliases.find(command); it != aliases.end()) + return command_descriptions.at(it->second); + + return command_descriptions.at(command); + } + catch (std::out_of_range &) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "The command `{}` is unknown", command); + } } -size_t DisksApp::findCommandPos(std::vector & common_arguments) +std::vector DisksApp::getEmptyCompletion(String command_name) const { - for (size_t i = 0; i < common_arguments.size(); i++) - if (supported_commands.contains(common_arguments[i])) - return i + 1; - return common_arguments.size(); + auto command_ptr = command_descriptions.at(command_name); + std::vector answer{}; + if (multidisk_commands.contains(command_ptr->command_name)) + { + answer = client->getAllFilesByPatternFromAllDisks(""); + } + else + { + answer = client->getCurrentDiskWithPath().getAllFilesByPattern(""); + } + for (const auto & disk_name : client->getAllDiskNames()) + { + answer.push_back(disk_name); + } + for (const auto & option : command_ptr->options_description.options()) + { + answer.push_back("--" + option->long_name()); + } + if (command_name == "help") + { + for (const auto & [current_command_name, description] : command_descriptions) + { + answer.push_back(current_command_name); + } + } + std::sort(answer.begin(), answer.end()); + return answer; } -void DisksApp::printHelpMessage(ProgramOptionsDescription & command_option_description) +std::vector DisksApp::getCommandsToComplete(const String & command_prefix) const { - std::optional help_description = - createOptionsDescription("Help Message for clickhouse-disks", getTerminalWidth()); - - help_description->add(command_option_description); - - std::cout << "ClickHouse disk management tool\n"; - std::cout << "Usage: ./clickhouse-disks [OPTION]\n"; - std::cout << "clickhouse-disks\n\n"; - - for (const auto & current_command : supported_commands) - std::cout << command_descriptions[current_command]->command_name - << "\t" - << command_descriptions[current_command]->description - << "\n\n"; - - std::cout << command_option_description << '\n'; + std::vector answer{}; + for (const auto & [word, _] : command_descriptions) + { + if (word.starts_with(command_prefix)) + { + answer.push_back(word); + } + } + if (!answer.empty()) + { + std::sort(answer.begin(), answer.end()); + return answer; + } + for (const auto & [word, _] : aliases) + { + if (word.starts_with(command_prefix)) + { + answer.push_back(word); + } + } + if (!answer.empty()) + { + std::sort(answer.begin(), answer.end()); + return answer; + } + return {command_prefix}; } -String DisksApp::getDefaultConfigFileName() +std::vector DisksApp::getCompletions(const String & prefix) const { - return "/etc/clickhouse-server/config.xml"; + auto arguments = po::split_unix(prefix, word_break_characters); + if (arguments.empty()) + { + return {}; + } + if (word_break_characters.contains(prefix.back())) + { + CommandPtr command; + try + { + command = getCommandByName(arguments[0]); + } + catch (...) + { + return {arguments.back()}; + } + return getEmptyCompletion(command->command_name); + } + else if (arguments.size() == 1) + { + String command_prefix = arguments[0]; + return getCommandsToComplete(command_prefix); + } + else + { + String last_token = arguments.back(); + CommandPtr command; + try + { + command = getCommandByName(arguments[0]); + } + catch (...) + { + return {last_token}; + } + std::vector answer = {}; + if (command->command_name == "help") + { + return getCommandsToComplete(last_token); + } + else + { + answer = [&]() -> std::vector + { + if (multidisk_commands.contains(command->command_name)) + { + return client->getAllFilesByPatternFromAllDisks(last_token); + } + else + { + return client->getCurrentDiskWithPath().getAllFilesByPattern(last_token); + } + }(); + + for (const auto & disk_name : client->getAllDiskNames()) + { + if (disk_name.starts_with(last_token)) + { + answer.push_back(disk_name); + } + } + for (const auto & option : command->options_description.options()) + { + String option_sign = "--" + option->long_name(); + if (option_sign.starts_with(last_token)) + { + answer.push_back(option_sign); + } + } + } + if (!answer.empty()) + { + std::sort(answer.begin(), answer.end()); + return answer; + } + else + { + return {last_token}; + } + } } -void DisksApp::addOptions( - ProgramOptionsDescription & options_description_, - boost::program_options::positional_options_description & positional_options_description -) +bool DisksApp::processQueryText(const String & text) { - options_description_.add_options() - ("help,h", "Print common help message") - ("config-file,C", po::value(), "Set config file") - ("disk", po::value(), "Set disk name") - ("command_name", po::value(), "Name for command to do") - ("save-logs", "Save logs to a file") - ("log-level", po::value(), "Logging level") - ; + if (text.find_first_not_of(word_break_characters) == std::string::npos) + { + return true; + } + if (exit_strings.find(text) != exit_strings.end()) + return false; + CommandPtr command; + try + { + auto arguments = po::split_unix(text, word_break_characters); + command = getCommandByName(arguments[0]); + arguments.erase(arguments.begin()); + command->execute(arguments, *client); + } + catch (DB::Exception & err) + { + int code = getCurrentExceptionCode(); + if (code == ErrorCodes::LOGICAL_ERROR) + { + throw std::move(err); + } + else if (code == ErrorCodes::BAD_ARGUMENTS) + { + std::cerr << err.message() << "\n" + << "\n"; + if (command.get()) + { + std::cerr << "COMMAND: " << command->command_name << "\n"; + std::cerr << command->options_description << "\n"; + } + else + { + printAvailableCommandsHelpMessage(); + } + } + else + { + std::cerr << err.message() << "\n"; + } + } + catch (std::exception & err) + { + std::cerr << err.what() << "\n"; + } - positional_options_description.add("command_name", 1); + return true; +} - supported_commands = {"list-disks", "list", "move", "remove", "link", "copy", "write", "read", "mkdir"}; -#ifdef CLICKHOUSE_CLOUD - supported_commands.insert("packed-io"); -#endif +void DisksApp::runInteractiveReplxx() +{ + ReplxxLineReader lr( + suggest, + history_file, + /* multiline= */ false, + query_extenders, + query_delimiters, + word_break_characters.c_str(), + /* highlighter_= */ {}); + lr.enableBracketedPaste(); + + while (true) + { + DiskWithPath disk_with_path = client->getCurrentDiskWithPath(); + String prompt = "\x1b[1;34m" + disk_with_path.getDisk()->getName() + "\x1b[0m:" + "\x1b[1;31m" + disk_with_path.getCurrentPath() + + "\x1b[0m$ "; + + auto input = lr.readLine(prompt, "\x1b[1;31m:-] \x1b[0m"); + if (input.empty()) + break; + + if (!processQueryText(input)) + break; + } +} + +void DisksApp::parseAndCheckOptions( + const std::vector & arguments, const ProgramOptionsDescription & options_description, CommandLineOptions & options) +{ + auto parser = po::command_line_parser(arguments).options(options_description).allow_unregistered(); + po::parsed_options parsed = parser.run(); + po::store(parsed, options); +} + +void DisksApp::addOptions() +{ + options_description.add_options()("help,h", "Print common help message")("config-file,C", po::value(), "Set config file")( + "disk", po::value(), "Set disk name")("save-logs", "Save logs to a file")( + "log-level", po::value(), "Logging level")("query,q", po::value(), "Query for a non-interactive mode")( + "test-mode", "Interactive interface in test regyme"); command_descriptions.emplace("list-disks", makeCommandListDisks()); + command_descriptions.emplace("copy", makeCommandCopy()); command_descriptions.emplace("list", makeCommandList()); + command_descriptions.emplace("cd", makeCommandChangeDirectory()); command_descriptions.emplace("move", makeCommandMove()); command_descriptions.emplace("remove", makeCommandRemove()); command_descriptions.emplace("link", makeCommandLink()); - command_descriptions.emplace("copy", makeCommandCopy()); command_descriptions.emplace("write", makeCommandWrite()); command_descriptions.emplace("read", makeCommandRead()); command_descriptions.emplace("mkdir", makeCommandMkDir()); + command_descriptions.emplace("switch-disk", makeCommandSwitchDisk()); + command_descriptions.emplace("current_disk_with_path", makeCommandGetCurrentDiskAndPath()); + command_descriptions.emplace("touch", makeCommandTouch()); + command_descriptions.emplace("help", makeCommandHelp(*this)); #ifdef CLICKHOUSE_CLOUD command_descriptions.emplace("packed-io", makeCommandPackedIO()); #endif + for (const auto & [command_name, command_ptr] : command_descriptions) + { + if (command_name != command_ptr->command_name) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "Command name inside map doesn't coincide with actual command name"); + } + } } void DisksApp::processOptions() @@ -93,76 +319,122 @@ void DisksApp::processOptions() config().setBool("save-logs", true); if (options.count("log-level")) config().setString("log-level", options["log-level"].as()); + if (options.count("test-mode")) + config().setBool("test-mode", true); + if (options.count("query")) + query = std::optional{options["query"].as()}; } -DisksApp::~DisksApp() + +void DisksApp::printEntryHelpMessage() const { - if (global_context) - global_context->shutdown(); + std::cout << "\x1b[1;33m ClickHouse disk management tool \x1b[0m \n"; + std::cout << options_description << '\n'; } -void DisksApp::init(std::vector & common_arguments) + +void DisksApp::printAvailableCommandsHelpMessage() const { - stopOptionsProcessing(); + std::cout << "\x1b[1;32mAvailable commands:\x1b[0m\n"; + std::vector> commands_with_aliases_and_descrtiptions{}; + size_t maximal_command_length = 0; + for (const auto & [command_name, command_ptr] : command_descriptions) + { + std::string command_string = getCommandLineWithAliases(command_ptr); + maximal_command_length = std::max(maximal_command_length, command_string.size()); + commands_with_aliases_and_descrtiptions.push_back({std::move(command_string), command_descriptions.at(command_name)}); + } + for (const auto & [command_with_aliases, command_ptr] : commands_with_aliases_and_descrtiptions) + { + std::cout << "\x1b[1;33m" << command_with_aliases << "\x1b[0m" << std::string(5, ' ') << "\x1b[1;33m" << command_ptr->description + << "\x1b[0m \n"; + std::cout << command_ptr->options_description; + std::cout << std::endl; + } +} - ProgramOptionsDescription options_description{createOptionsDescription("clickhouse-disks", getTerminalWidth())}; +void DisksApp::printCommandHelpMessage(CommandPtr command) const +{ + String command_name_with_aliases = getCommandLineWithAliases(command); + std::cout << "\x1b[1;32m" << command_name_with_aliases << "\x1b[0m" << std::string(2, ' ') << command->description << "\n"; + std::cout << command->options_description; +} - po::positional_options_description positional_options_description; +void DisksApp::printCommandHelpMessage(String command_name) const +{ + printCommandHelpMessage(getCommandByName(command_name)); +} - addOptions(options_description, positional_options_description); +String DisksApp::getCommandLineWithAliases(CommandPtr command) const +{ + String command_string = command->command_name; + bool need_comma = false; + for (const auto & [alias_name, alias_command_name] : aliases) + { + if (alias_command_name == command->command_name) + { + if (std::exchange(need_comma, true)) + command_string += ","; + else + command_string += "("; + command_string += alias_name; + } + } + command_string += (need_comma ? ")" : ""); + return command_string; +} - size_t command_pos = findCommandPos(common_arguments); - std::vector global_flags(command_pos); - command_arguments.resize(common_arguments.size() - command_pos); - copy(common_arguments.begin(), common_arguments.begin() + command_pos, global_flags.begin()); - copy(common_arguments.begin() + command_pos, common_arguments.end(), command_arguments.begin()); +void DisksApp::initializeHistoryFile() +{ + String home_path; + const char * home_path_cstr = getenv("HOME"); // NOLINT(concurrency-mt-unsafe) + if (home_path_cstr) + home_path = home_path_cstr; + if (config().has("history-file")) + history_file = config().getString("history-file"); + else + history_file = home_path + "/.disks-file-history"; - parseAndCheckOptions(options_description, positional_options_description, global_flags); + if (!history_file.empty() && !fs::exists(history_file)) + { + try + { + FS::createFile(history_file); + } + catch (const ErrnoException & e) + { + if (e.getErrno() != EEXIST) + throw; + } + } +} + +void DisksApp::init(const std::vector & common_arguments) +{ + addOptions(); + parseAndCheckOptions(common_arguments, options_description, options); po::notify(options); if (options.count("help")) { - printHelpMessage(options_description); + printEntryHelpMessage(); + printAvailableCommandsHelpMessage(); exit(0); // NOLINT(concurrency-mt-unsafe) } - if (!supported_commands.contains(command_name)) - { - std::cerr << "Unknown command name: " << command_name << "\n"; - printHelpMessage(options_description); - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Bad Arguments"); - } - processOptions(); } -void DisksApp::parseAndCheckOptions( - ProgramOptionsDescription & options_description_, - boost::program_options::positional_options_description & positional_options_description, - std::vector & arguments) +String DisksApp::getDefaultConfigFileName() { - auto parser = po::command_line_parser(arguments) - .options(options_description_) - .positional(positional_options_description) - .allow_unregistered(); - - po::parsed_options parsed = parser.run(); - po::store(parsed, options); - - auto positional_arguments = po::collect_unrecognized(parsed.options, po::collect_unrecognized_mode::include_positional); - for (const auto & arg : positional_arguments) - { - if (command_descriptions.contains(arg)) - { - command_name = arg; - break; - } - } + return "/etc/clickhouse-server/config.xml"; } int DisksApp::main(const std::vector & /*args*/) { + std::vector keys; + config().keys(keys); if (config().has("config-file") || fs::exists(getDefaultConfigFileName())) { String config_path = config().getString("config-file", getDefaultConfigFileName()); @@ -173,9 +445,13 @@ int DisksApp::main(const std::vector & /*args*/) } else { + printEntryHelpMessage(); throw Exception(ErrorCodes::BAD_ARGUMENTS, "No config-file specified"); } + config().keys(keys); + initializeHistoryFile(); + if (config().has("save-logs")) { auto log_level = config().getString("log-level", "trace"); @@ -200,61 +476,68 @@ int DisksApp::main(const std::vector & /*args*/) global_context->setApplicationType(Context::ApplicationType::DISKS); String path = config().getString("path", DBMS_DEFAULT_PATH); + global_context->setPath(path); - auto & command = command_descriptions[command_name]; + String main_disk = config().getString("disk", "default"); - auto command_options = command->getCommandOptions(); - std::vector args; - if (command_options) + auto validator = [](const Poco::Util::AbstractConfiguration &, const std::string &, const std::string &) { return true; }; + + constexpr auto config_prefix = "storage_configuration.disks"; + auto disk_selector = std::make_shared(std::unordered_set{"cache", "encrypted"}); + disk_selector->initialize(config(), config_prefix, global_context, validator); + + std::vector>> disks_with_path; + + for (const auto & [_, disk_ptr] : disk_selector->getDisksMap()) { - auto parser = po::command_line_parser(command_arguments).options(*command_options).allow_unregistered(); - po::parsed_options parsed = parser.run(); - po::store(parsed, options); - po::notify(options); + disks_with_path.emplace_back( + disk_ptr, (disk_ptr->getName() == "local") ? std::optional{fs::current_path().string()} : std::nullopt); + } - args = po::collect_unrecognized(parsed.options, po::collect_unrecognized_mode::include_positional); - command->processOptions(config(), options); + + client = std::make_unique(std::move(disks_with_path), main_disk); + + suggest.setCompletionsCallback([&](const String & prefix, size_t /* prefix_length */) { return getCompletions(prefix); }); + + if (!query.has_value()) + { + runInteractive(); } else { - auto parser = po::command_line_parser(command_arguments).options({}).allow_unregistered(); - po::parsed_options parsed = parser.run(); - args = po::collect_unrecognized(parsed.options, po::collect_unrecognized_mode::include_positional); + processQueryText(query.value()); } - std::unordered_set disks - { - config().getString("disk", "default"), - config().getString("disk-from", config().getString("disk", "default")), - config().getString("disk-to", config().getString("disk", "default")), - }; - - auto validator = [&disks]( - const Poco::Util::AbstractConfiguration & config, - const std::string & disk_config_prefix, - const std::string & disk_name) - { - if (!disks.contains(disk_name)) - return false; - - const auto disk_type = config.getString(disk_config_prefix + ".type", "local"); - - if (disk_type == "cache") - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Disk type 'cache' of disk {} is not supported by clickhouse-disks", disk_name); - - return true; - }; - - constexpr auto config_prefix = "storage_configuration.disks"; - auto disk_selector = std::make_shared(); - disk_selector->initialize(config(), config_prefix, global_context, validator); - - command->execute(args, disk_selector, config()); - return Application::EXIT_OK; } +DisksApp::~DisksApp() +{ + client.reset(nullptr); + if (global_context) + global_context->shutdown(); +} + +void DisksApp::runInteractiveTestMode() +{ + for (String input; std::getline(std::cin, input);) + { + if (!processQueryText(input)) + break; + + std::cout << "\a\a\a\a" << std::endl; + std::cerr << std::flush; + } +} + +void DisksApp::runInteractive() +{ + if (config().hasOption("test-mode")) + runInteractiveTestMode(); + else + runInteractiveReplxx(); +} } int mainEntryClickHouseDisks(int argc, char ** argv) @@ -269,16 +552,16 @@ int mainEntryClickHouseDisks(int argc, char ** argv) catch (const DB::Exception & e) { std::cerr << DB::getExceptionMessage(e, false) << std::endl; - return 1; + return 0; } catch (const boost::program_options::error & e) { std::cerr << "Bad arguments: " << e.what() << std::endl; - return DB::ErrorCodes::BAD_ARGUMENTS; + return 0; } catch (...) { std::cerr << DB::getCurrentExceptionMessage(true) << std::endl; - return 1; + return 0; } } diff --git a/programs/disks/DisksApp.h b/programs/disks/DisksApp.h index 51bc3f58dc4..f8167884c62 100644 --- a/programs/disks/DisksApp.h +++ b/programs/disks/DisksApp.h @@ -1,61 +1,107 @@ #pragma once +#include +#include +#include #include +#include "DisksClient.h" +#include "ICommand_fwd.h" #include +#include +#include #include -#include - namespace DB { -class ICommand; -using CommandPtr = std::unique_ptr; - -namespace po = boost::program_options; using ProgramOptionsDescription = boost::program_options::options_description; using CommandLineOptions = boost::program_options::variables_map; -class DisksApp : public Poco::Util::Application, public Loggers +class DisksApp : public Poco::Util::Application { public: - DisksApp() = default; - ~DisksApp() override; + void addOptions(); - void init(std::vector & common_arguments); - - int main(const std::vector & args) override; - -protected: - static String getDefaultConfigFileName(); - - void addOptions( - ProgramOptionsDescription & options_description, - boost::program_options::positional_options_description & positional_options_description); void processOptions(); - void printHelpMessage(ProgramOptionsDescription & command_option_description); + bool processQueryText(const String & text); - size_t findCommandPos(std::vector & common_arguments); + void init(const std::vector & common_arguments); + + int main(const std::vector & /*args*/) override; + + CommandPtr getCommandByName(const String & command) const; + + void initializeHistoryFile(); + + static void parseAndCheckOptions( + const std::vector & arguments, const ProgramOptionsDescription & options_description, CommandLineOptions & options); + + void printEntryHelpMessage() const; + void printAvailableCommandsHelpMessage() const; + void printCommandHelpMessage(String command_name) const; + void printCommandHelpMessage(CommandPtr command) const; + String getCommandLineWithAliases(CommandPtr command) const; + + + std::vector getCompletions(const String & prefix) const; + + std::vector getEmptyCompletion(String command_name) const; + + ~DisksApp() override; private: - void parseAndCheckOptions( - ProgramOptionsDescription & options_description, - boost::program_options::positional_options_description & positional_options_description, - std::vector & arguments); + void runInteractive(); + void runInteractiveReplxx(); + void runInteractiveTestMode(); + + String getDefaultConfigFileName(); + + std::vector getCommandsToComplete(const String & command_prefix) const; + + // Fields responsible for the REPL work + String history_file; + LineReader::Suggest suggest; + static LineReader::Patterns query_extenders; + static LineReader::Patterns query_delimiters; + static String word_break_characters; + + // General command line arguments parsing fields -protected: - ContextMutablePtr global_context; SharedContextHolder shared_context; - - String command_name; - std::vector command_arguments; - - std::unordered_set supported_commands; + ContextMutablePtr global_context; + ProgramOptionsDescription options_description; + CommandLineOptions options; std::unordered_map command_descriptions; - po::variables_map options; -}; + std::optional query; + const std::unordered_map aliases + = {{"cp", "copy"}, + {"mv", "move"}, + {"ls", "list"}, + {"list_disks", "list-disks"}, + {"ln", "link"}, + {"rm", "remove"}, + {"cat", "read"}, + {"r", "read"}, + {"w", "write"}, + {"create", "touch"}, + {"delete", "remove"}, + {"ls-disks", "list-disks"}, + {"ls_disks", "list-disks"}, + {"packed_io", "packed-io"}, + {"change-dir", "cd"}, + {"change_dir", "cd"}, + {"switch_disk", "switch-disk"}, + {"current", "current_disk_with_path"}, + {"current_disk", "current_disk_with_path"}, + {"current_path", "current_disk_with_path"}, + {"cur", "current_disk_with_path"}}; + + std::set multidisk_commands = {"copy", "packed-io", "switch-disk", "cd"}; + + std::unique_ptr client{}; +}; } diff --git a/programs/disks/DisksClient.cpp b/programs/disks/DisksClient.cpp new file mode 100644 index 00000000000..7e36c7911ab --- /dev/null +++ b/programs/disks/DisksClient.cpp @@ -0,0 +1,263 @@ +#include "DisksClient.h" +#include +#include +#include +#include + +#include + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; + extern const int LOGICAL_ERROR; +}; + +namespace DB +{ +DiskWithPath::DiskWithPath(DiskPtr disk_, std::optional path_) : disk(disk_) +{ + if (path_.has_value()) + { + if (!fs::path{path_.value()}.is_absolute()) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Initializing path {} is not absolute", path_.value()); + } + path = path_.value(); + } + else + { + path = String{"/"}; + } + + String relative_path = normalizePathAndGetAsRelative(path); + if (disk->isDirectory(relative_path) || (relative_path.empty() && (disk->isDirectory("/")))) + { + return; + } + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Initializing path {} (normalized path: {}) at disk {} is not a directory", + path, + relative_path, + disk->getName()); +} + +std::vector DiskWithPath::listAllFilesByPath(const String & any_path) const +{ + if (isDirectory(any_path)) + { + std::vector file_names; + disk->listFiles(getRelativeFromRoot(any_path), file_names); + return file_names; + } + else + { + return {}; + } +} + +std::vector DiskWithPath::getAllFilesByPattern(const String & pattern) const +{ + auto [path_before, path_after] = [&]() -> std::pair + { + auto slash_pos = pattern.find_last_of('/'); + if (slash_pos >= pattern.size()) + { + return {"", pattern}; + } + else + { + return {pattern.substr(0, slash_pos + 1), pattern.substr(slash_pos + 1, pattern.size() - slash_pos - 1)}; + } + }(); + + if (!isDirectory(path_before)) + { + return {}; + } + else + { + std::vector file_names = listAllFilesByPath(path_before); + + std::vector answer; + + for (const auto & file_name : file_names) + { + if (file_name.starts_with(path_after)) + { + String file_pattern = path_before + file_name; + if (isDirectory(file_pattern)) + { + file_pattern = file_pattern + "/"; + } + answer.push_back(file_pattern); + } + } + return answer; + } +}; + +void DiskWithPath::setPath(const String & any_path) +{ + if (isDirectory(any_path)) + { + path = getAbsolutePath(any_path); + } + else + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Path {} at disk {} is not a directory", any_path, disk->getName()); + } +} + +String DiskWithPath::validatePathAndGetAsRelative(const String & path) +{ + String lexically_normal_path = fs::path(path).lexically_normal(); + if (lexically_normal_path.find("..") != std::string::npos) + throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Path {} is not normalized", path); + + /// If path is absolute we should keep it as relative inside disk, so disk will look like + /// an ordinary filesystem with root. + if (fs::path(lexically_normal_path).is_absolute()) + return lexically_normal_path.substr(1); + + return lexically_normal_path; +} + +String DiskWithPath::normalizePathAndGetAsRelative(const String & messyPath) +{ + std::filesystem::path path(messyPath); + std::filesystem::path canonical_path = std::filesystem::weakly_canonical(path); + String npath = canonical_path.make_preferred().string(); + return validatePathAndGetAsRelative(npath); +} + +String DiskWithPath::normalizePath(const String & path) +{ + std::filesystem::path canonical_path = std::filesystem::weakly_canonical(path); + return canonical_path.make_preferred().string(); +} + +DisksClient::DisksClient(std::vector>> && disks_with_paths, std::optional begin_disk) +{ + if (disks_with_paths.empty()) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Initializing array of disks is empty"); + } + if (!begin_disk.has_value()) + { + begin_disk = disks_with_paths[0].first->getName(); + } + bool has_begin_disk = false; + for (auto & [disk, path] : disks_with_paths) + { + addDisk(disk, path); + if (disk->getName() == begin_disk.value()) + { + has_begin_disk = true; + } + } + if (!has_begin_disk) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "There is no begin_disk '{}' in initializing array", begin_disk.value()); + } + current_disk = std::move(begin_disk.value()); +} + +const DiskWithPath & DisksClient::getDiskWithPath(const String & disk) const +{ + try + { + return disks.at(disk); + } + catch (...) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "The disk '{}' is unknown", disk); + } +} + +DiskWithPath & DisksClient::getDiskWithPath(const String & disk) +{ + try + { + return disks.at(disk); + } + catch (...) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "The disk '{}' is unknown", disk); + } +} + +const DiskWithPath & DisksClient::getCurrentDiskWithPath() const +{ + try + { + return disks.at(current_disk); + } + catch (...) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "There is no current disk in client"); + } +} + +DiskWithPath & DisksClient::getCurrentDiskWithPath() +{ + try + { + return disks.at(current_disk); + } + catch (...) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "There is no current disk in client"); + } +} + +void DisksClient::switchToDisk(const String & disk_, const std::optional & path_) +{ + if (disks.contains(disk_)) + { + if (path_.has_value()) + { + disks.at(disk_).setPath(path_.value()); + } + current_disk = disk_; + } + else + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "The disk '{}' is unknown", disk_); + } +} + +std::vector DisksClient::getAllDiskNames() const +{ + std::vector answer{}; + answer.reserve(disks.size()); + for (const auto & [disk_name, _] : disks) + { + answer.push_back(disk_name); + } + return answer; +} + +std::vector DisksClient::getAllFilesByPatternFromAllDisks(const String & pattern) const +{ + std::vector answer{}; + for (const auto & [_, disk] : disks) + { + for (auto & word : disk.getAllFilesByPattern(pattern)) + { + answer.push_back(word); + } + } + return answer; +} + +void DisksClient::addDisk(DiskPtr disk_, const std::optional & path_) +{ + String disk_name = disk_->getName(); + if (disks.contains(disk_->getName())) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "The disk '{}' already exists", disk_name); + } + disks.emplace(disk_name, DiskWithPath{disk_, path_}); +} +} diff --git a/programs/disks/DisksClient.h b/programs/disks/DisksClient.h new file mode 100644 index 00000000000..8a55d22af93 --- /dev/null +++ b/programs/disks/DisksClient.h @@ -0,0 +1,89 @@ +#pragma once + +#include +#include +#include +#include +#include +#include +#include +#include "Disks/IDisk.h" + +#include +#include +#include + +namespace fs = std::filesystem; + +namespace DB +{ + +std::vector split(const String & text, const String & delimiters); + +using ProgramOptionsDescription = boost::program_options::options_description; +using CommandLineOptions = boost::program_options::variables_map; + +class DiskWithPath +{ +public: + explicit DiskWithPath(DiskPtr disk_, std::optional path_ = std::nullopt); + + String getAbsolutePath(const String & any_path) const { return normalizePath(fs::path(path) / any_path); } + + String getCurrentPath() const { return path; } + + bool isDirectory(const String & any_path) const + { + return disk->isDirectory(getRelativeFromRoot(any_path)) || (getRelativeFromRoot(any_path).empty() && (disk->isDirectory("/"))); + } + + std::vector listAllFilesByPath(const String & any_path) const; + + std::vector getAllFilesByPattern(const String & pattern) const; + + DiskPtr getDisk() const { return disk; } + + void setPath(const String & any_path); + + String getRelativeFromRoot(const String & any_path) const { return normalizePathAndGetAsRelative(getAbsolutePath(any_path)); } + +private: + static String validatePathAndGetAsRelative(const String & path); + static std::string normalizePathAndGetAsRelative(const std::string & messyPath); + static std::string normalizePath(const std::string & messyPath); + + const DiskPtr disk; + String path; +}; + +class DisksClient +{ +public: + explicit DisksClient(std::vector>> && disks_with_paths, std::optional begin_disk); + + const DiskWithPath & getDiskWithPath(const String & disk) const; + + DiskWithPath & getDiskWithPath(const String & disk); + + const DiskWithPath & getCurrentDiskWithPath() const; + + DiskWithPath & getCurrentDiskWithPath(); + + DiskPtr getCurrentDisk() const { return getCurrentDiskWithPath().getDisk(); } + + DiskPtr getDisk(const String & disk) const { return getDiskWithPath(disk).getDisk(); } + + void switchToDisk(const String & disk_, const std::optional & path_); + + std::vector getAllDiskNames() const; + + std::vector getAllFilesByPatternFromAllDisks(const String & pattern) const; + + +private: + void addDisk(DiskPtr disk_, const std::optional & path_); + + String current_disk; + std::unordered_map disks; +}; +} diff --git a/programs/disks/ICommand.cpp b/programs/disks/ICommand.cpp index 86188fb6db1..f622bcad3c6 100644 --- a/programs/disks/ICommand.cpp +++ b/programs/disks/ICommand.cpp @@ -1,5 +1,5 @@ #include "ICommand.h" -#include +#include "DisksClient.h" namespace DB @@ -10,43 +10,42 @@ namespace ErrorCodes extern const int BAD_ARGUMENTS; } -void ICommand::printHelpMessage() const +CommandLineOptions ICommand::processCommandLineArguments(const Strings & commands) { - std::cout << "Command: " << command_name << '\n'; - std::cout << "Description: " << description << '\n'; - std::cout << "Usage: " << usage << '\n'; + CommandLineOptions options; + auto parser = po::command_line_parser(commands); + parser.options(options_description).positional(positional_options_description); - if (command_option_description) + po::parsed_options parsed = parser.run(); + po::store(parsed, options); + + return options; +} + +void ICommand::execute(const Strings & commands, DisksClient & client) +{ + try { - auto options = *command_option_description; - if (!options.options().empty()) - std::cout << options << '\n'; + processCommandLineArguments(commands); + } + catch (std::exception & exc) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "{}", exc.what()); + } + executeImpl(processCommandLineArguments(commands), client); +} + +DiskWithPath & ICommand::getDiskWithPath(DisksClient & client, const CommandLineOptions & options, const String & name) +{ + auto disk_name = getValueFromCommandLineOptionsWithOptional(options, name); + if (disk_name.has_value()) + { + return client.getDiskWithPath(disk_name.value()); + } + else + { + return client.getCurrentDiskWithPath(); } } -void ICommand::addOptions(ProgramOptionsDescription & options_description) -{ - if (!command_option_description || command_option_description->options().empty()) - return; - - options_description.add(*command_option_description); -} - -String ICommand::validatePathAndGetAsRelative(const String & path) -{ - /// If path contain non-normalized symbols like . we will normalized them. If the resulting normalized path - /// still contain '..' it can be dangerous, disallow such paths. Also since clickhouse-disks - /// is not an interactive program (don't track you current path) it's OK to disallow .. paths. - String lexically_normal_path = fs::path(path).lexically_normal(); - if (lexically_normal_path.find("..") != std::string::npos) - throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Path {} is not normalized", path); - - /// If path is absolute we should keep it as relative inside disk, so disk will look like - /// an ordinary filesystem with root. - if (fs::path(lexically_normal_path).is_absolute()) - return lexically_normal_path.substr(1); - - return lexically_normal_path; -} - } diff --git a/programs/disks/ICommand.h b/programs/disks/ICommand.h index efe350fe87b..6faf90e2b52 100644 --- a/programs/disks/ICommand.h +++ b/programs/disks/ICommand.h @@ -1,66 +1,146 @@ #pragma once -#include +#include #include +#include +#include #include -#include #include +#include "Common/Exception.h" +#include -#include +#include + +#include "DisksApp.h" + +#include "DisksClient.h" + +#include "ICommand_fwd.h" namespace DB { namespace po = boost::program_options; -using ProgramOptionsDescription = boost::program_options::options_description; -using CommandLineOptions = boost::program_options::variables_map; +using ProgramOptionsDescription = po::options_description; +using PositionalProgramOptionsDescription = po::positional_options_description; +using CommandLineOptions = po::variables_map; + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; +} class ICommand { public: - ICommand() = default; + explicit ICommand() = default; virtual ~ICommand() = default; - virtual void execute( - const std::vector & command_arguments, - std::shared_ptr & disk_selector, - Poco::Util::LayeredConfiguration & config) = 0; + void execute(const Strings & commands, DisksClient & client); - const std::optional & getCommandOptions() const { return command_option_description; } + virtual void executeImpl(const CommandLineOptions & options, DisksClient & client) = 0; - void addOptions(ProgramOptionsDescription & options_description); - - virtual void processOptions(Poco::Util::LayeredConfiguration & config, po::variables_map & options) const = 0; + CommandLineOptions processCommandLineArguments(const Strings & commands); protected: - void printHelpMessage() const; + template + static T getValueFromCommandLineOptions(const CommandLineOptions & options, const String & name) + { + try + { + return options[name].as(); + } + catch (boost::bad_any_cast &) + { + throw DB::Exception(ErrorCodes::BAD_ARGUMENTS, "Argument '{}' has wrong type and can't be parsed", name); + } + } + + template + static T getValueFromCommandLineOptionsThrow(const CommandLineOptions & options, const String & name) + { + if (options.count(name)) + { + return getValueFromCommandLineOptions(options, name); + } + else + { + throw DB::Exception(ErrorCodes::BAD_ARGUMENTS, "Mandatory argument '{}' is missing", name); + } + } + + template + static T getValueFromCommandLineOptionsWithDefault(const CommandLineOptions & options, const String & name, const T & default_value) + { + if (options.count(name)) + { + return getValueFromCommandLineOptions(options, name); + } + else + { + return default_value; + } + } + + template + static std::optional getValueFromCommandLineOptionsWithOptional(const CommandLineOptions & options, const String & name) + { + if (options.count(name)) + { + return std::optional{getValueFromCommandLineOptions(options, name)}; + } + else + { + return std::nullopt; + } + } + + DiskWithPath & getDiskWithPath(DisksClient & client, const CommandLineOptions & options, const String & name); + + String getTargetLocation(const String & path_from, DiskWithPath & disk_to, const String & path_to) + { + if (!disk_to.getDisk()->isDirectory(path_to)) + { + return path_to; + } + String copied_path_from = path_from; + if (copied_path_from.ends_with('/')) + { + copied_path_from.pop_back(); + } + String plain_filename = fs::path(copied_path_from).filename(); + + return fs::path{path_to} / plain_filename; + } - static String validatePathAndGetAsRelative(const String & path); public: String command_name; String description; + ProgramOptionsDescription options_description; protected: - std::optional command_option_description; - String usage; - po::positional_options_description positional_options_description; + PositionalProgramOptionsDescription positional_options_description; }; -using CommandPtr = std::unique_ptr; - -} - DB::CommandPtr makeCommandCopy(); -DB::CommandPtr makeCommandLink(); -DB::CommandPtr makeCommandList(); DB::CommandPtr makeCommandListDisks(); +DB::CommandPtr makeCommandList(); +DB::CommandPtr makeCommandChangeDirectory(); +DB::CommandPtr makeCommandLink(); DB::CommandPtr makeCommandMove(); DB::CommandPtr makeCommandRead(); DB::CommandPtr makeCommandRemove(); DB::CommandPtr makeCommandWrite(); DB::CommandPtr makeCommandMkDir(); +DB::CommandPtr makeCommandSwitchDisk(); +DB::CommandPtr makeCommandGetCurrentDiskAndPath(); +DB::CommandPtr makeCommandHelp(const DisksApp & disks_app); +DB::CommandPtr makeCommandTouch(); +#ifdef CLICKHOUSE_CLOUD DB::CommandPtr makeCommandPackedIO(); +#endif +} diff --git a/programs/disks/ICommand_fwd.h b/programs/disks/ICommand_fwd.h new file mode 100644 index 00000000000..84310b4a18d --- /dev/null +++ b/programs/disks/ICommand_fwd.h @@ -0,0 +1,10 @@ +#pragma once + +#include + +namespace DB +{ +class ICommand; + +using CommandPtr = std::shared_ptr; +} diff --git a/programs/keeper-client/KeeperClient.cpp b/programs/keeper-client/KeeperClient.cpp index 68adc2c2aac..a20c1f686f3 100644 --- a/programs/keeper-client/KeeperClient.cpp +++ b/programs/keeper-client/KeeperClient.cpp @@ -383,6 +383,9 @@ int KeeperClient::main(const std::vector & /* args */) for (const auto & key : keys) { + if (key != "node") + continue; + String prefix = "zookeeper." + key; String host = clickhouse_config.configuration->getString(prefix + ".host"); String port = clickhouse_config.configuration->getString(prefix + ".port"); @@ -401,6 +404,7 @@ int KeeperClient::main(const std::vector & /* args */) zk_args.hosts.push_back(host + ":" + port); } + zk_args.availability_zones.resize(zk_args.hosts.size()); zk_args.connection_timeout_ms = config().getInt("connection-timeout", 10) * 1000; zk_args.session_timeout_ms = config().getInt("session-timeout", 10) * 1000; zk_args.operation_timeout_ms = config().getInt("operation-timeout", 10) * 1000; diff --git a/programs/keeper-converter/KeeperConverter.cpp b/programs/keeper-converter/KeeperConverter.cpp index 639e0957f49..5f2787f8930 100644 --- a/programs/keeper-converter/KeeperConverter.cpp +++ b/programs/keeper-converter/KeeperConverter.cpp @@ -45,16 +45,17 @@ int mainEntryClickHouseKeeperConverter(int argc, char ** argv) keeper_context->setDigestEnabled(true); keeper_context->setSnapshotDisk(std::make_shared("Keeper-snapshots", options["output-dir"].as())); - DB::KeeperStorage storage(/* tick_time_ms */ 500, /* superdigest */ "", keeper_context, /* initialize_system_nodes */ false); + /// TODO(hanfei): support rocksdb here + DB::KeeperMemoryStorage storage(/* tick_time_ms */ 500, /* superdigest */ "", keeper_context, /* initialize_system_nodes */ false); DB::deserializeKeeperStorageFromSnapshotsDir(storage, options["zookeeper-snapshots-dir"].as(), logger); storage.initializeSystemNodes(); DB::deserializeLogsAndApplyToStorage(storage, options["zookeeper-logs-dir"].as(), logger); DB::SnapshotMetadataPtr snapshot_meta = std::make_shared(storage.getZXID(), 1, std::make_shared()); - DB::KeeperStorageSnapshot snapshot(&storage, snapshot_meta); + DB::KeeperStorageSnapshot snapshot(&storage, snapshot_meta); - DB::KeeperSnapshotManager manager(1, keeper_context); + DB::KeeperSnapshotManager manager(1, keeper_context); auto snp = manager.serializeSnapshotToBuffer(snapshot); auto file_info = manager.serializeSnapshotBufferToDisk(*snp, storage.getZXID()); std::cout << "Snapshot serialized to path:" << fs::path(file_info->disk->getPath()) / file_info->path << std::endl; diff --git a/programs/keeper/CMakeLists.txt b/programs/keeper/CMakeLists.txt index 079951be55e..9b931c49c24 100644 --- a/programs/keeper/CMakeLists.txt +++ b/programs/keeper/CMakeLists.txt @@ -1,4 +1,5 @@ set(CLICKHOUSE_KEEPER_SOURCES + keeper_main.cpp Keeper.cpp ) @@ -8,6 +9,9 @@ set (CLICKHOUSE_KEEPER_LINK clickhouse_common_io clickhouse_common_zookeeper daemon + clickhouse-keeper-converter-lib + clickhouse-keeper-client-lib + clickhouse_functions dbms ) @@ -17,199 +21,11 @@ install(FILES keeper_config.xml DESTINATION "${CLICKHOUSE_ETC_DIR}/clickhouse-ke if (BUILD_STANDALONE_KEEPER) # Straight list of all required sources - set(CLICKHOUSE_KEEPER_STANDALONE_SOURCES - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperReconfiguration.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/RaftServerConfig.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/ACLMap.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/Changelog.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/CoordinationSettings.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/FourLetterCommand.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/InMemoryLogStore.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperConnectionStats.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperDispatcher.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperLogStore.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperServer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperContext.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperFeatureFlags.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperSnapshotManager.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperSnapshotManagerS3.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperStateMachine.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperContext.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperStateManager.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperStorage.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperConstants.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperAsynchronousMetrics.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/KeeperCommon.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/SessionExpiryQueue.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/SummingStateMachine.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/WriteBufferFromNuraftBuffer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/ZooKeeperDataReader.cpp + clickhouse_add_executable(clickhouse-keeper ${CLICKHOUSE_KEEPER_SOURCES}) - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Core/SettingsFields.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Core/BaseSettings.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Core/ServerSettings.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Core/Field.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Core/SettingsEnums.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Core/ServerUUID.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Core/UUID.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Core/BackgroundSchedulePool.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/IO/ReadBuffer.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTPPathHints.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/KeeperTCPHandler.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/TCPServer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/NotFoundHandler.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/ProtocolServerAdapter.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/CertificateReloader.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/PrometheusRequestHandler.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/PrometheusMetricsWriter.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/waitServersToFinish.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/ServerType.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTPRequestHandlerFactoryMain.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/KeeperReadinessHandler.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/CloudPlacementInfo.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTP/HTTPServer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTP/ReadHeaders.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTP/HTTPServerConnection.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTP/HTTPServerRequest.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTP/HTTPServerResponse.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTP/HTTPServerConnectionFactory.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Server/HTTP/WriteBufferFromHTTPServerResponse.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CachedCompressedReadBuffer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CheckingCompressedReadBuffer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressedReadBufferBase.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressedReadBuffer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressedReadBufferFromFile.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressedWriteBuffer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecEncrypted.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecLZ4.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecMultiple.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecNone.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecZSTD.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionFactory.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/ICompressionCodec.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/LZ4_decompress_faster.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/CurrentThread.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/NamedCollections/NamedCollections.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/NamedCollections/NamedCollectionConfiguration.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/Jemalloc.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/IKeeper.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/TestKeeper.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/ZooKeeperCommon.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/ZooKeeperConstants.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/ZooKeeper.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/ZooKeeperImpl.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/ZooKeeperIO.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/ZooKeeperLock.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Common/ZooKeeper/ZooKeeperNodeCache.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/registerDisks.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IDisk.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/DiskFactory.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/DiskSelector.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/DiskLocal.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/DiskLocalCheckThread.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/LocalDirectorySyncGuard.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/TemporaryFileOnDisk.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/loadLocalDiskConfig.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/DiskType.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/IObjectStorage.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/MetadataOperationsHolder.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorageOperations.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/MetadataStorageFromPlainRewritableObjectStorage.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/MetadataStorageFromDisk.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/MetadataStorageTransactionState.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/MetadataStorageFromDiskTransactionOperations.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/DiskObjectStorage.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/DiskObjectStorageRemoteMetadataRestoreHelper.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/ObjectStorageIteratorAsync.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/ObjectStorageIterator.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/StoredObject.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/S3/S3Capabilities.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/S3/diskSettings.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/S3/DiskS3Utils.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/CommonPathPrefixKeyGenerator.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/ObjectStorageFactory.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/MetadataStorageFactory.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/ObjectStorages/RegisterDiskObjectStorage.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/createReadBufferFromFileBase.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/ReadBufferFromRemoteFSGather.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/IOUringReader.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/getIOUringReader.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/WriteBufferFromTemporaryFile.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/WriteBufferWithFinalizeCallback.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/AsynchronousBoundedReadBuffer.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/getThreadPoolReader.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/ThreadPoolRemoteFSReader.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Disks/IO/ThreadPoolReader.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Daemon/BaseDaemon.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Daemon/SentryWriter.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Daemon/GraphiteWriter.cpp - ${CMAKE_CURRENT_BINARY_DIR}/../../src/Daemon/GitHash.generated.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/Standalone/Context.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/Standalone/Settings.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Coordination/Standalone/ThreadStatusExt.cpp - - Keeper.cpp - clickhouse-keeper.cpp - ) - - # List of resources for clickhouse-keeper client - if (ENABLE_CLICKHOUSE_KEEPER_CLIENT) - list(APPEND CLICKHOUSE_KEEPER_STANDALONE_SOURCES - ${CMAKE_CURRENT_SOURCE_DIR}/../../programs/keeper-client/KeeperClient.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../programs/keeper-client/Commands.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../programs/keeper-client/Parser.cpp - - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Client/LineReader.cpp - ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Client/ReplxxLineReader.cpp - ) - endif() - - clickhouse_add_executable(clickhouse-keeper ${CLICKHOUSE_KEEPER_STANDALONE_SOURCES}) - - # Remove some redundant dependencies - target_compile_definitions (clickhouse-keeper PRIVATE -DCLICKHOUSE_KEEPER_STANDALONE_BUILD) - target_compile_definitions (clickhouse-keeper PUBLIC -DWITHOUT_TEXT_LOG) - - if (ENABLE_CLICKHOUSE_KEEPER_CLIENT AND TARGET ch_rust::skim) - target_link_libraries(clickhouse-keeper PRIVATE ch_rust::skim) - endif() - - target_link_libraries(clickhouse-keeper - PRIVATE - ch_contrib::abseil_swiss_tables - ch_contrib::nuraft - ch_contrib::lz4 - ch_contrib::zstd - ch_contrib::cityhash - ch_contrib::jemalloc - common ch_contrib::double_conversion - ch_contrib::dragonbox_to_chars - pcg_random - ch_contrib::pdqsort - ch_contrib::miniselect - clickhouse_common_config_no_zookeeper_log - loggers_no_text_log - clickhouse_common_io - clickhouse_parsers # Otherwise compression will not built. FIXME. - ) + target_link_libraries(clickhouse-keeper PUBLIC ${CLICKHOUSE_KEEPER_LINK}) set_target_properties(clickhouse-keeper PROPERTIES RUNTIME_OUTPUT_DIRECTORY ../) - if (SPLIT_DEBUG_SYMBOLS) clickhouse_split_debug_symbols(TARGET clickhouse-keeper DESTINATION_DIR ${CMAKE_CURRENT_BINARY_DIR}/../${SPLITTED_DEBUG_SYMBOLS_DIR} BINARY_PATH ../clickhouse-keeper) else() diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp index bb04ff88936..8cf1a4d1999 100644 --- a/programs/keeper/Keeper.cpp +++ b/programs/keeper/Keeper.cpp @@ -27,6 +27,8 @@ #include #include +#include + #include #include @@ -50,6 +52,10 @@ # include #endif +#if USE_GWP_ASAN +# include +#endif + #include #include @@ -75,16 +81,6 @@ int mainEntryClickHouseKeeper(int argc, char ** argv) } } -#ifdef CLICKHOUSE_KEEPER_STANDALONE_BUILD - -// Weak symbols don't work correctly on Darwin -// so we have a stub implementation to avoid linker errors -void collectCrashLog( - Int32, UInt64, const String &, const StackTrace &) -{} - -#endif - namespace DB { @@ -272,11 +268,43 @@ HTTPContextPtr httpContext() return std::make_shared(Context::getGlobalContextInstance()); } +String getKeeperPath(Poco::Util::LayeredConfiguration & config) +{ + String path; + if (config.has("keeper_server.storage_path")) + { + path = config.getString("keeper_server.storage_path"); + } + else if (config.has("keeper_server.log_storage_path")) + { + path = std::filesystem::path(config.getString("keeper_server.log_storage_path")).parent_path(); + } + else if (config.has("keeper_server.snapshot_storage_path")) + { + path = std::filesystem::path(config.getString("keeper_server.snapshot_storage_path")).parent_path(); + } + else if (std::filesystem::is_directory(std::filesystem::path{config.getString("path", DBMS_DEFAULT_PATH)} / "coordination")) + { + throw Exception(ErrorCodes::NO_ELEMENTS_IN_CONFIG, + "By default 'keeper_server.storage_path' could be assigned to {}, but the directory {} already exists. Please specify 'keeper_server.storage_path' in the keeper configuration explicitly", + KEEPER_DEFAULT_PATH, String{std::filesystem::path{config.getString("path", DBMS_DEFAULT_PATH)} / "coordination"}); + } + else + { + path = KEEPER_DEFAULT_PATH; + } + return path; +} + + } int Keeper::main(const std::vector & /*args*/) try { +#if USE_JEMALLOC + setJemallocBackgroundThreads(true); +#endif Poco::Logger * log = &logger(); UseSSL use_ssl; @@ -321,31 +349,7 @@ try updateMemorySoftLimitInConfig(config()); - std::string path; - - if (config().has("keeper_server.storage_path")) - { - path = config().getString("keeper_server.storage_path"); - } - else if (config().has("keeper_server.log_storage_path")) - { - path = std::filesystem::path(config().getString("keeper_server.log_storage_path")).parent_path(); - } - else if (config().has("keeper_server.snapshot_storage_path")) - { - path = std::filesystem::path(config().getString("keeper_server.snapshot_storage_path")).parent_path(); - } - else if (std::filesystem::is_directory(std::filesystem::path{config().getString("path", DBMS_DEFAULT_PATH)} / "coordination")) - { - throw Exception(ErrorCodes::NO_ELEMENTS_IN_CONFIG, - "By default 'keeper_server.storage_path' could be assigned to {}, but the directory {} already exists. Please specify 'keeper_server.storage_path' in the keeper configuration explicitly", - KEEPER_DEFAULT_PATH, String{std::filesystem::path{config().getString("path", DBMS_DEFAULT_PATH)} / "coordination"}); - } - else - { - path = KEEPER_DEFAULT_PATH; - } - + std::string path = getKeeperPath(config()); std::filesystem::create_directories(path); /// Check that the process user id matches the owner of the data. @@ -355,10 +359,7 @@ try std::string include_from_path = config().getString("include_from", "/etc/metrika.xml"); - if (config().has(DB::PlacementInfo::PLACEMENT_CONFIG_PREFIX)) - { - PlacementInfo::PlacementInfo::instance().initialize(config()); - } + PlacementInfo::PlacementInfo::instance().initialize(config()); GlobalThreadPool::initialize( /// We need to have sufficient amount of threads for connections + nuraft workers + keeper workers, 1000 is an estimation @@ -562,7 +563,7 @@ try auto main_config_reloader = std::make_unique( config_path, extra_paths, - config().getString("path", KEEPER_DEFAULT_PATH), + getKeeperPath(config()), std::move(unused_cache), unused_event, [&](ConfigurationPtr config, bool /* initial_loading */) @@ -642,6 +643,10 @@ try tryLogCurrentException(log, "Disabling cgroup memory observer because of an error during initialization"); } +#if USE_GWP_ASAN + GWPAsan::initFinished(); +#endif + LOG_INFO(log, "Ready for connections."); diff --git a/programs/keeper/clickhouse-keeper.cpp b/programs/keeper/clickhouse-keeper.cpp deleted file mode 100644 index f2f91930ac0..00000000000 --- a/programs/keeper/clickhouse-keeper.cpp +++ /dev/null @@ -1,30 +0,0 @@ -#include -#include "config_tools.h" - - -int mainEntryClickHouseKeeper(int argc, char ** argv); - -#if ENABLE_CLICKHOUSE_KEEPER_CLIENT -int mainEntryClickHouseKeeperClient(int argc, char ** argv); -#endif - -int main(int argc_, char ** argv_) -{ -#if ENABLE_CLICKHOUSE_KEEPER_CLIENT - - if (argc_ >= 2) - { - /// 'clickhouse-keeper --client ...' and 'clickhouse-keeper client ...' are OK - if (strcmp(argv_[1], "--client") == 0 || strcmp(argv_[1], "client") == 0) - { - argv_[1] = argv_[0]; - return mainEntryClickHouseKeeperClient(--argc_, argv_ + 1); - } - } - - if (argc_ > 0 && (strcmp(argv_[0], "clickhouse-keeper-client") == 0 || endsWith(argv_[0], "/clickhouse-keeper-client"))) - return mainEntryClickHouseKeeperClient(argc_, argv_); -#endif - - return mainEntryClickHouseKeeper(argc_, argv_); -} diff --git a/programs/keeper/conf.d/local.yaml b/programs/keeper/conf.d/local.yaml new file mode 100644 index 00000000000..722e90e374a --- /dev/null +++ b/programs/keeper/conf.d/local.yaml @@ -0,0 +1,9 @@ +logger: + log: + "@remove": remove + errorlog: + "@remove": remove + console: 1 +keeper_server: + log_storage_path: ./logs + snapshot_storage_path: ./snapshots diff --git a/programs/keeper/keeper_main.cpp b/programs/keeper/keeper_main.cpp new file mode 100644 index 00000000000..a240f9699f2 --- /dev/null +++ b/programs/keeper/keeper_main.cpp @@ -0,0 +1,189 @@ +#include +#include + +#include +#include +#include +#include +#include /// pair + +#include + +#include "config.h" +#include "config_tools.h" + +#include +#include + +#include +#include +#include + +#include +#include + + +int mainEntryClickHouseKeeper(int argc, char ** argv); +#if ENABLE_CLICKHOUSE_KEEPER_CONVERTER +int mainEntryClickHouseKeeperConverter(int argc, char ** argv); +#endif +#if ENABLE_CLICKHOUSE_KEEPER_CLIENT +int mainEntryClickHouseKeeperClient(int argc, char ** argv); +#endif + +namespace +{ + +using MainFunc = int (*)(int, char**); + +/// Add an item here to register new application +std::pair clickhouse_applications[] = +{ + // keeper + {"keeper", mainEntryClickHouseKeeper}, +#if ENABLE_CLICKHOUSE_KEEPER_CONVERTER + {"converter", mainEntryClickHouseKeeperConverter}, + {"keeper-converter", mainEntryClickHouseKeeperConverter}, +#endif +#if ENABLE_CLICKHOUSE_KEEPER_CLIENT + {"client", mainEntryClickHouseKeeperClient}, + {"keeper-client", mainEntryClickHouseKeeperClient}, +#endif + +}; + +int printHelp(int, char **) +{ + std::cerr << "Use one of the following commands:" << std::endl; + for (auto & application : clickhouse_applications) + std::cerr << "clickhouse " << application.first << " [args] " << std::endl; + return -1; +} + +} + + +bool isClickhouseApp(std::string_view app_suffix, std::vector & argv) +{ + /// Use app if the first arg 'app' is passed (the arg should be quietly removed) + if (argv.size() >= 2) + { + auto first_arg = argv.begin() + 1; + + /// 'clickhouse --client ...' and 'clickhouse client ...' are Ok + if (*first_arg == app_suffix + || (std::string_view(*first_arg).starts_with("--") && std::string_view(*first_arg).substr(2) == app_suffix)) + { + argv.erase(first_arg); + return true; + } + } + + /// keeper suffix is default which will be used if no other app is detected + if (app_suffix == "keeper") + return false; + + /// Use app if clickhouse binary is run through symbolic link with name clickhouse-app + std::string app_name = "clickhouse-" + std::string(app_suffix); + return !argv.empty() && (app_name == argv[0] || endsWith(argv[0], "/" + app_name)); +} + +/// Don't allow dlopen in the main ClickHouse binary, because it is harmful and insecure. +/// We don't use it. But it can be used by some libraries for implementation of "plugins". +/// We absolutely discourage the ancient technique of loading +/// 3rd-party uncontrolled dangerous libraries into the process address space, +/// because it is insane. + +#if !defined(USE_MUSL) +extern "C" +{ + void * dlopen(const char *, int) + { + return nullptr; + } + + void * dlmopen(long, const char *, int) // NOLINT + { + return nullptr; + } + + int dlclose(void *) + { + return 0; + } + + const char * dlerror() + { + return "ClickHouse does not allow dynamic library loading"; + } +} +#endif + +/// Prevent messages from JeMalloc in the release build. +/// Some of these messages are non-actionable for the users, such as: +/// : Number of CPUs detected is not deterministic. Per-CPU arena disabled. +#if USE_JEMALLOC && defined(NDEBUG) && !defined(SANITIZER) +extern "C" void (*malloc_message)(void *, const char *s); +__attribute__((constructor(0))) void init_je_malloc_message() { malloc_message = [](void *, const char *){}; } +#endif + +/// This allows to implement assert to forbid initialization of a class in static constructors. +/// Usage: +/// +/// extern bool inside_main; +/// class C { C() { assert(inside_main); } }; +bool inside_main = false; + +int main(int argc_, char ** argv_) +{ + inside_main = true; + SCOPE_EXIT({ inside_main = false; }); + + /// PHDR cache is required for query profiler to work reliably + /// It also speed up exception handling, but exceptions from dynamically loaded libraries (dlopen) + /// will work only after additional call of this function. + /// Note: we forbid dlopen in our code. + updatePHDRCache(); + +#if !defined(USE_MUSL) + checkHarmfulEnvironmentVariables(argv_); +#endif + + /// This is used for testing. For example, + /// clickhouse-local should be able to run a simple query without throw/catch. + if (getenv("CLICKHOUSE_TERMINATE_ON_ANY_EXCEPTION")) // NOLINT(concurrency-mt-unsafe) + DB::terminate_on_any_exception = true; + + /// Reset new handler to default (that throws std::bad_alloc) + /// It is needed because LLVM library clobbers it. + std::set_new_handler(nullptr); + + std::vector argv(argv_, argv_ + argc_); + + /// Print a basic help if nothing was matched + MainFunc main_func = mainEntryClickHouseKeeper; + + if (isClickhouseApp("help", argv)) + { + main_func = printHelp; + } + else + { + for (auto & application : clickhouse_applications) + { + if (isClickhouseApp(application.first, argv)) + { + main_func = application.second; + break; + } + } + } + + int exit_code = main_func(static_cast(argv.size()), argv.data()); + +#if defined(SANITIZE_COVERAGE) + dumpCoverage(); +#endif + + return exit_code; +} diff --git a/programs/library-bridge/CMakeLists.txt b/programs/library-bridge/CMakeLists.txt index 2fca10ce4d7..86410d712ec 100644 --- a/programs/library-bridge/CMakeLists.txt +++ b/programs/library-bridge/CMakeLists.txt @@ -11,7 +11,6 @@ set (CLICKHOUSE_LIBRARY_BRIDGE_SOURCES LibraryBridgeHandlers.cpp SharedLibrary.cpp library-bridge.cpp - createFunctionBaseCast.cpp ) clickhouse_add_executable(clickhouse-library-bridge ${CLICKHOUSE_LIBRARY_BRIDGE_SOURCES}) @@ -20,6 +19,7 @@ target_link_libraries(clickhouse-library-bridge PRIVATE daemon dbms bridge + clickhouse_functions ) set_target_properties(clickhouse-library-bridge PROPERTIES RUNTIME_OUTPUT_DIRECTORY ..) diff --git a/programs/library-bridge/LibraryBridge.cpp b/programs/library-bridge/LibraryBridge.cpp index 8a07ca57104..d35e7a9b832 100644 --- a/programs/library-bridge/LibraryBridge.cpp +++ b/programs/library-bridge/LibraryBridge.cpp @@ -1,5 +1,7 @@ #include "LibraryBridge.h" +#include + int mainEntryClickHouseLibraryBridge(int argc, char ** argv) { DB::LibraryBridge app; diff --git a/programs/library-bridge/LibraryBridgeHandlers.cpp b/programs/library-bridge/LibraryBridgeHandlers.cpp index 8d116e537aa..53f4b31ca06 100644 --- a/programs/library-bridge/LibraryBridgeHandlers.cpp +++ b/programs/library-bridge/LibraryBridgeHandlers.cpp @@ -6,6 +6,7 @@ #include "ExternalDictionaryLibraryHandlerFactory.h" #include +#include #include #include #include diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp index ab2311c7196..414cf65b12b 100644 --- a/programs/local/LocalServer.cpp +++ b/programs/local/LocalServer.cpp @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include @@ -59,8 +60,13 @@ # include #endif + namespace fs = std::filesystem; +namespace CurrentMetrics +{ + extern const Metric MemoryTracking; +} namespace DB { @@ -82,6 +88,11 @@ void applySettingsOverridesForLocal(ContextMutablePtr context) context->setSettings(settings); } +Poco::Util::LayeredConfiguration & LocalServer::getClientConfiguration() +{ + return config(); +} + void LocalServer::processError(const String &) const { if (ignore_error) @@ -117,20 +128,21 @@ void LocalServer::initialize(Poco::Util::Application & self) Poco::Util::Application::initialize(self); /// Load config files if exists - if (config().has("config-file") || fs::exists("config.xml")) + if (getClientConfiguration().has("config-file") || fs::exists("config.xml")) { - const auto config_path = config().getString("config-file", "config.xml"); + const auto config_path = getClientConfiguration().getString("config-file", "config.xml"); ConfigProcessor config_processor(config_path, false, true); ConfigProcessor::setConfigPath(fs::path(config_path).parent_path()); auto loaded_config = config_processor.loadConfig(); - config().add(loaded_config.configuration.duplicate(), PRIO_DEFAULT, false); + getClientConfiguration().add(loaded_config.configuration.duplicate(), PRIO_DEFAULT, false); } + server_settings.loadSettingsFromConfig(config()); + GlobalThreadPool::initialize( - config().getUInt("max_thread_pool_size", 10000), - config().getUInt("max_thread_pool_free_size", 1000), - config().getUInt("thread_pool_queue_size", 10000) - ); + server_settings.max_thread_pool_size, + server_settings.max_thread_pool_free_size, + server_settings.thread_pool_queue_size); #if USE_AZURE_BLOB_STORAGE /// See the explanation near the same line in Server.cpp @@ -141,18 +153,17 @@ void LocalServer::initialize(Poco::Util::Application & self) #endif getIOThreadPool().initialize( - config().getUInt("max_io_thread_pool_size", 100), - config().getUInt("max_io_thread_pool_free_size", 0), - config().getUInt("io_thread_pool_queue_size", 10000)); + server_settings.max_io_thread_pool_size, + server_settings.max_io_thread_pool_free_size, + server_settings.io_thread_pool_queue_size); - - const size_t active_parts_loading_threads = config().getUInt("max_active_parts_loading_thread_pool_size", 64); + const size_t active_parts_loading_threads = server_settings.max_active_parts_loading_thread_pool_size; getActivePartsLoadingThreadPool().initialize( active_parts_loading_threads, 0, // We don't need any threads one all the parts will be loaded active_parts_loading_threads); - const size_t outdated_parts_loading_threads = config().getUInt("max_outdated_parts_loading_thread_pool_size", 32); + const size_t outdated_parts_loading_threads = server_settings.max_outdated_parts_loading_thread_pool_size; getOutdatedPartsLoadingThreadPool().initialize( outdated_parts_loading_threads, 0, // We don't need any threads one all the parts will be loaded @@ -160,7 +171,7 @@ void LocalServer::initialize(Poco::Util::Application & self) getOutdatedPartsLoadingThreadPool().setMaxTurboThreads(active_parts_loading_threads); - const size_t unexpected_parts_loading_threads = config().getUInt("max_unexpected_parts_loading_thread_pool_size", 32); + const size_t unexpected_parts_loading_threads = server_settings.max_unexpected_parts_loading_thread_pool_size; getUnexpectedPartsLoadingThreadPool().initialize( unexpected_parts_loading_threads, 0, // We don't need any threads one all the parts will be loaded @@ -168,7 +179,7 @@ void LocalServer::initialize(Poco::Util::Application & self) getUnexpectedPartsLoadingThreadPool().setMaxTurboThreads(active_parts_loading_threads); - const size_t cleanup_threads = config().getUInt("max_parts_cleaning_thread_pool_size", 128); + const size_t cleanup_threads = server_settings.max_parts_cleaning_thread_pool_size; getPartsCleaningThreadPool().initialize( cleanup_threads, 0, // We don't need any threads one all the parts will be deleted @@ -201,10 +212,10 @@ void LocalServer::tryInitPath() { std::string path; - if (config().has("path")) + if (getClientConfiguration().has("path")) { // User-supplied path. - path = config().getString("path"); + path = getClientConfiguration().getString("path"); Poco::trimInPlace(path); if (path.empty()) @@ -263,13 +274,13 @@ void LocalServer::tryInitPath() global_context->setUserFilesPath(""); /// user's files are everywhere - std::string user_scripts_path = config().getString("user_scripts_path", fs::path(path) / "user_scripts/"); + std::string user_scripts_path = getClientConfiguration().getString("user_scripts_path", fs::path(path) / "user_scripts/"); global_context->setUserScriptsPath(user_scripts_path); /// top_level_domains_lists - const std::string & top_level_domains_path = config().getString("top_level_domains_path", fs::path(path) / "top_level_domains/"); + const std::string & top_level_domains_path = getClientConfiguration().getString("top_level_domains_path", fs::path(path) / "top_level_domains/"); if (!top_level_domains_path.empty()) - TLDListsHolder::getInstance().parseConfig(fs::path(top_level_domains_path) / "", config()); + TLDListsHolder::getInstance().parseConfig(fs::path(top_level_domains_path) / "", getClientConfiguration()); } @@ -311,14 +322,14 @@ void LocalServer::cleanup() std::string LocalServer::getInitialCreateTableQuery() { - if (!config().has("table-structure") && !config().has("table-file") && !config().has("table-data-format") && (!isRegularFile(STDIN_FILENO) || queries.empty())) + if (!getClientConfiguration().has("table-structure") && !getClientConfiguration().has("table-file") && !getClientConfiguration().has("table-data-format") && (!isRegularFile(STDIN_FILENO) || queries.empty())) return {}; - auto table_name = backQuoteIfNeed(config().getString("table-name", "table")); - auto table_structure = config().getString("table-structure", "auto"); + auto table_name = backQuoteIfNeed(getClientConfiguration().getString("table-name", "table")); + auto table_structure = getClientConfiguration().getString("table-structure", "auto"); String table_file; - if (!config().has("table-file") || config().getString("table-file") == "-") + if (!getClientConfiguration().has("table-file") || getClientConfiguration().getString("table-file") == "-") { /// Use Unix tools stdin naming convention table_file = "stdin"; @@ -326,7 +337,7 @@ std::string LocalServer::getInitialCreateTableQuery() else { /// Use regular file - auto file_name = config().getString("table-file"); + auto file_name = getClientConfiguration().getString("table-file"); table_file = quoteString(file_name); } @@ -365,6 +376,7 @@ void LocalServer::setupUsers() "
" " default" " default" + " 1" " " " " " " @@ -374,18 +386,18 @@ void LocalServer::setupUsers() ConfigurationPtr users_config; auto & access_control = global_context->getAccessControl(); - access_control.setNoPasswordAllowed(config().getBool("allow_no_password", true)); - access_control.setPlaintextPasswordAllowed(config().getBool("allow_plaintext_password", true)); - if (config().has("config-file") || fs::exists("config.xml")) + access_control.setNoPasswordAllowed(getClientConfiguration().getBool("allow_no_password", true)); + access_control.setPlaintextPasswordAllowed(getClientConfiguration().getBool("allow_plaintext_password", true)); + if (getClientConfiguration().has("config-file") || fs::exists("config.xml")) { - String config_path = config().getString("config-file", ""); - bool has_user_directories = config().has("user_directories"); + String config_path = getClientConfiguration().getString("config-file", ""); + bool has_user_directories = getClientConfiguration().has("user_directories"); const auto config_dir = fs::path{config_path}.remove_filename().string(); - String users_config_path = config().getString("users_config", ""); + String users_config_path = getClientConfiguration().getString("users_config", ""); if (users_config_path.empty() && has_user_directories) { - users_config_path = config().getString("user_directories.users_xml.path"); + users_config_path = getClientConfiguration().getString("user_directories.users_xml.path"); if (fs::path(users_config_path).is_relative() && fs::exists(fs::path(config_dir) / users_config_path)) users_config_path = fs::path(config_dir) / users_config_path; } @@ -409,10 +421,11 @@ void LocalServer::setupUsers() void LocalServer::connect() { - connection_parameters = ConnectionParameters(config(), "localhost"); + connection_parameters = ConnectionParameters(getClientConfiguration(), "localhost"); + /// This is needed for table function input(...). ReadBuffer * in; - auto table_file = config().getString("table-file", "-"); + auto table_file = getClientConfiguration().getString("table-file", "-"); if (table_file == "-" || table_file == "stdin") { in = &std_in; @@ -433,7 +446,7 @@ try UseSSL use_ssl; thread_status.emplace(); - StackTrace::setShowAddresses(config().getBool("show_addresses_in_stack_traces", true)); + StackTrace::setShowAddresses(server_settings.show_addresses_in_stack_traces); setupSignalHandler(); @@ -448,7 +461,7 @@ try if (rlim.rlim_cur < rlim.rlim_max) { - rlim.rlim_cur = config().getUInt("max_open_files", static_cast(rlim.rlim_max)); + rlim.rlim_cur = getClientConfiguration().getUInt("max_open_files", static_cast(rlim.rlim_max)); int rc = setrlimit(RLIMIT_NOFILE, &rlim); if (rc != 0) std::cerr << fmt::format("Cannot set max number of file descriptors to {}. Try to specify max_open_files according to your system limits. error: {}", rlim.rlim_cur, errnoToString()) << '\n'; @@ -456,8 +469,8 @@ try } is_interactive = stdin_is_a_tty - && (config().hasOption("interactive") - || (queries.empty() && !config().has("table-structure") && queries_files.empty() && !config().has("table-file"))); + && (getClientConfiguration().hasOption("interactive") + || (queries.empty() && !getClientConfiguration().has("table-structure") && queries_files.empty() && !getClientConfiguration().has("table-file"))); if (!is_interactive) { @@ -481,7 +494,7 @@ try SCOPE_EXIT({ cleanup(); }); - initTTYBuffer(toProgressOption(config().getString("progress", "default"))); + initTTYBuffer(toProgressOption(getClientConfiguration().getString("progress", "default"))); ASTAlterCommand::setFormatAlterCommandsWithParentheses(true); applyCmdSettings(global_context); @@ -489,7 +502,7 @@ try /// try to load user defined executable functions, throw on error and die try { - global_context->loadOrReloadUserDefinedExecutableFunctions(config()); + global_context->loadOrReloadUserDefinedExecutableFunctions(getClientConfiguration()); } catch (...) { @@ -530,7 +543,7 @@ try } catch (const DB::Exception & e) { - bool need_print_stack_trace = config().getBool("stacktrace", false); + bool need_print_stack_trace = getClientConfiguration().getBool("stacktrace", false); std::cerr << getExceptionMessage(e, need_print_stack_trace, true) << std::endl; return e.code() ? e.code() : -1; } @@ -542,42 +555,42 @@ catch (...) void LocalServer::updateLoggerLevel(const String & logs_level) { - config().setString("logger.level", logs_level); - updateLevels(config(), logger()); + getClientConfiguration().setString("logger.level", logs_level); + updateLevels(getClientConfiguration(), logger()); } void LocalServer::processConfig() { - if (!queries.empty() && config().has("queries-file")) + if (!queries.empty() && getClientConfiguration().has("queries-file")) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Options '--query' and '--queries-file' cannot be specified at the same time"); - if (config().has("multiquery")) + if (getClientConfiguration().has("multiquery")) is_multiquery = true; - pager = config().getString("pager", ""); + pager = getClientConfiguration().getString("pager", ""); - delayed_interactive = config().has("interactive") && (!queries.empty() || config().has("queries-file")); + delayed_interactive = getClientConfiguration().has("interactive") && (!queries.empty() || getClientConfiguration().has("queries-file")); if (!is_interactive || delayed_interactive) { - echo_queries = config().hasOption("echo") || config().hasOption("verbose"); - ignore_error = config().getBool("ignore-error", false); + echo_queries = getClientConfiguration().hasOption("echo") || getClientConfiguration().hasOption("verbose"); + ignore_error = getClientConfiguration().getBool("ignore-error", false); } - print_stack_trace = config().getBool("stacktrace", false); + print_stack_trace = getClientConfiguration().getBool("stacktrace", false); const std::string clickhouse_dialect{"clickhouse"}; - load_suggestions = (is_interactive || delayed_interactive) && !config().getBool("disable_suggestion", false) - && config().getString("dialect", clickhouse_dialect) == clickhouse_dialect; - wait_for_suggestions_to_load = config().getBool("wait_for_suggestions_to_load", false); + load_suggestions = (is_interactive || delayed_interactive) && !getClientConfiguration().getBool("disable_suggestion", false) + && getClientConfiguration().getString("dialect", clickhouse_dialect) == clickhouse_dialect; + wait_for_suggestions_to_load = getClientConfiguration().getBool("wait_for_suggestions_to_load", false); - auto logging = (config().has("logger.console") - || config().has("logger.level") - || config().has("log-level") - || config().has("send_logs_level") - || config().has("logger.log")); + auto logging = (getClientConfiguration().has("logger.console") + || getClientConfiguration().has("logger.level") + || getClientConfiguration().has("log-level") + || getClientConfiguration().has("send_logs_level") + || getClientConfiguration().has("logger.log")); - auto level = config().getString("log-level", "trace"); + auto level = getClientConfiguration().getString("log-level", "trace"); - if (config().has("server_logs_file")) + if (getClientConfiguration().has("server_logs_file")) { auto poco_logs_level = Poco::Logger::parseLevel(level); Poco::Logger::root().setLevel(poco_logs_level); @@ -587,10 +600,10 @@ void LocalServer::processConfig() } else { - config().setString("logger", "logger"); + getClientConfiguration().setString("logger", "logger"); auto log_level_default = logging ? level : "fatal"; - config().setString("logger.level", config().getString("log-level", config().getString("send_logs_level", log_level_default))); - buildLoggers(config(), logger(), "clickhouse-local"); + getClientConfiguration().setString("logger.level", getClientConfiguration().getString("log-level", getClientConfiguration().getString("send_logs_level", log_level_default))); + buildLoggers(getClientConfiguration(), logger(), "clickhouse-local"); } shared_context = Context::createShared(); @@ -604,13 +617,13 @@ void LocalServer::processConfig() LoggerRawPtr log = &logger(); /// Maybe useless - if (config().has("macros")) - global_context->setMacros(std::make_unique(config(), "macros", log)); + if (getClientConfiguration().has("macros")) + global_context->setMacros(std::make_unique(getClientConfiguration(), "macros", log)); setDefaultFormatsAndCompressionFromConfiguration(); /// Sets external authenticators config (LDAP, Kerberos). - global_context->setExternalAuthenticatorsConfig(config()); + global_context->setExternalAuthenticatorsConfig(getClientConfiguration()); setupUsers(); @@ -619,12 +632,43 @@ void LocalServer::processConfig() global_context->getProcessList().setMaxSize(0); const size_t physical_server_memory = getMemoryAmount(); - const double cache_size_to_ram_max_ratio = config().getDouble("cache_size_to_ram_max_ratio", 0.5); + + size_t max_server_memory_usage = server_settings.max_server_memory_usage; + double max_server_memory_usage_to_ram_ratio = server_settings.max_server_memory_usage_to_ram_ratio; + + size_t default_max_server_memory_usage = static_cast(physical_server_memory * max_server_memory_usage_to_ram_ratio); + + if (max_server_memory_usage == 0) + { + max_server_memory_usage = default_max_server_memory_usage; + LOG_INFO(log, "Setting max_server_memory_usage was set to {}" + " ({} available * {:.2f} max_server_memory_usage_to_ram_ratio)", + formatReadableSizeWithBinarySuffix(max_server_memory_usage), + formatReadableSizeWithBinarySuffix(physical_server_memory), + max_server_memory_usage_to_ram_ratio); + } + else if (max_server_memory_usage > default_max_server_memory_usage) + { + max_server_memory_usage = default_max_server_memory_usage; + LOG_INFO(log, "Setting max_server_memory_usage was lowered to {}" + " because the system has low amount of memory. The amount was" + " calculated as {} available" + " * {:.2f} max_server_memory_usage_to_ram_ratio", + formatReadableSizeWithBinarySuffix(max_server_memory_usage), + formatReadableSizeWithBinarySuffix(physical_server_memory), + max_server_memory_usage_to_ram_ratio); + } + + total_memory_tracker.setHardLimit(max_server_memory_usage); + total_memory_tracker.setDescription("(total)"); + total_memory_tracker.setMetric(CurrentMetrics::MemoryTracking); + + const double cache_size_to_ram_max_ratio = server_settings.cache_size_to_ram_max_ratio; const size_t max_cache_size = static_cast(physical_server_memory * cache_size_to_ram_max_ratio); - String uncompressed_cache_policy = config().getString("uncompressed_cache_policy", DEFAULT_UNCOMPRESSED_CACHE_POLICY); - size_t uncompressed_cache_size = config().getUInt64("uncompressed_cache_size", DEFAULT_UNCOMPRESSED_CACHE_MAX_SIZE); - double uncompressed_cache_size_ratio = config().getDouble("uncompressed_cache_size_ratio", DEFAULT_UNCOMPRESSED_CACHE_SIZE_RATIO); + String uncompressed_cache_policy = server_settings.uncompressed_cache_policy; + size_t uncompressed_cache_size = server_settings.uncompressed_cache_size; + double uncompressed_cache_size_ratio = server_settings.uncompressed_cache_size_ratio; if (uncompressed_cache_size > max_cache_size) { uncompressed_cache_size = max_cache_size; @@ -632,9 +676,9 @@ void LocalServer::processConfig() } global_context->setUncompressedCache(uncompressed_cache_policy, uncompressed_cache_size, uncompressed_cache_size_ratio); - String mark_cache_policy = config().getString("mark_cache_policy", DEFAULT_MARK_CACHE_POLICY); - size_t mark_cache_size = config().getUInt64("mark_cache_size", DEFAULT_MARK_CACHE_MAX_SIZE); - double mark_cache_size_ratio = config().getDouble("mark_cache_size_ratio", DEFAULT_MARK_CACHE_SIZE_RATIO); + String mark_cache_policy = server_settings.mark_cache_policy; + size_t mark_cache_size = server_settings.mark_cache_size; + double mark_cache_size_ratio = server_settings.mark_cache_size_ratio; if (!mark_cache_size) LOG_ERROR(log, "Too low mark cache size will lead to severe performance degradation."); if (mark_cache_size > max_cache_size) @@ -644,9 +688,9 @@ void LocalServer::processConfig() } global_context->setMarkCache(mark_cache_policy, mark_cache_size, mark_cache_size_ratio); - String index_uncompressed_cache_policy = config().getString("index_uncompressed_cache_policy", DEFAULT_INDEX_UNCOMPRESSED_CACHE_POLICY); - size_t index_uncompressed_cache_size = config().getUInt64("index_uncompressed_cache_size", DEFAULT_INDEX_UNCOMPRESSED_CACHE_MAX_SIZE); - double index_uncompressed_cache_size_ratio = config().getDouble("index_uncompressed_cache_size_ratio", DEFAULT_INDEX_UNCOMPRESSED_CACHE_SIZE_RATIO); + String index_uncompressed_cache_policy = server_settings.index_uncompressed_cache_policy; + size_t index_uncompressed_cache_size = server_settings.index_uncompressed_cache_size; + double index_uncompressed_cache_size_ratio = server_settings.index_uncompressed_cache_size_ratio; if (index_uncompressed_cache_size > max_cache_size) { index_uncompressed_cache_size = max_cache_size; @@ -654,9 +698,9 @@ void LocalServer::processConfig() } global_context->setIndexUncompressedCache(index_uncompressed_cache_policy, index_uncompressed_cache_size, index_uncompressed_cache_size_ratio); - String index_mark_cache_policy = config().getString("index_mark_cache_policy", DEFAULT_INDEX_MARK_CACHE_POLICY); - size_t index_mark_cache_size = config().getUInt64("index_mark_cache_size", DEFAULT_INDEX_MARK_CACHE_MAX_SIZE); - double index_mark_cache_size_ratio = config().getDouble("index_mark_cache_size_ratio", DEFAULT_INDEX_MARK_CACHE_SIZE_RATIO); + String index_mark_cache_policy = server_settings.index_mark_cache_policy; + size_t index_mark_cache_size = server_settings.index_mark_cache_size; + double index_mark_cache_size_ratio = server_settings.index_mark_cache_size_ratio; if (index_mark_cache_size > max_cache_size) { index_mark_cache_size = max_cache_size; @@ -664,7 +708,7 @@ void LocalServer::processConfig() } global_context->setIndexMarkCache(index_mark_cache_policy, index_mark_cache_size, index_mark_cache_size_ratio); - size_t mmap_cache_size = config().getUInt64("mmap_cache_size", DEFAULT_MMAP_CACHE_MAX_SIZE); + size_t mmap_cache_size = server_settings.mmap_cache_size; if (mmap_cache_size > max_cache_size) { mmap_cache_size = max_cache_size; @@ -676,8 +720,8 @@ void LocalServer::processConfig() global_context->setQueryCache(0, 0, 0, 0); #if USE_EMBEDDED_COMPILER - size_t compiled_expression_cache_max_size_in_bytes = config().getUInt64("compiled_expression_cache_size", DEFAULT_COMPILED_EXPRESSION_CACHE_MAX_SIZE); - size_t compiled_expression_cache_max_elements = config().getUInt64("compiled_expression_cache_elements_size", DEFAULT_COMPILED_EXPRESSION_CACHE_MAX_ENTRIES); + size_t compiled_expression_cache_max_size_in_bytes = server_settings.compiled_expression_cache_size; + size_t compiled_expression_cache_max_elements = server_settings.compiled_expression_cache_elements_size; CompiledExpressionCacheFactory::instance().init(compiled_expression_cache_max_size_in_bytes, compiled_expression_cache_max_elements); #endif @@ -689,16 +733,16 @@ void LocalServer::processConfig() applyCmdOptions(global_context); /// Load global settings from default_profile and system_profile. - global_context->setDefaultProfiles(config()); + global_context->setDefaultProfiles(getClientConfiguration()); /// We load temporary database first, because projections need it. DatabaseCatalog::instance().initializeAndLoadTemporaryDatabase(); - std::string default_database = config().getString("default_database", "default"); + std::string default_database = server_settings.default_database; DatabaseCatalog::instance().attachDatabase(default_database, createClickHouseLocalDatabaseOverlay(default_database, global_context)); global_context->setCurrentDatabase(default_database); - if (config().has("path")) + if (getClientConfiguration().has("path")) { String path = global_context->getPath(); fs::create_directories(fs::path(path)); @@ -713,7 +757,7 @@ void LocalServer::processConfig() attachInformationSchema(global_context, *createMemoryDatabaseIfNotExists(global_context, DatabaseCatalog::INFORMATION_SCHEMA_UPPERCASE)); waitLoad(TablesLoaderForegroundPoolId, startup_system_tasks); - if (!config().has("only-system-tables")) + if (!getClientConfiguration().has("only-system-tables")) { DatabaseCatalog::instance().createBackgroundTasks(); waitLoad(loadMetadata(global_context)); @@ -725,15 +769,15 @@ void LocalServer::processConfig() LOG_DEBUG(log, "Loaded metadata."); } - else if (!config().has("no-system-tables")) + else if (!getClientConfiguration().has("no-system-tables")) { attachSystemTablesServer(global_context, *createMemoryDatabaseIfNotExists(global_context, DatabaseCatalog::SYSTEM_DATABASE), false); attachInformationSchema(global_context, *createMemoryDatabaseIfNotExists(global_context, DatabaseCatalog::INFORMATION_SCHEMA)); attachInformationSchema(global_context, *createMemoryDatabaseIfNotExists(global_context, DatabaseCatalog::INFORMATION_SCHEMA_UPPERCASE)); } - server_display_name = config().getString("display_name", ""); - prompt_by_server_display_name = config().getRawString("prompt_by_server_display_name.default", ":) "); + server_display_name = getClientConfiguration().getString("display_name", ""); + prompt_by_server_display_name = getClientConfiguration().getRawString("prompt_by_server_display_name.default", ":) "); global_context->setQueryKindInitial(); global_context->setQueryKind(query_kind); @@ -811,7 +855,7 @@ void LocalServer::applyCmdSettings(ContextMutablePtr context) void LocalServer::applyCmdOptions(ContextMutablePtr context) { - context->setDefaultFormat(config().getString("output-format", config().getString("format", is_interactive ? "PrettyCompact" : "TSV"))); + context->setDefaultFormat(getClientConfiguration().getString("output-format", getClientConfiguration().getString("format", is_interactive ? "PrettyCompact" : "TSV"))); applyCmdSettings(context); } @@ -819,33 +863,33 @@ void LocalServer::applyCmdOptions(ContextMutablePtr context) void LocalServer::processOptions(const OptionsDescription &, const CommandLineOptions & options, const std::vector &, const std::vector &) { if (options.count("table")) - config().setString("table-name", options["table"].as()); + getClientConfiguration().setString("table-name", options["table"].as()); if (options.count("file")) - config().setString("table-file", options["file"].as()); + getClientConfiguration().setString("table-file", options["file"].as()); if (options.count("structure")) - config().setString("table-structure", options["structure"].as()); + getClientConfiguration().setString("table-structure", options["structure"].as()); if (options.count("no-system-tables")) - config().setBool("no-system-tables", true); + getClientConfiguration().setBool("no-system-tables", true); if (options.count("only-system-tables")) - config().setBool("only-system-tables", true); + getClientConfiguration().setBool("only-system-tables", true); if (options.count("database")) - config().setString("default_database", options["database"].as()); + getClientConfiguration().setString("default_database", options["database"].as()); if (options.count("input-format")) - config().setString("table-data-format", options["input-format"].as()); + getClientConfiguration().setString("table-data-format", options["input-format"].as()); if (options.count("output-format")) - config().setString("output-format", options["output-format"].as()); + getClientConfiguration().setString("output-format", options["output-format"].as()); if (options.count("logger.console")) - config().setBool("logger.console", options["logger.console"].as()); + getClientConfiguration().setBool("logger.console", options["logger.console"].as()); if (options.count("logger.log")) - config().setString("logger.log", options["logger.log"].as()); + getClientConfiguration().setString("logger.log", options["logger.log"].as()); if (options.count("logger.level")) - config().setString("logger.level", options["logger.level"].as()); + getClientConfiguration().setString("logger.level", options["logger.level"].as()); if (options.count("send_logs_level")) - config().setString("send_logs_level", options["send_logs_level"].as()); + getClientConfiguration().setString("send_logs_level", options["send_logs_level"].as()); if (options.count("wait_for_suggestions_to_load")) - config().setBool("wait_for_suggestions_to_load", true); + getClientConfiguration().setBool("wait_for_suggestions_to_load", true); } void LocalServer::readArguments(int argc, char ** argv, Arguments & common_arguments, std::vector &, std::vector &) diff --git a/programs/local/LocalServer.h b/programs/local/LocalServer.h index 4856e68ff9b..0715f358313 100644 --- a/programs/local/LocalServer.h +++ b/programs/local/LocalServer.h @@ -3,11 +3,12 @@ #include #include -#include -#include -#include +#include #include #include +#include +#include +#include #include #include @@ -30,6 +31,9 @@ public: int main(const std::vector & /*args*/) override; protected: + + Poco::Util::LayeredConfiguration & getClientConfiguration() override; + void connect() override; void processError(const String & query) const override; @@ -63,6 +67,8 @@ private: void applyCmdOptions(ContextMutablePtr context); void applyCmdSettings(ContextMutablePtr context); + ServerSettings server_settings; + std::optional status; std::optional temporary_directory_to_delete; diff --git a/programs/main.cpp b/programs/main.cpp index c270388f17f..02ea1471108 100644 --- a/programs/main.cpp +++ b/programs/main.cpp @@ -1,5 +1,3 @@ -#include -#include #include #include @@ -7,14 +5,16 @@ #include #include #include -#include #include #include /// pair #include +#include "config.h" #include "config_tools.h" +#include +#include #include #include #include @@ -118,268 +118,6 @@ std::pair clickhouse_short_names[] = {"chc", "client"}, }; - -enum class InstructionFail : uint8_t -{ - NONE = 0, - SSE3 = 1, - SSSE3 = 2, - SSE4_1 = 3, - SSE4_2 = 4, - POPCNT = 5, - AVX = 6, - AVX2 = 7, - AVX512 = 8 -}; - -auto instructionFailToString(InstructionFail fail) -{ - switch (fail) - { -#define ret(x) return std::make_tuple(STDERR_FILENO, x, sizeof(x) - 1) - case InstructionFail::NONE: - ret("NONE"); - case InstructionFail::SSE3: - ret("SSE3"); - case InstructionFail::SSSE3: - ret("SSSE3"); - case InstructionFail::SSE4_1: - ret("SSE4.1"); - case InstructionFail::SSE4_2: - ret("SSE4.2"); - case InstructionFail::POPCNT: - ret("POPCNT"); - case InstructionFail::AVX: - ret("AVX"); - case InstructionFail::AVX2: - ret("AVX2"); - case InstructionFail::AVX512: - ret("AVX512"); -#undef ret - } -} - - -sigjmp_buf jmpbuf; - -[[noreturn]] void sigIllCheckHandler(int, siginfo_t *, void *) -{ - siglongjmp(jmpbuf, 1); -} - -/// Check if necessary SSE extensions are available by trying to execute some sse instructions. -/// If instruction is unavailable, SIGILL will be sent by kernel. -void checkRequiredInstructionsImpl(volatile InstructionFail & fail) -{ -#if defined(__SSE3__) - fail = InstructionFail::SSE3; - __asm__ volatile ("addsubpd %%xmm0, %%xmm0" : : : "xmm0"); -#endif - -#if defined(__SSSE3__) - fail = InstructionFail::SSSE3; - __asm__ volatile ("pabsw %%xmm0, %%xmm0" : : : "xmm0"); - -#endif - -#if defined(__SSE4_1__) - fail = InstructionFail::SSE4_1; - __asm__ volatile ("pmaxud %%xmm0, %%xmm0" : : : "xmm0"); -#endif - -#if defined(__SSE4_2__) - fail = InstructionFail::SSE4_2; - __asm__ volatile ("pcmpgtq %%xmm0, %%xmm0" : : : "xmm0"); -#endif - - /// Defined by -msse4.2 -#if defined(__POPCNT__) - fail = InstructionFail::POPCNT; - { - uint64_t a = 0; - uint64_t b = 0; - __asm__ volatile ("popcnt %1, %0" : "=r"(a) :"r"(b) :); - } -#endif - -#if defined(__AVX__) - fail = InstructionFail::AVX; - __asm__ volatile ("vaddpd %%ymm0, %%ymm0, %%ymm0" : : : "ymm0"); -#endif - -#if defined(__AVX2__) - fail = InstructionFail::AVX2; - __asm__ volatile ("vpabsw %%ymm0, %%ymm0" : : : "ymm0"); -#endif - -#if defined(__AVX512__) - fail = InstructionFail::AVX512; - __asm__ volatile ("vpabsw %%zmm0, %%zmm0" : : : "zmm0"); -#endif - - fail = InstructionFail::NONE; -} - -/// Macros to avoid using strlen(), since it may fail if SSE is not supported. -#define writeError(data) do \ - { \ - static_assert(__builtin_constant_p(data)); \ - if (!writeRetry(STDERR_FILENO, data, sizeof(data) - 1)) \ - _Exit(1); \ - } while (false) - -/// Check SSE and others instructions availability. Calls exit on fail. -/// This function must be called as early as possible, even before main, because static initializers may use unavailable instructions. -void checkRequiredInstructions() -{ - struct sigaction sa{}; - struct sigaction sa_old{}; - sa.sa_sigaction = sigIllCheckHandler; - sa.sa_flags = SA_SIGINFO; - auto signal = SIGILL; - if (sigemptyset(&sa.sa_mask) != 0 - || sigaddset(&sa.sa_mask, signal) != 0 - || sigaction(signal, &sa, &sa_old) != 0) - { - /// You may wonder about strlen. - /// Typical implementation of strlen is using SSE4.2 or AVX2. - /// But this is not the case because it's compiler builtin and is executed at compile time. - - writeError("Can not set signal handler\n"); - _Exit(1); - } - - volatile InstructionFail fail = InstructionFail::NONE; - - if (sigsetjmp(jmpbuf, 1)) - { - writeError("Instruction check fail. The CPU does not support "); - if (!std::apply(writeRetry, instructionFailToString(fail))) - _Exit(1); - writeError(" instruction set.\n"); - _Exit(1); - } - - checkRequiredInstructionsImpl(fail); - - if (sigaction(signal, &sa_old, nullptr)) - { - writeError("Can not set signal handler\n"); - _Exit(1); - } -} - -struct Checker -{ - Checker() - { - checkRequiredInstructions(); - } -} checker -#ifndef OS_DARWIN - __attribute__((init_priority(101))) /// Run before other static initializers. -#endif -; - - -#if !defined(USE_MUSL) -/// NOTE: We will migrate to full static linking or our own dynamic loader to make this code obsolete. -void checkHarmfulEnvironmentVariables(char ** argv) -{ - std::initializer_list harmful_env_variables = { - /// The list is a selection from "man ld-linux". - "LD_PRELOAD", - "LD_LIBRARY_PATH", - "LD_ORIGIN_PATH", - "LD_AUDIT", - "LD_DYNAMIC_WEAK", - /// The list is a selection from "man dyld" (osx). - "DYLD_LIBRARY_PATH", - "DYLD_FALLBACK_LIBRARY_PATH", - "DYLD_VERSIONED_LIBRARY_PATH", - "DYLD_INSERT_LIBRARIES", - }; - - bool require_reexec = false; - for (const auto * var : harmful_env_variables) - { - if (const char * value = getenv(var); value && value[0]) // NOLINT(concurrency-mt-unsafe) - { - /// NOTE: setenv() is used over unsetenv() since unsetenv() marked as harmful - if (setenv(var, "", true)) // NOLINT(concurrency-mt-unsafe) // this is safe if not called concurrently - { - fmt::print(stderr, "Cannot override {} environment variable", var); - _exit(1); - } - require_reexec = true; - } - } - - if (require_reexec) - { - /// Use execvp() over execv() to search in PATH. - /// - /// This should be safe, since: - /// - if argv[0] is relative path - it is OK - /// - if argv[0] has only basename, the it will search in PATH, like shell will do. - /// - /// Also note, that this (search in PATH) because there is no easy and - /// portable way to get absolute path of argv[0]. - /// - on linux there is /proc/self/exec and AT_EXECFN - /// - but on other OSes there is no such thing (especially on OSX). - /// - /// And since static linking will be done someday anyway, - /// let's not pollute the code base with special cases. - int error = execvp(argv[0], argv); - _exit(error); - } -} -#endif - - -#if defined(SANITIZE_COVERAGE) -__attribute__((no_sanitize("coverage"))) -void dumpCoverage() -{ - /// A user can request to dump the coverage information into files at exit. - /// This is useful for non-server applications such as clickhouse-format or clickhouse-client, - /// that cannot introspect it with SQL functions at runtime. - - /// The CLICKHOUSE_WRITE_COVERAGE environment variable defines a prefix for a filename 'prefix.pid' - /// containing the list of addresses of covered . - - /// The format is even simpler than Clang's "sancov": an array of 64-bit addresses, native byte order, no header. - - if (const char * coverage_filename_prefix = getenv("CLICKHOUSE_WRITE_COVERAGE")) // NOLINT(concurrency-mt-unsafe) - { - auto dump = [](const std::string & name, auto span) - { - /// Write only non-zeros. - std::vector data; - data.reserve(span.size()); - for (auto addr : span) - if (addr) - data.push_back(addr); - - int fd = ::open(name.c_str(), O_WRONLY | O_CREAT | O_TRUNC | O_CLOEXEC, 0400); - if (-1 == fd) - { - writeError("Cannot open a file to write the coverage data\n"); - } - else - { - if (!writeRetry(fd, reinterpret_cast(data.data()), data.size() * sizeof(data[0]))) - writeError("Cannot write the coverage data to a file\n"); - if (0 != ::close(fd)) - writeError("Cannot close the file with coverage data\n"); - } - }; - - dump(fmt::format("{}.{}", coverage_filename_prefix, getpid()), getCumulativeCoverage()); - } -} -#endif - } bool isClickhouseApp(std::string_view app_suffix, std::vector & argv) @@ -439,6 +177,14 @@ extern "C" } #endif +/// Prevent messages from JeMalloc in the release build. +/// Some of these messages are non-actionable for the users, such as: +/// : Number of CPUs detected is not deterministic. Per-CPU arena disabled. +#if USE_JEMALLOC && defined(NDEBUG) && !defined(SANITIZER) +extern "C" void (*malloc_message)(void *, const char *s); +__attribute__((constructor(0))) void init_je_malloc_message() { malloc_message = [](void *, const char *){}; } +#endif + /// This allows to implement assert to forbid initialization of a class in static constructors. /// Usage: /// diff --git a/programs/odbc-bridge/CMakeLists.txt b/programs/odbc-bridge/CMakeLists.txt index 83839cc21ac..14af330f788 100644 --- a/programs/odbc-bridge/CMakeLists.txt +++ b/programs/odbc-bridge/CMakeLists.txt @@ -13,7 +13,6 @@ set (CLICKHOUSE_ODBC_BRIDGE_SOURCES getIdentifierQuote.cpp odbc-bridge.cpp validateODBCConnectionString.cpp - createFunctionBaseCast.cpp ) clickhouse_add_executable(clickhouse-odbc-bridge ${CLICKHOUSE_ODBC_BRIDGE_SOURCES}) @@ -25,6 +24,7 @@ target_link_libraries(clickhouse-odbc-bridge PRIVATE clickhouse_parsers ch_contrib::nanodbc ch_contrib::unixodbc + clickhouse_functions ) set_target_properties(clickhouse-odbc-bridge PROPERTIES RUNTIME_OUTPUT_DIRECTORY ..) diff --git a/programs/odbc-bridge/ColumnInfoHandler.cpp b/programs/odbc-bridge/ColumnInfoHandler.cpp index 5ff985b3d12..a738d5e8c12 100644 --- a/programs/odbc-bridge/ColumnInfoHandler.cpp +++ b/programs/odbc-bridge/ColumnInfoHandler.cpp @@ -2,6 +2,7 @@ #if USE_ODBC +#include #include #include #include diff --git a/programs/odbc-bridge/IdentifierQuoteHandler.cpp b/programs/odbc-bridge/IdentifierQuoteHandler.cpp index cf5acdc4534..2f00e1171bb 100644 --- a/programs/odbc-bridge/IdentifierQuoteHandler.cpp +++ b/programs/odbc-bridge/IdentifierQuoteHandler.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include "getIdentifierQuote.h" #include "validateODBCConnectionString.h" #include "ODBCPooledConnectionFactory.h" diff --git a/programs/odbc-bridge/MainHandler.cpp b/programs/odbc-bridge/MainHandler.cpp index 2cf1576ccd7..404f66eab3b 100644 --- a/programs/odbc-bridge/MainHandler.cpp +++ b/programs/odbc-bridge/MainHandler.cpp @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include diff --git a/programs/odbc-bridge/ODBCBridge.cpp b/programs/odbc-bridge/ODBCBridge.cpp index e91cc3158df..155e26c8a27 100644 --- a/programs/odbc-bridge/ODBCBridge.cpp +++ b/programs/odbc-bridge/ODBCBridge.cpp @@ -1,5 +1,7 @@ #include "ODBCBridge.h" +#include + int mainEntryClickHouseODBCBridge(int argc, char ** argv) { DB::ODBCBridge app; diff --git a/programs/odbc-bridge/ODBCSource.cpp b/programs/odbc-bridge/ODBCSource.cpp index 940970f36ab..41a9813ce50 100644 --- a/programs/odbc-bridge/ODBCSource.cpp +++ b/programs/odbc-bridge/ODBCSource.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include @@ -47,9 +48,17 @@ Chunk ODBCSource::generate() for (int idx = 0; idx < result.columns(); ++idx) { const auto & sample = description.sample_block.getByPosition(idx); - if (!result.is_null(idx)) - insertValue(*columns[idx], removeNullable(sample.type), description.types[idx].first, result, idx); + { + if (columns[idx]->isNullable()) + { + ColumnNullable & column_nullable = assert_cast(*columns[idx]); + insertValue(column_nullable.getNestedColumn(), removeNullable(sample.type), description.types[idx].first, result, idx); + column_nullable.getNullMapData().emplace_back(0); + } + else + insertValue(*columns[idx], removeNullable(sample.type), description.types[idx].first, result, idx); + } else insertDefaultValue(*columns[idx], *sample.column); } diff --git a/programs/odbc-bridge/SchemaAllowedHandler.cpp b/programs/odbc-bridge/SchemaAllowedHandler.cpp index c7025ca4311..dfb4333240b 100644 --- a/programs/odbc-bridge/SchemaAllowedHandler.cpp +++ b/programs/odbc-bridge/SchemaAllowedHandler.cpp @@ -2,8 +2,10 @@ #if USE_ODBC +#include #include #include +#include #include #include #include diff --git a/programs/odbc-bridge/tests/CMakeLists.txt b/programs/odbc-bridge/tests/CMakeLists.txt index edf364ea192..f1411dbb554 100644 --- a/programs/odbc-bridge/tests/CMakeLists.txt +++ b/programs/odbc-bridge/tests/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable (validate-odbc-connection-string validate-odbc-connection-string.cpp ../validateODBCConnectionString.cpp) -target_link_libraries (validate-odbc-connection-string PRIVATE clickhouse_common_io) +target_link_libraries (validate-odbc-connection-string PRIVATE clickhouse_common_io clickhouse_common_config) diff --git a/programs/self-extracting/CMakeLists.txt b/programs/self-extracting/CMakeLists.txt index 4b6dd07f618..32b686d40dd 100644 --- a/programs/self-extracting/CMakeLists.txt +++ b/programs/self-extracting/CMakeLists.txt @@ -10,9 +10,24 @@ else () set (COMPRESSOR "${PROJECT_BINARY_DIR}/utils/self-extracting-executable/compressor") endif () -add_custom_target (self-extracting ALL +add_custom_target (self-extracting-server ALL ${CMAKE_COMMAND} -E remove clickhouse clickhouse-stripped COMMAND ${COMPRESSOR} ${DECOMPRESSOR} clickhouse ../clickhouse COMMAND ${COMPRESSOR} ${DECOMPRESSOR} clickhouse-stripped ../clickhouse-stripped DEPENDS clickhouse clickhouse-stripped compressor ) + +set(self_extracting_deps "self-extracting-server") + +if (BUILD_STANDALONE_KEEPER) + add_custom_target (self-extracting-keeper ALL + ${CMAKE_COMMAND} -E remove clickhouse-keeper + COMMAND ${COMPRESSOR} ${DECOMPRESSOR} clickhouse-keeper ../clickhouse-keeper + DEPENDS compressor clickhouse-keeper + ) + list(APPEND self_extracting_deps "self-extracting-keeper") +endif() + +add_custom_target (self-extracting ALL + DEPENDS ${self_extracting_deps} +) diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp index d0ebc139adb..e4bb3abdc84 100644 --- a/programs/server/Server.cpp +++ b/programs/server/Server.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -67,6 +68,7 @@ #include #include #include +#include #include #include #include @@ -133,10 +135,6 @@ # include #endif -#if USE_JEMALLOC -# include -#endif - #if USE_AZURE_BLOB_STORAGE # include # include @@ -176,34 +174,10 @@ namespace ProfileEvents namespace fs = std::filesystem; -#if USE_JEMALLOC -static bool jemallocOptionEnabled(const char *name) -{ - bool value; - size_t size = sizeof(value); - - if (mallctl(name, reinterpret_cast(&value), &size, /* newp= */ nullptr, /* newlen= */ 0)) - throw Poco::SystemException("mallctl() failed"); - - return value; -} -#else -static bool jemallocOptionEnabled(const char *) { return false; } -#endif - int mainEntryClickHouseServer(int argc, char ** argv) { DB::Server app; - if (jemallocOptionEnabled("opt.background_thread")) - { - LOG_ERROR(&app.logger(), - "jemalloc.background_thread was requested, " - "however ClickHouse uses percpu_arena and background_thread most likely will not give any benefits, " - "and also background_thread is not compatible with ClickHouse watchdog " - "(that can be disabled with CLICKHOUSE_WATCHDOG_ENABLE=0)"); - } - /// Do not fork separate process from watchdog if we attached to terminal. /// Otherwise it breaks gdb usage. /// Can be overridden by environment variable (cannot use server config at this moment). @@ -614,6 +588,54 @@ static void sanityChecks(Server & server) } } +void loadStartupScripts(const Poco::Util::AbstractConfiguration & config, ContextMutablePtr context, Poco::Logger * log) +{ + try + { + Poco::Util::AbstractConfiguration::Keys keys; + config.keys("startup_scripts", keys); + + SetResultDetailsFunc callback; + for (const auto & key : keys) + { + std::string full_prefix = "startup_scripts." + key; + + if (config.has(full_prefix + ".condition")) + { + auto condition = config.getString(full_prefix + ".condition"); + auto condition_read_buffer = ReadBufferFromString(condition); + auto condition_write_buffer = WriteBufferFromOwnString(); + + LOG_DEBUG(log, "Checking startup query condition `{}`", condition); + executeQuery(condition_read_buffer, condition_write_buffer, true, context, callback, QueryFlags{ .internal = true }, std::nullopt, {}); + + auto result = condition_write_buffer.str(); + + if (result != "1\n" && result != "true\n") + { + if (result != "0\n" && result != "false\n") + context->addWarningMessage(fmt::format("The condition query returned `{}`, which can't be interpreted as a boolean (`0`, `false`, `1`, `true`). Will skip this query.", result)); + + continue; + } + + LOG_DEBUG(log, "Condition is true, will execute the query next"); + } + + auto query = config.getString(full_prefix + ".query"); + auto read_buffer = ReadBufferFromString(query); + auto write_buffer = WriteBufferFromOwnString(); + + LOG_DEBUG(log, "Executing query `{}`", query); + executeQuery(read_buffer, write_buffer, true, context, callback, QueryFlags{ .internal = true }, std::nullopt, {}); + } + } + catch (...) + { + tryLogCurrentException(log, "Failed to parse startup scripts file"); + } +} + static void initializeAzureSDKLogger( [[ maybe_unused ]] const ServerSettings & server_settings, [[ maybe_unused ]] int server_logs_level) @@ -653,9 +675,35 @@ static void initializeAzureSDKLogger( #endif } +#if defined(SANITIZER) +static std::vector getSanitizerNames() +{ + std::vector names; + +#if defined(ADDRESS_SANITIZER) + names.push_back("address"); +#endif +#if defined(THREAD_SANITIZER) + names.push_back("thread"); +#endif +#if defined(MEMORY_SANITIZER) + names.push_back("memory"); +#endif +#if defined(UNDEFINED_BEHAVIOR_SANITIZER) + names.push_back("undefined behavior"); +#endif + + return names; +} +#endif + int Server::main(const std::vector & /*args*/) try { +#if USE_JEMALLOC + setJemallocBackgroundThreads(true); +#endif + Stopwatch startup_watch; Poco::Logger * log = &logger(); @@ -739,7 +787,17 @@ try global_context->addWarningMessage("ThreadFuzzer is enabled. Application will run slowly and unstable."); #if defined(SANITIZER) - global_context->addWarningMessage("Server was built with sanitizer. It will work slowly."); + auto sanitizers = getSanitizerNames(); + + String log_message; + if (sanitizers.empty()) + log_message = "sanitizer"; + else if (sanitizers.size() == 1) + log_message = fmt::format("{} sanitizer", sanitizers.front()); + else + log_message = fmt::format("sanitizers ({})", fmt::join(sanitizers, ", ")); + + global_context->addWarningMessage(fmt::format("Server was built with {}. It will work slowly.", log_message)); #endif #if defined(SANITIZE_COVERAGE) || WITH_COVERAGE @@ -1003,6 +1061,8 @@ try ServerUUID::load(path / "uuid", log); + PlacementInfo::PlacementInfo::instance().initialize(config()); + zkutil::validateZooKeeperConfig(config()); bool has_zookeeper = zkutil::hasZooKeeperConfig(config()); @@ -1817,11 +1877,6 @@ try } - if (config().has(DB::PlacementInfo::PLACEMENT_CONFIG_PREFIX)) - { - PlacementInfo::PlacementInfo::instance().initialize(config()); - } - { std::lock_guard lock(servers_lock); /// We should start interserver communications before (and more important shutdown after) tables. @@ -1984,6 +2039,11 @@ try /// otherwise there is a race condition between the system database initialization /// and creation of new tables in the database. waitLoad(TablesLoaderForegroundPoolId, system_startup_tasks); + + /// Startup scripts can depend on the system log tables. + if (config().has("startup_scripts") && !server_settings.prepare_system_log_tables_on_startup.changed) + global_context->setServerSetting("prepare_system_log_tables_on_startup", true); + /// After attaching system databases we can initialize system log. global_context->initializeSystemLogs(); global_context->setSystemZooKeeperLogAfterInitializationIfNeeded(); @@ -2132,6 +2192,9 @@ try load_metadata_tasks.clear(); load_metadata_tasks.shrink_to_fit(); + if (config().has("startup_scripts")) + loadStartupScripts(config(), global_context, log); + { std::lock_guard lock(servers_lock); for (auto & server : servers) @@ -2150,6 +2213,7 @@ try CannotAllocateThreadFaultInjector::setFaultProbability(server_settings.cannot_allocate_thread_fault_injection_probability); #if USE_GWP_ASAN + GWPAsan::initFinished(); GWPAsan::setForceSampleProbability(server_settings.gwp_asan_force_sample_probability); #endif @@ -2668,8 +2732,7 @@ void Server::createInterserverServers( void Server::stopServers( std::vector & servers, - const ServerType & server_type -) const + const ServerType & server_type) const { LoggerRawPtr log = &logger(); diff --git a/programs/server/Server.h b/programs/server/Server.h index 3f03dd137ef..feaf61f1ffd 100644 --- a/programs/server/Server.h +++ b/programs/server/Server.h @@ -129,8 +129,7 @@ private: void stopServers( std::vector & servers, - const ServerType & server_type - ) const; + const ServerType & server_type) const; }; } diff --git a/programs/server/dashboard.html b/programs/server/dashboard.html index b21d4b86314..45f988f7b1e 100644 --- a/programs/server/dashboard.html +++ b/programs/server/dashboard.html @@ -506,6 +506,14 @@ let user = 'default'; let password = ''; let add_http_cors_header = (location.protocol != 'file:'); +const current_url = new URL(window.location); +/// Substitute user name if it's specified in the query string +const user_from_url = current_url.searchParams.get('user'); +if (user_from_url) { + user = user_from_url; +} + + const errorCodeMessageMap = { 516: 'Error authenticating with database. Please check your connection params and try again.' } diff --git a/programs/server/play.html b/programs/server/play.html index 507a96382a7..9590a65524c 100644 --- a/programs/server/play.html +++ b/programs/server/play.html @@ -516,6 +516,9 @@ /// Save query in history only if it is different. let previous_query = ''; + /// Start of the last query + let last_query_start = 0; + const current_url = new URL(window.location); const opened_locally = location.protocol == 'file:'; @@ -567,6 +570,8 @@ '&password=' + encodeURIComponent(password) } + last_query_start = performance.now(); + const xhr = new XMLHttpRequest; xhr.open('POST', url, true); @@ -579,7 +584,8 @@ if (posted_request_num != request_num) { return; } else if (this.readyState === XMLHttpRequest.DONE) { - renderResponse(this.status, this.response); + const elapsed_msec = performance.now() - last_query_start; + renderResponse(this.status, this.response, elapsed_msec); /// The query is saved in browser history (in state JSON object) /// as well as in URL fragment identifier. @@ -587,7 +593,8 @@ const state = { query: query, status: this.status, - response: this.response.length > 100000 ? null : this.response /// Lower than the browser's limit. + response: this.response.length > 100000 ? null : this.response, /// Lower than the browser's limit. + elapsed_msec: elapsed_msec, }; const title = "ClickHouse Query: " + query; @@ -617,7 +624,7 @@ xhr.send(query); } - function renderResponse(status, response) { + function renderResponse(status, response, elapsed_msec) { document.getElementById('hourglass').style.display = 'none'; if (status === 200) { @@ -632,6 +639,7 @@ renderChart(json); } else { renderUnparsedResult(response); + stats.innerText = `Elapsed (client-side): ${(elapsed_msec / 1000).toFixed(3)} sec.`; } document.getElementById('check-mark').style.display = 'inline'; } else { @@ -651,7 +659,7 @@ clear(); return; } - renderResponse(event.state.status, event.state.response); + renderResponse(event.state.status, event.state.response, event.state.elapsed_msec); }; if (window.location.hash) { diff --git a/pyproject.toml b/pyproject.toml index 279d077a695..c89d46c0929 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -17,6 +17,8 @@ src_paths = ["src", "tests/ci", "tests/sqllogic"] [tool.pylint.'MESSAGES CONTROL'] # pytest.mark.parametrize is not callable (not-callable) disable = ''' + pointless-string-statement, + line-too-long, missing-docstring, too-few-public-methods, invalid-name, @@ -35,10 +37,10 @@ disable = ''' broad-except, bare-except, no-else-return, - global-statement + global-statement, + f-string-without-interpolation, ''' [tool.pylint.SIMILARITIES] # due to SQL min-similarity-lines=1000 - diff --git a/src/Access/AccessControl.cpp b/src/Access/AccessControl.cpp index 353358fac65..95a467bbbe5 100644 --- a/src/Access/AccessControl.cpp +++ b/src/Access/AccessControl.cpp @@ -21,7 +21,6 @@ #include #include #include -#include #include #include #include diff --git a/src/Access/Authentication.cpp b/src/Access/Authentication.cpp index f8df56516ec..6b9a6e05cf6 100644 --- a/src/Access/Authentication.cpp +++ b/src/Access/Authentication.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include "config.h" @@ -238,7 +239,15 @@ bool Authentication::areCredentialsValid( throw Authentication::Require(auth_data.getKerberosRealm()); case AuthenticationType::SSL_CERTIFICATE: - return auth_data.getSSLCertificateCommonNames().contains(ssl_certificate_credentials->getCommonName()); + for (SSLCertificateSubjects::Type type : {SSLCertificateSubjects::Type::CN, SSLCertificateSubjects::Type::SAN}) + { + for (const auto & subject : auth_data.getSSLCertificateSubjects().at(type)) + { + if (ssl_certificate_credentials->getSSLCertificateSubjects().at(type).contains(subject)) + return true; + } + } + return false; case AuthenticationType::SSH_KEY: #if USE_SSH diff --git a/src/Access/AuthenticationData.cpp b/src/Access/AuthenticationData.cpp index e9bc111e18a..5a35eeefe5b 100644 --- a/src/Access/AuthenticationData.cpp +++ b/src/Access/AuthenticationData.cpp @@ -15,6 +15,7 @@ #include #include +#include #include "config.h" #if USE_SSL @@ -107,7 +108,7 @@ bool operator ==(const AuthenticationData & lhs, const AuthenticationData & rhs) { return (lhs.type == rhs.type) && (lhs.password_hash == rhs.password_hash) && (lhs.ldap_server_name == rhs.ldap_server_name) && (lhs.kerberos_realm == rhs.kerberos_realm) - && (lhs.ssl_certificate_common_names == rhs.ssl_certificate_common_names) + && (lhs.ssl_certificate_subjects == rhs.ssl_certificate_subjects) #if USE_SSH && (lhs.ssh_keys == rhs.ssh_keys) #endif @@ -277,11 +278,16 @@ String AuthenticationData::getSalt() const return salt; } -void AuthenticationData::setSSLCertificateCommonNames(boost::container::flat_set common_names_) +void AuthenticationData::setSSLCertificateSubjects(SSLCertificateSubjects && ssl_certificate_subjects_) { - if (common_names_.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "The 'SSL CERTIFICATE' authentication type requires a non-empty list of common names."); - ssl_certificate_common_names = std::move(common_names_); + if (ssl_certificate_subjects_.empty()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "The 'SSL CERTIFICATE' authentication type requires a non-empty list of subjects."); + ssl_certificate_subjects = std::move(ssl_certificate_subjects_); +} + +void AuthenticationData::addSSLCertificateSubject(SSLCertificateSubjects::Type type_, String && subject_) +{ + ssl_certificate_subjects.insert(type_, std::move(subject_)); } std::shared_ptr AuthenticationData::toAST() const @@ -339,7 +345,14 @@ std::shared_ptr AuthenticationData::toAST() const } case AuthenticationType::SSL_CERTIFICATE: { - for (const auto & name : getSSLCertificateCommonNames()) + using SSLCertificateSubjects::Type::CN; + using SSLCertificateSubjects::Type::SAN; + + const auto &subjects = getSSLCertificateSubjects(); + SSLCertificateSubjects::Type cert_subject_type = !subjects.at(SAN).empty() ? SAN : CN; + + node->ssl_cert_subject_type = toString(cert_subject_type); + for (const auto & name : getSSLCertificateSubjects().at(cert_subject_type)) node->children.push_back(std::make_shared(name)); break; @@ -513,11 +526,9 @@ AuthenticationData AuthenticationData::fromAST(const ASTAuthenticationData & que } else if (query.type == AuthenticationType::SSL_CERTIFICATE) { - boost::container::flat_set common_names; + auto ssl_cert_subject_type = parseSSLCertificateSubjectType(*query.ssl_cert_subject_type); for (const auto & arg : args) - common_names.insert(checkAndGetLiteralArgument(arg, "common_name")); - - auth_data.setSSLCertificateCommonNames(std::move(common_names)); + auth_data.addSSLCertificateSubject(ssl_cert_subject_type, checkAndGetLiteralArgument(arg, "ssl_certificate_subject")); } else if (query.type == AuthenticationType::HTTP) { diff --git a/src/Access/AuthenticationData.h b/src/Access/AuthenticationData.h index c97e0327b56..8093fe1d888 100644 --- a/src/Access/AuthenticationData.h +++ b/src/Access/AuthenticationData.h @@ -2,13 +2,14 @@ #include #include +#include #include #include #include #include #include -#include + #include "config.h" @@ -58,8 +59,9 @@ public: const String & getKerberosRealm() const { return kerberos_realm; } void setKerberosRealm(const String & realm) { kerberos_realm = realm; } - const boost::container::flat_set & getSSLCertificateCommonNames() const { return ssl_certificate_common_names; } - void setSSLCertificateCommonNames(boost::container::flat_set common_names_); + const SSLCertificateSubjects & getSSLCertificateSubjects() const { return ssl_certificate_subjects; } + void setSSLCertificateSubjects(SSLCertificateSubjects && ssl_certificate_subjects_); + void addSSLCertificateSubject(SSLCertificateSubjects::Type type_, String && subject_); #if USE_SSH const std::vector & getSSHKeys() const { return ssh_keys; } @@ -96,7 +98,7 @@ private: Digest password_hash; String ldap_server_name; String kerberos_realm; - boost::container::flat_set ssl_certificate_common_names; + SSLCertificateSubjects ssl_certificate_subjects; String salt; #if USE_SSH std::vector ssh_keys; diff --git a/src/Access/Common/SSLCertificateSubjects.cpp b/src/Access/Common/SSLCertificateSubjects.cpp new file mode 100644 index 00000000000..ca7001a31a2 --- /dev/null +++ b/src/Access/Common/SSLCertificateSubjects.cpp @@ -0,0 +1,95 @@ +#include +#include + +#if USE_SSL +#include +#endif + +namespace DB +{ +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; +} + +#if USE_SSL +SSLCertificateSubjects extractSSLCertificateSubjects(const Poco::Net::X509Certificate & certificate) +{ + + SSLCertificateSubjects subjects; + if (!certificate.commonName().empty()) + { + subjects.insert(SSLCertificateSubjects::Type::CN, certificate.commonName()); + } + +#pragma clang diagnostic push +#pragma clang diagnostic ignored "-Wused-but-marked-unused" + auto stackof_general_name_deleter = [](void * ptr) { GENERAL_NAMES_free(static_cast(ptr)); }; + std::unique_ptr cert_names( + X509_get_ext_d2i(const_cast(certificate.certificate()), NID_subject_alt_name, nullptr, nullptr), + stackof_general_name_deleter); + + if (STACK_OF(GENERAL_NAME) * names = static_cast(cert_names.get())) + { + for (int i = 0; i < sk_GENERAL_NAME_num(names); ++i) + { + const GENERAL_NAME * name = sk_GENERAL_NAME_value(names, i); + if (name->type == GEN_DNS || name->type == GEN_URI) + { + const char * data = reinterpret_cast(ASN1_STRING_get0_data(name->d.ia5)); + std::size_t len = ASN1_STRING_length(name->d.ia5); + std::string subject = (name->type == GEN_DNS ? "DNS:" : "URI:") + std::string(data, len); + subjects.insert(SSLCertificateSubjects::Type::SAN, std::move(subject)); + } + } + } + +#pragma clang diagnostic pop + return subjects; +} +#endif + + +void SSLCertificateSubjects::insert(const String & subject_type_, String && subject) +{ + insert(parseSSLCertificateSubjectType(subject_type_), std::move(subject)); +} + +void SSLCertificateSubjects::insert(Type subject_type_, String && subject) +{ + subjects[static_cast(subject_type_)].insert(std::move(subject)); +} + +SSLCertificateSubjects::Type parseSSLCertificateSubjectType(const String & type_) +{ + if (type_ == "CN") + return SSLCertificateSubjects::Type::CN; + if (type_ == "SAN") + return SSLCertificateSubjects::Type::SAN; + + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unknown SSL Certificate Subject Type: {}", type_); +} + +String toString(SSLCertificateSubjects::Type type_) +{ + switch (type_) + { + case SSLCertificateSubjects::Type::CN: + return "CN"; + case SSLCertificateSubjects::Type::SAN: + return "SAN"; + } +} + +bool operator==(const SSLCertificateSubjects & lhs, const SSLCertificateSubjects & rhs) +{ + for (SSLCertificateSubjects::Type type : {SSLCertificateSubjects::Type::CN, SSLCertificateSubjects::Type::SAN}) + { + if (lhs.at(type) != rhs.at(type)) + return false; + } + return true; +} + +} + diff --git a/src/Access/Common/SSLCertificateSubjects.h b/src/Access/Common/SSLCertificateSubjects.h new file mode 100644 index 00000000000..ec11714d48a --- /dev/null +++ b/src/Access/Common/SSLCertificateSubjects.h @@ -0,0 +1,48 @@ +#pragma once + +#include "config.h" +#include +#include + +#if USE_SSL +# include +#endif + +namespace DB +{ +class SSLCertificateSubjects +{ +public: + using container = boost::container::flat_set; + enum class Type + { + CN, + SAN + }; + +private: + std::array subjects; + +public: + inline const container & at(Type type_) const { return subjects[static_cast(type_)]; } + inline bool empty() + { + for (auto & subject_list : subjects) + { + if (!subject_list.empty()) + return false; + } + return true; + } + void insert(const String & subject_type_, String && subject); + void insert(Type type_, String && subject); + friend bool operator==(const SSLCertificateSubjects & lhs, const SSLCertificateSubjects & rhs); +}; + +String toString(SSLCertificateSubjects::Type type_); +SSLCertificateSubjects::Type parseSSLCertificateSubjectType(const String & type_); + +#if USE_SSL +SSLCertificateSubjects extractSSLCertificateSubjects(const Poco::Net::X509Certificate & certificate); +#endif +} diff --git a/src/Access/Credentials.cpp b/src/Access/Credentials.cpp index f9886c0182b..f01700b6e46 100644 --- a/src/Access/Credentials.cpp +++ b/src/Access/Credentials.cpp @@ -1,7 +1,7 @@ #include +#include #include - namespace DB { @@ -48,18 +48,18 @@ void AlwaysAllowCredentials::setUserName(const String & user_name_) user_name = user_name_; } -SSLCertificateCredentials::SSLCertificateCredentials(const String & user_name_, const String & common_name_) +SSLCertificateCredentials::SSLCertificateCredentials(const String & user_name_, SSLCertificateSubjects && subjects_) : Credentials(user_name_) - , common_name(common_name_) + , certificate_subjects(subjects_) { is_ready = true; } -const String & SSLCertificateCredentials::getCommonName() const +const SSLCertificateSubjects & SSLCertificateCredentials::getSSLCertificateSubjects() const { if (!isReady()) throwNotReady(); - return common_name; + return certificate_subjects; } BasicCredentials::BasicCredentials() diff --git a/src/Access/Credentials.h b/src/Access/Credentials.h index d04f8a66541..5f6b0269eef 100644 --- a/src/Access/Credentials.h +++ b/src/Access/Credentials.h @@ -1,6 +1,8 @@ #pragma once #include +#include +#include #include #include "config.h" @@ -42,11 +44,11 @@ class SSLCertificateCredentials : public Credentials { public: - explicit SSLCertificateCredentials(const String & user_name_, const String & common_name_); - const String & getCommonName() const; + explicit SSLCertificateCredentials(const String & user_name_, SSLCertificateSubjects && subjects_); + const SSLCertificateSubjects & getSSLCertificateSubjects() const; private: - String common_name; + SSLCertificateSubjects certificate_subjects; }; class BasicCredentials diff --git a/src/Access/UsersConfigAccessStorage.cpp b/src/Access/UsersConfigAccessStorage.cpp index 5e36fe1ad84..a030ae96cbb 100644 --- a/src/Access/UsersConfigAccessStorage.cpp +++ b/src/Access/UsersConfigAccessStorage.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include @@ -194,18 +195,23 @@ namespace /// Fill list of allowed certificates. Poco::Util::AbstractConfiguration::Keys keys; config.keys(certificates_config, keys); - boost::container::flat_set common_names; for (const String & key : keys) { if (key.starts_with("common_name")) { String value = config.getString(certificates_config + "." + key); - common_names.insert(std::move(value)); + user->auth_data.addSSLCertificateSubject(SSLCertificateSubjects::Type::CN, std::move(value)); + } + else if (key.starts_with("subject_alt_name")) + { + String value = config.getString(certificates_config + "." + key); + if (value.empty()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected ssl_certificates.subject_alt_name to not be empty"); + user->auth_data.addSSLCertificateSubject(SSLCertificateSubjects::Type::SAN, std::move(value)); } else throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unknown certificate pattern type: {}", key); } - user->auth_data.setSSLCertificateCommonNames(std::move(common_names)); } else if (has_ssh_keys) { diff --git a/src/AggregateFunctions/AggregateFunctionAnalysisOfVariance.cpp b/src/AggregateFunctions/AggregateFunctionAnalysisOfVariance.cpp index 934a8dffd90..5d833796510 100644 --- a/src/AggregateFunctions/AggregateFunctionAnalysisOfVariance.cpp +++ b/src/AggregateFunctions/AggregateFunctionAnalysisOfVariance.cpp @@ -118,10 +118,10 @@ AggregateFunctionPtr createAggregateFunctionAnalysisOfVariance(const std::string void registerAggregateFunctionAnalysisOfVariance(AggregateFunctionFactory & factory) { AggregateFunctionProperties properties = { .is_order_dependent = false }; - factory.registerFunction("analysisOfVariance", {createAggregateFunctionAnalysisOfVariance, properties}, AggregateFunctionFactory::CaseInsensitive); + factory.registerFunction("analysisOfVariance", {createAggregateFunctionAnalysisOfVariance, properties}, AggregateFunctionFactory::Case::Insensitive); /// This is widely used term - factory.registerAlias("anova", "analysisOfVariance", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("anova", "analysisOfVariance", AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionAny.cpp b/src/AggregateFunctions/AggregateFunctionAny.cpp index f727ab04aa9..2bcee0fdd5f 100644 --- a/src/AggregateFunctions/AggregateFunctionAny.cpp +++ b/src/AggregateFunctions/AggregateFunctionAny.cpp @@ -361,9 +361,9 @@ void registerAggregateFunctionsAny(AggregateFunctionFactory & factory) AggregateFunctionProperties default_properties = {.returns_default_when_only_null = false, .is_order_dependent = true}; factory.registerFunction("any", {createAggregateFunctionAny, default_properties}); - factory.registerAlias("any_value", "any", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("first_value", "any", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("any_value", "any", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("first_value", "any", AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("anyLast", {createAggregateFunctionAnyLast, default_properties}); - factory.registerAlias("last_value", "anyLast", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("last_value", "anyLast", AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionAnyRespectNulls.cpp b/src/AggregateFunctions/AggregateFunctionAnyRespectNulls.cpp index 7275409c151..0b6642bffac 100644 --- a/src/AggregateFunctions/AggregateFunctionAnyRespectNulls.cpp +++ b/src/AggregateFunctions/AggregateFunctionAnyRespectNulls.cpp @@ -221,11 +221,11 @@ void registerAggregateFunctionsAnyRespectNulls(AggregateFunctionFactory & factor = {.returns_default_when_only_null = false, .is_order_dependent = true, .is_window_function = true}; factory.registerFunction("any_respect_nulls", {createAggregateFunctionAnyRespectNulls, default_properties_for_respect_nulls}); - factory.registerAlias("any_value_respect_nulls", "any_respect_nulls", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("first_value_respect_nulls", "any_respect_nulls", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("any_value_respect_nulls", "any_respect_nulls", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("first_value_respect_nulls", "any_respect_nulls", AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("anyLast_respect_nulls", {createAggregateFunctionAnyLastRespectNulls, default_properties_for_respect_nulls}); - factory.registerAlias("last_value_respect_nulls", "anyLast_respect_nulls", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("last_value_respect_nulls", "anyLast_respect_nulls", AggregateFunctionFactory::Case::Insensitive); /// Must happen after registering any and anyLast factory.registerNullsActionTransformation("any", "any_respect_nulls"); diff --git a/src/AggregateFunctions/AggregateFunctionAvg.cpp b/src/AggregateFunctions/AggregateFunctionAvg.cpp index ac6d2cf7fb4..57b14921c99 100644 --- a/src/AggregateFunctions/AggregateFunctionAvg.cpp +++ b/src/AggregateFunctions/AggregateFunctionAvg.cpp @@ -46,6 +46,6 @@ AggregateFunctionPtr createAggregateFunctionAvg(const std::string & name, const void registerAggregateFunctionAvg(AggregateFunctionFactory & factory) { - factory.registerFunction("avg", createAggregateFunctionAvg, AggregateFunctionFactory::CaseInsensitive); + factory.registerFunction("avg", createAggregateFunctionAvg, AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionBitwise.cpp b/src/AggregateFunctions/AggregateFunctionBitwise.cpp index 619251552e4..ecced5f3e32 100644 --- a/src/AggregateFunctions/AggregateFunctionBitwise.cpp +++ b/src/AggregateFunctions/AggregateFunctionBitwise.cpp @@ -234,9 +234,9 @@ void registerAggregateFunctionsBitwise(AggregateFunctionFactory & factory) factory.registerFunction("groupBitXor", createAggregateFunctionBitwise); /// Aliases for compatibility with MySQL. - factory.registerAlias("BIT_OR", "groupBitOr", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("BIT_AND", "groupBitAnd", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("BIT_XOR", "groupBitXor", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("BIT_OR", "groupBitOr", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("BIT_AND", "groupBitAnd", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("BIT_XOR", "groupBitXor", AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionCorr.cpp b/src/AggregateFunctions/AggregateFunctionCorr.cpp index 2e8ff3af933..02d3a4aa912 100644 --- a/src/AggregateFunctions/AggregateFunctionCorr.cpp +++ b/src/AggregateFunctions/AggregateFunctionCorr.cpp @@ -9,7 +9,7 @@ template using AggregateFunctionCorr = AggregateFunct void registerAggregateFunctionsStatisticsCorr(AggregateFunctionFactory & factory) { - factory.registerFunction("corr", createAggregateFunctionStatisticsBinary, AggregateFunctionFactory::CaseInsensitive); + factory.registerFunction("corr", createAggregateFunctionStatisticsBinary, AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionCount.cpp b/src/AggregateFunctions/AggregateFunctionCount.cpp index 25f991ab693..ad3aee90c37 100644 --- a/src/AggregateFunctions/AggregateFunctionCount.cpp +++ b/src/AggregateFunctions/AggregateFunctionCount.cpp @@ -37,7 +37,7 @@ AggregateFunctionPtr createAggregateFunctionCount(const std::string & name, cons void registerAggregateFunctionCount(AggregateFunctionFactory & factory) { AggregateFunctionProperties properties = { .returns_default_when_only_null = true, .is_order_dependent = false }; - factory.registerFunction("count", {createAggregateFunctionCount, properties}, AggregateFunctionFactory::CaseInsensitive); + factory.registerFunction("count", {createAggregateFunctionCount, properties}, AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionCovar.cpp b/src/AggregateFunctions/AggregateFunctionCovar.cpp index 9645685483f..e4877a0aed3 100644 --- a/src/AggregateFunctions/AggregateFunctionCovar.cpp +++ b/src/AggregateFunctions/AggregateFunctionCovar.cpp @@ -13,8 +13,8 @@ void registerAggregateFunctionsStatisticsCovar(AggregateFunctionFactory & factor factory.registerFunction("covarPop", createAggregateFunctionStatisticsBinary); /// Synonyms for compatibility. - factory.registerAlias("COVAR_SAMP", "covarSamp", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("COVAR_POP", "covarPop", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("COVAR_SAMP", "covarSamp", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("COVAR_POP", "covarPop", AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionFactory.cpp b/src/AggregateFunctions/AggregateFunctionFactory.cpp index 6555ae63128..082fa11ca8a 100644 --- a/src/AggregateFunctions/AggregateFunctionFactory.cpp +++ b/src/AggregateFunctions/AggregateFunctionFactory.cpp @@ -6,6 +6,7 @@ #include #include #include +#include static constexpr size_t MAX_AGGREGATE_FUNCTION_NAME_LENGTH = 1000; @@ -28,7 +29,7 @@ const String & getAggregateFunctionCanonicalNameIfAny(const String & name) return AggregateFunctionFactory::instance().getCanonicalNameIfAny(name); } -void AggregateFunctionFactory::registerFunction(const String & name, Value creator_with_properties, CaseSensitiveness case_sensitiveness) +void AggregateFunctionFactory::registerFunction(const String & name, Value creator_with_properties, Case case_sensitiveness) { if (creator_with_properties.creator == nullptr) throw Exception(ErrorCodes::LOGICAL_ERROR, "AggregateFunctionFactory: " @@ -38,7 +39,7 @@ void AggregateFunctionFactory::registerFunction(const String & name, Value creat throw Exception(ErrorCodes::LOGICAL_ERROR, "AggregateFunctionFactory: the aggregate function name '{}' is not unique", name); - if (case_sensitiveness == CaseInsensitive) + if (case_sensitiveness == Case::Insensitive) { auto key = Poco::toLower(name); if (!case_insensitive_aggregate_functions.emplace(key, creator_with_properties).second) diff --git a/src/AggregateFunctions/AggregateFunctionFactory.h b/src/AggregateFunctions/AggregateFunctionFactory.h index b1dc422fcb0..a5fa3424543 100644 --- a/src/AggregateFunctions/AggregateFunctionFactory.h +++ b/src/AggregateFunctions/AggregateFunctionFactory.h @@ -60,7 +60,7 @@ public: void registerFunction( const String & name, Value creator, - CaseSensitiveness case_sensitiveness = CaseSensitive); + Case case_sensitiveness = Case::Sensitive); /// Register how to transform from one aggregate function to other based on NullsAction /// Registers them both ways: diff --git a/src/AggregateFunctions/AggregateFunctionGroupArray.cpp b/src/AggregateFunctions/AggregateFunctionGroupArray.cpp index 16907e0f24f..7034e6373b1 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupArray.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupArray.cpp @@ -840,8 +840,8 @@ void registerAggregateFunctionGroupArray(AggregateFunctionFactory & factory) AggregateFunctionProperties properties = { .returns_default_when_only_null = false, .is_order_dependent = true }; factory.registerFunction("groupArray", { createAggregateFunctionGroupArray, properties }); - factory.registerAlias("array_agg", "groupArray", AggregateFunctionFactory::CaseInsensitive); - factory.registerAliasUnchecked("array_concat_agg", "groupArrayArray", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("array_agg", "groupArray", AggregateFunctionFactory::Case::Insensitive); + factory.registerAliasUnchecked("array_concat_agg", "groupArrayArray", AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("groupArraySample", { createAggregateFunctionGroupArraySample, properties }); factory.registerFunction("groupArrayLast", { createAggregateFunctionGroupArray, properties }); } diff --git a/src/AggregateFunctions/AggregateFunctionGroupArrayIntersect.cpp b/src/AggregateFunctions/AggregateFunctionGroupArrayIntersect.cpp index 903adf5c547..38f2fcb9fb9 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupArrayIntersect.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupArrayIntersect.cpp @@ -1,12 +1,12 @@ -#include -#include - #include #include #include #include -#include +#include +#include +#include +#include #include #include @@ -15,18 +15,14 @@ #include #include -#include -#include - #include #include -#include #include -#include -#include -#include -#include +#include +#include + +#include namespace DB @@ -51,7 +47,7 @@ struct AggregateFunctionGroupArrayIntersectData }; -/// Puts all values to the hash set. Returns an array of unique values. Implemented for numeric types. +/// Puts all values to the hash set. Returns an array of unique values present in all inputs. Implemented for numeric types. template class AggregateFunctionGroupArrayIntersect : public IAggregateFunctionDataHelper, AggregateFunctionGroupArrayIntersect> @@ -69,7 +65,7 @@ public: : IAggregateFunctionDataHelper, AggregateFunctionGroupArrayIntersect>({argument_type}, parameters_, result_type_) {} - String getName() const override { return "GroupArrayIntersect"; } + String getName() const override { return "groupArrayIntersect"; } bool allocatesMemoryInArena() const override { return false; } @@ -150,8 +146,18 @@ public: void deserialize(AggregateDataPtr __restrict place, ReadBuffer & buf, std::optional /* version */, Arena *) const override { - readVarUInt(this->data(place).version, buf); - this->data(place).value.read(buf); + auto & set = this->data(place).value; + auto & version = this->data(place).version; + size_t size; + readVarUInt(version, buf); + readVarUInt(size, buf); + set.reserve(size); + for (size_t i = 0; i < size; ++i) + { + T key; + readIntBinary(key, buf); + set.insert(key); + } } void insertResultInto(AggregateDataPtr __restrict place, IColumn & to, Arena *) const override @@ -203,7 +209,7 @@ public: : IAggregateFunctionDataHelper>({input_data_type_}, parameters_, result_type_) , input_data_type(result_type_) {} - String getName() const override { return "GroupArrayIntersect"; } + String getName() const override { return "groupArrayIntersect"; } bool allocatesMemoryInArena() const override { return true; } @@ -230,7 +236,7 @@ public: { const char * begin = nullptr; StringRef serialized = data_column->serializeValueIntoArena(offset + i, *arena, begin); - assert(serialized.data != nullptr); + chassert(serialized.data != nullptr); set.emplace(SerializedKeyHolder{serialized, *arena}, it, inserted); } } @@ -250,7 +256,7 @@ public: { const char * begin = nullptr; StringRef serialized = data_column->serializeValueIntoArena(offset + i, *arena, begin); - assert(serialized.data != nullptr); + chassert(serialized.data != nullptr); it = set.find(serialized); if (it != nullptr) @@ -292,7 +298,7 @@ public: } return new_map; }; - auto new_map = rhs_value.size() < set.size() ? create_new_map(rhs_value, set) : create_new_map(set, rhs_value); + auto new_map = create_new_map(set, rhs_value); set = std::move(new_map); } } @@ -316,11 +322,9 @@ public: readVarUInt(version, buf); readVarUInt(size, buf); set.reserve(size); - UInt64 elem_version; for (size_t i = 0; i < size; ++i) { auto key = readStringBinaryInto(*arena, buf); - readVarUInt(elem_version, buf); set.insert(key); } } diff --git a/src/AggregateFunctions/AggregateFunctionGroupConcat.cpp b/src/AggregateFunctions/AggregateFunctionGroupConcat.cpp index 1c059dc52aa..5494ef74705 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupConcat.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupConcat.cpp @@ -277,7 +277,7 @@ void registerAggregateFunctionGroupConcat(AggregateFunctionFactory & factory) AggregateFunctionProperties properties = { .returns_default_when_only_null = false, .is_order_dependent = true }; factory.registerFunction("groupConcat", { createAggregateFunctionGroupConcat, properties }); - factory.registerAlias("group_concat", "groupConcat", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("group_concat", "groupConcat", AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionKolmogorovSmirnovTest.cpp b/src/AggregateFunctions/AggregateFunctionKolmogorovSmirnovTest.cpp index 736cca11f1e..04eebe9f485 100644 --- a/src/AggregateFunctions/AggregateFunctionKolmogorovSmirnovTest.cpp +++ b/src/AggregateFunctions/AggregateFunctionKolmogorovSmirnovTest.cpp @@ -350,7 +350,7 @@ AggregateFunctionPtr createAggregateFunctionKolmogorovSmirnovTest( void registerAggregateFunctionKolmogorovSmirnovTest(AggregateFunctionFactory & factory) { - factory.registerFunction("kolmogorovSmirnovTest", createAggregateFunctionKolmogorovSmirnovTest, AggregateFunctionFactory::CaseInsensitive); + factory.registerFunction("kolmogorovSmirnovTest", createAggregateFunctionKolmogorovSmirnovTest, AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionSecondMoment.cpp b/src/AggregateFunctions/AggregateFunctionSecondMoment.cpp index 80fbe2511d9..4aa6a0a4429 100644 --- a/src/AggregateFunctions/AggregateFunctionSecondMoment.cpp +++ b/src/AggregateFunctions/AggregateFunctionSecondMoment.cpp @@ -15,11 +15,11 @@ void registerAggregateFunctionsStatisticsSecondMoment(AggregateFunctionFactory & factory.registerFunction("stddevPop", createAggregateFunctionStatisticsUnary); /// Synonyms for compatibility. - factory.registerAlias("VAR_SAMP", "varSamp", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("VAR_POP", "varPop", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("STDDEV_SAMP", "stddevSamp", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("STDDEV_POP", "stddevPop", AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("STD", "stddevPop", AggregateFunctionFactory::CaseInsensitive); + factory.registerAlias("VAR_SAMP", "varSamp", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("VAR_POP", "varPop", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("STDDEV_SAMP", "stddevSamp", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("STDDEV_POP", "stddevPop", AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("STD", "stddevPop", AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionSum.cpp b/src/AggregateFunctions/AggregateFunctionSum.cpp index e393cb6dd38..910e49f388d 100644 --- a/src/AggregateFunctions/AggregateFunctionSum.cpp +++ b/src/AggregateFunctions/AggregateFunctionSum.cpp @@ -72,7 +72,7 @@ AggregateFunctionPtr createAggregateFunctionSum(const std::string & name, const void registerAggregateFunctionSum(AggregateFunctionFactory & factory) { - factory.registerFunction("sum", createAggregateFunctionSum, AggregateFunctionFactory::CaseInsensitive); + factory.registerFunction("sum", createAggregateFunctionSum, AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("sumWithOverflow", createAggregateFunctionSum); factory.registerFunction("sumKahan", createAggregateFunctionSum); } diff --git a/src/AggregateFunctions/AggregateFunctionTopK.cpp b/src/AggregateFunctions/AggregateFunctionTopK.cpp index 26f756abe18..f949f6b7e4a 100644 --- a/src/AggregateFunctions/AggregateFunctionTopK.cpp +++ b/src/AggregateFunctions/AggregateFunctionTopK.cpp @@ -535,9 +535,9 @@ void registerAggregateFunctionTopK(AggregateFunctionFactory & factory) factory.registerFunction("topK", { createAggregateFunctionTopK, properties }); factory.registerFunction("topKWeighted", { createAggregateFunctionTopK, properties }); - factory.registerFunction("approx_top_k", { createAggregateFunctionTopK, properties }, AggregateFunctionFactory::CaseInsensitive); - factory.registerFunction("approx_top_sum", { createAggregateFunctionTopK, properties }, AggregateFunctionFactory::CaseInsensitive); - factory.registerAlias("approx_top_count", "approx_top_k", AggregateFunctionFactory::CaseInsensitive); + factory.registerFunction("approx_top_k", { createAggregateFunctionTopK, properties }, AggregateFunctionFactory::Case::Insensitive); + factory.registerFunction("approx_top_sum", { createAggregateFunctionTopK, properties }, AggregateFunctionFactory::Case::Insensitive); + factory.registerAlias("approx_top_count", "approx_top_k", AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/AggregateFunctionsMinMax.cpp b/src/AggregateFunctions/AggregateFunctionsMinMax.cpp index 03e21c15a75..5fa9a4ff5d1 100644 --- a/src/AggregateFunctions/AggregateFunctionsMinMax.cpp +++ b/src/AggregateFunctions/AggregateFunctionsMinMax.cpp @@ -195,8 +195,8 @@ AggregateFunctionPtr createAggregateFunctionMinMax( void registerAggregateFunctionsMinMax(AggregateFunctionFactory & factory) { - factory.registerFunction("min", createAggregateFunctionMinMax, AggregateFunctionFactory::CaseInsensitive); - factory.registerFunction("max", createAggregateFunctionMinMax, AggregateFunctionFactory::CaseInsensitive); + factory.registerFunction("min", createAggregateFunctionMinMax, AggregateFunctionFactory::Case::Insensitive); + factory.registerFunction("max", createAggregateFunctionMinMax, AggregateFunctionFactory::Case::Insensitive); } } diff --git a/src/AggregateFunctions/SingleValueData.cpp b/src/AggregateFunctions/SingleValueData.cpp index a14caf00f73..11931acbbc8 100644 --- a/src/AggregateFunctions/SingleValueData.cpp +++ b/src/AggregateFunctions/SingleValueData.cpp @@ -195,7 +195,7 @@ bool SingleValueDataFixed::isEqualTo(const IColumn & column, size_t index) co template bool SingleValueDataFixed::isEqualTo(const SingleValueDataFixed & to) const { - return has() && to.value == value; + return has() && to.has() && to.value == value; } template @@ -904,6 +904,7 @@ bool SingleValueDataNumeric::isEqualTo(const DB::IColumn & column, size_t ind template bool SingleValueDataNumeric::isEqualTo(const DB::SingleValueDataBase & to) const { + /// to.has() is checked in memory.get().isEqualTo auto const & other = assert_cast(to); return memory.get().isEqualTo(other.memory.get()); } @@ -917,6 +918,7 @@ void SingleValueDataNumeric::set(const DB::IColumn & column, size_t row_num, template void SingleValueDataNumeric::set(const DB::SingleValueDataBase & to, DB::Arena * arena) { + /// to.has() is checked in memory.get().set auto const & other = assert_cast(to); return memory.get().set(other.memory.get(), arena); } @@ -924,6 +926,7 @@ void SingleValueDataNumeric::set(const DB::SingleValueDataBase & to, DB::Aren template bool SingleValueDataNumeric::setIfSmaller(const DB::SingleValueDataBase & to, DB::Arena * arena) { + /// to.has() is checked in memory.get().setIfSmaller auto const & other = assert_cast(to); return memory.get().setIfSmaller(other.memory.get(), arena); } @@ -931,6 +934,7 @@ bool SingleValueDataNumeric::setIfSmaller(const DB::SingleValueDataBase & to, template bool SingleValueDataNumeric::setIfGreater(const DB::SingleValueDataBase & to, DB::Arena * arena) { + /// to.has() is checked in memory.get().setIfGreater auto const & other = assert_cast(to); return memory.get().setIfGreater(other.memory.get(), arena); } @@ -1191,7 +1195,7 @@ bool SingleValueDataString::isEqualTo(const DB::IColumn & column, size_t row_num bool SingleValueDataString::isEqualTo(const SingleValueDataBase & other) const { auto const & to = assert_cast(other); - return has() && to.getStringRef() == getStringRef(); + return has() && to.has() && to.getStringRef() == getStringRef(); } void SingleValueDataString::set(const IColumn & column, size_t row_num, Arena * arena) @@ -1291,7 +1295,7 @@ bool SingleValueDataGeneric::isEqualTo(const IColumn & column, size_t row_num) c bool SingleValueDataGeneric::isEqualTo(const DB::SingleValueDataBase & other) const { auto const & to = assert_cast(other); - return has() && to.value == value; + return has() && to.has() && to.value == value; } void SingleValueDataGeneric::set(const IColumn & column, size_t row_num, Arena *) diff --git a/src/Analyzer/IQueryTreeNode.h b/src/Analyzer/IQueryTreeNode.h index df3687f8fd9..b36c1401798 100644 --- a/src/Analyzer/IQueryTreeNode.h +++ b/src/Analyzer/IQueryTreeNode.h @@ -49,7 +49,7 @@ enum class QueryTreeNodeType : uint8_t /// Convert query tree node type to string const char * toString(QueryTreeNodeType type); -/** Query tree is semantical representation of query. +/** Query tree is a semantic representation of query. * Query tree node represent node in query tree. * IQueryTreeNode is base class for all query tree nodes. * diff --git a/src/Analyzer/InDepthQueryTreeVisitor.h b/src/Analyzer/InDepthQueryTreeVisitor.h index 62ddc06659c..67e3b75d8b5 100644 --- a/src/Analyzer/InDepthQueryTreeVisitor.h +++ b/src/Analyzer/InDepthQueryTreeVisitor.h @@ -3,7 +3,6 @@ #include #include -#include #include #include diff --git a/src/Analyzer/Passes/AggregateFunctionOfGroupByKeysPass.cpp b/src/Analyzer/Passes/AggregateFunctionOfGroupByKeysPass.cpp index a419bbcc133..b68635ab953 100644 --- a/src/Analyzer/Passes/AggregateFunctionOfGroupByKeysPass.cpp +++ b/src/Analyzer/Passes/AggregateFunctionOfGroupByKeysPass.cpp @@ -10,6 +10,8 @@ #include #include +#include + namespace DB { diff --git a/src/Analyzer/Passes/AggregateFunctionsArithmericOperationsPass.cpp b/src/Analyzer/Passes/AggregateFunctionsArithmericOperationsPass.cpp index e6798a792dd..6578a6c5bd7 100644 --- a/src/Analyzer/Passes/AggregateFunctionsArithmericOperationsPass.cpp +++ b/src/Analyzer/Passes/AggregateFunctionsArithmericOperationsPass.cpp @@ -9,6 +9,9 @@ #include #include #include +#include + +#include namespace DB { @@ -164,32 +167,15 @@ private: auto aggregate_function_clone = aggregate_function->clone(); auto & aggregate_function_clone_typed = aggregate_function_clone->as(); + aggregate_function_clone_typed.getArguments().getNodes() = { arithmetic_function_clone_argument }; - resolveAggregateFunctionNode(aggregate_function_clone_typed, arithmetic_function_clone_argument, result_aggregate_function_name); + resolveAggregateFunctionNodeByName(aggregate_function_clone_typed, result_aggregate_function_name); arithmetic_function_clone_arguments_nodes[arithmetic_function_argument_index] = std::move(aggregate_function_clone); - resolveOrdinaryFunctionNode(arithmetic_function_clone_typed, arithmetic_function_clone_typed.getFunctionName()); + resolveOrdinaryFunctionNodeByName(arithmetic_function_clone_typed, arithmetic_function_clone_typed.getFunctionName(), getContext()); return arithmetic_function_clone; } - - void resolveOrdinaryFunctionNode(FunctionNode & function_node, const String & function_name) const - { - auto function = FunctionFactory::instance().get(function_name, getContext()); - function_node.resolveAsFunction(function->build(function_node.getArgumentColumns())); - } - - static void resolveAggregateFunctionNode(FunctionNode & function_node, const QueryTreeNodePtr & argument, const String & aggregate_function_name) - { - auto function_aggregate_function = function_node.getAggregateFunction(); - - AggregateFunctionProperties properties; - auto action = NullsAction::EMPTY; - auto aggregate_function = AggregateFunctionFactory::instance().get( - aggregate_function_name, action, {argument->getResultType()}, function_aggregate_function->getParameters(), properties); - - function_node.resolveAsAggregateFunction(std::move(aggregate_function)); - } }; } diff --git a/src/Analyzer/Passes/ArrayExistsToHasPass.cpp b/src/Analyzer/Passes/ArrayExistsToHasPass.cpp index 62db502e1dc..bf4c1232d79 100644 --- a/src/Analyzer/Passes/ArrayExistsToHasPass.cpp +++ b/src/Analyzer/Passes/ArrayExistsToHasPass.cpp @@ -11,6 +11,8 @@ #include #include +#include + namespace DB { diff --git a/src/Analyzer/Passes/AutoFinalOnQueryPass.cpp b/src/Analyzer/Passes/AutoFinalOnQueryPass.cpp index 70aa1a41548..f7dd542dbf4 100644 --- a/src/Analyzer/Passes/AutoFinalOnQueryPass.cpp +++ b/src/Analyzer/Passes/AutoFinalOnQueryPass.cpp @@ -8,6 +8,8 @@ #include #include +#include + namespace DB { diff --git a/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp b/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp index ebefc12ae53..180470952cd 100644 --- a/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp +++ b/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp @@ -11,6 +11,7 @@ #include #include #include +#include namespace DB { @@ -18,19 +19,18 @@ namespace DB namespace { -class ComparisonTupleEliminationPassVisitor : public InDepthQueryTreeVisitor +class ComparisonTupleEliminationPassVisitor : public InDepthQueryTreeVisitorWithContext { public: - explicit ComparisonTupleEliminationPassVisitor(ContextPtr context_) - : context(std::move(context_)) - {} + using Base = InDepthQueryTreeVisitorWithContext; + using Base::Base; static bool needChildVisit(QueryTreeNodePtr &, QueryTreeNodePtr & child) { return child->getNodeType() != QueryTreeNodeType::TABLE_FUNCTION; } - void visitImpl(QueryTreeNodePtr & node) const + void enterImpl(QueryTreeNodePtr & node) const { auto * function_node = node->as(); if (!function_node) @@ -171,13 +171,13 @@ private: { auto result_function = std::make_shared("and"); result_function->getArguments().getNodes() = std::move(tuple_arguments_equals_functions); - resolveOrdinaryFunctionNode(*result_function, result_function->getFunctionName()); + resolveOrdinaryFunctionNodeByName(*result_function, result_function->getFunctionName(), getContext()); if (comparison_function_name == "notEquals") { auto not_function = std::make_shared("not"); not_function->getArguments().getNodes().push_back(std::move(result_function)); - resolveOrdinaryFunctionNode(*not_function, not_function->getFunctionName()); + resolveOrdinaryFunctionNodeByName(*not_function, not_function->getFunctionName(), getContext()); result_function = std::move(not_function); } @@ -197,18 +197,10 @@ private: comparison_function->getArguments().getNodes().push_back(std::move(lhs_argument)); comparison_function->getArguments().getNodes().push_back(std::move(rhs_argument)); - resolveOrdinaryFunctionNode(*comparison_function, comparison_function->getFunctionName()); + resolveOrdinaryFunctionNodeByName(*comparison_function, comparison_function->getFunctionName(), getContext()); return comparison_function; } - - void resolveOrdinaryFunctionNode(FunctionNode & function_node, const String & function_name) const - { - auto function = FunctionFactory::instance().get(function_name, context); - function_node.resolveAsFunction(function->build(function_node.getArgumentColumns())); - } - - ContextPtr context; }; } diff --git a/src/Analyzer/Passes/ConvertOrLikeChainPass.cpp b/src/Analyzer/Passes/ConvertOrLikeChainPass.cpp index eb897ef8746..2b2ac95d7b9 100644 --- a/src/Analyzer/Passes/ConvertOrLikeChainPass.cpp +++ b/src/Analyzer/Passes/ConvertOrLikeChainPass.cpp @@ -22,6 +22,8 @@ #include #include +#include + namespace DB { diff --git a/src/Analyzer/Passes/ConvertQueryToCNFPass.cpp b/src/Analyzer/Passes/ConvertQueryToCNFPass.cpp index 5951e8fc5ea..f66d092429f 100644 --- a/src/Analyzer/Passes/ConvertQueryToCNFPass.cpp +++ b/src/Analyzer/Passes/ConvertQueryToCNFPass.cpp @@ -10,6 +10,8 @@ #include #include +#include + #include #include diff --git a/src/Analyzer/Passes/CountDistinctPass.cpp b/src/Analyzer/Passes/CountDistinctPass.cpp index 3307c440f42..3f9185a07ca 100644 --- a/src/Analyzer/Passes/CountDistinctPass.cpp +++ b/src/Analyzer/Passes/CountDistinctPass.cpp @@ -9,6 +9,9 @@ #include #include #include +#include + +#include namespace DB { @@ -77,11 +80,9 @@ public: /// Replace `countDistinct` of initial query into `count` auto result_type = function_node->getResultType(); - AggregateFunctionProperties properties; - auto action = NullsAction::EMPTY; - auto aggregate_function = AggregateFunctionFactory::instance().get("count", action, {}, {}, properties); - function_node->resolveAsAggregateFunction(std::move(aggregate_function)); + function_node->getArguments().getNodes().clear(); + resolveAggregateFunctionNodeByName(*function_node, "count"); } }; diff --git a/src/Analyzer/Passes/CrossToInnerJoinPass.cpp b/src/Analyzer/Passes/CrossToInnerJoinPass.cpp index 3e2a2055fdb..eb615879893 100644 --- a/src/Analyzer/Passes/CrossToInnerJoinPass.cpp +++ b/src/Analyzer/Passes/CrossToInnerJoinPass.cpp @@ -9,13 +9,14 @@ #include #include #include +#include #include #include #include #include -#include +#include namespace DB diff --git a/src/Analyzer/Passes/FunctionToSubcolumnsPass.cpp b/src/Analyzer/Passes/FunctionToSubcolumnsPass.cpp index 15ac8d642a4..b968f43c6a6 100644 --- a/src/Analyzer/Passes/FunctionToSubcolumnsPass.cpp +++ b/src/Analyzer/Passes/FunctionToSubcolumnsPass.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -16,6 +17,11 @@ #include #include #include +#include +#include +#include + +#include namespace DB { @@ -23,202 +29,410 @@ namespace DB namespace { -class FunctionToSubcolumnsVisitor : public InDepthQueryTreeVisitorWithContext +struct ColumnContext +{ + NameAndTypePair column; + QueryTreeNodePtr column_source; + ContextPtr context; +}; + +using NodeToSubcolumnTransformer = std::function; + +void optimizeFunctionLength(QueryTreeNodePtr & node, FunctionNode &, ColumnContext & ctx) +{ + /// Replace `length(argument)` with `argument.size0` + /// `argument` may be Array or Map. + + NameAndTypePair column{ctx.column.name + ".size0", std::make_shared()}; + node = std::make_shared(column, ctx.column_source); +} + +template +void optimizeFunctionEmpty(QueryTreeNodePtr &, FunctionNode & function_node, ColumnContext & ctx) +{ + /// Replace `empty(argument)` with `equals(argument.size0, 0)` if positive + /// Replace `notEmpty(argument)` with `notEquals(argument.size0, 0)` if not positive + /// `argument` may be Array or Map. + + NameAndTypePair column{ctx.column.name + ".size0", std::make_shared()}; + auto & function_arguments_nodes = function_node.getArguments().getNodes(); + + function_arguments_nodes.clear(); + function_arguments_nodes.push_back(std::make_shared(column, ctx.column_source)); + function_arguments_nodes.push_back(std::make_shared(static_cast(0))); + + const auto * function_name = positive ? "equals" : "notEquals"; + resolveOrdinaryFunctionNodeByName(function_node, function_name, ctx.context); +} + +String getSubcolumnNameForElement(const Field & value, const DataTypeTuple & data_type_tuple) +{ + if (value.getType() == Field::Types::String) + return value.get(); + + if (value.getType() == Field::Types::UInt64) + return data_type_tuple.getNameByPosition(value.get()); + + return ""; +} + +String getSubcolumnNameForElement(const Field & value, const DataTypeVariant &) +{ + if (value.getType() == Field::Types::String) + return value.get(); + + return ""; +} + +template +void optimizeTupleOrVariantElement(QueryTreeNodePtr & node, FunctionNode & function_node, ColumnContext & ctx) +{ + /// Replace `tupleElement(tuple_argument, string_literal)`, `tupleElement(tuple_argument, integer_literal)` with `tuple_argument.column_name`. + /// Replace `variantElement(variant_argument, string_literal)` with `variant_argument.column_name`. + + auto & function_arguments_nodes = function_node.getArguments().getNodes(); + if (function_arguments_nodes.size() != 2) + return; + + const auto * second_argument_constant_node = function_arguments_nodes[1]->as(); + if (!second_argument_constant_node) + return; + + const auto & data_type_concrete = assert_cast(*ctx.column.type); + auto subcolumn_name = getSubcolumnNameForElement(second_argument_constant_node->getValue(), data_type_concrete); + + if (subcolumn_name.empty()) + return; + + NameAndTypePair column{ctx.column.name + "." + subcolumn_name, function_node.getResultType()}; + node = std::make_shared(column, ctx.column_source); +} + +std::map, NodeToSubcolumnTransformer> node_transformers = +{ + { + {TypeIndex::Array, "length"}, optimizeFunctionLength, + }, + { + {TypeIndex::Array, "empty"}, optimizeFunctionEmpty, + }, + { + {TypeIndex::Array, "notEmpty"}, optimizeFunctionEmpty, + }, + { + {TypeIndex::Map, "length"}, optimizeFunctionLength, + }, + { + {TypeIndex::Map, "empty"}, optimizeFunctionEmpty, + }, + { + {TypeIndex::Map, "notEmpty"}, optimizeFunctionEmpty, + }, + { + {TypeIndex::Map, "mapKeys"}, + [](QueryTreeNodePtr & node, FunctionNode & function_node, ColumnContext & ctx) + { + /// Replace `mapKeys(map_argument)` with `map_argument.keys` + NameAndTypePair column{ctx.column.name + ".keys", function_node.getResultType()}; + node = std::make_shared(column, ctx.column_source); + }, + }, + { + {TypeIndex::Map, "mapValues"}, + [](QueryTreeNodePtr & node, FunctionNode & function_node, ColumnContext & ctx) + { + /// Replace `mapValues(map_argument)` with `map_argument.values` + NameAndTypePair column{ctx.column.name + ".values", function_node.getResultType()}; + node = std::make_shared(column, ctx.column_source); + }, + }, + { + {TypeIndex::Map, "mapContains"}, + [](QueryTreeNodePtr &, FunctionNode & function_node, ColumnContext & ctx) + { + /// Replace `mapContains(map_argument, argument)` with `has(map_argument.keys, argument)` + const auto & data_type_map = assert_cast(*ctx.column.type); + + NameAndTypePair column{ctx.column.name + ".keys", std::make_shared(data_type_map.getKeyType())}; + auto & function_arguments_nodes = function_node.getArguments().getNodes(); + + auto has_function_argument = std::make_shared(column, ctx.column_source); + function_arguments_nodes[0] = std::move(has_function_argument); + + resolveOrdinaryFunctionNodeByName(function_node, "has", ctx.context); + }, + }, + { + {TypeIndex::Nullable, "count"}, + [](QueryTreeNodePtr &, FunctionNode & function_node, ColumnContext & ctx) + { + /// Replace `count(nullable_argument)` with `sum(not(nullable_argument.null))` + NameAndTypePair column{ctx.column.name + ".null", std::make_shared()}; + auto & function_arguments_nodes = function_node.getArguments().getNodes(); + + auto new_column_node = std::make_shared(column, ctx.column_source); + auto function_node_not = std::make_shared("not"); + + function_node_not->getArguments().getNodes().push_back(std::move(new_column_node)); + resolveOrdinaryFunctionNodeByName(*function_node_not, "not", ctx.context); + + function_arguments_nodes = {std::move(function_node_not)}; + resolveAggregateFunctionNodeByName(function_node, "sum"); + }, + }, + { + {TypeIndex::Nullable, "isNull"}, + [](QueryTreeNodePtr & node, FunctionNode &, ColumnContext & ctx) + { + /// Replace `isNull(nullable_argument)` with `nullable_argument.null` + NameAndTypePair column{ctx.column.name + ".null", std::make_shared()}; + node = std::make_shared(column, ctx.column_source); + }, + }, + { + {TypeIndex::Nullable, "isNotNull"}, + [](QueryTreeNodePtr &, FunctionNode & function_node, ColumnContext & ctx) + { + /// Replace `isNotNull(nullable_argument)` with `not(nullable_argument.null)` + NameAndTypePair column{ctx.column.name + ".null", std::make_shared()}; + auto & function_arguments_nodes = function_node.getArguments().getNodes(); + + function_arguments_nodes = {std::make_shared(column, ctx.column_source)}; + resolveOrdinaryFunctionNodeByName(function_node, "not", ctx.context); + }, + }, + { + {TypeIndex::Tuple, "tupleElement"}, optimizeTupleOrVariantElement, + }, + { + {TypeIndex::Variant, "variantElement"}, optimizeTupleOrVariantElement, + }, +}; + +std::tuple getTypedNodesForOptimization(const QueryTreeNodePtr & node) +{ + auto * function_node = node->as(); + if (!function_node) + return {}; + + auto & function_arguments_nodes = function_node->getArguments().getNodes(); + if (function_arguments_nodes.empty() || function_arguments_nodes.size() > 2) + return {}; + + auto * first_argument_column_node = function_arguments_nodes.front()->as(); + if (!first_argument_column_node || first_argument_column_node->getColumnName() == "__grouping_set") + return {}; + + auto column_source = first_argument_column_node->getColumnSource(); + auto * table_node = column_source->as(); + if (!table_node) + return {}; + + const auto & storage = table_node->getStorage(); + const auto & storage_snapshot = table_node->getStorageSnapshot(); + auto column = first_argument_column_node->getColumn(); + + if (!storage->supportsOptimizationToSubcolumns() || storage->isVirtualColumn(column.name, storage_snapshot->metadata)) + return {}; + + auto column_in_table = storage_snapshot->tryGetColumn(GetColumnsOptions::All, column.name); + if (!column_in_table || !column_in_table->type->equals(*column.type)) + return {}; + + return std::make_tuple(function_node, first_argument_column_node, table_node); +} + +/// First pass collects info about identifiers to determine which identifiers are allowed to optimize. +class FunctionToSubcolumnsVisitorFirstPass : public InDepthQueryTreeVisitorWithContext { public: - using Base = InDepthQueryTreeVisitorWithContext; + using Base = InDepthQueryTreeVisitorWithContext; using Base::Base; + void enterImpl(const QueryTreeNodePtr & node) + { + if (!getSettings().optimize_functions_to_subcolumns) + return; + + if (auto * table_node = node->as()) + { + enterImpl(*table_node); + return; + } + + if (auto * column_node = node->as()) + { + enterImpl(*column_node); + return; + } + + auto [function_node, first_argument_node, table_node] = getTypedNodesForOptimization(node); + if (function_node && first_argument_node && table_node) + { + enterImpl(*function_node, *first_argument_node, *table_node); + return; + } + + if (const auto * join_node = node->as()) + { + can_wrap_result_columns_with_nullable |= getContext()->getSettingsRef().join_use_nulls; + return; + } + + if (const auto * query_node = node->as()) + { + if (query_node->isGroupByWithCube() || query_node->isGroupByWithRollup() || query_node->isGroupByWithGroupingSets()) + can_wrap_result_columns_with_nullable |= getContext()->getSettingsRef().group_by_use_nulls; + return; + } + } + + std::unordered_set getIdentifiersToOptimize() const + { + if (can_wrap_result_columns_with_nullable) + { + /// Do not optimize if we have JOIN with setting join_use_null. + /// Do not optimize if we have GROUP BY WITH ROLLUP/CUBE/GROUPING SETS with setting group_by_use_nulls. + /// It may change the behaviour if subcolumn can be converted + /// to Nullable while the original column cannot (e.g. for Array type). + return {}; + } + + /// Do not optimize if full column is requested in other context. + /// It doesn't make sense because it doesn't reduce amount of read data + /// and optimized functions are not computation heavy. But introducing + /// new identifier complicates query analysis and may break it. + /// + /// E.g. query: + /// SELECT n FROM table GROUP BY n HAVING isNotNull(n) + /// may be optimized to incorrect query: + /// SELECT n FROM table GROUP BY n HAVING not(n.null) + /// Will produce: `n.null` is not under aggregate function and not in GROUP BY keys) + /// + /// Do not optimize index columns (primary, min-max, secondary), + /// because otherwise analysis of indexes may be broken. + /// TODO: handle subcolumns in index analysis. + + std::unordered_set identifiers_to_optimize; + for (const auto & [identifier, count] : optimized_identifiers_count) + { + if (all_key_columns.contains(identifier)) + continue; + + auto it = identifiers_count.find(identifier); + if (it != identifiers_count.end() && it->second == count) + identifiers_to_optimize.insert(identifier); + } + + return identifiers_to_optimize; + } + +private: + std::unordered_set all_key_columns; + std::unordered_map identifiers_count; + std::unordered_map optimized_identifiers_count; + + NameSet processed_tables; + bool can_wrap_result_columns_with_nullable = false; + + void enterImpl(const TableNode & table_node) + { + auto table_name = table_node.getStorage()->getStorageID().getFullTableName(); + if (processed_tables.emplace(table_name).second) + return; + + auto add_key_columns = [&](const auto & key_columns) + { + for (const auto & column_name : key_columns) + { + Identifier identifier({table_name, column_name}); + all_key_columns.insert(identifier); + } + }; + + const auto & metadata_snapshot = table_node.getStorageSnapshot()->metadata; + const auto & primary_key_columns = metadata_snapshot->getColumnsRequiredForPrimaryKey(); + const auto & partition_key_columns = metadata_snapshot->getColumnsRequiredForPartitionKey(); + + add_key_columns(primary_key_columns); + add_key_columns(partition_key_columns); + + for (const auto & index : metadata_snapshot->getSecondaryIndices()) + { + const auto & index_columns = index.expression->getRequiredColumns(); + add_key_columns(index_columns); + } + } + + void enterImpl(const ColumnNode & column_node) + { + if (column_node.getColumnName() == "__grouping_set") + return; + + auto column_source = column_node.getColumnSource(); + auto * table_node = column_source->as(); + if (!table_node) + return; + + auto table_name = table_node->getStorage()->getStorageID().getFullTableName(); + Identifier qualified_name({table_name, column_node.getColumnName()}); + + ++identifiers_count[qualified_name]; + } + + void enterImpl(const FunctionNode & function_node, const ColumnNode & first_argument_column_node, const TableNode & table_node) + { + /// For queries with FINAL converting function to subcolumn may alter + /// special merging algorithms and produce wrong result of query. + if (table_node.hasTableExpressionModifiers() && table_node.getTableExpressionModifiers()->hasFinal()) + return; + + const auto & column = first_argument_column_node.getColumn(); + auto table_name = table_node.getStorage()->getStorageID().getFullTableName(); + Identifier qualified_name({table_name, column.name}); + + if (node_transformers.contains({column.type->getTypeId(), function_node.getFunctionName()})) + ++optimized_identifiers_count[qualified_name]; + } +}; + +/// Second pass optimizes functions to subcolumns for allowed identifiers. +class FunctionToSubcolumnsVisitorSecondPass : public InDepthQueryTreeVisitorWithContext +{ +private: + std::unordered_set identifiers_to_optimize; + +public: + using Base = InDepthQueryTreeVisitorWithContext; + using Base::Base; + + FunctionToSubcolumnsVisitorSecondPass(ContextPtr context_, std::unordered_set identifiers_to_optimize_) + : Base(std::move(context_)), identifiers_to_optimize(std::move(identifiers_to_optimize_)) + { + } + void enterImpl(QueryTreeNodePtr & node) const { if (!getSettings().optimize_functions_to_subcolumns) return; - auto * function_node = node->as(); - if (!function_node) - return; - - auto & function_arguments_nodes = function_node->getArguments().getNodes(); - size_t function_arguments_nodes_size = function_arguments_nodes.size(); - - if (function_arguments_nodes.empty() || function_arguments_nodes_size > 2) - return; - - auto * first_argument_column_node = function_arguments_nodes.front()->as(); - - if (!first_argument_column_node) - return; - - if (first_argument_column_node->getColumnName() == "__grouping_set") - return; - - auto column_source = first_argument_column_node->getColumnSource(); - auto * table_node = column_source->as(); - - if (!table_node) - return; - - const auto & storage = table_node->getStorage(); - if (!storage->supportsSubcolumns()) + auto [function_node, first_argument_column_node, table_node] = getTypedNodesForOptimization(node); + if (!function_node || !first_argument_column_node || !table_node) return; auto column = first_argument_column_node->getColumn(); - WhichDataType column_type(column.type); + auto table_name = table_node->getStorage()->getStorageID().getFullTableName(); - const auto & function_name = function_node->getFunctionName(); + Identifier qualified_name({table_name, column.name}); + if (!identifiers_to_optimize.contains(qualified_name)) + return; - if (function_arguments_nodes_size == 1) + auto transformer_it = node_transformers.find({column.type->getTypeId(), function_node->getFunctionName()}); + if (transformer_it != node_transformers.end()) { - if (column_type.isArray()) - { - if (function_name == "length") - { - /// Replace `length(array_argument)` with `array_argument.size0` - column.name += ".size0"; - column.type = std::make_shared(); - - node = std::make_shared(column, column_source); - } - else if (function_name == "empty") - { - /// Replace `empty(array_argument)` with `equals(array_argument.size0, 0)` - column.name += ".size0"; - column.type = std::make_shared(); - - function_arguments_nodes.clear(); - function_arguments_nodes.push_back(std::make_shared(column, column_source)); - function_arguments_nodes.push_back(std::make_shared(static_cast(0))); - - resolveOrdinaryFunctionNode(*function_node, "equals"); - } - else if (function_name == "notEmpty") - { - /// Replace `notEmpty(array_argument)` with `notEquals(array_argument.size0, 0)` - column.name += ".size0"; - column.type = std::make_shared(); - - function_arguments_nodes.clear(); - function_arguments_nodes.push_back(std::make_shared(column, column_source)); - function_arguments_nodes.push_back(std::make_shared(static_cast(0))); - - resolveOrdinaryFunctionNode(*function_node, "notEquals"); - } - } - else if (column_type.isNullable()) - { - if (function_name == "isNull") - { - /// Replace `isNull(nullable_argument)` with `nullable_argument.null` - column.name += ".null"; - column.type = std::make_shared(); - - node = std::make_shared(column, column_source); - } - else if (function_name == "isNotNull") - { - /// Replace `isNotNull(nullable_argument)` with `not(nullable_argument.null)` - column.name += ".null"; - column.type = std::make_shared(); - - function_arguments_nodes = {std::make_shared(column, column_source)}; - - resolveOrdinaryFunctionNode(*function_node, "not"); - } - } - else if (column_type.isMap()) - { - if (function_name == "mapKeys") - { - /// Replace `mapKeys(map_argument)` with `map_argument.keys` - column.name += ".keys"; - column.type = function_node->getResultType(); - - node = std::make_shared(column, column_source); - } - else if (function_name == "mapValues") - { - /// Replace `mapValues(map_argument)` with `map_argument.values` - column.name += ".values"; - column.type = function_node->getResultType(); - - node = std::make_shared(column, column_source); - } - } + ColumnContext ctx{std::move(column), first_argument_column_node->getColumnSource(), getContext()}; + transformer_it->second(node, *function_node, ctx); } - else - { - const auto * second_argument_constant_node = function_arguments_nodes[1]->as(); - - if (function_name == "tupleElement" && column_type.isTuple() && second_argument_constant_node) - { - /** Replace `tupleElement(tuple_argument, string_literal)`, `tupleElement(tuple_argument, integer_literal)` - * with `tuple_argument.column_name`. - */ - const auto & tuple_element_constant_value = second_argument_constant_node->getValue(); - const auto & tuple_element_constant_value_type = tuple_element_constant_value.getType(); - - const auto & data_type_tuple = assert_cast(*column.type); - - String subcolumn_name; - - if (tuple_element_constant_value_type == Field::Types::String) - { - subcolumn_name = tuple_element_constant_value.get(); - } - else if (tuple_element_constant_value_type == Field::Types::UInt64) - { - auto tuple_column_index = tuple_element_constant_value.get(); - subcolumn_name = data_type_tuple.getNameByPosition(tuple_column_index); - } - else - { - return; - } - - column.name += '.'; - column.name += subcolumn_name; - column.type = function_node->getResultType(); - - node = std::make_shared(column, column_source); - } - else if (function_name == "variantElement" && isVariant(column_type) && second_argument_constant_node) - { - /// Replace `variantElement(variant_argument, type_name)` with `variant_argument.type_name`. - const auto & variant_element_constant_value = second_argument_constant_node->getValue(); - String subcolumn_name; - - if (variant_element_constant_value.getType() != Field::Types::String) - return; - - subcolumn_name = variant_element_constant_value.get(); - - column.name += '.'; - column.name += subcolumn_name; - column.type = function_node->getResultType(); - - node = std::make_shared(column, column_source); - } - else if (function_name == "mapContains" && column_type.isMap()) - { - const auto & data_type_map = assert_cast(*column.type); - - /// Replace `mapContains(map_argument, argument)` with `has(map_argument.keys, argument)` - column.name += ".keys"; - column.type = std::make_shared(data_type_map.getKeyType()); - - auto has_function_argument = std::make_shared(column, column_source); - function_arguments_nodes[0] = std::move(has_function_argument); - - resolveOrdinaryFunctionNode(*function_node, "has"); - } - } - } - -private: - void resolveOrdinaryFunctionNode(FunctionNode & function_node, const String & function_name) const - { - auto function = FunctionFactory::instance().get(function_name, getContext()); - function_node.resolveAsFunction(function->build(function_node.getArgumentColumns())); } }; @@ -226,8 +440,15 @@ private: void FunctionToSubcolumnsPass::run(QueryTreeNodePtr & query_tree_node, ContextPtr context) { - FunctionToSubcolumnsVisitor visitor(context); - visitor.visit(query_tree_node); + FunctionToSubcolumnsVisitorFirstPass first_visitor(context); + first_visitor.visit(query_tree_node); + auto identifiers_to_optimize = first_visitor.getIdentifiersToOptimize(); + + if (identifiers_to_optimize.empty()) + return; + + FunctionToSubcolumnsVisitorSecondPass second_visitor(std::move(context), std::move(identifiers_to_optimize)); + second_visitor.visit(query_tree_node); } } diff --git a/src/Analyzer/Passes/FuseFunctionsPass.cpp b/src/Analyzer/Passes/FuseFunctionsPass.cpp index ac4678fb362..0175e304a2b 100644 --- a/src/Analyzer/Passes/FuseFunctionsPass.cpp +++ b/src/Analyzer/Passes/FuseFunctionsPass.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include diff --git a/src/Analyzer/Passes/GroupingFunctionsResolvePass.cpp b/src/Analyzer/Passes/GroupingFunctionsResolvePass.cpp index e259ffa64a6..91f06183523 100644 --- a/src/Analyzer/Passes/GroupingFunctionsResolvePass.cpp +++ b/src/Analyzer/Passes/GroupingFunctionsResolvePass.cpp @@ -1,6 +1,7 @@ #include #include +#include #include diff --git a/src/Analyzer/Passes/IfChainToMultiIfPass.cpp b/src/Analyzer/Passes/IfChainToMultiIfPass.cpp index beb2247607e..cff98adb6af 100644 --- a/src/Analyzer/Passes/IfChainToMultiIfPass.cpp +++ b/src/Analyzer/Passes/IfChainToMultiIfPass.cpp @@ -4,6 +4,7 @@ #include #include +#include #include #include diff --git a/src/Analyzer/Passes/IfTransformStringsToEnumPass.cpp b/src/Analyzer/Passes/IfTransformStringsToEnumPass.cpp index 29b392a0951..d966f129d08 100644 --- a/src/Analyzer/Passes/IfTransformStringsToEnumPass.cpp +++ b/src/Analyzer/Passes/IfTransformStringsToEnumPass.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include diff --git a/src/Analyzer/Passes/LogicalExpressionOptimizerPass.cpp b/src/Analyzer/Passes/LogicalExpressionOptimizerPass.cpp index ac221bd66e7..e136440556f 100644 --- a/src/Analyzer/Passes/LogicalExpressionOptimizerPass.cpp +++ b/src/Analyzer/Passes/LogicalExpressionOptimizerPass.cpp @@ -8,8 +8,11 @@ #include #include #include +#include #include +#include +#include namespace DB { @@ -26,12 +29,145 @@ static constexpr std::array boolean_functions{ "like"sv, "notLike"sv, "ilike"sv, "notILike"sv, "empty"sv, "notEmpty"sv, "not"sv, "and"sv, "or"sv}; -static bool isBooleanFunction(const String & func_name) + +bool isBooleanFunction(const String & func_name) { return std::any_of( boolean_functions.begin(), boolean_functions.end(), [&](const auto boolean_func) { return func_name == boolean_func; }); } +bool isNodeFunction(const QueryTreeNodePtr & node, const String & func_name) +{ + if (const auto * function_node = node->as()) + return function_node->getFunctionName() == func_name; + return false; +} + +QueryTreeNodePtr getFunctionArgument(const QueryTreeNodePtr & node, size_t idx) +{ + if (const auto * function_node = node->as()) + { + const auto & args = function_node->getArguments().getNodes(); + if (idx < args.size()) + return args[idx]; + } + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected '{}' to be a function with at least {} arguments", node->formatASTForErrorMessage(), idx + 1); +} + +QueryTreeNodePtr findEqualsFunction(const QueryTreeNodes & nodes) +{ + for (const auto & node : nodes) + { + const auto * function_node = node->as(); + if (function_node && function_node->getFunctionName() == "equals" && + function_node->getArguments().getNodes().size() == 2) + { + return node; + } + } + return nullptr; +} + +/// Checks if the node is combination of isNull and notEquals functions of two the same arguments: +/// [ (a <> b AND) ] (a IS NULL) AND (b IS NULL) +bool matchIsNullOfTwoArgs(const QueryTreeNodes & nodes, QueryTreeNodePtr & lhs, QueryTreeNodePtr & rhs) +{ + QueryTreeNodePtrWithHashSet all_arguments; + QueryTreeNodePtrWithHashSet is_null_arguments; + + for (const auto & node : nodes) + { + const auto * func_node = node->as(); + if (!func_node) + return false; + + const auto & arguments = func_node->getArguments().getNodes(); + if (func_node->getFunctionName() == "isNull" && arguments.size() == 1) + { + all_arguments.insert(QueryTreeNodePtrWithHash(arguments[0])); + is_null_arguments.insert(QueryTreeNodePtrWithHash(arguments[0])); + } + + else if (func_node->getFunctionName() == "notEquals" && arguments.size() == 2) + { + if (arguments[0]->isEqual(*arguments[1])) + return false; + all_arguments.insert(QueryTreeNodePtrWithHash(arguments[0])); + all_arguments.insert(QueryTreeNodePtrWithHash(arguments[1])); + } + else + return false; + + if (all_arguments.size() > 2) + return false; + } + + if (all_arguments.size() != 2 || is_null_arguments.size() != 2) + return false; + + lhs = all_arguments.begin()->node; + rhs = std::next(all_arguments.begin())->node; + return true; +} + +bool isBooleanConstant(const QueryTreeNodePtr & node, bool expected_value) +{ + const auto * constant_node = node->as(); + if (!constant_node || !constant_node->getResultType()->equals(DataTypeUInt8())) + return false; + + UInt64 constant_value; + return (constant_node->getValue().tryGet(constant_value) && constant_value == expected_value); +} + +/// Returns true if expression consists of only conjunctions of functions with the specified name or true constants +bool isOnlyConjunctionOfFunctions( + const QueryTreeNodePtr & node, + const String & func_name, + const QueryTreeNodePtrWithHashSet & allowed_arguments) +{ + if (isBooleanConstant(node, true)) + return true; + + const auto * node_function = node->as(); + if (!node_function) + return false; + + if (node_function->getFunctionName() == func_name + && allowed_arguments.contains(node_function->getArgumentsNode())) + return true; + + if (node_function->getFunctionName() == "and") + { + for (const auto & and_argument : node_function->getArguments().getNodes()) + { + if (!isOnlyConjunctionOfFunctions(and_argument, func_name, allowed_arguments)) + return false; + } + return true; + } + return false; +} + +/// We can rewrite to a <=> b only if we are joining on a and b, +/// because the function is not yet implemented for other cases. +bool isTwoArgumentsFromDifferentSides(const FunctionNode & node_function, const JoinNode & join_node) +{ + const auto & argument_nodes = node_function.getArguments().getNodes(); + if (argument_nodes.size() != 2) + return false; + + auto first_src = getExpressionSource(argument_nodes[0]); + auto second_src = getExpressionSource(argument_nodes[1]); + if (!first_src || !second_src) + return false; + + const auto & lhs_join = *join_node.getLeftTableExpression(); + const auto & rhs_join = *join_node.getRightTableExpression(); + return (first_src->isEqual(lhs_join) && second_src->isEqual(rhs_join)) || + (first_src->isEqual(rhs_join) && second_src->isEqual(lhs_join)); +} + /// Visitor that optimizes logical expressions _only_ in JOIN ON section class JoinOnLogicalExpressionOptimizerVisitor : public InDepthQueryTreeVisitorWithContext { @@ -47,15 +183,16 @@ public: { auto * function_node = node->as(); - if (!function_node) - return; + QueryTreeNodePtr new_node = nullptr; + if (function_node && function_node->getFunctionName() == "or") + new_node = tryOptimizeJoinOnNulls(function_node->getArguments().getNodes(), getContext()); + else + new_node = tryOptimizeJoinOnNulls({node}, getContext()); - if (function_node->getFunctionName() == "or") + if (new_node) { - bool is_argument_type_changed = tryOptimizeIsNotDistinctOrIsNull(node, getContext()); - if (is_argument_type_changed) - need_rerun_resolve = true; - return; + need_rerun_resolve |= !new_node->getResultType()->equals(*node->getResultType()); + node = new_node; } } @@ -72,15 +209,11 @@ private: const JoinNode * join_node; bool need_rerun_resolve = false; - /// Returns true if type of some operand is changed and parent function needs to be re-resolved - bool tryOptimizeIsNotDistinctOrIsNull(QueryTreeNodePtr & node, const ContextPtr & context) + /// Returns optimized node or nullptr if nothing have been changed + QueryTreeNodePtr tryOptimizeJoinOnNulls(const QueryTreeNodes & nodes, const ContextPtr & context) { - auto & function_node = node->as(); - chassert(function_node.getFunctionName() == "or"); - - QueryTreeNodes or_operands; - or_operands.reserve(function_node.getArguments().getNodes().size()); + or_operands.reserve(nodes.size()); /// Indices of `equals` or `isNotDistinctFrom` functions in the vector above std::vector equals_functions_indices; @@ -93,47 +226,76 @@ private: * b => [(a IS NULL AND b IS NULL)] * c => [(a IS NULL AND c IS NULL)] * } - * Then for each a <=> b we can find all operands that contains both a IS NULL and b IS NULL + * Then for each equality a = b we can check if we have operand (a IS NULL AND b IS NULL) */ QueryTreeNodePtrWithHashMap> is_null_argument_to_indices; - for (const auto & argument : function_node.getArguments()) - { - or_operands.push_back(argument); + bool is_anything_changed = false; - auto * argument_function = argument->as(); + for (const auto & node : nodes) + { + if (isBooleanConstant(node, false)) + { + /// Remove false constants from OR + is_anything_changed = true; + continue; + } + + or_operands.push_back(node); + auto * argument_function = node->as(); if (!argument_function) continue; const auto & func_name = argument_function->getFunctionName(); if (func_name == "equals" || func_name == "isNotDistinctFrom") { - const auto & argument_nodes = argument_function->getArguments().getNodes(); - if (argument_nodes.size() != 2) - continue; - /// We can rewrite to a <=> b only if we are joining on a and b, - /// because the function is not yet implemented for other cases. - auto first_src = getExpressionSource(argument_nodes[0]); - auto second_src = getExpressionSource(argument_nodes[1]); - if (!first_src || !second_src) - continue; - const auto & lhs_join = *join_node->getLeftTableExpression(); - const auto & rhs_join = *join_node->getRightTableExpression(); - bool arguments_from_both_sides = (first_src->isEqual(lhs_join) && second_src->isEqual(rhs_join)) || - (first_src->isEqual(rhs_join) && second_src->isEqual(lhs_join)); - if (!arguments_from_both_sides) - continue; - equals_functions_indices.push_back(or_operands.size() - 1); + if (isTwoArgumentsFromDifferentSides(*argument_function, *join_node)) + equals_functions_indices.push_back(or_operands.size() - 1); } else if (func_name == "and") { - for (const auto & and_argument : argument_function->getArguments().getNodes()) + const auto & and_arguments = argument_function->getArguments().getNodes(); + + QueryTreeNodePtr is_null_lhs_arg; + QueryTreeNodePtr is_null_rhs_arg; + if (matchIsNullOfTwoArgs(and_arguments, is_null_lhs_arg, is_null_rhs_arg)) { - auto * and_argument_function = and_argument->as(); - if (and_argument_function && and_argument_function->getFunctionName() == "isNull") + is_null_argument_to_indices[is_null_lhs_arg].push_back(or_operands.size() - 1); + is_null_argument_to_indices[is_null_rhs_arg].push_back(or_operands.size() - 1); + continue; + } + + /// Expression `a = b AND (a IS NOT NULL) AND true AND (b IS NOT NULL)` we can be replaced with `a = b` + /// Even though this expression are not equivalent (first is NULL on NULLs, while second is FALSE), + /// it is still correct since for JOIN ON condition NULL is treated as FALSE + if (const auto & equals_function = findEqualsFunction(and_arguments)) + { + const auto & equals_arguments = equals_function->as()->getArguments().getNodes(); + /// Expected isNotNull arguments + QueryTreeNodePtrWithHashSet allowed_arguments; + allowed_arguments.insert(QueryTreeNodePtrWithHash(std::make_shared(QueryTreeNodes{equals_arguments[0]}))); + allowed_arguments.insert(QueryTreeNodePtrWithHash(std::make_shared(QueryTreeNodes{equals_arguments[1]}))); + + bool can_be_optimized = true; + for (const auto & and_argument : and_arguments) { - const auto & is_null_argument = and_argument_function->getArguments().getNodes()[0]; - is_null_argument_to_indices[is_null_argument].push_back(or_operands.size() - 1); + if (and_argument.get() == equals_function.get()) + continue; + + if (isOnlyConjunctionOfFunctions(and_argument, "isNotNull", allowed_arguments)) + continue; + + can_be_optimized = false; + break; + } + + if (can_be_optimized) + { + is_anything_changed = true; + or_operands.pop_back(); + or_operands.push_back(equals_function); + if (isTwoArgumentsFromDifferentSides(equals_function->as(), *join_node)) + equals_functions_indices.push_back(or_operands.size() - 1); } } } @@ -144,9 +306,9 @@ private: for (size_t equals_function_idx : equals_functions_indices) { - auto * equals_function = or_operands[equals_function_idx]->as(); + const auto * equals_function = or_operands[equals_function_idx]->as(); - /// For a <=> b we are looking for expressions containing both `a IS NULL` and `b IS NULL` combined with AND + /// For a = b we are looking for all expressions `a IS NULL AND b IS NULL` const auto & argument_nodes = equals_function->getArguments().getNodes(); const auto & lhs_is_null_parents = is_null_argument_to_indices[argument_nodes[0]]; const auto & rhs_is_null_parents = is_null_argument_to_indices[argument_nodes[1]]; @@ -161,60 +323,40 @@ private: for (size_t to_optimize_idx : operands_to_optimize) { - /// We are looking for operand `a IS NULL AND b IS NULL AND ...` - auto * operand_to_optimize = or_operands[to_optimize_idx]->as(); - - /// Remove `a IS NULL` and `b IS NULL` arguments from AND - QueryTreeNodes new_arguments; - for (const auto & and_argument : operand_to_optimize->getArguments().getNodes()) - { - bool to_eliminate = false; - - const auto * and_argument_function = and_argument->as(); - if (and_argument_function && and_argument_function->getFunctionName() == "isNull") - { - const auto & is_null_argument = and_argument_function->getArguments().getNodes()[0]; - to_eliminate = (is_null_argument->isEqual(*argument_nodes[0]) || is_null_argument->isEqual(*argument_nodes[1])); - } - - if (to_eliminate) - arguments_to_reresolve.insert(to_optimize_idx); - else - new_arguments.emplace_back(and_argument); - } - /// If less than two arguments left, we will remove or replace the whole AND below - operand_to_optimize->getArguments().getNodes() = std::move(new_arguments); + /// Remove `a IS NULL AND b IS NULL` + or_operands[to_optimize_idx] = nullptr; + is_anything_changed = true; } } - if (arguments_to_reresolve.empty()) + if (arguments_to_reresolve.empty() && !is_anything_changed) /// Nothing have been changed - return false; + return nullptr; auto and_function_resolver = FunctionFactory::instance().get("and", context); auto strict_equals_function_resolver = FunctionFactory::instance().get("isNotDistinctFrom", context); - bool need_reresolve = false; QueryTreeNodes new_or_operands; for (size_t i = 0; i < or_operands.size(); ++i) { if (arguments_to_reresolve.contains(i)) { - auto * function = or_operands[i]->as(); + const auto * function = or_operands[i]->as(); if (function->getFunctionName() == "equals") { /// We should replace `a = b` with `a <=> b` because we removed checks for IS NULL - need_reresolve |= function->getResultType()->isNullable(); - function->resolveAsFunction(strict_equals_function_resolver); - new_or_operands.emplace_back(std::move(or_operands[i])); + auto new_function = or_operands[i]->clone(); + new_function->as()->resolveAsFunction(strict_equals_function_resolver); + new_or_operands.emplace_back(std::move(new_function)); } else if (function->getFunctionName() == "and") { const auto & and_arguments = function->getArguments().getNodes(); if (and_arguments.size() > 1) { - function->resolveAsFunction(and_function_resolver); - new_or_operands.emplace_back(std::move(or_operands[i])); + auto new_function = or_operands[i]->clone(); + new_function->as()->resolveAsFunction(and_function_resolver); + new_or_operands.emplace_back(std::move(new_function)); } else if (and_arguments.size() == 1) { @@ -223,25 +365,26 @@ private: } } else - throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected function name: '{}'", function->getFunctionName()); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected function '{}'", function->getFunctionName()); } - else + else if (or_operands[i]) { new_or_operands.emplace_back(std::move(or_operands[i])); } } + if (new_or_operands.empty()) + return nullptr; + if (new_or_operands.size() == 1) - { - node = std::move(new_or_operands[0]); - return need_reresolve; - } + return new_or_operands[0]; /// Rebuild OR function auto or_function_resolver = FunctionFactory::instance().get("or", context); - function_node.getArguments().getNodes() = std::move(new_or_operands); - function_node.resolveAsFunction(or_function_resolver); - return need_reresolve; + auto function_node = std::make_shared("or"); + function_node->getArguments().getNodes() = std::move(new_or_operands); + function_node->resolveAsFunction(or_function_resolver); + return function_node; } }; @@ -519,6 +662,7 @@ private: bool is_any_nullable = false; Tuple args; args.reserve(equals_functions.size()); + DataTypes tuple_element_types; /// first we create tuple from RHS of equals functions for (const auto & equals : equals_functions) { @@ -531,16 +675,18 @@ private: if (const auto * rhs_literal = equals_arguments[1]->as()) { args.push_back(rhs_literal->getValue()); + tuple_element_types.push_back(rhs_literal->getResultType()); } else { const auto * lhs_literal = equals_arguments[0]->as(); assert(lhs_literal); args.push_back(lhs_literal->getValue()); + tuple_element_types.push_back(lhs_literal->getResultType()); } } - auto rhs_node = std::make_shared(std::move(args)); + auto rhs_node = std::make_shared(std::move(args), std::make_shared(std::move(tuple_element_types))); auto in_function = std::make_shared("in"); diff --git a/src/Analyzer/Passes/MultiIfToIfPass.cpp b/src/Analyzer/Passes/MultiIfToIfPass.cpp index c42ea61b34a..0b441032b02 100644 --- a/src/Analyzer/Passes/MultiIfToIfPass.cpp +++ b/src/Analyzer/Passes/MultiIfToIfPass.cpp @@ -2,6 +2,7 @@ #include #include +#include #include #include diff --git a/src/Analyzer/Passes/NormalizeCountVariantsPass.cpp b/src/Analyzer/Passes/NormalizeCountVariantsPass.cpp index e70e08e65f4..3a8b6e75d40 100644 --- a/src/Analyzer/Passes/NormalizeCountVariantsPass.cpp +++ b/src/Analyzer/Passes/NormalizeCountVariantsPass.cpp @@ -6,6 +6,8 @@ #include #include #include +#include +#include #include #include @@ -47,25 +49,17 @@ public: if (function_node->getFunctionName() == "count" && !first_argument_constant_literal.isNull()) { - resolveAsCountAggregateFunction(*function_node); function_node->getArguments().getNodes().clear(); + resolveAggregateFunctionNodeByName(*function_node, "count"); } else if (function_node->getFunctionName() == "sum" && first_argument_constant_literal.getType() == Field::Types::UInt64 && first_argument_constant_literal.get() == 1) { - resolveAsCountAggregateFunction(*function_node); function_node->getArguments().getNodes().clear(); + resolveAggregateFunctionNodeByName(*function_node, "count"); } } -private: - static void resolveAsCountAggregateFunction(FunctionNode & function_node) - { - AggregateFunctionProperties properties; - auto aggregate_function = AggregateFunctionFactory::instance().get("count", NullsAction::EMPTY, {}, {}, properties); - - function_node.resolveAsAggregateFunction(std::move(aggregate_function)); - } }; } diff --git a/src/Analyzer/Passes/OptimizeDateOrDateTimeConverterWithPreimagePass.cpp b/src/Analyzer/Passes/OptimizeDateOrDateTimeConverterWithPreimagePass.cpp index 0c37749c706..feb8bcc792d 100644 --- a/src/Analyzer/Passes/OptimizeDateOrDateTimeConverterWithPreimagePass.cpp +++ b/src/Analyzer/Passes/OptimizeDateOrDateTimeConverterWithPreimagePass.cpp @@ -5,9 +5,11 @@ #include #include #include +#include #include #include #include +#include namespace DB { diff --git a/src/Analyzer/Passes/OptimizeGroupByFunctionKeysPass.cpp b/src/Analyzer/Passes/OptimizeGroupByFunctionKeysPass.cpp index fd8c3e6ee6c..bfb6aa1dd07 100644 --- a/src/Analyzer/Passes/OptimizeGroupByFunctionKeysPass.cpp +++ b/src/Analyzer/Passes/OptimizeGroupByFunctionKeysPass.cpp @@ -8,6 +8,7 @@ #include #include #include +#include namespace DB { diff --git a/src/Analyzer/Passes/OptimizeGroupByInjectiveFunctionsPass.cpp b/src/Analyzer/Passes/OptimizeGroupByInjectiveFunctionsPass.cpp index a30ad2a1590..fc989003c02 100644 --- a/src/Analyzer/Passes/OptimizeGroupByInjectiveFunctionsPass.cpp +++ b/src/Analyzer/Passes/OptimizeGroupByInjectiveFunctionsPass.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include diff --git a/src/Analyzer/Passes/OptimizeRedundantFunctionsInOrderByPass.cpp b/src/Analyzer/Passes/OptimizeRedundantFunctionsInOrderByPass.cpp index b13af7f00ae..4f0271b4914 100644 --- a/src/Analyzer/Passes/OptimizeRedundantFunctionsInOrderByPass.cpp +++ b/src/Analyzer/Passes/OptimizeRedundantFunctionsInOrderByPass.cpp @@ -8,6 +8,7 @@ #include #include #include +#include namespace DB { diff --git a/src/Analyzer/Passes/RewriteAggregateFunctionWithIfPass.cpp b/src/Analyzer/Passes/RewriteAggregateFunctionWithIfPass.cpp index 3500d8327ac..c1adf05ac76 100644 --- a/src/Analyzer/Passes/RewriteAggregateFunctionWithIfPass.cpp +++ b/src/Analyzer/Passes/RewriteAggregateFunctionWithIfPass.cpp @@ -6,6 +6,8 @@ #include #include +#include + #include #include @@ -74,8 +76,7 @@ public: new_arguments[1] = std::move(if_arguments_nodes[0]); function_arguments_nodes = std::move(new_arguments); - resolveAsAggregateFunctionWithIf( - *function_node, {function_arguments_nodes[0]->getResultType(), function_arguments_nodes[1]->getResultType()}); + resolveAggregateFunctionNodeByName(*function_node, function_node->getFunctionName() + "If"); } } else if (first_const_node) @@ -104,27 +105,10 @@ public: new_arguments[1] = std::move(not_function); function_arguments_nodes = std::move(new_arguments); - resolveAsAggregateFunctionWithIf( - *function_node, {function_arguments_nodes[0]->getResultType(), function_arguments_nodes[1]->getResultType()}); + resolveAggregateFunctionNodeByName(*function_node, function_node->getFunctionName() + "If"); } } } - -private: - static void resolveAsAggregateFunctionWithIf(FunctionNode & function_node, const DataTypes & argument_types) - { - auto result_type = function_node.getResultType(); - - AggregateFunctionProperties properties; - auto aggregate_function = AggregateFunctionFactory::instance().get( - function_node.getFunctionName() + "If", - function_node.getNullsAction(), - argument_types, - function_node.getAggregateFunction()->getParameters(), - properties); - - function_node.resolveAsAggregateFunction(std::move(aggregate_function)); - } }; } diff --git a/src/Analyzer/Passes/RewriteSumFunctionWithSumAndCountPass.cpp b/src/Analyzer/Passes/RewriteSumFunctionWithSumAndCountPass.cpp index 5646d26f7f6..a42b4750f47 100644 --- a/src/Analyzer/Passes/RewriteSumFunctionWithSumAndCountPass.cpp +++ b/src/Analyzer/Passes/RewriteSumFunctionWithSumAndCountPass.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include namespace DB @@ -73,23 +74,24 @@ public: const auto lhs = std::make_shared("sum"); lhs->getArguments().getNodes().push_back(func_plus_minus_nodes[column_id]); - resolveAsAggregateFunctionNode(*lhs, column_type); + resolveAggregateFunctionNodeByName(*lhs, lhs->getFunctionName()); const auto rhs_count = std::make_shared("count"); rhs_count->getArguments().getNodes().push_back(func_plus_minus_nodes[column_id]); - resolveAsAggregateFunctionNode(*rhs_count, column_type); + resolveAggregateFunctionNodeByName(*rhs_count, rhs_count->getFunctionName()); const auto rhs = std::make_shared("multiply"); rhs->getArguments().getNodes().push_back(func_plus_minus_nodes[literal_id]); rhs->getArguments().getNodes().push_back(rhs_count); - resolveOrdinaryFunctionNode(*rhs, rhs->getFunctionName()); + resolveOrdinaryFunctionNodeByName(*rhs, rhs->getFunctionName(), getContext()); auto new_node = std::make_shared(Poco::toLower(func_plus_minus_node->getFunctionName())); if (column_id == 0) new_node->getArguments().getNodes() = {lhs, rhs}; else if (column_id == 1) new_node->getArguments().getNodes() = {rhs, lhs}; - resolveOrdinaryFunctionNode(*new_node, new_node->getFunctionName()); + + resolveOrdinaryFunctionNodeByName(*new_node, new_node->getFunctionName(), getContext()); if (!new_node) return; @@ -100,28 +102,7 @@ public: res = createCastFunction(res, function_node->getResultType(), getContext()); node = std::move(res); - } - -private: - void resolveOrdinaryFunctionNode(FunctionNode & function_node, const String & function_name) const - { - const auto function = FunctionFactory::instance().get(function_name, getContext()); - function_node.resolveAsFunction(function->build(function_node.getArgumentColumns())); - } - - static void resolveAsAggregateFunctionNode(FunctionNode & function_node, const DataTypePtr & argument_type) - { - AggregateFunctionProperties properties; - const auto aggregate_function = AggregateFunctionFactory::instance().get(function_node.getFunctionName(), - NullsAction::EMPTY, - {argument_type}, - {}, - properties); - - function_node.resolveAsAggregateFunction(aggregate_function); - } - }; } diff --git a/src/Analyzer/Passes/SumIfToCountIfPass.cpp b/src/Analyzer/Passes/SumIfToCountIfPass.cpp index 852cbe75c4a..1524629dc81 100644 --- a/src/Analyzer/Passes/SumIfToCountIfPass.cpp +++ b/src/Analyzer/Passes/SumIfToCountIfPass.cpp @@ -1,18 +1,16 @@ #include -#include -#include - #include #include - -#include - -#include - #include #include #include +#include +#include +#include +#include +#include +#include namespace DB { @@ -32,7 +30,7 @@ public: return; auto * function_node = node->as(); - if (!function_node || !function_node->isAggregateFunction()) + if (!function_node || !function_node->isAggregateFunction() || !function_node->getResultType()->equals(DataTypeUInt64())) return; auto function_name = function_node->getFunctionName(); @@ -65,7 +63,8 @@ public: auto multiplier_node = function_node_arguments_nodes[0]; function_node_arguments_nodes[0] = std::move(function_node_arguments_nodes[1]); function_node_arguments_nodes.resize(1); - resolveAsCountIfAggregateFunction(*function_node, function_node_arguments_nodes[0]->getResultType()); + + resolveAggregateFunctionNodeByName(*function_node, "countIf"); if (constant_value_literal.get() != 1) { @@ -115,7 +114,7 @@ public: function_node_arguments_nodes[0] = nested_if_function_arguments_nodes[0]; function_node_arguments_nodes.resize(1); - resolveAsCountIfAggregateFunction(*function_node, function_node_arguments_nodes[0]->getResultType()); + resolveAggregateFunctionNodeByName(*function_node, "countIf"); if (if_true_condition_value != 1) { @@ -144,7 +143,7 @@ public: function_node_arguments_nodes[0] = std::move(not_function); function_node_arguments_nodes.resize(1); - resolveAsCountIfAggregateFunction(*function_node, function_node_arguments_nodes[0]->getResultType()); + resolveAggregateFunctionNodeByName(*function_node, "countIf"); if (if_false_condition_value != 1) { @@ -156,15 +155,6 @@ public: } private: - static void resolveAsCountIfAggregateFunction(FunctionNode & function_node, const DataTypePtr & argument_type) - { - AggregateFunctionProperties properties; - auto aggregate_function = AggregateFunctionFactory::instance().get( - "countIf", NullsAction::EMPTY, {argument_type}, function_node.getAggregateFunction()->getParameters(), properties); - - function_node.resolveAsAggregateFunction(std::move(aggregate_function)); - } - QueryTreeNodePtr getMultiplyFunction(QueryTreeNodePtr left, QueryTreeNodePtr right) { auto multiply_function_node = std::make_shared("multiply"); diff --git a/src/Analyzer/Passes/UniqInjectiveFunctionsEliminationPass.cpp b/src/Analyzer/Passes/UniqInjectiveFunctionsEliminationPass.cpp index 91186db0e0c..cb8ef35fc13 100644 --- a/src/Analyzer/Passes/UniqInjectiveFunctionsEliminationPass.cpp +++ b/src/Analyzer/Passes/UniqInjectiveFunctionsEliminationPass.cpp @@ -7,6 +7,9 @@ #include #include +#include + +#include namespace DB diff --git a/src/Analyzer/Passes/UniqToCountPass.cpp b/src/Analyzer/Passes/UniqToCountPass.cpp index b801865c9a5..3f73322d1fa 100644 --- a/src/Analyzer/Passes/UniqToCountPass.cpp +++ b/src/Analyzer/Passes/UniqToCountPass.cpp @@ -7,6 +7,9 @@ #include #include #include +#include + +#include namespace DB { @@ -184,11 +187,8 @@ public: /// Replace uniq of initial query to count if (match_subquery_with_distinct() || match_subquery_with_group_by()) { - AggregateFunctionProperties properties; - auto aggregate_function = AggregateFunctionFactory::instance().get("count", NullsAction::EMPTY, {}, {}, properties); - function_node->getArguments().getNodes().clear(); - function_node->resolveAsAggregateFunction(std::move(aggregate_function)); + resolveAggregateFunctionNodeByName(*function_node, "count"); } } }; diff --git a/src/Analyzer/QueryTreeBuilder.cpp b/src/Analyzer/QueryTreeBuilder.cpp index 6a5db4bc1de..a62b6e56ac5 100644 --- a/src/Analyzer/QueryTreeBuilder.cpp +++ b/src/Analyzer/QueryTreeBuilder.cpp @@ -40,6 +40,8 @@ #include #include +#include + #include #include @@ -940,6 +942,7 @@ QueryTreeNodePtr QueryTreeBuilder::buildJoinTree(const ASTPtr & tables_in_select table_join.locality, result_join_strictness, result_join_kind); + join_node->setOriginalAST(table_element.table_join); /** Original AST is not set because it will contain only join part and does * not include left table expression. diff --git a/src/Analyzer/QueryTreePassManager.cpp b/src/Analyzer/QueryTreePassManager.cpp index f7919b6422c..4443f83596f 100644 --- a/src/Analyzer/QueryTreePassManager.cpp +++ b/src/Analyzer/QueryTreePassManager.cpp @@ -62,7 +62,7 @@ namespace ErrorCodes namespace { -#if defined(ABORT_ON_LOGICAL_ERROR) +#if defined(DEBUG_OR_SANITIZER_BUILD) /** This visitor checks if Query Tree structure is valid after each pass * in debug build. @@ -183,7 +183,7 @@ void QueryTreePassManager::run(QueryTreeNodePtr query_tree_node) for (size_t i = 0; i < passes_size; ++i) { passes[i]->run(query_tree_node, current_context); -#if defined(ABORT_ON_LOGICAL_ERROR) +#if defined(DEBUG_OR_SANITIZER_BUILD) ValidationChecker(passes[i]->getName()).visit(query_tree_node); #endif } @@ -208,7 +208,7 @@ void QueryTreePassManager::run(QueryTreeNodePtr query_tree_node, size_t up_to_pa for (size_t i = 0; i < up_to_pass_index; ++i) { passes[i]->run(query_tree_node, current_context); -#if defined(ABORT_ON_LOGICAL_ERROR) +#if defined(DEBUG_OR_SANITIZER_BUILD) ValidationChecker(passes[i]->getName()).visit(query_tree_node); #endif } diff --git a/src/Analyzer/Resolve/IdentifierResolveScope.cpp b/src/Analyzer/Resolve/IdentifierResolveScope.cpp index ae363b57047..eb3e2179440 100644 --- a/src/Analyzer/Resolve/IdentifierResolveScope.cpp +++ b/src/Analyzer/Resolve/IdentifierResolveScope.cpp @@ -1,8 +1,9 @@ #include -#include #include #include +#include +#include namespace DB { diff --git a/src/Analyzer/Resolve/IdentifierResolver.cpp b/src/Analyzer/Resolve/IdentifierResolver.cpp index 692a31b66ba..447bf825836 100644 --- a/src/Analyzer/Resolve/IdentifierResolver.cpp +++ b/src/Analyzer/Resolve/IdentifierResolver.cpp @@ -1,6 +1,7 @@ #include #include #include +#include #include #include @@ -25,6 +26,8 @@ #include #include +#include + namespace DB { namespace ErrorCodes @@ -678,9 +681,33 @@ QueryTreeNodePtr IdentifierResolver::tryResolveIdentifierFromStorage( bool match_full_identifier = false; const auto & identifier_full_name = identifier_without_column_qualifier.getFullName(); - auto it = table_expression_data.column_name_to_column_node.find(identifier_full_name); - bool can_resolve_directly_from_storage = it != table_expression_data.column_name_to_column_node.end(); - if (can_resolve_directly_from_storage && table_expression_data.subcolumn_names.contains(identifier_full_name)) + + ColumnNodePtr result_column_node; + bool can_resolve_directly_from_storage = false; + bool is_subcolumn = false; + if (auto it = table_expression_data.column_name_to_column_node.find(identifier_full_name); it != table_expression_data.column_name_to_column_node.end()) + { + can_resolve_directly_from_storage = true; + is_subcolumn = table_expression_data.subcolumn_names.contains(identifier_full_name); + result_column_node = it->second; + } + /// Check if it's a dynamic subcolumn + else + { + auto [column_name, dynamic_subcolumn_name] = Nested::splitName(identifier_full_name); + auto jt = table_expression_data.column_name_to_column_node.find(column_name); + if (jt != table_expression_data.column_name_to_column_node.end() && jt->second->getColumnType()->hasDynamicSubcolumns()) + { + if (auto dynamic_subcolumn_type = jt->second->getColumnType()->tryGetSubcolumnType(dynamic_subcolumn_name)) + { + result_column_node = std::make_shared(NameAndTypePair{identifier_full_name, dynamic_subcolumn_type}, jt->second->getColumnSource()); + can_resolve_directly_from_storage = true; + is_subcolumn = true; + } + } + } + + if (can_resolve_directly_from_storage && is_subcolumn) { /** In the case when we have an ARRAY JOIN, we should not resolve subcolumns directly from storage. * For example, consider the following SQL query: @@ -696,11 +723,11 @@ QueryTreeNodePtr IdentifierResolver::tryResolveIdentifierFromStorage( if (can_resolve_directly_from_storage) { match_full_identifier = true; - result_expression = it->second; + result_expression = result_column_node; } else { - it = table_expression_data.column_name_to_column_node.find(identifier_without_column_qualifier.at(0)); + auto it = table_expression_data.column_name_to_column_node.find(identifier_without_column_qualifier.at(0)); if (it != table_expression_data.column_name_to_column_node.end()) result_expression = it->second; } diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index 576c4943ccb..b1fe2554988 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -64,6 +64,8 @@ #include #include +#include + namespace ProfileEvents { extern const Event ScalarSubqueriesGlobalCacheHit; @@ -1490,6 +1492,10 @@ void QueryAnalyzer::qualifyColumnNodesWithProjectionNames(const QueryTreeNodes & additional_column_qualification_parts = {table_expression_node->getAlias()}; else if (auto * table_node = table_expression_node->as()) additional_column_qualification_parts = {table_node->getStorageID().getDatabaseName(), table_node->getStorageID().getTableName()}; + else if (auto * query_node = table_expression_node->as(); query_node && query_node->isCTE()) + additional_column_qualification_parts = {query_node->getCTEName()}; + else if (auto * union_node = table_expression_node->as(); union_node && union_node->isCTE()) + additional_column_qualification_parts = {union_node->getCTEName()}; size_t additional_column_qualification_parts_size = additional_column_qualification_parts.size(); const auto & table_expression_data = scope.getTableExpressionDataOrThrow(table_expression_node); @@ -2913,6 +2919,17 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi resolveExpressionNode(in_second_argument, scope, false /*allow_lambda_expression*/, true /*allow_table_expression*/); } + + /// Edge case when the first argument of IN is scalar subquery. + auto & in_first_argument = function_in_arguments_nodes[0]; + auto first_argument_type = in_first_argument->getNodeType(); + if (first_argument_type == QueryTreeNodeType::QUERY || first_argument_type == QueryTreeNodeType::UNION) + { + IdentifierResolveScope subquery_scope(in_first_argument, &scope /*parent_scope*/); + subquery_scope.subquery_depth = scope.subquery_depth + 1; + + evaluateScalarSubqueryIfNeeded(in_first_argument, subquery_scope); + } } /// Initialize function argument columns @@ -3410,14 +3427,14 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi function_base = function->build(argument_columns); /// Do not constant fold get scalar functions - bool disable_constant_folding = function_name == "__getScalar" || function_name == "shardNum" || - function_name == "shardCount" || function_name == "hostName" || function_name == "tcpPort"; + // bool disable_constant_folding = function_name == "__getScalar" || function_name == "shardNum" || + // function_name == "shardCount" || function_name == "hostName" || function_name == "tcpPort"; /** If function is suitable for constant folding try to convert it to constant. * Example: SELECT plus(1, 1); * Result: SELECT 2; */ - if (function_base->isSuitableForConstantFolding() && !disable_constant_folding) + if (function_base->isSuitableForConstantFolding()) // && !disable_constant_folding) { auto result_type = function_base->getResultType(); auto executable_function = function_base->prepare(argument_columns); @@ -3826,6 +3843,10 @@ ProjectionNames QueryAnalyzer::resolveExpressionNode( node->convertToNullable(); break; } + + /// Check parent scopes until find current query scope. + if (scope_ptr->scope_node->getNodeType() == QueryTreeNodeType::QUERY) + break; } } @@ -4455,9 +4476,8 @@ void QueryAnalyzer::initializeTableExpressionData(const QueryTreeNodePtr & table { auto left_table_expression = extractLeftTableExpression(scope_query_node->getJoinTree()); if (table_expression_node.get() == left_table_expression.get() && - scope.joins_count == 1 && - scope.context->getSettingsRef().single_join_prefer_left_table) - table_expression_data.should_qualify_columns = false; + scope.joins_count == 1 && scope.context->getSettingsRef().single_join_prefer_left_table) + table_expression_data.should_qualify_columns = false; } scope.table_expression_node_to_data.emplace(table_expression_node, std::move(table_expression_data)); diff --git a/src/Analyzer/TableNode.cpp b/src/Analyzer/TableNode.cpp index 1c4d022542d..d5a4b06f652 100644 --- a/src/Analyzer/TableNode.cpp +++ b/src/Analyzer/TableNode.cpp @@ -10,6 +10,8 @@ #include +#include + namespace DB { diff --git a/src/Analyzer/Utils.cpp b/src/Analyzer/Utils.cpp index 3c3489681f6..d10bbd9bd23 100644 --- a/src/Analyzer/Utils.cpp +++ b/src/Analyzer/Utils.cpp @@ -1,5 +1,7 @@ #include +#include + #include #include #include @@ -636,16 +638,16 @@ private: bool has_function = false; }; -inline AggregateFunctionPtr resolveAggregateFunction(FunctionNode * function_node) +inline AggregateFunctionPtr resolveAggregateFunction(FunctionNode & function_node, const String & function_name) { Array parameters; - for (const auto & param : function_node->getParameters()) + for (const auto & param : function_node.getParameters()) { auto * constant = param->as(); parameters.push_back(constant->getValue()); } - const auto & function_node_argument_nodes = function_node->getArguments().getNodes(); + const auto & function_node_argument_nodes = function_node.getArguments().getNodes(); DataTypes argument_types; argument_types.reserve(function_node_argument_nodes.size()); @@ -655,7 +657,7 @@ inline AggregateFunctionPtr resolveAggregateFunction(FunctionNode * function_nod AggregateFunctionProperties properties; auto action = NullsAction::EMPTY; - return AggregateFunctionFactory::instance().get(function_node->getFunctionName(), action, argument_types, parameters, properties); + return AggregateFunctionFactory::instance().get(function_name, action, argument_types, parameters, properties); } } @@ -736,11 +738,11 @@ void rerunFunctionResolve(FunctionNode * function_node, ContextPtr context) { if (name == "nothing" || name == "nothingUInt64" || name == "nothingNull") return; - function_node->resolveAsAggregateFunction(resolveAggregateFunction(function_node)); + function_node->resolveAsAggregateFunction(resolveAggregateFunction(*function_node, function_node->getFunctionName())); } else if (function_node->isWindowFunction()) { - function_node->resolveAsWindowFunction(resolveAggregateFunction(function_node)); + function_node->resolveAsWindowFunction(resolveAggregateFunction(*function_node, function_node->getFunctionName())); } } @@ -793,6 +795,18 @@ QueryTreeNodePtr createCastFunction(QueryTreeNodePtr node, DataTypePtr result_ty return function_node; } +void resolveOrdinaryFunctionNodeByName(FunctionNode & function_node, const String & function_name, const ContextPtr & context) +{ + auto function = FunctionFactory::instance().get(function_name, context); + function_node.resolveAsFunction(function->build(function_node.getArgumentColumns())); +} + +void resolveAggregateFunctionNodeByName(FunctionNode & function_node, const String & function_name) +{ + auto aggregate_function = resolveAggregateFunction(function_node, function_name); + function_node.resolveAsAggregateFunction(std::move(aggregate_function)); +} + /** Returns: * {_, false} - multiple sources * {nullptr, true} - no sources (for constants) diff --git a/src/Analyzer/Utils.h b/src/Analyzer/Utils.h index f64b724abeb..f2e2c500384 100644 --- a/src/Analyzer/Utils.h +++ b/src/Analyzer/Utils.h @@ -112,6 +112,14 @@ NameSet collectIdentifiersFullNames(const QueryTreeNodePtr & node); /// Wrap node into `_CAST` function QueryTreeNodePtr createCastFunction(QueryTreeNodePtr node, DataTypePtr result_type, ContextPtr context); +/// Resolves function node as ordinary function with given name. +/// Arguments and parameters are taken from the node. +void resolveOrdinaryFunctionNodeByName(FunctionNode & function_node, const String & function_name, const ContextPtr & context); + +/// Resolves function node as aggregate function with given name. +/// Arguments and parameters are taken from the node. +void resolveAggregateFunctionNodeByName(FunctionNode & function_node, const String & function_name); + /// Checks that node has only one source and returns it QueryTreeNodePtr getExpressionSource(const QueryTreeNodePtr & node); diff --git a/src/Backups/BackupEntriesCollector.cpp b/src/Backups/BackupEntriesCollector.cpp index 64ab2112326..3ee7a05600e 100644 --- a/src/Backups/BackupEntriesCollector.cpp +++ b/src/Backups/BackupEntriesCollector.cpp @@ -17,6 +17,7 @@ #include #include #include +#include #include #include diff --git a/src/Backups/BackupIO_AzureBlobStorage.cpp b/src/Backups/BackupIO_AzureBlobStorage.cpp index 44e7808babc..cee41861d70 100644 --- a/src/Backups/BackupIO_AzureBlobStorage.cpp +++ b/src/Backups/BackupIO_AzureBlobStorage.cpp @@ -29,48 +29,49 @@ namespace ErrorCodes } BackupReaderAzureBlobStorage::BackupReaderAzureBlobStorage( - const StorageAzureConfiguration & configuration_, + const AzureBlobStorage::ConnectionParams & connection_params_, + const String & blob_path_, bool allow_azure_native_copy, const ReadSettings & read_settings_, const WriteSettings & write_settings_, const ContextPtr & context_) : BackupReaderDefault(read_settings_, write_settings_, getLogger("BackupReaderAzureBlobStorage")) - , data_source_description{DataSourceType::ObjectStorage, ObjectStorageType::Azure, MetadataStorageType::None, configuration_.getConnectionURL().toString(), false, false} - , configuration(configuration_) + , data_source_description{DataSourceType::ObjectStorage, ObjectStorageType::Azure, MetadataStorageType::None, connection_params_.getConnectionURL(), false, false} + , connection_params(connection_params_) + , blob_path(blob_path_) { - auto client_ptr = configuration.createClient(/* is_readonly */false, /* attempt_to_create_container */true); - client_ptr->SetClickhouseOptions(Azure::Storage::Blobs::ClickhouseClientOptions{.IsClientForDisk=true}); + auto client_ptr = AzureBlobStorage::getContainerClient(connection_params, /*readonly=*/ false); + auto settings_ptr = AzureBlobStorage::getRequestSettingsForBackup(context_->getSettingsRef(), allow_azure_native_copy); - object_storage = std::make_unique("BackupReaderAzureBlobStorage", - std::move(client_ptr), - configuration.createSettings(context_), - configuration_.container, - configuration.getConnectionURL().toString()); + object_storage = std::make_unique( + "BackupReaderAzureBlobStorage", + std::move(client_ptr), + std::move(settings_ptr), + connection_params.getContainer(), + connection_params.getConnectionURL()); client = object_storage->getAzureBlobStorageClient(); - auto settings_copy = *object_storage->getSettings(); - settings_copy.use_native_copy = allow_azure_native_copy; - settings = std::make_unique(settings_copy); + settings = object_storage->getSettings(); } BackupReaderAzureBlobStorage::~BackupReaderAzureBlobStorage() = default; bool BackupReaderAzureBlobStorage::fileExists(const String & file_name) { - String key = fs::path(configuration.blob_path) / file_name; + String key = fs::path(blob_path) / file_name; return object_storage->exists(StoredObject(key)); } UInt64 BackupReaderAzureBlobStorage::getFileSize(const String & file_name) { - String key = fs::path(configuration.blob_path) / file_name; + String key = fs::path(blob_path) / file_name; ObjectMetadata object_metadata = object_storage->getObjectMetadata(key); return object_metadata.size_bytes; } std::unique_ptr BackupReaderAzureBlobStorage::readFile(const String & file_name) { - String key = fs::path(configuration.blob_path) / file_name; + String key = fs::path(blob_path) / file_name; return std::make_unique( client, key, read_settings, settings->max_single_read_retries, settings->max_single_download_retries); @@ -85,23 +86,23 @@ void BackupReaderAzureBlobStorage::copyFileToDisk(const String & path_in_backup, && destination_data_source_description.is_encrypted == encrypted_in_backup) { LOG_TRACE(log, "Copying {} from AzureBlobStorage to disk {}", path_in_backup, destination_disk->getName()); - auto write_blob_function = [&](const Strings & blob_path, WriteMode mode, const std::optional &) -> size_t + auto write_blob_function = [&](const Strings & dst_blob_path, WriteMode mode, const std::optional &) -> size_t { /// Object storage always uses mode `Rewrite` because it simulates append using metadata and different files. - if (blob_path.size() != 2 || mode != WriteMode::Rewrite) + if (dst_blob_path.size() != 2 || mode != WriteMode::Rewrite) throw Exception(ErrorCodes::LOGICAL_ERROR, "Blob writing function called with unexpected blob_path.size={} or mode={}", - blob_path.size(), mode); + dst_blob_path.size(), mode); copyAzureBlobStorageFile( client, destination_disk->getObjectStorage()->getAzureBlobStorageClient(), - configuration.container, - fs::path(configuration.blob_path) / path_in_backup, + connection_params.getContainer(), + fs::path(blob_path) / path_in_backup, 0, file_size, - /* dest_container */ blob_path[1], - /* dest_path */ blob_path[0], + /* dest_container */ dst_blob_path[1], + /* dest_path */ dst_blob_path[0], settings, read_settings, threadPoolCallbackRunnerUnsafe(getBackupsIOThreadPool().get(), "BackupRDAzure")); @@ -119,28 +120,33 @@ void BackupReaderAzureBlobStorage::copyFileToDisk(const String & path_in_backup, BackupWriterAzureBlobStorage::BackupWriterAzureBlobStorage( - const StorageAzureConfiguration & configuration_, + const AzureBlobStorage::ConnectionParams & connection_params_, + const String & blob_path_, bool allow_azure_native_copy, const ReadSettings & read_settings_, const WriteSettings & write_settings_, const ContextPtr & context_, bool attempt_to_create_container) : BackupWriterDefault(read_settings_, write_settings_, getLogger("BackupWriterAzureBlobStorage")) - , data_source_description{DataSourceType::ObjectStorage, ObjectStorageType::Azure, MetadataStorageType::None, configuration_.getConnectionURL().toString(), false, false} - , configuration(configuration_) + , data_source_description{DataSourceType::ObjectStorage, ObjectStorageType::Azure, MetadataStorageType::None, connection_params_.getConnectionURL(), false, false} + , connection_params(connection_params_) + , blob_path(blob_path_) { - auto client_ptr = configuration.createClient(/* is_readonly */false, attempt_to_create_container); - client_ptr->SetClickhouseOptions(Azure::Storage::Blobs::ClickhouseClientOptions{.IsClientForDisk=true}); + if (!attempt_to_create_container) + connection_params.endpoint.container_already_exists = true; + + auto client_ptr = AzureBlobStorage::getContainerClient(connection_params, /*readonly=*/ false); + auto settings_ptr = AzureBlobStorage::getRequestSettingsForBackup(context_->getSettingsRef(), allow_azure_native_copy); + + object_storage = std::make_unique( + "BackupWriterAzureBlobStorage", + std::move(client_ptr), + std::move(settings_ptr), + connection_params.getContainer(), + connection_params.getConnectionURL()); - object_storage = std::make_unique("BackupWriterAzureBlobStorage", - std::move(client_ptr), - configuration.createSettings(context_), - configuration.container, - configuration_.getConnectionURL().toString()); client = object_storage->getAzureBlobStorageClient(); - auto settings_copy = *object_storage->getSettings(); - settings_copy.use_native_copy = allow_azure_native_copy; - settings = std::make_unique(settings_copy); + settings = object_storage->getSettings(); } void BackupWriterAzureBlobStorage::copyFileFromDisk( @@ -159,18 +165,18 @@ void BackupWriterAzureBlobStorage::copyFileFromDisk( { /// getBlobPath() can return more than 3 elements if the file is stored as multiple objects in AzureBlobStorage container. /// In this case we can't use the native copy. - if (auto blob_path = src_disk->getBlobPath(src_path); blob_path.size() == 2) + if (auto src_blob_path = src_disk->getBlobPath(src_path); src_blob_path.size() == 2) { LOG_TRACE(log, "Copying file {} from disk {} to AzureBlobStorag", src_path, src_disk->getName()); copyAzureBlobStorageFile( src_disk->getObjectStorage()->getAzureBlobStorageClient(), client, - /* src_container */ blob_path[1], - /* src_path */ blob_path[0], + /* src_container */ src_blob_path[1], + /* src_path */ src_blob_path[0], start_pos, length, - configuration.container, - fs::path(configuration.blob_path) / path_in_backup, + connection_params.getContainer(), + fs::path(blob_path) / path_in_backup, settings, read_settings, threadPoolCallbackRunnerUnsafe(getBackupsIOThreadPool().get(), "BackupWRAzure")); @@ -188,11 +194,11 @@ void BackupWriterAzureBlobStorage::copyFile(const String & destination, const St copyAzureBlobStorageFile( client, client, - configuration.container, - fs::path(configuration.blob_path)/ source, + connection_params.getContainer(), + fs::path(blob_path)/ source, 0, size, - /* dest_container */ configuration.container, + /* dest_container */ connection_params.getContainer(), /* dest_path */ destination, settings, read_settings, @@ -206,22 +212,28 @@ void BackupWriterAzureBlobStorage::copyDataToFile( UInt64 length) { copyDataToAzureBlobStorageFile( - create_read_buffer, start_pos, length, client, configuration.container, - fs::path(configuration.blob_path) / path_in_backup, settings, - threadPoolCallbackRunnerUnsafe(getBackupsIOThreadPool().get(), "BackupWRAzure")); + create_read_buffer, + start_pos, + length, + client, + connection_params.getContainer(), + fs::path(blob_path) / path_in_backup, + settings, + threadPoolCallbackRunnerUnsafe(getBackupsIOThreadPool().get(), + "BackupWRAzure")); } BackupWriterAzureBlobStorage::~BackupWriterAzureBlobStorage() = default; bool BackupWriterAzureBlobStorage::fileExists(const String & file_name) { - String key = fs::path(configuration.blob_path) / file_name; + String key = fs::path(blob_path) / file_name; return object_storage->exists(StoredObject(key)); } UInt64 BackupWriterAzureBlobStorage::getFileSize(const String & file_name) { - String key = fs::path(configuration.blob_path) / file_name; + String key = fs::path(blob_path) / file_name; RelativePathsWithMetadata children; object_storage->listObjects(key,children,/*max_keys*/0); if (children.empty()) @@ -231,7 +243,7 @@ UInt64 BackupWriterAzureBlobStorage::getFileSize(const String & file_name) std::unique_ptr BackupWriterAzureBlobStorage::readFile(const String & file_name, size_t /*expected_file_size*/) { - String key = fs::path(configuration.blob_path) / file_name; + String key = fs::path(blob_path) / file_name; return std::make_unique( client, key, read_settings, settings->max_single_read_retries, settings->max_single_download_retries); @@ -239,7 +251,7 @@ std::unique_ptr BackupWriterAzureBlobStorage::readFile(const String std::unique_ptr BackupWriterAzureBlobStorage::writeFile(const String & file_name) { - String key = fs::path(configuration.blob_path) / file_name; + String key = fs::path(blob_path) / file_name; return std::make_unique( client, key, @@ -251,7 +263,7 @@ std::unique_ptr BackupWriterAzureBlobStorage::writeFile(const Strin void BackupWriterAzureBlobStorage::removeFile(const String & file_name) { - String key = fs::path(configuration.blob_path) / file_name; + String key = fs::path(blob_path) / file_name; StoredObject object(key); object_storage->removeObjectIfExists(object); } @@ -260,7 +272,7 @@ void BackupWriterAzureBlobStorage::removeFiles(const Strings & file_names) { StoredObjects objects; for (const auto & file_name : file_names) - objects.emplace_back(fs::path(configuration.blob_path) / file_name); + objects.emplace_back(fs::path(blob_path) / file_name); object_storage->removeObjectsIfExist(objects); @@ -270,7 +282,7 @@ void BackupWriterAzureBlobStorage::removeFilesBatch(const Strings & file_names) { StoredObjects objects; for (const auto & file_name : file_names) - objects.emplace_back(fs::path(configuration.blob_path) / file_name); + objects.emplace_back(fs::path(blob_path) / file_name); object_storage->removeObjectsIfExist(objects); } diff --git a/src/Backups/BackupIO_AzureBlobStorage.h b/src/Backups/BackupIO_AzureBlobStorage.h index 61688107839..c3b88f245ab 100644 --- a/src/Backups/BackupIO_AzureBlobStorage.h +++ b/src/Backups/BackupIO_AzureBlobStorage.h @@ -1,12 +1,10 @@ #pragma once - #include "config.h" #if USE_AZURE_BLOB_STORAGE #include #include -#include -#include +#include namespace DB @@ -17,7 +15,8 @@ class BackupReaderAzureBlobStorage : public BackupReaderDefault { public: BackupReaderAzureBlobStorage( - const StorageAzureConfiguration & configuration_, + const AzureBlobStorage::ConnectionParams & connection_params_, + const String & blob_path_, bool allow_azure_native_copy, const ReadSettings & read_settings_, const WriteSettings & write_settings_, @@ -40,16 +39,18 @@ public: private: const DataSourceDescription data_source_description; std::shared_ptr client; - StorageAzureConfiguration configuration; + AzureBlobStorage::ConnectionParams connection_params; + String blob_path; std::unique_ptr object_storage; - std::shared_ptr settings; + std::shared_ptr settings; }; class BackupWriterAzureBlobStorage : public BackupWriterDefault { public: BackupWriterAzureBlobStorage( - const StorageAzureConfiguration & configuration_, + const AzureBlobStorage::ConnectionParams & connection_params_, + const String & blob_path_, bool allow_azure_native_copy, const ReadSettings & read_settings_, const WriteSettings & write_settings_, @@ -87,9 +88,10 @@ private: const DataSourceDescription data_source_description; std::shared_ptr client; - StorageAzureConfiguration configuration; + AzureBlobStorage::ConnectionParams connection_params; + String blob_path; std::unique_ptr object_storage; - std::shared_ptr settings; + std::shared_ptr settings; }; } diff --git a/src/Backups/BackupIO_S3.cpp b/src/Backups/BackupIO_S3.cpp index 56544312c26..687096d0404 100644 --- a/src/Backups/BackupIO_S3.cpp +++ b/src/Backups/BackupIO_S3.cpp @@ -1,6 +1,7 @@ #include #if USE_AWS_S3 +#include #include #include #include @@ -54,9 +55,9 @@ namespace S3::PocoHTTPClientConfiguration client_configuration = S3::ClientFactory::instance().createClientConfiguration( settings.auth_settings.region, context->getRemoteHostFilter(), - static_cast(global_settings.s3_max_redirects), - static_cast(global_settings.s3_retry_attempts), - global_settings.enable_s3_requests_logging, + static_cast(local_settings.s3_max_redirects), + static_cast(local_settings.backup_restore_s3_retry_attempts), + local_settings.enable_s3_requests_logging, /* for_disk_s3 = */ false, request_settings.get_request_throttler, request_settings.put_request_throttler, diff --git a/src/Backups/BackupOperationInfo.h b/src/Backups/BackupOperationInfo.h index 21b5284458c..71589ec3b30 100644 --- a/src/Backups/BackupOperationInfo.h +++ b/src/Backups/BackupOperationInfo.h @@ -3,6 +3,8 @@ #include #include +#include + namespace DB { diff --git a/src/Backups/BackupUtils.cpp b/src/Backups/BackupUtils.cpp index fa8ed5855dd..cd3f963b15d 100644 --- a/src/Backups/BackupUtils.cpp +++ b/src/Backups/BackupUtils.cpp @@ -105,7 +105,7 @@ bool compareRestoredTableDef(const IAST & restored_table_create_query, const IAS auto new_query = query.clone(); adjustCreateQueryForBackup(new_query, global_context); ASTCreateQuery & create = typeid_cast(*new_query); - create.setUUID({}); + create.resetUUIDs(); create.if_not_exists = false; return new_query; }; diff --git a/src/Backups/BackupsWorker.cpp b/src/Backups/BackupsWorker.cpp index 69d9c52ebd9..15a7d7c1eca 100644 --- a/src/Backups/BackupsWorker.cpp +++ b/src/Backups/BackupsWorker.cpp @@ -26,6 +26,7 @@ #include #include #include +#include #include diff --git a/src/Backups/DDLAdjustingForBackupVisitor.cpp b/src/Backups/DDLAdjustingForBackupVisitor.cpp index 7e5ce91629b..910831195a3 100644 --- a/src/Backups/DDLAdjustingForBackupVisitor.cpp +++ b/src/Backups/DDLAdjustingForBackupVisitor.cpp @@ -1,11 +1,11 @@ #include +#include +#include #include #include #include -#include -#include - #include +#include namespace DB diff --git a/src/Backups/RestoreCoordinationLocal.cpp b/src/Backups/RestoreCoordinationLocal.cpp index f51d6c0c1d8..9fe22f874b4 100644 --- a/src/Backups/RestoreCoordinationLocal.cpp +++ b/src/Backups/RestoreCoordinationLocal.cpp @@ -1,4 +1,5 @@ #include +#include #include #include @@ -67,7 +68,7 @@ void RestoreCoordinationLocal::generateUUIDForTable(ASTCreateQuery & create_quer auto it = create_query_uuids.find(query_str); if (it != create_query_uuids.end()) { - create_query.setUUID(it->second); + it->second.copyToQuery(create_query); return true; } return false; @@ -79,7 +80,8 @@ void RestoreCoordinationLocal::generateUUIDForTable(ASTCreateQuery & create_quer return; } - auto new_uuids = create_query.generateRandomUUID(/* always_generate_new_uuid= */ true); + CreateQueryUUIDs new_uuids{create_query, /* generate_random= */ true, /* force_random= */ true}; + new_uuids.copyToQuery(create_query); { std::lock_guard lock{mutex}; diff --git a/src/Backups/RestoreCoordinationLocal.h b/src/Backups/RestoreCoordinationLocal.h index 5e51b719d63..35f93574b68 100644 --- a/src/Backups/RestoreCoordinationLocal.h +++ b/src/Backups/RestoreCoordinationLocal.h @@ -1,16 +1,17 @@ #pragma once #include -#include +#include +#include #include #include #include -namespace Poco { class Logger; } - namespace DB { +class ASTCreateQuery; + /// Implementation of the IRestoreCoordination interface performing coordination in memory. class RestoreCoordinationLocal : public IRestoreCoordination @@ -55,7 +56,7 @@ private: std::set> acquired_tables_in_replicated_databases; std::unordered_set acquired_data_in_replicated_tables; - std::unordered_map create_query_uuids; + std::unordered_map create_query_uuids; std::unordered_set acquired_data_in_keeper_map_tables; mutable std::mutex mutex; diff --git a/src/Backups/RestoreCoordinationRemote.cpp b/src/Backups/RestoreCoordinationRemote.cpp index 84106737fc9..44214d00be5 100644 --- a/src/Backups/RestoreCoordinationRemote.cpp +++ b/src/Backups/RestoreCoordinationRemote.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -269,7 +270,8 @@ bool RestoreCoordinationRemote::acquireInsertingDataForKeeperMap(const String & void RestoreCoordinationRemote::generateUUIDForTable(ASTCreateQuery & create_query) { String query_str = serializeAST(create_query); - String new_uuids_str = create_query.generateRandomUUID(/* always_generate_new_uuid= */ true).toString(); + CreateQueryUUIDs new_uuids{create_query, /* generate_random= */ true, /* force_random= */ true}; + String new_uuids_str = new_uuids.toString(); auto holder = with_retries.createRetriesControlHolder("generateUUIDForTable"); holder.retries_ctl.retryLoop( @@ -281,11 +283,14 @@ void RestoreCoordinationRemote::generateUUIDForTable(ASTCreateQuery & create_que Coordination::Error res = zk->tryCreate(path, new_uuids_str, zkutil::CreateMode::Persistent); if (res == Coordination::Error::ZOK) + { + new_uuids.copyToQuery(create_query); return; + } if (res == Coordination::Error::ZNODEEXISTS) { - create_query.setUUID(ASTCreateQuery::UUIDs::fromString(zk->get(path))); + CreateQueryUUIDs::fromString(zk->get(path)).copyToQuery(create_query); return; } diff --git a/src/Backups/RestorerFromBackup.cpp b/src/Backups/RestorerFromBackup.cpp index 1a3fdf58cc4..3056f9fe421 100644 --- a/src/Backups/RestorerFromBackup.cpp +++ b/src/Backups/RestorerFromBackup.cpp @@ -23,8 +23,9 @@ #include #include #include -#include +#include +#include #include #include @@ -438,7 +439,7 @@ void RestorerFromBackup::findTableInBackupImpl(const QualifiedTableName & table_ String create_table_query_str = serializeAST(*create_table_query); bool is_predefined_table = DatabaseCatalog::instance().isPredefinedTable(StorageID{table_name.database, table_name.table}); - auto table_dependencies = getDependenciesFromCreateQuery(context, table_name, create_table_query); + auto table_dependencies = getDependenciesFromCreateQuery(context, table_name, create_table_query, context->getCurrentDatabase()); bool table_has_data = backup->hasFiles(data_path_in_backup); std::lock_guard lock{mutex}; diff --git a/src/Backups/WithRetries.cpp b/src/Backups/WithRetries.cpp index 66851fa42ce..181e6331ac9 100644 --- a/src/Backups/WithRetries.cpp +++ b/src/Backups/WithRetries.cpp @@ -1,5 +1,7 @@ -#include #include +#include + +#include namespace DB { diff --git a/src/Backups/registerBackupEngineAzureBlobStorage.cpp b/src/Backups/registerBackupEngineAzureBlobStorage.cpp index 03d156d1009..626df99b00c 100644 --- a/src/Backups/registerBackupEngineAzureBlobStorage.cpp +++ b/src/Backups/registerBackupEngineAzureBlobStorage.cpp @@ -5,6 +5,7 @@ #if USE_AZURE_BLOB_STORAGE #include +#include #include #include #include @@ -49,7 +50,9 @@ void registerBackupEngineAzureBlobStorage(BackupFactory & factory) const String & id_arg = params.backup_info.id_arg; const auto & args = params.backup_info.args; - StorageAzureConfiguration configuration; + String blob_path; + AzureBlobStorage::ConnectionParams connection_params; + auto request_settings = AzureBlobStorage::getRequestSettings(params.context->getSettingsRef()); if (!id_arg.empty()) { @@ -59,55 +62,42 @@ void registerBackupEngineAzureBlobStorage(BackupFactory & factory) if (!config.has(config_prefix)) throw Exception(ErrorCodes::BAD_ARGUMENTS, "There is no collection named `{}` in config", id_arg); - if (config.has(config_prefix + ".connection_string")) + connection_params = { - configuration.connection_url = config.getString(config_prefix + ".connection_string"); - configuration.is_connection_string = true; - configuration.container = config.getString(config_prefix + ".container"); - } - else - { - configuration.connection_url = config.getString(config_prefix + ".storage_account_url"); - configuration.is_connection_string = false; - configuration.container = config.getString(config_prefix + ".container"); - configuration.account_name = config.getString(config_prefix + ".account_name"); - configuration.account_key = config.getString(config_prefix + ".account_key"); - - if (config.has(config_prefix + ".account_name") && config.has(config_prefix + ".account_key")) - { - configuration.account_name = config.getString(config_prefix + ".account_name"); - configuration.account_key = config.getString(config_prefix + ".account_key"); - } - } + .endpoint = AzureBlobStorage::processEndpoint(config, config_prefix), + .auth_method = AzureBlobStorage::getAuthMethod(config, config_prefix), + .client_options = AzureBlobStorage::getClientOptions(*request_settings, /*for_disk=*/ true), + }; if (args.size() > 1) throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Backup AzureBlobStorage requires 1 or 2 arguments: named_collection, [filename]"); if (args.size() == 1) - configuration.setPath(args[0].safeGet()); - + blob_path = args[0].safeGet(); } else { if (args.size() == 3) { - configuration.connection_url = args[0].safeGet(); - configuration.is_connection_string = !configuration.connection_url.starts_with("http"); + auto connection_url = args[0].safeGet(); + auto container_name = args[1].safeGet(); + blob_path = args[2].safeGet(); - configuration.container = args[1].safeGet(); - configuration.blob_path = args[2].safeGet(); + AzureBlobStorage::processURL(connection_url, container_name, connection_params.endpoint, connection_params.auth_method); + connection_params.client_options = AzureBlobStorage::getClientOptions(*request_settings, /*for_disk=*/ true); } else if (args.size() == 5) { - configuration.connection_url = args[0].safeGet(); - configuration.is_connection_string = false; + connection_params.endpoint.storage_account_url = args[0].safeGet(); + connection_params.endpoint.container_name = args[1].safeGet(); + blob_path = args[2].safeGet(); - configuration.container = args[1].safeGet(); - configuration.blob_path = args[2].safeGet(); - configuration.account_name = args[3].safeGet(); - configuration.account_key = args[4].safeGet(); + auto account_name = args[3].safeGet(); + auto account_key = args[4].safeGet(); + connection_params.auth_method = std::make_shared(account_name, account_key); + connection_params.client_options = AzureBlobStorage::getClientOptions(*request_settings, /*for_disk=*/ true); } else { @@ -117,16 +107,12 @@ void registerBackupEngineAzureBlobStorage(BackupFactory & factory) } BackupImpl::ArchiveParams archive_params; - if (hasRegisteredArchiveFileExtension(configuration.getPath())) + if (hasRegisteredArchiveFileExtension(blob_path)) { if (params.is_internal_backup) throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "Using archives with backups on clusters is disabled"); - auto path = configuration.getPath(); - auto filename = removeFileNameFromURL(path); - configuration.setPath(path); - - archive_params.archive_name = filename; + archive_params.archive_name = removeFileNameFromURL(blob_path); archive_params.compression_method = params.compression_method; archive_params.compression_level = params.compression_level; archive_params.password = params.password; @@ -141,7 +127,8 @@ void registerBackupEngineAzureBlobStorage(BackupFactory & factory) if (params.open_mode == IBackup::OpenMode::READ) { auto reader = std::make_shared( - configuration, + connection_params, + blob_path, params.allow_azure_native_copy, params.read_settings, params.write_settings, @@ -159,7 +146,8 @@ void registerBackupEngineAzureBlobStorage(BackupFactory & factory) else { auto writer = std::make_shared( - configuration, + connection_params, + blob_path, params.allow_azure_native_copy, params.read_settings, params.write_settings, diff --git a/src/Bridge/IBridge.cpp b/src/Bridge/IBridge.cpp index c25d7bd2fed..de48a4f2b84 100644 --- a/src/Bridge/IBridge.cpp +++ b/src/Bridge/IBridge.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include diff --git a/src/BridgeHelper/LibraryBridgeHelper.cpp b/src/BridgeHelper/LibraryBridgeHelper.cpp index 84bfe096e79..12ca5564e80 100644 --- a/src/BridgeHelper/LibraryBridgeHelper.cpp +++ b/src/BridgeHelper/LibraryBridgeHelper.cpp @@ -1,5 +1,7 @@ #include "LibraryBridgeHelper.h" +#include +#include #include namespace DB diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index 8a14d10c9d9..4f5ff82d753 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -210,6 +210,7 @@ add_object_library(clickhouse_analyzer_passes Analyzer/Resolve) add_object_library(clickhouse_planner Planner) add_object_library(clickhouse_interpreters Interpreters) add_object_library(clickhouse_interpreters_cache Interpreters/Cache) +add_object_library(clickhouse_interpreters_hash_join Interpreters/HashJoin) add_object_library(clickhouse_interpreters_access Interpreters/Access) add_object_library(clickhouse_interpreters_mysql Interpreters/MySQL) add_object_library(clickhouse_interpreters_clusterproxy Interpreters/ClusterProxy) @@ -222,9 +223,12 @@ add_object_library(clickhouse_storages_mergetree Storages/MergeTree) add_object_library(clickhouse_storages_statistics Storages/Statistics) add_object_library(clickhouse_storages_liveview Storages/LiveView) add_object_library(clickhouse_storages_windowview Storages/WindowView) -add_object_library(clickhouse_storages_s3queue Storages/S3Queue) +add_object_library(clickhouse_storages_s3queue Storages/ObjectStorageQueue) add_object_library(clickhouse_storages_materializedview Storages/MaterializedView) add_object_library(clickhouse_client Client) +# Always compile this file with the highest possible level of optimizations, even in Debug builds. +# https://github.com/ClickHouse/ClickHouse/issues/65745 +set_source_files_properties(Client/ClientBaseOptimizedParts.cpp PROPERTIES COMPILE_FLAGS "-O3") add_object_library(clickhouse_bridge BridgeHelper) add_object_library(clickhouse_server Server) add_object_library(clickhouse_server_http Server/HTTP) @@ -431,9 +435,6 @@ dbms_target_link_libraries ( boost::circular_buffer boost::heap) -target_include_directories(clickhouse_common_io PUBLIC "${CMAKE_CURRENT_BINARY_DIR}/Core/include") # uses some includes from core -dbms_target_include_directories(PUBLIC "${CMAKE_CURRENT_BINARY_DIR}/Core/include") - target_link_libraries(clickhouse_common_io PUBLIC ch_contrib::miniselect ch_contrib::pdqsort) @@ -558,7 +559,7 @@ if (TARGET ch_contrib::libpqxx) endif() if (TARGET ch_contrib::datasketches) - target_link_libraries (clickhouse_aggregate_functions PRIVATE ch_contrib::datasketches) + dbms_target_link_libraries(PUBLIC ch_contrib::datasketches) endif () target_link_libraries (clickhouse_common_io PRIVATE ch_contrib::lz4) @@ -619,6 +620,10 @@ if (TARGET ch_contrib::usearch) dbms_target_link_libraries(PUBLIC ch_contrib::usearch) endif() +if (TARGET ch_contrib::prometheus_protobufs) + dbms_target_link_libraries (PUBLIC ch_contrib::prometheus_protobufs) +endif() + if (TARGET ch_rust::skim) dbms_target_include_directories(PRIVATE $) dbms_target_link_libraries(PUBLIC ch_rust::skim) diff --git a/src/Client/ClientBase.cpp b/src/Client/ClientBase.cpp index c97837b685d..04af9db7afe 100644 --- a/src/Client/ClientBase.cpp +++ b/src/Client/ClientBase.cpp @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include @@ -80,6 +81,11 @@ #include #include "config.h" +#if USE_GWP_ASAN +# include +#endif + + namespace fs = std::filesystem; using namespace std::literals; @@ -102,7 +108,6 @@ namespace ErrorCodes extern const int UNEXPECTED_PACKET_FROM_SERVER; extern const int INVALID_USAGE_OF_INPUT; extern const int CANNOT_SET_SIGNAL_HANDLER; - extern const int UNRECOGNIZED_ARGUMENTS; extern const int LOGICAL_ERROR; extern const int CANNOT_OPEN_FILE; extern const int FILE_ALREADY_EXISTS; @@ -301,9 +306,43 @@ public: }; -ClientBase::~ClientBase() = default; -ClientBase::ClientBase() = default; +ClientBase::~ClientBase() +{ + try + { + writeSignalIDtoSignalPipe(SignalListener::StopThread); + signal_listener_thread.join(); + HandledSignals::instance().reset(); + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + } +} +ClientBase::ClientBase( + int in_fd_, + int out_fd_, + int err_fd_, + std::istream & input_stream_, + std::ostream & output_stream_, + std::ostream & error_stream_ +) + : std_in(in_fd_) + , std_out(out_fd_) + , progress_indication(output_stream_, in_fd_, err_fd_) + , in_fd(in_fd_) + , out_fd(out_fd_) + , err_fd(err_fd_) + , input_stream(input_stream_) + , output_stream(output_stream_) + , error_stream(error_stream_) +{ + stdin_is_a_tty = isatty(in_fd); + stdout_is_a_tty = isatty(out_fd); + stderr_is_a_tty = isatty(err_fd); + terminal_width = getTerminalWidth(in_fd, err_fd); +} void ClientBase::setupSignalHandler() { @@ -330,7 +369,7 @@ void ClientBase::setupSignalHandler() } -ASTPtr ClientBase::parseQuery(const char *& pos, const char * end, const Settings & settings, bool allow_multi_statements, bool is_interactive, bool ignore_error) +ASTPtr ClientBase::parseQuery(const char *& pos, const char * end, const Settings & settings, bool allow_multi_statements) { std::unique_ptr parser; ASTPtr res; @@ -359,7 +398,7 @@ ASTPtr ClientBase::parseQuery(const char *& pos, const char * end, const Setting if (!res) { - std::cerr << std::endl << message << std::endl << std::endl; + error_stream << std::endl << message << std::endl << std::endl; return nullptr; } } @@ -373,11 +412,11 @@ ASTPtr ClientBase::parseQuery(const char *& pos, const char * end, const Setting if (is_interactive) { - std::cout << std::endl; - WriteBufferFromOStream res_buf(std::cout, 4096); + output_stream << std::endl; + WriteBufferFromOStream res_buf(output_stream, 4096); formatAST(*res, res_buf); res_buf.finalize(); - std::cout << std::endl << std::endl; + output_stream << std::endl << std::endl; } return res; @@ -481,7 +520,7 @@ void ClientBase::onData(Block & block, ASTPtr parsed_query) if (need_render_progress && tty_buf) { if (select_into_file && !select_into_file_and_stdout) - std::cerr << "\r"; + error_stream << "\r"; progress_indication.writeProgress(*tty_buf); } } @@ -741,17 +780,17 @@ bool ClientBase::isRegularFile(int fd) void ClientBase::setDefaultFormatsAndCompressionFromConfiguration() { - if (config().has("output-format")) + if (getClientConfiguration().has("output-format")) { - default_output_format = config().getString("output-format"); + default_output_format = getClientConfiguration().getString("output-format"); is_default_format = false; } - else if (config().has("format")) + else if (getClientConfiguration().has("format")) { - default_output_format = config().getString("format"); + default_output_format = getClientConfiguration().getString("format"); is_default_format = false; } - else if (config().has("vertical")) + else if (getClientConfiguration().has("vertical")) { default_output_format = "Vertical"; is_default_format = false; @@ -777,17 +816,17 @@ void ClientBase::setDefaultFormatsAndCompressionFromConfiguration() default_output_format = "TSV"; } - if (config().has("input-format")) + if (getClientConfiguration().has("input-format")) { - default_input_format = config().getString("input-format"); + default_input_format = getClientConfiguration().getString("input-format"); } - else if (config().has("format")) + else if (getClientConfiguration().has("format")) { - default_input_format = config().getString("format"); + default_input_format = getClientConfiguration().getString("format"); } - else if (config().getString("table-file", "-") != "-") + else if (getClientConfiguration().getString("table-file", "-") != "-") { - auto file_name = config().getString("table-file"); + auto file_name = getClientConfiguration().getString("table-file"); std::optional format_from_file_name = FormatFactory::instance().tryGetFormatFromFileName(file_name); if (format_from_file_name) default_input_format = *format_from_file_name; @@ -803,7 +842,7 @@ void ClientBase::setDefaultFormatsAndCompressionFromConfiguration() default_input_format = "TSV"; } - format_max_block_size = config().getUInt64("format_max_block_size", + format_max_block_size = getClientConfiguration().getUInt64("format_max_block_size", global_context->getSettingsRef().max_block_size); /// Setting value from cmd arg overrides one from config @@ -813,7 +852,7 @@ void ClientBase::setDefaultFormatsAndCompressionFromConfiguration() } else { - insert_format_max_block_size = config().getUInt64("insert_format_max_block_size", + insert_format_max_block_size = getClientConfiguration().getUInt64("insert_format_max_block_size", global_context->getSettingsRef().max_insert_block_size); } } @@ -924,9 +963,7 @@ void ClientBase::processTextAsSingleQuery(const String & full_query) const char * begin = full_query.data(); auto parsed_query = parseQuery(begin, begin + full_query.size(), global_context->getSettingsRef(), - /*allow_multi_statements=*/ false, - is_interactive, - ignore_error); + /*allow_multi_statements=*/ false); if (!parsed_query) return; @@ -1100,7 +1137,7 @@ void ClientBase::processOrdinaryQuery(const String & query_to_execute, ASTPtr pa /// has been received yet. if (processed_rows == 0 && e.code() == ErrorCodes::DEADLOCK_AVOIDED && --retries_left) { - std::cerr << "Got a transient error from the server, will" + error_stream << "Got a transient error from the server, will" << " retry (" << retries_left << " retries left)"; } else @@ -1154,7 +1191,7 @@ void ClientBase::receiveResult(ASTPtr parsed_query, Int32 signals_before_stop, b double elapsed = receive_watch.elapsedSeconds(); if (break_on_timeout && elapsed > receive_timeout.totalSeconds()) { - std::cout << "Timeout exceeded while receiving data from server." + output_stream << "Timeout exceeded while receiving data from server." << " Waited for " << static_cast(elapsed) << " seconds," << " timeout is " << receive_timeout.totalSeconds() << " seconds." << std::endl; @@ -1187,11 +1224,8 @@ void ClientBase::receiveResult(ASTPtr parsed_query, Int32 signals_before_stop, b if (local_format_error) std::rethrow_exception(local_format_error); - if (cancelled && is_interactive) - { - std::cout << "Query was cancelled." << std::endl; - cancelled_printed = true; - } + if (cancelled && is_interactive && !cancelled_printed.exchange(true)) + output_stream << "Query was cancelled." << std::endl; } @@ -1307,10 +1341,10 @@ void ClientBase::onEndOfStream() if (is_interactive) { - if (cancelled && !cancelled_printed) - std::cout << "Query was cancelled." << std::endl; + if (cancelled && !cancelled_printed.exchange(true)) + output_stream << "Query was cancelled." << std::endl; else if (!written_first_block) - std::cout << "Ok." << std::endl; + output_stream << "Ok." << std::endl; } } @@ -1863,7 +1897,7 @@ void ClientBase::cancelQuery() progress_indication.clearProgressOutput(*tty_buf); if (is_interactive) - std::cout << "Cancelling query." << std::endl; + output_stream << "Cancelling query." << std::endl; cancelled = true; } @@ -2026,7 +2060,7 @@ void ClientBase::processParsedSingleQuery(const String & full_query, const Strin { const String & new_database = use_query->getDatabase(); /// If the client initiates the reconnection, it takes the settings from the config. - config().setString("database", new_database); + getClientConfiguration().setString("database", new_database); /// If the connection initiates the reconnection, it uses its variable. connection->setDefaultDatabase(new_database); } @@ -2046,21 +2080,30 @@ void ClientBase::processParsedSingleQuery(const String & full_query, const Strin if (is_interactive) { - std::cout << std::endl; + output_stream << std::endl; if (!server_exception || processed_rows != 0) - std::cout << processed_rows << " row" << (processed_rows == 1 ? "" : "s") << " in set. "; - std::cout << "Elapsed: " << progress_indication.elapsedSeconds() << " sec. "; + output_stream << processed_rows << " row" << (processed_rows == 1 ? "" : "s") << " in set. "; + output_stream << "Elapsed: " << progress_indication.elapsedSeconds() << " sec. "; progress_indication.writeFinalProgress(); - std::cout << std::endl << std::endl; + output_stream << std::endl << std::endl; } - else if (print_time_to_stderr) + else { - std::cerr << progress_indication.elapsedSeconds() << "\n"; + const auto & config = getClientConfiguration(); + if (config.getBool("print-time-to-stderr", false)) + error_stream << progress_indication.elapsedSeconds() << "\n"; + + const auto & print_memory_mode = config.getString("print-memory-to-stderr", ""); + auto peak_memeory_usage = std::max(progress_indication.getMemoryUsage().peak, 0); + if (print_memory_mode == "default") + error_stream << peak_memeory_usage << "\n"; + else if (print_memory_mode == "readable") + error_stream << formatReadableSizeWithBinarySuffix(peak_memeory_usage) << "\n"; } - if (!is_interactive && print_num_processed_rows) + if (!is_interactive && getClientConfiguration().getBool("print-num-processed-rows", false)) { - std::cout << "Processed rows: " << processed_rows << "\n"; + output_stream << "Processed rows: " << processed_rows << "\n"; } if (have_error && report_error) @@ -2110,9 +2153,7 @@ MultiQueryProcessingStage ClientBase::analyzeMultiQueryText( { parsed_query = parseQuery(this_query_end, all_queries_end, global_context->getSettingsRef(), - /*allow_multi_statements=*/ true, - is_interactive, - ignore_error); + /*allow_multi_statements=*/ true); } catch (const Exception & e) { @@ -2199,6 +2240,8 @@ bool ClientBase::executeMultiQuery(const String & all_queries_text) ASTPtr parsed_query; std::unique_ptr current_exception; + size_t retries_count = 0; + while (true) { auto stage = analyzeMultiQueryText(this_query_begin, this_query_end, all_queries_end, @@ -2279,7 +2322,12 @@ bool ClientBase::executeMultiQuery(const String & all_queries_text) // Check whether the error (or its absence) matches the test hints // (or their absence). bool error_matches_hint = true; - if (have_error) + bool need_retry = test_hint.needRetry(server_exception, &retries_count); + if (need_retry) + { + std::this_thread::sleep_for(std::chrono::seconds(1)); + } + else if (have_error) { if (test_hint.hasServerErrors()) { @@ -2373,7 +2421,8 @@ bool ClientBase::executeMultiQuery(const String & all_queries_text) if (have_error && !ignore_error) return is_interactive; - this_query_begin = this_query_end; + if (!need_retry) + this_query_begin = this_query_end; break; } } @@ -2428,12 +2477,12 @@ void ClientBase::initQueryIdFormats() return; /// Initialize query_id_formats if any - if (config().has("query_id_formats")) + if (getClientConfiguration().has("query_id_formats")) { Poco::Util::AbstractConfiguration::Keys keys; - config().keys("query_id_formats", keys); + getClientConfiguration().keys("query_id_formats", keys); for (const auto & name : keys) - query_id_formats.emplace_back(name + ":", config().getString("query_id_formats." + name)); + query_id_formats.emplace_back(name + ":", getClientConfiguration().getString("query_id_formats." + name)); } if (query_id_formats.empty()) @@ -2478,9 +2527,9 @@ bool ClientBase::addMergeTreeSettings(ASTCreateQuery & ast_create) void ClientBase::runInteractive() { - if (config().has("query_id")) + if (getClientConfiguration().has("query_id")) throw Exception(ErrorCodes::BAD_ARGUMENTS, "query_id could be specified only in non-interactive mode"); - if (print_time_to_stderr) + if (getClientConfiguration().getBool("print-time-to-stderr", false)) throw Exception(ErrorCodes::BAD_ARGUMENTS, "time option could be specified only in non-interactive mode"); initQueryIdFormats(); @@ -2493,9 +2542,9 @@ void ClientBase::runInteractive() { /// Load suggestion data from the server. if (global_context->getApplicationType() == Context::ApplicationType::CLIENT) - suggest->load(global_context, connection_parameters, config().getInt("suggestion_limit"), wait_for_suggestions_to_load); + suggest->load(global_context, connection_parameters, getClientConfiguration().getInt("suggestion_limit"), wait_for_suggestions_to_load); else if (global_context->getApplicationType() == Context::ApplicationType::LOCAL) - suggest->load(global_context, connection_parameters, config().getInt("suggestion_limit"), wait_for_suggestions_to_load); + suggest->load(global_context, connection_parameters, getClientConfiguration().getInt("suggestion_limit"), wait_for_suggestions_to_load); } if (home_path.empty()) @@ -2506,8 +2555,8 @@ void ClientBase::runInteractive() } /// Load command history if present. - if (config().has("history_file")) - history_file = config().getString("history_file"); + if (getClientConfiguration().has("history_file")) + history_file = getClientConfiguration().getString("history_file"); else { auto * history_file_from_env = getenv("CLICKHOUSE_HISTORY_FILE"); // NOLINT(concurrency-mt-unsafe) @@ -2528,7 +2577,7 @@ void ClientBase::runInteractive() { if (e.getErrno() != EEXIST) { - std::cerr << getCurrentExceptionMessage(false) << '\n'; + error_stream << getCurrentExceptionMessage(false) << '\n'; } } } @@ -2539,24 +2588,24 @@ void ClientBase::runInteractive() #if USE_REPLXX replxx::Replxx::highlighter_callback_t highlight_callback{}; - if (config().getBool("highlight", true)) + if (getClientConfiguration().getBool("highlight", true)) highlight_callback = highlight; ReplxxLineReader lr( *suggest, history_file, - config().has("multiline"), + getClientConfiguration().has("multiline"), query_extenders, query_delimiters, word_break_characters, highlight_callback); #else + (void)word_break_characters; LineReader lr( history_file, - config().has("multiline"), + getClientConfiguration().has("multiline"), query_extenders, - query_delimiters, - word_break_characters); + query_delimiters); #endif static const std::initializer_list> backslash_aliases = @@ -2633,7 +2682,7 @@ void ClientBase::runInteractive() { // If a separate connection loading suggestions failed to open a new session, // use the main session to receive them. - suggest->load(*connection, connection_parameters.timeouts, config().getInt("suggestion_limit"), global_context->getClientInfo()); + suggest->load(*connection, connection_parameters.timeouts, getClientConfiguration().getInt("suggestion_limit"), global_context->getClientInfo()); } try @@ -2648,7 +2697,7 @@ void ClientBase::runInteractive() break; /// We don't need to handle the test hints in the interactive mode. - std::cerr << "Exception on client:" << std::endl << getExceptionMessage(e, print_stack_trace, true) << std::endl << std::endl; + error_stream << "Exception on client:" << std::endl << getExceptionMessage(e, print_stack_trace, true) << std::endl << std::endl; client_exception.reset(e.clone()); } @@ -2665,11 +2714,11 @@ void ClientBase::runInteractive() while (true); if (isNewYearMode()) - std::cout << "Happy new year." << std::endl; + output_stream << "Happy new year." << std::endl; else if (isChineseNewYearMode(local_tz)) - std::cout << "Happy Chinese new year. 春节快乐!" << std::endl; + output_stream << "Happy Chinese new year. 春节快乐!" << std::endl; else - std::cout << "Bye." << std::endl; + output_stream << "Bye." << std::endl; } @@ -2680,7 +2729,7 @@ bool ClientBase::processMultiQueryFromFile(const String & file_name) ReadBufferFromFile in(file_name); readStringUntilEOF(queries_from_file, in); - if (!has_log_comment) + if (!getClientConfiguration().has("log_comment")) { Settings settings = global_context->getSettings(); /// NOTE: cannot use even weakly_canonical() since it fails for /dev/stdin due to resolving of "pipe:[X]" @@ -2789,174 +2838,15 @@ void ClientBase::clearTerminal() /// It is needed if garbage is left in terminal. /// Show cursor. It can be left hidden by invocation of previous programs. /// A test for this feature: perl -e 'print "x"x100000'; echo -ne '\033[0;0H\033[?25l'; clickhouse-client - std::cout << "\033[0J" "\033[?25h"; + output_stream << "\033[0J" "\033[?25h"; } void ClientBase::showClientVersion() { - std::cout << VERSION_NAME << " " + getName() + " version " << VERSION_STRING << VERSION_OFFICIAL << "." << std::endl; + output_stream << VERSION_NAME << " " + getName() + " version " << VERSION_STRING << VERSION_OFFICIAL << "." << std::endl; } -namespace -{ - -/// Define transparent hash to we can use -/// std::string_view with the containers -struct TransparentStringHash -{ - using is_transparent = void; - size_t operator()(std::string_view txt) const - { - return std::hash{}(txt); - } -}; - -/* - * This functor is used to parse command line arguments and replace dashes with underscores, - * allowing options to be specified using either dashes or underscores. - */ -class OptionsAliasParser -{ -public: - explicit OptionsAliasParser(const boost::program_options::options_description& options) - { - options_names.reserve(options.options().size()); - for (const auto& option : options.options()) - options_names.insert(option->long_name()); - } - - /* - * Parses arguments by replacing dashes with underscores, and matches the resulting name with known options - * Implements boost::program_options::ext_parser logic - */ - std::pair operator()(const std::string & token) const - { - if (!token.starts_with("--")) - return {}; - std::string arg = token.substr(2); - - // divide token by '=' to separate key and value if options style=long_allow_adjacent - auto pos_eq = arg.find('='); - std::string key = arg.substr(0, pos_eq); - - if (options_names.contains(key)) - // option does not require any changes, because it is already correct - return {}; - - std::replace(key.begin(), key.end(), '-', '_'); - if (!options_names.contains(key)) - // after replacing '-' with '_' argument is still unknown - return {}; - - std::string value; - if (pos_eq != std::string::npos && pos_eq < arg.size()) - value = arg.substr(pos_eq + 1); - - return {key, value}; - } - -private: - std::unordered_set options_names; -}; - -} - - -void ClientBase::parseAndCheckOptions(OptionsDescription & options_description, po::variables_map & options, Arguments & arguments) -{ - if (allow_repeated_settings) - addProgramOptionsAsMultitokens(cmd_settings, options_description.main_description.value()); - else - addProgramOptions(cmd_settings, options_description.main_description.value()); - - if (allow_merge_tree_settings) - { - /// Add merge tree settings manually, because names of some settings - /// may clash. Query settings have higher priority and we just - /// skip ambiguous merge tree settings. - auto & main_options = options_description.main_description.value(); - - std::unordered_set> main_option_names; - for (const auto & option : main_options.options()) - main_option_names.insert(option->long_name()); - - for (const auto & setting : cmd_merge_tree_settings.all()) - { - const auto add_setting = [&](const std::string_view name) - { - if (auto it = main_option_names.find(name); it != main_option_names.end()) - return; - - if (allow_repeated_settings) - addProgramOptionAsMultitoken(cmd_merge_tree_settings, main_options, name, setting); - else - addProgramOption(cmd_merge_tree_settings, main_options, name, setting); - }; - - const auto & setting_name = setting.getName(); - - add_setting(setting_name); - - const auto & settings_to_aliases = MergeTreeSettings::Traits::settingsToAliases(); - if (auto it = settings_to_aliases.find(setting_name); it != settings_to_aliases.end()) - { - for (const auto alias : it->second) - { - add_setting(alias); - } - } - } - } - - /// Parse main commandline options. - auto parser = po::command_line_parser(arguments) - .options(options_description.main_description.value()) - .extra_parser(OptionsAliasParser(options_description.main_description.value())) - .allow_unregistered(); - po::parsed_options parsed = parser.run(); - - /// Check unrecognized options without positional options. - auto unrecognized_options = po::collect_unrecognized(parsed.options, po::collect_unrecognized_mode::exclude_positional); - if (!unrecognized_options.empty()) - { - auto hints = this->getHints(unrecognized_options[0]); - if (!hints.empty()) - throw Exception(ErrorCodes::UNRECOGNIZED_ARGUMENTS, "Unrecognized option '{}'. Maybe you meant {}", - unrecognized_options[0], toString(hints)); - - throw Exception(ErrorCodes::UNRECOGNIZED_ARGUMENTS, "Unrecognized option '{}'", unrecognized_options[0]); - } - - /// Check positional options. - for (const auto & op : parsed.options) - { - if (!op.unregistered && op.string_key.empty() && !op.original_tokens[0].starts_with("--") - && !op.original_tokens[0].empty() && !op.value.empty()) - { - /// Two special cases for better usability: - /// - if the option contains a whitespace, it might be a query: clickhouse "SELECT 1" - /// These are relevant for interactive usage - user-friendly, but questionable in general. - /// In case of ambiguity or for scripts, prefer using proper options. - - const auto & token = op.original_tokens[0]; - po::variable_value value(boost::any(op.value), false); - - const char * option; - if (token.contains(' ')) - option = "query"; - else - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Positional option `{}` is not supported.", token); - - if (!options.emplace(option, value).second) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Positional option `{}` is not supported.", token); - } - } - - po::store(parsed, options); -} - - void ClientBase::init(int argc, char ** argv) { namespace po = boost::program_options; @@ -3018,6 +2908,7 @@ void ClientBase::init(int argc, char ** argv) ("disable_suggestion,A", "Disable loading suggestion data. Note that suggestion data is loaded asynchronously through a second connection to ClickHouse server. Also it is reasonable to disable suggestion if you want to paste a query with TAB characters. Shorthand option -A is for those who get used to mysql client.") ("wait_for_suggestions_to_load", "Load suggestion data synchonously.") ("time,t", "print query execution time to stderr in non-interactive mode (for benchmarks)") + ("memory-usage", po::value()->implicit_value("default")->default_value("none"), "print memory usage to stderr in non-interactive mode (for benchmarks). Values: 'none', 'default', 'readable'") ("echo", "in batch mode, print query before execution") @@ -3044,6 +2935,8 @@ void ClientBase::init(int argc, char ** argv) ("max_memory_usage_in_client", po::value(), "Set memory limit in client/local server") ("fuzzer-args", po::value(), "Command line arguments for the LLVM's libFuzzer driver. Only relevant if the application is compiled with libFuzzer.") + + ("client_logs_file", po::value(), "Path to a file for writing client logs. Currently we only have fatal logs (when the client crashes)") ; addOptions(options_description); @@ -3080,18 +2973,18 @@ void ClientBase::init(int argc, char ** argv) if (options.count("version-clean")) { - std::cout << VERSION_STRING; + output_stream << VERSION_STRING; exit(0); // NOLINT(concurrency-mt-unsafe) } if (options.count("verbose")) - config().setBool("verbose", true); + getClientConfiguration().setBool("verbose", true); /// Output of help message. if (options.count("help") || (options.count("host") && options["host"].as() == "elp")) /// If user writes -help instead of --help. { - if (config().getBool("verbose", false)) + if (getClientConfiguration().getBool("verbose", false)) printHelpMessage(options_description, true); else printHelpMessage(options_description_non_verbose, false); @@ -3099,72 +2992,83 @@ void ClientBase::init(int argc, char ** argv) } /// Common options for clickhouse-client and clickhouse-local. + + /// Output execution time to stderr in batch mode. if (options.count("time")) - print_time_to_stderr = true; + getClientConfiguration().setBool("print-time-to-stderr", true); + if (options.count("memory-usage")) + { + const auto & memory_usage_mode = options["memory-usage"].as(); + if (memory_usage_mode != "none" && memory_usage_mode != "default" && memory_usage_mode != "readable") + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unknown memory-usage mode: {}", memory_usage_mode); + getClientConfiguration().setString("print-memory-to-stderr", memory_usage_mode); + } + if (options.count("query")) queries = options["query"].as>(); if (options.count("query_id")) - config().setString("query_id", options["query_id"].as()); + getClientConfiguration().setString("query_id", options["query_id"].as()); if (options.count("database")) - config().setString("database", options["database"].as()); + getClientConfiguration().setString("database", options["database"].as()); if (options.count("config-file")) - config().setString("config-file", options["config-file"].as()); + getClientConfiguration().setString("config-file", options["config-file"].as()); if (options.count("queries-file")) queries_files = options["queries-file"].as>(); if (options.count("multiline")) - config().setBool("multiline", true); + getClientConfiguration().setBool("multiline", true); if (options.count("multiquery")) - config().setBool("multiquery", true); + getClientConfiguration().setBool("multiquery", true); if (options.count("ignore-error")) - config().setBool("ignore-error", true); + getClientConfiguration().setBool("ignore-error", true); if (options.count("format")) - config().setString("format", options["format"].as()); + getClientConfiguration().setString("format", options["format"].as()); if (options.count("output-format")) - config().setString("output-format", options["output-format"].as()); + getClientConfiguration().setString("output-format", options["output-format"].as()); if (options.count("vertical")) - config().setBool("vertical", true); + getClientConfiguration().setBool("vertical", true); if (options.count("stacktrace")) - config().setBool("stacktrace", true); + getClientConfiguration().setBool("stacktrace", true); if (options.count("print-profile-events")) - config().setBool("print-profile-events", true); + getClientConfiguration().setBool("print-profile-events", true); if (options.count("profile-events-delay-ms")) - config().setUInt64("profile-events-delay-ms", options["profile-events-delay-ms"].as()); + getClientConfiguration().setUInt64("profile-events-delay-ms", options["profile-events-delay-ms"].as()); + /// Whether to print the number of processed rows at if (options.count("processed-rows")) - print_num_processed_rows = true; + getClientConfiguration().setBool("print-num-processed-rows", true); if (options.count("progress")) { switch (options["progress"].as()) { case DEFAULT: - config().setString("progress", "default"); + getClientConfiguration().setString("progress", "default"); break; case OFF: - config().setString("progress", "off"); + getClientConfiguration().setString("progress", "off"); break; case TTY: - config().setString("progress", "tty"); + getClientConfiguration().setString("progress", "tty"); break; case ERR: - config().setString("progress", "err"); + getClientConfiguration().setString("progress", "err"); break; } } if (options.count("echo")) - config().setBool("echo", true); + getClientConfiguration().setBool("echo", true); if (options.count("disable_suggestion")) - config().setBool("disable_suggestion", true); + getClientConfiguration().setBool("disable_suggestion", true); if (options.count("wait_for_suggestions_to_load")) - config().setBool("wait_for_suggestions_to_load", true); + getClientConfiguration().setBool("wait_for_suggestions_to_load", true); if (options.count("suggestion_limit")) - config().setInt("suggestion_limit", options["suggestion_limit"].as()); + getClientConfiguration().setInt("suggestion_limit", options["suggestion_limit"].as()); if (options.count("highlight")) - config().setBool("highlight", options["highlight"].as()); + getClientConfiguration().setBool("highlight", options["highlight"].as()); if (options.count("history_file")) - config().setString("history_file", options["history_file"].as()); + getClientConfiguration().setString("history_file", options["history_file"].as()); if (options.count("interactive")) - config().setBool("interactive", true); + getClientConfiguration().setBool("interactive", true); if (options.count("pager")) - config().setString("pager", options["pager"].as()); + getClientConfiguration().setString("pager", options["pager"].as()); if (options.count("log-level")) Poco::Logger::root().setLevel(options["log-level"].as()); @@ -3182,13 +3086,13 @@ void ClientBase::init(int argc, char ** argv) alias_names.reserve(options_description.main_description->options().size()); for (const auto& option : options_description.main_description->options()) alias_names.insert(option->long_name()); - argsToConfig(common_arguments, config(), 100, &alias_names); + argsToConfig(common_arguments, getClientConfiguration(), 100, &alias_names); } clearPasswordFromCommandLine(argc, argv); /// Limit on total memory usage - std::string max_client_memory_usage = config().getString("max_memory_usage_in_client", "0" /*default value*/); + std::string max_client_memory_usage = getClientConfiguration().getString("max_memory_usage_in_client", "0" /*default value*/); if (max_client_memory_usage != "0") { UInt64 max_client_memory_usage_int = parseWithSizeSuffix(max_client_memory_usage.c_str(), max_client_memory_usage.length()); @@ -3198,7 +3102,29 @@ void ClientBase::init(int argc, char ** argv) total_memory_tracker.setMetric(CurrentMetrics::MemoryTracking); } - has_log_comment = config().has("log_comment"); + /// Print stacktrace in case of crash + HandledSignals::instance().setupTerminateHandler(); + HandledSignals::instance().setupCommonDeadlySignalHandlers(); + /// We don't setup signal handlers for SIGINT, SIGQUIT, SIGTERM because we don't + /// have an option for client to shutdown gracefully. + + fatal_channel_ptr = new Poco::SplitterChannel; + fatal_console_channel_ptr = new Poco::ConsoleChannel; + fatal_channel_ptr->addChannel(fatal_console_channel_ptr); + if (options.count("client_logs_file")) + { + fatal_file_channel_ptr = new Poco::SimpleFileChannel(options["client_logs_file"].as()); + fatal_channel_ptr->addChannel(fatal_file_channel_ptr); + } + + fatal_log = createLogger("ClientBase", fatal_channel_ptr.get(), Poco::Message::PRIO_FATAL); + signal_listener = std::make_unique(nullptr, fatal_log); + signal_listener_thread.start(*signal_listener); + +#if USE_GWP_ASAN + GWPAsan::initFinished(); +#endif + } } diff --git a/src/Client/ClientBase.h b/src/Client/ClientBase.h index 5202b57040f..986990aecaa 100644 --- a/src/Client/ClientBase.h +++ b/src/Client/ClientBase.h @@ -9,7 +9,11 @@ #include #include #include +#include #include +#include +#include +#include #include #include #include @@ -18,7 +22,6 @@ #include #include - namespace po = boost::program_options; @@ -67,13 +70,22 @@ class ClientBase : public Poco::Util::Application, public IHints<2> public: using Arguments = std::vector; - ClientBase(); + explicit ClientBase + ( + int in_fd_ = STDIN_FILENO, + int out_fd_ = STDOUT_FILENO, + int err_fd_ = STDERR_FILENO, + std::istream & input_stream_ = std::cin, + std::ostream & output_stream_ = std::cout, + std::ostream & error_stream_ = std::cerr + ); + ~ClientBase() override; void init(int argc, char ** argv); std::vector getAllRegisteredNames() const override { return cmd_options; } - static ASTPtr parseQuery(const char *& pos, const char * end, const Settings & settings, bool allow_multi_statements, bool is_interactive, bool ignore_error); + ASTPtr parseQuery(const char *& pos, const char * end, const Settings & settings, bool allow_multi_statements); protected: void runInteractive(); @@ -82,6 +94,9 @@ protected: char * argv0 = nullptr; void runLibFuzzer(); + /// This is the analogue of Poco::Application::config() + virtual Poco::Util::LayeredConfiguration & getClientConfiguration() = 0; + virtual bool processWithFuzzing(const String &) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Query processing with fuzzing is not implemented"); @@ -107,7 +122,7 @@ protected: String & query_to_execute, ASTPtr & parsed_query, const String & all_queries_text, std::unique_ptr & current_exception); - static void clearTerminal(); + void clearTerminal(); void showClientVersion(); using ProgramOptionsDescription = boost::program_options::options_description; @@ -200,13 +215,19 @@ protected: SharedContextHolder shared_context; ContextMutablePtr global_context; + LoggerPtr fatal_log; + Poco::AutoPtr fatal_channel_ptr; + Poco::AutoPtr fatal_console_channel_ptr; + Poco::AutoPtr fatal_file_channel_ptr; + Poco::Thread signal_listener_thread; + std::unique_ptr signal_listener; + bool is_interactive = false; /// Use either interactive line editing interface or batch mode. bool is_multiquery = false; bool delayed_interactive = false; bool echo_queries = false; /// Print queries before execution in batch mode. bool ignore_error = false; /// In case of errors, don't print error message, continue to next query. Only applicable for non-interactive mode. - bool print_time_to_stderr = false; /// Output execution time to stderr in batch mode. std::optional suggest; bool load_suggestions = false; @@ -251,9 +272,9 @@ protected: ConnectionParameters connection_parameters; /// Buffer that reads from stdin in batch mode. - ReadBufferFromFileDescriptor std_in{STDIN_FILENO}; + ReadBufferFromFileDescriptor std_in; /// Console output. - WriteBufferFromFileDescriptor std_out{STDOUT_FILENO}; + WriteBufferFromFileDescriptor std_out; std::unique_ptr pager_cmd; /// The user can specify to redirect query output to a file. @@ -284,7 +305,6 @@ protected: bool need_render_profile_events = true; bool written_first_block = false; size_t processed_rows = 0; /// How many rows have been read or written. - bool print_num_processed_rows = false; /// Whether to print the number of processed rows at bool print_stack_trace = false; /// The last exception that was received from the server. Is used for the @@ -329,11 +349,17 @@ protected: bool allow_repeated_settings = false; bool allow_merge_tree_settings = false; - bool cancelled = false; - bool cancelled_printed = false; + std::atomic_bool cancelled = false; + std::atomic_bool cancelled_printed = false; + + /// Unpacked descriptors and streams for the ease of use. + int in_fd = STDIN_FILENO; + int out_fd = STDOUT_FILENO; + int err_fd = STDERR_FILENO; + std::istream & input_stream; + std::ostream & output_stream; + std::ostream & error_stream; - /// Does log_comment has specified by user? - bool has_log_comment = false; }; } diff --git a/src/Client/ClientBaseOptimizedParts.cpp b/src/Client/ClientBaseOptimizedParts.cpp new file mode 100644 index 00000000000..421843a0e79 --- /dev/null +++ b/src/Client/ClientBaseOptimizedParts.cpp @@ -0,0 +1,176 @@ +#include +#include + +namespace DB +{ + +/** + * Program options parsing is very slow in debug builds and it affects .sh tests + * causing them to timeout sporadically. + * It seems impossible to enable optimizations for a single function (only to disable them), so + * instead we extract the code to a separate source file and compile it with different options. + */ +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; + extern const int UNRECOGNIZED_ARGUMENTS; +} + +namespace +{ + +/// Define transparent hash to we can use +/// std::string_view with the containers +struct TransparentStringHash +{ + using is_transparent = void; + size_t operator()(std::string_view txt) const + { + return std::hash{}(txt); + } +}; + +/* + * This functor is used to parse command line arguments and replace dashes with underscores, + * allowing options to be specified using either dashes or underscores. + */ +class OptionsAliasParser +{ +public: + explicit OptionsAliasParser(const boost::program_options::options_description& options) + { + options_names.reserve(options.options().size()); + for (const auto& option : options.options()) + options_names.insert(option->long_name()); + } + + /* + * Parses arguments by replacing dashes with underscores, and matches the resulting name with known options + * Implements boost::program_options::ext_parser logic + */ + std::pair operator()(const std::string & token) const + { + if (!token.starts_with("--")) + return {}; + std::string arg = token.substr(2); + + // divide token by '=' to separate key and value if options style=long_allow_adjacent + auto pos_eq = arg.find('='); + std::string key = arg.substr(0, pos_eq); + + if (options_names.contains(key)) + // option does not require any changes, because it is already correct + return {}; + + std::replace(key.begin(), key.end(), '-', '_'); + if (!options_names.contains(key)) + // after replacing '-' with '_' argument is still unknown + return {}; + + std::string value; + if (pos_eq != std::string::npos && pos_eq < arg.size()) + value = arg.substr(pos_eq + 1); + + return {key, value}; + } + +private: + std::unordered_set options_names; +}; + +} + +void ClientBase::parseAndCheckOptions(OptionsDescription & options_description, po::variables_map & options, Arguments & arguments) +{ + if (allow_repeated_settings) + addProgramOptionsAsMultitokens(cmd_settings, options_description.main_description.value()); + else + addProgramOptions(cmd_settings, options_description.main_description.value()); + + if (allow_merge_tree_settings) + { + /// Add merge tree settings manually, because names of some settings + /// may clash. Query settings have higher priority and we just + /// skip ambiguous merge tree settings. + auto & main_options = options_description.main_description.value(); + + std::unordered_set> main_option_names; + for (const auto & option : main_options.options()) + main_option_names.insert(option->long_name()); + + for (const auto & setting : cmd_merge_tree_settings.all()) + { + const auto add_setting = [&](const std::string_view name) + { + if (auto it = main_option_names.find(name); it != main_option_names.end()) + return; + + if (allow_repeated_settings) + addProgramOptionAsMultitoken(cmd_merge_tree_settings, main_options, name, setting); + else + addProgramOption(cmd_merge_tree_settings, main_options, name, setting); + }; + + const auto & setting_name = setting.getName(); + + add_setting(setting_name); + + const auto & settings_to_aliases = MergeTreeSettings::Traits::settingsToAliases(); + if (auto it = settings_to_aliases.find(setting_name); it != settings_to_aliases.end()) + { + for (const auto alias : it->second) + { + add_setting(alias); + } + } + } + } + + /// Parse main commandline options. + auto parser = po::command_line_parser(arguments) + .options(options_description.main_description.value()) + .extra_parser(OptionsAliasParser(options_description.main_description.value())) + .allow_unregistered(); + po::parsed_options parsed = parser.run(); + + /// Check unrecognized options without positional options. + auto unrecognized_options = po::collect_unrecognized(parsed.options, po::collect_unrecognized_mode::exclude_positional); + if (!unrecognized_options.empty()) + { + auto hints = this->getHints(unrecognized_options[0]); + if (!hints.empty()) + throw Exception(ErrorCodes::UNRECOGNIZED_ARGUMENTS, "Unrecognized option '{}'. Maybe you meant {}", + unrecognized_options[0], toString(hints)); + + throw Exception(ErrorCodes::UNRECOGNIZED_ARGUMENTS, "Unrecognized option '{}'", unrecognized_options[0]); + } + + /// Check positional options. + for (const auto & op : parsed.options) + { + if (!op.unregistered && op.string_key.empty() && !op.original_tokens[0].starts_with("--") + && !op.original_tokens[0].empty() && !op.value.empty()) + { + /// Two special cases for better usability: + /// - if the option contains a whitespace, it might be a query: clickhouse "SELECT 1" + /// These are relevant for interactive usage - user-friendly, but questionable in general. + /// In case of ambiguity or for scripts, prefer using proper options. + + const auto & token = op.original_tokens[0]; + po::variable_value value(boost::any(op.value), false); + + const char * option; + if (token.contains(' ')) + option = "query"; + else + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Positional option `{}` is not supported.", token); + + if (!options.emplace(option, value).second) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Positional option `{}` is not supported.", token); + } + } + + po::store(parsed, options); +} + +} diff --git a/src/Client/ConnectionEstablisher.cpp b/src/Client/ConnectionEstablisher.cpp index 303105751ad..4bab643ed31 100644 --- a/src/Client/ConnectionEstablisher.cpp +++ b/src/Client/ConnectionEstablisher.cpp @@ -1,6 +1,7 @@ #include #include #include +#include namespace ProfileEvents { @@ -8,6 +9,7 @@ namespace ProfileEvents extern const Event DistributedConnectionUsable; extern const Event DistributedConnectionMissingTable; extern const Event DistributedConnectionStaleReplica; + extern const Event DistributedConnectionFailTry; } namespace DB @@ -97,6 +99,8 @@ void ConnectionEstablisher::run(ConnectionEstablisher::TryResult & result, std:: } catch (const Exception & e) { + ProfileEvents::increment(ProfileEvents::DistributedConnectionFailTry); + if (e.code() != ErrorCodes::NETWORK_ERROR && e.code() != ErrorCodes::SOCKET_TIMEOUT && e.code() != ErrorCodes::ATTEMPT_TO_READ_AFTER_EOF && e.code() != ErrorCodes::DNS_ERROR) throw; diff --git a/src/Client/ConnectionPool.cpp b/src/Client/ConnectionPool.cpp index 5cabb1465d1..ed2e7c3c725 100644 --- a/src/Client/ConnectionPool.cpp +++ b/src/Client/ConnectionPool.cpp @@ -1,4 +1,5 @@ #include +#include #include @@ -85,4 +86,15 @@ ConnectionPoolFactory & ConnectionPoolFactory::instance() return ret; } +IConnectionPool::Entry ConnectionPool::get(const DB::ConnectionTimeouts& timeouts, const DB::Settings& settings, + bool force_connected) +{ + Entry entry = Base::get(settings.connection_pool_max_wait_ms.totalMilliseconds()); + + if (force_connected) + entry->forceConnected(timeouts); + + return entry; +} + } diff --git a/src/Client/ConnectionPool.h b/src/Client/ConnectionPool.h index 725a5e91ac0..0fcb3c4e7e1 100644 --- a/src/Client/ConnectionPool.h +++ b/src/Client/ConnectionPool.h @@ -4,12 +4,13 @@ #include #include #include -#include #include namespace DB { +struct Settings; + /** Interface for connection pools. * * Usage (using the usual `ConnectionPool` example) @@ -102,15 +103,7 @@ public: Entry get(const ConnectionTimeouts & timeouts, /// NOLINT const Settings & settings, - bool force_connected = true) override - { - Entry entry = Base::get(settings.connection_pool_max_wait_ms.totalMilliseconds()); - - if (force_connected) - entry->forceConnected(timeouts); - - return entry; - } + bool force_connected) override; std::string getDescription() const { diff --git a/src/Client/HedgedConnections.cpp b/src/Client/HedgedConnections.cpp index 8c993f906e0..dd8348ea04f 100644 --- a/src/Client/HedgedConnections.cpp +++ b/src/Client/HedgedConnections.cpp @@ -3,6 +3,7 @@ #include #include +#include #include #include @@ -187,15 +188,22 @@ void HedgedConnections::sendQuery( modified_settings.group_by_two_level_threshold_bytes = 0; } - const bool enable_sample_offset_parallel_processing = settings.max_parallel_replicas > 1 && settings.allow_experimental_parallel_reading_from_replicas == 0; + const bool enable_offset_parallel_processing = context->canUseOffsetParallelReplicas(); - if (offset_states.size() > 1 && enable_sample_offset_parallel_processing) + if (offset_states.size() > 1 && enable_offset_parallel_processing) { modified_settings.parallel_replicas_count = offset_states.size(); modified_settings.parallel_replica_offset = fd_to_replica_location[replica.packet_receiver->getFileDescriptor()].offset; } - replica.connection->sendQuery(timeouts, query, /* query_parameters */ {}, query_id, stage, &modified_settings, &client_info, with_pending_data, {}); + /// FIXME: Remove once we will make `allow_experimental_analyzer` obsolete setting. + /// Make the analyzer being set, so it will be effectively applied on the remote server. + /// In other words, the initiator always controls whether the analyzer enabled or not for + /// all servers involved in the distributed query processing. + modified_settings.set("allow_experimental_analyzer", static_cast(modified_settings.allow_experimental_analyzer)); + + replica.connection->sendQuery( + timeouts, query, /* query_parameters */ {}, query_id, stage, &modified_settings, &client_info, with_pending_data, {}); replica.change_replica_timeout.setRelative(timeouts.receive_data_timeout); replica.packet_receiver->setTimeout(hedged_connections_factory.getConnectionTimeouts().receive_timeout); }; diff --git a/src/Client/HedgedConnectionsFactory.cpp b/src/Client/HedgedConnectionsFactory.cpp index 0fa2bc12924..be7397b0fad 100644 --- a/src/Client/HedgedConnectionsFactory.cpp +++ b/src/Client/HedgedConnectionsFactory.cpp @@ -7,7 +7,6 @@ namespace ProfileEvents { extern const Event HedgedRequestsChangeReplica; - extern const Event DistributedConnectionFailTry; extern const Event DistributedConnectionFailAtAll; } @@ -327,7 +326,6 @@ HedgedConnectionsFactory::State HedgedConnectionsFactory::processFinishedConnect { ShuffledPool & shuffled_pool = shuffled_pools[index]; LOG_INFO(log, "Connection failed at try №{}, reason: {}", (shuffled_pool.error_count + 1), fail_message); - ProfileEvents::increment(ProfileEvents::DistributedConnectionFailTry); shuffled_pool.error_count = std::min(pool->getMaxErrorCup(), shuffled_pool.error_count + 1); shuffled_pool.slowdown_count = 0; diff --git a/src/Client/HedgedConnectionsFactory.h b/src/Client/HedgedConnectionsFactory.h index 51da85ea178..63e39819f58 100644 --- a/src/Client/HedgedConnectionsFactory.h +++ b/src/Client/HedgedConnectionsFactory.h @@ -8,13 +8,14 @@ #include #include #include -#include #include #include namespace DB { +struct Settings; + /** Class for establishing hedged connections with replicas. * The process of establishing connection is divided on stages, on each stage if * replica doesn't respond for a long time, we start establishing connection with diff --git a/src/Client/IConnections.h b/src/Client/IConnections.h index ebc71511834..09211de53b0 100644 --- a/src/Client/IConnections.h +++ b/src/Client/IConnections.h @@ -54,8 +54,6 @@ public: struct ReplicaInfo { - bool collaborate_with_initiator{false}; - size_t all_replicas_count{0}; size_t number_of_current_replica{0}; }; diff --git a/src/Client/LineReader.cpp b/src/Client/LineReader.cpp index b3559657ced..487ef232fdc 100644 --- a/src/Client/LineReader.cpp +++ b/src/Client/LineReader.cpp @@ -23,14 +23,6 @@ void trim(String & s) s.erase(std::find_if(s.rbegin(), s.rend(), [](int ch) { return !std::isspace(ch); }).base(), s.end()); } -/// Check if multi-line query is inserted from the paste buffer. -/// Allows delaying the start of query execution until the entirety of query is inserted. -bool hasInputData() -{ - pollfd fd{STDIN_FILENO, POLLIN, 0}; - return poll(&fd, 1, 0) == 1; -} - struct NoCaseCompare { bool operator()(const std::string & str1, const std::string & str2) @@ -63,6 +55,14 @@ void addNewWords(Words & to, const Words & from, Compare comp) namespace DB { +/// Check if multi-line query is inserted from the paste buffer. +/// Allows delaying the start of query execution until the entirety of query is inserted. +bool LineReader::hasInputData() const +{ + pollfd fd{in_fd, POLLIN, 0}; + return poll(&fd, 1, 0) == 1; +} + replxx::Replxx::completions_t LineReader::Suggest::getCompletions(const String & prefix, size_t prefix_length, const char * word_break_characters) { std::string_view last_word; @@ -131,11 +131,22 @@ void LineReader::Suggest::addWords(Words && new_words) // NOLINT(cppcoreguidelin } } -LineReader::LineReader(const String & history_file_path_, bool multiline_, Patterns extenders_, Patterns delimiters_) +LineReader::LineReader( + const String & history_file_path_, + bool multiline_, + Patterns extenders_, + Patterns delimiters_, + std::istream & input_stream_, + std::ostream & output_stream_, + int in_fd_ +) : history_file_path(history_file_path_) , multiline(multiline_) , extenders(std::move(extenders_)) , delimiters(std::move(delimiters_)) + , input_stream(input_stream_) + , output_stream(output_stream_) + , in_fd(in_fd_) { /// FIXME: check extender != delimiter } @@ -212,9 +223,9 @@ LineReader::InputStatus LineReader::readOneLine(const String & prompt) input.clear(); { - std::cout << prompt; - std::getline(std::cin, input); - if (!std::cin.good()) + output_stream << prompt; + std::getline(input_stream, input); + if (!input_stream.good()) return ABORT; } diff --git a/src/Client/LineReader.h b/src/Client/LineReader.h index fc19eaa5667..8c101401190 100644 --- a/src/Client/LineReader.h +++ b/src/Client/LineReader.h @@ -1,5 +1,7 @@ #pragma once +#include +#include #include #include #include @@ -37,7 +39,15 @@ public: using Patterns = std::vector; - LineReader(const String & history_file_path, bool multiline, Patterns extenders, Patterns delimiters); + LineReader( + const String & history_file_path, + bool multiline, + Patterns extenders, + Patterns delimiters, + std::istream & input_stream_ = std::cin, + std::ostream & output_stream_ = std::cout, + int in_fd_ = STDIN_FILENO); + virtual ~LineReader() = default; /// Reads the whole line until delimiter (in multiline mode) or until the last line without extender. @@ -56,6 +66,8 @@ public: virtual void enableBracketedPaste() {} virtual void disableBracketedPaste() {} + bool hasInputData() const; + protected: enum InputStatus { @@ -77,6 +89,10 @@ protected: virtual InputStatus readOneLine(const String & prompt); virtual void addToHistory(const String &) {} + + std::istream & input_stream; + std::ostream & output_stream; + int in_fd; }; } diff --git a/src/Client/LocalConnection.cpp b/src/Client/LocalConnection.cpp index c7494e31605..072184e0a66 100644 --- a/src/Client/LocalConnection.cpp +++ b/src/Client/LocalConnection.cpp @@ -16,7 +16,10 @@ #include #include #include - +#include +#include +#include +#include namespace DB { @@ -151,12 +154,26 @@ void LocalConnection::sendQuery( state->block = sample; String current_format = "Values"; + + const auto & settings = context->getSettingsRef(); const char * begin = state->query.data(); - auto parsed_query = ClientBase::parseQuery(begin, begin + state->query.size(), - context->getSettingsRef(), - /*allow_multi_statements=*/ false, - /*is_interactive=*/ false, - /*ignore_error=*/ false); + const char * end = begin + state->query.size(); + const Dialect & dialect = settings.dialect; + + std::unique_ptr parser; + if (dialect == Dialect::kusto) + parser = std::make_unique(end, settings.allow_settings_after_format_in_insert); + else if (dialect == Dialect::prql) + parser = std::make_unique(settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks); + else + parser = std::make_unique(end, settings.allow_settings_after_format_in_insert); + + ASTPtr parsed_query; + if (dialect == Dialect::kusto) + parsed_query = parseKQLQueryAndMovePosition(*parser, begin, end, "", /*allow_multi_statements*/false, settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks); + else + parsed_query = parseQueryAndMovePosition(*parser, begin, end, "", /*allow_multi_statements*/false, settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks); + if (const auto * insert = parsed_query->as()) { if (!insert->format.empty()) @@ -341,22 +358,18 @@ bool LocalConnection::poll(size_t) if (!state->is_finished) { - if (send_progress && (state->after_send_progress.elapsedMicroseconds() >= query_context->getSettingsRef().interactive_delay)) - { - state->after_send_progress.restart(); - next_packet_type = Protocol::Server::Progress; + if (needSendProgressOrMetrics()) return true; - } - - if (send_profile_events && (state->after_send_profile_events.elapsedMicroseconds() >= query_context->getSettingsRef().interactive_delay)) - { - sendProfileEvents(); - return true; - } try { - pollImpl(); + while (pollImpl()) + { + LOG_DEBUG(&Poco::Logger::get("LocalConnection"), "Executor timeout encountered, will retry"); + + if (needSendProgressOrMetrics()) + return true; + } } catch (const Exception & e) { @@ -451,12 +464,34 @@ bool LocalConnection::poll(size_t) return false; } +bool LocalConnection::needSendProgressOrMetrics() +{ + if (send_progress && (state->after_send_progress.elapsedMicroseconds() >= query_context->getSettingsRef().interactive_delay)) + { + state->after_send_progress.restart(); + next_packet_type = Protocol::Server::Progress; + return true; + } + + if (send_profile_events && (state->after_send_profile_events.elapsedMicroseconds() >= query_context->getSettingsRef().interactive_delay)) + { + sendProfileEvents(); + return true; + } + + return false; +} + bool LocalConnection::pollImpl() { Block block; auto next_read = pullBlock(block); - if (block && !state->io.null_format) + if (!block && next_read) + { + return true; + } + else if (block && !state->io.null_format) { state->block.emplace(block); } @@ -465,7 +500,7 @@ bool LocalConnection::pollImpl() state->is_finished = true; } - return true; + return false; } Packet LocalConnection::receivePacket() diff --git a/src/Client/LocalConnection.h b/src/Client/LocalConnection.h index 899d134cce5..b424c5b5aa3 100644 --- a/src/Client/LocalConnection.h +++ b/src/Client/LocalConnection.h @@ -151,8 +151,11 @@ private: void sendProfileEvents(); + /// Returns true on executor timeout, meaning a retryable error. bool pollImpl(); + bool needSendProgressOrMetrics(); + ContextMutablePtr query_context; Session session; @@ -172,4 +175,5 @@ private: ReadBuffer * in; }; + } diff --git a/src/Client/MultiplexedConnections.cpp b/src/Client/MultiplexedConnections.cpp index 5d0fc8fd39e..244eccf1ed9 100644 --- a/src/Client/MultiplexedConnections.cpp +++ b/src/Client/MultiplexedConnections.cpp @@ -2,6 +2,8 @@ #include #include +#include +#include #include #include #include @@ -23,8 +25,8 @@ namespace ErrorCodes } -MultiplexedConnections::MultiplexedConnections(Connection & connection, const Settings & settings_, const ThrottlerPtr & throttler) - : settings(settings_) +MultiplexedConnections::MultiplexedConnections(Connection & connection, ContextPtr context_, const ThrottlerPtr & throttler) + : context(std::move(context_)), settings(context->getSettingsRef()) { connection.setThrottler(throttler); @@ -36,9 +38,9 @@ MultiplexedConnections::MultiplexedConnections(Connection & connection, const Se } -MultiplexedConnections::MultiplexedConnections(std::shared_ptr connection_ptr_, const Settings & settings_, const ThrottlerPtr & throttler) - : settings(settings_) - , connection_ptr(connection_ptr_) +MultiplexedConnections::MultiplexedConnections( + std::shared_ptr connection_ptr_, ContextPtr context_, const ThrottlerPtr & throttler) + : context(std::move(context_)), settings(context->getSettingsRef()), connection_ptr(connection_ptr_) { connection_ptr->setThrottler(throttler); @@ -50,9 +52,8 @@ MultiplexedConnections::MultiplexedConnections(std::shared_ptr conne } MultiplexedConnections::MultiplexedConnections( - std::vector && connections, - const Settings & settings_, const ThrottlerPtr & throttler) - : settings(settings_) + std::vector && connections, ContextPtr context_, const ThrottlerPtr & throttler) + : context(std::move(context_)), settings(context->getSettingsRef()) { /// If we didn't get any connections from pool and getMany() did not throw exceptions, this means that /// `skip_unavailable_shards` was set. Then just return. @@ -141,27 +142,32 @@ void MultiplexedConnections::sendQuery( modified_settings.group_by_two_level_threshold = 0; modified_settings.group_by_two_level_threshold_bytes = 0; } - - if (replica_info) - { - client_info.collaborate_with_initiator = true; - client_info.count_participating_replicas = replica_info->all_replicas_count; - client_info.number_of_current_replica = replica_info->number_of_current_replica; - } } - const bool enable_sample_offset_parallel_processing = settings.max_parallel_replicas > 1 && settings.allow_experimental_parallel_reading_from_replicas == 0; + if (replica_info) + { + client_info.collaborate_with_initiator = true; + client_info.number_of_current_replica = replica_info->number_of_current_replica; + } + + /// FIXME: Remove once we will make `allow_experimental_analyzer` obsolete setting. + /// Make the analyzer being set, so it will be effectively applied on the remote server. + /// In other words, the initiator always controls whether the analyzer enabled or not for + /// all servers involved in the distributed query processing. + modified_settings.set("allow_experimental_analyzer", static_cast(modified_settings.allow_experimental_analyzer)); + + const bool enable_offset_parallel_processing = context->canUseOffsetParallelReplicas(); size_t num_replicas = replica_states.size(); if (num_replicas > 1) { - if (enable_sample_offset_parallel_processing) + if (enable_offset_parallel_processing) /// Use multiple replicas for parallel query processing. modified_settings.parallel_replicas_count = num_replicas; for (size_t i = 0; i < num_replicas; ++i) { - if (enable_sample_offset_parallel_processing) + if (enable_offset_parallel_processing) modified_settings.parallel_replica_offset = i; replica_states[i].connection->sendQuery( diff --git a/src/Client/MultiplexedConnections.h b/src/Client/MultiplexedConnections.h index 9f7b47e0562..dec32e52d4f 100644 --- a/src/Client/MultiplexedConnections.h +++ b/src/Client/MultiplexedConnections.h @@ -10,7 +10,6 @@ namespace DB { - /** To retrieve data directly from multiple replicas (connections) from one shard * within a single thread. As a degenerate case, it can also work with one connection. * It is assumed that all functions except sendCancel are always executed in one thread. @@ -21,14 +20,12 @@ class MultiplexedConnections final : public IConnections { public: /// Accepts ready connection. - MultiplexedConnections(Connection & connection, const Settings & settings_, const ThrottlerPtr & throttler_); + MultiplexedConnections(Connection & connection, ContextPtr context_, const ThrottlerPtr & throttler_); /// Accepts ready connection and keep it alive before drain - MultiplexedConnections(std::shared_ptr connection_, const Settings & settings_, const ThrottlerPtr & throttler_); + MultiplexedConnections(std::shared_ptr connection_, ContextPtr context_, const ThrottlerPtr & throttler_); /// Accepts a vector of connections to replicas of one shard already taken from pool. - MultiplexedConnections( - std::vector && connections, - const Settings & settings_, const ThrottlerPtr & throttler_); + MultiplexedConnections(std::vector && connections, ContextPtr context_, const ThrottlerPtr & throttler_); void sendScalarsData(Scalars & data) override; void sendExternalTablesData(std::vector & data) override; @@ -86,6 +83,7 @@ private: /// Mark the replica as invalid. void invalidateReplica(ReplicaState & replica_state); + ContextPtr context; const Settings & settings; /// The current number of valid connections to the replicas of this shard. diff --git a/src/Client/ReplxxLineReader.cpp b/src/Client/ReplxxLineReader.cpp index 9e0f5946205..3b3508d1a58 100644 --- a/src/Client/ReplxxLineReader.cpp +++ b/src/Client/ReplxxLineReader.cpp @@ -297,8 +297,15 @@ ReplxxLineReader::ReplxxLineReader( Patterns extenders_, Patterns delimiters_, const char word_break_characters_[], - replxx::Replxx::highlighter_callback_t highlighter_) - : LineReader(history_file_path_, multiline_, std::move(extenders_), std::move(delimiters_)), highlighter(std::move(highlighter_)) + replxx::Replxx::highlighter_callback_t highlighter_, + [[ maybe_unused ]] std::istream & input_stream_, + [[ maybe_unused ]] std::ostream & output_stream_, + [[ maybe_unused ]] int in_fd_, + [[ maybe_unused ]] int out_fd_, + [[ maybe_unused ]] int err_fd_ +) + : LineReader(history_file_path_, multiline_, std::move(extenders_), std::move(delimiters_), input_stream_, output_stream_, in_fd_) + , highlighter(std::move(highlighter_)) , word_break_characters(word_break_characters_) , editor(getEditor()) { @@ -355,6 +362,9 @@ ReplxxLineReader::ReplxxLineReader( rx.bind_key(Replxx::KEY::control('N'), [this](char32_t code) { return rx.invoke(Replxx::ACTION::HISTORY_NEXT, code); }); rx.bind_key(Replxx::KEY::control('P'), [this](char32_t code) { return rx.invoke(Replxx::ACTION::HISTORY_PREVIOUS, code); }); + /// We don't want the default, "suspend" behavior, it confuses people. + rx.bind_key_internal(replxx::Replxx::KEY::control('Z'), "insert_character"); + auto commit_action = [this](char32_t code) { /// If we allow multiline and there is already something in the input, start a newline. @@ -471,7 +481,7 @@ ReplxxLineReader::ReplxxLineReader( ReplxxLineReader::~ReplxxLineReader() { - if (close(history_file_fd)) + if (history_file_fd >= 0 && close(history_file_fd)) rx.print("Close of history file failed: %s\n", errnoToString().c_str()); } @@ -496,7 +506,7 @@ void ReplxxLineReader::addToHistory(const String & line) // but replxx::Replxx::history_load() does not // and that is why flock() is added here. bool locked = false; - if (flock(history_file_fd, LOCK_EX)) + if (history_file_fd >= 0 && flock(history_file_fd, LOCK_EX)) rx.print("Lock of history file failed: %s\n", errnoToString().c_str()); else locked = true; @@ -507,7 +517,7 @@ void ReplxxLineReader::addToHistory(const String & line) if (!rx.history_save(history_file_path)) rx.print("Saving history failed: %s\n", errnoToString().c_str()); - if (locked && 0 != flock(history_file_fd, LOCK_UN)) + if (history_file_fd >= 0 && locked && 0 != flock(history_file_fd, LOCK_UN)) rx.print("Unlock of history file failed: %s\n", errnoToString().c_str()); } diff --git a/src/Client/ReplxxLineReader.h b/src/Client/ReplxxLineReader.h index 6ad149e38f2..c46080420ef 100644 --- a/src/Client/ReplxxLineReader.h +++ b/src/Client/ReplxxLineReader.h @@ -1,6 +1,7 @@ #pragma once -#include "LineReader.h" +#include +#include #include namespace DB @@ -9,14 +10,22 @@ namespace DB class ReplxxLineReader : public LineReader { public: - ReplxxLineReader( + ReplxxLineReader + ( Suggest & suggest, const String & history_file_path, bool multiline, Patterns extenders_, Patterns delimiters_, const char word_break_characters_[], - replxx::Replxx::highlighter_callback_t highlighter_); + replxx::Replxx::highlighter_callback_t highlighter_, + std::istream & input_stream_ = std::cin, + std::ostream & output_stream_ = std::cout, + int in_fd_ = STDIN_FILENO, + int out_fd_ = STDOUT_FILENO, + int err_fd_ = STDERR_FILENO + ); + ~ReplxxLineReader() override; void enableBracketedPaste() override; diff --git a/src/Client/TestHint.cpp b/src/Client/TestHint.cpp index b64882577ee..74c65009a73 100644 --- a/src/Client/TestHint.cpp +++ b/src/Client/TestHint.cpp @@ -10,6 +10,7 @@ namespace DB::ErrorCodes { extern const int CANNOT_PARSE_TEXT; + extern const int OK; } namespace DB @@ -62,9 +63,28 @@ bool TestHint::hasExpectedServerError(int error) return std::find(server_errors.begin(), server_errors.end(), error) != server_errors.end(); } +bool TestHint::needRetry(const std::unique_ptr & server_exception, size_t * retries_counter) +{ + chassert(retries_counter); + if (max_retries <= *retries_counter) + return false; + + ++*retries_counter; + + int error = ErrorCodes::OK; + if (server_exception) + error = server_exception->code(); + + + if (retry_until) + return !hasExpectedServerError(error); /// retry until we get the expected error + else + return hasExpectedServerError(error); /// retry while we have the expected error +} + void TestHint::parse(Lexer & comment_lexer, bool is_leading_hint) { - std::unordered_set commands{"echo", "echoOn", "echoOff"}; + std::unordered_set commands{"echo", "echoOn", "echoOff", "retry"}; std::unordered_set command_errors{ "serverError", @@ -73,6 +93,9 @@ void TestHint::parse(Lexer & comment_lexer, bool is_leading_hint) for (Token token = comment_lexer.nextToken(); !token.isEnd(); token = comment_lexer.nextToken()) { + if (token.type == TokenType::Whitespace) + continue; + String item = String(token.begin, token.end); if (token.type == TokenType::BareWord && commands.contains(item)) { @@ -82,6 +105,30 @@ void TestHint::parse(Lexer & comment_lexer, bool is_leading_hint) echo.emplace(true); if (item == "echoOff") echo.emplace(false); + + if (item == "retry") + { + token = comment_lexer.nextToken(); + while (token.type == TokenType::Whitespace) + token = comment_lexer.nextToken(); + + if (token.type != TokenType::Number) + throw DB::Exception(DB::ErrorCodes::CANNOT_PARSE_TEXT, "Could not parse the number of retries: {}", + std::string_view(token.begin, token.end)); + + max_retries = std::stoul(std::string(token.begin, token.end)); + + token = comment_lexer.nextToken(); + while (token.type == TokenType::Whitespace) + token = comment_lexer.nextToken(); + + if (token.type != TokenType::BareWord || + (std::string_view(token.begin, token.end) != "until" && + std::string_view(token.begin, token.end) != "while")) + throw DB::Exception(DB::ErrorCodes::CANNOT_PARSE_TEXT, "Expected 'until' or 'while' after the number of retries, got: {}", + std::string_view(token.begin, token.end)); + retry_until = std::string_view(token.begin, token.end) == "until"; + } } else if (!is_leading_hint && token.type == TokenType::BareWord && command_errors.contains(item)) { @@ -133,6 +180,9 @@ void TestHint::parse(Lexer & comment_lexer, bool is_leading_hint) break; } } + + if (max_retries && server_errors.size() != 1) + throw DB::Exception(DB::ErrorCodes::CANNOT_PARSE_TEXT, "Expected one serverError after the 'retry N while|until' command"); } } diff --git a/src/Client/TestHint.h b/src/Client/TestHint.h index b76c4245df4..bbe7873c08b 100644 --- a/src/Client/TestHint.h +++ b/src/Client/TestHint.h @@ -6,6 +6,7 @@ #include #include +#include namespace DB @@ -65,12 +66,17 @@ public: bool hasExpectedClientError(int error); bool hasExpectedServerError(int error); + bool needRetry(const std::unique_ptr & server_exception, size_t * retries_counter); + private: const String & query; ErrorVector server_errors{}; ErrorVector client_errors{}; std::optional echo; + size_t max_retries = 0; + bool retry_until = false; + void parse(Lexer & comment_lexer, bool is_leading_hint); bool allErrorsExpected(int actual_server_error, int actual_client_error) const diff --git a/src/Columns/ColumnAggregateFunction.cpp b/src/Columns/ColumnAggregateFunction.cpp index f7e6b1a1ccc..e26fe790a8e 100644 --- a/src/Columns/ColumnAggregateFunction.cpp +++ b/src/Columns/ColumnAggregateFunction.cpp @@ -267,7 +267,11 @@ bool ColumnAggregateFunction::structureEquals(const IColumn & to) const } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnAggregateFunction::insertRangeFrom(const IColumn & from, size_t start, size_t length) +#else +void ColumnAggregateFunction::doInsertRangeFrom(const IColumn & from, size_t start, size_t length) +#endif { const ColumnAggregateFunction & from_concrete = assert_cast(from); @@ -362,13 +366,10 @@ void ColumnAggregateFunction::updateHashWithValue(size_t n, SipHash & hash) cons hash.update(wbuf.str().c_str(), wbuf.str().size()); } -void ColumnAggregateFunction::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnAggregateFunction::getWeakHash32() const { auto s = data.size(); - if (hash.getData().size() != data.size()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), hash.getData().size()); - + WeakHash32 hash(s); auto & hash_data = hash.getData(); std::vector v; @@ -379,6 +380,8 @@ void ColumnAggregateFunction::updateWeakHash32(WeakHash32 & hash) const wbuf.finalize(); hash_data[i] = ::updateWeakHash32(v.data(), v.size(), hash_data[i]); } + + return hash; } void ColumnAggregateFunction::updateHashFast(SipHash & hash) const @@ -462,7 +465,11 @@ void ColumnAggregateFunction::insertFromWithOwnership(const IColumn & from, size insertMergeFrom(from, n); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnAggregateFunction::insertFrom(const IColumn & from, size_t n) +#else +void ColumnAggregateFunction::doInsertFrom(const IColumn & from, size_t n) +#endif { insertRangeFrom(from, n, 1); } diff --git a/src/Columns/ColumnAggregateFunction.h b/src/Columns/ColumnAggregateFunction.h index a75b27e835c..b581c3ba3b4 100644 --- a/src/Columns/ColumnAggregateFunction.h +++ b/src/Columns/ColumnAggregateFunction.h @@ -145,7 +145,14 @@ public: void insertData(const char * pos, size_t length) override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & from, size_t n) override; +#else + using IColumn::insertFrom; + + void doInsertFrom(const IColumn & from, size_t n) override; +#endif + void insertFrom(ConstAggregateDataPtr place); @@ -170,7 +177,7 @@ public: void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; @@ -182,7 +189,11 @@ public: void protect() override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & from, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & from, size_t start, size_t length) override; +#endif void popBack(size_t n) override; @@ -201,7 +212,11 @@ public: MutableColumns scatter(ColumnIndex num_columns, const Selector & selector) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t, size_t, const IColumn &, int) const override +#else + int doCompareAt(size_t, size_t, const IColumn &, int) const override +#endif { return 0; } diff --git a/src/Columns/ColumnArray.cpp b/src/Columns/ColumnArray.cpp index 0b7e6541560..19cce678cc7 100644 --- a/src/Columns/ColumnArray.cpp +++ b/src/Columns/ColumnArray.cpp @@ -271,15 +271,12 @@ void ColumnArray::updateHashWithValue(size_t n, SipHash & hash) const getData().updateHashWithValue(offset + i, hash); } -void ColumnArray::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnArray::getWeakHash32() const { auto s = offsets->size(); - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", s, hash.getData().size()); + WeakHash32 hash(s); - WeakHash32 internal_hash(data->size()); - data->updateWeakHash32(internal_hash); + WeakHash32 internal_hash = data->getWeakHash32(); Offset prev_offset = 0; const auto & offsets_data = getOffsets(); @@ -300,6 +297,8 @@ void ColumnArray::updateWeakHash32(WeakHash32 & hash) const prev_offset = offsets_data[i]; } + + return hash; } void ColumnArray::updateHashFast(SipHash & hash) const @@ -337,7 +336,11 @@ bool ColumnArray::tryInsert(const Field & x) return true; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnArray::insertFrom(const IColumn & src_, size_t n) +#else +void ColumnArray::doInsertFrom(const IColumn & src_, size_t n) +#endif { const ColumnArray & src = assert_cast(src_); size_t size = src.sizeAt(n); @@ -392,7 +395,11 @@ int ColumnArray::compareAtImpl(size_t n, size_t m, const IColumn & rhs_, int nan : 1); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnArray::compareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const +#else +int ColumnArray::doCompareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const +#endif { return compareAtImpl(n, m, rhs_, nan_direction_hint); } @@ -535,7 +542,11 @@ void ColumnArray::getExtremes(Field & min, Field & max) const } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnArray::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnArray::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { if (length == 0) return; diff --git a/src/Columns/ColumnArray.h b/src/Columns/ColumnArray.h index 53eb5166df8..63affb86d9d 100644 --- a/src/Columns/ColumnArray.h +++ b/src/Columns/ColumnArray.h @@ -82,12 +82,20 @@ public: const char * deserializeAndInsertFromArena(const char * pos) override; const char * skipSerializedInArena(const char * pos) const override; void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif void insert(const Field & x) override; bool tryInsert(const Field & x) override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src_, size_t n) override; +#else + void doInsertFrom(const IColumn & src_, size_t n) override; +#endif void insertDefault() override; void popBack(size_t n) override; ColumnPtr filter(const Filter & filt, ssize_t result_size_hint) const override; @@ -95,7 +103,11 @@ public: ColumnPtr permute(const Permutation & perm, size_t limit) const override; ColumnPtr index(const IColumn & indexes, size_t limit) const override; template ColumnPtr indexImpl(const PaddedPODArray & indexes, size_t limit) const; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const override; +#endif int compareAtWithCollation(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint, const Collator & collator) const override; void getPermutation(PermutationSortDirection direction, PermutationSortStability stability, size_t limit, int nan_direction_hint, Permutation & res) const override; diff --git a/src/Columns/ColumnCompressed.h b/src/Columns/ColumnCompressed.h index 934adf07cf4..c4270e8216b 100644 --- a/src/Columns/ColumnCompressed.h +++ b/src/Columns/ColumnCompressed.h @@ -3,6 +3,7 @@ #include #include #include +#include #include @@ -85,7 +86,11 @@ public: bool isDefaultAt(size_t) const override { throwMustBeDecompressed(); } void insert(const Field &) override { throwMustBeDecompressed(); } bool tryInsert(const Field &) override { throwMustBeDecompressed(); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn &, size_t, size_t) override { throwMustBeDecompressed(); } +#else + void doInsertRangeFrom(const IColumn &, size_t, size_t) override { throwMustBeDecompressed(); } +#endif void insertData(const char *, size_t) override { throwMustBeDecompressed(); } void insertDefault() override { throwMustBeDecompressed(); } void popBack(size_t) override { throwMustBeDecompressed(); } @@ -94,13 +99,17 @@ public: const char * deserializeAndInsertFromArena(const char *) override { throwMustBeDecompressed(); } const char * skipSerializedInArena(const char *) const override { throwMustBeDecompressed(); } void updateHashWithValue(size_t, SipHash &) const override { throwMustBeDecompressed(); } - void updateWeakHash32(WeakHash32 &) const override { throwMustBeDecompressed(); } + WeakHash32 getWeakHash32() const override { throwMustBeDecompressed(); } void updateHashFast(SipHash &) const override { throwMustBeDecompressed(); } ColumnPtr filter(const Filter &, ssize_t) const override { throwMustBeDecompressed(); } void expand(const Filter &, bool) override { throwMustBeDecompressed(); } ColumnPtr permute(const Permutation &, size_t) const override { throwMustBeDecompressed(); } ColumnPtr index(const IColumn &, size_t) const override { throwMustBeDecompressed(); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t, size_t, const IColumn &, int) const override { throwMustBeDecompressed(); } +#else + int doCompareAt(size_t, size_t, const IColumn &, int) const override { throwMustBeDecompressed(); } +#endif void compareColumn(const IColumn &, size_t, PaddedPODArray *, PaddedPODArray &, int, int) const override { throwMustBeDecompressed(); diff --git a/src/Columns/ColumnConst.cpp b/src/Columns/ColumnConst.cpp index f2cea83db0e..84427e7be2b 100644 --- a/src/Columns/ColumnConst.cpp +++ b/src/Columns/ColumnConst.cpp @@ -137,18 +137,10 @@ void ColumnConst::updatePermutation(PermutationSortDirection /*direction*/, Perm { } -void ColumnConst::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnConst::getWeakHash32() const { - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); - - WeakHash32 element_hash(1); - data->updateWeakHash32(element_hash); - size_t data_hash = element_hash.getData()[0]; - - for (auto & value : hash.getData()) - value = static_cast(intHashCRC32(data_hash, value)); + WeakHash32 element_hash = data->getWeakHash32(); + return WeakHash32(s, element_hash.getData()[0]); } void ColumnConst::compareColumn( diff --git a/src/Columns/ColumnConst.h b/src/Columns/ColumnConst.h index c2c0fa3027c..ca38e76ea57 100644 --- a/src/Columns/ColumnConst.h +++ b/src/Columns/ColumnConst.h @@ -32,6 +32,8 @@ private: ColumnConst(const ColumnConst & src) = default; public: + bool isConst() const override { return true; } + ColumnPtr convertToFullColumn() const; ColumnPtr convertToFullColumnIfConst() const override @@ -121,7 +123,11 @@ public: return data->isNullAt(0); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn &, size_t /*start*/, size_t length) override +#else + void doInsertRangeFrom(const IColumn &, size_t /*start*/, size_t length) override +#endif { s += length; } @@ -145,12 +151,20 @@ public: ++s; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn &, size_t) override +#else + void doInsertFrom(const IColumn &, size_t) override +#endif { ++s; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertManyFrom(const IColumn & /*src*/, size_t /* position */, size_t length) override { s += length; } +#else + void doInsertManyFrom(const IColumn & /*src*/, size_t /* position */, size_t length) override { s += length; } +#endif void insertDefault() override { @@ -190,7 +204,7 @@ public: data->updateHashWithValue(0, hash); } - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override { @@ -223,7 +237,11 @@ public: return data->allocatedBytes() + sizeof(s); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t, size_t, const IColumn & rhs, int nan_direction_hint) const override +#else + int doCompareAt(size_t, size_t, const IColumn & rhs, int nan_direction_hint) const override +#endif { return data->compareAt(0, 0, *assert_cast(rhs).data, nan_direction_hint); } diff --git a/src/Columns/ColumnDecimal.cpp b/src/Columns/ColumnDecimal.cpp index eb9784c14dd..bb4433f8956 100644 --- a/src/Columns/ColumnDecimal.cpp +++ b/src/Columns/ColumnDecimal.cpp @@ -28,11 +28,14 @@ namespace ErrorCodes extern const int PARAMETER_OUT_OF_BOUND; extern const int SIZES_OF_COLUMNS_DOESNT_MATCH; extern const int NOT_IMPLEMENTED; - extern const int LOGICAL_ERROR; } template +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnDecimal::compareAt(size_t n, size_t m, const IColumn & rhs_, int) const +#else +int ColumnDecimal::doCompareAt(size_t n, size_t m, const IColumn & rhs_, int) const +#endif { auto & other = static_cast(rhs_); const T & a = data[n]; @@ -72,13 +75,10 @@ void ColumnDecimal::updateHashWithValue(size_t n, SipHash & hash) const } template -void ColumnDecimal::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnDecimal::getWeakHash32() const { auto s = data.size(); - - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); + WeakHash32 hash(s); const T * begin = data.data(); const T * end = begin + s; @@ -90,6 +90,8 @@ void ColumnDecimal::updateWeakHash32(WeakHash32 & hash) const ++begin; ++hash_data; } + + return hash; } template @@ -331,7 +333,11 @@ void ColumnDecimal::insertData(const char * src, size_t /*length*/) } template +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnDecimal::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnDecimal::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { const ColumnDecimal & src_vec = assert_cast(src); diff --git a/src/Columns/ColumnDecimal.h b/src/Columns/ColumnDecimal.h index c4510ba2922..59bfbd2159c 100644 --- a/src/Columns/ColumnDecimal.h +++ b/src/Columns/ColumnDecimal.h @@ -55,9 +55,17 @@ public: void reserve(size_t n) override { data.reserve_exact(n); } void shrinkToFit() override { data.shrink_to_fit(); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src, size_t n) override { data.push_back(static_cast(src).getData()[n]); } +#else + void doInsertFrom(const IColumn & src, size_t n) override { data.push_back(static_cast(src).getData()[n]); } +#endif +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertManyFrom(const IColumn & src, size_t position, size_t length) override +#else + void doInsertManyFrom(const IColumn & src, size_t position, size_t length) override +#endif { ValueType v = assert_cast(src).getData()[position]; data.resize_fill(data.size() + length, v); @@ -68,7 +76,11 @@ public: void insertManyDefaults(size_t length) override { data.resize_fill(data.size() + length); } void insert(const Field & x) override { data.push_back(x.get()); } bool tryInsert(const Field & x) override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif void popBack(size_t n) override { @@ -90,9 +102,13 @@ public: const char * deserializeAndInsertFromArena(const char * pos) override; const char * skipSerializedInArena(const char * pos) const override; void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const override; +#endif void getPermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, size_t limit, int nan_direction_hint, IColumn::Permutation & res) const override; void updatePermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, diff --git a/src/Columns/ColumnDynamic.cpp b/src/Columns/ColumnDynamic.cpp index 3c147b6f123..a92d54dd675 100644 --- a/src/Columns/ColumnDynamic.cpp +++ b/src/Columns/ColumnDynamic.cpp @@ -4,7 +4,9 @@ #include #include #include +#include #include +#include #include #include #include @@ -213,7 +215,11 @@ bool ColumnDynamic::tryInsert(const DB::Field & x) } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnDynamic::insertFrom(const DB::IColumn & src_, size_t n) +#else +void ColumnDynamic::doInsertFrom(const DB::IColumn & src_, size_t n) +#endif { const auto & dynamic_src = assert_cast(src_); @@ -263,7 +269,11 @@ void ColumnDynamic::insertFrom(const DB::IColumn & src_, size_t n) variant_col.insertIntoVariantFrom(string_variant_discr, *tmp_string_column, 0); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnDynamic::insertRangeFrom(const DB::IColumn & src_, size_t start, size_t length) +#else +void ColumnDynamic::doInsertRangeFrom(const DB::IColumn & src_, size_t start, size_t length) +#endif { if (start + length > src_.size()) throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND, "Parameter out of bound in ColumnDynamic::insertRangeFrom method. " @@ -429,7 +439,11 @@ void ColumnDynamic::insertRangeFrom(const DB::IColumn & src_, size_t start, size } } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnDynamic::insertManyFrom(const DB::IColumn & src_, size_t position, size_t length) +#else +void ColumnDynamic::doInsertManyFrom(const DB::IColumn & src_, size_t position, size_t length) +#endif { const auto & dynamic_src = assert_cast(src_); @@ -481,7 +495,7 @@ StringRef ColumnDynamic::serializeValueIntoArena(size_t n, DB::Arena & arena, co /// We cannot use Variant serialization here as it serializes discriminator + value, /// but Dynamic doesn't have fixed mapping discriminator <-> variant type /// as different Dynamic column can have different Variants. - /// Instead, we serialize null bit + variant type name (size + bytes) + value. + /// Instead, we serialize null bit + variant type in binary format (size + bytes) + value. const auto & variant_col = assert_cast(*variant_column); auto discr = variant_col.globalDiscriminatorAt(n); StringRef res; @@ -495,14 +509,15 @@ StringRef ColumnDynamic::serializeValueIntoArena(size_t n, DB::Arena & arena, co return res; } - const auto & variant_name = variant_info.variant_names[discr]; - size_t variant_name_size = variant_name.size(); - char * pos = arena.allocContinue(sizeof(UInt8) + sizeof(size_t) + variant_name.size(), begin); + const auto & variant_type = assert_cast(*variant_info.variant_type).getVariant(discr); + String variant_type_binary_data = encodeDataType(variant_type); + size_t variant_type_binary_data_size = variant_type_binary_data.size(); + char * pos = arena.allocContinue(sizeof(UInt8) + sizeof(size_t) + variant_type_binary_data.size(), begin); memcpy(pos, &null_bit, sizeof(UInt8)); - memcpy(pos + sizeof(UInt8), &variant_name_size, sizeof(size_t)); - memcpy(pos + sizeof(UInt8) + sizeof(size_t), variant_name.data(), variant_name.size()); + memcpy(pos + sizeof(UInt8), &variant_type_binary_data_size, sizeof(size_t)); + memcpy(pos + sizeof(UInt8) + sizeof(size_t), variant_type_binary_data.data(), variant_type_binary_data.size()); res.data = pos; - res.size = sizeof(UInt8) + sizeof(size_t) + variant_name.size(); + res.size = sizeof(UInt8) + sizeof(size_t) + variant_type_binary_data.size(); auto value_ref = variant_col.getVariantByGlobalDiscriminator(discr).serializeValueIntoArena(variant_col.offsetAt(n), arena, begin); res.data = value_ref.data - res.size; @@ -521,13 +536,15 @@ const char * ColumnDynamic::deserializeAndInsertFromArena(const char * pos) return pos; } - /// Read variant type name. - const size_t variant_name_size = unalignedLoad(pos); - pos += sizeof(variant_name_size); - String variant_name; - variant_name.resize(variant_name_size); - memcpy(variant_name.data(), pos, variant_name_size); - pos += variant_name_size; + /// Read variant type in binary format. + const size_t variant_type_binary_data_size = unalignedLoad(pos); + pos += sizeof(variant_type_binary_data_size); + String variant_type_binary_data; + variant_type_binary_data.resize(variant_type_binary_data_size); + memcpy(variant_type_binary_data.data(), pos, variant_type_binary_data_size); + pos += variant_type_binary_data_size; + auto variant_type = decodeDataType(variant_type_binary_data); + auto variant_name = variant_type->getName(); /// If we already have such variant, just deserialize it into corresponding variant column. auto it = variant_info.variant_name_to_discriminator.find(variant_name); if (it != variant_info.variant_name_to_discriminator.end()) @@ -537,7 +554,6 @@ const char * ColumnDynamic::deserializeAndInsertFromArena(const char * pos) } /// If we don't have such variant, add it. - auto variant_type = DataTypeFactory::instance().get(variant_name); if (likely(addNewVariant(variant_type))) { auto discr = variant_info.variant_name_to_discriminator[variant_name]; @@ -563,13 +579,13 @@ const char * ColumnDynamic::skipSerializedInArena(const char * pos) const if (null_bit) return pos; - const size_t variant_name_size = unalignedLoad(pos); - pos += sizeof(variant_name_size); - String variant_name; - variant_name.resize(variant_name_size); - memcpy(variant_name.data(), pos, variant_name_size); - pos += variant_name_size; - auto tmp_variant_column = DataTypeFactory::instance().get(variant_name)->createColumn(); + const size_t variant_type_binary_data_size = unalignedLoad(pos); + pos += sizeof(variant_type_binary_data_size); + String variant_type_binary_data; + variant_type_binary_data.resize(variant_type_binary_data_size); + memcpy(variant_type_binary_data.data(), pos, variant_type_binary_data_size); + pos += variant_type_binary_data_size; + auto tmp_variant_column = decodeDataType(variant_type_binary_data)->createColumn(); return tmp_variant_column->skipSerializedInArena(pos); } @@ -587,7 +603,11 @@ void ColumnDynamic::updateHashWithValue(size_t n, SipHash & hash) const variant_col.getVariantByGlobalDiscriminator(discr).updateHashWithValue(variant_col.offsetAt(n), hash); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnDynamic::compareAt(size_t n, size_t m, const DB::IColumn & rhs, int nan_direction_hint) const +#else +int ColumnDynamic::doCompareAt(size_t n, size_t m, const DB::IColumn & rhs, int nan_direction_hint) const +#endif { const auto & left_variant = assert_cast(*variant_column); const auto & right_dynamic = assert_cast(rhs); diff --git a/src/Columns/ColumnDynamic.h b/src/Columns/ColumnDynamic.h index 27ad0dd583f..e92cabd3db9 100644 --- a/src/Columns/ColumnDynamic.h +++ b/src/Columns/ColumnDynamic.h @@ -4,6 +4,7 @@ #include #include #include +#include namespace DB @@ -142,9 +143,16 @@ public: void insert(const Field & x) override; bool tryInsert(const Field & x) override; + +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src_, size_t n) override; void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; void insertManyFrom(const IColumn & src, size_t position, size_t length) override; +#else + void doInsertFrom(const IColumn & src_, size_t n) override; + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; + void doInsertManyFrom(const IColumn & src, size_t position, size_t length) override; +#endif void insertDefault() override { @@ -167,9 +175,9 @@ public: void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override + WeakHash32 getWeakHash32() const override { - variant_column->updateWeakHash32(hash); + return variant_column->getWeakHash32(); } void updateHashFast(SipHash & hash) const override @@ -213,7 +221,11 @@ public: return scattered_columns; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#endif bool hasEqualValues() const override { diff --git a/src/Columns/ColumnFixedString.cpp b/src/Columns/ColumnFixedString.cpp index d7e4eff2727..0bb3f7edb14 100644 --- a/src/Columns/ColumnFixedString.cpp +++ b/src/Columns/ColumnFixedString.cpp @@ -74,7 +74,11 @@ bool ColumnFixedString::tryInsert(const Field & x) return true; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnFixedString::insertFrom(const IColumn & src_, size_t index) +#else +void ColumnFixedString::doInsertFrom(const IColumn & src_, size_t index) +#endif { const ColumnFixedString & src = assert_cast(src_); @@ -86,7 +90,11 @@ void ColumnFixedString::insertFrom(const IColumn & src_, size_t index) memcpySmallAllowReadWriteOverflow15(chars.data() + old_size, &src.chars[n * index], n); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnFixedString::insertManyFrom(const IColumn & src, size_t position, size_t length) +#else +void ColumnFixedString::doInsertManyFrom(const IColumn & src, size_t position, size_t length) +#endif { const ColumnFixedString & src_concrete = assert_cast(src); if (n != src_concrete.getN()) @@ -129,14 +137,10 @@ void ColumnFixedString::updateHashWithValue(size_t index, SipHash & hash) const hash.update(reinterpret_cast(&chars[n * index]), n); } -void ColumnFixedString::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnFixedString::getWeakHash32() const { auto s = size(); - - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, " - "hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); + WeakHash32 hash(s); const UInt8 * pos = chars.data(); UInt32 * hash_data = hash.getData().data(); @@ -148,6 +152,8 @@ void ColumnFixedString::updateWeakHash32(WeakHash32 & hash) const pos += n; ++hash_data; } + + return hash; } void ColumnFixedString::updateHashFast(SipHash & hash) const @@ -219,7 +225,11 @@ size_t ColumnFixedString::estimateCardinalityInPermutedRange(const Permutation & return elements.size(); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnFixedString::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnFixedString::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { const ColumnFixedString & src_concrete = assert_cast(src); chassert(this->n == src_concrete.n); diff --git a/src/Columns/ColumnFixedString.h b/src/Columns/ColumnFixedString.h index 7b46dc11cd6..676ac7712ba 100644 --- a/src/Columns/ColumnFixedString.h +++ b/src/Columns/ColumnFixedString.h @@ -98,9 +98,17 @@ public: bool tryInsert(const Field & x) override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src_, size_t index) override; +#else + void doInsertFrom(const IColumn & src_, size_t index) override; +#endif +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertManyFrom(const IColumn & src, size_t position, size_t length) override; +#else + void doInsertManyFrom(const IColumn & src, size_t position, size_t length) override; +#endif void insertData(const char * pos, size_t length) override; @@ -125,11 +133,15 @@ public: void updateHashWithValue(size_t index, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t p1, size_t p2, const IColumn & rhs_, int /*nan_direction_hint*/) const override +#else + int doCompareAt(size_t p1, size_t p2, const IColumn & rhs_, int /*nan_direction_hint*/) const override +#endif { const ColumnFixedString & rhs = assert_cast(rhs_); chassert(this->n == rhs.n); @@ -144,7 +156,11 @@ public: size_t estimateCardinalityInPermutedRange(const Permutation & permutation, const EqualRange & equal_range) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif ColumnPtr filter(const IColumn::Filter & filt, ssize_t result_size_hint) const override; diff --git a/src/Columns/ColumnFunction.cpp b/src/Columns/ColumnFunction.cpp index 0ab9d15ad50..fc81efaac0c 100644 --- a/src/Columns/ColumnFunction.cpp +++ b/src/Columns/ColumnFunction.cpp @@ -72,7 +72,11 @@ ColumnPtr ColumnFunction::cut(size_t start, size_t length) const return ColumnFunction::create(length, function, capture, is_short_circuit_argument, is_function_compiled); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnFunction::insertFrom(const IColumn & src, size_t n) +#else +void ColumnFunction::doInsertFrom(const IColumn & src, size_t n) +#endif { const ColumnFunction & src_func = assert_cast(src); @@ -89,7 +93,11 @@ void ColumnFunction::insertFrom(const IColumn & src, size_t n) ++elements_size; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnFunction::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnFunction::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { const ColumnFunction & src_func = assert_cast(src); diff --git a/src/Columns/ColumnFunction.h b/src/Columns/ColumnFunction.h index 6fdc6679d3e..b62c6bf70eb 100644 --- a/src/Columns/ColumnFunction.h +++ b/src/Columns/ColumnFunction.h @@ -4,6 +4,7 @@ #include #include #include +#include namespace DB @@ -94,8 +95,16 @@ public: throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Cannot insert into {}", getName()); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src, size_t n) override; +#else + void doInsertFrom(const IColumn & src, size_t n) override; +#endif +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn &, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn &, size_t start, size_t length) override; +#endif void insertData(const char *, size_t) override { @@ -122,9 +131,9 @@ public: throw Exception(ErrorCodes::NOT_IMPLEMENTED, "updateHashWithValue is not implemented for {}", getName()); } - void updateWeakHash32(WeakHash32 &) const override + WeakHash32 getWeakHash32() const override { - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "updateWeakHash32 is not implemented for {}", getName()); + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "getWeakHash32 is not implemented for {}", getName()); } void updateHashFast(SipHash &) const override @@ -137,7 +146,11 @@ public: throw Exception(ErrorCodes::NOT_IMPLEMENTED, "popBack is not implemented for {}", getName()); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t, size_t, const IColumn &, int) const override +#else + int doCompareAt(size_t, size_t, const IColumn &, int) const override +#endif { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "compareAt is not implemented for {}", getName()); } diff --git a/src/Columns/ColumnLowCardinality.cpp b/src/Columns/ColumnLowCardinality.cpp index 208326fe629..a977046b07f 100644 --- a/src/Columns/ColumnLowCardinality.cpp +++ b/src/Columns/ColumnLowCardinality.cpp @@ -7,8 +7,7 @@ #include #include #include -#include "Storages/IndicesDescription.h" -#include "base/types.h" +#include #include #include @@ -159,7 +158,11 @@ void ColumnLowCardinality::insertDefault() idx.insertPosition(getDictionary().getDefaultValueIndex()); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnLowCardinality::insertFrom(const IColumn & src, size_t n) +#else +void ColumnLowCardinality::doInsertFrom(const IColumn & src, size_t n) +#endif { const auto * low_cardinality_src = typeid_cast(&src); @@ -187,7 +190,11 @@ void ColumnLowCardinality::insertFromFullColumn(const IColumn & src, size_t n) idx.insertPosition(getDictionary().uniqueInsertFrom(src, n)); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnLowCardinality::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnLowCardinality::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { const auto * low_cardinality_src = typeid_cast(&src); @@ -312,19 +319,10 @@ const char * ColumnLowCardinality::skipSerializedInArena(const char * pos) const return getDictionary().skipSerializedInArena(pos); } -void ColumnLowCardinality::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnLowCardinality::getWeakHash32() const { - auto s = size(); - - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); - - const auto & dict = getDictionary().getNestedColumn(); - WeakHash32 dict_hash(dict->size()); - dict->updateWeakHash32(dict_hash); - - idx.updateWeakHash(hash, dict_hash); + WeakHash32 dict_hash = getDictionary().getNestedColumn()->getWeakHash32(); + return idx.getWeakHash(dict_hash); } void ColumnLowCardinality::updateHashFast(SipHash & hash) const @@ -364,7 +362,11 @@ int ColumnLowCardinality::compareAtImpl(size_t n, size_t m, const IColumn & rhs, return getDictionary().compareAt(n_index, m_index, low_cardinality_column.getDictionary(), nan_direction_hint); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnLowCardinality::compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#else +int ColumnLowCardinality::doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#endif { return compareAtImpl(n, m, rhs, nan_direction_hint); } @@ -820,10 +822,11 @@ bool ColumnLowCardinality::Index::containsDefault() const return contains; } -void ColumnLowCardinality::Index::updateWeakHash(WeakHash32 & hash, WeakHash32 & dict_hash) const +WeakHash32 ColumnLowCardinality::Index::getWeakHash(const WeakHash32 & dict_hash) const { + WeakHash32 hash(positions->size()); auto & hash_data = hash.getData(); - auto & dict_hash_data = dict_hash.getData(); + const auto & dict_hash_data = dict_hash.getData(); auto update_weak_hash = [&](auto x) { @@ -832,10 +835,11 @@ void ColumnLowCardinality::Index::updateWeakHash(WeakHash32 & hash, WeakHash32 & auto size = data.size(); for (size_t i = 0; i < size; ++i) - hash_data[i] = static_cast(intHashCRC32(dict_hash_data[data[i]], hash_data[i])); + hash_data[i] = dict_hash_data[data[i]]; }; callForType(std::move(update_weak_hash), size_of_type); + return hash; } void ColumnLowCardinality::Index::collectSerializedValueSizes( diff --git a/src/Columns/ColumnLowCardinality.h b/src/Columns/ColumnLowCardinality.h index ac3b725b22f..3766b247d60 100644 --- a/src/Columns/ColumnLowCardinality.h +++ b/src/Columns/ColumnLowCardinality.h @@ -78,10 +78,18 @@ public: bool tryInsert(const Field & x) override; void insertDefault() override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src, size_t n) override; +#else + void doInsertFrom(const IColumn & src, size_t n) override; +#endif void insertFromFullColumn(const IColumn & src, size_t n); +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif void insertRangeFromFullColumn(const IColumn & src, size_t start, size_t length); void insertRangeFromDictionaryEncodedColumn(const IColumn & keys, const IColumn & positions); @@ -103,7 +111,7 @@ public: getDictionary().updateHashWithValue(getIndexes().getUInt(n), hash); } - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash &) const override; @@ -127,7 +135,11 @@ public: return ColumnLowCardinality::create(dictionary.getColumnUniquePtr(), getIndexes().index(indexes_, limit)); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#endif int compareAtWithCollation(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint, const Collator &) const override; @@ -313,7 +325,7 @@ public: bool containsDefault() const; - void updateWeakHash(WeakHash32 & hash, WeakHash32 & dict_hash) const; + WeakHash32 getWeakHash(const WeakHash32 & dict_hash) const; void collectSerializedValueSizes(PaddedPODArray & sizes, const PaddedPODArray & dict_sizes) const; diff --git a/src/Columns/ColumnMap.cpp b/src/Columns/ColumnMap.cpp index eecea1a273f..1025b4e77b9 100644 --- a/src/Columns/ColumnMap.cpp +++ b/src/Columns/ColumnMap.cpp @@ -143,9 +143,9 @@ void ColumnMap::updateHashWithValue(size_t n, SipHash & hash) const nested->updateHashWithValue(n, hash); } -void ColumnMap::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnMap::getWeakHash32() const { - nested->updateWeakHash32(hash); + return nested->getWeakHash32(); } void ColumnMap::updateHashFast(SipHash & hash) const @@ -153,17 +153,29 @@ void ColumnMap::updateHashFast(SipHash & hash) const nested->updateHashFast(hash); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnMap::insertFrom(const IColumn & src, size_t n) +#else +void ColumnMap::doInsertFrom(const IColumn & src, size_t n) +#endif { nested->insertFrom(assert_cast(src).getNestedColumn(), n); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnMap::insertManyFrom(const IColumn & src, size_t position, size_t length) +#else +void ColumnMap::doInsertManyFrom(const IColumn & src, size_t position, size_t length) +#endif { assert_cast(*nested).insertManyFrom(assert_cast(src).getNestedColumn(), position, length); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnMap::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnMap::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { nested->insertRangeFrom( assert_cast(src).getNestedColumn(), @@ -210,7 +222,11 @@ MutableColumns ColumnMap::scatter(ColumnIndex num_columns, const Selector & sele return res; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnMap::compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#else +int ColumnMap::doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#endif { const auto & rhs_map = assert_cast(rhs); return nested->compareAt(n, m, rhs_map.getNestedColumn(), nan_direction_hint); diff --git a/src/Columns/ColumnMap.h b/src/Columns/ColumnMap.h index 52165d0d74e..3eaaa0ad562 100644 --- a/src/Columns/ColumnMap.h +++ b/src/Columns/ColumnMap.h @@ -64,18 +64,30 @@ public: const char * deserializeAndInsertFromArena(const char * pos) override; const char * skipSerializedInArena(const char * pos) const override; void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; + +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src_, size_t n) override; void insertManyFrom(const IColumn & src, size_t position, size_t length) override; void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertFrom(const IColumn & src_, size_t n) override; + void doInsertManyFrom(const IColumn & src, size_t position, size_t length) override; + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif + ColumnPtr filter(const Filter & filt, ssize_t result_size_hint) const override; void expand(const Filter & mask, bool inverted) override; ColumnPtr permute(const Permutation & perm, size_t limit) const override; ColumnPtr index(const IColumn & indexes, size_t limit) const override; ColumnPtr replicate(const Offsets & offsets) const override; MutableColumns scatter(ColumnIndex num_columns, const Selector & selector) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#endif void getExtremes(Field & min, Field & max) const override; void getPermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, size_t limit, int nan_direction_hint, IColumn::Permutation & res) const override; diff --git a/src/Columns/ColumnNullable.cpp b/src/Columns/ColumnNullable.cpp index 1d12a59fd59..6529f0b78db 100644 --- a/src/Columns/ColumnNullable.cpp +++ b/src/Columns/ColumnNullable.cpp @@ -56,25 +56,21 @@ void ColumnNullable::updateHashWithValue(size_t n, SipHash & hash) const getNestedColumn().updateHashWithValue(n, hash); } -void ColumnNullable::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnNullable::getWeakHash32() const { auto s = size(); - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); - - WeakHash32 old_hash = hash; - nested_column->updateWeakHash32(hash); + WeakHash32 hash = nested_column->getWeakHash32(); const auto & null_map_data = getNullMapData(); auto & hash_data = hash.getData(); - auto & old_hash_data = old_hash.getData(); - /// Use old data for nulls. + /// Use default for nulls. for (size_t row = 0; row < s; ++row) if (null_map_data[row]) - hash_data[row] = old_hash_data[row]; + hash_data[row] = WeakHash32::kDefaultInitialValue; + + return hash; } void ColumnNullable::updateHashFast(SipHash & hash) const @@ -221,7 +217,11 @@ const char * ColumnNullable::skipSerializedInArena(const char * pos) const return pos; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnNullable::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnNullable::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { const ColumnNullable & nullable_col = assert_cast(src); getNullMapColumn().insertRangeFrom(*nullable_col.null_map, start, length); @@ -258,7 +258,11 @@ bool ColumnNullable::tryInsert(const Field & x) return true; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnNullable::insertFrom(const IColumn & src, size_t n) +#else +void ColumnNullable::doInsertFrom(const IColumn & src, size_t n) +#endif { const ColumnNullable & src_concrete = assert_cast(src); getNestedColumn().insertFrom(src_concrete.getNestedColumn(), n); @@ -266,7 +270,11 @@ void ColumnNullable::insertFrom(const IColumn & src, size_t n) } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnNullable::insertManyFrom(const IColumn & src, size_t position, size_t length) +#else +void ColumnNullable::doInsertManyFrom(const IColumn & src, size_t position, size_t length) +#endif { const ColumnNullable & src_concrete = assert_cast(src); getNestedColumn().insertManyFrom(src_concrete.getNestedColumn(), position, length); @@ -402,7 +410,11 @@ int ColumnNullable::compareAtImpl(size_t n, size_t m, const IColumn & rhs_, int return getNestedColumn().compareAt(n, m, nested_rhs, null_direction_hint); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnNullable::compareAt(size_t n, size_t m, const IColumn & rhs_, int null_direction_hint) const +#else +int ColumnNullable::doCompareAt(size_t n, size_t m, const IColumn & rhs_, int null_direction_hint) const +#endif { return compareAtImpl(n, m, rhs_, null_direction_hint); } diff --git a/src/Columns/ColumnNullable.h b/src/Columns/ColumnNullable.h index 510a4cacf1e..fe9f5b6dcc2 100644 --- a/src/Columns/ColumnNullable.h +++ b/src/Columns/ColumnNullable.h @@ -69,11 +69,21 @@ public: char * serializeValueIntoMemory(size_t n, char * memory) const override; const char * deserializeAndInsertFromArena(const char * pos) override; const char * skipSerializedInArena(const char * pos) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif void insert(const Field & x) override; bool tryInsert(const Field & x) override; + +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src, size_t n) override; void insertManyFrom(const IColumn & src, size_t position, size_t length) override; +#else + void doInsertFrom(const IColumn & src, size_t n) override; + void doInsertManyFrom(const IColumn & src, size_t position, size_t length) override; +#endif void insertFromNotNullable(const IColumn & src, size_t n); void insertRangeFromNotNullable(const IColumn & src, size_t start, size_t length); @@ -90,7 +100,11 @@ public: void expand(const Filter & mask, bool inverted) override; ColumnPtr permute(const Permutation & perm, size_t limit) const override; ColumnPtr index(const IColumn & indexes, size_t limit) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs_, int null_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs_, int null_direction_hint) const override; +#endif #if USE_EMBEDDED_COMPILER @@ -119,7 +133,7 @@ public: void protect() override; ColumnPtr replicate(const Offsets & replicate_offsets) const override; void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; void getExtremes(Field & min, Field & max) const override; // Special function for nullable minmax index diff --git a/src/Columns/ColumnObject.cpp b/src/Columns/ColumnObject.cpp index 90ef974010c..a6431007cb6 100644 --- a/src/Columns/ColumnObject.cpp +++ b/src/Columns/ColumnObject.cpp @@ -763,12 +763,20 @@ void ColumnObject::get(size_t n, Field & res) const } } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnObject::insertFrom(const IColumn & src, size_t n) +#else +void ColumnObject::doInsertFrom(const IColumn & src, size_t n) +#endif { insert(src[n]); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnObject::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnObject::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { const auto & src_object = assert_cast(src); @@ -1093,4 +1101,10 @@ void ColumnObject::finalize() checkObjectHasNoAmbiguosPaths(getKeys()); } +void ColumnObject::updateHashFast(SipHash & hash) const +{ + for (const auto & entry : subcolumns) + for (auto & part : entry->data.data) + part->updateHashFast(hash); +} } diff --git a/src/Columns/ColumnObject.h b/src/Columns/ColumnObject.h index e2936b27994..25cfaa550f6 100644 --- a/src/Columns/ColumnObject.h +++ b/src/Columns/ColumnObject.h @@ -5,6 +5,7 @@ #include #include #include +#include #include @@ -209,8 +210,15 @@ public: void insert(const Field & field) override; bool tryInsert(const Field & field) override; void insertDefault() override; + +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src, size_t n) override; void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertFrom(const IColumn & src, size_t n) override; + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif + void popBack(size_t length) override; Field operator[](size_t n) const override; void get(size_t n, Field & res) const override; @@ -228,7 +236,11 @@ public: /// Order of rows in ColumnObject is undefined. void getPermutation(PermutationSortDirection, PermutationSortStability, size_t, int, Permutation & res) const override; void updatePermutation(PermutationSortDirection, PermutationSortStability, size_t, int, Permutation &, EqualRanges &) const override {} +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t, size_t, const IColumn &, int) const override { return 0; } +#else + int doCompareAt(size_t, size_t, const IColumn &, int) const override { return 0; } +#endif void getExtremes(Field & min, Field & max) const override; /// All other methods throw exception. @@ -241,8 +253,8 @@ public: const char * deserializeAndInsertFromArena(const char *) override { throwMustBeConcrete(); } const char * skipSerializedInArena(const char *) const override { throwMustBeConcrete(); } void updateHashWithValue(size_t, SipHash &) const override { throwMustBeConcrete(); } - void updateWeakHash32(WeakHash32 &) const override { throwMustBeConcrete(); } - void updateHashFast(SipHash &) const override { throwMustBeConcrete(); } + WeakHash32 getWeakHash32() const override { throwMustBeConcrete(); } + void updateHashFast(SipHash & hash) const override; void expand(const Filter &, bool) override { throwMustBeConcrete(); } bool hasEqualValues() const override { throwMustBeConcrete(); } size_t byteSizeAt(size_t) const override { throwMustBeConcrete(); } diff --git a/src/Columns/ColumnSparse.cpp b/src/Columns/ColumnSparse.cpp index 5190ceb49e5..a908d970a15 100644 --- a/src/Columns/ColumnSparse.cpp +++ b/src/Columns/ColumnSparse.cpp @@ -174,7 +174,11 @@ const char * ColumnSparse::skipSerializedInArena(const char * pos) const return values->skipSerializedInArena(pos); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnSparse::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnSparse::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { if (length == 0) return; @@ -248,7 +252,11 @@ bool ColumnSparse::tryInsert(const Field & x) return true; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnSparse::insertFrom(const IColumn & src, size_t n) +#else +void ColumnSparse::doInsertFrom(const IColumn & src, size_t n) +#endif { if (const auto * src_sparse = typeid_cast(&src)) { @@ -446,7 +454,11 @@ ColumnPtr ColumnSparse::indexImpl(const PaddedPODArray & indexes, size_t l return ColumnSparse::create(std::move(res_values), std::move(res_offsets), limit); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnSparse::compareAt(size_t n, size_t m, const IColumn & rhs_, int null_direction_hint) const +#else +int ColumnSparse::doCompareAt(size_t n, size_t m, const IColumn & rhs_, int null_direction_hint) const +#endif { if (const auto * rhs_sparse = typeid_cast(&rhs_)) return values->compareAt(getValueIndex(n), rhs_sparse->getValueIndex(m), rhs_sparse->getValuesColumn(), null_direction_hint); @@ -666,20 +678,22 @@ void ColumnSparse::updateHashWithValue(size_t n, SipHash & hash) const values->updateHashWithValue(getValueIndex(n), hash); } -void ColumnSparse::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnSparse::getWeakHash32() const { - if (hash.getData().size() != _size) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", _size, hash.getData().size()); + WeakHash32 values_hash = values->getWeakHash32(); + WeakHash32 hash(size()); + + auto & hash_data = hash.getData(); + auto & values_hash_data = values_hash.getData(); auto offset_it = begin(); - auto & hash_data = hash.getData(); for (size_t i = 0; i < _size; ++i, ++offset_it) { size_t value_index = offset_it.getValueIndex(); - auto data_ref = values->getDataAt(value_index); - hash_data[i] = ::updateWeakHash32(reinterpret_cast(data_ref.data), data_ref.size, hash_data[i]); + hash_data[i] = values_hash_data[value_index]; } + + return hash; } void ColumnSparse::updateHashFast(SipHash & hash) const diff --git a/src/Columns/ColumnSparse.h b/src/Columns/ColumnSparse.h index 12b2def7cf1..7a4d914e62a 100644 --- a/src/Columns/ColumnSparse.h +++ b/src/Columns/ColumnSparse.h @@ -81,10 +81,18 @@ public: char * serializeValueIntoMemory(size_t n, char * memory) const override; const char * deserializeAndInsertFromArena(const char * pos) override; const char * skipSerializedInArena(const char *) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif void insert(const Field & x) override; bool tryInsert(const Field & x) override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src, size_t n) override; +#else + void doInsertFrom(const IColumn & src, size_t n) override; +#endif void insertDefault() override; void insertManyDefaults(size_t length) override; @@ -98,7 +106,11 @@ public: template ColumnPtr indexImpl(const PaddedPODArray & indexes, size_t limit) const; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs_, int null_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs_, int null_direction_hint) const override; +#endif void compareColumn(const IColumn & rhs, size_t rhs_row_num, PaddedPODArray * row_indexes, PaddedPODArray & compare_results, int direction, int nan_direction_hint) const override; @@ -127,7 +139,7 @@ public: void protect() override; ColumnPtr replicate(const Offsets & replicate_offsets) const override; void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; void getExtremes(Field & min, Field & max) const override; diff --git a/src/Columns/ColumnString.cpp b/src/Columns/ColumnString.cpp index a84aea73486..7cfa2571f5a 100644 --- a/src/Columns/ColumnString.cpp +++ b/src/Columns/ColumnString.cpp @@ -39,7 +39,11 @@ ColumnString::ColumnString(const ColumnString & src) last_offset, chars.size()); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnString::insertManyFrom(const IColumn & src, size_t position, size_t length) +#else +void ColumnString::doInsertManyFrom(const IColumn & src, size_t position, size_t length) +#endif { const ColumnString & src_concrete = assert_cast(src); const UInt8 * src_buf = &src_concrete.chars[src_concrete.offsets[position - 1]]; @@ -104,13 +108,10 @@ MutableColumnPtr ColumnString::cloneResized(size_t to_size) const return res; } -void ColumnString::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnString::getWeakHash32() const { auto s = offsets.size(); - - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); + WeakHash32 hash(s); const UInt8 * pos = chars.data(); UInt32 * hash_data = hash.getData().data(); @@ -126,10 +127,16 @@ void ColumnString::updateWeakHash32(WeakHash32 & hash) const prev_offset = offset; ++hash_data; } + + return hash; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnString::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnString::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { if (length == 0) return; diff --git a/src/Columns/ColumnString.h b/src/Columns/ColumnString.h index 39d4684fd89..c1012e1e55e 100644 --- a/src/Columns/ColumnString.h +++ b/src/Columns/ColumnString.h @@ -142,7 +142,11 @@ public: return true; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src_, size_t n) override +#else + void doInsertFrom(const IColumn & src_, size_t n) override +#endif { const ColumnString & src = assert_cast(src_); const size_t size_to_append = src.offsets[n] - src.offsets[n - 1]; /// -1th index is Ok, see PaddedPODArray. @@ -165,7 +169,11 @@ public: } } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertManyFrom(const IColumn & src, size_t position, size_t length) override; +#else + void doInsertManyFrom(const IColumn & src, size_t position, size_t length) override; +#endif void insertData(const char * pos, size_t length) override { @@ -204,7 +212,7 @@ public: hash.update(reinterpret_cast(&chars[offset]), string_size); } - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override { @@ -212,7 +220,11 @@ public: hash.update(reinterpret_cast(chars.data()), chars.size() * sizeof(chars[0])); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif ColumnPtr filter(const Filter & filt, ssize_t result_size_hint) const override; @@ -238,7 +250,11 @@ public: offsets.push_back(offsets.back() + 1); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs_, int /*nan_direction_hint*/) const override +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs_, int /*nan_direction_hint*/) const override +#endif { const ColumnString & rhs = assert_cast(rhs_); return memcmpSmallAllowOverflow15(chars.data() + offsetAt(n), sizeAt(n) - 1, rhs.chars.data() + rhs.offsetAt(m), rhs.sizeAt(m) - 1); diff --git a/src/Columns/ColumnTuple.cpp b/src/Columns/ColumnTuple.cpp index 2159495b68f..4fc3f88a87c 100644 --- a/src/Columns/ColumnTuple.cpp +++ b/src/Columns/ColumnTuple.cpp @@ -201,11 +201,16 @@ bool ColumnTuple::tryInsert(const Field & x) return false; } } + ++column_length; return true; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnTuple::insertFrom(const IColumn & src_, size_t n) +#else +void ColumnTuple::doInsertFrom(const IColumn & src_, size_t n) +#endif { const ColumnTuple & src = assert_cast(src_); @@ -218,7 +223,11 @@ void ColumnTuple::insertFrom(const IColumn & src_, size_t n) columns[i]->insertFrom(*src.columns[i], n); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnTuple::insertManyFrom(const IColumn & src, size_t position, size_t length) +#else +void ColumnTuple::doInsertManyFrom(const IColumn & src, size_t position, size_t length) +#endif { const ColumnTuple & src_tuple = assert_cast(src); @@ -228,6 +237,7 @@ void ColumnTuple::insertManyFrom(const IColumn & src, size_t position, size_t le for (size_t i = 0; i < tuple_size; ++i) columns[i]->insertManyFrom(*src_tuple.columns[i], position, length); + column_length += length; } void ColumnTuple::insertDefault() @@ -300,16 +310,15 @@ void ColumnTuple::updateHashWithValue(size_t n, SipHash & hash) const column->updateHashWithValue(n, hash); } -void ColumnTuple::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnTuple::getWeakHash32() const { auto s = size(); - - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); + WeakHash32 hash(s); for (const auto & column : columns) - column->updateWeakHash32(hash); + hash.update(column->getWeakHash32()); + + return hash; } void ColumnTuple::updateHashFast(SipHash & hash) const @@ -318,7 +327,11 @@ void ColumnTuple::updateHashFast(SipHash & hash) const column->updateHashFast(hash); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnTuple::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnTuple::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { column_length += length; const size_t tuple_size = columns.size(); @@ -470,7 +483,11 @@ int ColumnTuple::compareAtImpl(size_t n, size_t m, const IColumn & rhs, int nan_ return 0; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnTuple::compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#else +int ColumnTuple::doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#endif { return compareAtImpl(n, m, rhs, nan_direction_hint); } @@ -711,7 +728,13 @@ void ColumnTuple::takeDynamicStructureFromSourceColumns(const Columns & source_c ColumnPtr ColumnTuple::compress() const { if (columns.empty()) - return Ptr(); + { + return ColumnCompressed::create(size(), 0, + [n = column_length] + { + return ColumnTuple::create(n); + }); + } size_t byte_size = 0; Columns compressed; diff --git a/src/Columns/ColumnTuple.h b/src/Columns/ColumnTuple.h index 0103f81b242..16b47a993f6 100644 --- a/src/Columns/ColumnTuple.h +++ b/src/Columns/ColumnTuple.h @@ -65,8 +65,15 @@ public: void insertData(const char * pos, size_t length) override; void insert(const Field & x) override; bool tryInsert(const Field & x) override; + +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src_, size_t n) override; void insertManyFrom(const IColumn & src, size_t position, size_t length) override; +#else + void doInsertFrom(const IColumn & src_, size_t n) override; + void doInsertManyFrom(const IColumn & src, size_t position, size_t length) override; +#endif + void insertDefault() override; void popBack(size_t n) override; StringRef serializeValueIntoArena(size_t n, Arena & arena, char const *& begin) const override; @@ -74,16 +81,24 @@ public: const char * deserializeAndInsertFromArena(const char * pos) override; const char * skipSerializedInArena(const char * pos) const override; void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif ColumnPtr filter(const Filter & filt, ssize_t result_size_hint) const override; void expand(const Filter & mask, bool inverted) override; ColumnPtr permute(const Permutation & perm, size_t limit) const override; ColumnPtr index(const IColumn & indexes, size_t limit) const override; ColumnPtr replicate(const Offsets & offsets) const override; MutableColumns scatter(ColumnIndex num_columns, const Selector & selector) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#endif int compareAtWithCollation(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint, const Collator & collator) const override; void getExtremes(Field & min, Field & max) const override; void getPermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, diff --git a/src/Columns/ColumnUnique.h b/src/Columns/ColumnUnique.h index 0311efd4c83..d6cb75679be 100644 --- a/src/Columns/ColumnUnique.h +++ b/src/Columns/ColumnUnique.h @@ -90,7 +90,11 @@ public: return getNestedColumn()->updateHashWithValue(n, hash_func); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#endif void getExtremes(Field & min, Field & max) const override { column_holder->getExtremes(min, max); } bool valuesHaveFixedSize() const override { return column_holder->valuesHaveFixedSize(); } @@ -488,7 +492,11 @@ const char * ColumnUnique::skipSerializedInArena(const char *) const } template +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnUnique::compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#else +int ColumnUnique::doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#endif { if (is_nullable) { diff --git a/src/Columns/ColumnVariant.cpp b/src/Columns/ColumnVariant.cpp index ec47f5dfa74..de7efb41d19 100644 --- a/src/Columns/ColumnVariant.cpp +++ b/src/Columns/ColumnVariant.cpp @@ -595,17 +595,29 @@ void ColumnVariant::insertManyFromImpl(const DB::IColumn & src_, size_t position } } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnVariant::insertFrom(const IColumn & src_, size_t n) +#else +void ColumnVariant::doInsertFrom(const IColumn & src_, size_t n) +#endif { insertFromImpl(src_, n, nullptr); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnVariant::insertRangeFrom(const IColumn & src_, size_t start, size_t length) +#else +void ColumnVariant::doInsertRangeFrom(const IColumn & src_, size_t start, size_t length) +#endif { insertRangeFromImpl(src_, start, length, nullptr); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnVariant::insertManyFrom(const DB::IColumn & src_, size_t position, size_t length) +#else +void ColumnVariant::doInsertManyFrom(const DB::IColumn & src_, size_t position, size_t length) +#endif { insertManyFromImpl(src_, position, length, nullptr); } @@ -777,36 +789,26 @@ void ColumnVariant::updateHashWithValue(size_t n, SipHash & hash) const variants[localDiscriminatorByGlobal(global_discr)]->updateHashWithValue(offsetAt(n), hash); } -void ColumnVariant::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnVariant::getWeakHash32() const { auto s = size(); - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); - /// If we have only NULLs, keep hash unchanged. if (hasOnlyNulls()) - return; + return WeakHash32(s); /// Optimization for case when there is only 1 non-empty variant and no NULLs. /// In this case we can just calculate weak hash for this variant. if (auto non_empty_local_discr = getLocalDiscriminatorOfOneNoneEmptyVariantNoNulls()) - { - variants[*non_empty_local_discr]->updateWeakHash32(hash); - return; - } + return variants[*non_empty_local_discr]->getWeakHash32(); /// Calculate weak hash for all variants. std::vector nested_hashes; for (const auto & variant : variants) - { - WeakHash32 nested_hash(variant->size()); - variant->updateWeakHash32(nested_hash); - nested_hashes.emplace_back(std::move(nested_hash)); - } + nested_hashes.emplace_back(variant->getWeakHash32()); /// For each row hash is a hash of corresponding row from corresponding variant. + WeakHash32 hash(s); auto & hash_data = hash.getData(); const auto & local_discriminators_data = getLocalDiscriminators(); const auto & offsets_data = getOffsets(); @@ -815,11 +817,10 @@ void ColumnVariant::updateWeakHash32(WeakHash32 & hash) const Discriminator discr = local_discriminators_data[i]; /// Update hash only for non-NULL values if (discr != NULL_DISCRIMINATOR) - { - auto nested_hash = nested_hashes[local_discriminators_data[i]].getData()[offsets_data[i]]; - hash_data[i] = static_cast(hashCRC32(nested_hash, hash_data[i])); - } + hash_data[i] = nested_hashes[discr].getData()[offsets_data[i]]; } + + return hash; } void ColumnVariant::updateHashFast(SipHash & hash) const @@ -1174,7 +1175,11 @@ bool ColumnVariant::hasEqualValues() const return local_discriminators->hasEqualValues() && variants[localDiscriminatorAt(0)]->hasEqualValues(); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int ColumnVariant::compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#else +int ColumnVariant::doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const +#endif { const auto & rhs_variant = assert_cast(rhs); Discriminator left_discr = globalDiscriminatorAt(n); diff --git a/src/Columns/ColumnVariant.h b/src/Columns/ColumnVariant.h index e5a4498f340..34c24b5428d 100644 --- a/src/Columns/ColumnVariant.h +++ b/src/Columns/ColumnVariant.h @@ -180,9 +180,19 @@ public: void insert(const Field & x) override; bool tryInsert(const Field & x) override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src_, size_t n) override; void insertRangeFrom(const IColumn & src_, size_t start, size_t length) override; void insertManyFrom(const IColumn & src_, size_t position, size_t length) override; +#else + using IColumn::insertFrom; + using IColumn::insertManyFrom; + using IColumn::insertRangeFrom; + + void doInsertFrom(const IColumn & src_, size_t n) override; + void doInsertRangeFrom(const IColumn & src_, size_t start, size_t length) override; + void doInsertManyFrom(const IColumn & src_, size_t position, size_t length) override; +#endif /// Methods for insertion from another Variant but with known mapping between global discriminators. void insertFrom(const IColumn & src_, size_t n, const std::vector & global_discriminators_mapping); @@ -203,7 +213,7 @@ public: const char * deserializeVariantAndInsertFromArena(Discriminator global_discr, const char * pos); const char * skipSerializedInArena(const char * pos) const override; void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; ColumnPtr filter(const Filter & filt, ssize_t result_size_hint) const override; void expand(const Filter & mask, bool inverted) override; @@ -213,7 +223,11 @@ public: ColumnPtr indexImpl(const PaddedPODArray & indexes, size_t limit) const; ColumnPtr replicate(const Offsets & replicate_offsets) const override; MutableColumns scatter(ColumnIndex num_columns, const Selector & selector) const override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const override; +#endif bool hasEqualValues() const override; void getExtremes(Field & min, Field & max) const override; void getPermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, diff --git a/src/Columns/ColumnVector.cpp b/src/Columns/ColumnVector.cpp index 35d9f5386ed..c474efe35bd 100644 --- a/src/Columns/ColumnVector.cpp +++ b/src/Columns/ColumnVector.cpp @@ -73,13 +73,10 @@ void ColumnVector::updateHashWithValue(size_t n, SipHash & hash) const } template -void ColumnVector::updateWeakHash32(WeakHash32 & hash) const +WeakHash32 ColumnVector::getWeakHash32() const { auto s = data.size(); - - if (hash.getData().size() != s) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match size of column: " - "column size is {}, hash size is {}", std::to_string(s), std::to_string(hash.getData().size())); + WeakHash32 hash(s); const T * begin = data.data(); const T * end = begin + s; @@ -91,6 +88,8 @@ void ColumnVector::updateWeakHash32(WeakHash32 & hash) const ++begin; ++hash_data; } + + return hash; } template @@ -503,7 +502,11 @@ bool ColumnVector::tryInsert(const DB::Field & x) } template +#if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnVector::insertRangeFrom(const IColumn & src, size_t start, size_t length) +#else +void ColumnVector::doInsertRangeFrom(const IColumn & src, size_t start, size_t length) +#endif { const ColumnVector & src_vec = assert_cast(src); diff --git a/src/Columns/ColumnVector.h b/src/Columns/ColumnVector.h index bbd27c91a70..2fe5b635bd2 100644 --- a/src/Columns/ColumnVector.h +++ b/src/Columns/ColumnVector.h @@ -64,12 +64,20 @@ public: return data.size(); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn & src, size_t n) override +#else + void doInsertFrom(const IColumn & src, size_t n) override +#endif { data.push_back(assert_cast(src).getData()[n]); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertManyFrom(const IColumn & src, size_t position, size_t length) override +#else + void doInsertManyFrom(const IColumn & src, size_t position, size_t length) override +#endif { ValueType v = assert_cast(src).getData()[position]; data.resize_fill(data.size() + length, v); @@ -106,7 +114,7 @@ public: void updateHashWithValue(size_t n, SipHash & hash) const override; - void updateWeakHash32(WeakHash32 & hash) const override; + WeakHash32 getWeakHash32() const override; void updateHashFast(SipHash & hash) const override; @@ -142,7 +150,11 @@ public: } /// This method implemented in header because it could be possibly devirtualized. +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const override +#else + int doCompareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const override +#endif { return CompareHelper::compare(data[n], assert_cast(rhs_).data[m], nan_direction_hint); } @@ -228,7 +240,11 @@ public: bool tryInsert(const DB::Field & x) override; +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#else + void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) override; +#endif ColumnPtr filter(const IColumn::Filter & filt, ssize_t result_size_hint) const override; diff --git a/src/Columns/IColumn.cpp b/src/Columns/IColumn.cpp index 90cccef2b03..a189903b617 100644 --- a/src/Columns/IColumn.cpp +++ b/src/Columns/IColumn.cpp @@ -46,7 +46,11 @@ String IColumn::dumpStructure() const return res.str(); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void IColumn::insertFrom(const IColumn & src, size_t n) +#else +void IColumn::doInsertFrom(const IColumn & src, size_t n) +#endif { insert(src[n]); } diff --git a/src/Columns/IColumn.h b/src/Columns/IColumn.h index afa301d5c1c..f9c1a3e7034 100644 --- a/src/Columns/IColumn.h +++ b/src/Columns/IColumn.h @@ -1,15 +1,14 @@ #pragma once -#include -#include -#include -#include -#include #include +#include +#include +#include +#include +#include #include "config.h" - class SipHash; class Collator; @@ -180,18 +179,42 @@ public: /// Appends n-th element from other column with the same type. /// Is used in merge-sort and merges. It could be implemented in inherited classes more optimally than default implementation. +#if !defined(DEBUG_OR_SANITIZER_BUILD) virtual void insertFrom(const IColumn & src, size_t n); +#else + void insertFrom(const IColumn & src, size_t n) + { + assertTypeEquality(src); + doInsertFrom(src, n); + } +#endif /// Appends range of elements from other column with the same type. /// Could be used to concatenate columns. +#if !defined(DEBUG_OR_SANITIZER_BUILD) virtual void insertRangeFrom(const IColumn & src, size_t start, size_t length) = 0; +#else + void insertRangeFrom(const IColumn & src, size_t start, size_t length) + { + assertTypeEquality(src); + doInsertRangeFrom(src, start, length); + } +#endif /// Appends one element from other column with the same type multiple times. +#if !defined(DEBUG_OR_SANITIZER_BUILD) virtual void insertManyFrom(const IColumn & src, size_t position, size_t length) { for (size_t i = 0; i < length; ++i) insertFrom(src, position); } +#else + void insertManyFrom(const IColumn & src, size_t position, size_t length) + { + assertTypeEquality(src); + doInsertManyFrom(src, position, length); + } +#endif /// Appends one field multiple times. Can be optimized in inherited classes. virtual void insertMany(const Field & field, size_t length) @@ -277,10 +300,10 @@ public: /// passed bytes to hash must identify sequence of values unambiguously. virtual void updateHashWithValue(size_t n, SipHash & hash) const = 0; - /// Update hash function value. Hash is calculated for each element. + /// Get hash function value. Hash is calculated for each element. /// It's a fast weak hash function. Mainly need to scatter data between threads. /// WeakHash32 must have the same size as column. - virtual void updateWeakHash32(WeakHash32 & hash) const = 0; + virtual WeakHash32 getWeakHash32() const = 0; /// Update state of hash with all column. virtual void updateHashFast(SipHash & hash) const = 0; @@ -322,7 +345,15 @@ public: * * For non Nullable and non floating point types, nan_direction_hint is ignored. */ +#if !defined(DEBUG_OR_SANITIZER_BUILD) [[nodiscard]] virtual int compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const = 0; +#else + [[nodiscard]] int compareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const + { + assertTypeEquality(rhs); + return doCompareAt(n, m, rhs, nan_direction_hint); + } +#endif #if USE_EMBEDDED_COMPILER @@ -610,6 +641,8 @@ public: [[nodiscard]] virtual bool isSparse() const { return false; } + [[nodiscard]] virtual bool isConst() const { return false; } + [[nodiscard]] virtual bool isCollationSupported() const { return false; } virtual ~IColumn() = default; @@ -633,6 +666,29 @@ protected: Equals equals, Sort full_sort, PartialSort partial_sort) const; + +#if defined(DEBUG_OR_SANITIZER_BUILD) + virtual void doInsertFrom(const IColumn & src, size_t n); + + virtual void doInsertRangeFrom(const IColumn & src, size_t start, size_t length) = 0; + + virtual void doInsertManyFrom(const IColumn & src, size_t position, size_t length) + { + for (size_t i = 0; i < length; ++i) + insertFrom(src, position); + } + + virtual int doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const = 0; + +private: + void assertTypeEquality(const IColumn & rhs) const + { + /// For Sparse and Const columns, we can compare only internal types. It is considered normal to e.g. insert from normal vector column to a sparse vector column. + /// This case is specifically handled in ColumnSparse implementation. Similar situation with Const column. + /// For the rest of column types we can compare the types directly. + chassert((isConst() || isSparse()) ? getDataType() == rhs.getDataType() : typeid(*this) == typeid(rhs)); + } +#endif }; using ColumnPtr = IColumn::Ptr; diff --git a/src/Columns/IColumnDummy.h b/src/Columns/IColumnDummy.h index 27f420fbc71..40d410e207d 100644 --- a/src/Columns/IColumnDummy.h +++ b/src/Columns/IColumnDummy.h @@ -1,6 +1,7 @@ #pragma once #include +#include namespace DB @@ -26,7 +27,11 @@ public: size_t byteSize() const override { return 0; } size_t byteSizeAt(size_t) const override { return 0; } size_t allocatedBytes() const override { return 0; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) int compareAt(size_t, size_t, const IColumn &, int) const override { return 0; } +#else + int doCompareAt(size_t, size_t, const IColumn &, int) const override { return 0; } +#endif void compareColumn(const IColumn &, size_t, PaddedPODArray *, PaddedPODArray &, int, int) const override { } @@ -59,20 +64,29 @@ public: { } - void updateWeakHash32(WeakHash32 & /*hash*/) const override + WeakHash32 getWeakHash32() const override { + return WeakHash32(s); } void updateHashFast(SipHash & /*hash*/) const override { } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertFrom(const IColumn &, size_t) override +#else + void doInsertFrom(const IColumn &, size_t) override +#endif { ++s; } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & /*src*/, size_t /*start*/, size_t length) override +#else + void doInsertRangeFrom(const IColumn & /*src*/, size_t /*start*/, size_t length) override +#endif { s += length; } diff --git a/src/Columns/IColumnUnique.h b/src/Columns/IColumnUnique.h index f71f19a5da6..a8e10e5e2b2 100644 --- a/src/Columns/IColumnUnique.h +++ b/src/Columns/IColumnUnique.h @@ -1,6 +1,7 @@ #pragma once #include #include +#include namespace DB { @@ -85,7 +86,11 @@ public: throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method tryInsert is not supported for ColumnUnique."); } +#if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn &, size_t, size_t) override +#else + void doInsertRangeFrom(const IColumn &, size_t, size_t) override +#endif { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method insertRangeFrom is not supported for ColumnUnique."); } @@ -162,9 +167,9 @@ public: throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method scatter is not supported for ColumnUnique."); } - void updateWeakHash32(WeakHash32 &) const override + WeakHash32 getWeakHash32() const override { - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method updateWeakHash32 is not supported for ColumnUnique."); + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method getWeakHash32 is not supported for ColumnUnique."); } void updateHashFast(SipHash &) const override diff --git a/src/Columns/benchmarks/benchmark_column_insert_many_from.cpp b/src/Columns/benchmarks/benchmark_column_insert_many_from.cpp index 325cf5559cd..240099f0ae5 100644 --- a/src/Columns/benchmarks/benchmark_column_insert_many_from.cpp +++ b/src/Columns/benchmarks/benchmark_column_insert_many_from.cpp @@ -52,7 +52,11 @@ static ColumnPtr mockColumn(const DataTypePtr & type, size_t rows) } +#if !defined(DEBUG_OR_SANITIZER_BUILD) static NO_INLINE void insertManyFrom(IColumn & dst, const IColumn & src) +#else +static NO_INLINE void doInsertManyFrom(IColumn & dst, const IColumn & src) +#endif { size_t size = src.size(); dst.insertManyFrom(src, size / 2, size); diff --git a/src/Columns/tests/gtest_weak_hash_32.cpp b/src/Columns/tests/gtest_weak_hash_32.cpp index 2c95998761b..3143d0ff83c 100644 --- a/src/Columns/tests/gtest_weak_hash_32.cpp +++ b/src/Columns/tests/gtest_weak_hash_32.cpp @@ -60,8 +60,7 @@ TEST(WeakHash32, ColumnVectorU8) data.push_back(i); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -77,8 +76,7 @@ TEST(WeakHash32, ColumnVectorI8) data.push_back(i); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -94,8 +92,7 @@ TEST(WeakHash32, ColumnVectorU16) data.push_back(i); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -111,8 +108,7 @@ TEST(WeakHash32, ColumnVectorI16) data.push_back(i); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -128,8 +124,7 @@ TEST(WeakHash32, ColumnVectorU32) data.push_back(i << 16u); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -145,8 +140,7 @@ TEST(WeakHash32, ColumnVectorI32) data.push_back(i << 16); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -162,8 +156,7 @@ TEST(WeakHash32, ColumnVectorU64) data.push_back(i << 32u); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -179,8 +172,7 @@ TEST(WeakHash32, ColumnVectorI64) data.push_back(i << 32); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -204,8 +196,7 @@ TEST(WeakHash32, ColumnVectorU128) } } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), eq_data); } @@ -221,8 +212,7 @@ TEST(WeakHash32, ColumnVectorI128) data.push_back(i << 32); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -238,8 +228,7 @@ TEST(WeakHash32, ColumnDecimal32) data.push_back(i << 16); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -255,8 +244,7 @@ TEST(WeakHash32, ColumnDecimal64) data.push_back(i << 32); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -272,8 +260,7 @@ TEST(WeakHash32, ColumnDecimal128) data.push_back(i << 32); } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), col->getData()); } @@ -294,8 +281,7 @@ TEST(WeakHash32, ColumnString1) } } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), data); } @@ -331,8 +317,7 @@ TEST(WeakHash32, ColumnString2) } } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), data); } @@ -369,8 +354,7 @@ TEST(WeakHash32, ColumnString3) } } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), data); } @@ -397,8 +381,7 @@ TEST(WeakHash32, ColumnFixedString) } } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), data); } @@ -444,8 +427,7 @@ TEST(WeakHash32, ColumnArray) auto col_arr = ColumnArray::create(std::move(val), std::move(off)); - WeakHash32 hash(col_arr->size()); - col_arr->updateWeakHash32(hash); + WeakHash32 hash = col_arr->getWeakHash32(); checkColumn(hash.getData(), eq_data); } @@ -479,8 +461,7 @@ TEST(WeakHash32, ColumnArray2) auto col_arr = ColumnArray::create(std::move(val), std::move(off)); - WeakHash32 hash(col_arr->size()); - col_arr->updateWeakHash32(hash); + WeakHash32 hash = col_arr->getWeakHash32(); checkColumn(hash.getData(), eq_data); } @@ -536,8 +517,7 @@ TEST(WeakHash32, ColumnArrayArray) auto col_arr = ColumnArray::create(std::move(val), std::move(off)); auto col_arr_arr = ColumnArray::create(std::move(col_arr), std::move(off2)); - WeakHash32 hash(col_arr_arr->size()); - col_arr_arr->updateWeakHash32(hash); + WeakHash32 hash = col_arr_arr->getWeakHash32(); checkColumn(hash.getData(), eq_data); } @@ -555,8 +535,7 @@ TEST(WeakHash32, ColumnConst) auto col_const = ColumnConst::create(std::move(inner_col), 256); - WeakHash32 hash(col_const->size()); - col_const->updateWeakHash32(hash); + WeakHash32 hash = col_const->getWeakHash32(); checkColumn(hash.getData(), data); } @@ -576,8 +555,7 @@ TEST(WeakHash32, ColumnLowcardinality) } } - WeakHash32 hash(col->size()); - col->updateWeakHash32(hash); + WeakHash32 hash = col->getWeakHash32(); checkColumn(hash.getData(), data); } @@ -602,8 +580,7 @@ TEST(WeakHash32, ColumnNullable) auto col_null = ColumnNullable::create(std::move(col), std::move(mask)); - WeakHash32 hash(col_null->size()); - col_null->updateWeakHash32(hash); + WeakHash32 hash = col_null->getWeakHash32(); checkColumn(hash.getData(), eq); } @@ -633,8 +610,7 @@ TEST(WeakHash32, ColumnTupleUInt64UInt64) columns.emplace_back(std::move(col2)); auto col_tuple = ColumnTuple::create(std::move(columns)); - WeakHash32 hash(col_tuple->size()); - col_tuple->updateWeakHash32(hash); + WeakHash32 hash = col_tuple->getWeakHash32(); checkColumn(hash.getData(), eq); } @@ -671,8 +647,7 @@ TEST(WeakHash32, ColumnTupleUInt64String) columns.emplace_back(std::move(col2)); auto col_tuple = ColumnTuple::create(std::move(columns)); - WeakHash32 hash(col_tuple->size()); - col_tuple->updateWeakHash32(hash); + WeakHash32 hash = col_tuple->getWeakHash32(); checkColumn(hash.getData(), eq); } @@ -709,8 +684,7 @@ TEST(WeakHash32, ColumnTupleUInt64FixedString) columns.emplace_back(std::move(col2)); auto col_tuple = ColumnTuple::create(std::move(columns)); - WeakHash32 hash(col_tuple->size()); - col_tuple->updateWeakHash32(hash); + WeakHash32 hash = col_tuple->getWeakHash32(); checkColumn(hash.getData(), eq); } @@ -756,8 +730,7 @@ TEST(WeakHash32, ColumnTupleUInt64Array) columns.emplace_back(ColumnArray::create(std::move(val), std::move(off))); auto col_tuple = ColumnTuple::create(std::move(columns)); - WeakHash32 hash(col_tuple->size()); - col_tuple->updateWeakHash32(hash); + WeakHash32 hash = col_tuple->getWeakHash32(); checkColumn(hash.getData(), eq_data); } diff --git a/src/Common/Allocator.cpp b/src/Common/Allocator.cpp index bfc85559fe8..1b43e746c69 100644 --- a/src/Common/Allocator.cpp +++ b/src/Common/Allocator.cpp @@ -68,7 +68,7 @@ void * allocNoTrack(size_t size, size_t alignment) { void * buf; #if USE_GWP_ASAN - if (unlikely(GWPAsan::GuardedAlloc.shouldSample())) + if (unlikely(GWPAsan::shouldSample())) { if (void * ptr = GWPAsan::GuardedAlloc.allocate(size, alignment)) { @@ -185,14 +185,11 @@ void * Allocator::realloc(void * buf, size_t old_size, } #if USE_GWP_ASAN - if (unlikely(GWPAsan::GuardedAlloc.shouldSample())) + if (unlikely(GWPAsan::shouldSample())) { + auto trace_alloc = CurrentMemoryTracker::alloc(new_size); if (void * ptr = GWPAsan::GuardedAlloc.allocate(new_size, alignment)) { - auto trace_free = CurrentMemoryTracker::free(old_size); - auto trace_alloc = CurrentMemoryTracker::alloc(new_size); - trace_free.onFree(buf, old_size); - memcpy(ptr, buf, std::min(old_size, new_size)); free(buf, old_size); trace_alloc.onAlloc(buf, new_size); @@ -209,6 +206,7 @@ void * Allocator::realloc(void * buf, size_t old_size, } else { + [[maybe_unused]] auto trace_free = CurrentMemoryTracker::free(new_size); ProfileEvents::increment(ProfileEvents::GWPAsanAllocateFailed); } } @@ -231,13 +229,17 @@ void * Allocator::realloc(void * buf, size_t old_size, if (alignment <= MALLOC_MIN_ALIGNMENT) { /// Resize malloc'd memory region with no special alignment requirement. - auto trace_free = CurrentMemoryTracker::free(old_size); + /// Realloc can do 2 possible things: + /// - expand existing memory region + /// - allocate new memory block and free the old one + /// Because we don't know which option will be picked we need to make sure there is enough + /// memory for all options auto trace_alloc = CurrentMemoryTracker::alloc(new_size); - trace_free.onFree(buf, old_size); void * new_buf = ::realloc(buf, new_size); if (nullptr == new_buf) { + [[maybe_unused]] auto trace_free = CurrentMemoryTracker::free(new_size); throw DB::ErrnoException( DB::ErrorCodes::CANNOT_ALLOCATE_MEMORY, "Allocator: Cannot realloc from {} to {}", @@ -246,6 +248,8 @@ void * Allocator::realloc(void * buf, size_t old_size, } buf = new_buf; + auto trace_free = CurrentMemoryTracker::free(old_size); + trace_free.onFree(buf, old_size); trace_alloc.onAlloc(buf, new_size); if constexpr (clear_memory) diff --git a/src/Common/AsyncLoader.cpp b/src/Common/AsyncLoader.cpp index cfb273b9058..6264eb03106 100644 --- a/src/Common/AsyncLoader.cpp +++ b/src/Common/AsyncLoader.cpp @@ -49,6 +49,7 @@ void logAboutProgress(LoggerPtr log, size_t processed, size_t total, AtomicStopw AsyncLoader::Pool::Pool(const AsyncLoader::PoolInitializer & init) : name(init.name) , priority(init.priority) + , max_threads(init.max_threads > 0 ? init.max_threads : getNumberOfPhysicalCPUCores()) , thread_pool(std::make_unique( init.metric_threads, init.metric_active_threads, @@ -56,17 +57,16 @@ AsyncLoader::Pool::Pool(const AsyncLoader::PoolInitializer & init) /* max_threads = */ std::numeric_limits::max(), // Unlimited number of threads, we do worker management ourselves /* max_free_threads = */ 0, // We do not require free threads /* queue_size = */0)) // Unlimited queue to avoid blocking during worker spawning - , max_threads(init.max_threads > 0 ? init.max_threads : getNumberOfPhysicalCPUCores()) {} AsyncLoader::Pool::Pool(Pool&& o) noexcept : name(o.name) , priority(o.priority) - , thread_pool(std::move(o.thread_pool)) , ready_queue(std::move(o.ready_queue)) , max_threads(o.max_threads) , workers(o.workers) , suspended_workers(o.suspended_workers.load()) // All these constructors are needed because std::atomic is neither copy-constructible, nor move-constructible. We never move pools after init, so it is safe. + , thread_pool(std::move(o.thread_pool)) {} void cancelOnDependencyFailure(const LoadJobPtr & self, const LoadJobPtr & dependency, std::exception_ptr & cancel) diff --git a/src/Common/AsyncLoader.h b/src/Common/AsyncLoader.h index 42707a4ee91..05b809aceae 100644 --- a/src/Common/AsyncLoader.h +++ b/src/Common/AsyncLoader.h @@ -365,11 +365,11 @@ private: { const String name; const Priority priority; - std::unique_ptr thread_pool; // NOTE: we avoid using a `ThreadPool` queue to be able to move jobs between pools. std::map ready_queue; // FIFO queue of jobs to be executed in this pool. Map is used for faster erasing. Key is `ready_seqno` size_t max_threads; // Max number of workers to be spawn size_t workers = 0; // Number of currently executing workers std::atomic suspended_workers{0}; // Number of workers that are blocked by `wait()` call on a job executing in the same pool (for deadlock resolution) + std::unique_ptr thread_pool; // NOTE: we avoid using a `ThreadPool` queue to be able to move jobs between pools. explicit Pool(const PoolInitializer & init); Pool(Pool&& o) noexcept; diff --git a/src/Common/AtomicLogger.h b/src/Common/AtomicLogger.h index 0ece9e8a09a..c1bbdb41866 100644 --- a/src/Common/AtomicLogger.h +++ b/src/Common/AtomicLogger.h @@ -1,5 +1,7 @@ #pragma once +#include +#include #include #include diff --git a/src/Common/BinStringDecodeHelper.h b/src/Common/BinStringDecodeHelper.h index df3e014cfad..03c175fd37f 100644 --- a/src/Common/BinStringDecodeHelper.h +++ b/src/Common/BinStringDecodeHelper.h @@ -5,7 +5,7 @@ namespace DB { -static void inline hexStringDecode(const char * pos, const char * end, char *& out, size_t word_size = 2) +static void inline hexStringDecode(const char * pos, const char * end, char *& out, size_t word_size) { if ((end - pos) & 1) { @@ -23,7 +23,7 @@ static void inline hexStringDecode(const char * pos, const char * end, char *& o ++out; } -static void inline binStringDecode(const char * pos, const char * end, char *& out) +static void inline binStringDecode(const char * pos, const char * end, char *& out, size_t word_size) { if (pos == end) { @@ -53,7 +53,7 @@ static void inline binStringDecode(const char * pos, const char * end, char *& o ++out; } - assert((end - pos) % 8 == 0); + chassert((end - pos) % word_size == 0); while (end - pos != 0) { diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp index 8a4792f0a5a..ef8bdfc1823 100644 --- a/src/Common/CgroupsMemoryUsageObserver.cpp +++ b/src/Common/CgroupsMemoryUsageObserver.cpp @@ -11,8 +11,11 @@ #include #include #include +#include +#include #include +#include #include #include "config.h" @@ -22,24 +25,189 @@ #define STRINGIFY(x) STRINGIFY_HELPER(x) #endif +using namespace DB; +namespace fs = std::filesystem; namespace DB { namespace ErrorCodes { - extern const int CANNOT_CLOSE_FILE; - extern const int CANNOT_OPEN_FILE; - extern const int FILE_DOESNT_EXIST; - extern const int INCORRECT_DATA; +extern const int FILE_DOESNT_EXIST; +extern const int INCORRECT_DATA; } -CgroupsMemoryUsageObserver::CgroupsMemoryUsageObserver(std::chrono::seconds wait_time_) - : log(getLogger("CgroupsMemoryUsageObserver")) - , wait_time(wait_time_) - , memory_usage_file(log) +} + +namespace { - LOG_INFO(log, "Initialized cgroups memory limit observer, wait time is {} sec", wait_time.count()); + +/// Format is +/// kernel 5 +/// rss 15 +/// [...] +using Metrics = std::map; + +Metrics readAllMetricsFromStatFile(ReadBufferFromFile & buf) +{ + Metrics metrics; + while (!buf.eof()) + { + std::string current_key; + readStringUntilWhitespace(current_key, buf); + + assertChar(' ', buf); + + uint64_t value = 0; + readIntText(value, buf); + assertChar('\n', buf); + + auto [_, inserted] = metrics.emplace(std::move(current_key), value); + chassert(inserted, "Duplicate keys in stat file"); + } + return metrics; +} + +uint64_t readMetricFromStatFile(ReadBufferFromFile & buf, const std::string & key) +{ + const auto all_metrics = readAllMetricsFromStatFile(buf); + if (const auto it = all_metrics.find(key); it != all_metrics.end()) + return it->second; + throw Exception(ErrorCodes::INCORRECT_DATA, "Cannot find '{}' in '{}'", key, buf.getFileName()); +} + +struct CgroupsV1Reader : ICgroupsReader +{ + explicit CgroupsV1Reader(const fs::path & stat_file_dir) : buf(stat_file_dir / "memory.stat") { } + + uint64_t readMemoryUsage() override + { + std::lock_guard lock(mutex); + buf.rewind(); + return readMetricFromStatFile(buf, "rss"); + } + + std::string dumpAllStats() override + { + std::lock_guard lock(mutex); + buf.rewind(); + return fmt::format("{}", readAllMetricsFromStatFile(buf)); + } + +private: + std::mutex mutex; + ReadBufferFromFile buf TSA_GUARDED_BY(mutex); +}; + +struct CgroupsV2Reader : ICgroupsReader +{ + explicit CgroupsV2Reader(const fs::path & stat_file_dir) + : current_buf(stat_file_dir / "memory.current"), stat_buf(stat_file_dir / "memory.stat") + { + } + + uint64_t readMemoryUsage() override + { + std::lock_guard lock(mutex); + current_buf.rewind(); + stat_buf.rewind(); + + int64_t mem_usage = 0; + /// memory.current contains a single number + /// the reason why we subtract it described here: https://github.com/ClickHouse/ClickHouse/issues/64652#issuecomment-2149630667 + readIntText(mem_usage, current_buf); + mem_usage -= readMetricFromStatFile(stat_buf, "inactive_file"); + chassert(mem_usage >= 0, "Negative memory usage"); + return mem_usage; + } + + std::string dumpAllStats() override + { + std::lock_guard lock(mutex); + stat_buf.rewind(); + return fmt::format("{}", readAllMetricsFromStatFile(stat_buf)); + } + +private: + std::mutex mutex; + ReadBufferFromFile current_buf TSA_GUARDED_BY(mutex); + ReadBufferFromFile stat_buf TSA_GUARDED_BY(mutex); +}; + +/// Caveats: +/// - All of the logic in this file assumes that the current process is the only process in the +/// containing cgroup (or more precisely: the only process with significant memory consumption). +/// If this is not the case, then other processe's memory consumption may affect the internal +/// memory tracker ... +/// - Cgroups v1 and v2 allow nested cgroup hierarchies. As v1 is deprecated for over half a +/// decade and will go away at some point, hierarchical detection is only implemented for v2. +/// - I did not test what happens if a host has v1 and v2 simultaneously enabled. I believe such +/// systems existed only for a short transition period. + +std::optional getCgroupsV2Path() +{ + if (!cgroupsV2Enabled()) + return {}; + + if (!cgroupsV2MemoryControllerEnabled()) + return {}; + + fs::path current_cgroup = cgroupV2PathOfProcess(); + if (current_cgroup.empty()) + return {}; + + /// Return the bottom-most nested current memory file. If there is no such file at the current + /// level, try again at the parent level as memory settings are inherited. + while (current_cgroup != default_cgroups_mount.parent_path()) + { + const auto current_path = current_cgroup / "memory.current"; + const auto stat_path = current_cgroup / "memory.stat"; + if (fs::exists(current_path) && fs::exists(stat_path)) + return {current_cgroup}; + current_cgroup = current_cgroup.parent_path(); + } + return {}; +} + +std::optional getCgroupsV1Path() +{ + auto path = default_cgroups_mount / "memory/memory.stat"; + if (!fs::exists(path)) + return {}; + return {default_cgroups_mount / "memory"}; +} + +std::pair getCgroupsPath() +{ + auto v2_path = getCgroupsV2Path(); + if (v2_path.has_value()) + return {*v2_path, CgroupsMemoryUsageObserver::CgroupsVersion::V2}; + + auto v1_path = getCgroupsV1Path(); + if (v1_path.has_value()) + return {*v1_path, CgroupsMemoryUsageObserver::CgroupsVersion::V1}; + + throw Exception(ErrorCodes::FILE_DOESNT_EXIST, "Cannot find cgroups v1 or v2 current memory file"); +} + +} + +namespace DB +{ + +CgroupsMemoryUsageObserver::CgroupsMemoryUsageObserver(std::chrono::seconds wait_time_) + : log(getLogger("CgroupsMemoryUsageObserver")), wait_time(wait_time_) +{ + const auto [cgroup_path, version] = getCgroupsPath(); + + cgroup_reader = createCgroupsReader(version, cgroup_path); + + LOG_INFO( + log, + "Will read the current memory usage from '{}' (cgroups version: {}), wait time is {} sec", + cgroup_path, + (version == CgroupsVersion::V1) ? "v1" : "v2", + wait_time.count()); } CgroupsMemoryUsageObserver::~CgroupsMemoryUsageObserver() @@ -79,12 +247,18 @@ void CgroupsMemoryUsageObserver::setMemoryUsageLimits(uint64_t hard_limit_, uint { LOG_WARNING(log, "Exceeded soft memory limit ({})", ReadableSize(soft_limit_)); -#if USE_JEMALLOC +# if USE_JEMALLOC LOG_INFO(log, "Purging jemalloc arenas"); mallctl("arena." STRINGIFY(MALLCTL_ARENAS_ALL) ".purge", nullptr, nullptr, nullptr, 0); -#endif +# endif /// Reset current usage in memory tracker. Expect zero for free_memory_in_allocator_arenas as we just purged them. - uint64_t memory_usage = memory_usage_file.readMemoryUsage(); + uint64_t memory_usage = cgroup_reader->readMemoryUsage(); + LOG_TRACE( + log, + "Read current memory usage {} bytes ({}) from cgroups, full available stats: {}", + memory_usage, + ReadableSize(memory_usage), + cgroup_reader->dumpAllStats()); MemoryTracker::setRSS(memory_usage, 0); LOG_INFO(log, "Purged jemalloc arenas. Current memory usage is {}", ReadableSize(memory_usage)); @@ -104,152 +278,6 @@ void CgroupsMemoryUsageObserver::setOnMemoryAmountAvailableChangedFn(OnMemoryAmo on_memory_amount_available_changed = on_memory_amount_available_changed_; } -namespace -{ - -/// Caveats: -/// - All of the logic in this file assumes that the current process is the only process in the -/// containing cgroup (or more precisely: the only process with significant memory consumption). -/// If this is not the case, then other processe's memory consumption may affect the internal -/// memory tracker ... -/// - Cgroups v1 and v2 allow nested cgroup hierarchies. As v1 is deprecated for over half a -/// decade and will go away at some point, hierarchical detection is only implemented for v2. -/// - I did not test what happens if a host has v1 and v2 simultaneously enabled. I believe such -/// systems existed only for a short transition period. - -std::optional getCgroupsV2FileName() -{ - if (!cgroupsV2Enabled()) - return {}; - - if (!cgroupsV2MemoryControllerEnabled()) - return {}; - - String cgroup = cgroupV2OfProcess(); - auto current_cgroup = cgroup.empty() ? default_cgroups_mount : (default_cgroups_mount / cgroup); - - /// Return the bottom-most nested current memory file. If there is no such file at the current - /// level, try again at the parent level as memory settings are inherited. - while (current_cgroup != default_cgroups_mount.parent_path()) - { - auto path = current_cgroup / "memory.current"; - if (std::filesystem::exists(path)) - return {path}; - current_cgroup = current_cgroup.parent_path(); - } - return {}; -} - -std::optional getCgroupsV1FileName() -{ - auto path = default_cgroups_mount / "memory/memory.stat"; - if (!std::filesystem::exists(path)) - return {}; - return {path}; -} - -std::pair getCgroupsFileName() -{ - auto v2_file_name = getCgroupsV2FileName(); - if (v2_file_name.has_value()) - return {*v2_file_name, CgroupsMemoryUsageObserver::CgroupsVersion::V2}; - - auto v1_file_name = getCgroupsV1FileName(); - if (v1_file_name.has_value()) - return {*v1_file_name, CgroupsMemoryUsageObserver::CgroupsVersion::V1}; - - throw Exception(ErrorCodes::FILE_DOESNT_EXIST, "Cannot find cgroups v1 or v2 current memory file"); -} - -} - -CgroupsMemoryUsageObserver::MemoryUsageFile::MemoryUsageFile(LoggerPtr log_) - : log(log_) -{ - std::tie(file_name, version) = getCgroupsFileName(); - - LOG_INFO(log, "Will read the current memory usage from '{}' (cgroups version: {})", file_name, (version == CgroupsVersion::V1) ? "v1" : "v2"); - - fd = ::open(file_name.data(), O_RDONLY); - if (fd == -1) - ErrnoException::throwFromPath( - (errno == ENOENT) ? ErrorCodes::FILE_DOESNT_EXIST : ErrorCodes::CANNOT_OPEN_FILE, - file_name, "Cannot open file '{}'", file_name); -} - -CgroupsMemoryUsageObserver::MemoryUsageFile::~MemoryUsageFile() -{ - assert(fd != -1); - if (::close(fd) != 0) - { - try - { - ErrnoException::throwFromPath( - ErrorCodes::CANNOT_CLOSE_FILE, - file_name, "Cannot close file '{}'", file_name); - } - catch (const ErrnoException &) - { - tryLogCurrentException(log, __PRETTY_FUNCTION__); - } - } -} - -uint64_t CgroupsMemoryUsageObserver::MemoryUsageFile::readMemoryUsage() const -{ - /// File read is probably not read is thread-safe, just to be sure - std::lock_guard lock(mutex); - - ReadBufferFromFileDescriptor buf(fd); - buf.rewind(); - - uint64_t mem_usage = 0; - - switch (version) - { - case CgroupsVersion::V1: - { - /// Format is - /// kernel 5 - /// rss 15 - /// [...] - std::string key; - bool found_rss = false; - - while (!buf.eof()) - { - readStringUntilWhitespace(key, buf); - if (key != "rss") - { - std::string dummy; - readStringUntilNewlineInto(dummy, buf); - buf.ignore(); - continue; - } - - assertChar(' ', buf); - readIntText(mem_usage, buf); - found_rss = true; - break; - } - - if (!found_rss) - throw Exception(ErrorCodes::INCORRECT_DATA, "Cannot find 'rss' in '{}'", file_name); - - break; - } - case CgroupsVersion::V2: - { - readIntText(mem_usage, buf); - break; - } - } - - LOG_TRACE(log, "Read current memory usage {} from cgroups", ReadableSize(mem_usage)); - - return mem_usage; -} - void CgroupsMemoryUsageObserver::startThread() { if (!thread.joinable()) @@ -301,7 +329,8 @@ void CgroupsMemoryUsageObserver::runThread() std::lock_guard limit_lock(limit_mutex); if (soft_limit > 0 && hard_limit > 0) { - uint64_t memory_usage = memory_usage_file.readMemoryUsage(); + uint64_t memory_usage = cgroup_reader->readMemoryUsage(); + LOG_TRACE(log, "Read current memory usage {} bytes ({}) from cgroups", memory_usage, ReadableSize(memory_usage)); if (memory_usage > hard_limit) { if (last_memory_usage <= hard_limit) @@ -333,6 +362,13 @@ void CgroupsMemoryUsageObserver::runThread() } } +std::unique_ptr createCgroupsReader(CgroupsMemoryUsageObserver::CgroupsVersion version, const fs::path & cgroup_path) +{ + if (version == CgroupsMemoryUsageObserver::CgroupsVersion::V2) + return std::make_unique(cgroup_path); + else + return std::make_unique(cgroup_path); +} } #endif diff --git a/src/Common/CgroupsMemoryUsageObserver.h b/src/Common/CgroupsMemoryUsageObserver.h index edc1cee750a..7f888fe631b 100644 --- a/src/Common/CgroupsMemoryUsageObserver.h +++ b/src/Common/CgroupsMemoryUsageObserver.h @@ -3,11 +3,21 @@ #include #include +#include #include namespace DB { +struct ICgroupsReader +{ + virtual ~ICgroupsReader() = default; + + virtual uint64_t readMemoryUsage() = 0; + + virtual std::string dumpAllStats() = 0; +}; + /// Does two things: /// 1. Periodically reads the memory usage of the process from Linux cgroups. /// You can specify soft or hard memory limits: @@ -61,33 +71,21 @@ private: uint64_t last_memory_usage = 0; /// how much memory does the process use uint64_t last_available_memory_amount; /// how much memory can the process use - /// Represents the cgroup virtual file that shows the memory consumption of the process's cgroup. - struct MemoryUsageFile - { - public: - explicit MemoryUsageFile(LoggerPtr log_); - ~MemoryUsageFile(); - uint64_t readMemoryUsage() const; - private: - LoggerPtr log; - mutable std::mutex mutex; - int fd TSA_GUARDED_BY(mutex) = -1; - CgroupsVersion version; - std::string file_name; - }; - - MemoryUsageFile memory_usage_file; - void stopThread(); void runThread(); + std::unique_ptr cgroup_reader; + std::mutex thread_mutex; std::condition_variable cond; ThreadFromGlobalPool thread; bool quit = false; }; +std::unique_ptr +createCgroupsReader(CgroupsMemoryUsageObserver::CgroupsVersion version, const std::filesystem::path & cgroup_path); + #else class CgroupsMemoryUsageObserver { diff --git a/src/Common/CollectionOfDerived.h b/src/Common/CollectionOfDerived.h new file mode 100644 index 00000000000..9f80ff727b4 --- /dev/null +++ b/src/Common/CollectionOfDerived.h @@ -0,0 +1,187 @@ +#pragma once + +#include + +#include + +#include +#include +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +/* This is a collections of objects derived from ItemBase. +* Collection contains no more than one instance for each derived type. +* The derived type is used to access the instance. +*/ + +template +class CollectionOfDerivedItems +{ +public: + using Self = CollectionOfDerivedItems; + using ItemPtr = std::shared_ptr; + +private: + struct Rec + { + std::type_index type_idx; + ItemPtr ptr; + + bool operator<(const Rec & other) const + { + return type_idx < other.type_idx; + } + + bool operator<(const std::type_index & value) const + { + return type_idx < value; + } + + bool operator==(const Rec & other) const + { + return type_idx == other.type_idx; + } + }; + using Records = std::vector; + +public: + void swap(Self & other) noexcept + { + records.swap(other.records); + } + + void clear() + { + records.clear(); + } + + bool empty() const + { + return records.empty(); + } + + size_t size() const + { + return records.size(); + } + + Self clone() const + { + Self result; + result.records.reserve(records.size()); + for (const auto & rec : records) + result.records.emplace_back(rec.type_idx, rec.ptr->clone()); + return result; + } + + // append items for other inscnace only if there is no such item in current instance + void appendIfUniq(Self && other) + { + auto middle_idx = records.size(); + std::move(other.records.begin(), other.records.end(), std::back_inserter(records)); + // merge is stable + std::inplace_merge(records.begin(), records.begin() + middle_idx, records.end()); + // remove duplicates + records.erase(std::unique(records.begin(), records.end()), records.end()); + + assert(std::is_sorted(records.begin(), records.end())); + assert(isUniqTypes()); + } + + template + void add(std::shared_ptr info) + { + static_assert(std::is_base_of_v, "Template parameter must inherit items base class"); + return addImpl(std::type_index(typeid(T)), std::move(info)); + } + + template + std::shared_ptr get() const + { + static_assert(std::is_base_of_v, "Template parameter must inherit items base class"); + auto it = getImpl(std::type_index(typeid(T))); + if (it == records.cend()) + return nullptr; + auto cast = std::dynamic_pointer_cast(it->ptr); + chassert(cast); + return cast; + } + + template + std::shared_ptr extract() + { + static_assert(std::is_base_of_v, "Template parameter must inherit items base class"); + auto it = getImpl(std::type_index(typeid(T))); + if (it == records.cend()) + return nullptr; + auto cast = std::dynamic_pointer_cast(it->ptr); + chassert(cast); + + records.erase(it); + return cast; + } + + std::string debug() const + { + std::string result; + + for (auto & rec : records) + { + result.append(rec.type_idx.name()); + result.append(" "); + } + + return result; + } + +private: + bool isUniqTypes() const + { + auto uniq_it = std::adjacent_find(records.begin(), records.end()); + return uniq_it == records.end(); + } + + void addImpl(std::type_index type_idx, ItemPtr item) + { + auto it = std::lower_bound(records.begin(), records.end(), type_idx); + + if (it == records.end()) + { + records.emplace_back(type_idx, item); + return; + } + + if (it->type_idx == type_idx) + throw Exception(ErrorCodes::LOGICAL_ERROR, "inserted items must be unique by their type, type {} is inserted twice", type_idx.name()); + + + records.emplace(it, type_idx, item); + } + + Records::const_iterator getImpl(std::type_index type_idx) const + { + auto it = std::lower_bound(records.cbegin(), records.cend(), type_idx); + + if (it == records.cend()) + return records.cend(); + + if (it->type_idx != type_idx) + return records.cend(); + + return it; + } + + Records records; +}; + +} diff --git a/src/Common/ConcurrencyControl.h b/src/Common/ConcurrencyControl.h index ba94502962c..9d35d7cb8b0 100644 --- a/src/Common/ConcurrencyControl.h +++ b/src/Common/ConcurrencyControl.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include #include diff --git a/src/Common/ConcurrentBoundedQueue.h b/src/Common/ConcurrentBoundedQueue.h index 922607da813..16b9488c98d 100644 --- a/src/Common/ConcurrentBoundedQueue.h +++ b/src/Common/ConcurrentBoundedQueue.h @@ -1,8 +1,6 @@ #pragma once #include -#include -#include #include #include #include @@ -200,22 +198,18 @@ public: */ bool finish() { - bool was_finished_before = false; - { std::lock_guard lock(queue_mutex); if (is_finished) return true; - was_finished_before = is_finished; is_finished = true; } pop_condition.notify_all(); push_condition.notify_all(); - - return was_finished_before; + return false; } /// Returns if queue is finished diff --git a/src/Common/Config/AbstractConfigurationComparison.cpp b/src/Common/Config/AbstractConfigurationComparison.cpp index 607b583cf31..80c837ed43b 100644 --- a/src/Common/Config/AbstractConfigurationComparison.cpp +++ b/src/Common/Config/AbstractConfigurationComparison.cpp @@ -38,7 +38,7 @@ namespace std::erase_if(left_subkeys, [&](const String & key) { return ignore_keys->contains(key); }); std::erase_if(right_subkeys, [&](const String & key) { return ignore_keys->contains(key); }); -#if defined(ABORT_ON_LOGICAL_ERROR) +#if defined(DEBUG_OR_SANITIZER_BUILD) /// Compound `ignore_keys` are not yet implemented. for (const auto & ignore_key : *ignore_keys) chassert(ignore_key.find('.') == std::string_view::npos); diff --git a/src/Common/Config/ConfigProcessor.cpp b/src/Common/Config/ConfigProcessor.cpp index c9832e8efd5..67d6036aa51 100644 --- a/src/Common/Config/ConfigProcessor.cpp +++ b/src/Common/Config/ConfigProcessor.cpp @@ -316,7 +316,6 @@ void ConfigProcessor::mergeRecursive(XMLDocumentPtr config, Node * config_root, } else if (replace) { - with_element.removeAttribute("replace"); NodePtr new_node = config->importNode(with_node, true); config_root->replaceChild(new_node, config_node); } diff --git a/src/Common/Coverage.cpp b/src/Common/Coverage.cpp new file mode 100644 index 00000000000..a21efe62fb6 --- /dev/null +++ b/src/Common/Coverage.cpp @@ -0,0 +1,65 @@ +#include + +#if defined(SANITIZE_COVERAGE) + +#include +#include + +#include +#include + +#include +#include + +#include + +/// Macros to avoid using strlen(), since it may fail if SSE is not supported. +#define writeError(data) do \ + { \ + static_assert(__builtin_constant_p(data)); \ + if (!writeRetry(STDERR_FILENO, data, sizeof(data) - 1)) \ + _Exit(1); \ + } while (false) + +__attribute__((no_sanitize("coverage"))) +void dumpCoverage() +{ + /// A user can request to dump the coverage information into files at exit. + /// This is useful for non-server applications such as clickhouse-format or clickhouse-client, + /// that cannot introspect it with SQL functions at runtime. + + /// The CLICKHOUSE_WRITE_COVERAGE environment variable defines a prefix for a filename 'prefix.pid' + /// containing the list of addresses of covered . + + /// The format is even simpler than Clang's "sancov": an array of 64-bit addresses, native byte order, no header. + + if (const char * coverage_filename_prefix = getenv("CLICKHOUSE_WRITE_COVERAGE")) // NOLINT(concurrency-mt-unsafe) + { + auto dump = [](const std::string & name, auto span) + { + /// Write only non-zeros. + std::vector data; + data.reserve(span.size()); + for (auto addr : span) + if (addr) + data.push_back(addr); + + int fd = ::open(name.c_str(), O_WRONLY | O_CREAT | O_TRUNC | O_CLOEXEC, 0400); + if (-1 == fd) + { + writeError("Cannot open a file to write the coverage data\n"); + } + else + { + if (!writeRetry(fd, reinterpret_cast(data.data()), data.size() * sizeof(data[0]))) + writeError("Cannot write the coverage data to a file\n"); + if (0 != ::close(fd)) + writeError("Cannot close the file with coverage data\n"); + } + }; + + dump(fmt::format("{}.{}", coverage_filename_prefix, getpid()), getCumulativeCoverage()); + } +} +#endif + diff --git a/src/Common/Coverage.h b/src/Common/Coverage.h new file mode 100644 index 00000000000..aa6dd2825ed --- /dev/null +++ b/src/Common/Coverage.h @@ -0,0 +1,5 @@ +#pragma once + +#if defined(SANITIZE_COVERAGE) +void dumpCoverage(); +#endif diff --git a/src/Common/CurrentMetrics.cpp b/src/Common/CurrentMetrics.cpp index 8516a88c7af..1011ab12d15 100644 --- a/src/Common/CurrentMetrics.cpp +++ b/src/Common/CurrentMetrics.cpp @@ -234,10 +234,10 @@ M(PartsCommitted, "Deprecated. See PartsActive.") \ M(PartsPreActive, "The part is in data_parts, but not used for SELECTs.") \ M(PartsActive, "Active data part, used by current and upcoming SELECTs.") \ - M(AttachedDatabase, "Active database, used by current and upcoming SELECTs.") \ - M(AttachedTable, "Active table, used by current and upcoming SELECTs.") \ - M(AttachedView, "Active view, used by current and upcoming SELECTs.") \ - M(AttachedDictionary, "Active dictionary, used by current and upcoming SELECTs.") \ + M(AttachedDatabase, "Active databases.") \ + M(AttachedTable, "Active tables.") \ + M(AttachedView, "Active views.") \ + M(AttachedDictionary, "Active dictionaries.") \ M(PartsOutdated, "Not active data part, but could be used by only current SELECTs, could be deleted after SELECTs finishes.") \ M(PartsDeleting, "Not active data part with identity refcounter, it is deleting right now by a cleaner.") \ M(PartsDeleteOnDestroy, "Part was moved to another disk and should be deleted in own destructor.") \ @@ -267,7 +267,7 @@ M(AsyncInsertCacheSize, "Number of async insert hash id in cache") \ M(S3Requests, "S3 requests count") \ M(KeeperAliveConnections, "Number of alive connections") \ - M(KeeperOutstandingRequets, "Number of outstanding requests") \ + M(KeeperOutstandingRequests, "Number of outstanding requests") \ M(ThreadsInOvercommitTracker, "Number of waiting threads inside of OvercommitTracker") \ M(IOUringPendingEvents, "Number of io_uring SQEs waiting to be submitted") \ M(IOUringInFlightEvents, "Number of io_uring SQEs in flight") \ diff --git a/src/Common/DateLUT.cpp b/src/Common/DateLUT.cpp index 3a20fb1a125..f9bbe2fbf40 100644 --- a/src/Common/DateLUT.cpp +++ b/src/Common/DateLUT.cpp @@ -2,7 +2,9 @@ #include #include +#include #include +#include #include #include diff --git a/src/Common/EnvironmentChecks.cpp b/src/Common/EnvironmentChecks.cpp new file mode 100644 index 00000000000..d69e8cbaa3d --- /dev/null +++ b/src/Common/EnvironmentChecks.cpp @@ -0,0 +1,234 @@ +#include +#include + +#include + +#include +#include +#include + +#include + +#include + +namespace +{ + +enum class InstructionFail : uint8_t +{ + NONE = 0, + SSE3 = 1, + SSSE3 = 2, + SSE4_1 = 3, + SSE4_2 = 4, + POPCNT = 5, + AVX = 6, + AVX2 = 7, + AVX512 = 8 +}; + +auto instructionFailToString(InstructionFail fail) +{ + switch (fail) + { +#define ret(x) return std::make_tuple(STDERR_FILENO, x, sizeof(x) - 1) + case InstructionFail::NONE: + ret("NONE"); + case InstructionFail::SSE3: + ret("SSE3"); + case InstructionFail::SSSE3: + ret("SSSE3"); + case InstructionFail::SSE4_1: + ret("SSE4.1"); + case InstructionFail::SSE4_2: + ret("SSE4.2"); + case InstructionFail::POPCNT: + ret("POPCNT"); + case InstructionFail::AVX: + ret("AVX"); + case InstructionFail::AVX2: + ret("AVX2"); + case InstructionFail::AVX512: + ret("AVX512"); +#undef ret + } +} + + +sigjmp_buf jmpbuf; + +[[noreturn]] void sigIllCheckHandler(int, siginfo_t *, void *) +{ + siglongjmp(jmpbuf, 1); +} + +/// Check if necessary SSE extensions are available by trying to execute some sse instructions. +/// If instruction is unavailable, SIGILL will be sent by kernel. +void checkRequiredInstructionsImpl(volatile InstructionFail & fail) +{ +#if defined(__SSE3__) + fail = InstructionFail::SSE3; + __asm__ volatile ("addsubpd %%xmm0, %%xmm0" : : : "xmm0"); +#endif + +#if defined(__SSSE3__) + fail = InstructionFail::SSSE3; + __asm__ volatile ("pabsw %%xmm0, %%xmm0" : : : "xmm0"); + +#endif + +#if defined(__SSE4_1__) + fail = InstructionFail::SSE4_1; + __asm__ volatile ("pmaxud %%xmm0, %%xmm0" : : : "xmm0"); +#endif + +#if defined(__SSE4_2__) + fail = InstructionFail::SSE4_2; + __asm__ volatile ("pcmpgtq %%xmm0, %%xmm0" : : : "xmm0"); +#endif + + /// Defined by -msse4.2 +#if defined(__POPCNT__) + fail = InstructionFail::POPCNT; + { + uint64_t a = 0; + uint64_t b = 0; + __asm__ volatile ("popcnt %1, %0" : "=r"(a) :"r"(b) :); + } +#endif + +#if defined(__AVX__) + fail = InstructionFail::AVX; + __asm__ volatile ("vaddpd %%ymm0, %%ymm0, %%ymm0" : : : "ymm0"); +#endif + +#if defined(__AVX2__) + fail = InstructionFail::AVX2; + __asm__ volatile ("vpabsw %%ymm0, %%ymm0" : : : "ymm0"); +#endif + +#if defined(__AVX512__) + fail = InstructionFail::AVX512; + __asm__ volatile ("vpabsw %%zmm0, %%zmm0" : : : "zmm0"); +#endif + + fail = InstructionFail::NONE; +} + +/// Macros to avoid using strlen(), since it may fail if SSE is not supported. +#define writeError(data) do \ + { \ + static_assert(__builtin_constant_p(data)); \ + if (!writeRetry(STDERR_FILENO, data, sizeof(data) - 1)) \ + _Exit(1); \ + } while (false) + +/// Check SSE and others instructions availability. Calls exit on fail. +/// This function must be called as early as possible, even before main, because static initializers may use unavailable instructions. +void checkRequiredInstructions() +{ + struct sigaction sa{}; + struct sigaction sa_old{}; + sa.sa_sigaction = sigIllCheckHandler; + sa.sa_flags = SA_SIGINFO; + auto signal = SIGILL; + if (sigemptyset(&sa.sa_mask) != 0 + || sigaddset(&sa.sa_mask, signal) != 0 + || sigaction(signal, &sa, &sa_old) != 0) + { + /// You may wonder about strlen. + /// Typical implementation of strlen is using SSE4.2 or AVX2. + /// But this is not the case because it's compiler builtin and is executed at compile time. + + writeError("Can not set signal handler\n"); + _Exit(1); + } + + volatile InstructionFail fail = InstructionFail::NONE; + + if (sigsetjmp(jmpbuf, 1)) + { + writeError("Instruction check fail. The CPU does not support "); + if (!std::apply(writeRetry, instructionFailToString(fail))) + _Exit(1); + writeError(" instruction set.\n"); + _Exit(1); + } + + checkRequiredInstructionsImpl(fail); + + if (sigaction(signal, &sa_old, nullptr)) + { + writeError("Can not set signal handler\n"); + _Exit(1); + } +} + +struct Checker +{ + Checker() + { + checkRequiredInstructions(); + } +} checker +#ifndef OS_DARWIN + __attribute__((init_priority(101))) /// Run before other static initializers. +#endif +; + +} + + +#if !defined(USE_MUSL) +/// NOTE: We will migrate to full static linking or our own dynamic loader to make this code obsolete. +void checkHarmfulEnvironmentVariables(char ** argv) +{ + std::initializer_list harmful_env_variables = { + /// The list is a selection from "man ld-linux". + "LD_PRELOAD", + "LD_LIBRARY_PATH", + "LD_ORIGIN_PATH", + "LD_AUDIT", + "LD_DYNAMIC_WEAK", + /// The list is a selection from "man dyld" (osx). + "DYLD_LIBRARY_PATH", + "DYLD_FALLBACK_LIBRARY_PATH", + "DYLD_VERSIONED_LIBRARY_PATH", + "DYLD_INSERT_LIBRARIES", + }; + + bool require_reexec = false; + for (const auto * var : harmful_env_variables) + { + if (const char * value = getenv(var); value && value[0]) // NOLINT(concurrency-mt-unsafe) + { + /// NOTE: setenv() is used over unsetenv() since unsetenv() marked as harmful + if (setenv(var, "", true)) // NOLINT(concurrency-mt-unsafe) // this is safe if not called concurrently + { + fmt::print(stderr, "Cannot override {} environment variable", var); + _exit(1); + } + require_reexec = true; + } + } + + if (require_reexec) + { + /// Use execvp() over execv() to search in PATH. + /// + /// This should be safe, since: + /// - if argv[0] is relative path - it is OK + /// - if argv[0] has only basename, the it will search in PATH, like shell will do. + /// + /// Also note, that this (search in PATH) because there is no easy and + /// portable way to get absolute path of argv[0]. + /// - on linux there is /proc/self/exec and AT_EXECFN + /// - but on other OSes there is no such thing (especially on OSX). + /// + /// And since static linking will be done someday anyway, + /// let's not pollute the code base with special cases. + int error = execvp(argv[0], argv); + _exit(error); + } +} +#endif diff --git a/src/Common/EnvironmentChecks.h b/src/Common/EnvironmentChecks.h new file mode 100644 index 00000000000..6d355a69ff9 --- /dev/null +++ b/src/Common/EnvironmentChecks.h @@ -0,0 +1,5 @@ +#pragma once + +#if !defined(USE_MUSL) +void checkHarmfulEnvironmentVariables(char ** argv); +#endif diff --git a/src/Common/Exception.cpp b/src/Common/Exception.cpp index 1f4b0aea8f2..d68537513da 100644 --- a/src/Common/Exception.cpp +++ b/src/Common/Exception.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -38,10 +39,19 @@ namespace ErrorCodes extern const int CANNOT_MREMAP; } +void abortOnFailedAssertion(const String & description, void * const * trace, size_t trace_offset, size_t trace_size) +{ + auto & logger = Poco::Logger::root(); + LOG_FATAL(&logger, "Logical error: '{}'.", description); + if (trace) + LOG_FATAL(&logger, "Stack trace (when copying this message, always include the lines below):\n\n{}", StackTrace::toString(trace, trace_offset, trace_size)); + abort(); +} + void abortOnFailedAssertion(const String & description) { - LOG_FATAL(&Poco::Logger::root(), "Logical error: '{}'.", description); - abort(); + StackTrace st; + abortOnFailedAssertion(description, st.getFramePointers().data(), st.getOffset(), st.getSize()); } bool terminate_on_any_exception = false; @@ -55,10 +65,10 @@ void handle_error_code(const std::string & msg, int code, bool remote, const Exc { // In debug builds and builds with sanitizers, treat LOGICAL_ERROR as an assertion failure. // Log the message before we fail. -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD if (code == ErrorCodes::LOGICAL_ERROR) { - abortOnFailedAssertion(msg); + abortOnFailedAssertion(msg, trace.data(), 0, trace.size()); } #endif @@ -91,7 +101,7 @@ Exception::Exception(const MessageMasked & msg_masked, int code, bool remote_) { if (terminate_on_any_exception) std::_Exit(terminate_status_code); - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); handle_error_code(msg_masked.msg, code, remote, getStackFramePointers()); } @@ -101,7 +111,7 @@ Exception::Exception(MessageMasked && msg_masked, int code, bool remote_) { if (terminate_on_any_exception) std::_Exit(terminate_status_code); - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); handle_error_code(message(), code, remote, getStackFramePointers()); } @@ -110,7 +120,7 @@ Exception::Exception(CreateFromPocoTag, const Poco::Exception & exc) { if (terminate_on_any_exception) std::_Exit(terminate_status_code); - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); #ifdef STD_EXCEPTION_HAS_STACK_TRACE auto * stack_trace_frames = exc.get_stack_trace_frames(); auto stack_trace_size = exc.get_stack_trace_size(); @@ -124,7 +134,7 @@ Exception::Exception(CreateFromSTDTag, const std::exception & exc) { if (terminate_on_any_exception) std::_Exit(terminate_status_code); - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); #ifdef STD_EXCEPTION_HAS_STACK_TRACE auto * stack_trace_frames = exc.get_stack_trace_frames(); auto stack_trace_size = exc.get_stack_trace_size(); @@ -214,10 +224,38 @@ Exception::FramePointers Exception::getStackFramePointers() const } thread_local bool Exception::enable_job_stack_trace = false; -thread_local std::vector Exception::thread_frame_pointers = {}; +thread_local bool Exception::can_use_thread_frame_pointers = false; +thread_local Exception::ThreadFramePointers Exception::thread_frame_pointers; + +Exception::ThreadFramePointers::ThreadFramePointers() +{ + can_use_thread_frame_pointers = true; +} + +Exception::ThreadFramePointers::~ThreadFramePointers() +{ + can_use_thread_frame_pointers = false; +} + +Exception::ThreadFramePointersBase Exception::getThreadFramePointers() +{ + if (can_use_thread_frame_pointers) + return thread_frame_pointers.frame_pointers; + + return {}; +} + +void Exception::setThreadFramePointers(ThreadFramePointersBase frame_pointers) +{ + if (can_use_thread_frame_pointers) + thread_frame_pointers.frame_pointers = std::move(frame_pointers); +} static void tryLogCurrentExceptionImpl(Poco::Logger * logger, const std::string & start_of_message) { + if (!isLoggingEnabled()) + return; + try { PreformattedMessage message = getCurrentExceptionMessageAndPattern(true); @@ -233,6 +271,9 @@ static void tryLogCurrentExceptionImpl(Poco::Logger * logger, const std::string void tryLogCurrentException(const char * log_name, const std::string & start_of_message) { + if (!isLoggingEnabled()) + return; + /// Under high memory pressure, new allocations throw a /// MEMORY_LIMIT_EXCEEDED exception. /// @@ -434,7 +475,7 @@ PreformattedMessage getCurrentExceptionMessageAndPattern(bool with_stacktrace, b } catch (...) {} // NOLINT(bugprone-empty-catch) -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD try { throw; diff --git a/src/Common/Exception.h b/src/Common/Exception.h index 87ef7101cdc..a4f55f41caa 100644 --- a/src/Common/Exception.h +++ b/src/Common/Exception.h @@ -10,7 +10,6 @@ #include #include -#include #include #include @@ -25,8 +24,6 @@ namespace DB class AtomicLogger; -[[noreturn]] void abortOnFailedAssertion(const String & description); - /// This flag can be set for testing purposes - to check that no exceptions are thrown. extern bool terminate_on_any_exception; @@ -51,14 +48,14 @@ public: { if (terminate_on_any_exception) std::terminate(); - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); } Exception(const PreformattedMessage & msg, int code): Exception(msg.text, code) { if (terminate_on_any_exception) std::terminate(); - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); message_format_string = msg.format_string; message_format_string_args = msg.format_string_args; } @@ -67,18 +64,36 @@ public: { if (terminate_on_any_exception) std::terminate(); - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); message_format_string = msg.format_string; message_format_string_args = msg.format_string_args; } /// Collect call stacks of all previous jobs' schedulings leading to this thread job's execution static thread_local bool enable_job_stack_trace; - static thread_local std::vector thread_frame_pointers; + static thread_local bool can_use_thread_frame_pointers; + /// Because of unknown order of static destructor calls, + /// thread_frame_pointers can already be uninitialized when a different destructor generates an exception. + /// To prevent such scenarios, a wrapper class is created and a function that will return empty vector + /// if its destructor is already called + using ThreadFramePointersBase = std::vector; + struct ThreadFramePointers + { + ThreadFramePointers(); + ~ThreadFramePointers(); + + ThreadFramePointersBase frame_pointers; + }; + + static ThreadFramePointersBase getThreadFramePointers(); + static void setThreadFramePointers(ThreadFramePointersBase frame_pointers); + /// Callback for any exception static std::function callback; protected: + static thread_local ThreadFramePointers thread_frame_pointers; + // used to remove the sensitive information from exceptions if query_masking_rules is configured struct MessageMasked { @@ -167,6 +182,8 @@ protected: mutable std::vector capture_thread_frame_pointers; }; +[[noreturn]] void abortOnFailedAssertion(const String & description, void * const * trace, size_t trace_offset, size_t trace_size); +[[noreturn]] void abortOnFailedAssertion(const String & description); std::string getExceptionStackTraceString(const std::exception & e); std::string getExceptionStackTraceString(std::exception_ptr e); @@ -178,7 +195,7 @@ class ErrnoException : public Exception public: ErrnoException(std::string && msg, int code, int with_errno) : Exception(msg, code), saved_errno(with_errno) { - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); addMessage(", {}", errnoToString(saved_errno)); } @@ -187,7 +204,7 @@ public: requires std::is_convertible_v ErrnoException(int code, T && message) : Exception(message, code), saved_errno(errno) { - capture_thread_frame_pointers = thread_frame_pointers; + capture_thread_frame_pointers = getThreadFramePointers(); addMessage(", {}", errnoToString(saved_errno)); } @@ -244,6 +261,15 @@ private: const char * className() const noexcept override { return "DB::ErrnoException"; } }; +/// An exception to use in unit tests to test interfaces. +/// It is distinguished from others, so it does not have to be logged. +class TestException : public Exception +{ +public: + using Exception::Exception; +}; + + using Exceptions = std::vector; /** Try to write an exception to the log (and forget about it). diff --git a/src/Common/FailPoint.cpp b/src/Common/FailPoint.cpp index 5454cba8e2e..f5ec8cf0356 100644 --- a/src/Common/FailPoint.cpp +++ b/src/Common/FailPoint.cpp @@ -57,7 +57,9 @@ static struct InitFiu PAUSEABLE_ONCE(finish_clean_quorum_failed_parts) \ PAUSEABLE(dummy_pausable_failpoint) \ ONCE(execute_query_calling_empty_set_result_func_on_exception) \ - ONCE(receive_timeout_on_table_status_response) + ONCE(receive_timeout_on_table_status_response) \ + REGULAR(keepermap_fail_drop_data) \ + REGULAR(lazy_pipe_fds_fail_close) \ namespace FailPoints diff --git a/src/Common/FieldBinaryEncoding.cpp b/src/Common/FieldBinaryEncoding.cpp new file mode 100644 index 00000000000..6c1a8496fe6 --- /dev/null +++ b/src/Common/FieldBinaryEncoding.cpp @@ -0,0 +1,389 @@ +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int UNSUPPORTED_METHOD; + extern const int INCORRECT_DATA; +} + +namespace +{ + +enum class FieldBinaryTypeIndex: uint8_t +{ + Null = 0x00, + UInt64 = 0x01, + Int64 = 0x02, + UInt128 = 0x03, + Int128 = 0x04, + UInt256 = 0x05, + Int256 = 0x06, + Float64 = 0x07, + Decimal32 = 0x08, + Decimal64 = 0x09, + Decimal128 = 0x0A, + Decimal256 = 0x0B, + String = 0x0C, + Array = 0x0D, + Tuple = 0x0E, + Map = 0x0F, + IPv4 = 0x10, + IPv6 = 0x11, + UUID = 0x12, + Bool = 0x13, + Object = 0x14, + AggregateFunctionState = 0x15, + + NegativeInfinity = 0xFE, + PositiveInfinity = 0xFF, +}; + +class FieldVisitorEncodeBinary +{ +public: + void operator() (const Null & x, WriteBuffer & buf) const; + void operator() (const UInt64 & x, WriteBuffer & buf) const; + void operator() (const UInt128 & x, WriteBuffer & buf) const; + void operator() (const UInt256 & x, WriteBuffer & buf) const; + void operator() (const Int64 & x, WriteBuffer & buf) const; + void operator() (const Int128 & x, WriteBuffer & buf) const; + void operator() (const Int256 & x, WriteBuffer & buf) const; + void operator() (const UUID & x, WriteBuffer & buf) const; + void operator() (const IPv4 & x, WriteBuffer & buf) const; + void operator() (const IPv6 & x, WriteBuffer & buf) const; + void operator() (const Float64 & x, WriteBuffer & buf) const; + void operator() (const String & x, WriteBuffer & buf) const; + void operator() (const Array & x, WriteBuffer & buf) const; + void operator() (const Tuple & x, WriteBuffer & buf) const; + void operator() (const Map & x, WriteBuffer & buf) const; + void operator() (const Object & x, WriteBuffer & buf) const; + void operator() (const DecimalField & x, WriteBuffer & buf) const; + void operator() (const DecimalField & x, WriteBuffer & buf) const; + void operator() (const DecimalField & x, WriteBuffer & buf) const; + void operator() (const DecimalField & x, WriteBuffer & buf) const; + void operator() (const AggregateFunctionStateData & x, WriteBuffer & buf) const; + [[noreturn]] void operator() (const CustomType & x, WriteBuffer & buf) const; + void operator() (const bool & x, WriteBuffer & buf) const; +}; + +void FieldVisitorEncodeBinary::operator() (const Null & x, WriteBuffer & buf) const +{ + if (x.isNull()) + writeBinary(UInt8(FieldBinaryTypeIndex::Null), buf); + else if (x.isPositiveInfinity()) + writeBinary(UInt8(FieldBinaryTypeIndex::PositiveInfinity), buf); + else if (x.isNegativeInfinity()) + writeBinary(UInt8(FieldBinaryTypeIndex::NegativeInfinity), buf); +} + +void FieldVisitorEncodeBinary::operator() (const UInt64 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::UInt64), buf); + writeVarUInt(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const Int64 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Int64), buf); + writeVarInt(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const Float64 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Float64), buf); + writeBinaryLittleEndian(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const String & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::String), buf); + writeStringBinary(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const UInt128 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::UInt128), buf); + writeBinaryLittleEndian(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const Int128 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Int128), buf); + writeBinaryLittleEndian(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const UInt256 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::UInt256), buf); + writeBinaryLittleEndian(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const Int256 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Int256), buf); + writeBinaryLittleEndian(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const UUID & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::UUID), buf); + writeBinaryLittleEndian(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const IPv4 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::IPv4), buf); + writeBinaryLittleEndian(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const IPv6 & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::IPv6), buf); + writeBinaryLittleEndian(x, buf); +} + +void FieldVisitorEncodeBinary::operator() (const DecimalField & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Decimal32), buf); + writeVarUInt(x.getScale(), buf); + writeBinaryLittleEndian(x.getValue(), buf); +} + +void FieldVisitorEncodeBinary::operator() (const DecimalField & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Decimal64), buf); + writeVarUInt(x.getScale(), buf); + writeBinaryLittleEndian(x.getValue(), buf); +} + +void FieldVisitorEncodeBinary::operator() (const DecimalField & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Decimal128), buf); + writeVarUInt(x.getScale(), buf); + writeBinaryLittleEndian(x.getValue(), buf); +} + +void FieldVisitorEncodeBinary::operator() (const DecimalField & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Decimal256), buf); + writeVarUInt(x.getScale(), buf); + writeBinaryLittleEndian(x.getValue(), buf); +} + +void FieldVisitorEncodeBinary::operator() (const AggregateFunctionStateData & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::AggregateFunctionState), buf); + writeStringBinary(x.name, buf); + writeStringBinary(x.data, buf); +} + +void FieldVisitorEncodeBinary::operator() (const Array & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Array), buf); + size_t size = x.size(); + writeVarUInt(size, buf); + for (size_t i = 0; i < size; ++i) + Field::dispatch([&buf] (const auto & value) { FieldVisitorEncodeBinary()(value, buf); }, x[i]); +} + +void FieldVisitorEncodeBinary::operator() (const Tuple & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Tuple), buf); + size_t size = x.size(); + writeVarUInt(size, buf); + for (size_t i = 0; i < size; ++i) + Field::dispatch([&buf] (const auto & value) { FieldVisitorEncodeBinary()(value, buf); }, x[i]); +} + +void FieldVisitorEncodeBinary::operator() (const Map & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Map), buf); + size_t size = x.size(); + writeVarUInt(size, buf); + for (size_t i = 0; i < size; ++i) + { + const Tuple & key_and_value = x[i].get(); + Field::dispatch([&buf] (const auto & value) { FieldVisitorEncodeBinary()(value, buf); }, key_and_value[0]); + Field::dispatch([&buf] (const auto & value) { FieldVisitorEncodeBinary()(value, buf); }, key_and_value[1]); + } +} + +void FieldVisitorEncodeBinary::operator() (const Object & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Object), buf); + + size_t size = x.size(); + writeVarUInt(size, buf); + for (const auto & [key, value] : x) + { + writeStringBinary(key, buf); + Field::dispatch([&buf] (const auto & val) { FieldVisitorEncodeBinary()(val, buf); }, value); + } +} + +void FieldVisitorEncodeBinary::operator()(const bool & x, WriteBuffer & buf) const +{ + writeBinary(UInt8(FieldBinaryTypeIndex::Bool), buf); + writeBinary(static_cast(x), buf); +} + +[[noreturn]] void FieldVisitorEncodeBinary::operator()(const CustomType &, WriteBuffer &) const +{ + /// TODO: Support binary encoding/decoding for custom types somehow. + throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "Binary encoding of Field with custom type is not supported"); +} + +template +Field decodeBigInteger(ReadBuffer & buf) +{ + T value; + readBinaryLittleEndian(value, buf); + return value; +} + +template +DecimalField decodeDecimal(ReadBuffer & buf) +{ + UInt32 scale; + readVarUInt(scale, buf); + T value; + readBinaryLittleEndian(value, buf); + return DecimalField(value, scale); +} + +template +T decodeValueLittleEndian(ReadBuffer & buf) +{ + T value; + readBinaryLittleEndian(value, buf); + return value; +} + +template +T decodeArrayLikeField(ReadBuffer & buf) +{ + size_t size; + readVarUInt(size, buf); + T value; + for (size_t i = 0; i != size; ++i) + value.push_back(decodeField(buf)); + return value; +} + +} +void encodeField(const Field & x, WriteBuffer & buf) +{ + Field::dispatch([&buf] (const auto & val) { FieldVisitorEncodeBinary()(val, buf); }, x); +} + +Field decodeField(ReadBuffer & buf) +{ + UInt8 type; + readBinary(type, buf); + switch (FieldBinaryTypeIndex(type)) + { + case FieldBinaryTypeIndex::Null: + return Null(); + case FieldBinaryTypeIndex::PositiveInfinity: + return POSITIVE_INFINITY; + case FieldBinaryTypeIndex::NegativeInfinity: + return NEGATIVE_INFINITY; + case FieldBinaryTypeIndex::Int64: + { + Int64 value; + readVarInt(value, buf); + return value; + } + case FieldBinaryTypeIndex::UInt64: + { + UInt64 value; + readVarUInt(value, buf); + return value; + } + case FieldBinaryTypeIndex::Int128: + return decodeBigInteger(buf); + case FieldBinaryTypeIndex::UInt128: + return decodeBigInteger(buf); + case FieldBinaryTypeIndex::Int256: + return decodeBigInteger(buf); + case FieldBinaryTypeIndex::UInt256: + return decodeBigInteger(buf); + case FieldBinaryTypeIndex::Float64: + return decodeValueLittleEndian(buf); + case FieldBinaryTypeIndex::Decimal32: + return decodeDecimal(buf); + case FieldBinaryTypeIndex::Decimal64: + return decodeDecimal(buf); + case FieldBinaryTypeIndex::Decimal128: + return decodeDecimal(buf); + case FieldBinaryTypeIndex::Decimal256: + return decodeDecimal(buf); + case FieldBinaryTypeIndex::String: + { + String value; + readStringBinary(value, buf); + return value; + } + case FieldBinaryTypeIndex::UUID: + return decodeValueLittleEndian(buf); + case FieldBinaryTypeIndex::IPv4: + return decodeValueLittleEndian(buf); + case FieldBinaryTypeIndex::IPv6: + return decodeValueLittleEndian(buf); + case FieldBinaryTypeIndex::Bool: + { + bool value; + readBinary(value, buf); + return value; + } + case FieldBinaryTypeIndex::Array: + return decodeArrayLikeField(buf); + case FieldBinaryTypeIndex::Tuple: + return decodeArrayLikeField(buf); + case FieldBinaryTypeIndex::Map: + { + size_t size; + readVarUInt(size, buf); + Map map; + for (size_t i = 0; i != size; ++i) + { + Tuple key_and_value; + key_and_value.push_back(decodeField(buf)); + key_and_value.push_back(decodeField(buf)); + map.push_back(key_and_value); + } + return map; + } + case FieldBinaryTypeIndex::Object: + { + size_t size; + readVarUInt(size, buf); + Object value; + for (size_t i = 0; i != size; ++i) + { + String name; + readStringBinary(name, buf); + value[name] = decodeField(buf); + } + return value; + } + case FieldBinaryTypeIndex::AggregateFunctionState: + { + String name; + readStringBinary(name, buf); + String data; + readStringBinary(data, buf); + return AggregateFunctionStateData{.name = name, .data = data}; + } + } + + throw Exception(ErrorCodes::INCORRECT_DATA, "Unknown Field type: {0:#04x}", UInt64(type)); +} + +} diff --git a/src/Common/FieldBinaryEncoding.h b/src/Common/FieldBinaryEncoding.h new file mode 100644 index 00000000000..aa6694cb03e --- /dev/null +++ b/src/Common/FieldBinaryEncoding.h @@ -0,0 +1,43 @@ +#pragma once + +#include + +namespace DB +{ + +/** +Binary encoding for Fields: +|--------------------------|--------------------------------------------------------------------------------------------------------------------------------| +| Field type | Binary encoding | +|--------------------------|--------------------------------------------------------------------------------------------------------------------------------| +| `Null` | `0x00` | +| `UInt64` | `0x01` | +| `Int64` | `0x02` | +| `UInt128` | `0x03` | +| `Int128` | `0x04` | +| `UInt128` | `0x05` | +| `Int128` | `0x06` | +| `Float64` | `0x07` | +| `Decimal32` | `0x08` | +| `Decimal64` | `0x09` | +| `Decimal128` | `0x0A` | +| `Decimal256` | `0x0B` | +| `String` | `0x0C` | +| `Array` | `0x0D...` | +| `Tuple` | `0x0E...` | +| `Map` | `0x0F...` | +| `IPv4` | `0x10` | +| `IPv6` | `0x11` | +| `UUID` | `0x12` | +| `Bool` | `0x13` | +| `Object` | `0x14...` | +| `AggregateFunctionState` | `0x15` | +| `Negative infinity` | `0xFE` | +| `Positive infinity` | `0xFF` | +|--------------------------|--------------------------------------------------------------------------------------------------------------------------------| +*/ + +void encodeField(const Field &, WriteBuffer & buf); +Field decodeField(ReadBuffer & buf); + +} diff --git a/src/Common/GWPAsan.cpp b/src/Common/GWPAsan.cpp index 488f8e2c5dc..de6991191ea 100644 --- a/src/Common/GWPAsan.cpp +++ b/src/Common/GWPAsan.cpp @@ -57,9 +57,12 @@ static bool guarded_alloc_initialized = [] opts.MaxSimultaneousAllocations = 1024; if (!env_options_raw || !std::string_view{env_options_raw}.contains("SampleRate")) - opts.SampleRate = 50000; + opts.SampleRate = 10000; + + const char * collect_stacktraces = std::getenv("GWP_ASAN_COLLECT_STACKTRACES"); // NOLINT(concurrency-mt-unsafe) + if (collect_stacktraces && std::string_view{collect_stacktraces} == "1") + opts.Backtrace = getBackTrace; - opts.Backtrace = getBackTrace; GuardedAlloc.init(opts); return true; @@ -214,6 +217,13 @@ void printReport([[maybe_unused]] uintptr_t fault_address) reinterpret_cast(trace.data()), 0, trace_length, [&](const auto line) { LOG_FATAL(logger, fmt::runtime(line)); }); } +std::atomic init_finished = false; + +void initFinished() +{ + init_finished.store(true, std::memory_order_relaxed); +} + std::atomic force_sample_probability = 0.0; void setForceSampleProbability(double value) diff --git a/src/Common/GWPAsan.h b/src/Common/GWPAsan.h index b3215c6157e..846c3417db4 100644 --- a/src/Common/GWPAsan.h +++ b/src/Common/GWPAsan.h @@ -19,12 +19,30 @@ bool isGWPAsanError(uintptr_t fault_address); void printReport(uintptr_t fault_address); +extern std::atomic init_finished; + +void initFinished(); + extern std::atomic force_sample_probability; void setForceSampleProbability(double value); +/** + * We'd like to postpone sampling allocations under the startup is finished. There are mainly + * two reasons for that: + * + * - To avoid complex issues with initialization order + * - Don't waste MaxSimultaneousAllocations on global objects as it's not useful +*/ +inline bool shouldSample() +{ + return init_finished.load(std::memory_order_relaxed) && GuardedAlloc.shouldSample(); +} + inline bool shouldForceSample() { + if (!init_finished.load(std::memory_order_relaxed)) + return false; std::bernoulli_distribution dist(force_sample_probability.load(std::memory_order_relaxed)); return dist(thread_local_rng); } diff --git a/src/Common/GetPriorityForLoadBalancing.cpp b/src/Common/GetPriorityForLoadBalancing.cpp index d4c6f89ff92..77b7867b45d 100644 --- a/src/Common/GetPriorityForLoadBalancing.cpp +++ b/src/Common/GetPriorityForLoadBalancing.cpp @@ -1,3 +1,4 @@ +#include #include #include @@ -60,4 +61,26 @@ GetPriorityForLoadBalancing::getPriorityFunc(LoadBalancing load_balance, size_t return get_priority; } +/// Some load balancing strategies (such as "nearest hostname") have preferred nodes to connect to. +/// Usually it's a node in the same data center/availability zone. +/// For other strategies there's no difference between nodes. +bool GetPriorityForLoadBalancing::hasOptimalNode() const +{ + switch (load_balancing) + { + case LoadBalancing::NEAREST_HOSTNAME: + return true; + case LoadBalancing::HOSTNAME_LEVENSHTEIN_DISTANCE: + return true; + case LoadBalancing::IN_ORDER: + return false; + case LoadBalancing::RANDOM: + return false; + case LoadBalancing::FIRST_OR_RANDOM: + return true; + case LoadBalancing::ROUND_ROBIN: + return false; + } +} + } diff --git a/src/Common/GetPriorityForLoadBalancing.h b/src/Common/GetPriorityForLoadBalancing.h index 0de99730977..d4c68803866 100644 --- a/src/Common/GetPriorityForLoadBalancing.h +++ b/src/Common/GetPriorityForLoadBalancing.h @@ -1,6 +1,10 @@ #pragma once -#include +#include +#include + +#include +#include namespace DB { @@ -30,6 +34,8 @@ public: Func getPriorityFunc(LoadBalancing load_balance, size_t offset, size_t pool_size) const; + bool hasOptimalNode() const; + std::vector hostname_prefix_distance; /// Prefix distances from name of this host to the names of hosts of pools. std::vector hostname_levenshtein_distance; /// Levenshtein Distances from name of this host to the names of hosts of pools. diff --git a/src/Common/IFactoryWithAliases.h b/src/Common/IFactoryWithAliases.h index 74d4b6e3bcb..431e5c7b733 100644 --- a/src/Common/IFactoryWithAliases.h +++ b/src/Common/IFactoryWithAliases.h @@ -39,16 +39,16 @@ protected: public: /// For compatibility with SQL, it's possible to specify that certain function name is case insensitive. - enum CaseSensitiveness + enum Case { - CaseSensitive, - CaseInsensitive + Sensitive, + Insensitive }; /** Register additional name for value * real_name have to be already registered. */ - void registerAlias(const String & alias_name, const String & real_name, CaseSensitiveness case_sensitiveness = CaseSensitive) + void registerAlias(const String & alias_name, const String & real_name, Case case_sensitiveness = Sensitive) { const auto & creator_map = getMap(); const auto & case_insensitive_creator_map = getCaseInsensitiveMap(); @@ -66,12 +66,12 @@ public: } /// We need sure the real_name exactly exists when call the function directly. - void registerAliasUnchecked(const String & alias_name, const String & real_name, CaseSensitiveness case_sensitiveness = CaseSensitive) + void registerAliasUnchecked(const String & alias_name, const String & real_name, Case case_sensitiveness = Sensitive) { String alias_name_lowercase = Poco::toLower(alias_name); const String factory_name = getFactoryName(); - if (case_sensitiveness == CaseInsensitive) + if (case_sensitiveness == Insensitive) { if (!case_insensitive_aliases.emplace(alias_name_lowercase, real_name).second) throw Exception(ErrorCodes::LOGICAL_ERROR, "{}: case insensitive alias name '{}' is not unique", factory_name, alias_name); diff --git a/src/Common/IntervalKind.h b/src/Common/IntervalKind.h index f8e1fe87276..d66ea6018d4 100644 --- a/src/Common/IntervalKind.h +++ b/src/Common/IntervalKind.h @@ -7,19 +7,20 @@ namespace DB /// Kind of a temporal interval. struct IntervalKind { + /// note: The order and numbers are important and used in binary encoding, append new interval kinds to the end of list. enum class Kind : uint8_t { - Nanosecond, - Microsecond, - Millisecond, - Second, - Minute, - Hour, - Day, - Week, - Month, - Quarter, - Year, + Nanosecond = 0x00, + Microsecond = 0x01, + Millisecond = 0x02, + Second = 0x03, + Minute = 0x04, + Hour = 0x05, + Day = 0x06, + Week = 0x07, + Month = 0x08, + Quarter = 0x09, + Year = 0x0A, }; Kind kind = Kind::Second; diff --git a/src/Common/JSONParsers/SimdJSONParser.h b/src/Common/JSONParsers/SimdJSONParser.h index 827d142266a..db679b14f52 100644 --- a/src/Common/JSONParsers/SimdJSONParser.h +++ b/src/Common/JSONParsers/SimdJSONParser.h @@ -14,6 +14,7 @@ namespace DB { + namespace ErrorCodes { extern const int CANNOT_ALLOCATE_MEMORY; diff --git a/src/Common/Jemalloc.cpp b/src/Common/Jemalloc.cpp index fbe2f62c944..d7cc246db6a 100644 --- a/src/Common/Jemalloc.cpp +++ b/src/Common/Jemalloc.cpp @@ -46,6 +46,20 @@ void checkJemallocProfilingEnabled() "set: MALLOC_CONF=background_thread:true,prof:true"); } +template +void setJemallocValue(const char * name, T value) +{ + T old_value; + size_t old_value_size = sizeof(T); + if (mallctl(name, &old_value, &old_value_size, reinterpret_cast(&value), sizeof(T))) + { + LOG_WARNING(getLogger("Jemalloc"), "mallctl for {} failed", name); + return; + } + + LOG_INFO(getLogger("Jemalloc"), "Value for {} set to {} (from {})", name, value, old_value); +} + void setJemallocProfileActive(bool value) { checkJemallocProfilingEnabled(); @@ -58,7 +72,7 @@ void setJemallocProfileActive(bool value) return; } - mallctl("prof.active", nullptr, nullptr, &value, sizeof(bool)); + setJemallocValue("prof.active", value); LOG_TRACE(getLogger("SystemJemalloc"), "Profiling is {}", value ? "enabled" : "disabled"); } @@ -84,6 +98,16 @@ std::string flushJemallocProfile(const std::string & file_prefix) return profile_dump_path; } +void setJemallocBackgroundThreads(bool enabled) +{ + setJemallocValue("background_thread", enabled); +} + +void setJemallocMaxBackgroundThreads(size_t max_threads) +{ + setJemallocValue("max_background_threads", max_threads); +} + } #endif diff --git a/src/Common/Jemalloc.h b/src/Common/Jemalloc.h index 80ff0f1a319..499a906fd3d 100644 --- a/src/Common/Jemalloc.h +++ b/src/Common/Jemalloc.h @@ -17,6 +17,10 @@ void setJemallocProfileActive(bool value); std::string flushJemallocProfile(const std::string & file_prefix); +void setJemallocBackgroundThreads(bool enabled); + +void setJemallocMaxBackgroundThreads(size_t max_threads); + } #endif diff --git a/src/Common/Logger.cpp b/src/Common/Logger.cpp index c8d557bc3a3..bd848abe353 100644 --- a/src/Common/Logger.cpp +++ b/src/Common/Logger.cpp @@ -25,3 +25,15 @@ bool hasLogger(const std::string & name) { return Poco::Logger::has(name); } + +static constinit std::atomic allow_logging{true}; + +bool isLoggingEnabled() +{ + return allow_logging; +} + +void disableLogging() +{ + allow_logging = false; +} diff --git a/src/Common/Logger.h b/src/Common/Logger.h index b54ccd33e72..7471e3dff9b 100644 --- a/src/Common/Logger.h +++ b/src/Common/Logger.h @@ -64,3 +64,7 @@ LoggerRawPtr createRawLogger(const std::string & name, Poco::Channel * channel, * Otherwise, returns false. */ bool hasLogger(const std::string & name); + +void disableLogging(); + +bool isLoggingEnabled(); diff --git a/src/Common/MemoryTracker.cpp b/src/Common/MemoryTracker.cpp index 28cfa98666a..ac412684cf7 100644 --- a/src/Common/MemoryTracker.cpp +++ b/src/Common/MemoryTracker.cpp @@ -192,7 +192,7 @@ void MemoryTracker::debugLogBigAllocationWithoutCheck(Int64 size [[maybe_unused] { /// Big allocations through allocNoThrow (without checking memory limits) may easily lead to OOM (and it's hard to debug). /// Let's find them. -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD if (size < 0) return; diff --git a/src/Common/NamedCollections/NamedCollectionsFactory.cpp b/src/Common/NamedCollections/NamedCollectionsFactory.cpp index 14105a8651d..bc283d2be1a 100644 --- a/src/Common/NamedCollections/NamedCollectionsFactory.cpp +++ b/src/Common/NamedCollections/NamedCollectionsFactory.cpp @@ -1,6 +1,7 @@ #include #include #include +#include #include namespace DB @@ -235,7 +236,7 @@ bool NamedCollectionFactory::loadIfNot(std::lock_guard & lock) loadFromConfig(context->getConfigRef(), lock); loadFromSQL(lock); - if (metadata_storage->supportsPeriodicUpdate()) + if (metadata_storage->isReplicated()) { update_task = context->getSchedulePool().createTask("NamedCollectionsMetadataStorage", [this]{ updateFunc(); }); update_task->activate(); @@ -357,6 +358,13 @@ void NamedCollectionFactory::reloadFromSQL() add(std::move(collections), lock); } +bool NamedCollectionFactory::usesReplicatedStorage() +{ + std::lock_guard lock(mutex); + loadIfNot(lock); + return metadata_storage->isReplicated(); +} + void NamedCollectionFactory::updateFunc() { LOG_TRACE(log, "Named collections background updating thread started"); diff --git a/src/Common/NamedCollections/NamedCollectionsFactory.h b/src/Common/NamedCollections/NamedCollectionsFactory.h index 6ee5940e686..a0721ad8a50 100644 --- a/src/Common/NamedCollections/NamedCollectionsFactory.h +++ b/src/Common/NamedCollections/NamedCollectionsFactory.h @@ -34,6 +34,8 @@ public: void updateFromSQL(const ASTAlterNamedCollectionQuery & query); + bool usesReplicatedStorage(); + void loadIfNot(); void shutdown(); diff --git a/src/Common/NamedCollections/NamedCollectionsMetadataStorage.cpp b/src/Common/NamedCollections/NamedCollectionsMetadataStorage.cpp index 32fdb25abd3..36191b89e86 100644 --- a/src/Common/NamedCollections/NamedCollectionsMetadataStorage.cpp +++ b/src/Common/NamedCollections/NamedCollectionsMetadataStorage.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -67,7 +68,7 @@ public: virtual bool removeIfExists(const std::string & path) = 0; - virtual bool supportsPeriodicUpdate() const = 0; + virtual bool isReplicated() const = 0; virtual bool waitUpdate(size_t /* timeout */) { return false; } }; @@ -89,7 +90,7 @@ public: ~LocalStorage() override = default; - bool supportsPeriodicUpdate() const override { return false; } + bool isReplicated() const override { return false; } std::vector list() const override { @@ -115,32 +116,32 @@ public: return elements; } - bool exists(const std::string & path) const override + bool exists(const std::string & file_name) const override { - return fs::exists(getPath(path)); + return fs::exists(getPath(file_name)); } - std::string read(const std::string & path) const override + std::string read(const std::string & file_name) const override { - ReadBufferFromFile in(getPath(path)); + ReadBufferFromFile in(getPath(file_name)); std::string data; readStringUntilEOF(data, in); return data; } - void write(const std::string & path, const std::string & data, bool replace) override + void write(const std::string & file_name, const std::string & data, bool replace) override { - if (!replace && fs::exists(path)) + if (!replace && fs::exists(file_name)) { throw Exception( ErrorCodes::NAMED_COLLECTION_ALREADY_EXISTS, "Metadata file {} for named collection already exists", - path); + file_name); } fs::create_directories(root_path); - auto tmp_path = getPath(path + ".tmp"); + auto tmp_path = getPath(file_name + ".tmp"); WriteBufferFromFile out(tmp_path, data.size(), O_WRONLY | O_CREAT | O_EXCL); writeString(data, out); @@ -149,28 +150,32 @@ public: out.sync(); out.close(); - fs::rename(tmp_path, getPath(path)); + fs::rename(tmp_path, getPath(file_name)); } - void remove(const std::string & path) override + void remove(const std::string & file_name) override { - if (!removeIfExists(getPath(path))) + if (!removeIfExists(file_name)) { throw Exception( ErrorCodes::NAMED_COLLECTION_DOESNT_EXIST, - "Cannot remove `{}`, because it doesn't exist", path); + "Cannot remove `{}`, because it doesn't exist", file_name); } } - bool removeIfExists(const std::string & path) override + bool removeIfExists(const std::string & file_name) override { - return fs::remove(getPath(path)); + return fs::remove(getPath(file_name)); } private: - std::string getPath(const std::string & path) const + std::string getPath(const std::string & file_name) const { - return fs::path(root_path) / path; + const auto file_name_as_path = fs::path(file_name); + if (file_name_as_path.is_absolute()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Filename {} cannot be an absolute path", file_name); + + return fs::path(root_path) / file_name_as_path; } /// Delete .tmp files. They could be left undeleted in case of @@ -221,7 +226,7 @@ public: ~ZooKeeperStorage() override = default; - bool supportsPeriodicUpdate() const override { return true; } + bool isReplicated() const override { return true; } /// Return true if children changed. bool waitUpdate(size_t timeout) override @@ -263,49 +268,49 @@ public: return children; } - bool exists(const std::string & path) const override + bool exists(const std::string & file_name) const override { - return getClient()->exists(getPath(path)); + return getClient()->exists(getPath(file_name)); } - std::string read(const std::string & path) const override + std::string read(const std::string & file_name) const override { - return getClient()->get(getPath(path)); + return getClient()->get(getPath(file_name)); } - void write(const std::string & path, const std::string & data, bool replace) override + void write(const std::string & file_name, const std::string & data, bool replace) override { if (replace) { - getClient()->createOrUpdate(getPath(path), data, zkutil::CreateMode::Persistent); + getClient()->createOrUpdate(getPath(file_name), data, zkutil::CreateMode::Persistent); } else { - auto code = getClient()->tryCreate(getPath(path), data, zkutil::CreateMode::Persistent); + auto code = getClient()->tryCreate(getPath(file_name), data, zkutil::CreateMode::Persistent); if (code == Coordination::Error::ZNODEEXISTS) { throw Exception( ErrorCodes::NAMED_COLLECTION_ALREADY_EXISTS, "Metadata file {} for named collection already exists", - path); + file_name); } } } - void remove(const std::string & path) override + void remove(const std::string & file_name) override { - getClient()->remove(getPath(path)); + getClient()->remove(getPath(file_name)); } - bool removeIfExists(const std::string & path) override + bool removeIfExists(const std::string & file_name) override { - auto code = getClient()->tryRemove(getPath(path)); + auto code = getClient()->tryRemove(getPath(file_name)); if (code == Coordination::Error::ZOK) return true; if (code == Coordination::Error::ZNONODE) return false; - throw Coordination::Exception::fromPath(code, getPath(path)); + throw Coordination::Exception::fromPath(code, getPath(file_name)); } private: @@ -319,9 +324,13 @@ private: return zookeeper_client; } - std::string getPath(const std::string & path) const + std::string getPath(const std::string & file_name) const { - return fs::path(root_path) / path; + const auto file_name_as_path = fs::path(file_name); + if (file_name_as_path.is_absolute()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Filename {} cannot be an absolute path", file_name); + + return fs::path(root_path) / file_name_as_path; } }; @@ -465,14 +474,14 @@ void NamedCollectionsMetadataStorage::writeCreateQuery(const ASTCreateNamedColle storage->write(getFileName(query.collection_name), serializeAST(*normalized_query), replace); } -bool NamedCollectionsMetadataStorage::supportsPeriodicUpdate() const +bool NamedCollectionsMetadataStorage::isReplicated() const { - return storage->supportsPeriodicUpdate(); + return storage->isReplicated(); } bool NamedCollectionsMetadataStorage::waitUpdate() { - if (!storage->supportsPeriodicUpdate()) + if (!storage->isReplicated()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Periodic updates are not supported"); const auto & config = Context::getGlobalContextInstance()->getConfigRef(); diff --git a/src/Common/NamedCollections/NamedCollectionsMetadataStorage.h b/src/Common/NamedCollections/NamedCollectionsMetadataStorage.h index 3c089fe2fa2..c3468fbc468 100644 --- a/src/Common/NamedCollections/NamedCollectionsMetadataStorage.h +++ b/src/Common/NamedCollections/NamedCollectionsMetadataStorage.h @@ -30,7 +30,7 @@ public: /// Return true if update was made bool waitUpdate(); - bool supportsPeriodicUpdate() const; + bool isReplicated() const; private: class INamedCollectionsStorage; diff --git a/src/Common/PageCache.cpp b/src/Common/PageCache.cpp index 56bd8c1a339..d719a387e14 100644 --- a/src/Common/PageCache.cpp +++ b/src/Common/PageCache.cpp @@ -424,7 +424,7 @@ static void logUnexpectedSyscallError(std::string name) { std::string message = fmt::format("{} failed: {}", name, errnoToString()); LOG_WARNING(&Poco::Logger::get("PageCache"), "{}", message); -#if defined(ABORT_ON_LOGICAL_ERROR) +#if defined(DEBUG_OR_SANITIZER_BUILD) volatile bool true_ = true; if (true_) // suppress warning about missing [[noreturn]] abortOnFailedAssertion(message); diff --git a/src/Common/PipeFDs.cpp b/src/Common/PipeFDs.cpp index ceadbb2f983..50eeda1bbe2 100644 --- a/src/Common/PipeFDs.cpp +++ b/src/Common/PipeFDs.cpp @@ -1,19 +1,23 @@ #include #include #include +#include #include #include #include #include -#include #include - namespace DB { +namespace FailPoints +{ + extern const char lazy_pipe_fds_fail_close[]; +} + namespace ErrorCodes { extern const int CANNOT_PIPE; @@ -42,6 +46,11 @@ void LazyPipeFDs::open() void LazyPipeFDs::close() { + fiu_do_on(FailPoints::lazy_pipe_fds_fail_close, + { + throw Exception(ErrorCodes::CANNOT_PIPE, "Manually triggered exception on close"); + }); + for (int & fd : fds_rw) { if (fd < 0) diff --git a/src/Common/PipeFDs.h b/src/Common/PipeFDs.h index 20bd847c077..b651176ee26 100644 --- a/src/Common/PipeFDs.h +++ b/src/Common/PipeFDs.h @@ -1,8 +1,5 @@ #pragma once -#include - - namespace DB { diff --git a/src/Common/PoolWithFailoverBase.h b/src/Common/PoolWithFailoverBase.h index 2359137012c..3d4de773a36 100644 --- a/src/Common/PoolWithFailoverBase.h +++ b/src/Common/PoolWithFailoverBase.h @@ -28,7 +28,6 @@ namespace ErrorCodes namespace ProfileEvents { - extern const Event DistributedConnectionFailTry; extern const Event DistributedConnectionFailAtAll; extern const Event DistributedConnectionSkipReadOnlyReplica; } @@ -285,7 +284,6 @@ PoolWithFailoverBase::getMany( else { LOG_WARNING(log, "Connection failed at try №{}, reason: {}", (shuffled_pool.error_count + 1), fail_message); - ProfileEvents::increment(ProfileEvents::DistributedConnectionFailTry); shuffled_pool.error_count = std::min(max_error_cap, shuffled_pool.error_count + 1); diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp index 8f181d55042..2b090136e2e 100644 --- a/src/Common/ProfileEvents.cpp +++ b/src/Common/ProfileEvents.cpp @@ -238,7 +238,12 @@ \ M(CannotRemoveEphemeralNode, "Number of times an error happened while trying to remove ephemeral node. This is not an issue, because our implementation of ZooKeeper library guarantee that the session will expire and the node will be removed.") \ \ - M(RegexpCreated, "Compiled regular expressions. Identical regular expressions compiled just once and cached forever.") \ + M(RegexpWithMultipleNeedlesCreated, "Regular expressions with multiple needles (VectorScan library) compiled.") \ + M(RegexpWithMultipleNeedlesGlobalCacheHit, "Number of times we fetched compiled regular expression with multiple needles (VectorScan library) from the global cache.") \ + M(RegexpWithMultipleNeedlesGlobalCacheMiss, "Number of times we failed to fetch compiled regular expression with multiple needles (VectorScan library) from the global cache.") \ + M(RegexpLocalCacheHit, "Number of times we fetched compiled regular expression from a local cache.") \ + M(RegexpLocalCacheMiss, "Number of times we failed to fetch compiled regular expression from a local cache.") \ + \ M(ContextLock, "Number of times the lock of Context was acquired or tried to acquire. This is global lock.") \ M(ContextLockWaitMicroseconds, "Context lock wait time in microseconds") \ \ @@ -437,8 +442,6 @@ The server successfully detected this situation and will download merged part fr M(ReadBufferFromS3InitMicroseconds, "Time spent initializing connection to S3.") \ M(ReadBufferFromS3Bytes, "Bytes read from S3.") \ M(ReadBufferFromS3RequestsErrors, "Number of exceptions while reading from S3.") \ - M(ReadBufferFromS3ResetSessions, "Number of HTTP sessions that were reset in ReadBufferFromS3.") \ - M(ReadBufferFromS3PreservedSessions, "Number of HTTP sessions that were preserved in ReadBufferFromS3.") \ \ M(WriteBufferFromS3Microseconds, "Time spent on writing to S3.") \ M(WriteBufferFromS3Bytes, "Bytes written to S3.") \ @@ -447,14 +450,18 @@ The server successfully detected this situation and will download merged part fr M(QueryMemoryLimitExceeded, "Number of times when memory limit exceeded for query.") \ \ M(AzureGetObject, "Number of Azure API GetObject calls.") \ - M(AzureUploadPart, "Number of Azure blob storage API UploadPart calls") \ + M(AzureUpload, "Number of Azure blob storage API Upload calls") \ + M(AzureStageBlock, "Number of Azure blob storage API StageBlock calls") \ + M(AzureCommitBlockList, "Number of Azure blob storage API CommitBlockList calls") \ M(AzureCopyObject, "Number of Azure blob storage API CopyObject calls") \ M(AzureDeleteObjects, "Number of Azure blob storage API DeleteObject(s) calls.") \ M(AzureListObjects, "Number of Azure blob storage API ListObjects calls.") \ M(AzureGetProperties, "Number of Azure blob storage API GetProperties calls.") \ \ M(DiskAzureGetObject, "Number of Disk Azure API GetObject calls.") \ - M(DiskAzureUploadPart, "Number of Disk Azure blob storage API UploadPart calls") \ + M(DiskAzureUpload, "Number of Disk Azure blob storage API Upload calls") \ + M(DiskAzureStageBlock, "Number of Disk Azure blob storage API StageBlock calls") \ + M(DiskAzureCommitBlockList, "Number of Disk Azure blob storage API CommitBlockList calls") \ M(DiskAzureCopyObject, "Number of Disk Azure blob storage API CopyObject calls") \ M(DiskAzureListObjects, "Number of Disk Azure blob storage API ListObjects calls.") \ M(DiskAzureDeleteObjects, "Number of Azure blob storage API DeleteObject(s) calls.") \ @@ -499,6 +506,7 @@ The server successfully detected this situation and will download merged part fr M(FileSegmentHolderCompleteMicroseconds, "File segments holder complete() time") \ M(FileSegmentFailToIncreasePriority, "Number of times the priority was not increased due to a high contention on the cache lock") \ M(FilesystemCacheFailToReserveSpaceBecauseOfLockContention, "Number of times space reservation was skipped due to a high contention on the cache lock") \ + M(FilesystemCacheFailToReserveSpaceBecauseOfCacheResize, "Number of times space reservation was skipped due to the cache is being resized") \ M(FilesystemCacheHoldFileSegments, "Filesystem cache file segments count, which were hold") \ M(FilesystemCacheUnusedHoldFileSegments, "Filesystem cache file segments count, which were hold, but not used (because of seek or LIMIT n, etc)") \ M(FilesystemCacheFreeSpaceKeepingThreadRun, "Number of times background thread executed free space keeping job") \ @@ -559,6 +567,7 @@ The server successfully detected this situation and will download merged part fr M(AggregationPreallocatedElementsInHashTables, "How many elements were preallocated in hash tables for aggregation.") \ M(AggregationHashTablesInitializedAsTwoLevel, "How many hash tables were inited as two-level for aggregation.") \ M(AggregationOptimizedEqualRangesOfKeys, "For how many blocks optimization of equal ranges of keys was applied") \ + M(HashJoinPreallocatedElementsInHashTables, "How many elements were preallocated in hash tables for hash join.") \ \ M(MetadataFromKeeperCacheHit, "Number of times an object storage metadata request was answered from cache without making request to Keeper") \ M(MetadataFromKeeperCacheMiss, "Number of times an object storage metadata request had to be answered from Keeper") \ @@ -611,6 +620,13 @@ The server successfully detected this situation and will download merged part fr M(KeeperPacketsReceived, "Packets received by keeper server") \ M(KeeperRequestTotal, "Total requests number on keeper server") \ M(KeeperLatency, "Keeper latency") \ + M(KeeperTotalElapsedMicroseconds, "Keeper total latency for a single request") \ + M(KeeperProcessElapsedMicroseconds, "Keeper commit latency for a single request") \ + M(KeeperPreprocessElapsedMicroseconds, "Keeper preprocessing latency for a single reuquest") \ + M(KeeperStorageLockWaitMicroseconds, "Time spent waiting for acquiring Keeper storage lock") \ + M(KeeperCommitWaitElapsedMicroseconds, "Time spent waiting for certain log to be committed") \ + M(KeeperBatchMaxCount, "Number of times the size of batch was limited by the amount") \ + M(KeeperBatchMaxTotalSize, "Number of times the size of batch was limited by the total bytes size") \ M(KeeperCommits, "Number of successful commits") \ M(KeeperCommitsFailed, "Number of failed commits") \ M(KeeperSnapshotCreations, "Number of snapshots creations")\ @@ -637,11 +653,11 @@ The server successfully detected this situation and will download merged part fr M(S3QueueSetFileProcessingMicroseconds, "Time spent to set file as processing")\ M(S3QueueSetFileProcessedMicroseconds, "Time spent to set file as processed")\ M(S3QueueSetFileFailedMicroseconds, "Time spent to set file as failed")\ - M(S3QueueFailedFiles, "Number of files which failed to be processed")\ - M(S3QueueProcessedFiles, "Number of files which were processed")\ - M(S3QueueCleanupMaxSetSizeOrTTLMicroseconds, "Time spent to set file as failed")\ - M(S3QueuePullMicroseconds, "Time spent to read file data")\ - M(S3QueueLockLocalFileStatusesMicroseconds, "Time spent to lock local file statuses")\ + M(ObjectStorageQueueFailedFiles, "Number of files which failed to be processed")\ + M(ObjectStorageQueueProcessedFiles, "Number of files which were processed")\ + M(ObjectStorageQueueCleanupMaxSetSizeOrTTLMicroseconds, "Time spent to set file as failed")\ + M(ObjectStorageQueuePullMicroseconds, "Time spent to read file data")\ + M(ObjectStorageQueueLockLocalFileStatusesMicroseconds, "Time spent to lock local file statuses")\ \ M(ServerStartupMilliseconds, "Time elapsed from starting server to listening to sockets in milliseconds")\ M(IOUringSQEsSubmitted, "Total number of io_uring SQEs submitted") \ diff --git a/src/Common/ProgressIndication.cpp b/src/Common/ProgressIndication.cpp index 7b07c72824a..8f0fb3cac6c 100644 --- a/src/Common/ProgressIndication.cpp +++ b/src/Common/ProgressIndication.cpp @@ -1,6 +1,7 @@ #include "ProgressIndication.h" #include #include +#include #include #include #include @@ -92,19 +93,19 @@ void ProgressIndication::writeFinalProgress() if (progress.read_rows < 1000) return; - std::cout << "Processed " << formatReadableQuantity(progress.read_rows) << " rows, " + output_stream << "Processed " << formatReadableQuantity(progress.read_rows) << " rows, " << formatReadableSizeWithDecimalSuffix(progress.read_bytes); UInt64 elapsed_ns = getElapsedNanoseconds(); if (elapsed_ns) - std::cout << " (" << formatReadableQuantity(progress.read_rows * 1000000000.0 / elapsed_ns) << " rows/s., " + output_stream << " (" << formatReadableQuantity(progress.read_rows * 1000000000.0 / elapsed_ns) << " rows/s., " << formatReadableSizeWithDecimalSuffix(progress.read_bytes * 1000000000.0 / elapsed_ns) << "/s.)"; else - std::cout << ". "; + output_stream << ". "; auto peak_memory_usage = getMemoryUsage().peak; if (peak_memory_usage >= 0) - std::cout << "\nPeak memory usage: " << formatReadableSizeWithBinarySuffix(peak_memory_usage) << "."; + output_stream << "\nPeak memory usage: " << formatReadableSizeWithBinarySuffix(peak_memory_usage) << "."; } void ProgressIndication::writeProgress(WriteBufferFromFileDescriptor & message) @@ -125,7 +126,7 @@ void ProgressIndication::writeProgress(WriteBufferFromFileDescriptor & message) const char * indicator = indicators[increment % 8]; - size_t terminal_width = getTerminalWidth(); + size_t terminal_width = getTerminalWidth(in_fd, err_fd); if (!written_progress_chars) { diff --git a/src/Common/ProgressIndication.h b/src/Common/ProgressIndication.h index a9965785889..9baa1c7ffc4 100644 --- a/src/Common/ProgressIndication.h +++ b/src/Common/ProgressIndication.h @@ -1,14 +1,15 @@ #pragma once -#include -#include -#include #include #include #include #include #include +#include +#include +#include +#include namespace DB { @@ -32,6 +33,19 @@ using HostToTimesMap = std::unordered_map; class ProgressIndication { public: + + explicit ProgressIndication + ( + std::ostream & output_stream_ = std::cout, + int in_fd_ = STDIN_FILENO, + int err_fd_ = STDERR_FILENO + ) + : output_stream(output_stream_), + in_fd(in_fd_), + err_fd(err_fd_) + { + } + /// Write progress bar. void writeProgress(WriteBufferFromFileDescriptor & message); void clearProgressOutput(WriteBufferFromFileDescriptor & message); @@ -58,11 +72,6 @@ public: /// How much seconds passed since query execution start. double elapsedSeconds() const { return getElapsedNanoseconds() / 1e9; } - void updateThreadEventData(HostToTimesMap & new_hosts_data); - -private: - double getCPUUsage(); - struct MemoryUsage { UInt64 total = 0; @@ -72,6 +81,11 @@ private: MemoryUsage getMemoryUsage() const; + void updateThreadEventData(HostToTimesMap & new_hosts_data); + +private: + double getCPUUsage(); + UInt64 getElapsedNanoseconds() const; /// This flag controls whether to show the progress bar. We start showing it after @@ -103,6 +117,10 @@ private: /// - hosts_data/cpu_usage_meter (guarded with profile_events_mutex) mutable std::mutex profile_events_mutex; mutable std::mutex progress_mutex; + + std::ostream & output_stream; + int in_fd; + int err_fd; }; } diff --git a/src/Common/RemoteHostFilter.h b/src/Common/RemoteHostFilter.h index 2b91306f405..4c8983205fa 100644 --- a/src/Common/RemoteHostFilter.h +++ b/src/Common/RemoteHostFilter.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include #include diff --git a/src/Common/SignalHandlers.cpp b/src/Common/SignalHandlers.cpp new file mode 100644 index 00000000000..e025e49e0a3 --- /dev/null +++ b/src/Common/SignalHandlers.cpp @@ -0,0 +1,642 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#pragma clang diagnostic ignored "-Wreserved-identifier" + +namespace DB +{ +namespace ErrorCodes +{ +extern const int CANNOT_SET_SIGNAL_HANDLER; +extern const int CANNOT_SEND_SIGNAL; +} +} + +using namespace DB; + + +void call_default_signal_handler(int sig) +{ + if (SIG_ERR == signal(sig, SIG_DFL)) + throw ErrnoException(ErrorCodes::CANNOT_SET_SIGNAL_HANDLER, "Cannot set signal handler"); + + if (0 != raise(sig)) + throw ErrnoException(ErrorCodes::CANNOT_SEND_SIGNAL, "Cannot send signal"); +} + + +void writeSignalIDtoSignalPipe(int sig) +{ + auto saved_errno = errno; /// We must restore previous value of errno in signal handler. + + char buf[signal_pipe_buf_size]; + auto & signal_pipe = HandledSignals::instance().signal_pipe; + WriteBufferFromFileDescriptor out(signal_pipe.fds_rw[1], signal_pipe_buf_size, buf); + writeBinary(sig, out); + out.next(); + + errno = saved_errno; +} + +void closeLogsSignalHandler(int sig, siginfo_t *, void *) +{ + DENY_ALLOCATIONS_IN_SCOPE; + writeSignalIDtoSignalPipe(sig); +} + +void terminateRequestedSignalHandler(int sig, siginfo_t *, void *) +{ + DENY_ALLOCATIONS_IN_SCOPE; + writeSignalIDtoSignalPipe(sig); +} + + +void signalHandler(int sig, siginfo_t * info, void * context) +{ + if (asynchronous_stack_unwinding && sig == SIGSEGV) + siglongjmp(asynchronous_stack_unwinding_signal_jump_buffer, 1); + + DENY_ALLOCATIONS_IN_SCOPE; + auto saved_errno = errno; /// We must restore previous value of errno in signal handler. + + char buf[signal_pipe_buf_size]; + auto & signal_pipe = HandledSignals::instance().signal_pipe; + WriteBufferFromFileDescriptorDiscardOnFailure out(signal_pipe.fds_rw[1], signal_pipe_buf_size, buf); + + const ucontext_t * signal_context = reinterpret_cast(context); + const StackTrace stack_trace(*signal_context); + +#if USE_GWP_ASAN + if (const auto fault_address = reinterpret_cast(info->si_addr); + GWPAsan::isGWPAsanError(fault_address)) + GWPAsan::printReport(fault_address); +#endif + + writeBinary(sig, out); + writePODBinary(*info, out); + writePODBinary(signal_context, out); + writePODBinary(stack_trace, out); + writeVectorBinary(Exception::enable_job_stack_trace ? Exception::getThreadFramePointers() : std::vector{}, out); + writeBinary(static_cast(getThreadId()), out); + writePODBinary(current_thread, out); + + out.next(); + + if (sig != SIGTSTP) /// This signal is used for debugging. + { + /// The time that is usually enough for separate thread to print info into log. + /// Under MSan full stack unwinding with DWARF info about inline functions takes 101 seconds in one case. + for (size_t i = 0; i < 300; ++i) + { + /// We will synchronize with the thread printing the messages with an atomic variable to finish earlier. + if (HandledSignals::instance().fatal_error_printed.test()) + break; + + /// This coarse method of synchronization is perfectly ok for fatal signals. + sleepForSeconds(1); + } + + /// Wait for all logs flush operations + sleepForSeconds(3); + call_default_signal_handler(sig); + } + + errno = saved_errno; +} + + +[[noreturn]] void terminate_handler() +{ + static thread_local bool terminating = false; + if (terminating) + abort(); + + terminating = true; + + std::string log_message; + + if (std::current_exception()) + log_message = "Terminate called for uncaught exception:\n" + getCurrentExceptionMessage(true); + else + log_message = "Terminate called without an active exception"; + + /// POSIX.1 says that write(2)s of less than PIPE_BUF bytes must be atomic - man 7 pipe + /// And the buffer should not be too small because our exception messages can be large. + static constexpr size_t buf_size = PIPE_BUF; + + if (log_message.size() > buf_size - 16) + log_message.resize(buf_size - 16); + + char buf[buf_size]; + auto & signal_pipe = HandledSignals::instance().signal_pipe; + WriteBufferFromFileDescriptor out(signal_pipe.fds_rw[1], buf_size, buf); + + writeBinary(static_cast(SignalListener::StdTerminate), out); + writeBinary(static_cast(getThreadId()), out); + writeBinary(log_message, out); + out.next(); + + abort(); +} + +#if defined(SANITIZER) +template +struct ValueHolder +{ + ValueHolder(T value_) : value(value_) + {} + + T value; +}; + +extern "C" void __sanitizer_set_death_callback(void (*)()); + +/// Sanitizers may not expect some function calls from death callback. +/// Let's try to disable instrumentation to avoid possible issues. +/// However, this callback may call other functions that are still instrumented. +/// We can try [[clang::always_inline]] attribute for statements in future (available in clang-15) +/// See https://github.com/google/sanitizers/issues/1543 and https://github.com/google/sanitizers/issues/1549. +static DISABLE_SANITIZER_INSTRUMENTATION void sanitizerDeathCallback() +{ + DENY_ALLOCATIONS_IN_SCOPE; + /// Also need to send data via pipe. Otherwise it may lead to deadlocks or failures in printing diagnostic info. + + char buf[signal_pipe_buf_size]; + auto & signal_pipe = HandledSignals::instance().signal_pipe; + WriteBufferFromFileDescriptorDiscardOnFailure out(signal_pipe.fds_rw[1], signal_pipe_buf_size, buf); + + const StackTrace stack_trace; + + writeBinary(SignalListener::SanitizerTrap, out); + writePODBinary(stack_trace, out); + /// We create a dummy struct with a constructor so DISABLE_SANITIZER_INSTRUMENTATION is not applied to it + /// otherwise, Memory sanitizer can't know that values initiialized inside this function are actually initialized + /// because instrumentations are disabled leading to false positives later on + ValueHolder thread_id{static_cast(getThreadId())}; + writeBinary(thread_id.value, out); + writePODBinary(current_thread, out); + + out.next(); + + /// The time that is usually enough for separate thread to print info into log. + sleepForSeconds(20); +} +#endif + + +void HandledSignals::addSignalHandler(const std::vector & signals, signal_function handler, bool register_signal) +{ + struct sigaction sa; + memset(&sa, 0, sizeof(sa)); + sa.sa_sigaction = handler; + sa.sa_flags = SA_SIGINFO; + +#if defined(OS_DARWIN) + sigemptyset(&sa.sa_mask); + for (auto signal : signals) + sigaddset(&sa.sa_mask, signal); +#else + if (sigemptyset(&sa.sa_mask)) + throw Poco::Exception("Cannot set signal handler."); + + for (auto signal : signals) + if (sigaddset(&sa.sa_mask, signal)) + throw Poco::Exception("Cannot set signal handler."); +#endif + + for (auto signal : signals) + if (sigaction(signal, &sa, nullptr)) + throw Poco::Exception("Cannot set signal handler."); + + if (register_signal) + std::copy(signals.begin(), signals.end(), std::back_inserter(handled_signals)); +} + +void blockSignals(const std::vector & signals) +{ + sigset_t sig_set; + +#if defined(OS_DARWIN) + sigemptyset(&sig_set); + for (auto signal : signals) + sigaddset(&sig_set, signal); +#else + if (sigemptyset(&sig_set)) + throw Poco::Exception("Cannot block signal."); + + for (auto signal : signals) + if (sigaddset(&sig_set, signal)) + throw Poco::Exception("Cannot block signal."); +#endif + + if (pthread_sigmask(SIG_BLOCK, &sig_set, nullptr)) + throw Poco::Exception("Cannot block signal."); +} + + +void SignalListener::run() +{ + static_assert(PIPE_BUF >= 512); + static_assert(signal_pipe_buf_size <= PIPE_BUF, "Only write of PIPE_BUF to pipe is atomic and the minimal known PIPE_BUF across supported platforms is 512"); + char buf[signal_pipe_buf_size]; + auto & signal_pipe = HandledSignals::instance().signal_pipe; + ReadBufferFromFileDescriptor in(signal_pipe.fds_rw[0], signal_pipe_buf_size, buf); + + while (!in.eof()) + { + int sig = 0; + readBinary(sig, in); + // We may log some specific signals afterwards, with different log + // levels and more info, but for completeness we log all signals + // here at trace level. + // Don't use strsignal here, because it's not thread-safe. + LOG_TRACE(log, "Received signal {}", sig); + + if (sig == StopThread) + { + LOG_INFO(log, "Stop SignalListener thread"); + break; + } + else if (sig == SIGHUP) + { + LOG_DEBUG(log, "Received signal to close logs."); + BaseDaemon::instance().closeLogs(BaseDaemon::instance().logger()); + LOG_INFO(log, "Opened new log file after received signal."); + } + else if (sig == StdTerminate) + { + UInt32 thread_num; + std::string message; + + readBinary(thread_num, in); + readBinary(message, in); + + onTerminate(message, thread_num); + } + else if (sig == SIGINT || + sig == SIGQUIT || + sig == SIGTERM) + { + if (daemon) + daemon->handleSignal(sig); + } + else + { + siginfo_t info{}; + ucontext_t * context{}; + StackTrace stack_trace(NoCapture{}); + std::vector thread_frame_pointers; + UInt32 thread_num{}; + ThreadStatus * thread_ptr{}; + + if (sig != SanitizerTrap) + { + readPODBinary(info, in); + readPODBinary(context, in); + } + + readPODBinary(stack_trace, in); + if (sig != SanitizerTrap) + readVectorBinary(thread_frame_pointers, in); + readBinary(thread_num, in); + readPODBinary(thread_ptr, in); + + /// This allows to receive more signals if failure happens inside onFault function. + /// Example: segfault while symbolizing stack trace. + try + { + std::thread([=, this] { onFault(sig, info, context, stack_trace, thread_frame_pointers, thread_num, thread_ptr); }).detach(); + } + catch (...) + { + /// Likely cannot allocate thread + onFault(sig, info, context, stack_trace, thread_frame_pointers, thread_num, thread_ptr); + } + } + } +} + +void SignalListener::onTerminate(std::string_view message, UInt32 thread_num) const +{ + size_t pos = message.find('\n'); + + LOG_FATAL(log, "(version {}{}, build id: {}, git hash: {}) (from thread {}) {}", + VERSION_STRING, VERSION_OFFICIAL, daemon ? daemon->build_id : "", daemon ? daemon->git_hash : "", thread_num, message.substr(0, pos)); + + /// Print trace from std::terminate exception line-by-line to make it easy for grep. + while (pos != std::string_view::npos) + { + ++pos; + size_t next_pos = message.find('\n', pos); + size_t size = next_pos; + if (next_pos != std::string_view::npos) + size = next_pos - pos; + + LOG_FATAL(log, fmt::runtime(message.substr(pos, size))); + pos = next_pos; + } +} + +void SignalListener::onFault( + int sig, + const siginfo_t & info, + ucontext_t * context, + const StackTrace & stack_trace, + const std::vector & thread_frame_pointers, + UInt32 thread_num, + DB::ThreadStatus * thread_ptr) const +try +{ + ThreadStatus thread_status; + + /// First log those fields that are safe to access and that should not cause new fault. + /// That way we will have some duplicated info in the log but we don't loose important info + /// in case of double fault. + + LOG_FATAL(log, "########## Short fault info ############"); + LOG_FATAL(log, "(version {}{}, build id: {}, git hash: {}) (from thread {}) Received signal {}", + VERSION_STRING, VERSION_OFFICIAL, daemon ? daemon->build_id : "", daemon ? daemon->git_hash : "", + thread_num, sig); + + std::string signal_description = "Unknown signal"; + + /// Some of these are not really signals, but our own indications on failure reason. + if (sig == StdTerminate) + signal_description = "std::terminate"; + else if (sig == SanitizerTrap) + signal_description = "sanitizer trap"; + else if (sig >= 0) + signal_description = strsignal(sig); // NOLINT(concurrency-mt-unsafe) // it is not thread-safe but ok in this context + + LOG_FATAL(log, "Signal description: {}", signal_description); + + String error_message; + + if (sig != SanitizerTrap) + error_message = signalToErrorMessage(sig, info, *context); + else + error_message = "Sanitizer trap."; + + LOG_FATAL(log, fmt::runtime(error_message)); + + String bare_stacktrace_str; + if (stack_trace.getSize()) + { + /// Write bare stack trace (addresses) just in case if we will fail to print symbolized stack trace. + /// NOTE: This still require memory allocations and mutex lock inside logger. + /// BTW we can also print it to stderr using write syscalls. + + WriteBufferFromOwnString bare_stacktrace; + writeString("Stack trace:", bare_stacktrace); + for (size_t i = stack_trace.getOffset(); i < stack_trace.getSize(); ++i) + { + writeChar(' ', bare_stacktrace); + writePointerHex(stack_trace.getFramePointers()[i], bare_stacktrace); + } + + LOG_FATAL(log, fmt::runtime(bare_stacktrace.str())); + bare_stacktrace_str = bare_stacktrace.str(); + } + + /// Now try to access potentially unsafe data in thread_ptr. + + String query_id; + String query; + + /// Send logs from this thread to client if possible. + /// It will allow client to see failure messages directly. + if (thread_ptr) + { + query_id = thread_ptr->getQueryId(); + query = thread_ptr->getQueryForLog(); + + if (auto logs_queue = thread_ptr->getInternalTextLogsQueue()) + { + CurrentThread::attachInternalTextLogsQueue(logs_queue, LogsLevel::trace); + } + } + + LOG_FATAL(log, "########################################"); + + if (query_id.empty()) + { + LOG_FATAL(log, "(version {}{}, build id: {}, git hash: {}) (from thread {}) (no query) Received signal {} ({})", + VERSION_STRING, VERSION_OFFICIAL, daemon ? daemon->build_id : "", daemon ? daemon->git_hash : "", + thread_num, signal_description, sig); + } + else + { + LOG_FATAL(log, "(version {}{}, build id: {}, git hash: {}) (from thread {}) (query_id: {}) (query: {}) Received signal {} ({})", + VERSION_STRING, VERSION_OFFICIAL, daemon ? daemon->build_id : "", daemon ? daemon->git_hash : "", + thread_num, query_id, query, signal_description, sig); + } + + LOG_FATAL(log, fmt::runtime(error_message)); + + if (!bare_stacktrace_str.empty()) + { + LOG_FATAL(log, fmt::runtime(bare_stacktrace_str)); + } + + /// Write symbolized stack trace line by line for better grep-ability. + stack_trace.toStringEveryLine([&](std::string_view s) { LOG_FATAL(log, fmt::runtime(s)); }); + + /// In case it's a scheduled job write all previous jobs origins call stacks + std::for_each(thread_frame_pointers.rbegin(), thread_frame_pointers.rend(), + [this](const StackTrace::FramePointers & frame_pointers) + { + if (size_t size = std::ranges::find(frame_pointers, nullptr) - frame_pointers.begin()) + { + LOG_FATAL(log, "========================================"); + WriteBufferFromOwnString bare_stacktrace; + writeString("Job's origin stack trace:", bare_stacktrace); + std::for_each_n(frame_pointers.begin(), size, + [&bare_stacktrace](const void * ptr) + { + writeChar(' ', bare_stacktrace); + writePointerHex(ptr, bare_stacktrace); + } + ); + + LOG_FATAL(log, fmt::runtime(bare_stacktrace.str())); + + StackTrace::toStringEveryLine(const_cast(frame_pointers.data()), 0, size, [this](std::string_view s) { LOG_FATAL(log, fmt::runtime(s)); }); + } + } + ); + + +#if defined(OS_LINUX) + /// Write information about binary checksum. It can be difficult to calculate, so do it only after printing stack trace. + /// Please keep the below log messages in-sync with the ones in programs/server/Server.cpp + + if (daemon && daemon->stored_binary_hash.empty()) + { + LOG_FATAL(log, "Integrity check of the executable skipped because the reference checksum could not be read."); + } + else if (daemon) + { + String calculated_binary_hash = getHashOfLoadedBinaryHex(); + if (calculated_binary_hash == daemon->stored_binary_hash) + { + LOG_FATAL(log, "Integrity check of the executable successfully passed (checksum: {})", calculated_binary_hash); + } + else + { + LOG_FATAL( + log, + "Calculated checksum of the executable ({0}) does not correspond" + " to the reference checksum stored in the executable ({1})." + " This may indicate one of the following:" + " - the executable was changed just after startup;" + " - the executable was corrupted on disk due to faulty hardware;" + " - the loaded executable was corrupted in memory due to faulty hardware;" + " - the file was intentionally modified;" + " - a logical error in the code.", + calculated_binary_hash, + daemon->stored_binary_hash); + } + } +#endif + + /// Write crash to system.crash_log table if available. + if (collectCrashLog) + collectCrashLog(sig, thread_num, query_id, stack_trace); + + Context::getGlobalContextInstance()->handleCrash(); + + /// Send crash report to developers (if configured) + if (sig != SanitizerTrap) + { + if (daemon) + { + if (auto * sentry = SentryWriter::getInstance()) + sentry->onSignal(sig, error_message, stack_trace.getFramePointers(), stack_trace.getOffset(), stack_trace.getSize()); + } + + /// Advice the user to send it manually. + if (std::string_view(VERSION_OFFICIAL).contains("official build")) + { + const auto & date_lut = DateLUT::instance(); + + /// Approximate support period, upper bound. + if (time(nullptr) - date_lut.makeDate(2000 + VERSION_MAJOR, VERSION_MINOR, 1) < (365 + 30) * 86400) + { + LOG_FATAL(log, "Report this error to https://github.com/ClickHouse/ClickHouse/issues"); + } + else + { + LOG_FATAL(log, "ClickHouse version {} is old and should be upgraded to the latest version.", VERSION_STRING); + } + } + else + { + LOG_FATAL(log, "This ClickHouse version is not official and should be upgraded to the official build."); + } + } + + /// List changed settings. + if (!query_id.empty()) + { + ContextPtr query_context = thread_ptr->getQueryContext(); + if (query_context) + { + String changed_settings = query_context->getSettingsRef().toString(); + + if (changed_settings.empty()) + LOG_FATAL(log, "No settings were changed"); + else + LOG_FATAL(log, "Changed settings: {}", changed_settings); + } + } + + /// When everything is done, we will try to send these error messages to the client. + if (thread_ptr) + thread_ptr->onFatalError(); + + HandledSignals::instance().fatal_error_printed.test_and_set(); +} +catch (...) +{ + /// onFault is called from the std::thread, and it should catch all exceptions; otherwise, you can get unrelated fatal errors. + PreformattedMessage message = getCurrentExceptionMessageAndPattern(true); + LOG_FATAL(log, message); +} + +HandledSignals::HandledSignals() +{ + signal_pipe.setNonBlockingWrite(); + signal_pipe.tryIncreaseSize(1 << 20); +} + +void HandledSignals::reset() +{ + /// Reset signals to SIG_DFL to avoid trying to write to the signal_pipe that will be closed after. + for (int sig : handled_signals) + { + if (SIG_ERR == signal(sig, SIG_DFL)) + { + try + { + throw ErrnoException(ErrorCodes::CANNOT_SET_SIGNAL_HANDLER, "Cannot set signal handler"); + } + catch (ErrnoException &) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + } + } + } + + signal_pipe.close(); +} + +HandledSignals::~HandledSignals() +{ + try + { + reset(); + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + } +}; + +HandledSignals & HandledSignals::instance() +{ + static HandledSignals res; + return res; +} + +void HandledSignals::setupTerminateHandler() +{ + std::set_terminate(terminate_handler); +} + +void HandledSignals::setupCommonDeadlySignalHandlers() +{ + /// SIGTSTP is added for debugging purposes. To output a stack trace of any running thread at anytime. + addSignalHandler({SIGABRT, SIGSEGV, SIGILL, SIGBUS, SIGSYS, SIGFPE, SIGPIPE, SIGTSTP, SIGTRAP}, signalHandler, true); + +#if defined(SANITIZER) + __sanitizer_set_death_callback(sanitizerDeathCallback); +#endif +} + +void HandledSignals::setupCommonTerminateRequestSignalHandlers() +{ + addSignalHandler({SIGINT, SIGQUIT, SIGTERM}, terminateRequestedSignalHandler, true); +} diff --git a/src/Common/SignalHandlers.h b/src/Common/SignalHandlers.h new file mode 100644 index 00000000000..e7519f7aee2 --- /dev/null +++ b/src/Common/SignalHandlers.h @@ -0,0 +1,110 @@ +#pragma once +#include + +#include +#include +#include +#include +#include +#include + +class BaseDaemon; + +/** Reset signal handler to the default and send signal to itself. + * It's called from user signal handler to write core dump. + */ +void call_default_signal_handler(int sig); + +const size_t signal_pipe_buf_size = + sizeof(int) + + sizeof(siginfo_t) + + sizeof(ucontext_t*) + + sizeof(StackTrace) + + sizeof(UInt64) + + sizeof(UInt32) + + sizeof(void*); + +using signal_function = void(int, siginfo_t*, void*); + +void writeSignalIDtoSignalPipe(int sig); + +/** Signal handler for HUP */ +void closeLogsSignalHandler(int sig, siginfo_t *, void *); + +void terminateRequestedSignalHandler(int sig, siginfo_t *, void *); + + +/** Handler for "fault" or diagnostic signals. Send data about fault to separate thread to write into log. + */ +void signalHandler(int sig, siginfo_t * info, void * context); + + +/** To use with std::set_terminate. + * Collects slightly more info than __gnu_cxx::__verbose_terminate_handler, + * and send it to pipe. Other thread will read this info from pipe and asynchronously write it to log. + * Look at libstdc++-v3/libsupc++/vterminate.cc for example. + */ +[[noreturn]] void terminate_handler(); + +/// Avoid link time dependency on DB/Interpreters - will use this function only when linked. +__attribute__((__weak__)) void collectCrashLog( + Int32 signal, UInt64 thread_id, const String & query_id, const StackTrace & stack_trace); + + +void blockSignals(const std::vector & signals); + + +/** The thread that read info about signal or std::terminate from pipe. + * On HUP, close log files (for new files to be opened later). + * On information about std::terminate, write it to log. + * On other signals, write info to log. + */ +class SignalListener : public Poco::Runnable +{ +public: + static constexpr int StdTerminate = -1; + static constexpr int StopThread = -2; + static constexpr int SanitizerTrap = -3; + + explicit SignalListener(BaseDaemon * daemon_, LoggerPtr log_) + : daemon(daemon_), log(log_) + { + } + + void run() override; + +private: + BaseDaemon * daemon; + LoggerPtr log; + + void onTerminate(std::string_view message, UInt32 thread_num) const; + + void onFault( + int sig, + const siginfo_t & info, + ucontext_t * context, + const StackTrace & stack_trace, + const std::vector & thread_frame_pointers, + UInt32 thread_num, + DB::ThreadStatus * thread_ptr) const; +}; + +struct HandledSignals +{ + std::vector handled_signals; + DB::PipeFDs signal_pipe; + std::atomic_flag fatal_error_printed; + + HandledSignals(); + ~HandledSignals(); + + void setupTerminateHandler(); + void setupCommonDeadlySignalHandlers(); + void setupCommonTerminateRequestSignalHandlers(); + + void addSignalHandler(const std::vector & signals, signal_function handler, bool register_signal); + + void reset(); + + static HandledSignals & instance(); +}; diff --git a/src/Common/StackTrace.cpp b/src/Common/StackTrace.cpp index 239e957bdfe..76277cbc993 100644 --- a/src/Common/StackTrace.cpp +++ b/src/Common/StackTrace.cpp @@ -489,13 +489,25 @@ struct CacheEntry using CacheEntryPtr = std::shared_ptr; -using StackTraceCache = std::map>; +static constinit bool can_use_cache = false; -static StackTraceCache & cacheInstance() +using StackTraceCacheBase = std::map>; + +struct StackTraceCache : public StackTraceCacheBase { - static StackTraceCache cache; - return cache; -} + StackTraceCache() + : StackTraceCacheBase() + { + can_use_cache = true; + } + + ~StackTraceCache() + { + can_use_cache = false; + } +}; + +static StackTraceCache cache; static DB::SharedMutex stacktrace_cache_mutex; @@ -503,10 +515,16 @@ String toStringCached(const StackTrace::FramePointers & pointers, size_t offset, { const StackTraceRefTriple key{pointers, offset, size}; + if (!can_use_cache) + { + DB::WriteBufferFromOwnString out; + toStringEveryLineImpl(false, key, [&](std::string_view str) { out << str << '\n'; }); + return out.str(); + } + /// Calculation of stack trace text is extremely slow. /// We use cache because otherwise the server could be overloaded by trash queries. /// Note that this cache can grow unconditionally, but practically it should be small. - StackTraceCache & cache = cacheInstance(); CacheEntryPtr cache_entry; // Optimistic try for cache hit to avoid any contention whatsoever, should be the main hot code route @@ -545,7 +563,7 @@ std::string StackTrace::toString() const return toStringCached(frame_pointers, offset, size); } -std::string StackTrace::toString(void ** frame_pointers_raw, size_t offset, size_t size) +std::string StackTrace::toString(void * const * frame_pointers_raw, size_t offset, size_t size) { __msan_unpoison(frame_pointers_raw, size * sizeof(*frame_pointers_raw)); @@ -558,7 +576,7 @@ std::string StackTrace::toString(void ** frame_pointers_raw, size_t offset, size void StackTrace::dropCache() { std::lock_guard lock{stacktrace_cache_mutex}; - cacheInstance().clear(); + cache.clear(); } diff --git a/src/Common/StackTrace.h b/src/Common/StackTrace.h index 4ce9a9281f3..2078828f3d7 100644 --- a/src/Common/StackTrace.h +++ b/src/Common/StackTrace.h @@ -59,7 +59,7 @@ public: const FramePointers & getFramePointers() const { return frame_pointers; } std::string toString() const; - static std::string toString(void ** frame_pointers, size_t offset, size_t size); + static std::string toString(void * const * frame_pointers, size_t offset, size_t size); static void dropCache(); /// @param fatal - if true, will process inline frames (slower) diff --git a/src/Common/SystemLogBase.cpp b/src/Common/SystemLogBase.cpp index 17fa8c90a95..a9307c3be99 100644 --- a/src/Common/SystemLogBase.cpp +++ b/src/Common/SystemLogBase.cpp @@ -11,7 +11,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/src/Common/SystemLogBase.h b/src/Common/SystemLogBase.h index 616f7c42bb9..b87fcf419d3 100644 --- a/src/Common/SystemLogBase.h +++ b/src/Common/SystemLogBase.h @@ -25,7 +25,7 @@ M(ZooKeeperLogElement) \ M(ProcessorProfileLogElement) \ M(TextLogElement) \ - M(S3QueueLogElement) \ + M(ObjectStorageQueueLogElement) \ M(FilesystemCacheLogElement) \ M(FilesystemReadPrefetchesLogElement) \ M(AsynchronousInsertLogElement) \ diff --git a/src/Common/TerminalSize.cpp b/src/Common/TerminalSize.cpp index bc5b4474384..8139f4f7616 100644 --- a/src/Common/TerminalSize.cpp +++ b/src/Common/TerminalSize.cpp @@ -13,17 +13,17 @@ namespace DB::ErrorCodes extern const int SYSTEM_ERROR; } -uint16_t getTerminalWidth() +uint16_t getTerminalWidth(int in_fd, int err_fd) { struct winsize terminal_size {}; - if (isatty(STDIN_FILENO)) + if (isatty(in_fd)) { - if (ioctl(STDIN_FILENO, TIOCGWINSZ, &terminal_size)) + if (ioctl(in_fd, TIOCGWINSZ, &terminal_size)) throw DB::ErrnoException(DB::ErrorCodes::SYSTEM_ERROR, "Cannot obtain terminal window size (ioctl TIOCGWINSZ)"); } - else if (isatty(STDERR_FILENO)) + else if (isatty(err_fd)) { - if (ioctl(STDERR_FILENO, TIOCGWINSZ, &terminal_size)) + if (ioctl(err_fd, TIOCGWINSZ, &terminal_size)) throw DB::ErrnoException(DB::ErrorCodes::SYSTEM_ERROR, "Cannot obtain terminal window size (ioctl TIOCGWINSZ)"); } /// Default - 0. diff --git a/src/Common/TerminalSize.h b/src/Common/TerminalSize.h index b5fc6de7921..f1334f2bcb9 100644 --- a/src/Common/TerminalSize.h +++ b/src/Common/TerminalSize.h @@ -1,16 +1,16 @@ #pragma once #include +#include #include namespace po = boost::program_options; -uint16_t getTerminalWidth(); +uint16_t getTerminalWidth(int in_fd = STDIN_FILENO, int err_fd = STDERR_FILENO); /** Creates po::options_description with name and an appropriate size for option displaying * when program is called with option --help * */ po::options_description createOptionsDescription(const std::string &caption, unsigned short terminal_width); /// NOLINT - diff --git a/src/Common/ThreadFuzzer.cpp b/src/Common/ThreadFuzzer.cpp index 7b96d1f075b..719ac9593ff 100644 --- a/src/Common/ThreadFuzzer.cpp +++ b/src/Common/ThreadFuzzer.cpp @@ -12,13 +12,14 @@ #include #include -#include +#include #include #include +#include +#include #include -#include #include "config.h" // USE_JEMALLOC diff --git a/src/Common/ThreadPool.cpp b/src/Common/ThreadPool.cpp index 01f561d573f..c8f1ae99969 100644 --- a/src/Common/ThreadPool.cpp +++ b/src/Common/ThreadPool.cpp @@ -51,7 +51,7 @@ public: if (!capture_frame_pointers) return; /// Save all previous jobs call stacks and append with current - frame_pointers = DB::Exception::thread_frame_pointers; + frame_pointers = DB::Exception::getThreadFramePointers(); frame_pointers.push_back(StackTrace().getFramePointers()); } @@ -455,7 +455,7 @@ void ThreadPoolImpl::worker(typename std::list::iterator thread_ try { if (DB::Exception::enable_job_stack_trace) - DB::Exception::thread_frame_pointers = std::move(job_data->frame_pointers); + DB::Exception::setThreadFramePointers(std::move(job_data->frame_pointers)); CurrentMetrics::Increment metric_active_pool_threads(metric_active_threads); diff --git a/src/Common/TraceSender.cpp b/src/Common/TraceSender.cpp index 91d07367a82..f1adf7c516a 100644 --- a/src/Common/TraceSender.cpp +++ b/src/Common/TraceSender.cpp @@ -23,8 +23,20 @@ namespace DB LazyPipeFDs TraceSender::pipe; +static thread_local bool inside_send = false; void TraceSender::send(TraceType trace_type, const StackTrace & stack_trace, Extras extras) { + /** The method shouldn't be called recursively or throw exceptions. + * There are several reasons: + * - avoid infinite recursion when some of subsequent functions invoke tracing; + * - avoid inconsistent writes if the method was interrupted by a signal handler in the middle of writing, + * and then another tracing is invoked (e.g., from query profiler). + */ + if (unlikely(inside_send)) + return; + inside_send = true; + DENY_ALLOCATIONS_IN_SCOPE; + constexpr size_t buf_size = sizeof(char) /// TraceCollector stop flag + sizeof(UInt8) /// String size + QUERY_ID_MAX_LEN /// Maximum query_id length @@ -80,6 +92,8 @@ void TraceSender::send(TraceType trace_type, const StackTrace & stack_trace, Ext writePODBinary(extras.increment, out); out.next(); + + inside_send = false; } } diff --git a/src/Common/WeakHash.cpp b/src/Common/WeakHash.cpp index 54d973b6296..cb12df84db1 100644 --- a/src/Common/WeakHash.cpp +++ b/src/Common/WeakHash.cpp @@ -1,2 +1,24 @@ #include +#include +#include +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +void WeakHash32::update(const WeakHash32 & other) +{ + size_t size = data.size(); + if (size != other.data.size()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of WeakHash32 does not match:" + "left size is {}, right size is {}", size, other.data.size()); + + for (size_t i = 0; i < size; ++i) + data[i] = static_cast(intHashCRC32(other.data[i], data[i])); +} + +} diff --git a/src/Common/WeakHash.h b/src/Common/WeakHash.h index b59624e64f2..d4a8d63868c 100644 --- a/src/Common/WeakHash.h +++ b/src/Common/WeakHash.h @@ -11,9 +11,8 @@ namespace DB /// The main purpose why this class needed is to support data initialization. Initially, every bit is 1. class WeakHash32 { - static constexpr UInt32 kDefaultInitialValue = ~UInt32(0); - public: + static constexpr UInt32 kDefaultInitialValue = ~UInt32(0); using Container = PaddedPODArray; @@ -22,6 +21,8 @@ public: void reset(size_t size, UInt32 initial_value = kDefaultInitialValue) { data.assign(size, initial_value); } + void update(const WeakHash32 & other); + const Container & getData() const { return data; } Container & getData() { return data; } diff --git a/src/Common/ZooKeeper/IKeeper.h b/src/Common/ZooKeeper/IKeeper.h index 7d574247aa5..2c6cbc4a5d5 100644 --- a/src/Common/ZooKeeper/IKeeper.h +++ b/src/Common/ZooKeeper/IKeeper.h @@ -559,6 +559,8 @@ public: /// Useful to check owner of ephemeral node. virtual int64_t getSessionID() const = 0; + virtual String tryGetAvailabilityZone() { return ""; } + /// If the method will throw an exception, callbacks won't be called. /// /// After the method is executed successfully, you must wait for callbacks @@ -635,10 +637,6 @@ public: virtual const DB::KeeperFeatureFlags * getKeeperFeatureFlags() const { return nullptr; } - /// A ZooKeeper session can have an optional deadline set on it. - /// After it has been reached, the session needs to be finalized. - virtual bool hasReachedDeadline() const = 0; - /// Expire session and finish all pending requests virtual void finalize(const String & reason) = 0; }; diff --git a/src/Common/ZooKeeper/TestKeeper.h b/src/Common/ZooKeeper/TestKeeper.h index 2774055652c..2194ad015bf 100644 --- a/src/Common/ZooKeeper/TestKeeper.h +++ b/src/Common/ZooKeeper/TestKeeper.h @@ -39,7 +39,6 @@ public: ~TestKeeper() override; bool isExpired() const override { return expired; } - bool hasReachedDeadline() const override { return false; } Int8 getConnectedNodeIdx() const override { return 0; } String getConnectedHostPort() const override { return "TestKeeper:0000"; } int32_t getConnectionXid() const override { return 0; } diff --git a/src/Common/ZooKeeper/ZooKeeper.cpp b/src/Common/ZooKeeper/ZooKeeper.cpp index 4ec44a39136..01bb508da95 100644 --- a/src/Common/ZooKeeper/ZooKeeper.cpp +++ b/src/Common/ZooKeeper/ZooKeeper.cpp @@ -8,18 +8,21 @@ #include #include #include +#include #include #include #include -#include #include +#include #include #include -#include "Common/ZooKeeper/IKeeper.h" -#include +#include +#include #include +#include #include +#include #include #include @@ -55,70 +58,124 @@ static void check(Coordination::Error code, const std::string & path) throw KeeperException::fromPath(code, path); } +UInt64 getSecondsUntilReconnect(const ZooKeeperArgs & args) +{ + std::uniform_int_distribution fallback_session_lifetime_distribution + { + args.fallback_session_lifetime.min_sec, + args.fallback_session_lifetime.max_sec, + }; + UInt32 session_lifetime_seconds = fallback_session_lifetime_distribution(thread_local_rng); + return session_lifetime_seconds; +} -void ZooKeeper::init(ZooKeeperArgs args_) +void ZooKeeper::updateAvailabilityZones() +{ + ShuffleHosts shuffled_hosts = shuffleHosts(); + + for (const auto & node : shuffled_hosts) + { + try + { + ShuffleHosts single_node{node}; + auto tmp_impl = std::make_unique(single_node, args, zk_log); + auto idx = node.original_index; + availability_zones[idx] = tmp_impl->tryGetAvailabilityZone(); + LOG_TEST(log, "Got availability zone for {}: {}", args.hosts[idx], availability_zones[idx]); + } + catch (...) + { + DB::tryLogCurrentException(log, "Failed to get availability zone for " + node.host); + } + } + LOG_DEBUG(log, "Updated availability zones: [{}]", fmt::join(availability_zones, ", ")); +} + +void ZooKeeper::init(ZooKeeperArgs args_, std::unique_ptr existing_impl) { args = std::move(args_); log = getLogger("ZooKeeper"); - if (args.implementation == "zookeeper") + if (existing_impl) + { + chassert(args.implementation == "zookeeper"); + impl = std::move(existing_impl); + LOG_INFO(log, "Switching to connection to a more optimal node {}", impl->getConnectedHostPort()); + } + else if (args.implementation == "zookeeper") { if (args.hosts.empty()) throw KeeperException::fromMessage(Coordination::Error::ZBADARGUMENTS, "No hosts passed to ZooKeeper constructor."); - Coordination::ZooKeeper::Nodes nodes; - nodes.reserve(args.hosts.size()); + chassert(args.availability_zones.size() == args.hosts.size()); + if (availability_zones.empty()) + { + /// availability_zones is empty on server startup or after config reloading + /// We will keep the az info when starting new sessions + availability_zones = args.availability_zones; + + LOG_TEST(log, "Availability zones from config: [{}], client: {}", + fmt::join(collections::map(availability_zones, [](auto s){ return DB::quoteString(s); }), ", "), + DB::quoteString(args.client_availability_zone)); + + if (args.availability_zone_autodetect) + updateAvailabilityZones(); + } + chassert(availability_zones.size() == args.hosts.size()); /// Shuffle the hosts to distribute the load among ZooKeeper nodes. - std::vector shuffled_hosts = shuffleHosts(); + ShuffleHosts shuffled_hosts = shuffleHosts(); - bool dns_error = false; - for (auto & host : shuffled_hosts) - { - auto & host_string = host.host; - try - { - const bool secure = startsWith(host_string, "secure://"); - - if (secure) - host_string.erase(0, strlen("secure://")); - - /// We want to resolve all hosts without DNS cache for keeper connection. - Coordination::DNSResolver::instance().removeHostFromCache(host_string); - - const Poco::Net::SocketAddress host_socket_addr{host_string}; - LOG_TEST(log, "Adding ZooKeeper host {} ({})", host_string, host_socket_addr.toString()); - nodes.emplace_back(Coordination::ZooKeeper::Node{host_socket_addr, host.original_index, secure}); - } - catch (const Poco::Net::HostNotFoundException & e) - { - /// Most likely it's misconfiguration and wrong hostname was specified - LOG_ERROR(log, "Cannot use ZooKeeper host {}, reason: {}", host_string, e.displayText()); - } - catch (const Poco::Net::DNSException & e) - { - /// Most likely DNS is not available now - dns_error = true; - LOG_ERROR(log, "Cannot use ZooKeeper host {} due to DNS error: {}", host_string, e.displayText()); - } - } - - if (nodes.empty()) - { - /// For DNS errors we throw exception with ZCONNECTIONLOSS code, so it will be considered as hardware error, not user error - if (dns_error) - throw KeeperException::fromMessage(Coordination::Error::ZCONNECTIONLOSS, "Cannot resolve any of provided ZooKeeper hosts due to DNS error"); - else - throw KeeperException::fromMessage(Coordination::Error::ZCONNECTIONLOSS, "Cannot use any of provided ZooKeeper nodes"); - } - - impl = std::make_unique(nodes, args, zk_log); + impl = std::make_unique(shuffled_hosts, args, zk_log); + Int8 node_idx = impl->getConnectedNodeIdx(); if (args.chroot.empty()) LOG_TRACE(log, "Initialized, hosts: {}", fmt::join(args.hosts, ",")); else LOG_TRACE(log, "Initialized, hosts: {}, chroot: {}", fmt::join(args.hosts, ","), args.chroot); + + + /// If the balancing strategy has an optimal node then it will be the first in the list + bool connected_to_suboptimal_node = node_idx != shuffled_hosts[0].original_index; + bool respect_az = args.prefer_local_availability_zone && !args.client_availability_zone.empty(); + bool may_benefit_from_reconnecting = respect_az || args.get_priority_load_balancing.hasOptimalNode(); + if (connected_to_suboptimal_node && may_benefit_from_reconnecting) + { + auto reconnect_timeout_sec = getSecondsUntilReconnect(args); + LOG_DEBUG(log, "Connected to a suboptimal ZooKeeper host ({}, index {})." + " To preserve balance in ZooKeeper usage, this ZooKeeper session will expire in {} seconds", + impl->getConnectedHostPort(), node_idx, reconnect_timeout_sec); + + auto reconnect_task_holder = DB::Context::getGlobalContextInstance()->getSchedulePool().createTask("ZKReconnect", [this, optimal_host = shuffled_hosts[0]]() + { + try + { + LOG_DEBUG(log, "Trying to connect to a more optimal node {}", optimal_host.host); + ShuffleHosts node{optimal_host}; + std::unique_ptr new_impl = std::make_unique(node, args, zk_log); + Int8 new_node_idx = new_impl->getConnectedNodeIdx(); + + /// Maybe the node was unavailable when getting AZs first time, update just in case + if (args.availability_zone_autodetect && availability_zones[new_node_idx].empty()) + { + availability_zones[new_node_idx] = new_impl->tryGetAvailabilityZone(); + LOG_DEBUG(log, "Got availability zone for {}: {}", optimal_host.host, availability_zones[new_node_idx]); + } + + optimal_impl = std::move(new_impl); + impl->finalize("Connected to a more optimal node"); + } + catch (...) + { + LOG_WARNING(log, "Failed to connect to a more optimal ZooKeeper, will try again later: {}", DB::getCurrentExceptionMessage(/*with_stacktrace*/ false)); + (*reconnect_task)->scheduleAfter(getSecondsUntilReconnect(args) * 1000); + } + }); + reconnect_task = std::make_unique(std::move(reconnect_task_holder)); + (*reconnect_task)->activate(); + (*reconnect_task)->scheduleAfter(reconnect_timeout_sec * 1000); + } } else if (args.implementation == "testkeeper") { @@ -152,29 +209,53 @@ void ZooKeeper::init(ZooKeeperArgs args_) } } +ZooKeeper::~ZooKeeper() +{ + if (reconnect_task) + (*reconnect_task)->deactivate(); +} ZooKeeper::ZooKeeper(const ZooKeeperArgs & args_, std::shared_ptr zk_log_) : zk_log(std::move(zk_log_)) { - init(args_); + init(args_, /*existing_impl*/ {}); +} + + +ZooKeeper::ZooKeeper(const ZooKeeperArgs & args_, std::shared_ptr zk_log_, Strings availability_zones_, std::unique_ptr existing_impl) + : availability_zones(std::move(availability_zones_)), zk_log(std::move(zk_log_)) +{ + if (availability_zones.size() != args_.hosts.size()) + throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Argument sizes mismatch: availability_zones count {} and hosts count {}", + availability_zones.size(), args_.hosts.size()); + init(args_, std::move(existing_impl)); } ZooKeeper::ZooKeeper(const Poco::Util::AbstractConfiguration & config, const std::string & config_name, std::shared_ptr zk_log_) : zk_log(std::move(zk_log_)) { - init(ZooKeeperArgs(config, config_name)); + init(ZooKeeperArgs(config, config_name), /*existing_impl*/ {}); } -std::vector ZooKeeper::shuffleHosts() const +ShuffleHosts ZooKeeper::shuffleHosts() const { - std::function get_priority = args.get_priority_load_balancing.getPriorityFunc(args.get_priority_load_balancing.load_balancing, 0, args.hosts.size()); - std::vector shuffle_hosts; + std::function get_priority = args.get_priority_load_balancing.getPriorityFunc( + args.get_priority_load_balancing.load_balancing, /* offset for first_or_random */ 0, args.hosts.size()); + ShuffleHosts shuffle_hosts; for (size_t i = 0; i < args.hosts.size(); ++i) { ShuffleHost shuffle_host; shuffle_host.host = args.hosts[i]; shuffle_host.original_index = static_cast(i); + + shuffle_host.secure = startsWith(shuffle_host.host, "secure://"); + if (shuffle_host.secure) + shuffle_host.host.erase(0, strlen("secure://")); + + if (!args.client_availability_zone.empty() && !availability_zones[i].empty()) + shuffle_host.az_info = availability_zones[i] == args.client_availability_zone ? ShuffleHost::SAME : ShuffleHost::OTHER; + if (get_priority) shuffle_host.priority = get_priority(i); shuffle_host.randomize(); @@ -1023,7 +1104,10 @@ ZooKeeperPtr ZooKeeper::create(const Poco::Util::AbstractConfiguration & config, ZooKeeperPtr ZooKeeper::startNewSession() const { - auto res = std::shared_ptr(new ZooKeeper(args, zk_log)); + if (reconnect_task) + (*reconnect_task)->deactivate(); + + auto res = std::shared_ptr(new ZooKeeper(args, zk_log, availability_zones, std::move(optimal_impl))); res->initSession(); return res; } @@ -1456,6 +1540,16 @@ int32_t ZooKeeper::getConnectionXid() const return impl->getConnectionXid(); } +String ZooKeeper::getConnectedHostAvailabilityZone() const +{ + if (args.implementation != "zookeeper" || !impl) + return ""; + Int8 idx = impl->getConnectedNodeIdx(); + if (idx < 0) + return ""; /// session expired + return availability_zones.at(idx); +} + size_t getFailedOpIndex(Coordination::Error exception_code, const Coordination::Responses & responses) { if (responses.empty()) diff --git a/src/Common/ZooKeeper/ZooKeeper.h b/src/Common/ZooKeeper/ZooKeeper.h index 08ff60a80cf..4ae2cfa6096 100644 --- a/src/Common/ZooKeeper/ZooKeeper.h +++ b/src/Common/ZooKeeper/ZooKeeper.h @@ -32,6 +32,7 @@ namespace DB { class ZooKeeperLog; class ZooKeeperWithFaultInjection; +class BackgroundSchedulePoolTaskHolder; namespace ErrorCodes { @@ -48,11 +49,23 @@ constexpr size_t MULTI_BATCH_SIZE = 100; struct ShuffleHost { + enum AvailabilityZoneInfo + { + SAME = 0, + UNKNOWN = 1, + OTHER = 2, + }; + String host; + bool secure = false; UInt8 original_index = 0; + AvailabilityZoneInfo az_info = UNKNOWN; Priority priority; UInt64 random = 0; + /// We should resolve it each time without caching + mutable std::optional address; + void randomize() { random = thread_local_rng(); @@ -60,11 +73,13 @@ struct ShuffleHost static bool compare(const ShuffleHost & lhs, const ShuffleHost & rhs) { - return std::forward_as_tuple(lhs.priority, lhs.random) - < std::forward_as_tuple(rhs.priority, rhs.random); + return std::forward_as_tuple(lhs.az_info, lhs.priority, lhs.random) + < std::forward_as_tuple(rhs.az_info, rhs.priority, rhs.random); } }; +using ShuffleHosts = std::vector; + struct RemoveException { explicit RemoveException(std::string_view path_ = "", bool remove_subtree_ = true) @@ -197,6 +212,9 @@ class ZooKeeper explicit ZooKeeper(const ZooKeeperArgs & args_, std::shared_ptr zk_log_ = nullptr); + /// Allows to keep info about availability zones when starting a new session + ZooKeeper(const ZooKeeperArgs & args_, std::shared_ptr zk_log_, Strings availability_zones_, std::unique_ptr existing_impl); + /** Config of the form: @@ -228,7 +246,9 @@ public: using Ptr = std::shared_ptr; using ErrorsList = std::initializer_list; - std::vector shuffleHosts() const; + ~ZooKeeper(); + + ShuffleHosts shuffleHosts() const; static Ptr create(const Poco::Util::AbstractConfiguration & config, const std::string & config_name, @@ -596,8 +616,6 @@ public: UInt32 getSessionUptime() const { return static_cast(session_uptime.elapsedSeconds()); } - bool hasReachedDeadline() const { return impl->hasReachedDeadline(); } - uint64_t getSessionTimeoutMS() const { return args.session_timeout_ms; } void setServerCompletelyStarted(); @@ -606,6 +624,8 @@ public: String getConnectedHostPort() const; int32_t getConnectionXid() const; + String getConnectedHostAvailabilityZone() const; + const DB::KeeperFeatureFlags * getKeeperFeatureFlags() const { return impl->getKeeperFeatureFlags(); } /// Checks that our session was not killed, and allows to avoid applying a request from an old lost session. @@ -625,7 +645,8 @@ public: void addCheckSessionOp(Coordination::Requests & requests) const; private: - void init(ZooKeeperArgs args_); + void init(ZooKeeperArgs args_, std::unique_ptr existing_impl); + void updateAvailabilityZones(); /// The following methods don't any throw exceptions but return error codes. Coordination::Error createImpl(const std::string & path, const std::string & data, int32_t mode, std::string & path_created); @@ -690,15 +711,20 @@ private: } std::unique_ptr impl; + mutable std::unique_ptr optimal_impl; ZooKeeperArgs args; + Strings availability_zones; + LoggerPtr log = nullptr; std::shared_ptr zk_log; AtomicStopwatch session_uptime; int32_t session_node_version; + + std::unique_ptr reconnect_task; }; diff --git a/src/Common/ZooKeeper/ZooKeeperArgs.cpp b/src/Common/ZooKeeper/ZooKeeperArgs.cpp index a581b6a7f38..18dff779a70 100644 --- a/src/Common/ZooKeeper/ZooKeeperArgs.cpp +++ b/src/Common/ZooKeeper/ZooKeeperArgs.cpp @@ -5,6 +5,9 @@ #include #include #include +#include +#include +#include #include namespace DB @@ -53,6 +56,7 @@ ZooKeeperArgs::ZooKeeperArgs(const Poco::Util::AbstractConfiguration & config, c ZooKeeperArgs::ZooKeeperArgs(const String & hosts_string) { splitInto<','>(hosts, hosts_string); + availability_zones.resize(hosts.size()); } void ZooKeeperArgs::initFromKeeperServerSection(const Poco::Util::AbstractConfiguration & config) @@ -103,8 +107,11 @@ void ZooKeeperArgs::initFromKeeperServerSection(const Poco::Util::AbstractConfig for (const auto & key : keys) { if (startsWith(key, "server")) + { hosts.push_back( (secure ? "secure://" : "") + config.getString(raft_configuration_key + "." + key + ".hostname") + ":" + tcp_port); + availability_zones.push_back(config.getString(raft_configuration_key + "." + key + ".availability_zone", "")); + } } static constexpr std::array load_balancing_keys @@ -123,11 +130,15 @@ void ZooKeeperArgs::initFromKeeperServerSection(const Poco::Util::AbstractConfig auto load_balancing = magic_enum::enum_cast(Poco::toUpper(load_balancing_str)); if (!load_balancing) throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Unknown load balancing: {}", load_balancing_str); - get_priority_load_balancing.load_balancing = *load_balancing; + get_priority_load_balancing = DB::GetPriorityForLoadBalancing(*load_balancing, thread_local_rng() % hosts.size()); break; } } + availability_zone_autodetect = config.getBool(std::string{config_name} + ".availability_zone_autodetect", false); + prefer_local_availability_zone = config.getBool(std::string{config_name} + ".prefer_local_availability_zone", false); + if (prefer_local_availability_zone) + client_availability_zone = DB::PlacementInfo::PlacementInfo::instance().getAvailabilityZone(); } void ZooKeeperArgs::initFromKeeperSection(const Poco::Util::AbstractConfiguration & config, const std::string & config_name) @@ -137,6 +148,8 @@ void ZooKeeperArgs::initFromKeeperSection(const Poco::Util::AbstractConfiguratio Poco::Util::AbstractConfiguration::Keys keys; config.keys(config_name, keys); + std::optional load_balancing; + for (const auto & key : keys) { if (key.starts_with("node")) @@ -144,6 +157,7 @@ void ZooKeeperArgs::initFromKeeperSection(const Poco::Util::AbstractConfiguratio hosts.push_back( (config.getBool(config_name + "." + key + ".secure", false) ? "secure://" : "") + config.getString(config_name + "." + key + ".host") + ":" + config.getString(config_name + "." + key + ".port", "2181")); + availability_zones.push_back(config.getString(config_name + "." + key + ".availability_zone", "")); } else if (key == "session_timeout_ms") { @@ -199,6 +213,10 @@ void ZooKeeperArgs::initFromKeeperSection(const Poco::Util::AbstractConfiguratio { sessions_path = config.getString(config_name + "." + key); } + else if (key == "prefer_local_availability_zone") + { + prefer_local_availability_zone = config.getBool(config_name + "." + key); + } else if (key == "implementation") { implementation = config.getString(config_name + "." + key); @@ -207,10 +225,9 @@ void ZooKeeperArgs::initFromKeeperSection(const Poco::Util::AbstractConfiguratio { String load_balancing_str = config.getString(config_name + "." + key); /// Use magic_enum to avoid dependency from dbms (`SettingFieldLoadBalancingTraits::fromString(...)`) - auto load_balancing = magic_enum::enum_cast(Poco::toUpper(load_balancing_str)); + load_balancing = magic_enum::enum_cast(Poco::toUpper(load_balancing_str)); if (!load_balancing) throw DB::Exception(DB::ErrorCodes::BAD_ARGUMENTS, "Unknown load balancing: {}", load_balancing_str); - get_priority_load_balancing.load_balancing = *load_balancing; } else if (key == "fallback_session_lifetime") { @@ -224,9 +241,19 @@ void ZooKeeperArgs::initFromKeeperSection(const Poco::Util::AbstractConfiguratio { use_compression = config.getBool(config_name + "." + key); } + else if (key == "availability_zone_autodetect") + { + availability_zone_autodetect = config.getBool(config_name + "." + key); + } else throw KeeperException(Coordination::Error::ZBADARGUMENTS, "Unknown key {} in config file", key); } + + if (load_balancing) + get_priority_load_balancing = DB::GetPriorityForLoadBalancing(*load_balancing, thread_local_rng() % hosts.size()); + + if (prefer_local_availability_zone) + client_availability_zone = DB::PlacementInfo::PlacementInfo::instance().getAvailabilityZone(); } } diff --git a/src/Common/ZooKeeper/ZooKeeperArgs.h b/src/Common/ZooKeeper/ZooKeeperArgs.h index 27ba173c0c3..945b77bf9c1 100644 --- a/src/Common/ZooKeeper/ZooKeeperArgs.h +++ b/src/Common/ZooKeeper/ZooKeeperArgs.h @@ -32,10 +32,12 @@ struct ZooKeeperArgs String zookeeper_name = "zookeeper"; String implementation = "zookeeper"; Strings hosts; + Strings availability_zones; String auth_scheme; String identity; String chroot; String sessions_path = "/clickhouse/sessions"; + String client_availability_zone; int32_t connection_timeout_ms = Coordination::DEFAULT_CONNECTION_TIMEOUT_MS; int32_t session_timeout_ms = Coordination::DEFAULT_SESSION_TIMEOUT_MS; int32_t operation_timeout_ms = Coordination::DEFAULT_OPERATION_TIMEOUT_MS; @@ -47,6 +49,8 @@ struct ZooKeeperArgs UInt64 send_sleep_ms = 0; UInt64 recv_sleep_ms = 0; bool use_compression = false; + bool prefer_local_availability_zone = false; + bool availability_zone_autodetect = false; SessionLifetimeConfiguration fallback_session_lifetime = {}; DB::GetPriorityForLoadBalancing get_priority_load_balancing; diff --git a/src/Common/ZooKeeper/ZooKeeperCommon.cpp b/src/Common/ZooKeeper/ZooKeeperCommon.cpp index 48bb510e589..dff14f74681 100644 --- a/src/Common/ZooKeeper/ZooKeeperCommon.cpp +++ b/src/Common/ZooKeeper/ZooKeeperCommon.cpp @@ -9,7 +9,6 @@ #include #include #include -#include namespace Coordination @@ -29,7 +28,7 @@ void ZooKeeperResponse::write(WriteBuffer & out) const Coordination::write(buf.str(), out); } -std::string ZooKeeperRequest::toString() const +std::string ZooKeeperRequest::toString(bool short_format) const { return fmt::format( "XID = {}\n" @@ -37,7 +36,7 @@ std::string ZooKeeperRequest::toString() const "Additional info:\n{}", xid, getOpNum(), - toStringImpl()); + toStringImpl(short_format)); } void ZooKeeperRequest::write(WriteBuffer & out) const @@ -60,7 +59,7 @@ void ZooKeeperSyncRequest::readImpl(ReadBuffer & in) Coordination::read(path, in); } -std::string ZooKeeperSyncRequest::toStringImpl() const +std::string ZooKeeperSyncRequest::toStringImpl(bool /*short_format*/) const { return fmt::format("path = {}", path); } @@ -91,7 +90,7 @@ void ZooKeeperReconfigRequest::readImpl(ReadBuffer & in) Coordination::read(version, in); } -std::string ZooKeeperReconfigRequest::toStringImpl() const +std::string ZooKeeperReconfigRequest::toStringImpl(bool /*short_format*/) const { return fmt::format( "joining = {}\nleaving = {}\nnew_members = {}\nversion = {}", @@ -145,7 +144,7 @@ void ZooKeeperAuthRequest::readImpl(ReadBuffer & in) Coordination::read(data, in); } -std::string ZooKeeperAuthRequest::toStringImpl() const +std::string ZooKeeperAuthRequest::toStringImpl(bool /*short_format*/) const { return fmt::format( "type = {}\n" @@ -191,7 +190,7 @@ void ZooKeeperCreateRequest::readImpl(ReadBuffer & in) is_sequential = true; } -std::string ZooKeeperCreateRequest::toStringImpl() const +std::string ZooKeeperCreateRequest::toStringImpl(bool /*short_format*/) const { return fmt::format( "path = {}\n" @@ -218,7 +217,7 @@ void ZooKeeperRemoveRequest::writeImpl(WriteBuffer & out) const Coordination::write(version, out); } -std::string ZooKeeperRemoveRequest::toStringImpl() const +std::string ZooKeeperRemoveRequest::toStringImpl(bool /*short_format*/) const { return fmt::format( "path = {}\n" @@ -245,7 +244,7 @@ void ZooKeeperExistsRequest::readImpl(ReadBuffer & in) Coordination::read(has_watch, in); } -std::string ZooKeeperExistsRequest::toStringImpl() const +std::string ZooKeeperExistsRequest::toStringImpl(bool /*short_format*/) const { return fmt::format("path = {}", path); } @@ -272,7 +271,7 @@ void ZooKeeperGetRequest::readImpl(ReadBuffer & in) Coordination::read(has_watch, in); } -std::string ZooKeeperGetRequest::toStringImpl() const +std::string ZooKeeperGetRequest::toStringImpl(bool /*short_format*/) const { return fmt::format("path = {}", path); } @@ -303,7 +302,7 @@ void ZooKeeperSetRequest::readImpl(ReadBuffer & in) Coordination::read(version, in); } -std::string ZooKeeperSetRequest::toStringImpl() const +std::string ZooKeeperSetRequest::toStringImpl(bool /*short_format*/) const { return fmt::format( "path = {}\n" @@ -334,7 +333,7 @@ void ZooKeeperListRequest::readImpl(ReadBuffer & in) Coordination::read(has_watch, in); } -std::string ZooKeeperListRequest::toStringImpl() const +std::string ZooKeeperListRequest::toStringImpl(bool /*short_format*/) const { return fmt::format("path = {}", path); } @@ -356,7 +355,7 @@ void ZooKeeperFilteredListRequest::readImpl(ReadBuffer & in) list_request_type = static_cast(read_request_type); } -std::string ZooKeeperFilteredListRequest::toStringImpl() const +std::string ZooKeeperFilteredListRequest::toStringImpl(bool /*short_format*/) const { return fmt::format( "path = {}\n" @@ -401,7 +400,7 @@ void ZooKeeperSetACLRequest::readImpl(ReadBuffer & in) Coordination::read(version, in); } -std::string ZooKeeperSetACLRequest::toStringImpl() const +std::string ZooKeeperSetACLRequest::toStringImpl(bool /*short_format*/) const { return fmt::format("path = {}\nversion = {}", path, version); } @@ -426,7 +425,7 @@ void ZooKeeperGetACLRequest::writeImpl(WriteBuffer & out) const Coordination::write(path, out); } -std::string ZooKeeperGetACLRequest::toStringImpl() const +std::string ZooKeeperGetACLRequest::toStringImpl(bool /*short_format*/) const { return fmt::format("path = {}", path); } @@ -455,7 +454,7 @@ void ZooKeeperCheckRequest::readImpl(ReadBuffer & in) Coordination::read(version, in); } -std::string ZooKeeperCheckRequest::toStringImpl() const +std::string ZooKeeperCheckRequest::toStringImpl(bool /*short_format*/) const { return fmt::format("path = {}\nversion = {}", path, version); } @@ -600,8 +599,11 @@ void ZooKeeperMultiRequest::readImpl(ReadBuffer & in) } } -std::string ZooKeeperMultiRequest::toStringImpl() const +std::string ZooKeeperMultiRequest::toStringImpl(bool short_format) const { + if (short_format) + return fmt::format("Subrequests size = {}", requests.size()); + auto out = fmt::memory_buffer(); for (const auto & request : requests) { diff --git a/src/Common/ZooKeeper/ZooKeeperCommon.h b/src/Common/ZooKeeper/ZooKeeperCommon.h index 490c2dce4f8..fd6ec3cd375 100644 --- a/src/Common/ZooKeeper/ZooKeeperCommon.h +++ b/src/Common/ZooKeeper/ZooKeeperCommon.h @@ -63,12 +63,12 @@ struct ZooKeeperRequest : virtual Request /// Writes length, xid, op_num, then the rest. void write(WriteBuffer & out) const; - std::string toString() const; + std::string toString(bool short_format = false) const; virtual void writeImpl(WriteBuffer &) const = 0; virtual void readImpl(ReadBuffer &) = 0; - virtual std::string toStringImpl() const { return ""; } + virtual std::string toStringImpl(bool /*short_format*/) const { return ""; } static std::shared_ptr read(ReadBuffer & in); @@ -98,7 +98,7 @@ struct ZooKeeperSyncRequest final : ZooKeeperRequest OpNum getOpNum() const override { return OpNum::Sync; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return false; } @@ -123,7 +123,7 @@ struct ZooKeeperReconfigRequest final : ZooKeeperRequest OpNum getOpNum() const override { return OpNum::Reconfig; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return false; } @@ -176,7 +176,7 @@ struct ZooKeeperAuthRequest final : ZooKeeperRequest OpNum getOpNum() const override { return OpNum::Auth; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return false; } @@ -229,7 +229,7 @@ struct ZooKeeperCreateRequest final : public CreateRequest, ZooKeeperRequest OpNum getOpNum() const override { return not_exists ? OpNum::CreateIfNotExists : OpNum::Create; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return false; } @@ -266,7 +266,7 @@ struct ZooKeeperRemoveRequest final : RemoveRequest, ZooKeeperRequest OpNum getOpNum() const override { return OpNum::Remove; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return false; } @@ -293,7 +293,7 @@ struct ZooKeeperExistsRequest final : ExistsRequest, ZooKeeperRequest OpNum getOpNum() const override { return OpNum::Exists; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return true; } @@ -320,7 +320,7 @@ struct ZooKeeperGetRequest final : GetRequest, ZooKeeperRequest OpNum getOpNum() const override { return OpNum::Get; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return true; } @@ -347,7 +347,7 @@ struct ZooKeeperSetRequest final : SetRequest, ZooKeeperRequest OpNum getOpNum() const override { return OpNum::Set; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return false; } @@ -375,7 +375,7 @@ struct ZooKeeperListRequest : ListRequest, ZooKeeperRequest OpNum getOpNum() const override { return OpNum::List; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return true; } @@ -395,7 +395,7 @@ struct ZooKeeperFilteredListRequest final : ZooKeeperListRequest OpNum getOpNum() const override { return OpNum::FilteredList; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; size_t bytesSize() const override { return ZooKeeperListRequest::bytesSize() + sizeof(list_request_type); } }; @@ -428,7 +428,7 @@ struct ZooKeeperCheckRequest : CheckRequest, ZooKeeperRequest OpNum getOpNum() const override { return not_exists ? OpNum::CheckNotExists : OpNum::Check; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return true; } @@ -469,7 +469,7 @@ struct ZooKeeperSetACLRequest final : SetACLRequest, ZooKeeperRequest OpNum getOpNum() const override { return OpNum::SetACL; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return false; } @@ -490,7 +490,7 @@ struct ZooKeeperGetACLRequest final : GetACLRequest, ZooKeeperRequest OpNum getOpNum() const override { return OpNum::GetACL; } void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override { return true; } @@ -516,7 +516,7 @@ struct ZooKeeperMultiRequest final : MultiRequest, ZooKeeperRequest void writeImpl(WriteBuffer & out) const override; void readImpl(ReadBuffer & in) override; - std::string toStringImpl() const override; + std::string toStringImpl(bool short_format) const override; ZooKeeperResponsePtr makeResponse() const override; bool isReadRequest() const override; diff --git a/src/Common/ZooKeeper/ZooKeeperImpl.cpp b/src/Common/ZooKeeper/ZooKeeperImpl.cpp index ed7498b1ac9..d01fc341a63 100644 --- a/src/Common/ZooKeeper/ZooKeeperImpl.cpp +++ b/src/Common/ZooKeeper/ZooKeeperImpl.cpp @@ -23,6 +23,9 @@ #include #include +#include +#include + #include "Coordination/KeeperConstants.h" #include "config.h" @@ -338,7 +341,7 @@ ZooKeeper::~ZooKeeper() ZooKeeper::ZooKeeper( - const Nodes & nodes, + const zkutil::ShuffleHosts & nodes, const zkutil::ZooKeeperArgs & args_, std::shared_ptr zk_log_) : args(args_) @@ -426,7 +429,7 @@ ZooKeeper::ZooKeeper( void ZooKeeper::connect( - const Nodes & nodes, + const zkutil::ShuffleHosts & nodes, Poco::Timespan connection_timeout) { if (nodes.empty()) @@ -434,15 +437,51 @@ void ZooKeeper::connect( static constexpr size_t num_tries = 3; bool connected = false; + bool dns_error = false; + + size_t resolved_count = 0; + for (const auto & node : nodes) + { + try + { + const Poco::Net::SocketAddress host_socket_addr{node.host}; + LOG_TRACE(log, "Adding ZooKeeper host {} ({}), az: {}, priority: {}", node.host, host_socket_addr.toString(), node.az_info, node.priority); + node.address = host_socket_addr; + ++resolved_count; + } + catch (const Poco::Net::HostNotFoundException & e) + { + /// Most likely it's misconfiguration and wrong hostname was specified + LOG_ERROR(log, "Cannot use ZooKeeper host {}, reason: {}", node.host, e.displayText()); + } + catch (const Poco::Net::DNSException & e) + { + /// Most likely DNS is not available now + dns_error = true; + LOG_ERROR(log, "Cannot use ZooKeeper host {} due to DNS error: {}", node.host, e.displayText()); + } + } + + if (resolved_count == 0) + { + /// For DNS errors we throw exception with ZCONNECTIONLOSS code, so it will be considered as hardware error, not user error + if (dns_error) + throw zkutil::KeeperException::fromMessage( + Coordination::Error::ZCONNECTIONLOSS, "Cannot resolve any of provided ZooKeeper hosts due to DNS error"); + else + throw zkutil::KeeperException::fromMessage(Coordination::Error::ZCONNECTIONLOSS, "Cannot use any of provided ZooKeeper nodes"); + } WriteBufferFromOwnString fail_reasons; for (size_t try_no = 0; try_no < num_tries; ++try_no) { - for (size_t i = 0; i < nodes.size(); ++i) + for (const auto & node : nodes) { - const auto & node = nodes[i]; try { + if (!node.address) + continue; + /// Reset the state of previous attempt. if (node.secure) { @@ -458,7 +497,7 @@ void ZooKeeper::connect( socket = Poco::Net::StreamSocket(); } - socket.connect(node.address, connection_timeout); + socket.connect(*node.address, connection_timeout); socket_address = socket.peerAddress(); socket.setReceiveTimeout(args.operation_timeout_ms * 1000); @@ -498,27 +537,11 @@ void ZooKeeper::connect( } original_index = static_cast(node.original_index); - - if (i != 0) - { - std::uniform_int_distribution fallback_session_lifetime_distribution - { - args.fallback_session_lifetime.min_sec, - args.fallback_session_lifetime.max_sec, - }; - UInt32 session_lifetime_seconds = fallback_session_lifetime_distribution(thread_local_rng); - client_session_deadline = clock::now() + std::chrono::seconds(session_lifetime_seconds); - - LOG_DEBUG(log, "Connected to a suboptimal ZooKeeper host ({}, index {})." - " To preserve balance in ZooKeeper usage, this ZooKeeper session will expire in {} seconds", - node.address.toString(), i, session_lifetime_seconds); - } - break; } catch (...) { - fail_reasons << "\n" << getCurrentExceptionMessage(false) << ", " << node.address.toString(); + fail_reasons << "\n" << getCurrentExceptionMessage(false) << ", " << node.address->toString(); } } @@ -532,6 +555,9 @@ void ZooKeeper::connect( bool first = true; for (const auto & node : nodes) { + if (!node.address) + continue; + if (first) first = false; else @@ -540,7 +566,7 @@ void ZooKeeper::connect( if (node.secure) message << "secure://"; - message << node.address.toString(); + message << node.address->toString(); } message << fail_reasons.str() << "\n"; @@ -970,6 +996,10 @@ void ZooKeeper::receiveEvent() if (request_info.callback) request_info.callback(*response); + + /// Finalize current session if we receive a hardware error from ZooKeeper + if (err != Error::ZOK && isHardwareError(err)) + finalize(/*error_send*/ false, /*error_receive*/ true, fmt::format("Hardware error: {}", err)); } @@ -984,9 +1014,6 @@ void ZooKeeper::finalize(bool error_send, bool error_receive, const String & rea LOG_INFO(log, "Finalizing session {}. finalization_started: {}, queue_finished: {}, reason: '{}'", session_id, already_started, requests_queue.isFinished(), reason); - /// Reset the original index. - original_index = -1; - auto expire_session_if_not_expired = [&] { /// No new requests will appear in queue after finish() @@ -1153,7 +1180,6 @@ void ZooKeeper::pushRequest(RequestInfo && info) { try { - checkSessionDeadline(); info.time = clock::now(); auto maybe_zk_log = std::atomic_load(&zk_log); if (maybe_zk_log) @@ -1201,44 +1227,44 @@ bool ZooKeeper::isFeatureEnabled(KeeperFeatureFlag feature_flag) const return keeper_feature_flags.isEnabled(feature_flag); } -void ZooKeeper::initFeatureFlags() +std::optional ZooKeeper::tryGetSystemZnode(const std::string & path, const std::string & description) { - const auto try_get = [&](const std::string & path, const std::string & description) -> std::optional + auto promise = std::make_shared>(); + auto future = promise->get_future(); + + auto callback = [promise](const Coordination::GetResponse & response) mutable { - auto promise = std::make_shared>(); - auto future = promise->get_future(); - - auto callback = [promise](const Coordination::GetResponse & response) mutable - { - promise->set_value(response); - }; - - get(path, std::move(callback), {}); - if (future.wait_for(std::chrono::milliseconds(args.operation_timeout_ms)) != std::future_status::ready) - throw Exception(Error::ZOPERATIONTIMEOUT, "Failed to get {}: timeout", description); - - auto response = future.get(); - - if (response.error == Coordination::Error::ZNONODE) - { - LOG_TRACE(log, "Failed to get {}", description); - return std::nullopt; - } - else if (response.error != Coordination::Error::ZOK) - { - throw Exception(response.error, "Failed to get {}", description); - } - - return std::move(response.data); + promise->set_value(response); }; - if (auto feature_flags = try_get(keeper_api_feature_flags_path, "feature flags"); feature_flags.has_value()) + get(path, std::move(callback), {}); + if (future.wait_for(std::chrono::milliseconds(args.operation_timeout_ms)) != std::future_status::ready) + throw Exception(Error::ZOPERATIONTIMEOUT, "Failed to get {}: timeout", description); + + auto response = future.get(); + + if (response.error == Coordination::Error::ZNONODE) + { + LOG_TRACE(log, "Failed to get {}", description); + return std::nullopt; + } + else if (response.error != Coordination::Error::ZOK) + { + throw Exception(response.error, "Failed to get {}", description); + } + + return std::move(response.data); +} + +void ZooKeeper::initFeatureFlags() +{ + if (auto feature_flags = tryGetSystemZnode(keeper_api_feature_flags_path, "feature flags"); feature_flags.has_value()) { keeper_feature_flags.setFeatureFlags(std::move(*feature_flags)); return; } - auto keeper_api_version_string = try_get(keeper_api_version_path, "API version"); + auto keeper_api_version_string = tryGetSystemZnode(keeper_api_version_path, "API version"); DB::KeeperApiVersion keeper_api_version{DB::KeeperApiVersion::ZOOKEEPER_COMPATIBLE}; @@ -1256,6 +1282,17 @@ void ZooKeeper::initFeatureFlags() keeper_feature_flags.fromApiVersion(keeper_api_version); } +String ZooKeeper::tryGetAvailabilityZone() +{ + auto res = tryGetSystemZnode(keeper_availability_zone_path, "availability zone"); + if (res) + { + LOG_TRACE(log, "Availability zone for ZooKeeper at {}: {}", getConnectedHostPort(), *res); + return *res; + } + return ""; +} + void ZooKeeper::executeGenericRequest( const ZooKeeperRequestPtr & request, @@ -1587,17 +1624,6 @@ void ZooKeeper::setupFaultDistributions() inject_setup.test_and_set(); } -void ZooKeeper::checkSessionDeadline() const -{ - if (unlikely(hasReachedDeadline())) - throw Exception::fromMessage(Error::ZSESSIONEXPIRED, "Session expired (force expiry client-side)"); -} - -bool ZooKeeper::hasReachedDeadline() const -{ - return client_session_deadline.has_value() && clock::now() >= client_session_deadline.value(); -} - void ZooKeeper::maybeInjectSendFault() { if (unlikely(inject_setup.test() && send_inject_fault && send_inject_fault.value()(thread_local_rng))) diff --git a/src/Common/ZooKeeper/ZooKeeperImpl.h b/src/Common/ZooKeeper/ZooKeeperImpl.h index 8fdf0f97d9d..0c88c35b381 100644 --- a/src/Common/ZooKeeper/ZooKeeperImpl.h +++ b/src/Common/ZooKeeper/ZooKeeperImpl.h @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -102,21 +103,12 @@ using namespace DB; class ZooKeeper final : public IKeeper { public: - struct Node - { - Poco::Net::SocketAddress address; - UInt8 original_index; - bool secure; - }; - - using Nodes = std::vector; - /** Connection to nodes is performed in order. If you want, shuffle them manually. * Operation timeout couldn't be greater than session timeout. * Operation timeout applies independently for network read, network write, waiting for events and synchronization. */ ZooKeeper( - const Nodes & nodes, + const zkutil::ShuffleHosts & nodes, const zkutil::ZooKeeperArgs & args_, std::shared_ptr zk_log_); @@ -130,9 +122,7 @@ public: String getConnectedHostPort() const override { return (original_index == -1) ? "" : args.hosts[original_index]; } int32_t getConnectionXid() const override { return next_xid.load(); } - /// A ZooKeeper session can have an optional deadline set on it. - /// After it has been reached, the session needs to be finalized. - bool hasReachedDeadline() const override; + String tryGetAvailabilityZone() override; /// Useful to check owner of ephemeral node. int64_t getSessionID() const override { return session_id; } @@ -271,7 +261,6 @@ private: clock::time_point time; }; - std::optional client_session_deadline {}; using RequestsQueue = ConcurrentBoundedQueue; RequestsQueue requests_queue{1024}; @@ -316,7 +305,7 @@ private: LoggerPtr log; void connect( - const Nodes & node, + const zkutil::ShuffleHosts & node, Poco::Timespan connection_timeout); void sendHandshake(); @@ -346,9 +335,10 @@ private: void logOperationIfNeeded(const ZooKeeperRequestPtr & request, const ZooKeeperResponsePtr & response = nullptr, bool finalize = false, UInt64 elapsed_microseconds = 0); + std::optional tryGetSystemZnode(const std::string & path, const std::string & description); + void initFeatureFlags(); - void checkSessionDeadline() const; CurrentMetrics::Increment active_session_metric_increment{CurrentMetrics::ZooKeeperSession}; std::shared_ptr zk_log; diff --git a/src/Common/ZooKeeper/examples/CMakeLists.txt b/src/Common/ZooKeeper/examples/CMakeLists.txt index 678b302a512..11669d765f7 100644 --- a/src/Common/ZooKeeper/examples/CMakeLists.txt +++ b/src/Common/ZooKeeper/examples/CMakeLists.txt @@ -1,15 +1,18 @@ clickhouse_add_executable(zkutil_test_commands zkutil_test_commands.cpp) target_link_libraries(zkutil_test_commands PRIVATE clickhouse_common_zookeeper_no_log + clickhouse_functions dbms) clickhouse_add_executable(zkutil_test_commands_new_lib zkutil_test_commands_new_lib.cpp) target_link_libraries(zkutil_test_commands_new_lib PRIVATE clickhouse_common_zookeeper_no_log clickhouse_compression + clickhouse_functions dbms) clickhouse_add_executable(zkutil_test_async zkutil_test_async.cpp) target_link_libraries(zkutil_test_async PRIVATE clickhouse_common_zookeeper_no_log + clickhouse_functions dbms) diff --git a/src/Common/ZooKeeper/examples/zkutil_test_commands_new_lib.cpp b/src/Common/ZooKeeper/examples/zkutil_test_commands_new_lib.cpp index 25d66b94b46..b3a1564b8ab 100644 --- a/src/Common/ZooKeeper/examples/zkutil_test_commands_new_lib.cpp +++ b/src/Common/ZooKeeper/examples/zkutil_test_commands_new_lib.cpp @@ -25,24 +25,24 @@ try Poco::Logger::root().setChannel(channel); Poco::Logger::root().setLevel("trace"); - std::string hosts_arg = argv[1]; - std::vector hosts_strings; - splitInto<','>(hosts_strings, hosts_arg); - ZooKeeper::Nodes nodes; - nodes.reserve(hosts_strings.size()); - for (size_t i = 0; i < hosts_strings.size(); ++i) + zkutil::ZooKeeperArgs args{argv[1]}; + zkutil::ShuffleHosts nodes; + nodes.reserve(args.hosts.size()); + for (size_t i = 0; i < args.hosts.size(); ++i) { - std::string host_string = hosts_strings[i]; - bool secure = startsWith(host_string, "secure://"); + zkutil::ShuffleHost node; + std::string host_string = args.hosts[i]; + node.secure = startsWith(host_string, "secure://"); - if (secure) + if (node.secure) host_string.erase(0, strlen("secure://")); - nodes.emplace_back(ZooKeeper::Node{Poco::Net::SocketAddress{host_string}, static_cast(i) , secure}); + node.host = host_string; + node.original_index = i; + + nodes.emplace_back(node); } - - zkutil::ZooKeeperArgs args; ZooKeeper zk(nodes, args, nullptr); Poco::Event event(true); diff --git a/src/Common/assert_cast.h b/src/Common/assert_cast.h index 0b73ba1cc12..7a04372ffad 100644 --- a/src/Common/assert_cast.h +++ b/src/Common/assert_cast.h @@ -25,7 +25,7 @@ namespace DB template inline To assert_cast(From && from) { -#ifndef NDEBUG +#ifdef DEBUG_OR_SANITIZER_BUILD try { if constexpr (std::is_pointer_v) diff --git a/src/Common/config.h.in b/src/Common/config.h.in index 066314dfba0..a8446dd3e2f 100644 --- a/src/Common/config.h.in +++ b/src/Common/config.h.in @@ -63,6 +63,7 @@ #cmakedefine01 USE_BCRYPT #cmakedefine01 USE_LIBARCHIVE #cmakedefine01 USE_POCKETFFT +#cmakedefine01 USE_PROMETHEUS_PROTOBUFS #cmakedefine01 USE_MONGODB /// This is needed for .incbin in assembly. For some reason, include paths don't work there in presence of LTO. diff --git a/src/Common/examples/CMakeLists.txt b/src/Common/examples/CMakeLists.txt index 73e1396fb35..69580d4ad0e 100644 --- a/src/Common/examples/CMakeLists.txt +++ b/src/Common/examples/CMakeLists.txt @@ -1,29 +1,29 @@ clickhouse_add_executable (hashes_test hashes_test.cpp) -target_link_libraries (hashes_test PRIVATE clickhouse_common_io ch_contrib::cityhash) +target_link_libraries (hashes_test PRIVATE clickhouse_common_io clickhouse_common_config ch_contrib::cityhash) if (TARGET OpenSSL::Crypto) target_link_libraries (hashes_test PRIVATE OpenSSL::Crypto) endif() clickhouse_add_executable (sip_hash_perf sip_hash_perf.cpp) -target_link_libraries (sip_hash_perf PRIVATE clickhouse_common_io) +target_link_libraries (sip_hash_perf PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (small_table small_table.cpp) -target_link_libraries (small_table PRIVATE clickhouse_common_io) +target_link_libraries (small_table PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (parallel_aggregation parallel_aggregation.cpp) -target_link_libraries (parallel_aggregation PRIVATE dbms) +target_link_libraries (parallel_aggregation PRIVATE dbms clickhouse_functions) clickhouse_add_executable (parallel_aggregation2 parallel_aggregation2.cpp) -target_link_libraries (parallel_aggregation2 PRIVATE dbms) +target_link_libraries (parallel_aggregation2 PRIVATE dbms clickhouse_functions) clickhouse_add_executable (int_hashes_perf int_hashes_perf.cpp) -target_link_libraries (int_hashes_perf PRIVATE clickhouse_common_io) +target_link_libraries (int_hashes_perf PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (compact_array compact_array.cpp) -target_link_libraries (compact_array PRIVATE clickhouse_common_io) +target_link_libraries (compact_array PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (radix_sort radix_sort.cpp) -target_link_libraries (radix_sort PRIVATE clickhouse_common_io ch_contrib::pdqsort) +target_link_libraries (radix_sort PRIVATE clickhouse_common_io clickhouse_common_config ch_contrib::pdqsort) clickhouse_add_executable (arena_with_free_lists arena_with_free_lists.cpp) target_link_libraries (arena_with_free_lists PRIVATE dbms) @@ -31,46 +31,48 @@ target_link_libraries (arena_with_free_lists PRIVATE dbms) clickhouse_add_executable (lru_hash_map_perf lru_hash_map_perf.cpp) target_link_libraries (lru_hash_map_perf PRIVATE dbms) -clickhouse_add_executable (thread_creation_latency thread_creation_latency.cpp) -target_link_libraries (thread_creation_latency PRIVATE clickhouse_common_io) +if (OS_LINUX) + clickhouse_add_executable (thread_creation_latency thread_creation_latency.cpp) + target_link_libraries (thread_creation_latency PRIVATE clickhouse_common_io clickhouse_common_config) +endif() clickhouse_add_executable (array_cache array_cache.cpp) -target_link_libraries (array_cache PRIVATE clickhouse_common_io) +target_link_libraries (array_cache PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (space_saving space_saving.cpp) -target_link_libraries (space_saving PRIVATE clickhouse_common_io) +target_link_libraries (space_saving PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (integer_hash_tables_benchmark integer_hash_tables_benchmark.cpp) target_link_libraries (integer_hash_tables_benchmark PRIVATE dbms ch_contrib::abseil_swiss_tables ch_contrib::sparsehash) clickhouse_add_executable (cow_columns cow_columns.cpp) -target_link_libraries (cow_columns PRIVATE clickhouse_common_io) +target_link_libraries (cow_columns PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (cow_compositions cow_compositions.cpp) -target_link_libraries (cow_compositions PRIVATE clickhouse_common_io) +target_link_libraries (cow_compositions PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (stopwatch stopwatch.cpp) -target_link_libraries (stopwatch PRIVATE clickhouse_common_io) +target_link_libraries (stopwatch PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (symbol_index symbol_index.cpp) -target_link_libraries (symbol_index PRIVATE clickhouse_common_io) +target_link_libraries (symbol_index PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (chaos_sanitizer chaos_sanitizer.cpp) -target_link_libraries (chaos_sanitizer PRIVATE clickhouse_common_io) +target_link_libraries (chaos_sanitizer PRIVATE clickhouse_common_io clickhouse_common_config) if (OS_LINUX) clickhouse_add_executable (memory_statistics_os_perf memory_statistics_os_perf.cpp) - target_link_libraries (memory_statistics_os_perf PRIVATE clickhouse_common_io) + target_link_libraries (memory_statistics_os_perf PRIVATE clickhouse_common_io clickhouse_common_config) endif() clickhouse_add_executable (procfs_metrics_provider_perf procfs_metrics_provider_perf.cpp) -target_link_libraries (procfs_metrics_provider_perf PRIVATE clickhouse_common_io) +target_link_libraries (procfs_metrics_provider_perf PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (average average.cpp) -target_link_libraries (average PRIVATE clickhouse_common_io) +target_link_libraries (average PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (shell_command_inout shell_command_inout.cpp) -target_link_libraries (shell_command_inout PRIVATE clickhouse_common_io) +target_link_libraries (shell_command_inout PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (executable_udf executable_udf.cpp) target_link_libraries (executable_udf PRIVATE dbms) @@ -85,8 +87,8 @@ target_link_libraries (interval_tree PRIVATE dbms) if (ENABLE_SSL) clickhouse_add_executable (encrypt_decrypt encrypt_decrypt.cpp) - target_link_libraries (encrypt_decrypt PRIVATE dbms) + target_link_libraries (encrypt_decrypt PRIVATE dbms clickhouse_functions) endif() clickhouse_add_executable (check_pointer_valid check_pointer_valid.cpp) -target_link_libraries (check_pointer_valid PRIVATE clickhouse_common_io) +target_link_libraries (check_pointer_valid PRIVATE clickhouse_common_io clickhouse_common_config) diff --git a/src/Common/formatIPv6.h b/src/Common/formatIPv6.h index bb83e0381ef..abeda95ed0d 100644 --- a/src/Common/formatIPv6.h +++ b/src/Common/formatIPv6.h @@ -1,6 +1,7 @@ #pragma once #include +#include #include #include #include diff --git a/src/Common/getNumberOfPhysicalCPUCores.cpp b/src/Common/getNumberOfPhysicalCPUCores.cpp index 7e18a93e6ed..34a1add2f0e 100644 --- a/src/Common/getNumberOfPhysicalCPUCores.cpp +++ b/src/Common/getNumberOfPhysicalCPUCores.cpp @@ -37,12 +37,12 @@ uint32_t getCGroupLimitedCPUCores(unsigned default_cpu_count) /// cgroupsv2 if (cgroupsV2Enabled()) { - /// First, we identify the cgroup the process belongs - std::string cgroup = cgroupV2OfProcess(); - if (cgroup.empty()) + /// First, we identify the path of the cgroup the process belongs + std::filesystem::path cgroup_path = cgroupV2PathOfProcess(); + if (cgroup_path.empty()) return default_cpu_count; - auto current_cgroup = cgroup.empty() ? default_cgroups_mount : (default_cgroups_mount / cgroup); + auto current_cgroup = cgroup_path; // Looking for cpu.max in directories from the current cgroup to the top level // It does not stop on the first time since the child could have a greater value than parent @@ -62,7 +62,7 @@ uint32_t getCGroupLimitedCPUCores(unsigned default_cpu_count) } current_cgroup = current_cgroup.parent_path(); } - current_cgroup = default_cgroups_mount / cgroup; + current_cgroup = cgroup_path; // Looking for cpuset.cpus.effective in directories from the current cgroup to the top level while (current_cgroup != default_cgroups_mount.parent_path()) { diff --git a/src/Common/getRandomASCIIString.cpp b/src/Common/getRandomASCIIString.cpp index 594b4cd3228..a295277b453 100644 --- a/src/Common/getRandomASCIIString.cpp +++ b/src/Common/getRandomASCIIString.cpp @@ -6,12 +6,17 @@ namespace DB { String getRandomASCIIString(size_t length) +{ + return getRandomASCIIString(length, thread_local_rng); +} + +String getRandomASCIIString(size_t length, pcg64 & rng) { std::uniform_int_distribution distribution('a', 'z'); String res; res.resize(length); for (auto & c : res) - c = distribution(thread_local_rng); + c = distribution(rng); return res; } diff --git a/src/Common/getRandomASCIIString.h b/src/Common/getRandomASCIIString.h index 627d2700ce3..19e1ff7120e 100644 --- a/src/Common/getRandomASCIIString.h +++ b/src/Common/getRandomASCIIString.h @@ -2,11 +2,14 @@ #include +#include + namespace DB { /// Slow random string. Useful for random names and things like this. Not for generating data. String getRandomASCIIString(size_t length); +String getRandomASCIIString(size_t length, pcg64 & rng); } diff --git a/src/Common/memory.h b/src/Common/memory.h index caa0418fa56..d673f954fb2 100644 --- a/src/Common/memory.h +++ b/src/Common/memory.h @@ -37,7 +37,7 @@ requires DB::OptionalArgument inline ALWAYS_INLINE void * newImpl(std::size_t size, TAlign... align) { #if USE_GWP_ASAN - if (unlikely(GWPAsan::GuardedAlloc.shouldSample())) + if (unlikely(GWPAsan::shouldSample())) { if constexpr (sizeof...(TAlign) == 1) { @@ -80,10 +80,10 @@ inline ALWAYS_INLINE void * newImpl(std::size_t size, TAlign... align) throw std::bad_alloc{}; } -inline ALWAYS_INLINE void * newNoExept(std::size_t size) noexcept +inline ALWAYS_INLINE void * newNoExcept(std::size_t size) noexcept { #if USE_GWP_ASAN - if (unlikely(GWPAsan::GuardedAlloc.shouldSample())) + if (unlikely(GWPAsan::shouldSample())) { if (void * ptr = GWPAsan::GuardedAlloc.allocate(size)) { @@ -99,10 +99,10 @@ inline ALWAYS_INLINE void * newNoExept(std::size_t size) noexcept return malloc(size); } -inline ALWAYS_INLINE void * newNoExept(std::size_t size, std::align_val_t align) noexcept +inline ALWAYS_INLINE void * newNoExcept(std::size_t size, std::align_val_t align) noexcept { #if USE_GWP_ASAN - if (unlikely(GWPAsan::GuardedAlloc.shouldSample())) + if (unlikely(GWPAsan::shouldSample())) { if (void * ptr = GWPAsan::GuardedAlloc.allocate(size, alignToSizeT(align))) { diff --git a/src/Common/mysqlxx/Pool.cpp b/src/Common/mysqlxx/Pool.cpp index cc5b18214c8..546e9e91dc7 100644 --- a/src/Common/mysqlxx/Pool.cpp +++ b/src/Common/mysqlxx/Pool.cpp @@ -228,7 +228,6 @@ Pool::Entry Pool::tryGet() for (auto connection_it = connections.cbegin(); connection_it != connections.cend();) { Connection * connection_ptr = *connection_it; - /// Fixme: There is a race condition here b/c we do not synchronize with Pool::Entry's copy-assignment operator if (connection_ptr->ref_count == 0) { { diff --git a/src/Common/mysqlxx/mysqlxx/Pool.h b/src/Common/mysqlxx/mysqlxx/Pool.h index 6e509d8bdd6..f1ef81e28dd 100644 --- a/src/Common/mysqlxx/mysqlxx/Pool.h +++ b/src/Common/mysqlxx/mysqlxx/Pool.h @@ -64,17 +64,6 @@ public: decrementRefCount(); } - Entry & operator= (const Entry & src) /// NOLINT - { - pool = src.pool; - if (data) - decrementRefCount(); - data = src.data; - if (data) - incrementRefCount(); - return * this; - } - bool isNull() const { return data == nullptr; diff --git a/src/Common/mysqlxx/tests/CMakeLists.txt b/src/Common/mysqlxx/tests/CMakeLists.txt index c560d0e3153..53bee778470 100644 --- a/src/Common/mysqlxx/tests/CMakeLists.txt +++ b/src/Common/mysqlxx/tests/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable (mysqlxx_pool_test mysqlxx_pool_test.cpp) -target_link_libraries (mysqlxx_pool_test PRIVATE mysqlxx) +target_link_libraries (mysqlxx_pool_test PRIVATE mysqlxx clickhouse_common_config loggers_no_text_log) diff --git a/src/Common/mysqlxx/tests/mysqlxx_pool_test.cpp b/src/Common/mysqlxx/tests/mysqlxx_pool_test.cpp index 61d6a117285..121767edc84 100644 --- a/src/Common/mysqlxx/tests/mysqlxx_pool_test.cpp +++ b/src/Common/mysqlxx/tests/mysqlxx_pool_test.cpp @@ -13,13 +13,11 @@ mysqlxx::Pool::Entry getWithFailover(mysqlxx::Pool & connections_pool) constexpr size_t max_tries = 3; - mysqlxx::Pool::Entry worker_connection; - for (size_t try_no = 1; try_no <= max_tries; ++try_no) { try { - worker_connection = connections_pool.tryGet(); + mysqlxx::Pool::Entry worker_connection = connections_pool.tryGet(); if (!worker_connection.isNull()) { diff --git a/src/Common/new_delete.cpp b/src/Common/new_delete.cpp index e8151fbe201..80e05fc4ea0 100644 --- a/src/Common/new_delete.cpp +++ b/src/Common/new_delete.cpp @@ -87,7 +87,7 @@ void * operator new(std::size_t size, const std::nothrow_t &) noexcept { AllocationTrace trace; std::size_t actual_size = Memory::trackMemory(size, trace); - void * ptr = Memory::newNoExept(size); + void * ptr = Memory::newNoExcept(size); trace.onAlloc(ptr, actual_size); return ptr; } @@ -96,7 +96,7 @@ void * operator new[](std::size_t size, const std::nothrow_t &) noexcept { AllocationTrace trace; std::size_t actual_size = Memory::trackMemory(size, trace); - void * ptr = Memory::newNoExept(size); + void * ptr = Memory::newNoExcept(size); trace.onAlloc(ptr, actual_size); return ptr; } @@ -105,7 +105,7 @@ void * operator new(std::size_t size, std::align_val_t align, const std::nothrow { AllocationTrace trace; std::size_t actual_size = Memory::trackMemory(size, trace, align); - void * ptr = Memory::newNoExept(size, align); + void * ptr = Memory::newNoExcept(size, align); trace.onAlloc(ptr, actual_size); return ptr; } @@ -114,7 +114,7 @@ void * operator new[](std::size_t size, std::align_val_t align, const std::nothr { AllocationTrace trace; std::size_t actual_size = Memory::trackMemory(size, trace, align); - void * ptr = Memory::newNoExept(size, align); + void * ptr = Memory::newNoExcept(size, align); trace.onAlloc(ptr, actual_size); return ptr; } diff --git a/src/Common/tests/gtest_cgroups_reader.cpp b/src/Common/tests/gtest_cgroups_reader.cpp new file mode 100644 index 00000000000..2de25bb42ce --- /dev/null +++ b/src/Common/tests/gtest_cgroups_reader.cpp @@ -0,0 +1,178 @@ +#if defined(OS_LINUX) + +#include + +#include +#include + +#include +#include +#include + +using namespace DB; + + +const std::string SAMPLE_FILE[2] = { + R"(cache 4673703936 +rss 2232029184 +rss_huge 0 +shmem 0 +mapped_file 344678400 +dirty 4730880 +writeback 135168 +swap 0 +pgpgin 2038569918 +pgpgout 2036883790 +pgfault 2055373287 +pgmajfault 0 +inactive_anon 2156335104 +active_anon 0 +inactive_file 2841305088 +active_file 1653915648 +unevictable 256008192 +hierarchical_memory_limit 8589934592 +hierarchical_memsw_limit 8589934592 +total_cache 4673703936 +total_rss 2232029184 +total_rss_huge 0 +total_shmem 0 +total_mapped_file 344678400 +total_dirty 4730880 +total_writeback 135168 +total_swap 0 +total_pgpgin 2038569918 +total_pgpgout 2036883790 +total_pgfault 2055373287 +total_pgmajfault 0 +total_inactive_anon 2156335104 +total_active_anon 0 +total_inactive_file 2841305088 +total_active_file 1653915648 +total_unevictable 256008192 +)", + R"(anon 10429399040 +file 17410793472 +kernel 1537789952 +kernel_stack 3833856 +pagetables 65441792 +sec_pagetables 0 +percpu 15232 +sock 0 +vmalloc 0 +shmem 0 +zswap 0 +zswapped 0 +file_mapped 344010752 +file_dirty 2060857344 +file_writeback 0 +swapcached 0 +anon_thp 0 +file_thp 0 +shmem_thp 0 +inactive_anon 0 +active_anon 10429370368 +inactive_file 8693084160 +active_file 8717561856 +unevictable 0 +slab_reclaimable 1460982504 +slab_unreclaimable 5152864 +slab 1466135368 +workingset_refault_anon 0 +workingset_refault_file 0 +workingset_activate_anon 0 +workingset_activate_file 0 +workingset_restore_anon 0 +workingset_restore_file 0 +workingset_nodereclaim 0 +pgscan 0 +pgsteal 0 +pgscan_kswapd 0 +pgscan_direct 0 +pgscan_khugepaged 0 +pgsteal_kswapd 0 +pgsteal_direct 0 +pgsteal_khugepaged 0 +pgfault 43026352 +pgmajfault 36762 +pgrefill 0 +pgactivate 0 +pgdeactivate 0 +pglazyfree 259 +pglazyfreed 0 +zswpin 0 +zswpout 0 +thp_fault_alloc 0 +thp_collapse_alloc 0 +)"}; + +const std::string EXPECTED[2] + = {"{\"active_anon\": 0, \"active_file\": 1653915648, \"cache\": 4673703936, \"dirty\": 4730880, \"hierarchical_memory_limit\": " + "8589934592, \"hierarchical_memsw_limit\": 8589934592, \"inactive_anon\": 2156335104, \"inactive_file\": 2841305088, " + "\"mapped_file\": 344678400, \"pgfault\": 2055373287, \"pgmajfault\": 0, \"pgpgin\": 2038569918, \"pgpgout\": 2036883790, \"rss\": " + "2232029184, \"rss_huge\": 0, \"shmem\": 0, \"swap\": 0, \"total_active_anon\": 0, \"total_active_file\": 1653915648, " + "\"total_cache\": 4673703936, \"total_dirty\": 4730880, \"total_inactive_anon\": 2156335104, \"total_inactive_file\": 2841305088, " + "\"total_mapped_file\": 344678400, \"total_pgfault\": 2055373287, \"total_pgmajfault\": 0, \"total_pgpgin\": 2038569918, " + "\"total_pgpgout\": 2036883790, \"total_rss\": 2232029184, \"total_rss_huge\": 0, \"total_shmem\": 0, \"total_swap\": 0, " + "\"total_unevictable\": 256008192, \"total_writeback\": 135168, \"unevictable\": 256008192, \"writeback\": 135168}", + "{\"active_anon\": 10429370368, \"active_file\": 8717561856, \"anon\": 10429399040, \"anon_thp\": 0, \"file\": 17410793472, " + "\"file_dirty\": 2060857344, \"file_mapped\": 344010752, \"file_thp\": 0, \"file_writeback\": 0, \"inactive_anon\": 0, " + "\"inactive_file\": 8693084160, \"kernel\": 1537789952, \"kernel_stack\": 3833856, \"pagetables\": 65441792, \"percpu\": 15232, " + "\"pgactivate\": 0, \"pgdeactivate\": 0, \"pgfault\": 43026352, \"pglazyfree\": 259, \"pglazyfreed\": 0, \"pgmajfault\": 36762, " + "\"pgrefill\": 0, \"pgscan\": 0, \"pgscan_direct\": 0, \"pgscan_khugepaged\": 0, \"pgscan_kswapd\": 0, \"pgsteal\": 0, " + "\"pgsteal_direct\": 0, \"pgsteal_khugepaged\": 0, \"pgsteal_kswapd\": 0, \"sec_pagetables\": 0, \"shmem\": 0, \"shmem_thp\": 0, " + "\"slab\": 1466135368, \"slab_reclaimable\": 1460982504, \"slab_unreclaimable\": 5152864, \"sock\": 0, \"swapcached\": 0, " + "\"thp_collapse_alloc\": 0, \"thp_fault_alloc\": 0, \"unevictable\": 0, \"vmalloc\": 0, \"workingset_activate_anon\": 0, " + "\"workingset_activate_file\": 0, \"workingset_nodereclaim\": 0, \"workingset_refault_anon\": 0, \"workingset_refault_file\": 0, " + "\"workingset_restore_anon\": 0, \"workingset_restore_file\": 0, \"zswap\": 0, \"zswapped\": 0, \"zswpin\": 0, \"zswpout\": 0}"}; + + +class CgroupsMemoryUsageObserverFixture : public ::testing::TestWithParam +{ + void SetUp() override + { + const uint8_t version = static_cast(GetParam()); + tmp_dir = fmt::format("./test_cgroups_{}", magic_enum::enum_name(GetParam())); + fs::create_directories(tmp_dir); + + auto stat_file = WriteBufferFromFile(tmp_dir + "/memory.stat"); + stat_file.write(SAMPLE_FILE[version].data(), SAMPLE_FILE[version].size()); + stat_file.sync(); + + if (GetParam() == CgroupsMemoryUsageObserver::CgroupsVersion::V2) + { + auto current_file = WriteBufferFromFile(tmp_dir + "/memory.current"); + current_file.write("29645422592", 11); + current_file.sync(); + } + } + +protected: + std::string tmp_dir; +}; + + +TEST_P(CgroupsMemoryUsageObserverFixture, ReadMemoryUsageTest) +{ + const auto version = GetParam(); + auto reader = createCgroupsReader(version, tmp_dir); + ASSERT_EQ( + reader->readMemoryUsage(), + version == CgroupsMemoryUsageObserver::CgroupsVersion::V1 ? /* rss from memory.stat */ 2232029184 + : /* value from memory.current - inactive_file */ 20952338432); +} + + +TEST_P(CgroupsMemoryUsageObserverFixture, DumpAllStatsTest) +{ + const auto version = GetParam(); + auto reader = createCgroupsReader(version, tmp_dir); + ASSERT_EQ(reader->dumpAllStats(), EXPECTED[static_cast(version)]); +} + + +INSTANTIATE_TEST_SUITE_P( + CgroupsMemoryUsageObserverTests, + CgroupsMemoryUsageObserverFixture, + ::testing::Values(CgroupsMemoryUsageObserver::CgroupsVersion::V1, CgroupsMemoryUsageObserver::CgroupsVersion::V2)); + +#endif diff --git a/src/Common/tests/gtest_resolve_pool.cpp b/src/Common/tests/gtest_resolve_pool.cpp index b760b9b1524..c443e961cc7 100644 --- a/src/Common/tests/gtest_resolve_pool.cpp +++ b/src/Common/tests/gtest_resolve_pool.cpp @@ -33,7 +33,7 @@ size_t toMilliseconds(auto duration) return std::chrono::duration_cast(duration).count(); } -const auto epsilon = 500us; +const auto epsilon = 1ms; class ResolvePoolMock : public DB::HostResolver { @@ -358,53 +358,59 @@ void check_no_failed_address(size_t iteration, auto & resolver, auto & addresses TEST_F(ResolvePoolTest, BannedForConsiquenceFail) { - auto history = 5ms; + auto history = 10ms; auto resolver = make_resolver(toMilliseconds(history)); auto failed_addr = resolver->resolve(); ASSERT_TRUE(addresses.contains(*failed_addr)); - auto start_at = now(); failed_addr.setFail(); + auto start_at = now(); + ASSERT_EQ(3, CurrentMetrics::get(metrics.active_count)); ASSERT_EQ(1, CurrentMetrics::get(metrics.banned_count)); check_no_failed_address(1, resolver, addresses, failed_addr, metrics, start_at + history - epsilon); sleep_until(start_at + history + epsilon); - start_at = now(); resolver->update(); ASSERT_EQ(3, CurrentMetrics::get(metrics.active_count)); ASSERT_EQ(0, CurrentMetrics::get(metrics.banned_count)); failed_addr.setFail(); + start_at = now(); + check_no_failed_address(2, resolver, addresses, failed_addr, metrics, start_at + history - epsilon); sleep_until(start_at + history + epsilon); - start_at = now(); resolver->update(); + + // too much time has passed + if (now() > start_at + 2*history - epsilon) + return; + ASSERT_EQ(3, CurrentMetrics::get(metrics.active_count)); ASSERT_EQ(1, CurrentMetrics::get(metrics.banned_count)); // ip still banned adter history_ms + update, because it was his second consiquent fail - check_no_failed_address(2, resolver, addresses, failed_addr, metrics, start_at + history - epsilon); + check_no_failed_address(2, resolver, addresses, failed_addr, metrics, start_at + 2*history - epsilon); } TEST_F(ResolvePoolTest, NoAditionalBannForConcurrentFail) { - auto history = 5ms; + auto history = 10ms; auto resolver = make_resolver(toMilliseconds(history)); auto failed_addr = resolver->resolve(); ASSERT_TRUE(addresses.contains(*failed_addr)); - auto start_at = now(); + failed_addr.setFail(); + failed_addr.setFail(); + failed_addr.setFail(); - failed_addr.setFail(); - failed_addr.setFail(); - failed_addr.setFail(); + auto start_at = now(); ASSERT_EQ(3, CurrentMetrics::get(metrics.active_count)); ASSERT_EQ(1, CurrentMetrics::get(metrics.banned_count)); @@ -413,6 +419,7 @@ TEST_F(ResolvePoolTest, NoAditionalBannForConcurrentFail) sleep_until(start_at + history + epsilon); resolver->update(); + // ip is cleared after just 1 history_ms interval. ASSERT_EQ(3, CurrentMetrics::get(metrics.active_count)); ASSERT_EQ(0, CurrentMetrics::get(metrics.banned_count)); diff --git a/src/Common/tests/gtest_rw_lock.cpp b/src/Common/tests/gtest_rw_lock.cpp index d8c6e9cb99d..9b0c9aeafbe 100644 --- a/src/Common/tests/gtest_rw_lock.cpp +++ b/src/Common/tests/gtest_rw_lock.cpp @@ -166,7 +166,7 @@ TEST(Common, RWLockRecursive) auto lock2 = fifo_lock->getLock(RWLockImpl::Read, "q2"); -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD /// It throws LOGICAL_ERROR EXPECT_ANY_THROW({fifo_lock->getLock(RWLockImpl::Write, "q2");}); #endif diff --git a/src/Common/tests/gtest_shell_command.cpp b/src/Common/tests/gtest_shell_command.cpp index d6d0a544e9b..0ea96da9da2 100644 --- a/src/Common/tests/gtest_shell_command.cpp +++ b/src/Common/tests/gtest_shell_command.cpp @@ -54,16 +54,3 @@ TEST(ShellCommand, ExecuteWithInput) EXPECT_EQ(res, "Hello, world!\n"); } - -TEST(ShellCommand, AutoWait) -{ - // hunting: - for (int i = 0; i < 1000; ++i) - { - auto command = ShellCommand::execute("echo " + std::to_string(i)); - //command->wait(); // now automatic - } - - // std::cerr << "inspect me: ps auxwwf\n"; - // std::this_thread::sleep_for(std::chrono::seconds(100)); -} diff --git a/src/Compression/CompressionFactory.cpp b/src/Compression/CompressionFactory.cpp index 68e0131c91b..2e7aa0d086f 100644 --- a/src/Compression/CompressionFactory.cpp +++ b/src/Compression/CompressionFactory.cpp @@ -185,7 +185,6 @@ void registerCodecDeflateQpl(CompressionCodecFactory & factory); /// Keeper use only general-purpose codecs, so we don't need these special codecs /// in standalone build -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD void registerCodecDelta(CompressionCodecFactory & factory); void registerCodecT64(CompressionCodecFactory & factory); void registerCodecDoubleDelta(CompressionCodecFactory & factory); @@ -193,7 +192,6 @@ void registerCodecGorilla(CompressionCodecFactory & factory); void registerCodecEncrypted(CompressionCodecFactory & factory); void registerCodecFPC(CompressionCodecFactory & factory); void registerCodecGCD(CompressionCodecFactory & factory); -#endif CompressionCodecFactory::CompressionCodecFactory() { @@ -205,7 +203,6 @@ CompressionCodecFactory::CompressionCodecFactory() #endif registerCodecLZ4HC(*this); registerCodecMultiple(*this); -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD registerCodecDelta(*this); registerCodecT64(*this); registerCodecDoubleDelta(*this); @@ -216,7 +213,6 @@ CompressionCodecFactory::CompressionCodecFactory() registerCodecDeflateQpl(*this); #endif registerCodecGCD(*this); -#endif default_codec = get("LZ4", {}); } diff --git a/src/Compression/examples/CMakeLists.txt b/src/Compression/examples/CMakeLists.txt index a7cc6bebf42..fee8cf89942 100644 --- a/src/Compression/examples/CMakeLists.txt +++ b/src/Compression/examples/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable (compressed_buffer compressed_buffer.cpp) -target_link_libraries (compressed_buffer PRIVATE clickhouse_common_io clickhouse_compression) +target_link_libraries (compressed_buffer PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression) diff --git a/src/Coordination/Changelog.cpp b/src/Coordination/Changelog.cpp index ad6f95b3902..9607c345a3b 100644 --- a/src/Coordination/Changelog.cpp +++ b/src/Coordination/Changelog.cpp @@ -808,7 +808,11 @@ void LogEntryStorage::startCommitLogsPrefetch(uint64_t last_committed_index) con for (; current_index <= max_index_for_prefetch; ++current_index) { - const auto & [changelog_description, position, size] = logs_location.at(current_index); + auto location_it = logs_location.find(current_index); + if (location_it == logs_location.end()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Location of log entry with index {} is missing", current_index); + + const auto & [changelog_description, position, size] = location_it->second; if (total_size == 0) current_file_info = &file_infos.emplace_back(changelog_description, position, /* count */ 1); else if (total_size + size > commit_logs_cache.size_threshold) @@ -1416,7 +1420,11 @@ LogEntriesPtr LogEntryStorage::getLogEntriesBetween(uint64_t start, uint64_t end } else { - const auto & log_location = logs_location.at(i); + auto location_it = logs_location.find(i); + if (location_it == logs_location.end()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Location of log entry with index {} is missing", i); + + const auto & log_location = location_it->second; if (!read_info) set_new_file(log_location); diff --git a/src/Coordination/Changelog.h b/src/Coordination/Changelog.h index 2e8dbe75e90..0f833c17e1b 100644 --- a/src/Coordination/Changelog.h +++ b/src/Coordination/Changelog.h @@ -5,6 +5,8 @@ #include #include +#include +#include #include #include diff --git a/src/Coordination/CoordinationSettings.cpp b/src/Coordination/CoordinationSettings.cpp index 05f691ca76b..d72d39fd7e1 100644 --- a/src/Coordination/CoordinationSettings.cpp +++ b/src/Coordination/CoordinationSettings.cpp @@ -169,6 +169,23 @@ void KeeperConfigurationAndSettings::dump(WriteBufferFromOwnString & buf) const writeText("async_replication=", buf); write_bool(coordination_settings->async_replication); + + writeText("latest_logs_cache_size_threshold=", buf); + write_int(coordination_settings->latest_logs_cache_size_threshold); + writeText("commit_logs_cache_size_threshold=", buf); + write_int(coordination_settings->commit_logs_cache_size_threshold); + + writeText("disk_move_retries_wait_ms=", buf); + write_int(coordination_settings->disk_move_retries_wait_ms); + writeText("disk_move_retries_during_init=", buf); + write_int(coordination_settings->disk_move_retries_during_init); + + writeText("log_slow_total_threshold_ms=", buf); + write_int(coordination_settings->log_slow_total_threshold_ms); + writeText("log_slow_cpu_threshold_ms=", buf); + write_int(coordination_settings->log_slow_cpu_threshold_ms); + writeText("log_slow_connection_operation_threshold_ms=", buf); + write_int(coordination_settings->log_slow_connection_operation_threshold_ms); } KeeperConfigurationAndSettingsPtr diff --git a/src/Coordination/CoordinationSettings.h b/src/Coordination/CoordinationSettings.h index a32552616ee..35a23fc9e78 100644 --- a/src/Coordination/CoordinationSettings.h +++ b/src/Coordination/CoordinationSettings.h @@ -55,10 +55,14 @@ struct Settings; M(UInt64, min_request_size_for_cache, 50 * 1024, "Minimal size of the request to cache the deserialization result. Caching can have negative effect on latency for smaller requests, set to 0 to disable", 0) \ M(UInt64, raft_limits_reconnect_limit, 50, "If connection to a peer is silent longer than this limit * (multiplied by heartbeat interval), we re-establish the connection.", 0) \ M(Bool, async_replication, false, "Enable async replication. All write and read guarantees are preserved while better performance is achieved. Settings is disabled by default to not break backwards compatibility.", 0) \ + M(Bool, experimental_use_rocksdb, false, "Use rocksdb as backend storage", 0) \ M(UInt64, latest_logs_cache_size_threshold, 1 * 1024 * 1024 * 1024, "Maximum total size of in-memory cache of latest log entries.", 0) \ M(UInt64, commit_logs_cache_size_threshold, 500 * 1024 * 1024, "Maximum total size of in-memory cache of log entries needed next for commit.", 0) \ M(UInt64, disk_move_retries_wait_ms, 1000, "How long to wait between retries after a failure which happened while a file was being moved between disks.", 0) \ - M(UInt64, disk_move_retries_during_init, 100, "The amount of retries after a failure which happened while a file was being moved between disks during initialization.", 0) + M(UInt64, disk_move_retries_during_init, 100, "The amount of retries after a failure which happened while a file was being moved between disks during initialization.", 0) \ + M(UInt64, log_slow_total_threshold_ms, 5000, "Requests for which the total latency is larger than this settings will be logged", 0) \ + M(UInt64, log_slow_cpu_threshold_ms, 100, "Requests for which the CPU (preprocessing and processing) latency is larger than this settings will be logged", 0) \ + M(UInt64, log_slow_connection_operation_threshold_ms, 1000, "Log message if a certain operation took too long inside a single connection", 0) DECLARE_SETTINGS_TRAITS(CoordinationSettingsTraits, LIST_OF_COORDINATION_SETTINGS) diff --git a/src/Coordination/FourLetterCommand.cpp b/src/Coordination/FourLetterCommand.cpp index 8de9f8dfa1c..d0ecbe744a0 100644 --- a/src/Coordination/FourLetterCommand.cpp +++ b/src/Coordination/FourLetterCommand.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include diff --git a/src/Coordination/FourLetterCommand.h b/src/Coordination/FourLetterCommand.h index 82b30a0b5f6..e3289982b0d 100644 --- a/src/Coordination/FourLetterCommand.h +++ b/src/Coordination/FourLetterCommand.h @@ -2,8 +2,11 @@ #include "config.h" +#include +#include #include #include +#include #include namespace DB diff --git a/src/Coordination/KeeperConstants.cpp b/src/Coordination/KeeperConstants.cpp index 51bf037c1c9..daa6a4bcb3f 100644 --- a/src/Coordination/KeeperConstants.cpp +++ b/src/Coordination/KeeperConstants.cpp @@ -150,12 +150,18 @@ M(S3PutObject) \ M(S3GetObject) \ \ - M(AzureUploadPart) \ - M(DiskAzureUploadPart) \ + M(AzureUpload) \ + M(DiskAzureUpload) \ + M(AzureStageBlock) \ + M(DiskAzureStageBlock) \ + M(AzureCommitBlockList) \ + M(DiskAzureCommitBlockList) \ M(AzureCopyObject) \ M(DiskAzureCopyObject) \ M(AzureDeleteObjects) \ + M(DiskAzureDeleteObjects) \ M(AzureListObjects) \ + M(DiskAzureListObjects) \ \ M(DiskS3DeleteObjects) \ M(DiskS3CopyObject) \ @@ -177,8 +183,6 @@ M(ReadBufferFromS3InitMicroseconds) \ M(ReadBufferFromS3Bytes) \ M(ReadBufferFromS3RequestsErrors) \ - M(ReadBufferFromS3ResetSessions) \ - M(ReadBufferFromS3PreservedSessions) \ \ M(WriteBufferFromS3Microseconds) \ M(WriteBufferFromS3Bytes) \ @@ -238,6 +242,13 @@ M(KeeperPacketsReceived) \ M(KeeperRequestTotal) \ M(KeeperLatency) \ + M(KeeperTotalElapsedMicroseconds) \ + M(KeeperProcessElapsedMicroseconds) \ + M(KeeperPreprocessElapsedMicroseconds) \ + M(KeeperStorageLockWaitMicroseconds) \ + M(KeeperCommitWaitElapsedMicroseconds) \ + M(KeeperBatchMaxCount) \ + M(KeeperBatchMaxTotalSize) \ M(KeeperCommits) \ M(KeeperCommitsFailed) \ M(KeeperSnapshotCreations) \ @@ -359,7 +370,7 @@ extern const std::vector keeper_profile_events M(AsynchronousReadWait) \ M(S3Requests) \ M(KeeperAliveConnections) \ - M(KeeperOutstandingRequets) \ + M(KeeperOutstandingRequests) \ M(ThreadsInOvercommitTracker) \ M(IOUringPendingEvents) \ M(IOUringInFlightEvents) \ diff --git a/src/Coordination/KeeperContext.cpp b/src/Coordination/KeeperContext.cpp index a36a074ce89..1f66882ecad 100644 --- a/src/Coordination/KeeperContext.cpp +++ b/src/Coordination/KeeperContext.cpp @@ -3,19 +3,29 @@ #include +#include #include +#include +#include +#include +#include +#include +#include #include #include #include -#include #include -#include -#include -#include #include #include +#include "config.h" +#if USE_ROCKSDB +#include +#include +#include +#endif + namespace DB { @@ -23,6 +33,8 @@ namespace ErrorCodes { extern const int BAD_ARGUMENTS; +extern const int LOGICAL_ERROR; +extern const int ROCKSDB_ERROR; } @@ -40,6 +52,95 @@ KeeperContext::KeeperContext(bool standalone_keeper_, CoordinationSettingsPtr co system_nodes_with_data[keeper_api_version_path] = toString(static_cast(KeeperApiVersion::WITH_MULTI_READ)); } +#if USE_ROCKSDB +using RocksDBOptions = std::unordered_map; + +static RocksDBOptions getOptionsFromConfig(const Poco::Util::AbstractConfiguration & config, const std::string & path) +{ + RocksDBOptions options; + + Poco::Util::AbstractConfiguration::Keys keys; + config.keys(path, keys); + + for (const auto & key : keys) + { + const String key_path = path + "." + key; + options[key] = config.getString(key_path); + } + + return options; +} + +static rocksdb::Options getRocksDBOptionsFromConfig(const Poco::Util::AbstractConfiguration & config) +{ + rocksdb::Status status; + rocksdb::Options base; + + base.create_if_missing = true; + base.compression = rocksdb::CompressionType::kZSTD; + base.statistics = rocksdb::CreateDBStatistics(); + /// It is too verbose by default, and in fact we don't care about rocksdb logs at all. + base.info_log_level = rocksdb::ERROR_LEVEL; + + rocksdb::Options merged = base; + rocksdb::BlockBasedTableOptions table_options; + + if (config.has("keeper_server.rocksdb.options")) + { + auto config_options = getOptionsFromConfig(config, "keeper_server.rocksdb.options"); + status = rocksdb::GetDBOptionsFromMap(merged, config_options, &merged); + if (!status.ok()) + { + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Fail to merge rocksdb options from 'rocksdb.options' : {}", + status.ToString()); + } + } + if (config.has("rocksdb.column_family_options")) + { + auto column_family_options = getOptionsFromConfig(config, "rocksdb.column_family_options"); + status = rocksdb::GetColumnFamilyOptionsFromMap(merged, column_family_options, &merged); + if (!status.ok()) + { + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Fail to merge rocksdb options from 'rocksdb.column_family_options' at: {}", status.ToString()); + } + } + if (config.has("rocksdb.block_based_table_options")) + { + auto block_based_table_options = getOptionsFromConfig(config, "rocksdb.block_based_table_options"); + status = rocksdb::GetBlockBasedTableOptionsFromMap(table_options, block_based_table_options, &table_options); + if (!status.ok()) + { + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Fail to merge rocksdb options from 'rocksdb.block_based_table_options' at: {}", status.ToString()); + } + } + + merged.table_factory.reset(rocksdb::NewBlockBasedTableFactory(table_options)); + return merged; +} +#endif + +KeeperContext::Storage KeeperContext::getRocksDBPathFromConfig(const Poco::Util::AbstractConfiguration & config) const +{ + const auto create_local_disk = [](const auto & path) + { + if (fs::exists(path)) + fs::remove_all(path); + fs::create_directories(path); + + return std::make_shared("LocalRocksDBDisk", path); + }; + if (config.has("keeper_server.rocksdb_path")) + return create_local_disk(config.getString("keeper_server.rocksdb_path")); + + if (config.has("keeper_server.storage_path")) + return create_local_disk(std::filesystem::path{config.getString("keeper_server.storage_path")} / "rocksdb"); + + if (standalone_keeper) + return create_local_disk(std::filesystem::path{config.getString("path", KEEPER_DEFAULT_PATH)} / "rocksdb"); + else + return create_local_disk(std::filesystem::path{config.getString("path", DBMS_DEFAULT_PATH)} / "coordination/rocksdb"); +} + void KeeperContext::initialize(const Poco::Util::AbstractConfiguration & config, KeeperDispatcher * dispatcher_) { dispatcher = dispatcher_; @@ -58,6 +159,14 @@ void KeeperContext::initialize(const Poco::Util::AbstractConfiguration & config, initializeFeatureFlags(config); initializeDisks(config); + + #if USE_ROCKSDB + if (config.getBool("keeper_server.coordination_settings.experimental_use_rocksdb", false)) + { + rocksdb_options = std::make_shared(getRocksDBOptionsFromConfig(config)); + digest_enabled = false; /// TODO: support digest + } + #endif } namespace @@ -93,6 +202,8 @@ void KeeperContext::initializeDisks(const Poco::Util::AbstractConfiguration & co { disk_selector->initialize(config, "storage_configuration.disks", Context::getGlobalContextInstance(), diskValidator); + rocksdb_storage = getRocksDBPathFromConfig(config); + log_storage = getLogsPathFromConfig(config); if (config.has("keeper_server.latest_log_storage_disk")) @@ -261,6 +372,37 @@ void KeeperContext::dumpConfiguration(WriteBufferFromOwnString & buf) const } } + +void KeeperContext::setRocksDBDisk(DiskPtr disk) +{ + rocksdb_storage = std::move(disk); +} + +DiskPtr KeeperContext::getTemporaryRocksDBDisk() const +{ + DiskPtr rocksdb_disk = getDisk(rocksdb_storage); + if (!rocksdb_disk) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "rocksdb storage is not initialized"); + } + auto uuid_str = formatUUID(UUIDHelpers::generateV4()); + String path_to_create = "rocks_" + std::string(uuid_str.data(), uuid_str.size()); + rocksdb_disk->createDirectory(path_to_create); + return std::make_shared("LocalTmpRocksDBDisk", fullPath(rocksdb_disk, path_to_create)); +} + +void KeeperContext::setRocksDBOptions(std::shared_ptr rocksdb_options_) +{ + if (rocksdb_options_ != nullptr) + rocksdb_options = rocksdb_options_; + else + { + #if USE_ROCKSDB + rocksdb_options = std::make_shared(getRocksDBOptionsFromConfig(Poco::Util::JSONConfiguration())); + #endif + } +} + KeeperContext::Storage KeeperContext::getLogsPathFromConfig(const Poco::Util::AbstractConfiguration & config) const { const auto create_local_disk = [](const auto & path) diff --git a/src/Coordination/KeeperContext.h b/src/Coordination/KeeperContext.h index e283e65dffa..38013725f56 100644 --- a/src/Coordination/KeeperContext.h +++ b/src/Coordination/KeeperContext.h @@ -6,6 +6,11 @@ #include #include +namespace rocksdb +{ +struct Options; +} + namespace DB { @@ -62,6 +67,12 @@ public: constexpr KeeperDispatcher * getDispatcher() const { return dispatcher; } + void setRocksDBDisk(DiskPtr disk); + DiskPtr getTemporaryRocksDBDisk() const; + + void setRocksDBOptions(std::shared_ptr rocksdb_options_ = nullptr); + std::shared_ptr getRocksDBOptions() const { return rocksdb_options; } + UInt64 getKeeperMemorySoftLimit() const { return memory_soft_limit; } void updateKeeperMemorySoftLimit(const Poco::Util::AbstractConfiguration & config); @@ -90,6 +101,7 @@ private: void initializeFeatureFlags(const Poco::Util::AbstractConfiguration & config); void initializeDisks(const Poco::Util::AbstractConfiguration & config); + Storage getRocksDBPathFromConfig(const Poco::Util::AbstractConfiguration & config) const; Storage getLogsPathFromConfig(const Poco::Util::AbstractConfiguration & config) const; Storage getSnapshotsPathFromConfig(const Poco::Util::AbstractConfiguration & config) const; Storage getStatePathFromConfig(const Poco::Util::AbstractConfiguration & config) const; @@ -111,12 +123,15 @@ private: std::shared_ptr disk_selector; + Storage rocksdb_storage; Storage log_storage; Storage latest_log_storage; Storage snapshot_storage; Storage latest_snapshot_storage; Storage state_file_storage; + std::shared_ptr rocksdb_options; + std::vector old_log_disk_names; std::vector old_snapshot_disk_names; diff --git a/src/Coordination/KeeperDispatcher.cpp b/src/Coordination/KeeperDispatcher.cpp index b4389da082d..4a350077596 100644 --- a/src/Coordination/KeeperDispatcher.cpp +++ b/src/Coordination/KeeperDispatcher.cpp @@ -5,7 +5,8 @@ #include #include -#include "Common/ZooKeeper/IKeeper.h" +#include +#include #include #include #include @@ -13,6 +14,7 @@ #include #include #include +#include #include @@ -28,7 +30,14 @@ namespace CurrentMetrics { extern const Metric KeeperAliveConnections; - extern const Metric KeeperOutstandingRequets; + extern const Metric KeeperOutstandingRequests; +} + +namespace ProfileEvents +{ + extern const Event KeeperCommitWaitElapsedMicroseconds; + extern const Event KeeperBatchMaxCount; + extern const Event KeeperBatchMaxTotalSize; } using namespace std::chrono_literals; @@ -108,17 +117,18 @@ void KeeperDispatcher::requestThread() RaftAppendResult prev_result = nullptr; /// Requests from previous iteration. We store them to be able /// to send errors to the client. - KeeperStorage::RequestsForSessions prev_batch; + KeeperStorageBase::RequestsForSessions prev_batch; const auto & shutdown_called = keeper_context->isShutdownCalled(); while (!shutdown_called) { - KeeperStorage::RequestForSession request; + KeeperStorageBase::RequestForSession request; auto coordination_settings = configuration_and_settings->coordination_settings; uint64_t max_wait = coordination_settings->operation_timeout_ms.totalMilliseconds(); uint64_t max_batch_bytes_size = coordination_settings->max_requests_batch_bytes_size; + size_t max_batch_size = coordination_settings->max_requests_batch_size; /// The code below do a very simple thing: batch all write (quorum) requests into vector until /// previous write batch is not finished or max_batch size achieved. The main complexity goes from @@ -131,7 +141,7 @@ void KeeperDispatcher::requestThread() { if (requests_queue->tryPop(request, max_wait)) { - CurrentMetrics::sub(CurrentMetrics::KeeperOutstandingRequets); + CurrentMetrics::sub(CurrentMetrics::KeeperOutstandingRequests); if (shutdown_called) break; @@ -143,7 +153,7 @@ void KeeperDispatcher::requestThread() continue; } - KeeperStorage::RequestsForSessions current_batch; + KeeperStorageBase::RequestsForSessions current_batch; size_t current_batch_bytes_size = 0; bool has_read_request = false; @@ -163,7 +173,7 @@ void KeeperDispatcher::requestThread() /// Trying to get batch requests as fast as possible if (requests_queue->tryPop(request)) { - CurrentMetrics::sub(CurrentMetrics::KeeperOutstandingRequets); + CurrentMetrics::sub(CurrentMetrics::KeeperOutstandingRequests); /// Don't append read request into batch, we have to process them separately if (!coordination_settings->quorum_reads && request.request->isReadRequest()) { @@ -188,7 +198,6 @@ void KeeperDispatcher::requestThread() return false; }; - size_t max_batch_size = coordination_settings->max_requests_batch_size; while (!shutdown_called && current_batch.size() < max_batch_size && !has_reconfig_request && current_batch_bytes_size < max_batch_bytes_size && try_get_request()) ; @@ -225,6 +234,12 @@ void KeeperDispatcher::requestThread() /// Process collected write requests batch if (!current_batch.empty()) { + if (current_batch.size() == max_batch_size) + ProfileEvents::increment(ProfileEvents::KeeperBatchMaxCount, 1); + + if (current_batch_bytes_size == max_batch_bytes_size) + ProfileEvents::increment(ProfileEvents::KeeperBatchMaxTotalSize, 1); + LOG_TRACE(log, "Processing requests batch, size: {}, bytes: {}", current_batch.size(), current_batch_bytes_size); auto result = server->putRequestBatch(current_batch); @@ -243,6 +258,8 @@ void KeeperDispatcher::requestThread() /// If we will execute read or reconfig next, we have to process result now if (execute_requests_after_write) { + Stopwatch watch; + SCOPE_EXIT(ProfileEvents::increment(ProfileEvents::KeeperCommitWaitElapsedMicroseconds, watch.elapsedMicroseconds())); if (prev_result) result_buf = forceWaitAndProcessResult( prev_result, prev_batch, /*clear_requests_on_success=*/!execute_requests_after_write); @@ -294,7 +311,7 @@ void KeeperDispatcher::responseThread() const auto & shutdown_called = keeper_context->isShutdownCalled(); while (!shutdown_called) { - KeeperStorage::ResponseForSession response_for_session; + KeeperStorageBase::ResponseForSession response_for_session; uint64_t max_wait = configuration_and_settings->coordination_settings->operation_timeout_ms.totalMilliseconds(); @@ -305,7 +322,7 @@ void KeeperDispatcher::responseThread() try { - setResponse(response_for_session.session_id, response_for_session.response); + setResponse(response_for_session.session_id, response_for_session.response, response_for_session.request); } catch (...) { @@ -319,19 +336,13 @@ void KeeperDispatcher::snapshotThread() { setThreadName("KeeperSnpT"); const auto & shutdown_called = keeper_context->isShutdownCalled(); - while (!shutdown_called) + CreateSnapshotTask task; + while (snapshots_queue.pop(task)) { - CreateSnapshotTask task; - if (!snapshots_queue.pop(task)) - break; - try { auto snapshot_file_info = task.create_snapshot(std::move(task.snapshot), /*execute_only_cleanup=*/shutdown_called); - if (shutdown_called) - break; - if (!snapshot_file_info) continue; @@ -346,7 +357,7 @@ void KeeperDispatcher::snapshotThread() } } -void KeeperDispatcher::setResponse(int64_t session_id, const Coordination::ZooKeeperResponsePtr & response) +void KeeperDispatcher::setResponse(int64_t session_id, const Coordination::ZooKeeperResponsePtr & response, Coordination::ZooKeeperRequestPtr request) { std::lock_guard lock(session_to_response_callback_mutex); @@ -360,7 +371,7 @@ void KeeperDispatcher::setResponse(int64_t session_id, const Coordination::ZooKe return; auto callback = new_session_id_response_callback[session_id_resp.internal_id]; - callback(response); + callback(response, request); new_session_id_response_callback.erase(session_id_resp.internal_id); } else /// Normal response, just write to client @@ -371,7 +382,7 @@ void KeeperDispatcher::setResponse(int64_t session_id, const Coordination::ZooKe if (session_response_callback == session_to_response_callback.end()) return; - session_response_callback->second(response); + session_response_callback->second(response, request); /// Session closed, no more writes if (response->xid != Coordination::WATCH_XID && response->getOpNum() == Coordination::OpNum::Close) @@ -391,7 +402,7 @@ bool KeeperDispatcher::putRequest(const Coordination::ZooKeeperRequestPtr & requ return false; } - KeeperStorage::RequestForSession request_info; + KeeperStorageBase::RequestForSession request_info; request_info.request = request; using namespace std::chrono; request_info.time = duration_cast(system_clock::now().time_since_epoch()).count(); @@ -410,7 +421,7 @@ bool KeeperDispatcher::putRequest(const Coordination::ZooKeeperRequestPtr & requ { throw Exception(ErrorCodes::TIMEOUT_EXCEEDED, "Cannot push request to queue within operation timeout"); } - CurrentMetrics::add(CurrentMetrics::KeeperOutstandingRequets); + CurrentMetrics::add(CurrentMetrics::KeeperOutstandingRequests); return true; } @@ -437,7 +448,7 @@ void KeeperDispatcher::initialize(const Poco::Util::AbstractConfiguration & conf snapshots_queue, keeper_context, snapshot_s3, - [this](uint64_t /*log_idx*/, const KeeperStorage::RequestForSession & request_for_session) + [this](uint64_t /*log_idx*/, const KeeperStorageBase::RequestForSession & request_for_session) { { /// check if we have queue of read requests depending on this request to be committed @@ -529,18 +540,18 @@ void KeeperDispatcher::shutdown() update_configuration_thread.join(); } - KeeperStorage::RequestForSession request_for_session; + KeeperStorageBase::RequestForSession request_for_session; /// Set session expired for all pending requests while (requests_queue && requests_queue->tryPop(request_for_session)) { - CurrentMetrics::sub(CurrentMetrics::KeeperOutstandingRequets); + CurrentMetrics::sub(CurrentMetrics::KeeperOutstandingRequests); auto response = request_for_session.request->makeResponse(); response->error = Coordination::Error::ZSESSIONEXPIRED; setResponse(request_for_session.session_id, response); } - KeeperStorage::RequestsForSessions close_requests; + KeeperStorageBase::RequestsForSessions close_requests; { /// Clear all registered sessions std::lock_guard lock(session_to_response_callback_mutex); @@ -554,7 +565,7 @@ void KeeperDispatcher::shutdown() auto request = Coordination::ZooKeeperRequestFactory::instance().get(Coordination::OpNum::Close); request->xid = Coordination::CLOSE_XID; using namespace std::chrono; - KeeperStorage::RequestForSession request_info + KeeperStorageBase::RequestForSession request_info { .session_id = session, .time = duration_cast(system_clock::now().time_since_epoch()).count(), @@ -652,7 +663,7 @@ void KeeperDispatcher::sessionCleanerTask() auto request = Coordination::ZooKeeperRequestFactory::instance().get(Coordination::OpNum::Close); request->xid = Coordination::CLOSE_XID; using namespace std::chrono; - KeeperStorage::RequestForSession request_info + KeeperStorageBase::RequestForSession request_info { .session_id = dead_session, .time = duration_cast(system_clock::now().time_since_epoch()).count(), @@ -661,7 +672,7 @@ void KeeperDispatcher::sessionCleanerTask() }; if (!requests_queue->push(std::move(request_info))) LOG_INFO(log, "Cannot push close request to queue while cleaning outdated sessions"); - CurrentMetrics::add(CurrentMetrics::KeeperOutstandingRequets); + CurrentMetrics::add(CurrentMetrics::KeeperOutstandingRequests); /// Remove session from registered sessions finishSession(dead_session); @@ -700,16 +711,16 @@ void KeeperDispatcher::finishSession(int64_t session_id) } } -void KeeperDispatcher::addErrorResponses(const KeeperStorage::RequestsForSessions & requests_for_sessions, Coordination::Error error) +void KeeperDispatcher::addErrorResponses(const KeeperStorageBase::RequestsForSessions & requests_for_sessions, Coordination::Error error) { for (const auto & request_for_session : requests_for_sessions) { - KeeperStorage::ResponsesForSessions responses; + KeeperStorageBase::ResponsesForSessions responses; auto response = request_for_session.request->makeResponse(); response->xid = request_for_session.request->xid; response->zxid = 0; response->error = error; - if (!responses_queue.push(DB::KeeperStorage::ResponseForSession{request_for_session.session_id, response})) + if (!responses_queue.push(DB::KeeperStorageBase::ResponseForSession{request_for_session.session_id, response})) throw Exception(ErrorCodes::SYSTEM_ERROR, "Could not push error response xid {} zxid {} error message {} to responses queue", response->xid, @@ -719,7 +730,7 @@ void KeeperDispatcher::addErrorResponses(const KeeperStorage::RequestsForSession } nuraft::ptr KeeperDispatcher::forceWaitAndProcessResult( - RaftAppendResult & result, KeeperStorage::RequestsForSessions & requests_for_sessions, bool clear_requests_on_success) + RaftAppendResult & result, KeeperStorageBase::RequestsForSessions & requests_for_sessions, bool clear_requests_on_success) { if (!result->has_result()) result->get(); @@ -744,7 +755,7 @@ int64_t KeeperDispatcher::getSessionID(int64_t session_timeout_ms) { /// New session id allocation is a special request, because we cannot process it in normal /// way: get request -> put to raft -> set response for registered callback. - KeeperStorage::RequestForSession request_info; + KeeperStorageBase::RequestForSession request_info; std::shared_ptr request = std::make_shared(); /// Internal session id. It's a temporary number which is unique for each client on this server /// but can be same on different servers. @@ -762,21 +773,27 @@ int64_t KeeperDispatcher::getSessionID(int64_t session_timeout_ms) { std::lock_guard lock(session_to_response_callback_mutex); - new_session_id_response_callback[request->internal_id] = [promise, internal_id = request->internal_id] (const Coordination::ZooKeeperResponsePtr & response) + new_session_id_response_callback[request->internal_id] + = [promise, internal_id = request->internal_id]( + const Coordination::ZooKeeperResponsePtr & response, Coordination::ZooKeeperRequestPtr /*request*/) { if (response->getOpNum() != Coordination::OpNum::SessionID) - promise->set_exception(std::make_exception_ptr(Exception(ErrorCodes::LOGICAL_ERROR, - "Incorrect response of type {} instead of SessionID response", response->getOpNum()))); + promise->set_exception(std::make_exception_ptr(Exception( + ErrorCodes::LOGICAL_ERROR, "Incorrect response of type {} instead of SessionID response", response->getOpNum()))); auto session_id_response = dynamic_cast(*response); if (session_id_response.internal_id != internal_id) { - promise->set_exception(std::make_exception_ptr(Exception(ErrorCodes::LOGICAL_ERROR, - "Incorrect response with internal id {} instead of {}", session_id_response.internal_id, internal_id))); + promise->set_exception(std::make_exception_ptr(Exception( + ErrorCodes::LOGICAL_ERROR, + "Incorrect response with internal id {} instead of {}", + session_id_response.internal_id, + internal_id))); } if (response->error != Coordination::Error::ZOK) - promise->set_exception(std::make_exception_ptr(zkutil::KeeperException::fromMessage(response->error, "SessionID request failed with error"))); + promise->set_exception( + std::make_exception_ptr(zkutil::KeeperException::fromMessage(response->error, "SessionID request failed with error"))); promise->set_value(session_id_response.session_id); }; @@ -785,7 +802,7 @@ int64_t KeeperDispatcher::getSessionID(int64_t session_timeout_ms) /// Push new session request to queue if (!requests_queue->tryPush(std::move(request_info), session_timeout_ms)) throw Exception(ErrorCodes::TIMEOUT_EXCEEDED, "Cannot push session id request to queue within session timeout"); - CurrentMetrics::add(CurrentMetrics::KeeperOutstandingRequets); + CurrentMetrics::add(CurrentMetrics::KeeperOutstandingRequests); if (future.wait_for(std::chrono::milliseconds(session_timeout_ms)) != std::future_status::ready) throw Exception(ErrorCodes::TIMEOUT_EXCEEDED, "Cannot receive session id within session timeout"); diff --git a/src/Coordination/KeeperDispatcher.h b/src/Coordination/KeeperDispatcher.h index 2e0c73131d5..651fd0e1c88 100644 --- a/src/Coordination/KeeperDispatcher.h +++ b/src/Coordination/KeeperDispatcher.h @@ -11,7 +11,6 @@ #include #include #include -#include #include #include #include @@ -20,14 +19,14 @@ namespace DB { -using ZooKeeperResponseCallback = std::function; +using ZooKeeperResponseCallback = std::function; /// Highlevel wrapper for ClickHouse Keeper. /// Process user requests via consensus and return responses. class KeeperDispatcher { private: - using RequestsQueue = ConcurrentBoundedQueue; + using RequestsQueue = ConcurrentBoundedQueue; using SessionToResponseCallback = std::unordered_map; using ClusterUpdateQueue = ConcurrentBoundedQueue; @@ -92,22 +91,22 @@ private: void clusterUpdateWithReconfigDisabledThread(); void clusterUpdateThread(); - void setResponse(int64_t session_id, const Coordination::ZooKeeperResponsePtr & response); + void setResponse(int64_t session_id, const Coordination::ZooKeeperResponsePtr & response, Coordination::ZooKeeperRequestPtr request = nullptr); /// Add error responses for requests to responses queue. /// Clears requests. - void addErrorResponses(const KeeperStorage::RequestsForSessions & requests_for_sessions, Coordination::Error error); + void addErrorResponses(const KeeperStorageBase::RequestsForSessions & requests_for_sessions, Coordination::Error error); /// Forcefully wait for result and sets errors if something when wrong. /// Clears both arguments nuraft::ptr forceWaitAndProcessResult( - RaftAppendResult & result, KeeperStorage::RequestsForSessions & requests_for_sessions, bool clear_requests_on_success); + RaftAppendResult & result, KeeperStorageBase::RequestsForSessions & requests_for_sessions, bool clear_requests_on_success); public: std::mutex read_request_queue_mutex; /// queue of read requests that can be processed after a request with specific session ID and XID is committed - std::unordered_map> read_request_queue; + std::unordered_map> read_request_queue; /// Just allocate some objects, real initialization is done by `intialize method` KeeperDispatcher(); @@ -193,7 +192,7 @@ public: Keeper4LWInfo getKeeper4LWInfo() const; - const KeeperStateMachine & getStateMachine() const + const IKeeperStateMachine & getStateMachine() const { return *server->getKeeperStateMachine(); } diff --git a/src/Coordination/KeeperServer.cpp b/src/Coordination/KeeperServer.cpp index 736a01443ce..f09ea56391a 100644 --- a/src/Coordination/KeeperServer.cpp +++ b/src/Coordination/KeeperServer.cpp @@ -3,15 +3,14 @@ #include "config.h" -#include -#include -#include +#include #include +#include #include #include -#include #include #include +#include #include #include #include @@ -27,7 +26,11 @@ #include #include #include -#include +#include + +#include +#include +#include #pragma clang diagnostic ignored "-Wdeprecated-declarations" #include @@ -120,7 +123,7 @@ KeeperServer::KeeperServer( SnapshotsQueue & snapshots_queue_, KeeperContextPtr keeper_context_, KeeperSnapshotManagerS3 & snapshot_manager_s3, - KeeperStateMachine::CommitCallback commit_callback) + IKeeperStateMachine::CommitCallback commit_callback) : server_id(configuration_and_settings_->server_id) , log(getLogger("KeeperServer")) , is_recovering(config.getBool("keeper_server.force_recovery", false)) @@ -131,13 +134,28 @@ KeeperServer::KeeperServer( if (keeper_context->getCoordinationSettings()->quorum_reads) LOG_WARNING(log, "Quorum reads enabled, Keeper will work slower."); - state_machine = nuraft::cs_new( - responses_queue_, - snapshots_queue_, - keeper_context, - config.getBool("keeper_server.upload_snapshot_on_exit", false) ? &snapshot_manager_s3 : nullptr, - commit_callback, - checkAndGetSuperdigest(configuration_and_settings_->super_digest)); +#if USE_ROCKSDB + const auto & coordination_settings = keeper_context->getCoordinationSettings(); + if (coordination_settings->experimental_use_rocksdb) + { + state_machine = nuraft::cs_new>( + responses_queue_, + snapshots_queue_, + keeper_context, + config.getBool("keeper_server.upload_snapshot_on_exit", false) ? &snapshot_manager_s3 : nullptr, + commit_callback, + checkAndGetSuperdigest(configuration_and_settings_->super_digest)); + LOG_WARNING(log, "Use RocksDB as Keeper backend storage."); + } + else +#endif + state_machine = nuraft::cs_new>( + responses_queue_, + snapshots_queue_, + keeper_context, + config.getBool("keeper_server.upload_snapshot_on_exit", false) ? &snapshot_manager_s3 : nullptr, + commit_callback, + checkAndGetSuperdigest(configuration_and_settings_->super_digest)); state_manager = nuraft::cs_new( server_id, @@ -365,6 +383,11 @@ void KeeperServer::launchRaftServer(const Poco::Util::AbstractConfiguration & co LockMemoryExceptionInThread::removeUniqueLock(); }; + /// At least 16 threads for network communication in asio. + /// asio is async framework, so even with 1 thread it should be ok, but + /// still as safeguard it's better to have some redundant capacity here + asio_opts.thread_pool_size_ = std::max(16U, getNumberOfPhysicalCPUCores()); + if (state_manager->isSecure()) { #if USE_SSL @@ -517,7 +540,7 @@ namespace { // Serialize the request for the log entry -nuraft::ptr getZooKeeperLogEntry(const KeeperStorage::RequestForSession & request_for_session) +nuraft::ptr getZooKeeperLogEntry(const KeeperStorageBase::RequestForSession & request_for_session) { DB::WriteBufferFromNuraftBuffer write_buf; DB::writeIntBinary(request_for_session.session_id, write_buf); @@ -525,7 +548,7 @@ nuraft::ptr getZooKeeperLogEntry(const KeeperStorage::RequestFor DB::writeIntBinary(request_for_session.time, write_buf); /// we fill with dummy values to eliminate unnecessary copy later on when we will write correct values DB::writeIntBinary(static_cast(0), write_buf); /// zxid - DB::writeIntBinary(KeeperStorage::DigestVersion::NO_DIGEST, write_buf); /// digest version or NO_DIGEST flag + DB::writeIntBinary(KeeperStorageBase::DigestVersion::NO_DIGEST, write_buf); /// digest version or NO_DIGEST flag DB::writeIntBinary(static_cast(0), write_buf); /// digest value /// if new fields are added, update KeeperStateMachine::ZooKeeperLogSerializationVersion along with parseRequest function and PreAppendLog callback handler return write_buf.getBuffer(); @@ -533,7 +556,7 @@ nuraft::ptr getZooKeeperLogEntry(const KeeperStorage::RequestFor } -void KeeperServer::putLocalReadRequest(const KeeperStorage::RequestForSession & request_for_session) +void KeeperServer::putLocalReadRequest(const KeeperStorageBase::RequestForSession & request_for_session) { if (!request_for_session.request->isReadRequest()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot process non-read request locally"); @@ -541,7 +564,7 @@ void KeeperServer::putLocalReadRequest(const KeeperStorage::RequestForSession & state_machine->processReadRequest(request_for_session); } -RaftAppendResult KeeperServer::putRequestBatch(const KeeperStorage::RequestsForSessions & requests_for_sessions) +RaftAppendResult KeeperServer::putRequestBatch(const KeeperStorageBase::RequestsForSessions & requests_for_sessions) { std::vector> entries; entries.reserve(requests_for_sessions.size()); @@ -784,7 +807,7 @@ nuraft::cb_func::ReturnCode KeeperServer::callbackFunc(nuraft::cb_func::Type typ auto entry_buf = entry->get_buf_ptr(); - KeeperStateMachine::ZooKeeperLogSerializationVersion serialization_version; + IKeeperStateMachine::ZooKeeperLogSerializationVersion serialization_version; auto request_for_session = state_machine->parseRequest(*entry_buf, /*final=*/false, &serialization_version); request_for_session->zxid = next_zxid; if (!state_machine->preprocess(*request_for_session)) @@ -794,10 +817,10 @@ nuraft::cb_func::ReturnCode KeeperServer::callbackFunc(nuraft::cb_func::Type typ /// older versions of Keeper can send logs that are missing some fields size_t bytes_missing = 0; - if (serialization_version < KeeperStateMachine::ZooKeeperLogSerializationVersion::WITH_TIME) + if (serialization_version < IKeeperStateMachine::ZooKeeperLogSerializationVersion::WITH_TIME) bytes_missing += sizeof(request_for_session->time); - if (serialization_version < KeeperStateMachine::ZooKeeperLogSerializationVersion::WITH_ZXID_DIGEST) + if (serialization_version < IKeeperStateMachine::ZooKeeperLogSerializationVersion::WITH_ZXID_DIGEST) bytes_missing += sizeof(request_for_session->zxid) + sizeof(request_for_session->digest->version) + sizeof(request_for_session->digest->value); if (bytes_missing != 0) @@ -811,19 +834,19 @@ nuraft::cb_func::ReturnCode KeeperServer::callbackFunc(nuraft::cb_func::Type typ size_t write_buffer_header_size = sizeof(request_for_session->zxid) + sizeof(request_for_session->digest->version) + sizeof(request_for_session->digest->value); - if (serialization_version < KeeperStateMachine::ZooKeeperLogSerializationVersion::WITH_TIME) + if (serialization_version < IKeeperStateMachine::ZooKeeperLogSerializationVersion::WITH_TIME) write_buffer_header_size += sizeof(request_for_session->time); auto * buffer_start = reinterpret_cast(entry_buf->data_begin() + entry_buf->size() - write_buffer_header_size); WriteBufferFromPointer write_buf(buffer_start, write_buffer_header_size); - if (serialization_version < KeeperStateMachine::ZooKeeperLogSerializationVersion::WITH_TIME) + if (serialization_version < IKeeperStateMachine::ZooKeeperLogSerializationVersion::WITH_TIME) writeIntBinary(request_for_session->time, write_buf); writeIntBinary(request_for_session->zxid, write_buf); writeIntBinary(request_for_session->digest->version, write_buf); - if (request_for_session->digest->version != KeeperStorage::NO_DIGEST) + if (request_for_session->digest->version != KeeperStorageBase::NO_DIGEST) writeIntBinary(request_for_session->digest->value, write_buf); write_buf.finalize(); diff --git a/src/Coordination/KeeperServer.h b/src/Coordination/KeeperServer.h index 5e45a552cba..f082b5d377e 100644 --- a/src/Coordination/KeeperServer.h +++ b/src/Coordination/KeeperServer.h @@ -1,6 +1,5 @@ #pragma once -#include #include #include #include @@ -17,12 +16,15 @@ namespace DB using RaftAppendResult = nuraft::ptr>>; +struct KeeperConfigurationAndSettings; +using KeeperConfigurationAndSettingsPtr = std::shared_ptr; + class KeeperServer { private: const int server_id; - nuraft::ptr state_machine; + nuraft::ptr state_machine; nuraft::ptr state_manager; @@ -77,26 +79,26 @@ public: SnapshotsQueue & snapshots_queue_, KeeperContextPtr keeper_context_, KeeperSnapshotManagerS3 & snapshot_manager_s3, - KeeperStateMachine::CommitCallback commit_callback); + IKeeperStateMachine::CommitCallback commit_callback); /// Load state machine from the latest snapshot and load log storage. Start NuRaft with required settings. void startup(const Poco::Util::AbstractConfiguration & config, bool enable_ipv6 = true); /// Put local read request and execute in state machine directly and response into /// responses queue - void putLocalReadRequest(const KeeperStorage::RequestForSession & request); + void putLocalReadRequest(const KeeperStorageBase::RequestForSession & request); bool isRecovering() const { return is_recovering; } bool reconfigEnabled() const { return enable_reconfiguration; } /// Put batch of requests into Raft and get result of put. Responses will be set separately into /// responses_queue. - RaftAppendResult putRequestBatch(const KeeperStorage::RequestsForSessions & requests); + RaftAppendResult putRequestBatch(const KeeperStorageBase::RequestsForSessions & requests); /// Return set of the non-active sessions std::vector getDeadSessions(); - nuraft::ptr getKeeperStateMachine() const { return state_machine; } + nuraft::ptr getKeeperStateMachine() const { return state_machine; } void forceRecovery(); diff --git a/src/Coordination/KeeperSnapshotManager.cpp b/src/Coordination/KeeperSnapshotManager.cpp index 8bf42b434cd..3f5ac055470 100644 --- a/src/Coordination/KeeperSnapshotManager.cpp +++ b/src/Coordination/KeeperSnapshotManager.cpp @@ -66,7 +66,8 @@ namespace return base; } - void writeNode(const KeeperStorage::Node & node, SnapshotVersion version, WriteBuffer & out) + template + void writeNode(const Node & node, SnapshotVersion version, WriteBuffer & out) { writeBinary(node.getData(), out); @@ -86,7 +87,7 @@ namespace writeBinary(node.aversion, out); writeBinary(node.ephemeralOwner(), out); if (version < SnapshotVersion::V6) - writeBinary(static_cast(node.data_size), out); + writeBinary(static_cast(node.getData().size()), out); writeBinary(node.numChildren(), out); writeBinary(node.pzxid, out); @@ -96,7 +97,8 @@ namespace writeBinary(node.sizeInBytes(), out); } - void readNode(KeeperStorage::Node & node, ReadBuffer & in, SnapshotVersion version, ACLMap & acl_map) + template + void readNode(Node & node, ReadBuffer & in, SnapshotVersion version, ACLMap & acl_map) { readVarUInt(node.data_size, in); if (node.data_size != 0) @@ -195,7 +197,8 @@ namespace } } -void KeeperStorageSnapshot::serialize(const KeeperStorageSnapshot & snapshot, WriteBuffer & out, KeeperContextPtr keeper_context) +template +void KeeperStorageSnapshot::serialize(const KeeperStorageSnapshot & snapshot, WriteBuffer & out, KeeperContextPtr keeper_context) { writeBinary(static_cast(snapshot.version), out); serializeSnapshotMetadata(snapshot.snapshot_meta, out); @@ -205,11 +208,11 @@ void KeeperStorageSnapshot::serialize(const KeeperStorageSnapshot & snapshot, Wr writeBinary(snapshot.zxid, out); if (keeper_context->digestEnabled()) { - writeBinary(static_cast(KeeperStorage::CURRENT_DIGEST_VERSION), out); + writeBinary(static_cast(Storage::CURRENT_DIGEST_VERSION), out); writeBinary(snapshot.nodes_digest, out); } else - writeBinary(static_cast(KeeperStorage::NO_DIGEST), out); + writeBinary(static_cast(Storage::NO_DIGEST), out); } writeBinary(snapshot.session_id, out); @@ -255,7 +258,6 @@ void KeeperStorageSnapshot::serialize(const KeeperStorageSnapshot & snapshot, Wr /// slightly bigger than required. if (node.mzxid > snapshot.zxid) break; - writeBinary(path, out); writeNode(node, snapshot.version, out); @@ -282,7 +284,7 @@ void KeeperStorageSnapshot::serialize(const KeeperStorageSnapshot & snapshot, Wr writeBinary(session_id, out); writeBinary(timeout, out); - KeeperStorage::AuthIDs ids; + KeeperStorageBase::AuthIDs ids; if (snapshot.session_and_auth.contains(session_id)) ids = snapshot.session_and_auth.at(session_id); @@ -303,7 +305,8 @@ void KeeperStorageSnapshot::serialize(const KeeperStorageSnapshot & snapshot, Wr } } -void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserialization_result, ReadBuffer & in, KeeperContextPtr keeper_context) +template +void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserialization_result, ReadBuffer & in, KeeperContextPtr keeper_context) { uint8_t version; readBinary(version, in); @@ -312,7 +315,7 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial throw Exception(ErrorCodes::UNKNOWN_FORMAT_VERSION, "Unsupported snapshot version {}", version); deserialization_result.snapshot_meta = deserializeSnapshotMetadata(in); - KeeperStorage & storage = *deserialization_result.storage; + Storage & storage = *deserialization_result.storage; bool recalculate_digest = keeper_context->digestEnabled(); if (version >= SnapshotVersion::V5) @@ -320,11 +323,11 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial readBinary(storage.zxid, in); uint8_t digest_version; readBinary(digest_version, in); - if (digest_version != KeeperStorage::DigestVersion::NO_DIGEST) + if (digest_version != Storage::DigestVersion::NO_DIGEST) { uint64_t nodes_digest; readBinary(nodes_digest, in); - if (digest_version == KeeperStorage::CURRENT_DIGEST_VERSION) + if (digest_version == Storage::CURRENT_DIGEST_VERSION) { storage.nodes_digest = nodes_digest; recalculate_digest = false; @@ -374,8 +377,8 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial size_t snapshot_container_size; readBinary(snapshot_container_size, in); - - storage.container.reserve(snapshot_container_size); + if constexpr (!use_rocksdb) + storage.container.reserve(snapshot_container_size); if (recalculate_digest) storage.nodes_digest = 0; @@ -389,7 +392,7 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial in.readStrict(path_data.get(), path_size); std::string_view path{path_data.get(), path_size}; - KeeperStorage::Node node{}; + typename Storage::Node node{}; readNode(node, in, current_version, storage.acl_map); using enum Coordination::PathMatchResult; @@ -421,7 +424,7 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial if (keeper_context->ignoreSystemPathOnStartup() || keeper_context->getServerState() != KeeperContext::Phase::INIT) { LOG_ERROR(getLogger("KeeperSnapshotManager"), "{}. Ignoring it", get_error_msg()); - node = KeeperStorage::Node{}; + node = typename Storage::Node{}; } else throw Exception( @@ -433,8 +436,9 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial } auto ephemeral_owner = node.ephemeralOwner(); - if (!node.isEphemeral() && node.numChildren() > 0) - node.getChildren().reserve(node.numChildren()); + if constexpr (!use_rocksdb) + if (!node.isEphemeral() && node.numChildren() > 0) + node.getChildren().reserve(node.numChildren()); if (ephemeral_owner != 0) storage.ephemerals[node.ephemeralOwner()].insert(std::string{path}); @@ -447,36 +451,38 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial LOG_TRACE(getLogger("KeeperSnapshotManager"), "Building structure for children nodes"); - for (const auto & itr : storage.container) + if constexpr (!use_rocksdb) { - if (itr.key != "/") + for (const auto & itr : storage.container) { - auto parent_path = parentNodePath(itr.key); - storage.container.updateValue( - parent_path, [path = itr.key](KeeperStorage::Node & value) { value.addChild(getBaseNodeName(path)); }); - } - } - - for (const auto & itr : storage.container) - { - if (itr.key != "/") - { - if (itr.value.numChildren() != static_cast(itr.value.getChildren().size())) + if (itr.key != "/") { + auto parent_path = parentNodePath(itr.key); + storage.container.updateValue( + parent_path, [path = itr.key](typename Storage::Node & value) { value.addChild(getBaseNodeName(path)); }); + } + } + + for (const auto & itr : storage.container) + { + if (itr.key != "/") + { + if (itr.value.numChildren() != static_cast(itr.value.getChildren().size())) + { #ifdef NDEBUG - /// TODO (alesapin) remove this, it should be always CORRUPTED_DATA. - LOG_ERROR(getLogger("KeeperSnapshotManager"), "Children counter in stat.numChildren {}" - " is different from actual children size {} for node {}", itr.value.numChildren(), itr.value.getChildren().size(), itr.key); + /// TODO (alesapin) remove this, it should be always CORRUPTED_DATA. + LOG_ERROR(getLogger("KeeperSnapshotManager"), "Children counter in stat.numChildren {}" + " is different from actual children size {} for node {}", itr.value.numChildren(), itr.value.getChildren().size(), itr.key); #else - throw Exception(ErrorCodes::LOGICAL_ERROR, "Children counter in stat.numChildren {}" - " is different from actual children size {} for node {}", - itr.value.numChildren(), itr.value.getChildren().size(), itr.key); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Children counter in stat.numChildren {}" + " is different from actual children size {} for node {}", + itr.value.numChildren(), itr.value.getChildren().size(), itr.key); #endif + } } } } - size_t active_sessions_size; readBinary(active_sessions_size, in); @@ -493,14 +499,14 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial size_t session_auths_size; readBinary(session_auths_size, in); - KeeperStorage::AuthIDs ids; + typename Storage::AuthIDs ids; size_t session_auth_counter = 0; while (session_auth_counter < session_auths_size) { String scheme, id; readBinary(scheme, in); readBinary(id, in); - ids.emplace_back(KeeperStorage::AuthID{scheme, id}); + ids.emplace_back(typename Storage::AuthID{scheme, id}); session_auth_counter++; } @@ -523,7 +529,8 @@ void KeeperStorageSnapshot::deserialize(SnapshotDeserializationResult & deserial } } -KeeperStorageSnapshot::KeeperStorageSnapshot(KeeperStorage * storage_, uint64_t up_to_log_idx_, const ClusterConfigPtr & cluster_config_) +template +KeeperStorageSnapshot::KeeperStorageSnapshot(Storage * storage_, uint64_t up_to_log_idx_, const ClusterConfigPtr & cluster_config_) : storage(storage_) , snapshot_meta(std::make_shared(up_to_log_idx_, 0, std::make_shared())) , session_id(storage->session_id_counter) @@ -540,8 +547,9 @@ KeeperStorageSnapshot::KeeperStorageSnapshot(KeeperStorage * storage_, uint64_t session_and_auth = storage->session_and_auth; } -KeeperStorageSnapshot::KeeperStorageSnapshot( - KeeperStorage * storage_, const SnapshotMetadataPtr & snapshot_meta_, const ClusterConfigPtr & cluster_config_) +template +KeeperStorageSnapshot::KeeperStorageSnapshot( + Storage * storage_, const SnapshotMetadataPtr & snapshot_meta_, const ClusterConfigPtr & cluster_config_) : storage(storage_) , snapshot_meta(snapshot_meta_) , session_id(storage->session_id_counter) @@ -558,12 +566,14 @@ KeeperStorageSnapshot::KeeperStorageSnapshot( session_and_auth = storage->session_and_auth; } -KeeperStorageSnapshot::~KeeperStorageSnapshot() +template +KeeperStorageSnapshot::~KeeperStorageSnapshot() { storage->disableSnapshotMode(); } -KeeperSnapshotManager::KeeperSnapshotManager( +template +KeeperSnapshotManager::KeeperSnapshotManager( size_t snapshots_to_keep_, const KeeperContextPtr & keeper_context_, bool compress_snapshots_zstd_, @@ -651,7 +661,8 @@ KeeperSnapshotManager::KeeperSnapshotManager( moveSnapshotsIfNeeded(); } -SnapshotFileInfoPtr KeeperSnapshotManager::serializeSnapshotBufferToDisk(nuraft::buffer & buffer, uint64_t up_to_log_idx) +template +SnapshotFileInfoPtr KeeperSnapshotManager::serializeSnapshotBufferToDisk(nuraft::buffer & buffer, uint64_t up_to_log_idx) { ReadBufferFromNuraftBuffer reader(buffer); @@ -680,7 +691,8 @@ SnapshotFileInfoPtr KeeperSnapshotManager::serializeSnapshotBufferToDisk(nuraft: return snapshot_file_info; } -nuraft::ptr KeeperSnapshotManager::deserializeLatestSnapshotBufferFromDisk() +template +nuraft::ptr KeeperSnapshotManager::deserializeLatestSnapshotBufferFromDisk() { while (!existing_snapshots.empty()) { @@ -701,7 +713,8 @@ nuraft::ptr KeeperSnapshotManager::deserializeLatestSnapshotBuff return nullptr; } -nuraft::ptr KeeperSnapshotManager::deserializeSnapshotBufferFromDisk(uint64_t up_to_log_idx) const +template +nuraft::ptr KeeperSnapshotManager::deserializeSnapshotBufferFromDisk(uint64_t up_to_log_idx) const { const auto & [snapshot_path, snapshot_disk, size] = *existing_snapshots.at(up_to_log_idx); WriteBufferFromNuraftBuffer writer; @@ -710,7 +723,8 @@ nuraft::ptr KeeperSnapshotManager::deserializeSnapshotBufferFrom return writer.getBuffer(); } -nuraft::ptr KeeperSnapshotManager::serializeSnapshotToBuffer(const KeeperStorageSnapshot & snapshot) const +template +nuraft::ptr KeeperSnapshotManager::serializeSnapshotToBuffer(const KeeperStorageSnapshot & snapshot) const { std::unique_ptr writer = std::make_unique(); auto * buffer_raw_ptr = writer.get(); @@ -720,13 +734,13 @@ nuraft::ptr KeeperSnapshotManager::serializeSnapshotToBuffer(con else compressed_writer = std::make_unique(*writer); - KeeperStorageSnapshot::serialize(snapshot, *compressed_writer, keeper_context); + KeeperStorageSnapshot::serialize(snapshot, *compressed_writer, keeper_context); compressed_writer->finalize(); return buffer_raw_ptr->getBuffer(); } - -bool KeeperSnapshotManager::isZstdCompressed(nuraft::ptr buffer) +template +bool KeeperSnapshotManager::isZstdCompressed(nuraft::ptr buffer) { static constexpr unsigned char ZSTD_COMPRESSED_MAGIC[4] = {0x28, 0xB5, 0x2F, 0xFD}; @@ -737,7 +751,8 @@ bool KeeperSnapshotManager::isZstdCompressed(nuraft::ptr buffer) return memcmp(magic_from_buffer, ZSTD_COMPRESSED_MAGIC, 4) == 0; } -SnapshotDeserializationResult KeeperSnapshotManager::deserializeSnapshotFromBuffer(nuraft::ptr buffer) const +template +SnapshotDeserializationResult KeeperSnapshotManager::deserializeSnapshotFromBuffer(nuraft::ptr buffer) const { bool is_zstd_compressed = isZstdCompressed(buffer); @@ -749,14 +764,15 @@ SnapshotDeserializationResult KeeperSnapshotManager::deserializeSnapshotFromBuff else compressed_reader = std::make_unique(*reader); - SnapshotDeserializationResult result; - result.storage = std::make_unique(storage_tick_time, superdigest, keeper_context, /* initialize_system_nodes */ false); - KeeperStorageSnapshot::deserialize(result, *compressed_reader, keeper_context); + SnapshotDeserializationResult result; + result.storage = std::make_unique(storage_tick_time, superdigest, keeper_context, /* initialize_system_nodes */ false); + KeeperStorageSnapshot::deserialize(result, *compressed_reader, keeper_context); result.storage->initializeSystemNodes(); return result; } -SnapshotDeserializationResult KeeperSnapshotManager::restoreFromLatestSnapshot() +template +SnapshotDeserializationResult KeeperSnapshotManager::restoreFromLatestSnapshot() { if (existing_snapshots.empty()) return {}; @@ -767,23 +783,27 @@ SnapshotDeserializationResult KeeperSnapshotManager::restoreFromLatestSnapshot() return deserializeSnapshotFromBuffer(buffer); } -DiskPtr KeeperSnapshotManager::getDisk() const +template +DiskPtr KeeperSnapshotManager::getDisk() const { return keeper_context->getSnapshotDisk(); } -DiskPtr KeeperSnapshotManager::getLatestSnapshotDisk() const +template +DiskPtr KeeperSnapshotManager::getLatestSnapshotDisk() const { return keeper_context->getLatestSnapshotDisk(); } -void KeeperSnapshotManager::removeOutdatedSnapshotsIfNeeded() +template +void KeeperSnapshotManager::removeOutdatedSnapshotsIfNeeded() { while (existing_snapshots.size() > snapshots_to_keep) removeSnapshot(existing_snapshots.begin()->first); } -void KeeperSnapshotManager::moveSnapshotsIfNeeded() +template +void KeeperSnapshotManager::moveSnapshotsIfNeeded() { /// move snapshots to correct disks @@ -813,7 +833,8 @@ void KeeperSnapshotManager::moveSnapshotsIfNeeded() } -void KeeperSnapshotManager::removeSnapshot(uint64_t log_idx) +template +void KeeperSnapshotManager::removeSnapshot(uint64_t log_idx) { auto itr = existing_snapshots.find(log_idx); if (itr == existing_snapshots.end()) @@ -823,7 +844,8 @@ void KeeperSnapshotManager::removeSnapshot(uint64_t log_idx) existing_snapshots.erase(itr); } -SnapshotFileInfoPtr KeeperSnapshotManager::serializeSnapshotToDisk(const KeeperStorageSnapshot & snapshot) +template +SnapshotFileInfoPtr KeeperSnapshotManager::serializeSnapshotToDisk(const KeeperStorageSnapshot & snapshot) { auto up_to_log_idx = snapshot.snapshot_meta->get_last_log_idx(); auto snapshot_file_name = getSnapshotFileName(up_to_log_idx, compress_snapshots_zstd); @@ -842,7 +864,7 @@ SnapshotFileInfoPtr KeeperSnapshotManager::serializeSnapshotToDisk(const KeeperS else compressed_writer = std::make_unique(*writer); - KeeperStorageSnapshot::serialize(snapshot, *compressed_writer, keeper_context); + KeeperStorageSnapshot::serialize(snapshot, *compressed_writer, keeper_context); compressed_writer->finalize(); compressed_writer->sync(); @@ -864,14 +886,16 @@ SnapshotFileInfoPtr KeeperSnapshotManager::serializeSnapshotToDisk(const KeeperS return snapshot_file_info; } -size_t KeeperSnapshotManager::getLatestSnapshotIndex() const +template +size_t KeeperSnapshotManager::getLatestSnapshotIndex() const { if (!existing_snapshots.empty()) return existing_snapshots.rbegin()->first; return 0; } -SnapshotFileInfoPtr KeeperSnapshotManager::getLatestSnapshotInfo() const +template +SnapshotFileInfoPtr KeeperSnapshotManager::getLatestSnapshotInfo() const { if (!existing_snapshots.empty()) { @@ -890,4 +914,10 @@ SnapshotFileInfoPtr KeeperSnapshotManager::getLatestSnapshotInfo() const return nullptr; } +template struct KeeperStorageSnapshot; +template class KeeperSnapshotManager; +#if USE_ROCKSDB +template struct KeeperStorageSnapshot; +template class KeeperSnapshotManager; +#endif } diff --git a/src/Coordination/KeeperSnapshotManager.h b/src/Coordination/KeeperSnapshotManager.h index 36bf6c8775c..be9ce386ab1 100644 --- a/src/Coordination/KeeperSnapshotManager.h +++ b/src/Coordination/KeeperSnapshotManager.h @@ -34,10 +34,11 @@ enum SnapshotVersion : uint8_t static constexpr auto CURRENT_SNAPSHOT_VERSION = SnapshotVersion::V6; /// What is stored in binary snapshot +template struct SnapshotDeserializationResult { /// Storage - KeeperStoragePtr storage; + std::unique_ptr storage; /// Snapshot metadata (up_to_log_idx and so on) SnapshotMetadataPtr snapshot_meta; /// Cluster config @@ -52,21 +53,31 @@ struct SnapshotDeserializationResult /// /// This representation of snapshot have to be serialized into NuRaft /// buffer and send over network or saved to file. +template struct KeeperStorageSnapshot { +#if USE_ROCKSDB + static constexpr bool use_rocksdb = std::is_same_v; +#else + static constexpr bool use_rocksdb = false; +#endif + public: - KeeperStorageSnapshot(KeeperStorage * storage_, uint64_t up_to_log_idx_, const ClusterConfigPtr & cluster_config_ = nullptr); + KeeperStorageSnapshot(Storage * storage_, uint64_t up_to_log_idx_, const ClusterConfigPtr & cluster_config_ = nullptr); KeeperStorageSnapshot( - KeeperStorage * storage_, const SnapshotMetadataPtr & snapshot_meta_, const ClusterConfigPtr & cluster_config_ = nullptr); + Storage * storage_, const SnapshotMetadataPtr & snapshot_meta_, const ClusterConfigPtr & cluster_config_ = nullptr); + + KeeperStorageSnapshot(const KeeperStorageSnapshot&) = delete; + KeeperStorageSnapshot(KeeperStorageSnapshot&&) = default; ~KeeperStorageSnapshot(); - static void serialize(const KeeperStorageSnapshot & snapshot, WriteBuffer & out, KeeperContextPtr keeper_context); + static void serialize(const KeeperStorageSnapshot & snapshot, WriteBuffer & out, KeeperContextPtr keeper_context); - static void deserialize(SnapshotDeserializationResult & deserialization_result, ReadBuffer & in, KeeperContextPtr keeper_context); + static void deserialize(SnapshotDeserializationResult & deserialization_result, ReadBuffer & in, KeeperContextPtr keeper_context); - KeeperStorage * storage; + Storage * storage; SnapshotVersion version = CURRENT_SNAPSHOT_VERSION; /// Snapshot metadata @@ -77,11 +88,11 @@ public: /// so we have for loop for (i = 0; i < snapshot_container_size; ++i) { doSmth(begin + i); } size_t snapshot_container_size; /// Iterator to the start of the storage - KeeperStorage::Container::const_iterator begin; + Storage::Container::const_iterator begin; /// Active sessions and their timeouts SessionAndTimeout session_and_timeout; /// Sessions credentials - KeeperStorage::SessionAndAuth session_and_auth; + Storage::SessionAndAuth session_and_auth; /// ACLs cache for better performance. Without we cannot deserialize storage. std::unordered_map acl_map; /// Cluster config from snapshot, can be empty @@ -105,14 +116,16 @@ struct SnapshotFileInfo }; using SnapshotFileInfoPtr = std::shared_ptr; - -using KeeperStorageSnapshotPtr = std::shared_ptr; -using CreateSnapshotCallback = std::function(KeeperStorageSnapshotPtr &&, bool)>; - -using SnapshotMetaAndStorage = std::pair; +#if USE_ROCKSDB +using KeeperStorageSnapshotPtr = std::variant>, std::shared_ptr>>; +#else +using KeeperStorageSnapshotPtr = std::variant>>; +#endif +using CreateSnapshotCallback = std::function; /// Class responsible for snapshots serialization and deserialization. Each snapshot /// has it's path on disk and log index. +template class KeeperSnapshotManager { public: @@ -124,18 +137,18 @@ public: size_t storage_tick_time_ = 500); /// Restore storage from latest available snapshot - SnapshotDeserializationResult restoreFromLatestSnapshot(); + SnapshotDeserializationResult restoreFromLatestSnapshot(); /// Compress snapshot and serialize it to buffer - nuraft::ptr serializeSnapshotToBuffer(const KeeperStorageSnapshot & snapshot) const; + nuraft::ptr serializeSnapshotToBuffer(const KeeperStorageSnapshot & snapshot) const; /// Serialize already compressed snapshot to disk (return path) SnapshotFileInfoPtr serializeSnapshotBufferToDisk(nuraft::buffer & buffer, uint64_t up_to_log_idx); /// Serialize snapshot directly to disk - SnapshotFileInfoPtr serializeSnapshotToDisk(const KeeperStorageSnapshot & snapshot); + SnapshotFileInfoPtr serializeSnapshotToDisk(const KeeperStorageSnapshot & snapshot); - SnapshotDeserializationResult deserializeSnapshotFromBuffer(nuraft::ptr buffer) const; + SnapshotDeserializationResult deserializeSnapshotFromBuffer(nuraft::ptr buffer) const; /// Deserialize snapshot with log index up_to_log_idx from disk into compressed nuraft buffer. nuraft::ptr deserializeSnapshotBufferFromDisk(uint64_t up_to_log_idx) const; diff --git a/src/Coordination/KeeperStateMachine.cpp b/src/Coordination/KeeperStateMachine.cpp index e4d661dfe17..a4aa1b18746 100644 --- a/src/Coordination/KeeperStateMachine.cpp +++ b/src/Coordination/KeeperStateMachine.cpp @@ -1,12 +1,15 @@ #include #include +#include +#include +#include +#include #include #include -#include #include -#include #include #include +#include #include #include #include @@ -17,7 +20,6 @@ #include #include #include -#include namespace ProfileEvents @@ -31,6 +33,7 @@ namespace ProfileEvents extern const Event KeeperSnapshotApplysFailed; extern const Event KeeperReadSnapshot; extern const Event KeeperSaveSnapshot; + extern const Event KeeperStorageLockWaitMicroseconds; } namespace DB @@ -41,7 +44,7 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } -KeeperStateMachine::KeeperStateMachine( +IKeeperStateMachine::IKeeperStateMachine( ResponsesQueue & responses_queue_, SnapshotsQueue & snapshots_queue_, const KeeperContextPtr & keeper_context_, @@ -49,12 +52,6 @@ KeeperStateMachine::KeeperStateMachine( CommitCallback commit_callback_, const std::string & superdigest_) : commit_callback(commit_callback_) - , snapshot_manager( - keeper_context_->getCoordinationSettings()->snapshots_to_keep, - keeper_context_, - keeper_context_->getCoordinationSettings()->compress_snapshots_with_zstd_format, - superdigest_, - keeper_context_->getCoordinationSettings()->dead_session_check_period_ms.totalMilliseconds()) , responses_queue(responses_queue_) , snapshots_queue(snapshots_queue_) , min_request_size_to_cache(keeper_context_->getCoordinationSettings()->min_request_size_for_cache) @@ -65,6 +62,32 @@ KeeperStateMachine::KeeperStateMachine( { } +template +KeeperStateMachine::KeeperStateMachine( + ResponsesQueue & responses_queue_, + SnapshotsQueue & snapshots_queue_, + // const CoordinationSettingsPtr & coordination_settings_, + const KeeperContextPtr & keeper_context_, + KeeperSnapshotManagerS3 * snapshot_manager_s3_, + IKeeperStateMachine::CommitCallback commit_callback_, + const std::string & superdigest_) + : IKeeperStateMachine( + responses_queue_, + snapshots_queue_, + /// coordination_settings_, + keeper_context_, + snapshot_manager_s3_, + commit_callback_, + superdigest_), + snapshot_manager( + keeper_context_->getCoordinationSettings()->snapshots_to_keep, + keeper_context_, + keeper_context_->getCoordinationSettings()->compress_snapshots_with_zstd_format, + superdigest_, + keeper_context_->getCoordinationSettings()->dead_session_check_period_ms.totalMilliseconds()) +{ +} + namespace { @@ -75,7 +98,8 @@ bool isLocalDisk(const IDisk & disk) } -void KeeperStateMachine::init() +template +void KeeperStateMachine::init() { /// Do everything without mutexes, no other threads exist. LOG_DEBUG(log, "Totally have {} snapshots", snapshot_manager.totalSnapshots()); @@ -120,7 +144,7 @@ void KeeperStateMachine::init() LOG_DEBUG(log, "No existing snapshots, last committed log index {}", last_committed_idx); if (!storage) - storage = std::make_unique( + storage = std::make_unique( keeper_context->getCoordinationSettings()->dead_session_check_period_ms.totalMilliseconds(), superdigest, keeper_context); } @@ -128,13 +152,13 @@ namespace { void assertDigest( - const KeeperStorage::Digest & expected, - const KeeperStorage::Digest & actual, + const KeeperStorageBase::Digest & expected, + const KeeperStorageBase::Digest & actual, const Coordination::ZooKeeperRequest & request, uint64_t log_idx, bool committing) { - if (!KeeperStorage::checkDigest(expected, actual)) + if (!KeeperStorageBase::checkDigest(expected, actual)) { LOG_FATAL( getLogger("KeeperStateMachine"), @@ -151,9 +175,24 @@ void assertDigest( } } +struct TSA_SCOPED_LOCKABLE LockGuardWithStats final +{ + std::unique_lock lock; + explicit LockGuardWithStats(std::mutex & mutex) TSA_ACQUIRE(mutex) + { + Stopwatch watch; + std::unique_lock l(mutex); + ProfileEvents::increment(ProfileEvents::KeeperStorageLockWaitMicroseconds, watch.elapsedMicroseconds()); + lock = std::move(l); + } + + ~LockGuardWithStats() TSA_RELEASE() = default; +}; + } -nuraft::ptr KeeperStateMachine::pre_commit(uint64_t log_idx, nuraft::buffer & data) +template +nuraft::ptr KeeperStateMachine::pre_commit(uint64_t log_idx, nuraft::buffer & data) { auto result = nuraft::buffer::alloc(sizeof(log_idx)); nuraft::buffer_serializer ss(result); @@ -174,10 +213,10 @@ nuraft::ptr KeeperStateMachine::pre_commit(uint64_t log_idx, nur return result; } -std::shared_ptr KeeperStateMachine::parseRequest(nuraft::buffer & data, bool final, ZooKeeperLogSerializationVersion * serialization_version) +std::shared_ptr IKeeperStateMachine::parseRequest(nuraft::buffer & data, bool final, ZooKeeperLogSerializationVersion * serialization_version) { ReadBufferFromNuraftBuffer buffer(data); - auto request_for_session = std::make_shared(); + auto request_for_session = std::make_shared(); readIntBinary(request_for_session->session_id, buffer); int32_t length; @@ -250,7 +289,7 @@ std::shared_ptr KeeperStateMachine::parseReque request_for_session->digest.emplace(); readIntBinary(request_for_session->digest->version, buffer); - if (request_for_session->digest->version != KeeperStorage::DigestVersion::NO_DIGEST || !buffer.eof()) + if (request_for_session->digest->version != KeeperStorageBase::DigestVersion::NO_DIGEST || !buffer.eof()) readIntBinary(request_for_session->digest->value, buffer); } @@ -266,13 +305,14 @@ std::shared_ptr KeeperStateMachine::parseReque return request_for_session; } -bool KeeperStateMachine::preprocess(const KeeperStorage::RequestForSession & request_for_session) +template +bool KeeperStateMachine::preprocess(const KeeperStorageBase::RequestForSession & request_for_session) { const auto op_num = request_for_session.request->getOpNum(); if (op_num == Coordination::OpNum::SessionID || op_num == Coordination::OpNum::Reconfig) return true; - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); if (storage->isFinalized()) return false; @@ -300,10 +340,11 @@ bool KeeperStateMachine::preprocess(const KeeperStorage::RequestForSession & req return true; } -void KeeperStateMachine::reconfigure(const KeeperStorage::RequestForSession& request_for_session) +template +void KeeperStateMachine::reconfigure(const KeeperStorageBase::RequestForSession& request_for_session) { - std::lock_guard _(storage_and_responses_lock); - KeeperStorage::ResponseForSession response = processReconfiguration(request_for_session); + LockGuardWithStats lock(storage_and_responses_lock); + KeeperStorageBase::ResponseForSession response = processReconfiguration(request_for_session); if (!responses_queue.push(response)) { ProfileEvents::increment(ProfileEvents::KeeperCommitsFailed); @@ -313,8 +354,9 @@ void KeeperStateMachine::reconfigure(const KeeperStorage::RequestForSession& req } } -KeeperStorage::ResponseForSession KeeperStateMachine::processReconfiguration( - const KeeperStorage::RequestForSession & request_for_session) +template +KeeperStorageBase::ResponseForSession KeeperStateMachine::processReconfiguration( + const KeeperStorageBase::RequestForSession & request_for_session) { ProfileEvents::increment(ProfileEvents::KeeperReconfigRequest); @@ -323,7 +365,7 @@ KeeperStorage::ResponseForSession KeeperStateMachine::processReconfiguration( const int64_t zxid = request_for_session.zxid; using enum Coordination::Error; - auto bad_request = [&](Coordination::Error code = ZBADARGUMENTS) -> KeeperStorage::ResponseForSession + auto bad_request = [&](Coordination::Error code = ZBADARGUMENTS) -> KeeperStorageBase::ResponseForSession { auto res = std::make_shared(); res->xid = request.xid; @@ -380,7 +422,8 @@ KeeperStorage::ResponseForSession KeeperStateMachine::processReconfiguration( return { session_id, std::move(response) }; } -nuraft::ptr KeeperStateMachine::commit(const uint64_t log_idx, nuraft::buffer & data) +template +nuraft::ptr KeeperStateMachine::commit(const uint64_t log_idx, nuraft::buffer & data) { auto request_for_session = parseRequest(data, true); if (!request_for_session->zxid) @@ -391,7 +434,7 @@ nuraft::ptr KeeperStateMachine::commit(const uint64_t log_idx, n if (!keeper_context->localLogsPreprocessed() && !preprocess(*request_for_session)) return nullptr; - auto try_push = [this](const KeeperStorage::ResponseForSession& response) + auto try_push = [&](const KeeperStorageBase::ResponseForSession & response) { if (!responses_queue.push(response)) { @@ -413,11 +456,12 @@ nuraft::ptr KeeperStateMachine::commit(const uint64_t log_idx, n std::shared_ptr response = std::make_shared(); response->internal_id = session_id_request.internal_id; response->server_id = session_id_request.server_id; - KeeperStorage::ResponseForSession response_for_session; + KeeperStorageBase::ResponseForSession response_for_session; response_for_session.session_id = -1; response_for_session.response = response; + response_for_session.request = request_for_session->request; - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); session_id = storage->getSessionID(session_id_request.session_timeout_ms); LOG_DEBUG(log, "Session ID response {} with timeout {}", session_id, session_id_request.session_timeout_ms); response->session_id = session_id; @@ -426,16 +470,23 @@ nuraft::ptr KeeperStateMachine::commit(const uint64_t log_idx, n else { if (op_num == Coordination::OpNum::Close) + { std::lock_guard lock(request_cache_mutex); parsed_request_cache.erase(request_for_session->session_id); } - std::lock_guard lock(storage_and_responses_lock); - KeeperStorage::ResponsesForSessions responses_for_sessions + LockGuardWithStats lock(storage_and_responses_lock); + KeeperStorageBase::ResponsesForSessions responses_for_sessions = storage->processRequest(request_for_session->request, request_for_session->session_id, request_for_session->zxid); + for (auto & response_for_session : responses_for_sessions) + { + if (response_for_session.response->xid != Coordination::WATCH_XID) + response_for_session.request = request_for_session->request; + try_push(response_for_session); + } if (keeper_context->digestEnabled() && request_for_session->digest) assertDigest(*request_for_session->digest, storage->getNodesDigest(true), *request_for_session->request, request_for_session->log_idx, true); @@ -457,7 +508,8 @@ nuraft::ptr KeeperStateMachine::commit(const uint64_t log_idx, n return nullptr; } -bool KeeperStateMachine::apply_snapshot(nuraft::snapshot & s) +template +bool KeeperStateMachine::apply_snapshot(nuraft::snapshot & s) { LOG_DEBUG(log, "Applying snapshot {}", s.get_last_log_idx()); nuraft::ptr latest_snapshot_ptr; @@ -482,9 +534,9 @@ bool KeeperStateMachine::apply_snapshot(nuraft::snapshot & s) } { /// deserialize and apply snapshot to storage - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); - SnapshotDeserializationResult snapshot_deserialization_result; + SnapshotDeserializationResult snapshot_deserialization_result; if (latest_snapshot_ptr) snapshot_deserialization_result = snapshot_manager.deserializeSnapshotFromBuffer(latest_snapshot_ptr); else @@ -505,7 +557,7 @@ bool KeeperStateMachine::apply_snapshot(nuraft::snapshot & s) } -void KeeperStateMachine::commit_config(const uint64_t log_idx, nuraft::ptr & new_conf) +void IKeeperStateMachine::commit_config(const uint64_t log_idx, nuraft::ptr & new_conf) { std::lock_guard lock(cluster_config_lock); auto tmp = new_conf->serialize(); @@ -513,7 +565,7 @@ void KeeperStateMachine::commit_config(const uint64_t log_idx, nuraft::ptrsetLastCommitIndex(log_idx); } -void KeeperStateMachine::rollback(uint64_t log_idx, nuraft::buffer & data) +void IKeeperStateMachine::rollback(uint64_t log_idx, nuraft::buffer & data) { /// Don't rollback anything until the first commit because nothing was preprocessed if (!keeper_context->localLogsPreprocessed()) @@ -529,16 +581,18 @@ void KeeperStateMachine::rollback(uint64_t log_idx, nuraft::buffer & data) rollbackRequest(*request_for_session, false); } -void KeeperStateMachine::rollbackRequest(const KeeperStorage::RequestForSession & request_for_session, bool allow_missing) +template +void KeeperStateMachine::rollbackRequest(const KeeperStorageBase::RequestForSession & request_for_session, bool allow_missing) { if (request_for_session.request->getOpNum() == Coordination::OpNum::SessionID) return; - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); storage->rollbackRequest(request_for_session.zxid, allow_missing); } -void KeeperStateMachine::rollbackRequestNoLock(const KeeperStorage::RequestForSession & request_for_session, bool allow_missing) +template +void KeeperStateMachine::rollbackRequestNoLock(const KeeperStorageBase::RequestForSession & request_for_session, bool allow_missing) { if (request_for_session.request->getOpNum() == Coordination::OpNum::SessionID) return; @@ -546,14 +600,15 @@ void KeeperStateMachine::rollbackRequestNoLock(const KeeperStorage::RequestForSe storage->rollbackRequest(request_for_session.zxid, allow_missing); } -nuraft::ptr KeeperStateMachine::last_snapshot() +nuraft::ptr IKeeperStateMachine::last_snapshot() { /// Just return the latest snapshot. std::lock_guard lock(snapshots_lock); return latest_snapshot_meta; } -void KeeperStateMachine::create_snapshot(nuraft::snapshot & s, nuraft::async_result::handler_type & when_done) +template +void KeeperStateMachine::create_snapshot(nuraft::snapshot & s, nuraft::async_result::handler_type & when_done) { LOG_DEBUG(log, "Creating snapshot {}", s.get_last_log_idx()); @@ -561,15 +616,16 @@ void KeeperStateMachine::create_snapshot(nuraft::snapshot & s, nuraft::async_res auto snapshot_meta_copy = nuraft::snapshot::deserialize(*snp_buf); CreateSnapshotTask snapshot_task; { /// lock storage for a short period time to turn on "snapshot mode". After that we can read consistent storage state without locking. - std::lock_guard lock(storage_and_responses_lock); - snapshot_task.snapshot = std::make_shared(storage.get(), snapshot_meta_copy, getClusterConfig()); + LockGuardWithStats lock(storage_and_responses_lock); + snapshot_task.snapshot = std::make_shared>(storage.get(), snapshot_meta_copy, getClusterConfig()); } /// create snapshot task for background execution (in snapshot thread) - snapshot_task.create_snapshot = [this, when_done](KeeperStorageSnapshotPtr && snapshot, bool execute_only_cleanup) + snapshot_task.create_snapshot = [this, when_done](KeeperStorageSnapshotPtr && snapshot_, bool execute_only_cleanup) { nuraft::ptr exception(nullptr); - bool ret = true; + bool ret = false; + auto && snapshot = std::get>>(std::move(snapshot_)); if (!execute_only_cleanup) { try @@ -599,7 +655,8 @@ void KeeperStateMachine::create_snapshot(nuraft::snapshot & s, nuraft::async_res else { auto snapshot_buf = snapshot_manager.serializeSnapshotToBuffer(*snapshot); - auto snapshot_info = snapshot_manager.serializeSnapshotBufferToDisk(*snapshot_buf, snapshot->snapshot_meta->get_last_log_idx()); + auto snapshot_info = snapshot_manager.serializeSnapshotBufferToDisk( + *snapshot_buf, snapshot->snapshot_meta->get_last_log_idx()); latest_snapshot_info = std::move(snapshot_info); latest_snapshot_buf = std::move(snapshot_buf); } @@ -612,18 +669,19 @@ void KeeperStateMachine::create_snapshot(nuraft::snapshot & s, nuraft::async_res latest_snapshot_info->path); } } + + ret = true; } catch (...) { ProfileEvents::increment(ProfileEvents::KeeperSnapshotCreationsFailed); LOG_TRACE(log, "Exception happened during snapshot"); tryLogCurrentException(log); - ret = false; } } { /// Destroy snapshot with lock - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); LOG_TRACE(log, "Clearing garbage after snapshot"); /// Turn off "snapshot mode" and clear outdate part of storage state storage->clearGarbageAfterSnapshot(); @@ -656,7 +714,8 @@ void KeeperStateMachine::create_snapshot(nuraft::snapshot & s, nuraft::async_res LOG_WARNING(log, "Cannot push snapshot task into queue"); } -void KeeperStateMachine::save_logical_snp_obj( +template +void KeeperStateMachine::save_logical_snp_obj( nuraft::snapshot & s, uint64_t & obj_id, nuraft::buffer & data, bool /*is_first_obj*/, bool /*is_last_obj*/) { LOG_DEBUG(log, "Saving snapshot {} obj_id {}", s.get_last_log_idx(), obj_id); @@ -721,7 +780,7 @@ static int bufferFromFile(LoggerPtr log, const std::string & path, nuraft::ptr & data_out, bool & is_last_obj) { LOG_DEBUG(log, "Reading snapshot {} obj_id {}", s.get_last_log_idx(), obj_id); @@ -761,114 +820,137 @@ int KeeperStateMachine::read_logical_snp_obj( return 1; } -void KeeperStateMachine::processReadRequest(const KeeperStorage::RequestForSession & request_for_session) +template +void KeeperStateMachine::processReadRequest(const KeeperStorageBase::RequestForSession & request_for_session) { /// Pure local request, just process it with storage - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); auto responses = storage->processRequest( request_for_session.request, request_for_session.session_id, std::nullopt, true /*check_acl*/, true /*is_local*/); - for (const auto & response : responses) - if (!responses_queue.push(response)) - LOG_WARNING(log, "Failed to push response with session id {} to the queue, probably because of shutdown", response.session_id); + + for (auto & response_for_session : responses) + { + if (response_for_session.response->xid != Coordination::WATCH_XID) + response_for_session.request = request_for_session.request; + if (!responses_queue.push(response_for_session)) + LOG_WARNING(log, "Failed to push response with session id {} to the queue, probably because of shutdown", response_for_session.session_id); + } } -void KeeperStateMachine::shutdownStorage() +template +void KeeperStateMachine::shutdownStorage() { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); storage->finalize(); } -std::vector KeeperStateMachine::getDeadSessions() +template +std::vector KeeperStateMachine::getDeadSessions() { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getDeadSessions(); } -int64_t KeeperStateMachine::getNextZxid() const +template +int64_t KeeperStateMachine::getNextZxid() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getNextZXID(); } -KeeperStorage::Digest KeeperStateMachine::getNodesDigest() const +template +KeeperStorageBase::Digest KeeperStateMachine::getNodesDigest() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getNodesDigest(false); } -uint64_t KeeperStateMachine::getLastProcessedZxid() const +template +uint64_t KeeperStateMachine::getLastProcessedZxid() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getZXID(); } -uint64_t KeeperStateMachine::getNodesCount() const +template +uint64_t KeeperStateMachine::getNodesCount() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getNodesCount(); } -uint64_t KeeperStateMachine::getTotalWatchesCount() const +template +uint64_t KeeperStateMachine::getTotalWatchesCount() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getTotalWatchesCount(); } -uint64_t KeeperStateMachine::getWatchedPathsCount() const +template +uint64_t KeeperStateMachine::getWatchedPathsCount() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getWatchedPathsCount(); } -uint64_t KeeperStateMachine::getSessionsWithWatchesCount() const +template +uint64_t KeeperStateMachine::getSessionsWithWatchesCount() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getSessionsWithWatchesCount(); } -uint64_t KeeperStateMachine::getTotalEphemeralNodesCount() const +template +uint64_t KeeperStateMachine::getTotalEphemeralNodesCount() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getTotalEphemeralNodesCount(); } -uint64_t KeeperStateMachine::getSessionWithEphemeralNodesCount() const +template +uint64_t KeeperStateMachine::getSessionWithEphemeralNodesCount() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getSessionWithEphemeralNodesCount(); } -void KeeperStateMachine::dumpWatches(WriteBufferFromOwnString & buf) const +template +void KeeperStateMachine::dumpWatches(WriteBufferFromOwnString & buf) const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); storage->dumpWatches(buf); } -void KeeperStateMachine::dumpWatchesByPath(WriteBufferFromOwnString & buf) const +template +void KeeperStateMachine::dumpWatchesByPath(WriteBufferFromOwnString & buf) const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); storage->dumpWatchesByPath(buf); } -void KeeperStateMachine::dumpSessionsAndEphemerals(WriteBufferFromOwnString & buf) const +template +void KeeperStateMachine::dumpSessionsAndEphemerals(WriteBufferFromOwnString & buf) const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); storage->dumpSessionsAndEphemerals(buf); } -uint64_t KeeperStateMachine::getApproximateDataSize() const +template +uint64_t KeeperStateMachine::getApproximateDataSize() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getApproximateDataSize(); } -uint64_t KeeperStateMachine::getKeyArenaSize() const +template +uint64_t KeeperStateMachine::getKeyArenaSize() const { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); return storage->getArenaDataSize(); } -uint64_t KeeperStateMachine::getLatestSnapshotSize() const +template +uint64_t KeeperStateMachine::getLatestSnapshotSize() const { auto snapshot_info = [&] { @@ -891,7 +973,7 @@ uint64_t KeeperStateMachine::getLatestSnapshotSize() const return size; } -ClusterConfigPtr KeeperStateMachine::getClusterConfig() const +ClusterConfigPtr IKeeperStateMachine::getClusterConfig() const { std::lock_guard lock(cluster_config_lock); if (cluster_config) @@ -903,11 +985,18 @@ ClusterConfigPtr KeeperStateMachine::getClusterConfig() const return nullptr; } -void KeeperStateMachine::recalculateStorageStats() +template +void KeeperStateMachine::recalculateStorageStats() { - std::lock_guard lock(storage_and_responses_lock); + LockGuardWithStats lock(storage_and_responses_lock); LOG_INFO(log, "Recalculating storage stats"); storage->recalculateStats(); LOG_INFO(log, "Done recalculating storage stats"); } + +template class KeeperStateMachine; +#if USE_ROCKSDB +template class KeeperStateMachine; +#endif + } diff --git a/src/Coordination/KeeperStateMachine.h b/src/Coordination/KeeperStateMachine.h index ee6109f0a17..6afd413d782 100644 --- a/src/Coordination/KeeperStateMachine.h +++ b/src/Coordination/KeeperStateMachine.h @@ -1,6 +1,5 @@ #pragma once -#include #include #include #include @@ -12,26 +11,24 @@ namespace DB { -using ResponsesQueue = ConcurrentBoundedQueue; +using ResponsesQueue = ConcurrentBoundedQueue; using SnapshotsQueue = ConcurrentBoundedQueue; -/// ClickHouse Keeper state machine. Wrapper for KeeperStorage. -/// Responsible for entries commit, snapshots creation and so on. -class KeeperStateMachine : public nuraft::state_machine +class IKeeperStateMachine : public nuraft::state_machine { public: - using CommitCallback = std::function; + using CommitCallback = std::function; - KeeperStateMachine( + IKeeperStateMachine( ResponsesQueue & responses_queue_, SnapshotsQueue & snapshots_queue_, const KeeperContextPtr & keeper_context_, KeeperSnapshotManagerS3 * snapshot_manager_s3_, - CommitCallback commit_callback_ = {}, - const std::string & superdigest_ = ""); + CommitCallback commit_callback_, + const std::string & superdigest_); /// Read state from the latest snapshot - void init(); + virtual void init() = 0; enum ZooKeeperLogSerializationVersion { @@ -48,89 +45,66 @@ public: /// /// final - whether it's the final time we will fetch the request so we can safely remove it from cache /// serialization_version - information about which fields were parsed from the buffer so we can modify the buffer accordingly - std::shared_ptr parseRequest(nuraft::buffer & data, bool final, ZooKeeperLogSerializationVersion * serialization_version = nullptr); + std::shared_ptr parseRequest(nuraft::buffer & data, bool final, ZooKeeperLogSerializationVersion * serialization_version = nullptr); - bool preprocess(const KeeperStorage::RequestForSession & request_for_session); + virtual bool preprocess(const KeeperStorageBase::RequestForSession & request_for_session) = 0; - nuraft::ptr pre_commit(uint64_t log_idx, nuraft::buffer & data) override; - - nuraft::ptr commit(const uint64_t log_idx, nuraft::buffer & data) override; /// NOLINT - - /// Save new cluster config to our snapshot (copy of the config stored in StateManager) void commit_config(const uint64_t log_idx, nuraft::ptr & new_conf) override; /// NOLINT void rollback(uint64_t log_idx, nuraft::buffer & data) override; // allow_missing - whether the transaction we want to rollback can be missing from storage // (can happen in case of exception during preprocessing) - void rollbackRequest(const KeeperStorage::RequestForSession & request_for_session, bool allow_missing); - - void rollbackRequestNoLock( - const KeeperStorage::RequestForSession & request_for_session, - bool allow_missing) TSA_NO_THREAD_SAFETY_ANALYSIS; + virtual void rollbackRequest(const KeeperStorageBase::RequestForSession & request_for_session, bool allow_missing) = 0; uint64_t last_commit_index() override { return keeper_context->lastCommittedIndex(); } - /// Apply preliminarily saved (save_logical_snp_obj) snapshot to our state. - bool apply_snapshot(nuraft::snapshot & s) override; - nuraft::ptr last_snapshot() override; /// Create new snapshot from current state. - void create_snapshot(nuraft::snapshot & s, nuraft::async_result::handler_type & when_done) override; + void create_snapshot(nuraft::snapshot & s, nuraft::async_result::handler_type & when_done) override = 0; /// Save snapshot which was send by leader to us. After that we will apply it in apply_snapshot. - void save_logical_snp_obj(nuraft::snapshot & s, uint64_t & obj_id, nuraft::buffer & data, bool is_first_obj, bool is_last_obj) override; + void save_logical_snp_obj(nuraft::snapshot & s, uint64_t & obj_id, nuraft::buffer & data, bool is_first_obj, bool is_last_obj) override = 0; - /// Better name is `serialize snapshot` -- save existing snapshot (created by create_snapshot) into - /// in-memory buffer data_out. int read_logical_snp_obj( nuraft::snapshot & s, void *& user_snp_ctx, uint64_t obj_id, nuraft::ptr & data_out, bool & is_last_obj) override; - // This should be used only for tests or keeper-data-dumper because it violates - // TSA -- we can't acquire the lock outside of this class or return a storage under lock - // in a reasonable way. - KeeperStorage & getStorageUnsafe() TSA_NO_THREAD_SAFETY_ANALYSIS - { - return *storage; - } - - void shutdownStorage(); + virtual void shutdownStorage() = 0; ClusterConfigPtr getClusterConfig() const; - /// Process local read request - void processReadRequest(const KeeperStorage::RequestForSession & request_for_session); + virtual void processReadRequest(const KeeperStorageBase::RequestForSession & request_for_session) = 0; - std::vector getDeadSessions(); + virtual std::vector getDeadSessions() = 0; - int64_t getNextZxid() const; + virtual int64_t getNextZxid() const = 0; - KeeperStorage::Digest getNodesDigest() const; + virtual KeeperStorageBase::Digest getNodesDigest() const = 0; /// Introspection functions for 4lw commands - uint64_t getLastProcessedZxid() const; + virtual uint64_t getLastProcessedZxid() const = 0; - uint64_t getNodesCount() const; - uint64_t getTotalWatchesCount() const; - uint64_t getWatchedPathsCount() const; - uint64_t getSessionsWithWatchesCount() const; + virtual uint64_t getNodesCount() const = 0; + virtual uint64_t getTotalWatchesCount() const = 0; + virtual uint64_t getWatchedPathsCount() const = 0; + virtual uint64_t getSessionsWithWatchesCount() const = 0; - void dumpWatches(WriteBufferFromOwnString & buf) const; - void dumpWatchesByPath(WriteBufferFromOwnString & buf) const; - void dumpSessionsAndEphemerals(WriteBufferFromOwnString & buf) const; + virtual void dumpWatches(WriteBufferFromOwnString & buf) const = 0; + virtual void dumpWatchesByPath(WriteBufferFromOwnString & buf) const = 0; + virtual void dumpSessionsAndEphemerals(WriteBufferFromOwnString & buf) const = 0; - uint64_t getSessionWithEphemeralNodesCount() const; - uint64_t getTotalEphemeralNodesCount() const; - uint64_t getApproximateDataSize() const; - uint64_t getKeyArenaSize() const; - uint64_t getLatestSnapshotSize() const; + virtual uint64_t getSessionWithEphemeralNodesCount() const = 0; + virtual uint64_t getTotalEphemeralNodesCount() const = 0; + virtual uint64_t getApproximateDataSize() const = 0; + virtual uint64_t getKeyArenaSize() const = 0; + virtual uint64_t getLatestSnapshotSize() const = 0; - void recalculateStorageStats(); + virtual void recalculateStorageStats() = 0; - void reconfigure(const KeeperStorage::RequestForSession& request_for_session); + virtual void reconfigure(const KeeperStorageBase::RequestForSession& request_for_session) = 0; -private: +protected: CommitCallback commit_callback; /// In our state machine we always have a single snapshot which is stored /// in memory in compressed (serialized) format. @@ -138,12 +112,9 @@ private: std::shared_ptr latest_snapshot_info; nuraft::ptr latest_snapshot_buf = nullptr; - /// Main state machine logic - KeeperStoragePtr storage TSA_PT_GUARDED_BY(storage_and_responses_lock); + CoordinationSettingsPtr coordination_settings; /// Save/Load and Serialize/Deserialize logic for snapshots. - KeeperSnapshotManager snapshot_manager; - /// Put processed responses into this queue ResponsesQueue & responses_queue; @@ -160,7 +131,7 @@ private: /// for request. mutable std::mutex storage_and_responses_lock; - std::unordered_map>> parsed_request_cache; + std::unordered_map>> parsed_request_cache; uint64_t min_request_size_to_cache{0}; /// we only need to protect the access to the map itself /// requests can be modified from anywhere without lock because a single request @@ -182,8 +153,104 @@ private: KeeperSnapshotManagerS3 * snapshot_manager_s3; - KeeperStorage::ResponseForSession processReconfiguration( - const KeeperStorage::RequestForSession& request_for_session) - TSA_REQUIRES(storage_and_responses_lock); + virtual KeeperStorageBase::ResponseForSession processReconfiguration( + const KeeperStorageBase::RequestForSession& request_for_session) + TSA_REQUIRES(storage_and_responses_lock) = 0; + +}; + +/// ClickHouse Keeper state machine. Wrapper for KeeperStorage. +/// Responsible for entries commit, snapshots creation and so on. +template +class KeeperStateMachine : public IKeeperStateMachine +{ +public: + /// using CommitCallback = std::function; + + KeeperStateMachine( + ResponsesQueue & responses_queue_, + SnapshotsQueue & snapshots_queue_, + /// const CoordinationSettingsPtr & coordination_settings_, + const KeeperContextPtr & keeper_context_, + KeeperSnapshotManagerS3 * snapshot_manager_s3_, + CommitCallback commit_callback_ = {}, + const std::string & superdigest_ = ""); + + /// Read state from the latest snapshot + void init() override; + + bool preprocess(const KeeperStorageBase::RequestForSession & request_for_session) override; + + nuraft::ptr pre_commit(uint64_t log_idx, nuraft::buffer & data) override; + + nuraft::ptr commit(const uint64_t log_idx, nuraft::buffer & data) override; /// NOLINT + + // allow_missing - whether the transaction we want to rollback can be missing from storage + // (can happen in case of exception during preprocessing) + void rollbackRequest(const KeeperStorageBase::RequestForSession & request_for_session, bool allow_missing) override; + + void rollbackRequestNoLock( + const KeeperStorageBase::RequestForSession & request_for_session, + bool allow_missing) TSA_NO_THREAD_SAFETY_ANALYSIS; + + /// Apply preliminarily saved (save_logical_snp_obj) snapshot to our state. + bool apply_snapshot(nuraft::snapshot & s) override; + + /// Create new snapshot from current state. + void create_snapshot(nuraft::snapshot & s, nuraft::async_result::handler_type & when_done) override; + + /// Save snapshot which was send by leader to us. After that we will apply it in apply_snapshot. + void save_logical_snp_obj(nuraft::snapshot & s, uint64_t & obj_id, nuraft::buffer & data, bool is_first_obj, bool is_last_obj) override; + + // This should be used only for tests or keeper-data-dumper because it violates + // TSA -- we can't acquire the lock outside of this class or return a storage under lock + // in a reasonable way. + Storage & getStorageUnsafe() TSA_NO_THREAD_SAFETY_ANALYSIS + { + return *storage; + } + + void shutdownStorage() override; + + /// Process local read request + void processReadRequest(const KeeperStorageBase::RequestForSession & request_for_session) override; + + std::vector getDeadSessions() override; + + int64_t getNextZxid() const override; + + KeeperStorageBase::Digest getNodesDigest() const override; + + /// Introspection functions for 4lw commands + uint64_t getLastProcessedZxid() const override; + + uint64_t getNodesCount() const override; + uint64_t getTotalWatchesCount() const override; + uint64_t getWatchedPathsCount() const override; + uint64_t getSessionsWithWatchesCount() const override; + + void dumpWatches(WriteBufferFromOwnString & buf) const override; + void dumpWatchesByPath(WriteBufferFromOwnString & buf) const override; + void dumpSessionsAndEphemerals(WriteBufferFromOwnString & buf) const override; + + uint64_t getSessionWithEphemeralNodesCount() const override; + uint64_t getTotalEphemeralNodesCount() const override; + uint64_t getApproximateDataSize() const override; + uint64_t getKeyArenaSize() const override; + uint64_t getLatestSnapshotSize() const override; + + void recalculateStorageStats() override; + + void reconfigure(const KeeperStorageBase::RequestForSession& request_for_session) override; + +private: + /// Main state machine logic + std::unique_ptr storage; //TSA_PT_GUARDED_BY(storage_and_responses_lock); + + /// Save/Load and Serialize/Deserialize logic for snapshots. + KeeperSnapshotManager snapshot_manager; + + KeeperStorageBase::ResponseForSession processReconfiguration(const KeeperStorageBase::RequestForSession & request_for_session) + TSA_REQUIRES(storage_and_responses_lock) override; }; } diff --git a/src/Coordination/KeeperStateManager.h b/src/Coordination/KeeperStateManager.h index 60f6dbe7b62..e5ab5fb4807 100644 --- a/src/Coordination/KeeperStateManager.h +++ b/src/Coordination/KeeperStateManager.h @@ -1,14 +1,13 @@ #pragma once -#include #include #include -#include +#include +#include #include #include #include "Coordination/KeeperStateMachine.h" #include "Coordination/RaftServerConfig.h" -#include namespace DB { diff --git a/src/Coordination/KeeperStorage.cpp b/src/Coordination/KeeperStorage.cpp index 9251624346b..acdf209baae 100644 --- a/src/Coordination/KeeperStorage.cpp +++ b/src/Coordination/KeeperStorage.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -20,11 +21,12 @@ #include #include +#include #include #include +#include #include #include -#include #include #include @@ -40,6 +42,8 @@ namespace ProfileEvents extern const Event KeeperGetRequest; extern const Event KeeperListRequest; extern const Event KeeperExistsRequest; + extern const Event KeeperPreprocessElapsedMicroseconds; + extern const Event KeeperProcessElapsedMicroseconds; } namespace DB @@ -61,10 +65,11 @@ String getSHA1(const String & userdata) return String{digest_id.begin(), digest_id.end()}; } +template bool fixupACL( const std::vector & request_acls, int64_t session_id, - const KeeperStorage::UncommittedState & uncommitted_state, + const UncommittedState & uncommitted_state, std::vector & result_acls) { if (request_acls.empty()) @@ -77,7 +82,7 @@ bool fixupACL( { uncommitted_state.forEachAuthInSession( session_id, - [&](const KeeperStorage::AuthID & auth_id) + [&](const KeeperStorageBase::AuthID & auth_id) { valid_found = true; Coordination::ACL new_acl = request_acl; @@ -108,10 +113,10 @@ bool fixupACL( return valid_found; } -KeeperStorage::ResponsesForSessions processWatchesImpl( - const String & path, KeeperStorage::Watches & watches, KeeperStorage::Watches & list_watches, Coordination::Event event_type) +KeeperStorageBase::ResponsesForSessions processWatchesImpl( + const String & path, KeeperStorageBase::Watches & watches, KeeperStorageBase::Watches & list_watches, Coordination::Event event_type) { - KeeperStorage::ResponsesForSessions result; + KeeperStorageBase::ResponsesForSessions result; auto watch_it = watches.find(path); if (watch_it != watches.end()) { @@ -122,7 +127,7 @@ KeeperStorage::ResponsesForSessions processWatchesImpl( watch_response->type = event_type; watch_response->state = Coordination::State::CONNECTED; for (auto watcher_session : watch_it->second) - result.push_back(KeeperStorage::ResponseForSession{watcher_session, watch_response}); + result.push_back(KeeperStorageBase::ResponseForSession{watcher_session, watch_response}); watches.erase(watch_it); } @@ -158,7 +163,7 @@ KeeperStorage::ResponsesForSessions processWatchesImpl( watch_list_response->state = Coordination::State::CONNECTED; for (auto watcher_session : watch_it->second) - result.push_back(KeeperStorage::ResponseForSession{watcher_session, watch_list_response}); + result.push_back(KeeperStorageBase::ResponseForSession{watcher_session, watch_list_response}); list_watches.erase(watch_it); } @@ -167,7 +172,8 @@ KeeperStorage::ResponsesForSessions processWatchesImpl( } // When this function is updated, update CURRENT_DIGEST_VERSION!! -uint64_t calculateDigest(std::string_view path, const KeeperStorage::Node & node) +template +uint64_t calculateDigest(std::string_view path, const Node & node) { SipHash hash; @@ -202,7 +208,71 @@ uint64_t calculateDigest(std::string_view path, const KeeperStorage::Node & node } -KeeperStorage::Node & KeeperStorage::Node::operator=(const Node & other) +void KeeperRocksNodeInfo::copyStats(const Coordination::Stat & stat) +{ + czxid = stat.czxid; + mzxid = stat.mzxid; + pzxid = stat.pzxid; + + mtime = stat.mtime; + setCtime(stat.ctime); + + version = stat.version; + cversion = stat.cversion; + aversion = stat.aversion; + + if (stat.ephemeralOwner == 0) + { + is_ephemeral_and_ctime.is_ephemeral = false; + setNumChildren(stat.numChildren); + } + else + { + setEphemeralOwner(stat.ephemeralOwner); + } +} + +void KeeperRocksNode::invalidateDigestCache() const +{ + if (serialized) + throw Exception(ErrorCodes::LOGICAL_ERROR, "We modify node after serialized it"); + digest = 0; +} + +UInt64 KeeperRocksNode::getDigest(std::string_view path) const +{ + if (!digest) + digest = calculateDigest(path, *this); + return digest; +} + +String KeeperRocksNode::getEncodedString() +{ + if (serialized) + throw Exception(ErrorCodes::LOGICAL_ERROR, "We modify node after serialized it"); + serialized = true; + + WriteBufferFromOwnString buffer; + const KeeperRocksNodeInfo & node_info = *this; + writePODBinary(node_info, buffer); + writeBinary(getData(), buffer); + return buffer.str(); +} + +void KeeperRocksNode::decodeFromString(const String &buffer_str) +{ + ReadBufferFromOwnString buffer(buffer_str); + KeeperRocksNodeInfo & node_info = *this; + readPODBinary(node_info, buffer); + readVarUInt(data_size, buffer); + if (data_size) + { + data = std::unique_ptr(new char[data_size]); + buffer.readStrict(data.get(), data_size); + } +} + +KeeperMemNode & KeeperMemNode::operator=(const KeeperMemNode & other) { if (this == &other) return *this; @@ -230,12 +300,12 @@ KeeperStorage::Node & KeeperStorage::Node::operator=(const Node & other) return *this; } -KeeperStorage::Node::Node(const Node & other) +KeeperMemNode::KeeperMemNode(const KeeperMemNode & other) { *this = other; } -KeeperStorage::Node & KeeperStorage::Node::operator=(Node && other) noexcept +KeeperMemNode & KeeperMemNode::operator=(KeeperMemNode && other) noexcept { if (this == &other) return *this; @@ -262,17 +332,17 @@ KeeperStorage::Node & KeeperStorage::Node::operator=(Node && other) noexcept return *this; } -KeeperStorage::Node::Node(Node && other) noexcept +KeeperMemNode::KeeperMemNode(KeeperMemNode && other) noexcept { *this = std::move(other); } -bool KeeperStorage::Node::empty() const +bool KeeperMemNode::empty() const { return data_size == 0 && mzxid == 0; } -void KeeperStorage::Node::copyStats(const Coordination::Stat & stat) +void KeeperMemNode::copyStats(const Coordination::Stat & stat) { czxid = stat.czxid; mzxid = stat.mzxid; @@ -296,7 +366,7 @@ void KeeperStorage::Node::copyStats(const Coordination::Stat & stat) } } -void KeeperStorage::Node::setResponseStat(Coordination::Stat & response_stat) const +void KeeperMemNode::setResponseStat(Coordination::Stat & response_stat) const { response_stat.czxid = czxid; response_stat.mzxid = mzxid; @@ -312,12 +382,12 @@ void KeeperStorage::Node::setResponseStat(Coordination::Stat & response_stat) co } -uint64_t KeeperStorage::Node::sizeInBytes() const +uint64_t KeeperMemNode::sizeInBytes() const { - return sizeof(Node) + children.size() * sizeof(StringRef) + data_size; + return sizeof(KeeperMemNode) + children.size() * sizeof(StringRef) + data_size; } -void KeeperStorage::Node::setData(const String & new_data) +void KeeperMemNode::setData(const String & new_data) { data_size = static_cast(new_data.size()); if (data_size != 0) @@ -327,22 +397,22 @@ void KeeperStorage::Node::setData(const String & new_data) } } -void KeeperStorage::Node::addChild(StringRef child_path) +void KeeperMemNode::addChild(StringRef child_path) { children.insert(child_path); } -void KeeperStorage::Node::removeChild(StringRef child_path) +void KeeperMemNode::removeChild(StringRef child_path) { children.erase(child_path); } -void KeeperStorage::Node::invalidateDigestCache() const +void KeeperMemNode::invalidateDigestCache() const { cached_digest = 0; } -UInt64 KeeperStorage::Node::getDigest(const std::string_view path) const +UInt64 KeeperMemNode::getDigest(const std::string_view path) const { if (cached_digest == 0) cached_digest = calculateDigest(path, *this); @@ -350,7 +420,7 @@ UInt64 KeeperStorage::Node::getDigest(const std::string_view path) const return cached_digest; }; -void KeeperStorage::Node::shallowCopy(const KeeperStorage::Node & other) +void KeeperMemNode::shallowCopy(const KeeperMemNode & other) { czxid = other.czxid; mzxid = other.mzxid; @@ -377,19 +447,25 @@ void KeeperStorage::Node::shallowCopy(const KeeperStorage::Node & other) cached_digest = other.cached_digest; } -KeeperStorage::KeeperStorage( + +template +KeeperStorage::KeeperStorage( int64_t tick_time_ms, const String & superdigest_, const KeeperContextPtr & keeper_context_, const bool initialize_system_nodes) : session_expiry_queue(tick_time_ms), keeper_context(keeper_context_), superdigest(superdigest_) { + if constexpr (use_rocksdb) + container.initialize(keeper_context); Node root_node; container.insert("/", root_node); - addDigest(root_node, "/"); + if constexpr (!use_rocksdb) + addDigest(root_node, "/"); if (initialize_system_nodes) initializeSystemNodes(); } -void KeeperStorage::initializeSystemNodes() +template +void KeeperStorage::initializeSystemNodes() { if (initialized) throw Exception(ErrorCodes::LOGICAL_ERROR, "KeeperStorage system nodes initialized twice"); @@ -401,21 +477,25 @@ void KeeperStorage::initializeSystemNodes() container.insert(keeper_system_path, system_node); // store digest for the empty node because we won't update // its stats - addDigest(system_node, keeper_system_path); + if constexpr (!use_rocksdb) + addDigest(system_node, keeper_system_path); // update root and the digest based on it auto current_root_it = container.find("/"); chassert(current_root_it != container.end()); - removeDigest(current_root_it->value, "/"); + if constexpr (!use_rocksdb) + removeDigest(current_root_it->value, "/"); auto updated_root_it = container.updateValue( "/", [](KeeperStorage::Node & node) { node.increaseNumChildren(); - node.addChild(getBaseNodeName(keeper_system_path)); + if constexpr (!use_rocksdb) + node.addChild(getBaseNodeName(keeper_system_path)); } ); - addDigest(updated_root_it->value, "/"); + if constexpr (!use_rocksdb) + addDigest(updated_root_it->value, "/"); } // insert child system nodes @@ -424,17 +504,22 @@ void KeeperStorage::initializeSystemNodes() chassert(path.starts_with(keeper_system_path)); Node child_system_node; child_system_node.setData(data); - auto [map_key, _] = container.insert(std::string{path}, child_system_node); - /// Take child path from key owned by map. - auto child_path = getBaseNodeName(map_key->getKey()); - container.updateValue( - parentNodePath(StringRef(path)), - [child_path](auto & parent) - { - // don't update stats so digest is okay - parent.addChild(child_path); - } - ); + if constexpr (use_rocksdb) + container.insert(std::string{path}, child_system_node); + else + { + auto [map_key, _] = container.insert(std::string{path}, child_system_node); + /// Take child path from key owned by map. + auto child_path = getBaseNodeName(map_key->getKey()); + container.updateValue( + parentNodePath(StringRef(path)), + [child_path](auto & parent) + { + // don't update stats so digest is okay + parent.addChild(child_path); + } + ); + } } initialized = true; @@ -451,12 +536,13 @@ struct Overloaded : Ts... template Overloaded(Ts...) -> Overloaded; -std::shared_ptr KeeperStorage::UncommittedState::tryGetNodeFromStorage(StringRef path) const +template +std::shared_ptr KeeperStorage::UncommittedState::tryGetNodeFromStorage(StringRef path) const { if (auto node_it = storage.container.find(path); node_it != storage.container.end()) { const auto & committed_node = node_it->value; - auto node = std::make_shared(); + auto node = std::make_shared::Node>(); node->shallowCopy(committed_node); return node; } @@ -464,7 +550,8 @@ std::shared_ptr KeeperStorage::UncommittedState::tryGetNode return nullptr; } -void KeeperStorage::UncommittedState::applyDelta(const Delta & delta) +template +void KeeperStorage::UncommittedState::applyDelta(const Delta & delta) { chassert(!delta.path.empty()); if (!nodes.contains(delta.path)) @@ -511,7 +598,8 @@ void KeeperStorage::UncommittedState::applyDelta(const Delta & delta) delta.operation); } -bool KeeperStorage::UncommittedState::hasACL(int64_t session_id, bool is_local, std::function predicate) const +template +bool KeeperStorage::UncommittedState::hasACL(int64_t session_id, bool is_local, std::function predicate) const { const auto check_auth = [&](const auto & auth_ids) { @@ -534,6 +622,10 @@ bool KeeperStorage::UncommittedState::hasACL(int64_t session_id, bool is_local, if (is_local) return check_auth(storage.session_and_auth[session_id]); + /// we want to close the session and with that we will remove all the auth related to the session + if (closed_sessions.contains(session_id)) + return false; + if (check_auth(storage.session_and_auth[session_id])) return true; @@ -545,7 +637,8 @@ bool KeeperStorage::UncommittedState::hasACL(int64_t session_id, bool is_local, return check_auth(auth_it->second); } -void KeeperStorage::UncommittedState::addDelta(Delta new_delta) +template +void KeeperStorage::UncommittedState::addDelta(Delta new_delta) { const auto & added_delta = deltas.emplace_back(std::move(new_delta)); @@ -559,15 +652,21 @@ void KeeperStorage::UncommittedState::addDelta(Delta new_delta) auto & uncommitted_auth = session_and_auth[auth_delta->session_id]; uncommitted_auth.emplace_back(&auth_delta->auth_id); } + else if (const auto * close_session_delta = std::get_if(&added_delta.operation)) + { + closed_sessions.insert(close_session_delta->session_id); + } } -void KeeperStorage::UncommittedState::addDeltas(std::vector new_deltas) +template +void KeeperStorage::UncommittedState::addDeltas(std::vector new_deltas) { for (auto & delta : new_deltas) addDelta(std::move(delta)); } -void KeeperStorage::UncommittedState::commit(int64_t commit_zxid) +template +void KeeperStorage::UncommittedState::commit(int64_t commit_zxid) { chassert(deltas.empty() || deltas.front().zxid >= commit_zxid); @@ -627,7 +726,8 @@ void KeeperStorage::UncommittedState::commit(int64_t commit_zxid) } } -void KeeperStorage::UncommittedState::rollback(int64_t rollback_zxid) +template +void KeeperStorage::UncommittedState::rollback(int64_t rollback_zxid) { // we can only rollback the last zxid (if there is any) // if there is a delta with a larger zxid, we have invalid state @@ -723,7 +823,8 @@ void KeeperStorage::UncommittedState::rollback(int64_t rollback_zxid) } } -std::shared_ptr KeeperStorage::UncommittedState::getNode(StringRef path) const +template +std::shared_ptr KeeperStorage::UncommittedState::getNode(StringRef path) const { if (auto node_it = nodes.find(path.toView()); node_it != nodes.end()) return node_it->second.node; @@ -731,7 +832,8 @@ std::shared_ptr KeeperStorage::UncommittedState::getNode(St return tryGetNodeFromStorage(path); } -Coordination::ACLs KeeperStorage::UncommittedState::getACLs(StringRef path) const +template +Coordination::ACLs KeeperStorage::UncommittedState::getACLs(StringRef path) const { if (auto node_it = nodes.find(path.toView()); node_it != nodes.end()) return node_it->second.acls; @@ -743,7 +845,8 @@ Coordination::ACLs KeeperStorage::UncommittedState::getACLs(StringRef path) cons return storage.acl_map.convertNumber(node_it->value.acl_id); } -void KeeperStorage::UncommittedState::forEachAuthInSession(int64_t session_id, std::function func) const +template +void KeeperStorage::UncommittedState::forEachAuthInSession(int64_t session_id, std::function func) const { const auto call_for_each_auth = [&func](const auto & auth_ids) { @@ -782,7 +885,8 @@ namespace } -void KeeperStorage::applyUncommittedState(KeeperStorage & other, int64_t last_log_idx) +template +void KeeperStorage::applyUncommittedState(KeeperStorage & other, int64_t last_log_idx) { std::unordered_set zxids_to_apply; for (const auto & transaction : uncommitted_transactions) @@ -808,7 +912,8 @@ void KeeperStorage::applyUncommittedState(KeeperStorage & other, int64_t last_lo } } -Coordination::Error KeeperStorage::commit(int64_t commit_zxid) +template +Coordination::Error KeeperStorage::commit(int64_t commit_zxid) { // Deltas are added with increasing ZXIDs // If there are no deltas for the commit_zxid (e.g. read requests), we instantly return @@ -822,7 +927,7 @@ Coordination::Error KeeperStorage::commit(int64_t commit_zxid) auto result = std::visit( [&, &path = delta.path](DeltaType & operation) -> Coordination::Error { - if constexpr (std::same_as) + if constexpr (std::same_as) { if (!createNode( path, @@ -833,7 +938,7 @@ Coordination::Error KeeperStorage::commit(int64_t commit_zxid) return Coordination::Error::ZOK; } - else if constexpr (std::same_as) + else if constexpr (std::same_as) { auto node_it = container.find(path); if (node_it == container.end()) @@ -842,20 +947,22 @@ Coordination::Error KeeperStorage::commit(int64_t commit_zxid) if (operation.version != -1 && operation.version != node_it->value.version) onStorageInconsistency(); - removeDigest(node_it->value, path); + if constexpr (!use_rocksdb) + removeDigest(node_it->value, path); auto updated_node = container.updateValue(path, operation.update_fn); - addDigest(updated_node->value, path); + if constexpr (!use_rocksdb) + addDigest(updated_node->value, path); return Coordination::Error::ZOK; } - else if constexpr (std::same_as) + else if constexpr (std::same_as) { if (!removeNode(path, operation.version)) onStorageInconsistency(); return Coordination::Error::ZOK; } - else if constexpr (std::same_as) + else if constexpr (std::same_as) { auto node_it = container.find(path); if (node_it == container.end()) @@ -869,18 +976,18 @@ Coordination::Error KeeperStorage::commit(int64_t commit_zxid) uint64_t acl_id = acl_map.convertACLs(operation.acls); acl_map.addUsage(acl_id); - container.updateValue(path, [acl_id](KeeperStorage::Node & node) { node.acl_id = acl_id; }); + container.updateValue(path, [acl_id](Node & node) { node.acl_id = acl_id; }); return Coordination::Error::ZOK; } - else if constexpr (std::same_as) + else if constexpr (std::same_as) return operation.error; - else if constexpr (std::same_as) + else if constexpr (std::same_as) { finish_subdelta = true; return Coordination::Error::ZOK; } - else if constexpr (std::same_as) + else if constexpr (std::same_as) { session_and_auth[operation.session_id].emplace_back(std::move(operation.auth_id)); return Coordination::Error::ZOK; @@ -907,7 +1014,8 @@ Coordination::Error KeeperStorage::commit(int64_t commit_zxid) return Coordination::Error::ZOK; } -bool KeeperStorage::createNode( +template +bool KeeperStorage::createNode( const std::string & path, String data, const Coordination::Stat & stat, @@ -925,7 +1033,7 @@ bool KeeperStorage::createNode( if (container.contains(path)) return false; - KeeperStorage::Node created_node; + Node created_node; uint64_t acl_id = acl_map.convertACLs(node_acls); acl_map.addUsage(acl_id); @@ -933,23 +1041,31 @@ bool KeeperStorage::createNode( created_node.acl_id = acl_id; created_node.copyStats(stat); created_node.setData(data); - auto [map_key, _] = container.insert(path, created_node); - /// Take child path from key owned by map. - auto child_path = getBaseNodeName(map_key->getKey()); - container.updateValue( - parent_path, - [child_path](KeeperStorage::Node & parent) - { - parent.addChild(child_path); - chassert(parent.numChildren() == static_cast(parent.getChildren().size())); - } - ); + if constexpr (use_rocksdb) + { + container.insert(path, created_node); + } + else + { + auto [map_key, _] = container.insert(path, created_node); + /// Take child path from key owned by map. + auto child_path = getBaseNodeName(map_key->getKey()); + container.updateValue( + parent_path, + [child_path](KeeperMemNode & parent) + { + parent.addChild(child_path); + chassert(parent.numChildren() == static_cast(parent.getChildren().size())); + } + ); - addDigest(map_key->getMapped()->value, map_key->getKey().toView()); + addDigest(map_key->getMapped()->value, map_key->getKey().toView()); + } return true; }; -bool KeeperStorage::removeNode(const std::string & path, int32_t version) +template +bool KeeperStorage::removeNode(const std::string & path, int32_t version) { auto node_it = container.find(path); if (node_it == container.end()) @@ -965,69 +1081,84 @@ bool KeeperStorage::removeNode(const std::string & path, int32_t version) prev_node.shallowCopy(node_it->value); acl_map.removeUsage(node_it->value.acl_id); - container.updateValue( - parentNodePath(path), - [child_basename = getBaseNodeName(node_it->key)](KeeperStorage::Node & parent) - { - parent.removeChild(child_basename); - chassert(parent.numChildren() == static_cast(parent.getChildren().size())); - } - ); + if constexpr (use_rocksdb) + container.erase(path); + else + { + container.updateValue( + parentNodePath(path), + [child_basename = getBaseNodeName(node_it->key)](KeeperMemNode & parent) + { + parent.removeChild(child_basename); + chassert(parent.numChildren() == static_cast(parent.getChildren().size())); + } + ); - container.erase(path); + container.erase(path); - removeDigest(prev_node, path); + removeDigest(prev_node, path); + } return true; } +template struct KeeperStorageRequestProcessor { Coordination::ZooKeeperRequestPtr zk_request; explicit KeeperStorageRequestProcessor(const Coordination::ZooKeeperRequestPtr & zk_request_) : zk_request(zk_request_) { } - virtual Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const = 0; - virtual std::vector - preprocess(KeeperStorage & /*storage*/, int64_t /*zxid*/, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const + + virtual Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const = 0; + + virtual std::vector + preprocess(Storage & /*storage*/, int64_t /*zxid*/, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const { return {}; } // process the request using locally committed data virtual Coordination::ZooKeeperResponsePtr - processLocal(KeeperStorage & /*storage*/, int64_t /*zxid*/) const + processLocal(Storage & /*storage*/, int64_t /*zxid*/) const { throw Exception{DB::ErrorCodes::LOGICAL_ERROR, "Cannot process the request locally"}; } - virtual KeeperStorage::ResponsesForSessions - processWatches(KeeperStorage::Watches & /*watches*/, KeeperStorage::Watches & /*list_watches*/) const + virtual KeeperStorageBase::ResponsesForSessions + processWatches(KeeperStorageBase::Watches & /*watches*/, KeeperStorageBase::Watches & /*list_watches*/) const { return {}; } - virtual bool checkAuth(KeeperStorage & /*storage*/, int64_t /*session_id*/, bool /*is_local*/) const { return true; } + + virtual bool checkAuth(Storage & /*storage*/, int64_t /*session_id*/, bool /*is_local*/) const { return true; } virtual ~KeeperStorageRequestProcessor() = default; }; -struct KeeperStorageHeartbeatRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageHeartbeatRequestProcessor final : public KeeperStorageRequestProcessor { - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + Coordination::ZooKeeperResponsePtr - process(KeeperStorage & /* storage */, int64_t /* zxid */) const override + process(Storage & storage, int64_t zxid) const override { - return zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); + response_ptr->error = storage.commit(zxid); + return response_ptr; } }; -struct KeeperStorageSyncRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageSyncRequestProcessor final : public KeeperStorageRequestProcessor { - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + Coordination::ZooKeeperResponsePtr - process(KeeperStorage & /* storage */, int64_t /* zxid */) const override + process(Storage & /* storage */, int64_t /* zxid */) const override { - auto response = zk_request->makeResponse(); + auto response = this->zk_request->makeResponse(); dynamic_cast(*response).path - = dynamic_cast(*zk_request).path; + = dynamic_cast(*this->zk_request).path; return response; } }; @@ -1035,7 +1166,8 @@ struct KeeperStorageSyncRequestProcessor final : public KeeperStorageRequestProc namespace { -Coordination::ACLs getNodeACLs(KeeperStorage & storage, StringRef path, bool is_local) +template +Coordination::ACLs getNodeACLs(Storage & storage, StringRef path, bool is_local) { if (is_local) { @@ -1063,7 +1195,8 @@ void handleSystemNodeModification(const KeeperContext & keeper_context, std::str } -bool KeeperStorage::checkACL(StringRef path, int32_t permission, int64_t session_id, bool is_local) +template +bool KeeperStorage::checkACL(StringRef path, int32_t permission, int64_t session_id, bool is_local) { const auto node_acls = getNodeACLs(*this, path, is_local); if (node_acls.empty()) @@ -1090,7 +1223,8 @@ bool KeeperStorage::checkACL(StringRef path, int32_t permission, int64_t session return false; } -void KeeperStorage::unregisterEphemeralPath(int64_t session_id, const std::string & path) +template +void KeeperStorage::unregisterEphemeralPath(int64_t session_id, const std::string & path) { auto ephemerals_it = ephemerals.find(session_id); if (ephemerals_it == ephemerals.end()) @@ -1101,43 +1235,44 @@ void KeeperStorage::unregisterEphemeralPath(int64_t session_id, const std::strin ephemerals.erase(ephemerals_it); } -struct KeeperStorageCreateRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageCreateRequestProcessor final : public KeeperStorageRequestProcessor { - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - KeeperStorage::ResponsesForSessions - processWatches(KeeperStorage::Watches & watches, KeeperStorage::Watches & list_watches) const override + KeeperStorageBase::ResponsesForSessions + processWatches(KeeperStorageBase::Watches & watches, KeeperStorageBase::Watches & list_watches) const override { - return processWatchesImpl(zk_request->getPath(), watches, list_watches, Coordination::Event::CREATED); + return processWatchesImpl(this->zk_request->getPath(), watches, list_watches, Coordination::Event::CREATED); } - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { - auto path = zk_request->getPath(); + auto path = this->zk_request->getPath(); return storage.checkACL(parentNodePath(path), Coordination::ACL::Create, session_id, is_local); } - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t session_id, int64_t time, uint64_t & digest, const KeeperContext & keeper_context) const override + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t session_id, int64_t time, uint64_t & digest, const KeeperContext & keeper_context) const override { ProfileEvents::increment(ProfileEvents::KeeperCreateRequest); - Coordination::ZooKeeperCreateRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperCreateRequest & request = dynamic_cast(*this->zk_request); - std::vector new_deltas; + std::vector new_deltas; auto parent_path = parentNodePath(request.path); auto parent_node = storage.uncommitted_state.getNode(parent_path); if (parent_node == nullptr) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; else if (parent_node->isEphemeral()) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNOCHILDRENFOREPHEMERALS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNOCHILDRENFOREPHEMERALS}}; std::string path_created = request.path; if (request.is_sequential) { if (request.not_exists) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; auto seq_num = parent_node->seqNum(); @@ -1153,30 +1288,30 @@ struct KeeperStorageCreateRequestProcessor final : public KeeperStorageRequestPr auto error_msg = fmt::format("Trying to create a node inside the internal Keeper path ({}) which is not allowed. Path: {}", keeper_system_path, path_created); handleSystemNodeModification(keeper_context, error_msg); - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; } if (storage.uncommitted_state.getNode(path_created)) { - if (zk_request->getOpNum() == Coordination::OpNum::CreateIfNotExists) + if (this->zk_request->getOpNum() == Coordination::OpNum::CreateIfNotExists) return new_deltas; - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNODEEXISTS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNODEEXISTS}}; } if (getBaseNodeName(path_created).size == 0) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; Coordination::ACLs node_acls; if (!fixupACL(request.acls, session_id, storage.uncommitted_state, node_acls)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZINVALIDACL}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZINVALIDACL}}; if (request.is_ephemeral) storage.ephemerals[session_id].emplace(path_created); int32_t parent_cversion = request.parent_cversion; - auto parent_update = [parent_cversion, zxid](KeeperStorage::Node & node) + auto parent_update = [parent_cversion, zxid](Storage::Node & node) { /// Increment sequential number even if node is not sequential node.increaseSeqNum(); @@ -1189,7 +1324,7 @@ struct KeeperStorageCreateRequestProcessor final : public KeeperStorageRequestPr node.increaseNumChildren(); }; - new_deltas.emplace_back(std::string{parent_path}, zxid, KeeperStorage::UpdateNodeDelta{std::move(parent_update)}); + new_deltas.emplace_back(std::string{parent_path}, zxid, typename Storage::UpdateNodeDelta{std::move(parent_update)}); Coordination::Stat stat; stat.czxid = zxid; @@ -1206,20 +1341,20 @@ struct KeeperStorageCreateRequestProcessor final : public KeeperStorageRequestPr new_deltas.emplace_back( std::move(path_created), zxid, - KeeperStorage::CreateNodeDelta{stat, std::move(node_acls), request.data}); + typename Storage::CreateNodeDelta{stat, std::move(node_acls), request.data}); digest = storage.calculateNodesDigest(digest, new_deltas); return new_deltas; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperCreateResponse & response = dynamic_cast(*response_ptr); if (storage.uncommitted_state.deltas.begin()->zxid != zxid) { - response.path_created = zk_request->getPath(); + response.path_created = this->zk_request->getPath(); response.error = Coordination::Error::ZOK; return response_ptr; } @@ -1235,7 +1370,7 @@ struct KeeperStorageCreateRequestProcessor final : public KeeperStorageRequestPr deltas.begin(), deltas.end(), [zxid](const auto & delta) - { return delta.zxid == zxid && std::holds_alternative(delta.operation); }); + { return delta.zxid == zxid && std::holds_alternative(delta.operation); }); response.path_created = create_delta_it->path; response.error = Coordination::Error::ZOK; @@ -1243,20 +1378,21 @@ struct KeeperStorageCreateRequestProcessor final : public KeeperStorageRequestPr } }; -struct KeeperStorageGetRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageGetRequestProcessor final : public KeeperStorageRequestProcessor { - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { - return storage.checkACL(zk_request->getPath(), Coordination::ACL::Read, session_id, is_local); + return storage.checkACL(this->zk_request->getPath(), Coordination::ACL::Read, session_id, is_local); } - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override { ProfileEvents::increment(ProfileEvents::KeeperGetRequest); - Coordination::ZooKeeperGetRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperGetRequest & request = dynamic_cast(*this->zk_request); if (request.path == Coordination::keeper_api_feature_flags_path || request.path == Coordination::keeper_config_path @@ -1264,17 +1400,17 @@ struct KeeperStorageGetRequestProcessor final : public KeeperStorageRequestProce return {}; if (!storage.uncommitted_state.getNode(request.path)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; return {}; } template - Coordination::ZooKeeperResponsePtr processImpl(KeeperStorage & storage, int64_t zxid) const + Coordination::ZooKeeperResponsePtr processImpl(Storage & storage, int64_t zxid) const { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperGetResponse & response = dynamic_cast(*response_ptr); - Coordination::ZooKeeperGetRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperGetRequest & request = dynamic_cast(*this->zk_request); if constexpr (!local) { @@ -1314,40 +1450,42 @@ struct KeeperStorageGetRequestProcessor final : public KeeperStorageRequestProce } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { return processImpl(storage, zxid); } - Coordination::ZooKeeperResponsePtr processLocal(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr processLocal(Storage & storage, int64_t zxid) const override { ProfileEvents::increment(ProfileEvents::KeeperGetRequest); return processImpl(storage, zxid); } }; -struct KeeperStorageRemoveRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageRemoveRequestProcessor final : public KeeperStorageRequestProcessor { - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { - return storage.checkACL(parentNodePath(zk_request->getPath()), Coordination::ACL::Delete, session_id, is_local); + return storage.checkACL(parentNodePath(this->zk_request->getPath()), Coordination::ACL::Delete, session_id, is_local); } - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & digest, const KeeperContext & keeper_context) const override + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & digest, const KeeperContext & keeper_context) const override { ProfileEvents::increment(ProfileEvents::KeeperRemoveRequest); - Coordination::ZooKeeperRemoveRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperRemoveRequest & request = dynamic_cast(*this->zk_request); - std::vector new_deltas; + std::vector new_deltas; if (Coordination::matchPath(request.path, keeper_system_path) != Coordination::PathMatchResult::NOT_MATCH) { auto error_msg = fmt::format("Trying to delete an internal Keeper path ({}) which is not allowed", request.path); handleSystemNodeModification(keeper_context, error_msg); - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; } const auto update_parent_pzxid = [&]() @@ -1359,9 +1497,9 @@ struct KeeperStorageRemoveRequestProcessor final : public KeeperStorageRequestPr new_deltas.emplace_back( std::string{parent_path}, zxid, - KeeperStorage::UpdateNodeDelta + typename Storage::UpdateNodeDelta { - [zxid](KeeperStorage::Node & parent) + [zxid](Storage::Node & parent) { parent.pzxid = std::max(parent.pzxid, zxid); } @@ -1375,12 +1513,12 @@ struct KeeperStorageRemoveRequestProcessor final : public KeeperStorageRequestPr { if (request.restored_from_zookeeper_log) update_parent_pzxid(); - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; } else if (request.version != -1 && request.version != node->version) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADVERSION}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADVERSION}}; else if (node->numChildren() != 0) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNOTEMPTY}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNOTEMPTY}}; if (request.restored_from_zookeeper_log) update_parent_pzxid(); @@ -1388,13 +1526,13 @@ struct KeeperStorageRemoveRequestProcessor final : public KeeperStorageRequestPr new_deltas.emplace_back( std::string{parentNodePath(request.path)}, zxid, - KeeperStorage::UpdateNodeDelta{[](KeeperStorage::Node & parent) + typename Storage::UpdateNodeDelta{[](typename Storage::Node & parent) { ++parent.cversion; parent.decreaseNumChildren(); }}); - new_deltas.emplace_back(request.path, zxid, KeeperStorage::RemoveNodeDelta{request.version, node->ephemeralOwner()}); + new_deltas.emplace_back(request.path, zxid, typename Storage::RemoveNodeDelta{request.version, node->ephemeralOwner()}); if (node->isEphemeral()) storage.unregisterEphemeralPath(node->ephemeralOwner(), request.path); @@ -1404,44 +1542,45 @@ struct KeeperStorageRemoveRequestProcessor final : public KeeperStorageRequestPr return new_deltas; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperRemoveResponse & response = dynamic_cast(*response_ptr); response.error = storage.commit(zxid); return response_ptr; } - KeeperStorage::ResponsesForSessions - processWatches(KeeperStorage::Watches & watches, KeeperStorage::Watches & list_watches) const override + KeeperStorageBase::ResponsesForSessions + processWatches(KeeperStorageBase::Watches & watches, KeeperStorageBase::Watches & list_watches) const override { - return processWatchesImpl(zk_request->getPath(), watches, list_watches, Coordination::Event::DELETED); + return processWatchesImpl(this->zk_request->getPath(), watches, list_watches, Coordination::Event::DELETED); } }; -struct KeeperStorageExistsRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageExistsRequestProcessor final : public KeeperStorageRequestProcessor { - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override { ProfileEvents::increment(ProfileEvents::KeeperExistsRequest); - Coordination::ZooKeeperExistsRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperExistsRequest & request = dynamic_cast(*this->zk_request); if (!storage.uncommitted_state.getNode(request.path)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; return {}; } template - Coordination::ZooKeeperResponsePtr processImpl(KeeperStorage & storage, int64_t zxid) const + Coordination::ZooKeeperResponsePtr processImpl(Storage & storage, int64_t zxid) const { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperExistsResponse & response = dynamic_cast(*response_ptr); - Coordination::ZooKeeperExistsRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperExistsRequest & request = dynamic_cast(*this->zk_request); if constexpr (!local) { @@ -1470,55 +1609,57 @@ struct KeeperStorageExistsRequestProcessor final : public KeeperStorageRequestPr return response_ptr; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { return processImpl(storage, zxid); } - Coordination::ZooKeeperResponsePtr processLocal(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr processLocal(Storage & storage, int64_t zxid) const override { ProfileEvents::increment(ProfileEvents::KeeperExistsRequest); return processImpl(storage, zxid); } }; -struct KeeperStorageSetRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageSetRequestProcessor final : public KeeperStorageRequestProcessor { - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { - return storage.checkACL(zk_request->getPath(), Coordination::ACL::Write, session_id, is_local); + return storage.checkACL(this->zk_request->getPath(), Coordination::ACL::Write, session_id, is_local); } - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t /*session_id*/, int64_t time, uint64_t & digest, const KeeperContext & keeper_context) const override + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t /*session_id*/, int64_t time, uint64_t & digest, const KeeperContext & keeper_context) const override { ProfileEvents::increment(ProfileEvents::KeeperSetRequest); - Coordination::ZooKeeperSetRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperSetRequest & request = dynamic_cast(*this->zk_request); - std::vector new_deltas; + std::vector new_deltas; if (Coordination::matchPath(request.path, keeper_system_path) != Coordination::PathMatchResult::NOT_MATCH) { auto error_msg = fmt::format("Trying to update an internal Keeper path ({}) which is not allowed", request.path); handleSystemNodeModification(keeper_context, error_msg); - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; } if (!storage.uncommitted_state.getNode(request.path)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; auto node = storage.uncommitted_state.getNode(request.path); if (request.version != -1 && request.version != node->version) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADVERSION}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADVERSION}}; new_deltas.emplace_back( request.path, zxid, - KeeperStorage::UpdateNodeDelta{ - [zxid, data = request.data, time](KeeperStorage::Node & value) + typename Storage::UpdateNodeDelta{ + [zxid, data = request.data, time](typename Storage::Node & value) { value.version++; value.mzxid = zxid; @@ -1530,9 +1671,9 @@ struct KeeperStorageSetRequestProcessor final : public KeeperStorageRequestProce new_deltas.emplace_back( parentNodePath(request.path).toString(), zxid, - KeeperStorage::UpdateNodeDelta + typename Storage::UpdateNodeDelta { - [](KeeperStorage::Node & parent) + [](Storage::Node & parent) { parent.cversion++; } @@ -1543,13 +1684,13 @@ struct KeeperStorageSetRequestProcessor final : public KeeperStorageRequestProce return new_deltas; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { auto & container = storage.container; - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperSetResponse & response = dynamic_cast(*response_ptr); - Coordination::ZooKeeperSetRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperSetRequest & request = dynamic_cast(*this->zk_request); if (const auto result = storage.commit(zxid); result != Coordination::Error::ZOK) { @@ -1567,40 +1708,41 @@ struct KeeperStorageSetRequestProcessor final : public KeeperStorageRequestProce return response_ptr; } - KeeperStorage::ResponsesForSessions - processWatches(KeeperStorage::Watches & watches, KeeperStorage::Watches & list_watches) const override + KeeperStorageBase::ResponsesForSessions + processWatches(typename Storage::Watches & watches, typename Storage::Watches & list_watches) const override { - return processWatchesImpl(zk_request->getPath(), watches, list_watches, Coordination::Event::CHANGED); + return processWatchesImpl(this->zk_request->getPath(), watches, list_watches, Coordination::Event::CHANGED); } }; -struct KeeperStorageListRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageListRequestProcessor final : public KeeperStorageRequestProcessor { - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { - return storage.checkACL(zk_request->getPath(), Coordination::ACL::Read, session_id, is_local); + return storage.checkACL(this->zk_request->getPath(), Coordination::ACL::Read, session_id, is_local); } - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override { ProfileEvents::increment(ProfileEvents::KeeperListRequest); - Coordination::ZooKeeperListRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperListRequest & request = dynamic_cast(*this->zk_request); if (!storage.uncommitted_state.getNode(request.path)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; return {}; } - template - Coordination::ZooKeeperResponsePtr processImpl(KeeperStorage & storage, int64_t zxid) const + Coordination::ZooKeeperResponsePtr processImpl(Storage & storage, int64_t zxid) const { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperListResponse & response = dynamic_cast(*response_ptr); - Coordination::ZooKeeperListRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperListRequest & request = dynamic_cast(*this->zk_request); if constexpr (!local) { @@ -1627,33 +1769,55 @@ struct KeeperStorageListRequestProcessor final : public KeeperStorageRequestProc if (path_prefix.empty()) throw DB::Exception(ErrorCodes::LOGICAL_ERROR, "Path cannot be empty"); - const auto & children = node_it->value.getChildren(); + const auto & get_children = [&]() + { + if constexpr (Storage::use_rocksdb) + return container.getChildren(request.path); + else + return node_it->value.getChildren(); + }; + const auto & children = get_children(); response.names.reserve(children.size()); - const auto add_child = [&](const auto child) + const auto add_child = [&](const auto & child) { using enum Coordination::ListRequestType; auto list_request_type = ALL; if (auto * filtered_list = dynamic_cast(&request)) + { list_request_type = filtered_list->list_request_type; + } if (list_request_type == ALL) return true; - auto child_path = (std::filesystem::path(request.path) / child.toView()).generic_string(); - auto child_it = container.find(child_path); - if (child_it == container.end()) - onStorageInconsistency(); + bool is_ephemeral; + if constexpr (!Storage::use_rocksdb) + { + auto child_path = (std::filesystem::path(request.path) / child.toView()).generic_string(); + auto child_it = container.find(child_path); + if (child_it == container.end()) + onStorageInconsistency(); + is_ephemeral = child_it->value.isEphemeral(); + } + else + { + is_ephemeral = child.second.isEphemeral(); + } - const auto is_ephemeral = child_it->value.isEphemeral(); return (is_ephemeral && list_request_type == EPHEMERAL_ONLY) || (!is_ephemeral && list_request_type == PERSISTENT_ONLY); }; - for (const auto child : children) + for (const auto & child : children) { if (add_child(child)) - response.names.push_back(child.toString()); + { + if constexpr (Storage::use_rocksdb) + response.names.push_back(child.first); + else + response.names.push_back(child.toString()); + } } node_it->value.setResponseStat(response.stat); @@ -1663,63 +1827,64 @@ struct KeeperStorageListRequestProcessor final : public KeeperStorageRequestProc return response_ptr; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { return processImpl(storage, zxid); } - Coordination::ZooKeeperResponsePtr processLocal(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr processLocal(Storage & storage, int64_t zxid) const override { ProfileEvents::increment(ProfileEvents::KeeperListRequest); return processImpl(storage, zxid); } }; -struct KeeperStorageCheckRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageCheckRequestProcessor final : public KeeperStorageRequestProcessor { explicit KeeperStorageCheckRequestProcessor(const Coordination::ZooKeeperRequestPtr & zk_request_) - : KeeperStorageRequestProcessor(zk_request_) + : KeeperStorageRequestProcessor(zk_request_) { - check_not_exists = zk_request->getOpNum() == Coordination::OpNum::CheckNotExists; + check_not_exists = this->zk_request->getOpNum() == Coordination::OpNum::CheckNotExists; } - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { - auto path = zk_request->getPath(); + auto path = this->zk_request->getPath(); return storage.checkACL(check_not_exists ? parentNodePath(path) : path, Coordination::ACL::Read, session_id, is_local); } - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override { ProfileEvents::increment(ProfileEvents::KeeperCheckRequest); - Coordination::ZooKeeperCheckRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperCheckRequest & request = dynamic_cast(*this->zk_request); auto node = storage.uncommitted_state.getNode(request.path); if (check_not_exists) { if (node && (request.version == -1 || request.version == node->version)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNODEEXISTS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNODEEXISTS}}; } else { if (!node) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; if (request.version != -1 && request.version != node->version) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADVERSION}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADVERSION}}; } return {}; } template - Coordination::ZooKeeperResponsePtr processImpl(KeeperStorage & storage, int64_t zxid) const + Coordination::ZooKeeperResponsePtr processImpl(Storage & storage, int64_t zxid) const { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperCheckResponse & response = dynamic_cast(*response_ptr); - Coordination::ZooKeeperCheckRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperCheckRequest & request = dynamic_cast(*this->zk_request); if constexpr (!local) { @@ -1761,12 +1926,12 @@ struct KeeperStorageCheckRequestProcessor final : public KeeperStorageRequestPro return response_ptr; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { return processImpl(storage, zxid); } - Coordination::ZooKeeperResponsePtr processLocal(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr processLocal(Storage & storage, int64_t zxid) const override { ProfileEvents::increment(ProfileEvents::KeeperCheckRequest); return processImpl(storage, zxid); @@ -1777,55 +1942,56 @@ private: }; -struct KeeperStorageSetACLRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageSetACLRequestProcessor final : public KeeperStorageRequestProcessor { - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { - return storage.checkACL(zk_request->getPath(), Coordination::ACL::Admin, session_id, is_local); + return storage.checkACL(this->zk_request->getPath(), Coordination::ACL::Admin, session_id, is_local); } - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t session_id, int64_t /*time*/, uint64_t & digest, const KeeperContext & keeper_context) const override + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t session_id, int64_t /*time*/, uint64_t & digest, const KeeperContext & keeper_context) const override { - Coordination::ZooKeeperSetACLRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperSetACLRequest & request = dynamic_cast(*this->zk_request); if (Coordination::matchPath(request.path, keeper_system_path) != Coordination::PathMatchResult::NOT_MATCH) { auto error_msg = fmt::format("Trying to update an internal Keeper path ({}) which is not allowed", request.path); handleSystemNodeModification(keeper_context, error_msg); - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADARGUMENTS}}; } auto & uncommitted_state = storage.uncommitted_state; if (!uncommitted_state.getNode(request.path)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; auto node = uncommitted_state.getNode(request.path); if (request.version != -1 && request.version != node->aversion) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZBADVERSION}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZBADVERSION}}; Coordination::ACLs node_acls; if (!fixupACL(request.acls, session_id, uncommitted_state, node_acls)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZINVALIDACL}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZINVALIDACL}}; - std::vector new_deltas + std::vector new_deltas { { request.path, zxid, - KeeperStorage::SetACLDelta{std::move(node_acls), request.version} + typename Storage::SetACLDelta{std::move(node_acls), request.version} }, { request.path, zxid, - KeeperStorage::UpdateNodeDelta + typename Storage::UpdateNodeDelta { - [](KeeperStorage::Node & n) { ++n.aversion; } + [](typename Storage::Node & n) { ++n.aversion; } } } }; @@ -1835,11 +2001,11 @@ struct KeeperStorageSetACLRequestProcessor final : public KeeperStorageRequestPr return new_deltas; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperSetACLResponse & response = dynamic_cast(*response_ptr); - Coordination::ZooKeeperSetACLRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperSetACLRequest & request = dynamic_cast(*this->zk_request); if (const auto result = storage.commit(zxid); result != Coordination::Error::ZOK) { @@ -1857,32 +2023,33 @@ struct KeeperStorageSetACLRequestProcessor final : public KeeperStorageRequestPr } }; -struct KeeperStorageGetACLRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageGetACLRequestProcessor final : public KeeperStorageRequestProcessor { - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { - return storage.checkACL(zk_request->getPath(), Coordination::ACL::Admin | Coordination::ACL::Read, session_id, is_local); + return storage.checkACL(this->zk_request->getPath(), Coordination::ACL::Admin | Coordination::ACL::Read, session_id, is_local); } - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t /*session_id*/, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override { - Coordination::ZooKeeperGetACLRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperGetACLRequest & request = dynamic_cast(*this->zk_request); if (!storage.uncommitted_state.getNode(request.path)) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZNONODE}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZNONODE}}; return {}; } template - Coordination::ZooKeeperResponsePtr processImpl(KeeperStorage & storage, int64_t zxid) const + Coordination::ZooKeeperResponsePtr processImpl(Storage & storage, int64_t zxid) const { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperGetACLResponse & response = dynamic_cast(*response_ptr); - Coordination::ZooKeeperGetACLRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperGetACLRequest & request = dynamic_cast(*this->zk_request); if constexpr (!local) { @@ -1911,23 +2078,24 @@ struct KeeperStorageGetACLRequestProcessor final : public KeeperStorageRequestPr return response_ptr; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { return processImpl(storage, zxid); } - Coordination::ZooKeeperResponsePtr processLocal(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr processLocal(Storage & storage, int64_t zxid) const override { return processImpl(storage, zxid); } }; -struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestProcessor { using OperationType = Coordination::ZooKeeperMultiRequest::OperationType; std::optional operation_type; - bool checkAuth(KeeperStorage & storage, int64_t session_id, bool is_local) const override + bool checkAuth(Storage & storage, int64_t session_id, bool is_local) const override { for (const auto & concrete_request : concrete_requests) if (!concrete_request->checkAuth(storage, session_id, is_local)) @@ -1935,11 +2103,11 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro return true; } - std::vector concrete_requests; + std::vector>> concrete_requests; explicit KeeperStorageMultiRequestProcessor(const Coordination::ZooKeeperRequestPtr & zk_request_) - : KeeperStorageRequestProcessor(zk_request_) + : KeeperStorageRequestProcessor(zk_request_) { - Coordination::ZooKeeperMultiRequest & request = dynamic_cast(*zk_request); + Coordination::ZooKeeperMultiRequest & request = dynamic_cast(*this->zk_request); concrete_requests.reserve(request.requests.size()); const auto check_operation_type = [&](OperationType type) @@ -1957,34 +2125,34 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro case Coordination::OpNum::Create: case Coordination::OpNum::CreateIfNotExists: check_operation_type(OperationType::Write); - concrete_requests.push_back(std::make_shared(sub_zk_request)); + concrete_requests.push_back(std::make_shared>(sub_zk_request)); break; case Coordination::OpNum::Remove: check_operation_type(OperationType::Write); - concrete_requests.push_back(std::make_shared(sub_zk_request)); + concrete_requests.push_back(std::make_shared>(sub_zk_request)); break; case Coordination::OpNum::Set: check_operation_type(OperationType::Write); - concrete_requests.push_back(std::make_shared(sub_zk_request)); + concrete_requests.push_back(std::make_shared>(sub_zk_request)); break; case Coordination::OpNum::Check: case Coordination::OpNum::CheckNotExists: check_operation_type(OperationType::Write); - concrete_requests.push_back(std::make_shared(sub_zk_request)); + concrete_requests.push_back(std::make_shared>(sub_zk_request)); break; case Coordination::OpNum::Get: check_operation_type(OperationType::Read); - concrete_requests.push_back(std::make_shared(sub_zk_request)); + concrete_requests.push_back(std::make_shared>(sub_zk_request)); break; case Coordination::OpNum::Exists: check_operation_type(OperationType::Read); - concrete_requests.push_back(std::make_shared(sub_zk_request)); + concrete_requests.push_back(std::make_shared>(sub_zk_request)); break; case Coordination::OpNum::List: case Coordination::OpNum::FilteredList: case Coordination::OpNum::SimpleList: check_operation_type(OperationType::Read); - concrete_requests.push_back(std::make_shared(sub_zk_request)); + concrete_requests.push_back(std::make_shared>(sub_zk_request)); break; default: throw DB::Exception( @@ -1997,8 +2165,8 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro chassert(request.requests.empty() || operation_type.has_value()); } - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t session_id, int64_t time, uint64_t & digest, const KeeperContext & keeper_context) const override + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t session_id, int64_t time, uint64_t & digest, const KeeperContext & keeper_context) const override { ProfileEvents::increment(ProfileEvents::KeeperMultiRequest); std::vector response_errors; @@ -2010,7 +2178,7 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro if (!new_deltas.empty()) { - if (auto * error = std::get_if(&new_deltas.back().operation); + if (auto * error = std::get_if(&new_deltas.back().operation); error && *operation_type == OperationType::Write) { storage.uncommitted_state.rollback(zxid); @@ -2021,10 +2189,10 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro response_errors.push_back(Coordination::Error::ZRUNTIMEINCONSISTENCY); } - return {KeeperStorage::Delta{zxid, KeeperStorage::FailedMultiDelta{std::move(response_errors)}}}; + return {typename Storage::Delta{zxid, typename Storage::FailedMultiDelta{std::move(response_errors)}}}; } } - new_deltas.emplace_back(zxid, KeeperStorage::SubDeltaEnd{}); + new_deltas.emplace_back(zxid, typename Storage::SubDeltaEnd{}); response_errors.push_back(Coordination::Error::ZOK); // manually add deltas so that the result of previous request in the transaction is used in the next request @@ -2036,15 +2204,15 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro return {}; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperMultiResponse & response = dynamic_cast(*response_ptr); auto & deltas = storage.uncommitted_state.deltas; // the deltas will have at least SubDeltaEnd or FailedMultiDelta chassert(!deltas.empty()); - if (auto * failed_multi = std::get_if(&deltas.front().operation)) + if (auto * failed_multi = std::get_if(&deltas.front().operation)) { for (size_t i = 0; i < concrete_requests.size(); ++i) { @@ -2066,10 +2234,10 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro return response_ptr; } - Coordination::ZooKeeperResponsePtr processLocal(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr processLocal(Storage & storage, int64_t zxid) const override { ProfileEvents::increment(ProfileEvents::KeeperMultiReadRequest); - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperMultiResponse & response = dynamic_cast(*response_ptr); for (size_t i = 0; i < concrete_requests.size(); ++i) @@ -2081,10 +2249,10 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro return response_ptr; } - KeeperStorage::ResponsesForSessions - processWatches(KeeperStorage::Watches & watches, KeeperStorage::Watches & list_watches) const override + KeeperStorageBase::ResponsesForSessions + processWatches(typename Storage::Watches & watches, typename Storage::Watches & list_watches) const override { - KeeperStorage::ResponsesForSessions result; + typename Storage::ResponsesForSessions result; for (const auto & generic_request : concrete_requests) { auto responses = generic_request->processWatches(watches, list_watches); @@ -2094,47 +2262,50 @@ struct KeeperStorageMultiRequestProcessor final : public KeeperStorageRequestPro } }; -struct KeeperStorageCloseRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageCloseRequestProcessor final : public KeeperStorageRequestProcessor { - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - Coordination::ZooKeeperResponsePtr process(KeeperStorage &, int64_t) const override + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + Coordination::ZooKeeperResponsePtr process(Storage &, int64_t) const override { throw DB::Exception(ErrorCodes::LOGICAL_ERROR, "Called process on close request"); } }; -struct KeeperStorageAuthRequestProcessor final : public KeeperStorageRequestProcessor +template +struct KeeperStorageAuthRequestProcessor final : public KeeperStorageRequestProcessor { - using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; - std::vector - preprocess(KeeperStorage & storage, int64_t zxid, int64_t session_id, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override + using KeeperStorageRequestProcessor::KeeperStorageRequestProcessor; + + std::vector + preprocess(Storage & storage, int64_t zxid, int64_t session_id, int64_t /*time*/, uint64_t & /*digest*/, const KeeperContext & /*keeper_context*/) const override { - Coordination::ZooKeeperAuthRequest & auth_request = dynamic_cast(*zk_request); - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperAuthRequest & auth_request = dynamic_cast(*this->zk_request); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); if (auth_request.scheme != "digest" || std::count(auth_request.data.begin(), auth_request.data.end(), ':') != 1) - return {KeeperStorage::Delta{zxid, Coordination::Error::ZAUTHFAILED}}; + return {typename Storage::Delta{zxid, Coordination::Error::ZAUTHFAILED}}; - std::vector new_deltas; - auto auth_digest = KeeperStorage::generateDigest(auth_request.data); + std::vector new_deltas; + auto auth_digest = Storage::generateDigest(auth_request.data); if (auth_digest == storage.superdigest) { - KeeperStorage::AuthID auth{"super", ""}; - new_deltas.emplace_back(zxid, KeeperStorage::AddAuthDelta{session_id, std::move(auth)}); + typename Storage::AuthID auth{"super", ""}; + new_deltas.emplace_back(zxid, typename Storage::AddAuthDelta{session_id, std::move(auth)}); } else { - KeeperStorage::AuthID new_auth{auth_request.scheme, auth_digest}; + typename Storage::AuthID new_auth{auth_request.scheme, auth_digest}; if (!storage.uncommitted_state.hasACL(session_id, false, [&](const auto & auth_id) { return new_auth == auth_id; })) - new_deltas.emplace_back(zxid, KeeperStorage::AddAuthDelta{session_id, std::move(new_auth)}); + new_deltas.emplace_back(zxid, typename Storage::AddAuthDelta{session_id, std::move(new_auth)}); } return new_deltas; } - Coordination::ZooKeeperResponsePtr process(KeeperStorage & storage, int64_t zxid) const override + Coordination::ZooKeeperResponsePtr process(Storage & storage, int64_t zxid) const override { - Coordination::ZooKeeperResponsePtr response_ptr = zk_request->makeResponse(); + Coordination::ZooKeeperResponsePtr response_ptr = this->zk_request->makeResponse(); Coordination::ZooKeeperAuthResponse & auth_response = dynamic_cast(*response_ptr); if (const auto result = storage.commit(zxid); result != Coordination::Error::ZOK) @@ -2144,7 +2315,8 @@ struct KeeperStorageAuthRequestProcessor final : public KeeperStorageRequestProc } }; -void KeeperStorage::finalize() +template +void KeeperStorage::finalize() { if (finalized) throw DB::Exception(ErrorCodes::LOGICAL_ERROR, "KeeperStorage already finalized"); @@ -2159,25 +2331,26 @@ void KeeperStorage::finalize() session_expiry_queue.clear(); } -bool KeeperStorage::isFinalized() const +template +bool KeeperStorage::isFinalized() const { return finalized; } - +template class KeeperStorageRequestProcessorsFactory final : private boost::noncopyable { public: - using Creator = std::function; + using Creator = std::function>(const Coordination::ZooKeeperRequestPtr &)>; using OpNumToRequest = std::unordered_map; - static KeeperStorageRequestProcessorsFactory & instance() + static KeeperStorageRequestProcessorsFactory & instance() { - static KeeperStorageRequestProcessorsFactory factory; + static KeeperStorageRequestProcessorsFactory factory; return factory; } - KeeperStorageRequestProcessorPtr get(const Coordination::ZooKeeperRequestPtr & zk_request) const + std::shared_ptr> get(const Coordination::ZooKeeperRequestPtr & zk_request) const { auto request_it = op_num_to_request.find(zk_request->getOpNum()); if (request_it == op_num_to_request.end()) @@ -2197,39 +2370,41 @@ private: KeeperStorageRequestProcessorsFactory(); }; -template -void registerKeeperRequestProcessor(KeeperStorageRequestProcessorsFactory & factory) +template +void registerKeeperRequestProcessor(Factory & factory) { factory.registerRequest( num, [](const Coordination::ZooKeeperRequestPtr & zk_request) { return std::make_shared(zk_request); }); } -KeeperStorageRequestProcessorsFactory::KeeperStorageRequestProcessorsFactory() +template +KeeperStorageRequestProcessorsFactory::KeeperStorageRequestProcessorsFactory() { - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); - registerKeeperRequestProcessor(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); + registerKeeperRequestProcessor>(*this); } -UInt64 KeeperStorage::calculateNodesDigest(UInt64 current_digest, const std::vector & new_deltas) const +template +UInt64 KeeperStorage::calculateNodesDigest(UInt64 current_digest, const std::vector & new_deltas) const { if (!keeper_context->digestEnabled()) return current_digest; @@ -2264,7 +2439,7 @@ UInt64 KeeperStorage::calculateNodesDigest(UInt64 current_digest, const std::vec auto updated_node_it = updated_nodes.find(delta.path); if (updated_node_it == updated_nodes.end()) { - node = std::make_shared(); + node = std::make_shared(); node->shallowCopy(*uncommitted_state.getNode(delta.path)); current_digest -= node->getDigest(delta.path); updated_nodes.emplace(delta.path, node); @@ -2290,7 +2465,8 @@ UInt64 KeeperStorage::calculateNodesDigest(UInt64 current_digest, const std::vec return current_digest; } -void KeeperStorage::preprocessRequest( +template +void KeeperStorage::preprocessRequest( const Coordination::ZooKeeperRequestPtr & zk_request, int64_t session_id, int64_t time, @@ -2299,6 +2475,20 @@ void KeeperStorage::preprocessRequest( std::optional digest, int64_t log_idx) { + Stopwatch watch; + SCOPE_EXIT({ + auto elapsed = watch.elapsedMicroseconds(); + if (auto elapsed_ms = elapsed / 1000; elapsed_ms > keeper_context->getCoordinationSettings()->log_slow_cpu_threshold_ms) + { + LOG_INFO( + getLogger("KeeperStorage"), + "Preprocessing a request took too long ({}ms).\nRequest info: {}", + elapsed_ms, + zk_request->toString(/*short_format=*/true)); + } + ProfileEvents::increment(ProfileEvents::KeeperPreprocessElapsedMicroseconds, elapsed); + }); + if (!initialized) throw Exception(ErrorCodes::LOGICAL_ERROR, "KeeperStorage system nodes are not initialized"); @@ -2349,7 +2539,7 @@ void KeeperStorage::preprocessRequest( uncommitted_state.addDeltas(std::move(new_deltas)); }); - KeeperStorageRequestProcessorPtr request_processor = KeeperStorageRequestProcessorsFactory::instance().get(zk_request); + auto request_processor = KeeperStorageRequestProcessorsFactory>::instance().get(zk_request); if (zk_request->getOpNum() == Coordination::OpNum::Close) /// Close request is special { @@ -2377,15 +2567,13 @@ void KeeperStorage::preprocessRequest( ephemerals.erase(session_ephemerals); } - new_deltas.emplace_back(transaction.zxid, CloseSessionDelta{session_id}); - uncommitted_state.closed_sessions.insert(session_id); + new_deltas.emplace_back(transaction.zxid, CloseSessionDelta{session_id}); new_digest = calculateNodesDigest(new_digest, new_deltas); return; } - if ((check_acl && !request_processor->checkAuth(*this, session_id, false)) || - uncommitted_state.closed_sessions.contains(session_id)) // Is session closed but not committed yet + if (check_acl && !request_processor->checkAuth(*this, session_id, false)) { uncommitted_state.deltas.emplace_back(new_last_zxid, Coordination::Error::ZNOAUTH); return; @@ -2394,13 +2582,28 @@ void KeeperStorage::preprocessRequest( new_deltas = request_processor->preprocess(*this, transaction.zxid, session_id, time, new_digest, *keeper_context); } -KeeperStorage::ResponsesForSessions KeeperStorage::processRequest( +template +KeeperStorage::ResponsesForSessions KeeperStorage::processRequest( const Coordination::ZooKeeperRequestPtr & zk_request, int64_t session_id, std::optional new_last_zxid, bool check_acl, bool is_local) { + Stopwatch watch; + SCOPE_EXIT({ + auto elapsed = watch.elapsedMicroseconds(); + if (auto elapsed_ms = elapsed / 1000; elapsed_ms > keeper_context->getCoordinationSettings()->log_slow_cpu_threshold_ms) + { + LOG_INFO( + getLogger("KeeperStorage"), + "Processing a request took too long ({}ms).\nRequest info: {}", + elapsed_ms, + zk_request->toString(/*short_format=*/true)); + } + ProfileEvents::increment(ProfileEvents::KeeperProcessElapsedMicroseconds, elapsed); + }); + if (!initialized) throw Exception(ErrorCodes::LOGICAL_ERROR, "KeeperStorage system nodes are not initialized"); @@ -2421,7 +2624,7 @@ KeeperStorage::ResponsesForSessions KeeperStorage::processRequest( uncommitted_transactions.pop_front(); } - KeeperStorage::ResponsesForSessions results; + ResponsesForSessions results; /// ZooKeeper update sessions expirity for each request, not only for heartbeats session_expiry_queue.addNewSessionOrUpdate(session_id, session_and_timeout[session_id]); @@ -2442,8 +2645,6 @@ KeeperStorage::ResponsesForSessions KeeperStorage::processRequest( } } - uncommitted_state.commit(zxid); - clearDeadWatches(session_id); auto auth_it = session_and_auth.find(session_id); if (auth_it != session_and_auth.end()) @@ -2459,7 +2660,7 @@ KeeperStorage::ResponsesForSessions KeeperStorage::processRequest( } else if (zk_request->getOpNum() == Coordination::OpNum::Heartbeat) /// Heartbeat request is also special { - KeeperStorageRequestProcessorPtr storage_request = KeeperStorageRequestProcessorsFactory::instance().get(zk_request); + auto storage_request = KeeperStorageRequestProcessorsFactory>::instance().get(zk_request); auto response = storage_request->process(*this, zxid); response->xid = zk_request->xid; response->zxid = getZXID(); @@ -2468,7 +2669,7 @@ KeeperStorage::ResponsesForSessions KeeperStorage::processRequest( } else /// normal requests proccession { - KeeperStorageRequestProcessorPtr request_processor = KeeperStorageRequestProcessorsFactory::instance().get(zk_request); + auto request_processor = KeeperStorageRequestProcessorsFactory>::instance().get(zk_request); Coordination::ZooKeeperResponsePtr response; if (is_local) @@ -2488,7 +2689,6 @@ KeeperStorage::ResponsesForSessions KeeperStorage::processRequest( else { response = request_processor->process(*this, zxid); - uncommitted_state.commit(zxid); } /// Watches for this requests are added to the watches lists @@ -2528,10 +2728,12 @@ KeeperStorage::ResponsesForSessions KeeperStorage::processRequest( results.push_back(ResponseForSession{session_id, response}); } + uncommitted_state.commit(zxid); return results; } -void KeeperStorage::rollbackRequest(int64_t rollback_zxid, bool allow_missing) +template +void KeeperStorage::rollbackRequest(int64_t rollback_zxid, bool allow_missing) { if (allow_missing && (uncommitted_transactions.empty() || uncommitted_transactions.back().zxid < rollback_zxid)) return; @@ -2557,7 +2759,8 @@ void KeeperStorage::rollbackRequest(int64_t rollback_zxid, bool allow_missing) } } -KeeperStorage::Digest KeeperStorage::getNodesDigest(bool committed) const +template +KeeperStorageBase::Digest KeeperStorage::getNodesDigest(bool committed) const { if (!keeper_context->digestEnabled()) return {.version = DigestVersion::NO_DIGEST}; @@ -2568,13 +2771,15 @@ KeeperStorage::Digest KeeperStorage::getNodesDigest(bool committed) const return uncommitted_transactions.back().nodes_digest; } -void KeeperStorage::removeDigest(const Node & node, const std::string_view path) +template +void KeeperStorage::removeDigest(const Node & node, const std::string_view path) { if (keeper_context->digestEnabled()) nodes_digest -= node.getDigest(path); } -void KeeperStorage::addDigest(const Node & node, const std::string_view path) +template +void KeeperStorage::addDigest(const Node & node, const std::string_view path) { if (keeper_context->digestEnabled()) { @@ -2583,7 +2788,8 @@ void KeeperStorage::addDigest(const Node & node, const std::string_view path) } } -void KeeperStorage::clearDeadWatches(int64_t session_id) +template +void KeeperStorage::clearDeadWatches(int64_t session_id) { /// Clear all watches for this session auto watches_it = sessions_and_watchers.find(session_id); @@ -2616,7 +2822,8 @@ void KeeperStorage::clearDeadWatches(int64_t session_id) } } -void KeeperStorage::dumpWatches(WriteBufferFromOwnString & buf) const +template +void KeeperStorage::dumpWatches(WriteBufferFromOwnString & buf) const { for (const auto & [session_id, watches_paths] : sessions_and_watchers) { @@ -2626,7 +2833,8 @@ void KeeperStorage::dumpWatches(WriteBufferFromOwnString & buf) const } } -void KeeperStorage::dumpWatchesByPath(WriteBufferFromOwnString & buf) const +template +void KeeperStorage::dumpWatchesByPath(WriteBufferFromOwnString & buf) const { auto write_int_container = [&buf](const auto & session_ids) { @@ -2649,7 +2857,8 @@ void KeeperStorage::dumpWatchesByPath(WriteBufferFromOwnString & buf) const } } -void KeeperStorage::dumpSessionsAndEphemerals(WriteBufferFromOwnString & buf) const +template +void KeeperStorage::dumpSessionsAndEphemerals(WriteBufferFromOwnString & buf) const { auto write_str_set = [&buf](const std::unordered_set & ephemeral_paths) { @@ -2674,7 +2883,8 @@ void KeeperStorage::dumpSessionsAndEphemerals(WriteBufferFromOwnString & buf) co } } -uint64_t KeeperStorage::getTotalWatchesCount() const +template +uint64_t KeeperStorage::getTotalWatchesCount() const { uint64_t ret = 0; for (const auto & [session, paths] : sessions_and_watchers) @@ -2683,12 +2893,14 @@ uint64_t KeeperStorage::getTotalWatchesCount() const return ret; } -uint64_t KeeperStorage::getSessionsWithWatchesCount() const +template +uint64_t KeeperStorage::getSessionsWithWatchesCount() const { return sessions_and_watchers.size(); } -uint64_t KeeperStorage::getTotalEphemeralNodesCount() const +template +uint64_t KeeperStorage::getTotalEphemeralNodesCount() const { uint64_t ret = 0; for (const auto & [session_id, nodes] : ephemerals) @@ -2697,12 +2909,13 @@ uint64_t KeeperStorage::getTotalEphemeralNodesCount() const return ret; } -void KeeperStorage::recalculateStats() +template +void KeeperStorage::recalculateStats() { container.recalculateDataSize(); } -bool KeeperStorage::checkDigest(const Digest & first, const Digest & second) +bool KeeperStorageBase::checkDigest(const Digest & first, const Digest & second) { if (first.version != second.version) return true; @@ -2713,13 +2926,18 @@ bool KeeperStorage::checkDigest(const Digest & first, const Digest & second) return first.value == second.value; } -String KeeperStorage::generateDigest(const String & userdata) +template +String KeeperStorage::generateDigest(const String & userdata) { std::vector user_password; boost::split(user_password, userdata, [](char character) { return character == ':'; }); return user_password[0] + ":" + base64Encode(getSHA1(userdata)); } +template class KeeperStorage>; +#if USE_ROCKSDB +template class KeeperStorage>; +#endif } diff --git a/src/Coordination/KeeperStorage.h b/src/Coordination/KeeperStorage.h index d5e9a64e69c..4a9286d4835 100644 --- a/src/Coordination/KeeperStorage.h +++ b/src/Coordination/KeeperStorage.h @@ -8,188 +8,384 @@ #include +#include "config.h" +#if USE_ROCKSDB +#include +#endif + namespace DB { class KeeperContext; using KeeperContextPtr = std::shared_ptr; -struct KeeperStorageRequestProcessor; -using KeeperStorageRequestProcessorPtr = std::shared_ptr; using ResponseCallback = std::function; using ChildrenSet = absl::flat_hash_set; using SessionAndTimeout = std::unordered_map; -struct KeeperStorageSnapshot; - -/// Keeper state machine almost equal to the ZooKeeper's state machine. -/// Implements all logic of operations, data changes, sessions allocation. -/// In-memory and not thread safe. -class KeeperStorage +/// KeeperRocksNodeInfo is used in RocksDB keeper. +/// It is serialized directly as POD to RocksDB. +struct KeeperRocksNodeInfo { -public: - /// Node should have as minimal size as possible to reduce memory footprint - /// of stored nodes - /// New fields should be added to the struct only if it's really necessary - struct Node + int64_t czxid{0}; + int64_t mzxid{0}; + int64_t pzxid{0}; + uint64_t acl_id = 0; /// 0 -- no ACL by default + + int64_t mtime{0}; + + int32_t version{0}; + int32_t cversion{0}; + int32_t aversion{0}; + + int32_t seq_num = 0; + mutable UInt64 digest = 0; /// we cached digest for this node. + + /// as ctime can't be negative because it stores the timestamp when the + /// node was created, we can use the MSB for a bool + struct { - int64_t czxid{0}; - int64_t mzxid{0}; - int64_t pzxid{0}; - uint64_t acl_id = 0; /// 0 -- no ACL by default + bool is_ephemeral : 1; + int64_t ctime : 63; + } is_ephemeral_and_ctime{false, 0}; - int64_t mtime{0}; - - std::unique_ptr data{nullptr}; - uint32_t data_size{0}; - - int32_t version{0}; - int32_t cversion{0}; - int32_t aversion{0}; - - mutable uint64_t cached_digest = 0; - - Node() = default; - - Node & operator=(const Node & other); - Node(const Node & other); - - Node & operator=(Node && other) noexcept; - Node(Node && other) noexcept; - - bool empty() const; - - bool isEphemeral() const - { - return is_ephemeral_and_ctime.is_ephemeral; - } - - int64_t ephemeralOwner() const - { - if (isEphemeral()) - return ephemeral_or_children_data.ephemeral_owner; - - return 0; - } - - void setEphemeralOwner(int64_t ephemeral_owner) - { - is_ephemeral_and_ctime.is_ephemeral = ephemeral_owner != 0; - ephemeral_or_children_data.ephemeral_owner = ephemeral_owner; - } - - int32_t numChildren() const - { - if (isEphemeral()) - return 0; - - return ephemeral_or_children_data.children_info.num_children; - } - - void setNumChildren(int32_t num_children) - { - ephemeral_or_children_data.children_info.num_children = num_children; - } - - void increaseNumChildren() - { - chassert(!isEphemeral()); - ++ephemeral_or_children_data.children_info.num_children; - } - - void decreaseNumChildren() - { - chassert(!isEphemeral()); - --ephemeral_or_children_data.children_info.num_children; - } - - int32_t seqNum() const - { - if (isEphemeral()) - return 0; - - return ephemeral_or_children_data.children_info.seq_num; - } - - void setSeqNum(int32_t seq_num) - { - ephemeral_or_children_data.children_info.seq_num = seq_num; - } - - void increaseSeqNum() - { - chassert(!isEphemeral()); - ++ephemeral_or_children_data.children_info.seq_num; - } - - int64_t ctime() const - { - return is_ephemeral_and_ctime.ctime; - } - - void setCtime(uint64_t ctime) - { - is_ephemeral_and_ctime.ctime = ctime; - } - - void copyStats(const Coordination::Stat & stat); - - void setResponseStat(Coordination::Stat & response_stat) const; - - /// Object memory size - uint64_t sizeInBytes() const; - - void setData(const String & new_data); - - std::string_view getData() const noexcept { return {data.get(), data_size}; } - - void addChild(StringRef child_path); - - void removeChild(StringRef child_path); - - const auto & getChildren() const noexcept { return children; } - auto & getChildren() { return children; } - - // Invalidate the calculated digest so it's recalculated again on the next - // getDigest call - void invalidateDigestCache() const; - - // get the calculated digest of the node - UInt64 getDigest(std::string_view path) const; - - // copy only necessary information for preprocessing and digest calculation - // (e.g. we don't need to copy list of children) - void shallowCopy(const Node & other); - private: - /// as ctime can't be negative because it stores the timestamp when the - /// node was created, we can use the MSB for a bool + /// ephemeral notes cannot have children so a node can set either + /// ephemeral_owner OR seq_num + num_children + union + { + int64_t ephemeral_owner; struct { - bool is_ephemeral : 1; - int64_t ctime : 63; - } is_ephemeral_and_ctime{false, 0}; + int32_t seq_num; + int32_t num_children; + } children_info; + } ephemeral_or_children_data{0}; - /// ephemeral notes cannot have children so a node can set either - /// ephemeral_owner OR seq_num + num_children - union - { - int64_t ephemeral_owner; - struct - { - int32_t seq_num; - int32_t num_children; - } children_info; - } ephemeral_or_children_data{0}; + bool isEphemeral() const + { + return is_ephemeral_and_ctime.is_ephemeral; + } - ChildrenSet children{}; - }; + int64_t ephemeralOwner() const + { + if (isEphemeral()) + return ephemeral_or_children_data.ephemeral_owner; -#if !defined(ADDRESS_SANITIZER) && !defined(MEMORY_SANITIZER) - static_assert( - sizeof(ListNode) <= 144, - "std::list node containing ListNode is > 160 bytes (sizeof(ListNode) + 16 bytes for pointers) which will increase " - "memory consumption"); + return 0; + } + + void setEphemeralOwner(int64_t ephemeral_owner) + { + is_ephemeral_and_ctime.is_ephemeral = ephemeral_owner != 0; + ephemeral_or_children_data.ephemeral_owner = ephemeral_owner; + } + + int32_t numChildren() const + { + if (isEphemeral()) + return 0; + + return ephemeral_or_children_data.children_info.num_children; + } + + void setNumChildren(int32_t num_children) + { + ephemeral_or_children_data.children_info.num_children = num_children; + } + + /// dummy interface for test + void addChild(StringRef) {} + auto getChildren() const + { + return std::vector(numChildren()); + } + + void increaseNumChildren() + { + chassert(!isEphemeral()); + ++ephemeral_or_children_data.children_info.num_children; + } + + void decreaseNumChildren() + { + chassert(!isEphemeral()); + --ephemeral_or_children_data.children_info.num_children; + } + + int32_t seqNum() const + { + if (isEphemeral()) + return 0; + + return ephemeral_or_children_data.children_info.seq_num; + } + + void setSeqNum(int32_t seq_num_) + { + ephemeral_or_children_data.children_info.seq_num = seq_num_; + } + + void increaseSeqNum() + { + chassert(!isEphemeral()); + ++ephemeral_or_children_data.children_info.seq_num; + } + + int64_t ctime() const + { + return is_ephemeral_and_ctime.ctime; + } + + void setCtime(uint64_t ctime) + { + is_ephemeral_and_ctime.ctime = ctime; + } + + void copyStats(const Coordination::Stat & stat); +}; + +/// KeeperRocksNode is the memory structure used by RocksDB +struct KeeperRocksNode : public KeeperRocksNodeInfo +{ +#if USE_ROCKSDB + friend struct RocksDBContainer; #endif + using Meta = KeeperRocksNodeInfo; + + uint64_t size_bytes = 0; // only for compatible, should be deprecated + + uint64_t sizeInBytes() const { return data_size + sizeof(KeeperRocksNodeInfo); } + void setData(String new_data) + { + data_size = static_cast(new_data.size()); + if (data_size != 0) + { + data = std::unique_ptr(new char[new_data.size()]); + memcpy(data.get(), new_data.data(), data_size); + } + } + + void shallowCopy(const KeeperRocksNode & other) + { + czxid = other.czxid; + mzxid = other.mzxid; + pzxid = other.pzxid; + acl_id = other.acl_id; /// 0 -- no ACL by default + + mtime = other.mtime; + + is_ephemeral_and_ctime = other.is_ephemeral_and_ctime; + + ephemeral_or_children_data = other.ephemeral_or_children_data; + + data_size = other.data_size; + if (data_size != 0) + { + data = std::unique_ptr(new char[data_size]); + memcpy(data.get(), other.data.get(), data_size); + } + + version = other.version; + cversion = other.cversion; + aversion = other.aversion; + + /// cached_digest = other.cached_digest; + } + void invalidateDigestCache() const; + UInt64 getDigest(std::string_view path) const; + String getEncodedString(); + void decodeFromString(const String & buffer_str); + void recalculateSize() {} + std::string_view getData() const noexcept { return {data.get(), data_size}; } + + void setResponseStat(Coordination::Stat & response_stat) const + { + response_stat.czxid = czxid; + response_stat.mzxid = mzxid; + response_stat.ctime = ctime(); + response_stat.mtime = mtime; + response_stat.version = version; + response_stat.cversion = cversion; + response_stat.aversion = aversion; + response_stat.ephemeralOwner = ephemeralOwner(); + response_stat.dataLength = static_cast(data_size); + response_stat.numChildren = numChildren(); + response_stat.pzxid = pzxid; + } + + void reset() + { + serialized = false; + } + bool empty() const + { + return data_size == 0 && mzxid == 0; + } + std::unique_ptr data{nullptr}; + uint32_t data_size{0}; +private: + bool serialized = false; +}; + +/// KeeperMemNode should have as minimal size as possible to reduce memory footprint +/// of stored nodes +/// New fields should be added to the struct only if it's really necessary +struct KeeperMemNode +{ + int64_t czxid{0}; + int64_t mzxid{0}; + int64_t pzxid{0}; + uint64_t acl_id = 0; /// 0 -- no ACL by default + + int64_t mtime{0}; + + std::unique_ptr data{nullptr}; + uint32_t data_size{0}; + + int32_t version{0}; + int32_t cversion{0}; + int32_t aversion{0}; + + mutable uint64_t cached_digest = 0; + + KeeperMemNode() = default; + + KeeperMemNode & operator=(const KeeperMemNode & other); + KeeperMemNode(const KeeperMemNode & other); + + KeeperMemNode & operator=(KeeperMemNode && other) noexcept; + KeeperMemNode(KeeperMemNode && other) noexcept; + + bool empty() const; + + bool isEphemeral() const + { + return is_ephemeral_and_ctime.is_ephemeral; + } + + int64_t ephemeralOwner() const + { + if (isEphemeral()) + return ephemeral_or_children_data.ephemeral_owner; + + return 0; + } + + void setEphemeralOwner(int64_t ephemeral_owner) + { + is_ephemeral_and_ctime.is_ephemeral = ephemeral_owner != 0; + ephemeral_or_children_data.ephemeral_owner = ephemeral_owner; + } + + int32_t numChildren() const + { + if (isEphemeral()) + return 0; + + return ephemeral_or_children_data.children_info.num_children; + } + + void setNumChildren(int32_t num_children) + { + ephemeral_or_children_data.children_info.num_children = num_children; + } + + void increaseNumChildren() + { + chassert(!isEphemeral()); + ++ephemeral_or_children_data.children_info.num_children; + } + + void decreaseNumChildren() + { + chassert(!isEphemeral()); + --ephemeral_or_children_data.children_info.num_children; + } + + int32_t seqNum() const + { + if (isEphemeral()) + return 0; + + return ephemeral_or_children_data.children_info.seq_num; + } + + void setSeqNum(int32_t seq_num) + { + ephemeral_or_children_data.children_info.seq_num = seq_num; + } + + void increaseSeqNum() + { + chassert(!isEphemeral()); + ++ephemeral_or_children_data.children_info.seq_num; + } + + int64_t ctime() const + { + return is_ephemeral_and_ctime.ctime; + } + + void setCtime(uint64_t ctime) + { + is_ephemeral_and_ctime.ctime = ctime; + } + + void copyStats(const Coordination::Stat & stat); + + void setResponseStat(Coordination::Stat & response_stat) const; + + /// Object memory size + uint64_t sizeInBytes() const; + + void setData(const String & new_data); + + std::string_view getData() const noexcept { return {data.get(), data_size}; } + + void addChild(StringRef child_path); + + void removeChild(StringRef child_path); + + const auto & getChildren() const noexcept { return children; } + auto & getChildren() { return children; } + + // Invalidate the calculated digest so it's recalculated again on the next + // getDigest call + void invalidateDigestCache() const; + + // get the calculated digest of the node + UInt64 getDigest(std::string_view path) const; + + // copy only necessary information for preprocessing and digest calculation + // (e.g. we don't need to copy list of children) + void shallowCopy(const KeeperMemNode & other); +private: + /// as ctime can't be negative because it stores the timestamp when the + /// node was created, we can use the MSB for a bool + struct + { + bool is_ephemeral : 1; + int64_t ctime : 63; + } is_ephemeral_and_ctime{false, 0}; + + /// ephemeral notes cannot have children so a node can set either + /// ephemeral_owner OR seq_num + num_children + union + { + int64_t ephemeral_owner; + struct + { + int32_t seq_num; + int32_t num_children; + } children_info; + } ephemeral_or_children_data{0}; + + ChildrenSet children{}; +}; + +class KeeperStorageBase +{ +public: enum DigestVersion : uint8_t { @@ -200,24 +396,19 @@ public: V4 = 4 // 0 is not a valid digest value }; - static constexpr auto CURRENT_DIGEST_VERSION = DigestVersion::V4; - - struct ResponseForSession - { - int64_t session_id; - Coordination::ZooKeeperResponsePtr response; - }; - using ResponsesForSessions = std::vector; - struct Digest { DigestVersion version{DigestVersion::NO_DIGEST}; uint64_t value{0}; }; - static bool checkDigest(const Digest & first, const Digest & second); - - static String generateDigest(const String & userdata); + struct ResponseForSession + { + int64_t session_id; + Coordination::ZooKeeperResponsePtr response; + Coordination::ZooKeeperRequestPtr request = nullptr; + }; + using ResponsesForSessions = std::vector; struct RequestForSession { @@ -228,6 +419,7 @@ public: std::optional digest; int64_t log_idx{0}; }; + using RequestsForSessions = std::vector; struct AuthID { @@ -237,9 +429,6 @@ public: bool operator==(const AuthID & other) const { return scheme == other.scheme && id == other.id; } }; - using RequestsForSessions = std::vector; - - using Container = SnapshotableHashTable; using Ephemerals = std::unordered_map>; using SessionAndWatcher = std::unordered_map>; using SessionIDs = std::unordered_set; @@ -249,6 +438,38 @@ public: using SessionAndAuth = std::unordered_map; using Watches = std::unordered_map; + static bool checkDigest(const Digest & first, const Digest & second); + +}; + +/// Keeper state machine almost equal to the ZooKeeper's state machine. +/// Implements all logic of operations, data changes, sessions allocation. +/// In-memory and not thread safe. +template +class KeeperStorage : public KeeperStorageBase +{ +public: + using Container = Container_; + using Node = Container::Node; + +#if !defined(ADDRESS_SANITIZER) && !defined(MEMORY_SANITIZER) + static_assert( + sizeof(ListNode) <= 144, + "std::list node containing ListNode is > 160 bytes (sizeof(ListNode) + 16 bytes for pointers) which will increase " + "memory consumption"); +#endif + + +#if USE_ROCKSDB + static constexpr bool use_rocksdb = std::is_same_v>; +#else + static constexpr bool use_rocksdb = false; +#endif + + static constexpr auto CURRENT_DIGEST_VERSION = DigestVersion::V4; + + static String generateDigest(const String & userdata); + int64_t session_id_counter{1}; SessionAndAuth session_and_auth; @@ -392,7 +613,7 @@ public: std::unordered_map, Hash, Equal> deltas_for_path; std::list deltas; - KeeperStorage & storage; + KeeperStorage & storage; }; UncommittedState uncommitted_state{*this}; @@ -529,10 +750,16 @@ public: /// Set of methods for creating snapshots /// Turn on snapshot mode, so data inside Container is not deleted, but replaced with new version. - void enableSnapshotMode(size_t up_to_version) { container.enableSnapshotMode(up_to_version); } + void enableSnapshotMode(size_t up_to_version) + { + container.enableSnapshotMode(up_to_version); + } /// Turn off snapshot mode. - void disableSnapshotMode() { container.disableSnapshotMode(); } + void disableSnapshotMode() + { + container.disableSnapshotMode(); + } Container::const_iterator getSnapshotIteratorBegin() const { return container.begin(); } @@ -571,6 +798,9 @@ private: void addDigest(const Node & node, std::string_view path); }; -using KeeperStoragePtr = std::unique_ptr; +using KeeperMemoryStorage = KeeperStorage>; +#if USE_ROCKSDB +using KeeperRocksStorage = KeeperStorage>; +#endif } diff --git a/src/Coordination/RocksDBContainer.h b/src/Coordination/RocksDBContainer.h new file mode 100644 index 00000000000..12b40bbb87e --- /dev/null +++ b/src/Coordination/RocksDBContainer.h @@ -0,0 +1,460 @@ +#pragma once +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int ROCKSDB_ERROR; + extern const int LOGICAL_ERROR; +} + +/// The key-value format of rocks db will be +/// - key: Int8 (depth of the path) + String (path) +/// - value: SizeOf(keeperRocksNodeInfo) (meta of the node) + String (data) + +template +struct RocksDBContainer +{ + using Node = Node_; + +private: + /// MockNode is only use in test to mock `getChildren()` and `getData()` + struct MockNode + { + std::vector children; + std::string data; + MockNode(size_t children_num, std::string_view data_) + : children(std::vector(children_num)), + data(data_) + { + } + + std::vector getChildren() { return children; } + std::string getData() { return data; } + }; + + UInt16 getKeyDepth(const std::string & key) + { + UInt16 depth = 0; + for (size_t i = 0; i < key.size(); i++) + { + if (key[i] == '/' && i + 1 != key.size()) + depth ++; + } + return depth; + } + + std::string getEncodedKey(const std::string & key, bool child_prefix = false) + { + WriteBufferFromOwnString key_buffer; + UInt16 depth = getKeyDepth(key) + (child_prefix ? 1 : 0); + writeIntBinary(depth, key_buffer); + writeString(key, key_buffer); + return key_buffer.str(); + } + + static std::string_view getDecodedKey(const std::string_view & key) + { + return std::string_view(key.begin() + 2, key.end()); + } + + + struct KVPair + { + StringRef key; + Node value; + }; + + using ValueUpdater = std::function; + +public: + + /// This is an iterator wrapping rocksdb iterator and the kv result. + struct const_iterator + { + std::shared_ptr iter; + + std::shared_ptr pair; + + const_iterator() = default; + + explicit const_iterator(std::shared_ptr pair_) : pair(std::move(pair_)) {} + + explicit const_iterator(rocksdb::Iterator * iter_) : iter(iter_) + { + updatePairFromIter(); + } + + const KVPair & operator * () const + { + return *pair; + } + + const KVPair * operator->() const + { + return pair.get(); + } + + bool operator != (const const_iterator & other) const + { + return !(*this == other); + } + + bool operator == (const const_iterator & other) const + { + if (pair == nullptr && other == nullptr) + return true; + if (pair == nullptr || other == nullptr) + return false; + return pair->key.toView() == other->key.toView() && iter == other.iter; + } + + bool operator == (std::nullptr_t) const + { + return iter == nullptr; + } + + bool operator != (std::nullptr_t) const + { + return iter != nullptr; + } + + explicit operator bool() const + { + return iter != nullptr; + } + + const_iterator & operator ++() + { + iter->Next(); + updatePairFromIter(); + return *this; + } + + private: + void updatePairFromIter() + { + if (iter && iter->Valid()) + { + auto new_pair = std::make_shared(); + new_pair->key = StringRef(getDecodedKey(iter->key().ToStringView())); + ReadBufferFromOwnString buffer(iter->value().ToStringView()); + typename Node::Meta & meta = new_pair->value; + readPODBinary(meta, buffer); + readVarUInt(new_pair->value.data_size, buffer); + if (new_pair->value.data_size) + { + new_pair->value.data = std::unique_ptr(new char[new_pair->value.data_size]); + buffer.readStrict(new_pair->value.data.get(), new_pair->value.data_size); + } + pair = new_pair; + } + else + { + pair = nullptr; + iter = nullptr; + } + } + }; + + bool initialized = false; + + const const_iterator end_ptr; + + void initialize(const KeeperContextPtr & context) + { + DiskPtr disk = context->getTemporaryRocksDBDisk(); + if (disk == nullptr) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot get rocksdb disk"); + } + auto options = context->getRocksDBOptions(); + if (options == nullptr) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot get rocksdb options"); + } + rocksdb_dir = disk->getPath(); + rocksdb::DB * db; + auto status = rocksdb::DB::Open(*options, rocksdb_dir, &db); + if (!status.ok()) + { + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Failed to open rocksdb path at: {}: {}", + rocksdb_dir, status.ToString()); + } + rocksdb_ptr = std::unique_ptr(db); + write_options.disableWAL = true; + initialized = true; + } + + ~RocksDBContainer() + { + if (initialized) + { + rocksdb_ptr->Close(); + rocksdb_ptr = nullptr; + + std::filesystem::remove_all(rocksdb_dir); + } + } + + std::vector> getChildren(const std::string & key_) + { + rocksdb::ReadOptions read_options; + read_options.total_order_seek = true; + + std::string key = key_; + if (!key.ends_with('/')) + key += '/'; + size_t len = key.size() + 2; + + auto iter = std::unique_ptr(rocksdb_ptr->NewIterator(read_options)); + std::string encoded_string = getEncodedKey(key, true); + rocksdb::Slice prefix(encoded_string); + std::vector> result; + for (iter->Seek(prefix); iter->Valid() && iter->key().starts_with(prefix); iter->Next()) + { + Node node; + ReadBufferFromOwnString buffer(iter->value().ToStringView()); + typename Node::Meta & meta = node; + /// We do not read data here + readPODBinary(meta, buffer); + std::string real_key(iter->key().data() + len, iter->key().size() - len); + // std::cout << "real key: " << real_key << std::endl; + result.emplace_back(std::move(real_key), std::move(node)); + } + + return result; + } + + bool contains(const std::string & path) + { + const std::string & encoded_key = getEncodedKey(path); + std::string buffer_str; + rocksdb::Status status = rocksdb_ptr->Get(rocksdb::ReadOptions(), encoded_key, &buffer_str); + if (status.IsNotFound()) + return false; + if (!status.ok()) + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Got rocksdb error during executing contains. The error message is {}.", status.ToString()); + return true; + } + + const_iterator find(StringRef key_) + { + /// rocksdb::PinnableSlice slice; + const std::string & encoded_key = getEncodedKey(key_.toString()); + std::string buffer_str; + rocksdb::Status status = rocksdb_ptr->Get(rocksdb::ReadOptions(), encoded_key, &buffer_str); + if (status.IsNotFound()) + return end(); + if (!status.ok()) + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Got rocksdb error during executing find. The error message is {}.", status.ToString()); + ReadBufferFromOwnString buffer(buffer_str); + auto kv = std::make_shared(); + kv->key = key_; + typename Node::Meta & meta = kv->value; + readPODBinary(meta, buffer); + /// TODO: Sometimes we don't need to load data. + readVarUInt(kv->value.data_size, buffer); + if (kv->value.data_size) + { + kv->value.data = std::unique_ptr(new char[kv->value.data_size]); + buffer.readStrict(kv->value.data.get(), kv->value.data_size); + } + return const_iterator(kv); + } + + MockNode getValue(StringRef key) + { + auto it = find(key); + chassert(it != end()); + return MockNode(it->value.numChildren(), it->value.getData()); + } + + const_iterator updateValue(StringRef key_, ValueUpdater updater) + { + /// rocksdb::PinnableSlice slice; + const std::string & key = key_.toString(); + const std::string & encoded_key = getEncodedKey(key); + std::string buffer_str; + rocksdb::Status status = rocksdb_ptr->Get(rocksdb::ReadOptions(), encoded_key, &buffer_str); + if (!status.ok()) + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Got rocksdb error during find. The error message is {}.", status.ToString()); + auto kv = std::make_shared(); + kv->key = key_; + kv->value.decodeFromString(buffer_str); + /// storage->removeDigest(node, key); + updater(kv->value); + insertOrReplace(key, kv->value); + return const_iterator(kv); + } + + bool insert(const std::string & key, Node & value) + { + std::string value_str; + const std::string & encoded_key = getEncodedKey(key); + rocksdb::Status status = rocksdb_ptr->Get(rocksdb::ReadOptions(), encoded_key, &value_str); + if (status.ok()) + { + return false; + } + else if (status.IsNotFound()) + { + status = rocksdb_ptr->Put(write_options, encoded_key, value.getEncodedString()); + if (status.ok()) + { + counter++; + return true; + } + } + + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Got rocksdb error during insert. The error message is {}.", status.ToString()); + } + + void insertOrReplace(const std::string & key, Node & value) + { + const std::string & encoded_key = getEncodedKey(key); + /// storage->addDigest(value, key); + std::string value_str; + rocksdb::Status status = rocksdb_ptr->Get(rocksdb::ReadOptions(), encoded_key, &value_str); + bool increase_counter = false; + if (status.IsNotFound()) + increase_counter = true; + else if (!status.ok()) + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Got rocksdb error during get. The error message is {}.", status.ToString()); + + status = rocksdb_ptr->Put(write_options, encoded_key, value.getEncodedString()); + if (status.ok()) + counter += increase_counter; + else + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Got rocksdb error during insert. The error message is {}.", status.ToString()); + } + + using KeyPtr = std::unique_ptr; + + /// To be compatible with SnapshotableHashTable, will remove later; + KeyPtr allocateKey(size_t size) + { + return KeyPtr{new char[size]}; + } + + void insertOrReplace(KeyPtr key_data, size_t key_size, Node value) + { + std::string key(key_data.get(), key_size); + insertOrReplace(key, value); + } + + bool erase(const std::string & key) + { + /// storage->removeDigest(value, key); + const std::string & encoded_key = getEncodedKey(key); + + auto status = rocksdb_ptr->Delete(write_options, encoded_key); + if (status.IsNotFound()) + return false; + if (status.ok()) + { + counter--; + return true; + } + throw Exception(ErrorCodes::ROCKSDB_ERROR, "Got rocksdb error during erase. The error message is {}.", status.ToString()); + } + + void recalculateDataSize() {} + void reverse(size_t size_) {(void)size_;} + + uint64_t getApproximateDataSize() const + { + /// use statistics from rocksdb + return counter * sizeof(Node); + } + + void enableSnapshotMode(size_t version) + { + chassert(!snapshot_mode); + snapshot_mode = true; + snapshot_up_to_version = version; + snapshot_size = counter; + ++current_version; + + snapshot = rocksdb_ptr->GetSnapshot(); + } + + void disableSnapshotMode() + { + chassert(snapshot_mode); + snapshot_mode = false; + rocksdb_ptr->ReleaseSnapshot(snapshot); + } + + void clearOutdatedNodes() {} + + std::pair snapshotSizeWithVersion() const + { + if (!snapshot_mode) + return std::make_pair(counter, current_version); + else + return std::make_pair(snapshot_size, current_version); + } + + const_iterator begin() const + { + rocksdb::ReadOptions read_options; + read_options.total_order_seek = true; + if (snapshot_mode) + read_options.snapshot = snapshot; + auto * iter = rocksdb_ptr->NewIterator(read_options); + iter->SeekToFirst(); + return const_iterator(iter); + } + + const_iterator end() const + { + return end_ptr; + } + + size_t size() const + { + return counter; + } + + uint64_t getArenaDataSize() const + { + return 0; + } + + uint64_t keyArenaSize() const + { + return 0; + } + +private: + String rocksdb_dir; + + std::unique_ptr rocksdb_ptr; + rocksdb::WriteOptions write_options; + + const rocksdb::Snapshot * snapshot; + + bool snapshot_mode{false}; + size_t current_version{0}; + size_t snapshot_up_to_version{0}; + size_t snapshot_size{0}; + size_t counter{0}; + +}; + +} diff --git a/src/Coordination/SnapshotableHashTable.h b/src/Coordination/SnapshotableHashTable.h index 70858930115..85452558496 100644 --- a/src/Coordination/SnapshotableHashTable.h +++ b/src/Coordination/SnapshotableHashTable.h @@ -3,6 +3,7 @@ #include #include +#include namespace DB { @@ -211,9 +212,9 @@ private: updateDataSize(INSERT_OR_REPLACE, key.size, new_value_size, old_value_size, !snapshot_mode); } - public: + using Node = V; using iterator = typename List::iterator; using const_iterator = typename List::const_iterator; using ValueUpdater = std::function; @@ -363,6 +364,7 @@ public: { auto map_it = map.find(key); if (map_it != map.end()) + /// return std::make_shared(KVPair{map_it->getMapped()->key, map_it->getMapped()->value}); return map_it->getMapped(); return list.end(); } diff --git a/src/Coordination/Standalone/Context.cpp b/src/Coordination/Standalone/Context.cpp deleted file mode 100644 index 2017adcc58d..00000000000 --- a/src/Coordination/Standalone/Context.cpp +++ /dev/null @@ -1,486 +0,0 @@ -#include - -#include -#include -#include -#include -#include -#include -#include - -#include - -#include - -#include -#include - -namespace ProfileEvents -{ - extern const Event ContextLock; - extern const Event ContextLockWaitMicroseconds; -} - -namespace CurrentMetrics -{ - extern const Metric ContextLockWait; - extern const Metric BackgroundSchedulePoolTask; - extern const Metric BackgroundSchedulePoolSize; - extern const Metric IOWriterThreads; - extern const Metric IOWriterThreadsActive; - extern const Metric IOWriterThreadsScheduled; -} - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int LOGICAL_ERROR; - extern const int UNSUPPORTED_METHOD; -} - -struct ContextSharedPart : boost::noncopyable -{ - ContextSharedPart() - : macros(std::make_unique()) - {} - - ~ContextSharedPart() - { - if (keeper_dispatcher) - { - try - { - keeper_dispatcher->shutdown(); - } - catch (...) - { - tryLogCurrentException(__PRETTY_FUNCTION__); - } - } - - /// Wait for thread pool for background reads and writes, - /// since it may use per-user MemoryTracker which will be destroyed here. - if (asynchronous_remote_fs_reader) - { - try - { - asynchronous_remote_fs_reader->wait(); - asynchronous_remote_fs_reader.reset(); - } - catch (...) - { - tryLogCurrentException(__PRETTY_FUNCTION__); - } - } - - if (asynchronous_local_fs_reader) - { - try - { - asynchronous_local_fs_reader->wait(); - asynchronous_local_fs_reader.reset(); - } - catch (...) - { - tryLogCurrentException(__PRETTY_FUNCTION__); - } - } - - if (synchronous_local_fs_reader) - { - try - { - synchronous_local_fs_reader->wait(); - synchronous_local_fs_reader.reset(); - } - catch (...) - { - tryLogCurrentException(__PRETTY_FUNCTION__); - } - } - - if (threadpool_writer) - { - try - { - threadpool_writer->wait(); - threadpool_writer.reset(); - } - catch (...) - { - tryLogCurrentException(__PRETTY_FUNCTION__); - } - } - } - - /// For access of most of shared objects. - mutable SharedMutex mutex; - - ServerSettings server_settings; - - String path; /// Path to the data directory, with a slash at the end. - ConfigurationPtr config; /// Global configuration settings. - MultiVersion macros; /// Substitutions extracted from config. - OnceFlag schedule_pool_initialized; - mutable std::unique_ptr schedule_pool; /// A thread pool that can run different jobs in background - RemoteHostFilter remote_host_filter; /// Allowed URL from config.xml - - mutable OnceFlag readers_initialized; - mutable std::unique_ptr asynchronous_remote_fs_reader; - mutable std::unique_ptr asynchronous_local_fs_reader; - mutable std::unique_ptr synchronous_local_fs_reader; - -#if USE_LIBURING - mutable OnceFlag io_uring_reader_initialized; - mutable std::unique_ptr io_uring_reader; -#endif - - mutable OnceFlag threadpool_writer_initialized; - mutable std::unique_ptr threadpool_writer; - - mutable ThrottlerPtr remote_read_throttler; /// A server-wide throttler for remote IO reads - mutable ThrottlerPtr remote_write_throttler; /// A server-wide throttler for remote IO writes - - mutable ThrottlerPtr local_read_throttler; /// A server-wide throttler for local IO reads - mutable ThrottlerPtr local_write_throttler; /// A server-wide throttler for local IO writes - - std::optional storage_s3_settings TSA_GUARDED_BY(mutex); /// Settings of S3 storage - - mutable std::mutex keeper_dispatcher_mutex; - mutable std::shared_ptr keeper_dispatcher TSA_GUARDED_BY(keeper_dispatcher_mutex); -}; - -ContextData::ContextData() = default; -ContextData::ContextData(const ContextData &) = default; - -Context::Context() = default; -Context::Context(const Context & rhs) : ContextData(rhs), std::enable_shared_from_this(rhs) {} -Context::~Context() = default; - -SharedContextHolder::SharedContextHolder(SharedContextHolder &&) noexcept = default; -SharedContextHolder & SharedContextHolder::operator=(SharedContextHolder &&) noexcept = default; -SharedContextHolder::SharedContextHolder() = default; -SharedContextHolder::~SharedContextHolder() = default; -SharedContextHolder::SharedContextHolder(std::unique_ptr shared_context) - : shared(std::move(shared_context)) {} - -void SharedContextHolder::reset() { shared.reset(); } - -void Context::makeGlobalContext() -{ - initGlobal(); - global_context = shared_from_this(); -} - -ContextMutablePtr Context::createGlobal(ContextSharedPart * shared_part) -{ - auto res = std::shared_ptr(new Context); - res->shared = shared_part; - return res; -} - -void Context::initGlobal() -{ - assert(!global_context_instance); - global_context_instance = shared_from_this(); -} - -SharedContextHolder Context::createShared() -{ - return SharedContextHolder(std::make_unique()); -} - - -ContextMutablePtr Context::getGlobalContext() const -{ - auto ptr = global_context.lock(); - if (!ptr) throw Exception(ErrorCodes::LOGICAL_ERROR, "There is no global context or global context has expired"); - return ptr; -} - -std::unique_lock Context::getGlobalLock() const -{ - ProfileEvents::increment(ProfileEvents::ContextLock); - CurrentMetrics::Increment increment{CurrentMetrics::ContextLockWait}; - Stopwatch watch; - auto lock = std::unique_lock(shared->mutex); - ProfileEvents::increment(ProfileEvents::ContextLockWaitMicroseconds, watch.elapsedMicroseconds()); - return lock; -} - -std::shared_lock Context::getGlobalSharedLock() const -{ - ProfileEvents::increment(ProfileEvents::ContextLock); - CurrentMetrics::Increment increment{CurrentMetrics::ContextLockWait}; - Stopwatch watch; - auto lock = std::shared_lock(shared->mutex); - ProfileEvents::increment(ProfileEvents::ContextLockWaitMicroseconds, watch.elapsedMicroseconds()); - return lock; -} - -std::unique_lock Context::getLocalLock() const -{ - ProfileEvents::increment(ProfileEvents::ContextLock); - CurrentMetrics::Increment increment{CurrentMetrics::ContextLockWait}; - Stopwatch watch; - auto lock = std::unique_lock(mutex); - ProfileEvents::increment(ProfileEvents::ContextLockWaitMicroseconds, watch.elapsedMicroseconds()); - return lock; -} - -std::shared_lock Context::getLocalSharedLock() const -{ - ProfileEvents::increment(ProfileEvents::ContextLock); - CurrentMetrics::Increment increment{CurrentMetrics::ContextLockWait}; - Stopwatch watch; - auto lock = std::shared_lock(mutex); - ProfileEvents::increment(ProfileEvents::ContextLockWaitMicroseconds, watch.elapsedMicroseconds()); - return lock; -} - -String Context::getPath() const -{ - auto lock = getGlobalSharedLock(); - return shared->path; -} - -void Context::setPath(const String & path) -{ - auto lock = getGlobalLock(); - shared->path = path; -} - -MultiVersion::Version Context::getMacros() const -{ - return shared->macros.get(); -} - -void Context::setMacros(std::unique_ptr && macros) -{ - shared->macros.set(std::move(macros)); -} - -BackgroundSchedulePool & Context::getSchedulePool() const -{ - callOnce(shared->schedule_pool_initialized, [&] { - shared->schedule_pool = std::make_unique( - shared->server_settings.background_schedule_pool_size, - CurrentMetrics::BackgroundSchedulePoolTask, - CurrentMetrics::BackgroundSchedulePoolSize, - "BgSchPool"); - }); - - return *shared->schedule_pool; -} - -void Context::setRemoteHostFilter(const Poco::Util::AbstractConfiguration & config) -{ - shared->remote_host_filter.setValuesFromConfig(config); -} - -const RemoteHostFilter & Context::getRemoteHostFilter() const -{ - return shared->remote_host_filter; -} - -IAsynchronousReader & Context::getThreadPoolReader(FilesystemReaderType type) const -{ - callOnce(shared->readers_initialized, [&] { - const auto & config = getConfigRef(); - shared->asynchronous_remote_fs_reader = createThreadPoolReader(FilesystemReaderType::ASYNCHRONOUS_REMOTE_FS_READER, config); - shared->asynchronous_local_fs_reader = createThreadPoolReader(FilesystemReaderType::ASYNCHRONOUS_LOCAL_FS_READER, config); - shared->synchronous_local_fs_reader = createThreadPoolReader(FilesystemReaderType::SYNCHRONOUS_LOCAL_FS_READER, config); - }); - - switch (type) - { - case FilesystemReaderType::ASYNCHRONOUS_REMOTE_FS_READER: - return *shared->asynchronous_remote_fs_reader; - case FilesystemReaderType::ASYNCHRONOUS_LOCAL_FS_READER: - return *shared->asynchronous_local_fs_reader; - case FilesystemReaderType::SYNCHRONOUS_LOCAL_FS_READER: - return *shared->synchronous_local_fs_reader; - } -} - -#if USE_LIBURING -IOUringReader & Context::getIOUringReader() const -{ - callOnce(shared->io_uring_reader_initialized, [&] { - shared->io_uring_reader = createIOUringReader(); - }); - - return *shared->io_uring_reader; -} -#endif - -std::shared_ptr Context::getFilesystemCacheLog() const -{ - return nullptr; -} - -std::shared_ptr Context::getFilesystemReadPrefetchesLog() const -{ - return nullptr; -} - -std::shared_ptr Context::getBlobStorageLog() const -{ - return nullptr; -} - -void Context::setConfig(const ConfigurationPtr & config) -{ - auto lock = getGlobalLock(); - shared->config = config; -} - -const Poco::Util::AbstractConfiguration & Context::getConfigRef() const -{ - auto lock = getGlobalSharedLock(); - return shared->config ? *shared->config : Poco::Util::Application::instance().config(); -} - -std::shared_ptr Context::getAsyncReadCounters() const -{ - auto lock = getLocalLock(); - if (!async_read_counters) - async_read_counters = std::make_shared(); - return async_read_counters; -} - -ThreadPool & Context::getThreadPoolWriter() const -{ - callOnce(shared->threadpool_writer_initialized, [&] { - const auto & config = getConfigRef(); - auto pool_size = config.getUInt(".threadpool_writer_pool_size", 100); - auto queue_size = config.getUInt(".threadpool_writer_queue_size", 1000000); - - shared->threadpool_writer = std::make_unique( - CurrentMetrics::IOWriterThreads, CurrentMetrics::IOWriterThreadsActive, CurrentMetrics::IOWriterThreadsScheduled, pool_size, pool_size, queue_size); - }); - - return *shared->threadpool_writer; -} - -ThrottlerPtr Context::getRemoteReadThrottler() const -{ - return nullptr; -} - -ThrottlerPtr Context::getRemoteWriteThrottler() const -{ - return nullptr; -} - -ThrottlerPtr Context::getLocalReadThrottler() const -{ - return nullptr; -} - -ThrottlerPtr Context::getLocalWriteThrottler() const -{ - return nullptr; -} - -ReadSettings Context::getReadSettings() const -{ - return ReadSettings{}; -} - -ResourceManagerPtr Context::getResourceManager() const -{ - return nullptr; -} - -ClassifierPtr Context::getWorkloadClassifier() const -{ - return nullptr; -} - -void Context::initializeKeeperDispatcher([[maybe_unused]] bool start_async) const -{ - const auto & config_ref = getConfigRef(); - - std::lock_guard lock(shared->keeper_dispatcher_mutex); - - if (shared->keeper_dispatcher) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Trying to initialize Keeper multiple times"); - - if (config_ref.has("keeper_server")) - { - shared->keeper_dispatcher = std::make_shared(); - shared->keeper_dispatcher->initialize(config_ref, true, start_async, getMacros()); - } -} - -std::shared_ptr Context::getKeeperDispatcher() const -{ - std::lock_guard lock(shared->keeper_dispatcher_mutex); - if (!shared->keeper_dispatcher) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Keeper must be initialized before requests"); - - return shared->keeper_dispatcher; -} - -std::shared_ptr Context::tryGetKeeperDispatcher() const -{ - std::lock_guard lock(shared->keeper_dispatcher_mutex); - return shared->keeper_dispatcher; -} - -void Context::shutdownKeeperDispatcher() const -{ - std::lock_guard lock(shared->keeper_dispatcher_mutex); - if (shared->keeper_dispatcher) - { - shared->keeper_dispatcher->shutdown(); - shared->keeper_dispatcher.reset(); - } -} - -void Context::updateKeeperConfiguration([[maybe_unused]] const Poco::Util::AbstractConfiguration & config_) -{ - std::lock_guard lock(shared->keeper_dispatcher_mutex); - if (!shared->keeper_dispatcher) - return; - - shared->keeper_dispatcher->updateConfiguration(config_, getMacros()); -} - -std::shared_ptr Context::getZooKeeper() const -{ - throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "Cannot connect to ZooKeeper from Keeper"); -} - -const S3SettingsByEndpoint & Context::getStorageS3Settings() const -{ - std::lock_guard lock(shared->mutex); - - if (!shared->storage_s3_settings) - { - const auto & config = shared->config ? *shared->config : Poco::Util::Application::instance().config(); - shared->storage_s3_settings.emplace().loadFromConfig(config, "s3", getSettingsRef()); - } - - return *shared->storage_s3_settings; -} - -const ServerSettings & Context::getServerSettings() const -{ - return shared->server_settings; -} - -bool Context::hasTraceCollector() const -{ - return false; -} - -bool Context::isBackgroundOperationContext() const -{ - return false; -} - -} diff --git a/src/Coordination/Standalone/Context.h b/src/Coordination/Standalone/Context.h deleted file mode 100644 index d3bbfececed..00000000000 --- a/src/Coordination/Standalone/Context.h +++ /dev/null @@ -1,178 +0,0 @@ -#pragma once - -#include - -#include - -#include -#include -#include - -#include - -#include -#include -#include - -#include -#include - -#include - -#include - -#include "config.h" -namespace zkutil -{ - class ZooKeeper; - using ZooKeeperPtr = std::shared_ptr; -} - -namespace DB -{ - -struct ContextSharedPart; -class Macros; -class FilesystemCacheLog; -class FilesystemReadPrefetchesLog; -class BlobStorageLog; -class IOUringReader; -class S3SettingsByEndpoint; - -/// A small class which owns ContextShared. -/// We don't use something like unique_ptr directly to allow ContextShared type to be incomplete. -struct SharedContextHolder -{ - ~SharedContextHolder(); - SharedContextHolder(); - explicit SharedContextHolder(std::unique_ptr shared_context); - SharedContextHolder(SharedContextHolder &&) noexcept; - - SharedContextHolder & operator=(SharedContextHolder &&) noexcept; - - ContextSharedPart * get() const { return shared.get(); } - void reset(); -private: - std::unique_ptr shared; -}; - -class ContextData -{ -protected: - ContextWeakMutablePtr global_context; - inline static ContextPtr global_context_instance; - ContextSharedPart * shared; - - /// Query metrics for reading data asynchronously with IAsynchronousReader. - mutable std::shared_ptr async_read_counters; - - Settings settings; /// Setting for query execution. - -public: - /// Use copy constructor or createGlobal() instead - ContextData(); - ContextData(const ContextData &); -}; - -class Context : public ContextData, public std::enable_shared_from_this -{ -private: - /// ContextData mutex - mutable SharedMutex mutex; - - Context(); - Context(const Context &); - - std::unique_lock getGlobalLock() const; - - std::shared_lock getGlobalSharedLock() const; - - std::unique_lock getLocalLock() const; - - std::shared_lock getLocalSharedLock() const; - -public: - /// Create initial Context with ContextShared and etc. - static ContextMutablePtr createGlobal(ContextSharedPart * shared_part); - static SharedContextHolder createShared(); - - ContextMutablePtr getGlobalContext() const; - static ContextPtr getGlobalContextInstance() { return global_context_instance; } - - void makeGlobalContext(); - void initGlobal(); - - ~Context(); - - using ConfigurationPtr = Poco::AutoPtr; - - /// Global application configuration settings. - void setConfig(const ConfigurationPtr & config); - const Poco::Util::AbstractConfiguration & getConfigRef() const; - - const Settings & getSettingsRef() const { return settings; } - - String getPath() const; - void setPath(const String & path); - - MultiVersion::Version getMacros() const; - void setMacros(std::unique_ptr && macros); - - BackgroundSchedulePool & getSchedulePool() const; - - /// Storage of allowed hosts from config.xml - void setRemoteHostFilter(const Poco::Util::AbstractConfiguration & config); - const RemoteHostFilter & getRemoteHostFilter() const; - - std::shared_ptr getFilesystemCacheLog() const; - std::shared_ptr getFilesystemReadPrefetchesLog() const; - std::shared_ptr getBlobStorageLog() const; - - enum class ApplicationType : uint8_t - { - KEEPER, - SERVER, - }; - - void setApplicationType(ApplicationType) {} - ApplicationType getApplicationType() const { return ApplicationType::KEEPER; } - - IAsynchronousReader & getThreadPoolReader(FilesystemReaderType type) const; -#if USE_LIBURING - IOUringReader & getIOUringReader() const; -#endif - std::shared_ptr getAsyncReadCounters() const; - ThreadPool & getThreadPoolWriter() const; - - ThrottlerPtr getRemoteReadThrottler() const; - ThrottlerPtr getRemoteWriteThrottler() const; - - ThrottlerPtr getLocalReadThrottler() const; - ThrottlerPtr getLocalWriteThrottler() const; - - ReadSettings getReadSettings() const; - - /// Resource management related - ResourceManagerPtr getResourceManager() const; - ClassifierPtr getWorkloadClassifier() const; - - std::shared_ptr getKeeperDispatcher() const; - std::shared_ptr tryGetKeeperDispatcher() const; - void initializeKeeperDispatcher(bool start_async) const; - void shutdownKeeperDispatcher() const; - void updateKeeperConfiguration(const Poco::Util::AbstractConfiguration & config); - - zkutil::ZooKeeperPtr getZooKeeper() const; - - const S3SettingsByEndpoint & getStorageS3Settings() const; - - const String & getUserName() const { static std::string user; return user; } - - const ServerSettings & getServerSettings() const; - - bool hasTraceCollector() const; - - bool isBackgroundOperationContext() const; -}; - -} diff --git a/src/Coordination/Standalone/Settings.cpp b/src/Coordination/Standalone/Settings.cpp deleted file mode 100644 index 12a7a42ffac..00000000000 --- a/src/Coordination/Standalone/Settings.cpp +++ /dev/null @@ -1,24 +0,0 @@ -#include - -namespace DB -{ - -IMPLEMENT_SETTINGS_TRAITS(SettingsTraits, LIST_OF_SETTINGS) - -std::vector Settings::getAllRegisteredNames() const -{ - std::vector all_settings; - for (const auto & setting_field : all()) - { - all_settings.push_back(setting_field.getName()); - } - return all_settings; -} - -void Settings::set(std::string_view name, const Field & value) -{ - BaseSettings::set(name, value); -} - - -} diff --git a/src/Coordination/Standalone/ThreadStatusExt.cpp b/src/Coordination/Standalone/ThreadStatusExt.cpp deleted file mode 100644 index fc78233d9dc..00000000000 --- a/src/Coordination/Standalone/ThreadStatusExt.cpp +++ /dev/null @@ -1,19 +0,0 @@ -#include -#include - -namespace DB -{ - -void CurrentThread::detachFromGroupIfNotDetached() -{ -} - -void CurrentThread::attachToGroup(const ThreadGroupPtr &) -{ -} - -void ThreadStatus::initGlobalProfiler(UInt64 /*global_profiler_real_time_period*/, UInt64 /*global_profiler_cpu_time_period*/) -{ -} - -} diff --git a/src/Coordination/ZooKeeperDataReader.cpp b/src/Coordination/ZooKeeperDataReader.cpp index c205db942b9..99d71b85e78 100644 --- a/src/Coordination/ZooKeeperDataReader.cpp +++ b/src/Coordination/ZooKeeperDataReader.cpp @@ -43,7 +43,8 @@ void deserializeSnapshotMagic(ReadBuffer & in) throw Exception(ErrorCodes::CORRUPTED_DATA, "Incorrect magic header in file, expected {}, got {}", SNP_HEADER, magic_header); } -int64_t deserializeSessionAndTimeout(KeeperStorage & storage, ReadBuffer & in) +template +int64_t deserializeSessionAndTimeout(Storage & storage, ReadBuffer & in) { int32_t count; Coordination::read(count, in); @@ -62,7 +63,8 @@ int64_t deserializeSessionAndTimeout(KeeperStorage & storage, ReadBuffer & in) return max_session_id; } -void deserializeACLMap(KeeperStorage & storage, ReadBuffer & in) +template +void deserializeACLMap(Storage & storage, ReadBuffer & in) { int32_t count; Coordination::read(count, in); @@ -90,7 +92,8 @@ void deserializeACLMap(KeeperStorage & storage, ReadBuffer & in) } } -int64_t deserializeStorageData(KeeperStorage & storage, ReadBuffer & in, LoggerPtr log) +template +int64_t deserializeStorageData(Storage & storage, ReadBuffer & in, LoggerPtr log) { int64_t max_zxid = 0; std::string path; @@ -98,7 +101,7 @@ int64_t deserializeStorageData(KeeperStorage & storage, ReadBuffer & in, LoggerP size_t count = 0; while (path != "/") { - KeeperStorage::Node node{}; + typename Storage::Node node{}; String data; Coordination::read(data, in); node.setData(data); @@ -146,14 +149,15 @@ int64_t deserializeStorageData(KeeperStorage & storage, ReadBuffer & in, LoggerP if (itr.key != "/") { auto parent_path = parentNodePath(itr.key); - storage.container.updateValue(parent_path, [my_path = itr.key] (KeeperStorage::Node & value) { value.addChild(getBaseNodeName(my_path)); value.increaseNumChildren(); }); + storage.container.updateValue(parent_path, [my_path = itr.key] (typename Storage::Node & value) { value.addChild(getBaseNodeName(my_path)); value.increaseNumChildren(); }); } } return max_zxid; } -void deserializeKeeperStorageFromSnapshot(KeeperStorage & storage, const std::string & snapshot_path, LoggerPtr log) +template +void deserializeKeeperStorageFromSnapshot(Storage & storage, const std::string & snapshot_path, LoggerPtr log) { LOG_INFO(log, "Deserializing storage snapshot {}", snapshot_path); int64_t zxid = getZxidFromName(snapshot_path); @@ -192,9 +196,11 @@ void deserializeKeeperStorageFromSnapshot(KeeperStorage & storage, const std::st LOG_INFO(log, "Finished, snapshot ZXID {}", storage.zxid); } -void deserializeKeeperStorageFromSnapshotsDir(KeeperStorage & storage, const std::string & path, LoggerPtr log) +namespace fs = std::filesystem; + +template +void deserializeKeeperStorageFromSnapshotsDir(Storage & storage, const std::string & path, LoggerPtr log) { - namespace fs = std::filesystem; std::map existing_snapshots; for (const auto & p : fs::directory_iterator(path)) { @@ -480,7 +486,8 @@ bool hasErrorsInMultiRequest(Coordination::ZooKeeperRequestPtr request) } -bool deserializeTxn(KeeperStorage & storage, ReadBuffer & in, LoggerPtr /*log*/) +template +bool deserializeTxn(Storage & storage, ReadBuffer & in, LoggerPtr /*log*/) { int64_t checksum; Coordination::read(checksum, in); @@ -535,7 +542,8 @@ bool deserializeTxn(KeeperStorage & storage, ReadBuffer & in, LoggerPtr /*log*/) return true; } -void deserializeLogAndApplyToStorage(KeeperStorage & storage, const std::string & log_path, LoggerPtr log) +template +void deserializeLogAndApplyToStorage(Storage & storage, const std::string & log_path, LoggerPtr log) { ReadBufferFromFile reader(log_path); @@ -559,9 +567,9 @@ void deserializeLogAndApplyToStorage(KeeperStorage & storage, const std::string LOG_INFO(log, "Finished {} deserialization, totally read {} records", log_path, counter); } -void deserializeLogsAndApplyToStorage(KeeperStorage & storage, const std::string & path, LoggerPtr log) +template +void deserializeLogsAndApplyToStorage(Storage & storage, const std::string & path, LoggerPtr log) { - namespace fs = std::filesystem; std::map existing_logs; for (const auto & p : fs::directory_iterator(path)) { @@ -595,4 +603,9 @@ void deserializeLogsAndApplyToStorage(KeeperStorage & storage, const std::string } } +template void deserializeKeeperStorageFromSnapshot(KeeperMemoryStorage & storage, const std::string & snapshot_path, LoggerPtr log); +template void deserializeKeeperStorageFromSnapshotsDir(KeeperMemoryStorage & storage, const std::string & path, LoggerPtr log); +template void deserializeLogAndApplyToStorage(KeeperMemoryStorage & storage, const std::string & log_path, LoggerPtr log); +template void deserializeLogsAndApplyToStorage(KeeperMemoryStorage & storage, const std::string & path, LoggerPtr log); + } diff --git a/src/Coordination/ZooKeeperDataReader.h b/src/Coordination/ZooKeeperDataReader.h index 648dc95adcf..5520d1d3215 100644 --- a/src/Coordination/ZooKeeperDataReader.h +++ b/src/Coordination/ZooKeeperDataReader.h @@ -5,12 +5,16 @@ namespace DB { -void deserializeKeeperStorageFromSnapshot(KeeperStorage & storage, const std::string & snapshot_path, LoggerPtr log); +template +void deserializeKeeperStorageFromSnapshot(Storage & storage, const std::string & snapshot_path, LoggerPtr log); -void deserializeKeeperStorageFromSnapshotsDir(KeeperStorage & storage, const std::string & path, LoggerPtr log); +template +void deserializeKeeperStorageFromSnapshotsDir(Storage & storage, const std::string & path, LoggerPtr log); -void deserializeLogAndApplyToStorage(KeeperStorage & storage, const std::string & log_path, LoggerPtr log); +template +void deserializeLogAndApplyToStorage(Storage & storage, const std::string & log_path, LoggerPtr log); -void deserializeLogsAndApplyToStorage(KeeperStorage & storage, const std::string & path, LoggerPtr log); +template +void deserializeLogsAndApplyToStorage(Storage & storage, const std::string & path, LoggerPtr log); } diff --git a/src/Coordination/tests/gtest_coordination.cpp b/src/Coordination/tests/gtest_coordination.cpp index 653b1fa0a84..beae6254562 100644 --- a/src/Coordination/tests/gtest_coordination.cpp +++ b/src/Coordination/tests/gtest_coordination.cpp @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include @@ -60,10 +61,22 @@ struct CompressionParam std::string extension; }; -class CoordinationTest : public ::testing::TestWithParam +template +struct TestParam { -protected: - DB::KeeperContextPtr keeper_context = std::make_shared(true, std::make_shared()); + using Storage = Storage_; + static constexpr bool enable_compression = enable_compression_; +}; + +template +class CoordinationTest : public ::testing::Test +{ +public: + using Storage = typename TestType::Storage; + static constexpr bool enable_compression = TestType::enable_compression; + std::string extension; + + DB::KeeperContextPtr keeper_context; LoggerPtr log{getLogger("CoordinationTest")}; void SetUp() override @@ -72,7 +85,12 @@ protected: Poco::Logger::root().setChannel(channel); Poco::Logger::root().setLevel("trace"); + auto settings = std::make_shared(); + settings->experimental_use_rocksdb = true; + keeper_context = std::make_shared(true, settings); keeper_context->setLocalLogsPreprocessed(); + keeper_context->setRocksDBOptions(); + extension = enable_compression ? ".zstd" : ""; } void setLogDirectory(const std::string & path) { keeper_context->setLogDisk(std::make_shared("LogDisk", path)); } @@ -82,13 +100,24 @@ protected: keeper_context->setSnapshotDisk(std::make_shared("SnapshotDisk", path)); } + void setRocksDBDirectory(const std::string & path) + { + keeper_context->setRocksDBDisk(std::make_shared("RocksDisk", path)); + } + void setStateFileDirectory(const std::string & path) { keeper_context->setStateFileDisk(std::make_shared("StateFile", path)); } }; -TEST_P(CoordinationTest, RaftServerConfigParse) +using Implementation = testing::Types, + TestParam, + TestParam, + TestParam>; +TYPED_TEST_SUITE(CoordinationTest, Implementation); + +TYPED_TEST(CoordinationTest, RaftServerConfigParse) { auto parse = Coordination::RaftServerConfig::parse; using Cfg = std::optional; @@ -113,7 +142,7 @@ TEST_P(CoordinationTest, RaftServerConfigParse) (Cfg{{1, "2001:0db8:85a3:0000:0000:8a2e:0370:7334:80"}})); } -TEST_P(CoordinationTest, RaftServerClusterConfigParse) +TYPED_TEST(CoordinationTest, RaftServerClusterConfigParse) { auto parse = Coordination::parseRaftServers; using Cfg = DB::RaftServerConfig; @@ -129,14 +158,14 @@ TEST_P(CoordinationTest, RaftServerClusterConfigParse) (Servers{Cfg{1, "host:80"}, Cfg{2, "host:81"}})); } -TEST_P(CoordinationTest, BuildTest) +TYPED_TEST(CoordinationTest, BuildTest) { DB::InMemoryLogStore store; DB::SummingStateMachine machine; EXPECT_EQ(1, 1); } -TEST_P(CoordinationTest, BufferSerde) +TYPED_TEST(CoordinationTest, BufferSerde) { Coordination::ZooKeeperRequestPtr request = Coordination::ZooKeeperRequestFactory::instance().get(Coordination::OpNum::Get); request->xid = 3; @@ -260,13 +289,13 @@ nuraft::ptr getBuffer(int64_t number) return ret; } -TEST_P(CoordinationTest, TestSummingRaft1) +TYPED_TEST(CoordinationTest, TestSummingRaft1) { ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); - setStateFileDirectory("."); + this->setLogDirectory("./logs"); + this->setStateFileDirectory("."); - SummingRaftServer s1(1, "localhost", 44444, keeper_context); + SummingRaftServer s1(1, "localhost", 44444, this->keeper_context); SCOPE_EXIT(if (std::filesystem::exists("./state")) std::filesystem::remove("./state");); /// Single node is leader @@ -279,7 +308,7 @@ TEST_P(CoordinationTest, TestSummingRaft1) while (s1.state_machine->getValue() != 143) { - LOG_INFO(log, "Waiting s1 to apply entry"); + LOG_INFO(this->log, "Waiting s1 to apply entry"); std::this_thread::sleep_for(std::chrono::milliseconds(100)); } @@ -295,16 +324,16 @@ DB::LogEntryPtr getLogEntry(const std::string & s, size_t term) return nuraft::cs_new(term, bufwriter.getBuffer()); } -TEST_P(CoordinationTest, ChangelogTestSimple) +TYPED_TEST(CoordinationTest, ChangelogTestSimple) { - auto params = GetParam(); + /// ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); auto entry = getLogEntry("hello world", 77); changelog.append(entry); @@ -327,16 +356,15 @@ void waitDurableLogs(nuraft::log_store & log_store) } -TEST_P(CoordinationTest, ChangelogTestFile) +TYPED_TEST(CoordinationTest, ChangelogTestFile) { - auto params = GetParam(); ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); auto entry = getLogEntry("hello world", 77); changelog.append(entry); @@ -344,9 +372,9 @@ TEST_P(CoordinationTest, ChangelogTestFile) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); for (const auto & p : fs::directory_iterator("./logs")) - EXPECT_EQ(p.path(), "./logs/changelog_1_5.bin" + params.extension); + EXPECT_EQ(p.path(), "./logs/changelog_1_5.bin" + this->extension); changelog.append(entry); changelog.append(entry); @@ -357,20 +385,20 @@ TEST_P(CoordinationTest, ChangelogTestFile) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); } -TEST_P(CoordinationTest, ChangelogReadWrite) +TYPED_TEST(CoordinationTest, ChangelogReadWrite) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 1000}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 1000}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 10; ++i) @@ -385,9 +413,9 @@ TEST_P(CoordinationTest, ChangelogReadWrite) waitDurableLogs(changelog); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 1000}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 1000}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(1, 0); EXPECT_EQ(changelog_reader.size(), 10); EXPECT_EQ(changelog_reader.last_entry()->get_term(), changelog.last_entry()->get_term()); @@ -403,16 +431,16 @@ TEST_P(CoordinationTest, ChangelogReadWrite) EXPECT_EQ(10, entries_from_range->size()); } -TEST_P(CoordinationTest, ChangelogWriteAt) +TYPED_TEST(CoordinationTest, ChangelogWriteAt) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 1000}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 1000}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 10; ++i) { @@ -435,9 +463,9 @@ TEST_P(CoordinationTest, ChangelogWriteAt) EXPECT_EQ(changelog.next_slot(), 8); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 1000}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 1000}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(1, 0); EXPECT_EQ(changelog_reader.size(), changelog.size()); @@ -447,16 +475,16 @@ TEST_P(CoordinationTest, ChangelogWriteAt) } -TEST_P(CoordinationTest, ChangelogTestAppendAfterRead) +TYPED_TEST(CoordinationTest, ChangelogTestAppendAfterRead) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 7; ++i) { @@ -469,13 +497,13 @@ TEST_P(CoordinationTest, ChangelogTestAppendAfterRead) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(1, 0); EXPECT_EQ(changelog_reader.size(), 7); @@ -488,8 +516,8 @@ TEST_P(CoordinationTest, ChangelogTestAppendAfterRead) EXPECT_EQ(changelog_reader.size(), 10); waitDurableLogs(changelog_reader); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); size_t logs_count = 0; for (const auto & _ [[maybe_unused]] : fs::directory_iterator("./logs")) @@ -504,9 +532,9 @@ TEST_P(CoordinationTest, ChangelogTestAppendAfterRead) waitDurableLogs(changelog_reader); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); logs_count = 0; for (const auto & _ [[maybe_unused]] : fs::directory_iterator("./logs")) @@ -533,16 +561,16 @@ void assertFileDeleted(std::string path) } -TEST_P(CoordinationTest, ChangelogTestCompaction) +TYPED_TEST(CoordinationTest, ChangelogTestCompaction) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 3; ++i) @@ -556,7 +584,7 @@ TEST_P(CoordinationTest, ChangelogTestCompaction) EXPECT_EQ(changelog.size(), 3); - keeper_context->setLastCommitIndex(2); + this->keeper_context->setLastCommitIndex(2); changelog.compact(2); EXPECT_EQ(changelog.size(), 1); @@ -564,7 +592,7 @@ TEST_P(CoordinationTest, ChangelogTestCompaction) EXPECT_EQ(changelog.next_slot(), 4); EXPECT_EQ(changelog.last_entry()->get_term(), 20); // nothing should be deleted - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); auto e1 = getLogEntry("hello world", 30); changelog.append(e1); @@ -578,15 +606,15 @@ TEST_P(CoordinationTest, ChangelogTestCompaction) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); - keeper_context->setLastCommitIndex(6); + this->keeper_context->setLastCommitIndex(6); changelog.compact(6); std::this_thread::sleep_for(std::chrono::microseconds(1000)); - assertFileDeleted("./logs/changelog_1_5.bin" + params.extension); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); + assertFileDeleted("./logs/changelog_1_5.bin" + this->extension); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); EXPECT_EQ(changelog.size(), 1); EXPECT_EQ(changelog.start_index(), 7); @@ -594,9 +622,9 @@ TEST_P(CoordinationTest, ChangelogTestCompaction) EXPECT_EQ(changelog.last_entry()->get_term(), 60); /// And we able to read it DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(7, 0); EXPECT_EQ(changelog_reader.size(), 1); @@ -605,16 +633,16 @@ TEST_P(CoordinationTest, ChangelogTestCompaction) EXPECT_EQ(changelog_reader.last_entry()->get_term(), 60); } -TEST_P(CoordinationTest, ChangelogTestBatchOperations) +TYPED_TEST(CoordinationTest, ChangelogTestBatchOperations) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 10; ++i) { @@ -630,9 +658,9 @@ TEST_P(CoordinationTest, ChangelogTestBatchOperations) auto entries = changelog.pack(1, 5); DB::KeeperLogStore apply_changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); apply_changelog.init(1, 0); for (size_t i = 0; i < 10; ++i) @@ -660,18 +688,18 @@ TEST_P(CoordinationTest, ChangelogTestBatchOperations) EXPECT_EQ(apply_changelog.entry_at(12)->get_term(), 40); } -TEST_P(CoordinationTest, ChangelogTestBatchOperationsEmpty) +TYPED_TEST(CoordinationTest, ChangelogTestBatchOperationsEmpty) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); nuraft::ptr entries; { DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 10; ++i) { @@ -688,11 +716,11 @@ TEST_P(CoordinationTest, ChangelogTestBatchOperationsEmpty) } ChangelogDirTest test1("./logs1"); - setLogDirectory("./logs1"); + this->setLogDirectory("./logs1"); DB::KeeperLogStore changelog_new( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_new.init(1, 0); EXPECT_EQ(changelog_new.size(), 0); @@ -715,23 +743,23 @@ TEST_P(CoordinationTest, ChangelogTestBatchOperationsEmpty) EXPECT_EQ(changelog_new.next_slot(), 11); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(5, 0); } -TEST_P(CoordinationTest, ChangelogTestWriteAtPreviousFile) +TYPED_TEST(CoordinationTest, ChangelogTestWriteAtPreviousFile) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 33; ++i) @@ -743,13 +771,13 @@ TEST_P(CoordinationTest, ChangelogTestWriteAtPreviousFile) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); EXPECT_EQ(changelog.size(), 33); @@ -763,19 +791,19 @@ TEST_P(CoordinationTest, ChangelogTestWriteAtPreviousFile) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); DB::KeeperLogStore changelog_read( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_read.init(1, 0); EXPECT_EQ(changelog_read.size(), 7); EXPECT_EQ(changelog_read.start_index(), 1); @@ -783,16 +811,16 @@ TEST_P(CoordinationTest, ChangelogTestWriteAtPreviousFile) EXPECT_EQ(changelog_read.last_entry()->get_term(), 5555); } -TEST_P(CoordinationTest, ChangelogTestWriteAtFileBorder) +TYPED_TEST(CoordinationTest, ChangelogTestWriteAtFileBorder) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 33; ++i) @@ -804,13 +832,13 @@ TEST_P(CoordinationTest, ChangelogTestWriteAtFileBorder) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); EXPECT_EQ(changelog.size(), 33); @@ -824,19 +852,19 @@ TEST_P(CoordinationTest, ChangelogTestWriteAtFileBorder) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); DB::KeeperLogStore changelog_read( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_read.init(1, 0); EXPECT_EQ(changelog_read.size(), 11); EXPECT_EQ(changelog_read.start_index(), 1); @@ -844,16 +872,16 @@ TEST_P(CoordinationTest, ChangelogTestWriteAtFileBorder) EXPECT_EQ(changelog_read.last_entry()->get_term(), 5555); } -TEST_P(CoordinationTest, ChangelogTestWriteAtAllFiles) +TYPED_TEST(CoordinationTest, ChangelogTestWriteAtAllFiles) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 33; ++i) { @@ -864,13 +892,13 @@ TEST_P(CoordinationTest, ChangelogTestWriteAtAllFiles) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); EXPECT_EQ(changelog.size(), 33); @@ -884,26 +912,26 @@ TEST_P(CoordinationTest, ChangelogTestWriteAtAllFiles) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); } -TEST_P(CoordinationTest, ChangelogTestStartNewLogAfterRead) +TYPED_TEST(CoordinationTest, ChangelogTestStartNewLogAfterRead) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 35; ++i) @@ -915,19 +943,19 @@ TEST_P(CoordinationTest, ChangelogTestStartNewLogAfterRead) EXPECT_EQ(changelog.size(), 35); waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./logs/changelog_36_40.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_36_40.bin" + this->extension)); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(1, 0); auto entry = getLogEntry("36_hello_world", 360); @@ -937,14 +965,14 @@ TEST_P(CoordinationTest, ChangelogTestStartNewLogAfterRead) EXPECT_EQ(changelog_reader.size(), 36); waitDurableLogs(changelog_reader); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_36_40.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_36_40.bin" + this->extension)); } namespace @@ -965,18 +993,18 @@ void assertBrokenFileRemoved(const fs::path & directory, const fs::path & filena } -TEST_P(CoordinationTest, ChangelogTestReadAfterBrokenTruncate) +TYPED_TEST(CoordinationTest, ChangelogTestReadAfterBrokenTruncate) { static const fs::path log_folder{"./logs"}; - auto params = GetParam(); + ChangelogDirTest test(log_folder); - setLogDirectory(log_folder); + this->setLogDirectory(log_folder); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 35; ++i) @@ -988,36 +1016,36 @@ TEST_P(CoordinationTest, ChangelogTestReadAfterBrokenTruncate) EXPECT_EQ(changelog.size(), 35); waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_16_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_25.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_26_30.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_31_35.bin" + this->extension)); DB::WriteBufferFromFile plain_buf( - "./logs/changelog_11_15.bin" + params.extension, DB::DBMS_DEFAULT_BUFFER_SIZE, O_APPEND | O_CREAT | O_WRONLY); + "./logs/changelog_11_15.bin" + this->extension, DB::DBMS_DEFAULT_BUFFER_SIZE, O_APPEND | O_CREAT | O_WRONLY); plain_buf.truncate(0); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(1, 0); changelog_reader.end_of_append_batch(0, 0); EXPECT_EQ(changelog_reader.size(), 10); EXPECT_EQ(changelog_reader.last_entry()->get_term(), 90); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); - assertBrokenFileRemoved(log_folder, "changelog_16_20.bin" + params.extension); - assertBrokenFileRemoved(log_folder, "changelog_21_25.bin" + params.extension); - assertBrokenFileRemoved(log_folder, "changelog_26_30.bin" + params.extension); - assertBrokenFileRemoved(log_folder, "changelog_31_35.bin" + params.extension); + assertBrokenFileRemoved(log_folder, "changelog_16_20.bin" + this->extension); + assertBrokenFileRemoved(log_folder, "changelog_21_25.bin" + this->extension); + assertBrokenFileRemoved(log_folder, "changelog_26_30.bin" + this->extension); + assertBrokenFileRemoved(log_folder, "changelog_31_35.bin" + this->extension); auto entry = getLogEntry("h", 7777); changelog_reader.append(entry); @@ -1027,35 +1055,35 @@ TEST_P(CoordinationTest, ChangelogTestReadAfterBrokenTruncate) waitDurableLogs(changelog_reader); - EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_5.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_6_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_15.bin" + this->extension)); - assertBrokenFileRemoved(log_folder, "changelog_16_20.bin" + params.extension); - assertBrokenFileRemoved(log_folder, "changelog_21_25.bin" + params.extension); - assertBrokenFileRemoved(log_folder, "changelog_26_30.bin" + params.extension); - assertBrokenFileRemoved(log_folder, "changelog_31_35.bin" + params.extension); + assertBrokenFileRemoved(log_folder, "changelog_16_20.bin" + this->extension); + assertBrokenFileRemoved(log_folder, "changelog_21_25.bin" + this->extension); + assertBrokenFileRemoved(log_folder, "changelog_26_30.bin" + this->extension); + assertBrokenFileRemoved(log_folder, "changelog_31_35.bin" + this->extension); DB::KeeperLogStore changelog_reader2( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader2.init(1, 0); EXPECT_EQ(changelog_reader2.size(), 11); EXPECT_EQ(changelog_reader2.last_entry()->get_term(), 7777); } /// Truncating all entries -TEST_P(CoordinationTest, ChangelogTestReadAfterBrokenTruncate2) +TYPED_TEST(CoordinationTest, ChangelogTestReadAfterBrokenTruncate2) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 20}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 35; ++i) @@ -1066,22 +1094,22 @@ TEST_P(CoordinationTest, ChangelogTestReadAfterBrokenTruncate2) changelog.end_of_append_batch(0, 0); waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_40.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_40.bin" + this->extension)); DB::WriteBufferFromFile plain_buf( - "./logs/changelog_1_20.bin" + params.extension, DB::DBMS_DEFAULT_BUFFER_SIZE, O_APPEND | O_CREAT | O_WRONLY); + "./logs/changelog_1_20.bin" + this->extension, DB::DBMS_DEFAULT_BUFFER_SIZE, O_APPEND | O_CREAT | O_WRONLY); plain_buf.truncate(30); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 20}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(1, 0); EXPECT_EQ(changelog_reader.size(), 0); - EXPECT_TRUE(fs::exists("./logs/changelog_1_20.bin" + params.extension)); - assertBrokenFileRemoved("./logs", "changelog_21_40.bin" + params.extension); + EXPECT_TRUE(fs::exists("./logs/changelog_1_20.bin" + this->extension)); + assertBrokenFileRemoved("./logs", "changelog_21_40.bin" + this->extension); auto entry = getLogEntry("hello_world", 7777); changelog_reader.append(entry); changelog_reader.end_of_append_batch(0, 0); @@ -1092,9 +1120,9 @@ TEST_P(CoordinationTest, ChangelogTestReadAfterBrokenTruncate2) EXPECT_EQ(changelog_reader.last_entry()->get_term(), 7777); DB::KeeperLogStore changelog_reader2( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 1}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 1}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader2.init(1, 0); EXPECT_EQ(changelog_reader2.size(), 1); EXPECT_EQ(changelog_reader2.last_entry()->get_term(), 7777); @@ -1103,15 +1131,15 @@ TEST_P(CoordinationTest, ChangelogTestReadAfterBrokenTruncate2) /// Truncating only some entries from the end /// For compressed logs we have no reliable way of knowing how many log entries were lost /// after we truncate some bytes from the end -TEST_F(CoordinationTest, ChangelogTestReadAfterBrokenTruncate3) +TYPED_TEST(CoordinationTest, ChangelogTestReadAfterBrokenTruncate3) { ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( DB::LogFileSettings{.force_sync = true, .compress_logs = false, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 35; ++i) @@ -1133,7 +1161,7 @@ TEST_F(CoordinationTest, ChangelogTestReadAfterBrokenTruncate3) DB::KeeperLogStore changelog_reader( DB::LogFileSettings{.force_sync = true, .compress_logs = false, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_reader.init(1, 0); EXPECT_EQ(changelog_reader.size(), 19); @@ -1150,10 +1178,10 @@ TEST_F(CoordinationTest, ChangelogTestReadAfterBrokenTruncate3) EXPECT_EQ(changelog_reader.last_entry()->get_term(), 7777); } -TEST_F(CoordinationTest, ChangelogTestMixedLogTypes) +TYPED_TEST(CoordinationTest, ChangelogTestMixedLogTypes) { ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); std::vector changelog_files; @@ -1185,7 +1213,7 @@ TEST_F(CoordinationTest, ChangelogTestMixedLogTypes) DB::KeeperLogStore changelog( DB::LogFileSettings{.force_sync = true, .compress_logs = false, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 35; ++i) @@ -1206,7 +1234,7 @@ TEST_F(CoordinationTest, ChangelogTestMixedLogTypes) DB::KeeperLogStore changelog_compressed( DB::LogFileSettings{.force_sync = true, .compress_logs = true, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_compressed.init(1, 0); verify_changelog_files(); @@ -1228,7 +1256,7 @@ TEST_F(CoordinationTest, ChangelogTestMixedLogTypes) DB::KeeperLogStore changelog( DB::LogFileSettings{.force_sync = true, .compress_logs = false, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); verify_changelog_files(); @@ -1246,16 +1274,16 @@ TEST_F(CoordinationTest, ChangelogTestMixedLogTypes) } } -TEST_P(CoordinationTest, ChangelogTestLostFiles) +TYPED_TEST(CoordinationTest, ChangelogTestLostFiles) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 20}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 35; ++i) @@ -1266,30 +1294,30 @@ TEST_P(CoordinationTest, ChangelogTestLostFiles) changelog.end_of_append_batch(0, 0); waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_40.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_40.bin" + this->extension)); - fs::remove("./logs/changelog_1_20.bin" + params.extension); + fs::remove("./logs/changelog_1_20.bin" + this->extension); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 20}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 20}, DB::FlushSettings(), - keeper_context); + this->keeper_context); /// It should print error message, but still able to start changelog_reader.init(5, 0); - assertBrokenFileRemoved("./logs", "changelog_21_40.bin" + params.extension); + assertBrokenFileRemoved("./logs", "changelog_21_40.bin" + this->extension); } -TEST_P(CoordinationTest, ChangelogTestLostFiles2) +TYPED_TEST(CoordinationTest, ChangelogTestLostFiles2) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 10}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 10}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t i = 0; i < 35; ++i) @@ -1301,24 +1329,24 @@ TEST_P(CoordinationTest, ChangelogTestLostFiles2) waitDurableLogs(changelog); - EXPECT_TRUE(fs::exists("./logs/changelog_1_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_20.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_21_30.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_31_40.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_20.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_21_30.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_31_40.bin" + this->extension)); // we have a gap in our logs, we need to remove all the logs after the gap - fs::remove("./logs/changelog_21_30.bin" + params.extension); + fs::remove("./logs/changelog_21_30.bin" + this->extension); DB::KeeperLogStore changelog_reader( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 10}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 10}, DB::FlushSettings(), - keeper_context); + this->keeper_context); /// It should print error message, but still able to start changelog_reader.init(5, 0); - EXPECT_TRUE(fs::exists("./logs/changelog_1_10.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_11_20.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_10.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_11_20.bin" + this->extension)); - assertBrokenFileRemoved("./logs", "changelog_31_40.bin" + params.extension); + assertBrokenFileRemoved("./logs", "changelog_31_40.bin" + this->extension); } struct IntNode { @@ -1334,7 +1362,7 @@ struct IntNode bool operator!=(const int & rhs) const { return rhs != this->value; } }; -TEST_P(CoordinationTest, SnapshotableHashMapSimple) +TYPED_TEST(CoordinationTest, SnapshotableHashMapSimple) { DB::SnapshotableHashTable hello; EXPECT_TRUE(hello.insert("hello", 5).second); @@ -1349,7 +1377,7 @@ TEST_P(CoordinationTest, SnapshotableHashMapSimple) EXPECT_EQ(hello.size(), 0); } -TEST_P(CoordinationTest, SnapshotableHashMapTrySnapshot) +TYPED_TEST(CoordinationTest, SnapshotableHashMapTrySnapshot) { DB::SnapshotableHashTable map_snp; EXPECT_TRUE(map_snp.insert("/hello", 7).second); @@ -1426,7 +1454,7 @@ TEST_P(CoordinationTest, SnapshotableHashMapTrySnapshot) map_snp.disableSnapshotMode(); } -TEST_P(CoordinationTest, SnapshotableHashMapDataSize) +TYPED_TEST(CoordinationTest, SnapshotableHashMapDataSize) { /// int DB::SnapshotableHashTable hello; @@ -1464,7 +1492,7 @@ TEST_P(CoordinationTest, SnapshotableHashMapDataSize) EXPECT_EQ(hello.getApproximateDataSize(), 0); /// Node - using Node = DB::KeeperStorage::Node; + using Node = DB::KeeperMemoryStorage::Node; DB::SnapshotableHashTable world; Node n1; n1.setData("1234"); @@ -1503,9 +1531,10 @@ TEST_P(CoordinationTest, SnapshotableHashMapDataSize) EXPECT_EQ(world.getApproximateDataSize(), 0); } -void addNode(DB::KeeperStorage & storage, const std::string & path, const std::string & data, int64_t ephemeral_owner = 0) +template +void addNode(Storage & storage, const std::string & path, const std::string & data, int64_t ephemeral_owner = 0) { - using Node = DB::KeeperStorage::Node; + using Node = typename Storage::Node; Node node{}; node.setData(data); node.setEphemeralOwner(ephemeral_owner); @@ -1521,15 +1550,20 @@ void addNode(DB::KeeperStorage & storage, const std::string & path, const std::s }); } -TEST_P(CoordinationTest, TestStorageSnapshotSimple) +TYPED_TEST(CoordinationTest, TestStorageSnapshotSimple) { - auto params = GetParam(); + ChangelogDirTest test("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); - DB::KeeperSnapshotManager manager(3, keeper_context, params.enable_compression); + using Storage = typename TestFixture::Storage; - DB::KeeperStorage storage(500, "", keeper_context); + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + DB::KeeperSnapshotManager manager(3, this->keeper_context, this->enable_compression); + + Storage storage(500, "", this->keeper_context); addNode(storage, "/hello1", "world", 1); addNode(storage, "/hello2", "somedata", 3); storage.session_id_counter = 5; @@ -1539,7 +1573,7 @@ TEST_P(CoordinationTest, TestStorageSnapshotSimple) storage.getSessionID(130); storage.getSessionID(130); - DB::KeeperStorageSnapshot snapshot(&storage, 2); + DB::KeeperStorageSnapshot snapshot(&storage, 2); EXPECT_EQ(snapshot.snapshot_meta->get_last_log_idx(), 2); EXPECT_EQ(snapshot.session_id, 7); @@ -1548,7 +1582,7 @@ TEST_P(CoordinationTest, TestStorageSnapshotSimple) auto buf = manager.serializeSnapshotToBuffer(snapshot); manager.serializeSnapshotBufferToDisk(*buf, 2); - EXPECT_TRUE(fs::exists("./snapshots/snapshot_2.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./snapshots/snapshot_2.bin" + this->extension)); auto debuf = manager.deserializeSnapshotBufferFromDisk(2); @@ -1571,15 +1605,20 @@ TEST_P(CoordinationTest, TestStorageSnapshotSimple) EXPECT_EQ(restored_storage->session_and_timeout.size(), 2); } -TEST_P(CoordinationTest, TestStorageSnapshotMoreWrites) +TYPED_TEST(CoordinationTest, TestStorageSnapshotMoreWrites) { - auto params = GetParam(); + ChangelogDirTest test("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); - DB::KeeperSnapshotManager manager(3, keeper_context, params.enable_compression); + using Storage = typename TestFixture::Storage; - DB::KeeperStorage storage(500, "", keeper_context); + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + DB::KeeperSnapshotManager manager(3, this->keeper_context, this->enable_compression); + + Storage storage(500, "", this->keeper_context); storage.getSessionID(130); for (size_t i = 0; i < 50; ++i) @@ -1587,7 +1626,7 @@ TEST_P(CoordinationTest, TestStorageSnapshotMoreWrites) addNode(storage, "/hello_" + std::to_string(i), "world_" + std::to_string(i)); } - DB::KeeperStorageSnapshot snapshot(&storage, 50); + DB::KeeperStorageSnapshot snapshot(&storage, 50); EXPECT_EQ(snapshot.snapshot_meta->get_last_log_idx(), 50); EXPECT_EQ(snapshot.snapshot_container_size, 54); @@ -1600,7 +1639,7 @@ TEST_P(CoordinationTest, TestStorageSnapshotMoreWrites) auto buf = manager.serializeSnapshotToBuffer(snapshot); manager.serializeSnapshotBufferToDisk(*buf, 50); - EXPECT_TRUE(fs::exists("./snapshots/snapshot_50.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./snapshots/snapshot_50.bin" + this->extension)); auto debuf = manager.deserializeSnapshotBufferFromDisk(50); @@ -1614,15 +1653,20 @@ TEST_P(CoordinationTest, TestStorageSnapshotMoreWrites) } -TEST_P(CoordinationTest, TestStorageSnapshotManySnapshots) +TYPED_TEST(CoordinationTest, TestStorageSnapshotManySnapshots) { - auto params = GetParam(); + ChangelogDirTest test("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); - DB::KeeperSnapshotManager manager(3, keeper_context, params.enable_compression); + using Storage = typename TestFixture::Storage; - DB::KeeperStorage storage(500, "", keeper_context); + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + DB::KeeperSnapshotManager manager(3, this->keeper_context, this->enable_compression); + + Storage storage(500, "", this->keeper_context); storage.getSessionID(130); for (size_t j = 1; j <= 5; ++j) @@ -1632,17 +1676,17 @@ TEST_P(CoordinationTest, TestStorageSnapshotManySnapshots) addNode(storage, "/hello_" + std::to_string(i), "world_" + std::to_string(i)); } - DB::KeeperStorageSnapshot snapshot(&storage, j * 50); + DB::KeeperStorageSnapshot snapshot(&storage, j * 50); auto buf = manager.serializeSnapshotToBuffer(snapshot); manager.serializeSnapshotBufferToDisk(*buf, j * 50); - EXPECT_TRUE(fs::exists(std::string{"./snapshots/snapshot_"} + std::to_string(j * 50) + ".bin" + params.extension)); + EXPECT_TRUE(fs::exists(std::string{"./snapshots/snapshot_"} + std::to_string(j * 50) + ".bin" + this->extension)); } - EXPECT_FALSE(fs::exists("./snapshots/snapshot_50.bin" + params.extension)); - EXPECT_FALSE(fs::exists("./snapshots/snapshot_100.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./snapshots/snapshot_150.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./snapshots/snapshot_200.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./snapshots/snapshot_250.bin" + params.extension)); + EXPECT_FALSE(fs::exists("./snapshots/snapshot_50.bin" + this->extension)); + EXPECT_FALSE(fs::exists("./snapshots/snapshot_100.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./snapshots/snapshot_150.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./snapshots/snapshot_200.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./snapshots/snapshot_250.bin" + this->extension)); auto [restored_storage, meta, _] = manager.restoreFromLatestSnapshot(); @@ -1655,21 +1699,26 @@ TEST_P(CoordinationTest, TestStorageSnapshotManySnapshots) } } -TEST_P(CoordinationTest, TestStorageSnapshotMode) +TYPED_TEST(CoordinationTest, TestStorageSnapshotMode) { - auto params = GetParam(); - ChangelogDirTest test("./snapshots"); - setSnapshotDirectory("./snapshots"); - DB::KeeperSnapshotManager manager(3, keeper_context, params.enable_compression); - DB::KeeperStorage storage(500, "", keeper_context); + ChangelogDirTest test("./snapshots"); + this->setSnapshotDirectory("./snapshots"); + + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + DB::KeeperSnapshotManager manager(3, this->keeper_context, this->enable_compression); + Storage storage(500, "", this->keeper_context); for (size_t i = 0; i < 50; ++i) { addNode(storage, "/hello_" + std::to_string(i), "world_" + std::to_string(i)); } { - DB::KeeperStorageSnapshot snapshot(&storage, 50); + DB::KeeperStorageSnapshot snapshot(&storage, 50); for (size_t i = 0; i < 50; ++i) { addNode(storage, "/hello_" + std::to_string(i), "wlrd_" + std::to_string(i)); @@ -1684,12 +1733,15 @@ TEST_P(CoordinationTest, TestStorageSnapshotMode) storage.container.erase("/hello_" + std::to_string(i)); } EXPECT_EQ(storage.container.size(), 29); - EXPECT_EQ(storage.container.snapshotSizeWithVersion().first, 105); + if constexpr (Storage::use_rocksdb) + EXPECT_EQ(storage.container.snapshotSizeWithVersion().first, 54); + else + EXPECT_EQ(storage.container.snapshotSizeWithVersion().first, 105); EXPECT_EQ(storage.container.snapshotSizeWithVersion().second, 1); auto buf = manager.serializeSnapshotToBuffer(snapshot); manager.serializeSnapshotBufferToDisk(*buf, 50); } - EXPECT_TRUE(fs::exists("./snapshots/snapshot_50.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./snapshots/snapshot_50.bin" + this->extension)); EXPECT_EQ(storage.container.size(), 29); storage.clearGarbageAfterSnapshot(); EXPECT_EQ(storage.container.snapshotSizeWithVersion().first, 29); @@ -1709,28 +1761,33 @@ TEST_P(CoordinationTest, TestStorageSnapshotMode) } } -TEST_P(CoordinationTest, TestStorageSnapshotBroken) +TYPED_TEST(CoordinationTest, TestStorageSnapshotBroken) { - auto params = GetParam(); - ChangelogDirTest test("./snapshots"); - setSnapshotDirectory("./snapshots"); - DB::KeeperSnapshotManager manager(3, keeper_context, params.enable_compression); - DB::KeeperStorage storage(500, "", keeper_context); + ChangelogDirTest test("./snapshots"); + this->setSnapshotDirectory("./snapshots"); + + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + DB::KeeperSnapshotManager manager(3, this->keeper_context, this->enable_compression); + Storage storage(500, "", this->keeper_context); for (size_t i = 0; i < 50; ++i) { addNode(storage, "/hello_" + std::to_string(i), "world_" + std::to_string(i)); } { - DB::KeeperStorageSnapshot snapshot(&storage, 50); + DB::KeeperStorageSnapshot snapshot(&storage, 50); auto buf = manager.serializeSnapshotToBuffer(snapshot); manager.serializeSnapshotBufferToDisk(*buf, 50); } - EXPECT_TRUE(fs::exists("./snapshots/snapshot_50.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./snapshots/snapshot_50.bin" + this->extension)); /// Let's corrupt file DB::WriteBufferFromFile plain_buf( - "./snapshots/snapshot_50.bin" + params.extension, DB::DBMS_DEFAULT_BUFFER_SIZE, O_APPEND | O_CREAT | O_WRONLY); + "./snapshots/snapshot_50.bin" + this->extension, DB::DBMS_DEFAULT_BUFFER_SIZE, O_APPEND | O_CREAT | O_WRONLY); plain_buf.truncate(34); plain_buf.sync(); @@ -1746,7 +1803,7 @@ nuraft::ptr getBufferFromZKRequest(int64_t session_id, int64_t z auto time = duration_cast(system_clock::now().time_since_epoch()).count(); DB::writeIntBinary(time, buf); DB::writeIntBinary(zxid, buf); - DB::writeIntBinary(DB::KeeperStorage::DigestVersion::NO_DIGEST, buf); + DB::writeIntBinary(DB::KeeperMemoryStorage::DigestVersion::NO_DIGEST, buf); return buf.getBuffer(); } @@ -1757,6 +1814,7 @@ getLogEntryFromZKRequest(size_t term, int64_t session_id, int64_t zxid, const Co return nuraft::cs_new(term, buffer); } +template void testLogAndStateMachine( DB::CoordinationSettingsPtr settings, uint64_t total_logs, @@ -1767,12 +1825,15 @@ void testLogAndStateMachine( ChangelogDirTest snapshots("./snapshots"); ChangelogDirTest logs("./logs"); + ChangelogDirTest rocks("./rocksdb"); auto get_keeper_context = [&] { auto local_keeper_context = std::make_shared(true, settings); local_keeper_context->setSnapshotDisk(std::make_shared("SnapshotDisk", "./snapshots")); local_keeper_context->setLogDisk(std::make_shared("LogDisk", "./logs")); + local_keeper_context->setRocksDBDisk(std::make_shared("RocksDisk", "./rocksdb")); + local_keeper_context->setRocksDBOptions(); return local_keeper_context; }; @@ -1780,7 +1841,7 @@ void testLogAndStateMachine( SnapshotsQueue snapshots_queue{1}; auto keeper_context = get_keeper_context(); - auto state_machine = std::make_shared(queue, snapshots_queue, keeper_context, nullptr); + auto state_machine = std::make_shared>(queue, snapshots_queue, keeper_context, nullptr); state_machine->init(); DB::KeeperLogStore changelog( @@ -1827,7 +1888,7 @@ void testLogAndStateMachine( SnapshotsQueue snapshots_queue1{1}; keeper_context = get_keeper_context(); - auto restore_machine = std::make_shared(queue, snapshots_queue1, keeper_context, nullptr); + auto restore_machine = std::make_shared>(queue, snapshots_queue1, keeper_context, nullptr); restore_machine->init(); EXPECT_EQ(restore_machine->last_commit_index(), total_logs - total_logs % settings->snapshot_distance); @@ -1863,11 +1924,12 @@ void testLogAndStateMachine( } } -TEST_P(CoordinationTest, TestStateMachineAndLogStore) +TYPED_TEST(CoordinationTest, TestStateMachineAndLogStore) { using namespace Coordination; using namespace DB; - auto params = GetParam(); + + using Storage = typename TestFixture::Storage; { CoordinationSettingsPtr settings = std::make_shared(); @@ -1875,78 +1937,83 @@ TEST_P(CoordinationTest, TestStateMachineAndLogStore) settings->reserved_log_items = 10; settings->rotate_log_storage_interval = 10; - testLogAndStateMachine(settings, 37, params.enable_compression); + testLogAndStateMachine(settings, 37, this->enable_compression); } { CoordinationSettingsPtr settings = std::make_shared(); settings->snapshot_distance = 10; settings->reserved_log_items = 10; settings->rotate_log_storage_interval = 10; - testLogAndStateMachine(settings, 11, params.enable_compression); + testLogAndStateMachine(settings, 11, this->enable_compression); } { CoordinationSettingsPtr settings = std::make_shared(); settings->snapshot_distance = 10; settings->reserved_log_items = 10; settings->rotate_log_storage_interval = 10; - testLogAndStateMachine(settings, 40, params.enable_compression); + testLogAndStateMachine(settings, 40, this->enable_compression); } { CoordinationSettingsPtr settings = std::make_shared(); settings->snapshot_distance = 10; settings->reserved_log_items = 20; settings->rotate_log_storage_interval = 30; - testLogAndStateMachine(settings, 40, params.enable_compression); + testLogAndStateMachine(settings, 40, this->enable_compression); } { CoordinationSettingsPtr settings = std::make_shared(); settings->snapshot_distance = 10; settings->reserved_log_items = 0; settings->rotate_log_storage_interval = 10; - testLogAndStateMachine(settings, 40, params.enable_compression); + testLogAndStateMachine(settings, 40, this->enable_compression); } { CoordinationSettingsPtr settings = std::make_shared(); settings->snapshot_distance = 1; settings->reserved_log_items = 1; settings->rotate_log_storage_interval = 32; - testLogAndStateMachine(settings, 32, params.enable_compression); + testLogAndStateMachine(settings, 32, this->enable_compression); } { CoordinationSettingsPtr settings = std::make_shared(); settings->snapshot_distance = 10; settings->reserved_log_items = 7; settings->rotate_log_storage_interval = 1; - testLogAndStateMachine(settings, 33, params.enable_compression); + testLogAndStateMachine(settings, 33, this->enable_compression); } { CoordinationSettingsPtr settings = std::make_shared(); settings->snapshot_distance = 37; settings->reserved_log_items = 1000; settings->rotate_log_storage_interval = 5000; - testLogAndStateMachine(settings, 33, params.enable_compression); + testLogAndStateMachine(settings, 33, this->enable_compression); } { CoordinationSettingsPtr settings = std::make_shared(); settings->snapshot_distance = 37; settings->reserved_log_items = 1000; settings->rotate_log_storage_interval = 5000; - testLogAndStateMachine(settings, 45, params.enable_compression); + testLogAndStateMachine(settings, 45, this->enable_compression); } } -TEST_P(CoordinationTest, TestEphemeralNodeRemove) +TYPED_TEST(CoordinationTest, TestEphemeralNodeRemove) { using namespace Coordination; using namespace DB; ChangelogDirTest snapshots("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); + + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); ResponsesQueue queue(std::numeric_limits::max()); SnapshotsQueue snapshots_queue{1}; - auto state_machine = std::make_shared(queue, snapshots_queue, keeper_context, nullptr); + auto state_machine = std::make_shared>(queue, snapshots_queue, this->keeper_context, nullptr); state_machine->init(); std::shared_ptr request_c = std::make_shared(); @@ -1969,21 +2036,27 @@ TEST_P(CoordinationTest, TestEphemeralNodeRemove) } -TEST_P(CoordinationTest, TestCreateNodeWithAuthSchemeForAclWhenAuthIsPrecommitted) +TYPED_TEST(CoordinationTest, TestCreateNodeWithAuthSchemeForAclWhenAuthIsPrecommitted) { using namespace Coordination; using namespace DB; ChangelogDirTest snapshots("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); + + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + ResponsesQueue queue(std::numeric_limits::max()); SnapshotsQueue snapshots_queue{1}; - auto state_machine = std::make_shared(queue, snapshots_queue, keeper_context, nullptr); + auto state_machine = std::make_shared>(queue, snapshots_queue, this->keeper_context, nullptr); state_machine->init(); String user_auth_data = "test_user:test_password"; - String digest = KeeperStorage::generateDigest(user_auth_data); + String digest = KeeperMemoryStorage::generateDigest(user_auth_data); std::shared_ptr auth_req = std::make_shared(); auth_req->scheme = "digest"; @@ -2019,83 +2092,211 @@ TEST_P(CoordinationTest, TestCreateNodeWithAuthSchemeForAclWhenAuthIsPrecommitte EXPECT_EQ(acls[0].permissions, 31); } -TEST_P(CoordinationTest, TestPreprocessWhenCloseSessionIsPrecommitted) +TYPED_TEST(CoordinationTest, TestPreprocessWhenCloseSessionIsPrecommitted) { using namespace Coordination; using namespace DB; ChangelogDirTest snapshots("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); + + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); ResponsesQueue queue(std::numeric_limits::max()); SnapshotsQueue snapshots_queue{1}; - int64_t session_id = 1; + int64_t session_without_auth = 1; + int64_t session_with_auth = 2; size_t term = 0; - auto state_machine = std::make_shared(queue, snapshots_queue, keeper_context, nullptr); + auto state_machine = std::make_shared>(queue, snapshots_queue, this->keeper_context, nullptr); state_machine->init(); - auto & storage = state_machine->getStorageUnsafe(); - const auto & uncommitted_state = storage.uncommitted_state; + auto & storage = state_machine->getStorageUnsafe(); + const auto & uncommitted_state = storage.uncommitted_state; - // Create first node for the session - String node_path_1 = "/node_1"; - std::shared_ptr create_req_1 = std::make_shared(); - create_req_1->path = node_path_1; - auto create_entry_1 = getLogEntryFromZKRequest(term, session_id, state_machine->getNextZxid(), create_req_1); + auto auth_req = std::make_shared(); + auth_req->scheme = "digest"; + auth_req->data = "test_user:test_password"; - state_machine->pre_commit(1, create_entry_1->get_buf()); - EXPECT_TRUE(uncommitted_state.nodes.contains(node_path_1)); + // Add auth data to the session + auto auth_entry = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), auth_req); + state_machine->pre_commit(1, auth_entry->get_buf()); + state_machine->commit(1, auth_entry->get_buf()); - state_machine->commit(1, create_entry_1->get_buf()); - EXPECT_TRUE(storage.container.contains(node_path_1)); + std::string node_without_acl = "/node_without_acl"; + { + auto create_req = std::make_shared(); + create_req->path = node_without_acl; + create_req->data = "notmodified"; + auto create_entry = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), create_req); + state_machine->pre_commit(2, create_entry->get_buf()); + state_machine->commit(2, create_entry->get_buf()); + ASSERT_TRUE(storage.container.contains(node_without_acl)); + } - // Close session - std::shared_ptr close_req = std::make_shared(); - auto close_entry = getLogEntryFromZKRequest(term, session_id, state_machine->getNextZxid(), close_req); - // Pre-commit close session - state_machine->pre_commit(2, close_entry->get_buf()); + std::string node_with_acl = "/node_with_acl"; + { + auto create_req = std::make_shared(); + create_req->path = node_with_acl; + create_req->data = "notmodified"; + create_req->acls = {{.permissions = ACL::All, .scheme = "auth", .id = ""}}; + auto create_entry = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), create_req); + state_machine->pre_commit(3, create_entry->get_buf()); + state_machine->commit(3, create_entry->get_buf()); + ASSERT_TRUE(storage.container.contains(node_with_acl)); + } - // Try to create second node after close session is pre-committed - String node_path_2 = "/node_2"; - std::shared_ptr create_req_2 = std::make_shared(); - create_req_2->path = node_path_2; - auto create_entry_2 = getLogEntryFromZKRequest(term, session_id, state_machine->getNextZxid(), create_req_2); - - // Pre-commit creating second node - state_machine->pre_commit(3, create_entry_2->get_buf()); - // Second node wasn't created - EXPECT_FALSE(uncommitted_state.nodes.contains(node_path_2)); + auto set_req_with_acl = std::make_shared(); + set_req_with_acl->path = node_with_acl; + set_req_with_acl->data = "modified"; - // Rollback pre-committed closing session - state_machine->rollback(3, create_entry_2->get_buf()); - state_machine->rollback(2, close_entry->get_buf()); + auto set_req_without_acl = std::make_shared(); + set_req_without_acl->path = node_without_acl; + set_req_without_acl->data = "modified"; - // Pre-commit creating second node - state_machine->pre_commit(2, create_entry_2->get_buf()); - // Now second node was created - EXPECT_TRUE(uncommitted_state.nodes.contains(node_path_2)); + const auto reset_node_value + = [&](const auto & path) { storage.container.updateValue(path, [](auto & node) { node.setData("notmodified"); }); }; - state_machine->commit(2, create_entry_2->get_buf()); - EXPECT_TRUE(storage.container.contains(node_path_1)); - EXPECT_TRUE(storage.container.contains(node_path_2)); + auto close_req = std::make_shared(); + + { + SCOPED_TRACE("Session with Auth"); + + // test we can modify both nodes + auto set_entry = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), set_req_with_acl); + state_machine->pre_commit(5, set_entry->get_buf()); + state_machine->commit(5, set_entry->get_buf()); + ASSERT_TRUE(storage.container.find(node_with_acl)->value.getData() == "modified"); + reset_node_value(node_with_acl); + + set_entry = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), set_req_without_acl); + state_machine->pre_commit(6, set_entry->get_buf()); + state_machine->commit(6, set_entry->get_buf()); + ASSERT_TRUE(storage.container.find(node_without_acl)->value.getData() == "modified"); + reset_node_value(node_without_acl); + + auto close_entry = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), close_req); + + // Pre-commit close session + state_machine->pre_commit(7, close_entry->get_buf()); + + /// will be rejected because we don't have required auth + auto set_entry_with_acl = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), set_req_with_acl); + state_machine->pre_commit(8, set_entry_with_acl->get_buf()); + + /// will be accepted because no ACL + auto set_entry_without_acl = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), set_req_without_acl); + state_machine->pre_commit(9, set_entry_without_acl->get_buf()); + + ASSERT_TRUE(uncommitted_state.getNode(node_with_acl)->getData() == "notmodified"); + ASSERT_TRUE(uncommitted_state.getNode(node_without_acl)->getData() == "modified"); + + state_machine->rollback(9, set_entry_without_acl->get_buf()); + state_machine->rollback(8, set_entry_with_acl->get_buf()); + + // let's commit close and verify we get same outcome + state_machine->commit(7, close_entry->get_buf()); + + /// will be rejected because we don't have required auth + set_entry_with_acl = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), set_req_with_acl); + state_machine->pre_commit(8, set_entry_with_acl->get_buf()); + + /// will be accepted because no ACL + set_entry_without_acl = getLogEntryFromZKRequest(term, session_with_auth, state_machine->getNextZxid(), set_req_without_acl); + state_machine->pre_commit(9, set_entry_without_acl->get_buf()); + + ASSERT_TRUE(uncommitted_state.getNode(node_with_acl)->getData() == "notmodified"); + ASSERT_TRUE(uncommitted_state.getNode(node_without_acl)->getData() == "modified"); + + state_machine->commit(8, set_entry_with_acl->get_buf()); + state_machine->commit(9, set_entry_without_acl->get_buf()); + + ASSERT_TRUE(storage.container.find(node_with_acl)->value.getData() == "notmodified"); + ASSERT_TRUE(storage.container.find(node_without_acl)->value.getData() == "modified"); + + reset_node_value(node_without_acl); + } + + { + SCOPED_TRACE("Session without Auth"); + + // test we can modify only node without acl + auto set_entry = getLogEntryFromZKRequest(term, session_without_auth, state_machine->getNextZxid(), set_req_with_acl); + state_machine->pre_commit(10, set_entry->get_buf()); + state_machine->commit(10, set_entry->get_buf()); + ASSERT_TRUE(storage.container.find(node_with_acl)->value.getData() == "notmodified"); + + set_entry = getLogEntryFromZKRequest(term, session_without_auth, state_machine->getNextZxid(), set_req_without_acl); + state_machine->pre_commit(11, set_entry->get_buf()); + state_machine->commit(11, set_entry->get_buf()); + ASSERT_TRUE(storage.container.find(node_without_acl)->value.getData() == "modified"); + reset_node_value(node_without_acl); + + auto close_entry = getLogEntryFromZKRequest(term, session_without_auth, state_machine->getNextZxid(), close_req); + + // Pre-commit close session + state_machine->pre_commit(12, close_entry->get_buf()); + + /// will be rejected because we don't have required auth + auto set_entry_with_acl = getLogEntryFromZKRequest(term, session_without_auth, state_machine->getNextZxid(), set_req_with_acl); + state_machine->pre_commit(13, set_entry_with_acl->get_buf()); + + /// will be accepted because no ACL + auto set_entry_without_acl = getLogEntryFromZKRequest(term, session_without_auth, state_machine->getNextZxid(), set_req_without_acl); + state_machine->pre_commit(14, set_entry_without_acl->get_buf()); + + ASSERT_TRUE(uncommitted_state.getNode(node_with_acl)->getData() == "notmodified"); + ASSERT_TRUE(uncommitted_state.getNode(node_without_acl)->getData() == "modified"); + + state_machine->rollback(14, set_entry_without_acl->get_buf()); + state_machine->rollback(13, set_entry_with_acl->get_buf()); + + // let's commit close and verify we get same outcome + state_machine->commit(12, close_entry->get_buf()); + + /// will be rejected because we don't have required auth + set_entry_with_acl = getLogEntryFromZKRequest(term, session_without_auth, state_machine->getNextZxid(), set_req_with_acl); + state_machine->pre_commit(13, set_entry_with_acl->get_buf()); + + /// will be accepted because no ACL + set_entry_without_acl = getLogEntryFromZKRequest(term, session_without_auth, state_machine->getNextZxid(), set_req_without_acl); + state_machine->pre_commit(14, set_entry_without_acl->get_buf()); + + ASSERT_TRUE(uncommitted_state.getNode(node_with_acl)->getData() == "notmodified"); + ASSERT_TRUE(uncommitted_state.getNode(node_without_acl)->getData() == "modified"); + + state_machine->commit(13, set_entry_with_acl->get_buf()); + state_machine->commit(14, set_entry_without_acl->get_buf()); + + ASSERT_TRUE(storage.container.find(node_with_acl)->value.getData() == "notmodified"); + ASSERT_TRUE(storage.container.find(node_without_acl)->value.getData() == "modified"); + + reset_node_value(node_without_acl); + } } -TEST_P(CoordinationTest, TestSetACLWithAuthSchemeForAclWhenAuthIsPrecommitted) +TYPED_TEST(CoordinationTest, TestSetACLWithAuthSchemeForAclWhenAuthIsPrecommitted) { using namespace Coordination; using namespace DB; ChangelogDirTest snapshots("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); ResponsesQueue queue(std::numeric_limits::max()); SnapshotsQueue snapshots_queue{1}; - auto state_machine = std::make_shared(queue, snapshots_queue, keeper_context, nullptr); + using Storage = typename TestFixture::Storage; + auto state_machine = std::make_shared>(queue, snapshots_queue, this->keeper_context, nullptr); state_machine->init(); String user_auth_data = "test_user:test_password"; - String digest = KeeperStorage::generateDigest(user_auth_data); + String digest = KeeperMemoryStorage::generateDigest(user_auth_data); std::shared_ptr auth_req = std::make_shared(); auth_req->scheme = "digest"; @@ -2138,17 +2339,17 @@ TEST_P(CoordinationTest, TestSetACLWithAuthSchemeForAclWhenAuthIsPrecommitted) } -TEST_P(CoordinationTest, TestRotateIntervalChanges) +TYPED_TEST(CoordinationTest, TestRotateIntervalChanges) { using namespace Coordination; - auto params = GetParam(); + ChangelogDirTest snapshots("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); { DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(0, 3); for (size_t i = 1; i < 55; ++i) @@ -2164,12 +2365,12 @@ TEST_P(CoordinationTest, TestRotateIntervalChanges) } - EXPECT_TRUE(fs::exists("./logs/changelog_1_100.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_100.bin" + this->extension)); DB::KeeperLogStore changelog_1( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 10}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 10}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_1.init(0, 50); for (size_t i = 0; i < 55; ++i) { @@ -2182,13 +2383,13 @@ TEST_P(CoordinationTest, TestRotateIntervalChanges) waitDurableLogs(changelog_1); - EXPECT_TRUE(fs::exists("./logs/changelog_1_100.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_101_110.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_1_100.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_101_110.bin" + this->extension)); DB::KeeperLogStore changelog_2( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 7}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 7}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_2.init(98, 55); for (size_t i = 0; i < 17; ++i) @@ -2202,20 +2403,20 @@ TEST_P(CoordinationTest, TestRotateIntervalChanges) waitDurableLogs(changelog_2); - keeper_context->setLastCommitIndex(105); + this->keeper_context->setLastCommitIndex(105); changelog_2.compact(105); std::this_thread::sleep_for(std::chrono::microseconds(1000)); - assertFileDeleted("./logs/changelog_1_100.bin" + params.extension); - EXPECT_TRUE(fs::exists("./logs/changelog_101_110.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_111_117.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_118_124.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_125_131.bin" + params.extension)); + assertFileDeleted("./logs/changelog_1_100.bin" + this->extension); + EXPECT_TRUE(fs::exists("./logs/changelog_101_110.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_111_117.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_118_124.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_125_131.bin" + this->extension)); DB::KeeperLogStore changelog_3( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 5}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 5}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog_3.init(116, 3); for (size_t i = 0; i < 17; ++i) { @@ -2228,20 +2429,20 @@ TEST_P(CoordinationTest, TestRotateIntervalChanges) waitDurableLogs(changelog_3); - keeper_context->setLastCommitIndex(125); + this->keeper_context->setLastCommitIndex(125); changelog_3.compact(125); std::this_thread::sleep_for(std::chrono::microseconds(1000)); - assertFileDeleted("./logs/changelog_101_110.bin" + params.extension); - assertFileDeleted("./logs/changelog_111_117.bin" + params.extension); - assertFileDeleted("./logs/changelog_118_124.bin" + params.extension); + assertFileDeleted("./logs/changelog_101_110.bin" + this->extension); + assertFileDeleted("./logs/changelog_111_117.bin" + this->extension); + assertFileDeleted("./logs/changelog_118_124.bin" + this->extension); - EXPECT_TRUE(fs::exists("./logs/changelog_125_131.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_132_136.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_137_141.bin" + params.extension)); - EXPECT_TRUE(fs::exists("./logs/changelog_142_146.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_125_131.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_132_136.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_137_141.bin" + this->extension)); + EXPECT_TRUE(fs::exists("./logs/changelog_142_146.bin" + this->extension)); } -TEST_P(CoordinationTest, TestSessionExpiryQueue) +TYPED_TEST(CoordinationTest, TestSessionExpiryQueue) { using namespace Coordination; SessionExpiryQueue queue(500); @@ -2259,16 +2460,15 @@ TEST_P(CoordinationTest, TestSessionExpiryQueue) } -TEST_P(CoordinationTest, TestCompressedLogsMultipleRewrite) +TYPED_TEST(CoordinationTest, TestCompressedLogsMultipleRewrite) { using namespace Coordination; - auto test_params = GetParam(); ChangelogDirTest logs("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = test_params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(0, 3); for (size_t i = 1; i < 55; ++i) @@ -2283,9 +2483,9 @@ TEST_P(CoordinationTest, TestCompressedLogsMultipleRewrite) waitDurableLogs(changelog); DB::KeeperLogStore changelog1( - DB::LogFileSettings{.force_sync = true, .compress_logs = test_params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog1.init(0, 3); for (size_t i = 55; i < 70; ++i) { @@ -2299,9 +2499,9 @@ TEST_P(CoordinationTest, TestCompressedLogsMultipleRewrite) waitDurableLogs(changelog1); DB::KeeperLogStore changelog2( - DB::LogFileSettings{.force_sync = true, .compress_logs = test_params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog2.init(0, 3); for (size_t i = 70; i < 80; ++i) { @@ -2313,16 +2513,21 @@ TEST_P(CoordinationTest, TestCompressedLogsMultipleRewrite) } } -TEST_P(CoordinationTest, TestStorageSnapshotDifferentCompressions) +TYPED_TEST(CoordinationTest, TestStorageSnapshotDifferentCompressions) { - auto params = GetParam(); + ChangelogDirTest test("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); - DB::KeeperSnapshotManager manager(3, keeper_context, params.enable_compression); + using Storage = typename TestFixture::Storage; - DB::KeeperStorage storage(500, "", keeper_context); + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + DB::KeeperSnapshotManager manager(3, this->keeper_context, this->enable_compression); + + Storage storage(500, "", this->keeper_context); addNode(storage, "/hello1", "world", 1); addNode(storage, "/hello2", "somedata", 3); storage.session_id_counter = 5; @@ -2332,13 +2537,13 @@ TEST_P(CoordinationTest, TestStorageSnapshotDifferentCompressions) storage.getSessionID(130); storage.getSessionID(130); - DB::KeeperStorageSnapshot snapshot(&storage, 2); + DB::KeeperStorageSnapshot snapshot(&storage, 2); auto buf = manager.serializeSnapshotToBuffer(snapshot); manager.serializeSnapshotBufferToDisk(*buf, 2); - EXPECT_TRUE(fs::exists("./snapshots/snapshot_2.bin" + params.extension)); + EXPECT_TRUE(fs::exists("./snapshots/snapshot_2.bin" + this->extension)); - DB::KeeperSnapshotManager new_manager(3, keeper_context, !params.enable_compression); + DB::KeeperSnapshotManager new_manager(3, this->keeper_context, !this->enable_compression); auto debuf = new_manager.deserializeSnapshotBufferFromDisk(2); @@ -2360,17 +2565,17 @@ TEST_P(CoordinationTest, TestStorageSnapshotDifferentCompressions) EXPECT_EQ(restored_storage->session_and_timeout.size(), 2); } -TEST_P(CoordinationTest, ChangelogInsertThreeTimesSmooth) +TYPED_TEST(CoordinationTest, ChangelogInsertThreeTimesSmooth) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); { - LOG_INFO(log, "================First time====================="); + LOG_INFO(this->log, "================First time====================="); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); auto entry = getLogEntry("hello_world", 1000); changelog.append(entry); @@ -2380,11 +2585,11 @@ TEST_P(CoordinationTest, ChangelogInsertThreeTimesSmooth) } { - LOG_INFO(log, "================Second time====================="); + LOG_INFO(this->log, "================Second time====================="); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); auto entry = getLogEntry("hello_world", 1000); changelog.append(entry); @@ -2394,11 +2599,11 @@ TEST_P(CoordinationTest, ChangelogInsertThreeTimesSmooth) } { - LOG_INFO(log, "================Third time====================="); + LOG_INFO(this->log, "================Third time====================="); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); auto entry = getLogEntry("hello_world", 1000); changelog.append(entry); @@ -2408,11 +2613,11 @@ TEST_P(CoordinationTest, ChangelogInsertThreeTimesSmooth) } { - LOG_INFO(log, "================Fourth time====================="); + LOG_INFO(this->log, "================Fourth time====================="); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); auto entry = getLogEntry("hello_world", 1000); changelog.append(entry); @@ -2423,18 +2628,18 @@ TEST_P(CoordinationTest, ChangelogInsertThreeTimesSmooth) } -TEST_P(CoordinationTest, ChangelogInsertMultipleTimesSmooth) +TYPED_TEST(CoordinationTest, ChangelogInsertMultipleTimesSmooth) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); for (size_t i = 0; i < 36; ++i) { - LOG_INFO(log, "================First time====================="); + LOG_INFO(this->log, "================First time====================="); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (size_t j = 0; j < 7; ++j) { @@ -2446,24 +2651,24 @@ TEST_P(CoordinationTest, ChangelogInsertMultipleTimesSmooth) } DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); EXPECT_EQ(changelog.next_slot(), 36 * 7 + 1); } -TEST_P(CoordinationTest, ChangelogInsertThreeTimesHard) +TYPED_TEST(CoordinationTest, ChangelogInsertThreeTimesHard) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); { - LOG_INFO(log, "================First time====================="); + LOG_INFO(this->log, "================First time====================="); DB::KeeperLogStore changelog1( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog1.init(1, 0); auto entry = getLogEntry("hello_world", 1000); changelog1.append(entry); @@ -2473,11 +2678,11 @@ TEST_P(CoordinationTest, ChangelogInsertThreeTimesHard) } { - LOG_INFO(log, "================Second time====================="); + LOG_INFO(this->log, "================Second time====================="); DB::KeeperLogStore changelog2( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog2.init(1, 0); auto entry = getLogEntry("hello_world", 1000); changelog2.append(entry); @@ -2487,11 +2692,11 @@ TEST_P(CoordinationTest, ChangelogInsertThreeTimesHard) } { - LOG_INFO(log, "================Third time====================="); + LOG_INFO(this->log, "================Third time====================="); DB::KeeperLogStore changelog3( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog3.init(1, 0); auto entry = getLogEntry("hello_world", 1000); changelog3.append(entry); @@ -2501,11 +2706,11 @@ TEST_P(CoordinationTest, ChangelogInsertThreeTimesHard) } { - LOG_INFO(log, "================Fourth time====================="); + LOG_INFO(this->log, "================Fourth time====================="); DB::KeeperLogStore changelog4( - DB::LogFileSettings{.force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog4.init(1, 0); auto entry = getLogEntry("hello_world", 1000); changelog4.append(entry); @@ -2515,18 +2720,23 @@ TEST_P(CoordinationTest, ChangelogInsertThreeTimesHard) } } -TEST_P(CoordinationTest, TestStorageSnapshotEqual) +TYPED_TEST(CoordinationTest, TestStorageSnapshotEqual) { - auto params = GetParam(); + ChangelogDirTest test("./snapshots"); - setSnapshotDirectory("./snapshots"); + this->setSnapshotDirectory("./snapshots"); + + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); std::optional snapshot_hash; for (size_t i = 0; i < 15; ++i) { - DB::KeeperSnapshotManager manager(3, keeper_context, params.enable_compression); + DB::KeeperSnapshotManager manager(3, this->keeper_context, this->enable_compression); - DB::KeeperStorage storage(500, "", keeper_context); + Storage storage(500, "", this->keeper_context); addNode(storage, "/hello", ""); for (size_t j = 0; j < 5000; ++j) { @@ -2542,7 +2752,7 @@ TEST_P(CoordinationTest, TestStorageSnapshotEqual) for (size_t j = 0; j < 3333; ++j) storage.getSessionID(130 * j); - DB::KeeperStorageSnapshot snapshot(&storage, storage.zxid); + DB::KeeperStorageSnapshot snapshot(&storage, storage.zxid); auto buf = manager.serializeSnapshotToBuffer(snapshot); @@ -2559,17 +2769,16 @@ TEST_P(CoordinationTest, TestStorageSnapshotEqual) } -TEST_P(CoordinationTest, TestLogGap) +TYPED_TEST(CoordinationTest, TestLogGap) { using namespace Coordination; - auto test_params = GetParam(); ChangelogDirTest logs("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); DB::KeeperLogStore changelog( - DB::LogFileSettings{.force_sync = true, .compress_logs = test_params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(0, 3); for (size_t i = 1; i < 55; ++i) @@ -2582,13 +2791,13 @@ TEST_P(CoordinationTest, TestLogGap) } DB::KeeperLogStore changelog1( - DB::LogFileSettings{.force_sync = true, .compress_logs = test_params.enable_compression, .rotate_interval = 100}, + DB::LogFileSettings{.force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog1.init(61, 3); /// Logs discarded - EXPECT_FALSE(fs::exists("./logs/changelog_1_100.bin" + test_params.extension)); + EXPECT_FALSE(fs::exists("./logs/changelog_1_100.bin" + this->extension)); EXPECT_EQ(changelog1.start_index(), 61); EXPECT_EQ(changelog1.next_slot(), 61); } @@ -2600,12 +2809,17 @@ ResponseType getSingleResponse(const auto & responses) return dynamic_cast(*responses[0].response); } -TEST_P(CoordinationTest, TestUncommittedStateBasicCrud) +TYPED_TEST(CoordinationTest, TestUncommittedStateBasicCrud) { using namespace DB; using namespace Coordination; - DB::KeeperStorage storage{500, "", keeper_context}; + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + Storage storage{500, "", this->keeper_context}; constexpr std::string_view path = "/test"; @@ -2717,12 +2931,17 @@ TEST_P(CoordinationTest, TestUncommittedStateBasicCrud) ASSERT_FALSE(get_committed_data()); } -TEST_P(CoordinationTest, TestListRequestTypes) +TYPED_TEST(CoordinationTest, TestListRequestTypes) { using namespace DB; using namespace Coordination; - KeeperStorage storage{500, "", keeper_context}; + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + Storage storage{500, "", this->keeper_context}; int32_t zxid = 0; @@ -2799,18 +3018,18 @@ TEST_P(CoordinationTest, TestListRequestTypes) } } -TEST_P(CoordinationTest, TestDurableState) +TYPED_TEST(CoordinationTest, TestDurableState) { ChangelogDirTest logs("./logs"); - setLogDirectory("./logs"); - setStateFileDirectory("."); + this->setLogDirectory("./logs"); + this->setStateFileDirectory("."); auto state = nuraft::cs_new(); std::optional state_manager; const auto reload_state_manager = [&] { - state_manager.emplace(1, "localhost", 9181, keeper_context); + state_manager.emplace(1, "localhost", 9181, this->keeper_context); state_manager->loadLogStore(1, 0); }; @@ -2873,10 +3092,15 @@ TEST_P(CoordinationTest, TestDurableState) } } -TEST_P(CoordinationTest, TestFeatureFlags) +TYPED_TEST(CoordinationTest, TestFeatureFlags) { using namespace Coordination; - KeeperStorage storage{500, "", keeper_context}; + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + Storage storage{500, "", this->keeper_context}; auto request = std::make_shared(); request->path = DB::keeper_api_feature_flags_path; auto responses = storage.processRequest(request, 0, std::nullopt, true, true); @@ -2888,14 +3112,19 @@ TEST_P(CoordinationTest, TestFeatureFlags) ASSERT_FALSE(feature_flags.isEnabled(KeeperFeatureFlag::CHECK_NOT_EXISTS)); } -TEST_P(CoordinationTest, TestSystemNodeModify) +TYPED_TEST(CoordinationTest, TestSystemNodeModify) { using namespace Coordination; int64_t zxid{0}; + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + // On INIT we abort when a system path is modified - keeper_context->setServerState(KeeperContext::Phase::RUNNING); - KeeperStorage storage{500, "", keeper_context}; + this->keeper_context->setServerState(KeeperContext::Phase::RUNNING); + Storage storage{500, "", this->keeper_context}; const auto assert_create = [&](const std::string_view path, const auto expected_code) { auto request = std::make_shared(); @@ -2920,11 +3149,11 @@ TEST_P(CoordinationTest, TestSystemNodeModify) assert_create("/keeper1/test", Error::ZOK); } -TEST_P(CoordinationTest, ChangelogTestMaxLogSize) +TYPED_TEST(CoordinationTest, ChangelogTestMaxLogSize) { - auto params = GetParam(); + ChangelogDirTest test("./logs"); - setLogDirectory("./logs"); + this->setLogDirectory("./logs"); uint64_t last_entry_index{0}; size_t i{0}; @@ -2932,9 +3161,9 @@ TEST_P(CoordinationTest, ChangelogTestMaxLogSize) SCOPED_TRACE("Small rotation interval, big size limit"); DB::KeeperLogStore changelog( DB::LogFileSettings{ - .force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 20, .max_size = 50 * 1024 * 1024}, + .force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 20, .max_size = 50 * 1024 * 1024}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); for (; i < 100; ++i) @@ -2952,9 +3181,9 @@ TEST_P(CoordinationTest, ChangelogTestMaxLogSize) SCOPED_TRACE("Large rotation interval, small size limit"); DB::KeeperLogStore changelog( DB::LogFileSettings{ - .force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100'000, .max_size = 4000}, + .force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100'000, .max_size = 4000}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); ASSERT_EQ(changelog.entry_at(last_entry_index)->get_term(), (i - 1 + 44) * 10); @@ -2974,20 +3203,25 @@ TEST_P(CoordinationTest, ChangelogTestMaxLogSize) SCOPED_TRACE("Final verify all logs"); DB::KeeperLogStore changelog( DB::LogFileSettings{ - .force_sync = true, .compress_logs = params.enable_compression, .rotate_interval = 100'000, .max_size = 4000}, + .force_sync = true, .compress_logs = this->enable_compression, .rotate_interval = 100'000, .max_size = 4000}, DB::FlushSettings(), - keeper_context); + this->keeper_context); changelog.init(1, 0); ASSERT_EQ(changelog.entry_at(last_entry_index)->get_term(), (i - 1 + 44) * 10); } } -TEST_P(CoordinationTest, TestCheckNotExistsRequest) +TYPED_TEST(CoordinationTest, TestCheckNotExistsRequest) { using namespace DB; using namespace Coordination; - KeeperStorage storage{500, "", keeper_context}; + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + + Storage storage{500, "", this->keeper_context}; int32_t zxid = 0; @@ -3055,18 +3289,23 @@ TEST_P(CoordinationTest, TestCheckNotExistsRequest) } } -TEST_P(CoordinationTest, TestReapplyingDeltas) +TYPED_TEST(CoordinationTest, TestReapplyingDeltas) { using namespace DB; using namespace Coordination; + using Storage = typename TestFixture::Storage; + + ChangelogDirTest rocks("./rocksdb"); + this->setRocksDBDirectory("./rocksdb"); + static constexpr int64_t initial_zxid = 100; const auto create_request = std::make_shared(); create_request->path = "/test/data"; create_request->is_sequential = true; - const auto process_create = [](KeeperStorage & storage, const auto & request, int64_t zxid) + const auto process_create = [](Storage & storage, const auto & request, int64_t zxid) { storage.preprocessRequest(request, 1, 0, zxid); auto responses = storage.processRequest(request, 1, zxid); @@ -3087,19 +3326,19 @@ TEST_P(CoordinationTest, TestReapplyingDeltas) process_create(storage, create_request, zxid); }; - KeeperStorage storage1{500, "", keeper_context}; + Storage storage1{500, "", this->keeper_context}; commit_initial_data(storage1); for (int64_t zxid = initial_zxid + 1; zxid < initial_zxid + 50; ++zxid) storage1.preprocessRequest(create_request, 1, 0, zxid, /*check_acl=*/true, /*digest=*/std::nullopt, /*log_idx=*/zxid); /// create identical new storage - KeeperStorage storage2{500, "", keeper_context}; + Storage storage2{500, "", this->keeper_context}; commit_initial_data(storage2); storage1.applyUncommittedState(storage2, initial_zxid); - const auto commit_unprocessed = [&](KeeperStorage & storage) + const auto commit_unprocessed = [&](Storage & storage) { for (int64_t zxid = initial_zxid + 1; zxid < initial_zxid + 50; ++zxid) { @@ -3112,7 +3351,7 @@ TEST_P(CoordinationTest, TestReapplyingDeltas) commit_unprocessed(storage1); commit_unprocessed(storage2); - const auto get_children = [&](KeeperStorage & storage) + const auto get_children = [&](Storage & storage) { const auto list_request = std::make_shared(); list_request->path = "/test"; @@ -3132,8 +3371,8 @@ TEST_P(CoordinationTest, TestReapplyingDeltas) ASSERT_TRUE(children1_set == children2_set); } -INSTANTIATE_TEST_SUITE_P(CoordinationTestSuite, - CoordinationTest, - ::testing::ValuesIn(std::initializer_list{CompressionParam{true, ".zstd"}, CompressionParam{false, ""}})); +/// INSTANTIATE_TEST_SUITE_P(CoordinationTestSuite, +/// CoordinationTest, +/// ::testing::ValuesIn(std::initializer_list{CompressionParam{true, ".zstd"}, CompressionParam{false, ""}})); #endif diff --git a/src/Core/ExternalTable.cpp b/src/Core/ExternalTable.cpp index bc72c996384..af681cd5639 100644 --- a/src/Core/ExternalTable.cpp +++ b/src/Core/ExternalTable.cpp @@ -16,6 +16,7 @@ #include #include +#include #include #include #include diff --git a/src/Core/LoadBalancing.h b/src/Core/LoadBalancing.h new file mode 100644 index 00000000000..23758895ce2 --- /dev/null +++ b/src/Core/LoadBalancing.h @@ -0,0 +1,27 @@ +#pragma once + +#include + +namespace DB +{ + +enum class LoadBalancing : uint8_t +{ + /// among replicas with a minimum number of errors selected randomly + RANDOM = 0, + /// a replica is selected among the replicas with the minimum number of errors + /// with the minimum number of distinguished characters in the replica name prefix and local hostname prefix + NEAREST_HOSTNAME, + /// just like NEAREST_HOSTNAME, but it count distinguished characters in a levenshtein distance manner + HOSTNAME_LEVENSHTEIN_DISTANCE, + // replicas with the same number of errors are accessed in the same order + // as they are specified in the configuration. + IN_ORDER, + /// if first replica one has higher number of errors, + /// pick a random one from replicas with minimum number of errors + FIRST_OR_RANDOM, + // round robin across replicas with the same number of errors. + ROUND_ROBIN, +}; + +} diff --git a/src/Core/PostgreSQL/PoolWithFailover.cpp b/src/Core/PostgreSQL/PoolWithFailover.cpp index a034c50094d..5014564dbe0 100644 --- a/src/Core/PostgreSQL/PoolWithFailover.cpp +++ b/src/Core/PostgreSQL/PoolWithFailover.cpp @@ -27,7 +27,8 @@ PoolWithFailover::PoolWithFailover( size_t pool_size, size_t pool_wait_timeout_, size_t max_tries_, - bool auto_close_connection_) + bool auto_close_connection_, + size_t connection_attempt_timeout_) : pool_wait_timeout(pool_wait_timeout_) , max_tries(max_tries_) , auto_close_connection(auto_close_connection_) @@ -39,8 +40,13 @@ PoolWithFailover::PoolWithFailover( { for (const auto & replica_configuration : configurations) { - auto connection_info = formatConnectionString(replica_configuration.database, - replica_configuration.host, replica_configuration.port, replica_configuration.username, replica_configuration.password); + auto connection_info = formatConnectionString( + replica_configuration.database, + replica_configuration.host, + replica_configuration.port, + replica_configuration.username, + replica_configuration.password, + connection_attempt_timeout_); replicas_with_priority[priority].emplace_back(connection_info, pool_size); } } @@ -51,7 +57,8 @@ PoolWithFailover::PoolWithFailover( size_t pool_size, size_t pool_wait_timeout_, size_t max_tries_, - bool auto_close_connection_) + bool auto_close_connection_, + size_t connection_attempt_timeout_) : pool_wait_timeout(pool_wait_timeout_) , max_tries(max_tries_) , auto_close_connection(auto_close_connection_) @@ -63,7 +70,13 @@ PoolWithFailover::PoolWithFailover( for (const auto & [host, port] : configuration.addresses) { LOG_DEBUG(getLogger("PostgreSQLPoolWithFailover"), "Adding address host: {}, port: {} to connection pool", host, port); - auto connection_string = formatConnectionString(configuration.database, host, port, configuration.username, configuration.password); + auto connection_string = formatConnectionString( + configuration.database, + host, + port, + configuration.username, + configuration.password, + connection_attempt_timeout_); replicas_with_priority[0].emplace_back(connection_string, pool_size); } } diff --git a/src/Core/PostgreSQL/PoolWithFailover.h b/src/Core/PostgreSQL/PoolWithFailover.h index 3c538fc3dea..502a9a9b7d7 100644 --- a/src/Core/PostgreSQL/PoolWithFailover.h +++ b/src/Core/PostgreSQL/PoolWithFailover.h @@ -14,7 +14,6 @@ static constexpr inline auto POSTGRESQL_POOL_DEFAULT_SIZE = 16; static constexpr inline auto POSTGRESQL_POOL_WAIT_TIMEOUT = 5000; -static constexpr inline auto POSTGRESQL_POOL_WITH_FAILOVER_DEFAULT_MAX_TRIES = 2; namespace postgres { @@ -30,14 +29,16 @@ public: size_t pool_size, size_t pool_wait_timeout, size_t max_tries_, - bool auto_close_connection_); + bool auto_close_connection_, + size_t connection_attempt_timeout_); explicit PoolWithFailover( const DB::StoragePostgreSQL::Configuration & configuration, size_t pool_size, size_t pool_wait_timeout, size_t max_tries_, - bool auto_close_connection_); + bool auto_close_connection_, + size_t connection_attempt_timeout_); PoolWithFailover(const PoolWithFailover & other) = delete; diff --git a/src/Core/PostgreSQL/Utils.cpp b/src/Core/PostgreSQL/Utils.cpp index 810bf62fdab..9dc010c1c69 100644 --- a/src/Core/PostgreSQL/Utils.cpp +++ b/src/Core/PostgreSQL/Utils.cpp @@ -8,7 +8,7 @@ namespace postgres { -ConnectionInfo formatConnectionString(String dbname, String host, UInt16 port, String user, String password) +ConnectionInfo formatConnectionString(String dbname, String host, UInt16 port, String user, String password, UInt64 timeout) { DB::WriteBufferFromOwnString out; out << "dbname=" << DB::quote << dbname @@ -16,7 +16,7 @@ ConnectionInfo formatConnectionString(String dbname, String host, UInt16 port, S << " port=" << port << " user=" << DB::quote << user << " password=" << DB::quote << password - << " connect_timeout=2"; + << " connect_timeout=" << timeout; return {out.str(), host + ':' + DB::toString(port)}; } diff --git a/src/Core/PostgreSQL/Utils.h b/src/Core/PostgreSQL/Utils.h index f179ab14c89..f2b8f1ac084 100644 --- a/src/Core/PostgreSQL/Utils.h +++ b/src/Core/PostgreSQL/Utils.h @@ -18,7 +18,7 @@ namespace pqxx namespace postgres { -ConnectionInfo formatConnectionString(String dbname, String host, UInt16 port, String user, String password); +ConnectionInfo formatConnectionString(String dbname, String host, UInt16 port, String user, String password, UInt64 timeout); String getConnectionForLog(const String & host, UInt16 port); diff --git a/src/Core/Protocol.h b/src/Core/Protocol.h index 4c0848c0706..2e5b91e9b1b 100644 --- a/src/Core/Protocol.h +++ b/src/Core/Protocol.h @@ -63,7 +63,7 @@ const char USER_INTERSERVER_MARKER[] = " INTERSERVER SECRET "; /// Marker for SSH-keys-based authentication (passed as the user name) const char SSH_KEY_AUTHENTICAION_MARKER[] = " SSH KEY AUTHENTICATION "; -/// Market for JSON Web Token authentication +/// Marker for JSON Web Token authentication const char JWT_AUTHENTICAION_MARKER[] = " JWT AUTHENTICATION "; }; diff --git a/src/Core/QueryLogElementType.h b/src/Core/QueryLogElementType.h new file mode 100644 index 00000000000..99361f945cc --- /dev/null +++ b/src/Core/QueryLogElementType.h @@ -0,0 +1,17 @@ +#pragma once + +#include + +namespace DB +{ + +// Make it signed for compatibility with DataTypeEnum8 +enum QueryLogElementType : int8_t +{ + QUERY_START = 1, + QUERY_FINISH = 2, + EXCEPTION_BEFORE_START = 3, + EXCEPTION_WHILE_PROCESSING = 4, +}; + +} diff --git a/src/Core/SchemaInferenceMode.h b/src/Core/SchemaInferenceMode.h new file mode 100644 index 00000000000..17d1d62a60e --- /dev/null +++ b/src/Core/SchemaInferenceMode.h @@ -0,0 +1,14 @@ +#pragma once + +#include + +namespace DB +{ + +enum class SchemaInferenceMode : uint8_t +{ + DEFAULT, + UNION, +}; + +} diff --git a/src/Core/ServerSettings.h b/src/Core/ServerSettings.h index 5e8f8ae1fb4..1a3d68729cf 100644 --- a/src/Core/ServerSettings.h +++ b/src/Core/ServerSettings.h @@ -151,11 +151,13 @@ namespace DB M(UInt64, global_profiler_real_time_period_ns, 0, "Period for real clock timer of global profiler (in nanoseconds). Set 0 value to turn off the real clock global profiler. Recommended value is at least 10000000 (100 times a second) for single queries or 1000000000 (once a second) for cluster-wide profiling.", 0) \ M(UInt64, global_profiler_cpu_time_period_ns, 0, "Period for CPU clock timer of global profiler (in nanoseconds). Set 0 value to turn off the CPU clock global profiler. Recommended value is at least 10000000 (100 times a second) for single queries or 1000000000 (once a second) for cluster-wide profiling.", 0) \ M(Bool, enable_azure_sdk_logging, false, "Enables logging from Azure sdk", 0) \ + M(UInt64, max_entries_for_hash_table_stats, 10'000, "How many entries hash table statistics collected during aggregation is allowed to have", 0) \ M(String, merge_workload, "default", "Name of workload to be used to access resources for all merges (may be overridden by a merge tree setting)", 0) \ M(String, mutation_workload, "default", "Name of workload to be used to access resources for all mutations (may be overridden by a merge tree setting)", 0) \ - M(Double, gwp_asan_force_sample_probability, 0, "Probability that an allocation from specific places will be sampled by GWP Asan (i.e. PODArray allocations)", 0) \ - M(Bool, use_legacy_mongodb_integration, false, "Use the legacy MongoDB integration implementation. Deprecated.", 0) \ + M(Bool, prepare_system_log_tables_on_startup, false, "If true, ClickHouse creates all configured `system.*_log` tables before the startup. It can be helpful if some startup scripts depend on these tables.", 0) \ + M(Double, gwp_asan_force_sample_probability, 0.0003, "Probability that an allocation from specific places will be sampled by GWP Asan (i.e. PODArray allocations)", 0) \ M(UInt64, config_reload_interval_ms, 2000, "How often clickhouse will reload config and check for new changes", 0) \ + M(Bool, use_legacy_mongodb_integration, false, "Use the legacy MongoDB integration implementation. Deprecated.", 0) \ /// If you add a setting which can be updated at runtime, please update 'changeable_settings' map in StorageSystemServerSettings.cpp diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 8257b94cd9f..9c9c9c1db00 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -142,6 +142,7 @@ void Settings::applyCompatibilitySetting(const String & compatibility_value) return; ClickHouseVersion version(compatibility_value); + const auto & settings_changes_history = getSettingsChangesHistory(); /// Iterate through ClickHouse version in descending order and apply reversed /// changes for each version that is higher that version from compatibility setting for (auto it = settings_changes_history.rbegin(); it != settings_changes_history.rend(); ++it) diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 692a5db5da4..a1488eb3a91 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -1,5 +1,8 @@ #pragma once +/// CLion freezes for a minute on every keypress in any file including this. +#if !defined(__CLION_IDE__) + #include #include #include @@ -36,7 +39,7 @@ class IColumn; M(Dialect, dialect, Dialect::clickhouse, "Which dialect will be used to parse query", 0)\ M(UInt64, min_compress_block_size, 65536, "The actual size of the block to compress, if the uncompressed data less than max_compress_block_size is no less than this value and no less than the volume of data for one mark.", 0) \ M(UInt64, max_compress_block_size, 1048576, "The maximum size of blocks of uncompressed data before compressing for writing to a table.", 0) \ - M(UInt64, max_block_size, DEFAULT_BLOCK_SIZE, "Maximum block size for reading", 0) \ + M(UInt64, max_block_size, DEFAULT_BLOCK_SIZE, "Maximum block size in rows for reading", 0) \ M(UInt64, max_insert_block_size, DEFAULT_INSERT_BLOCK_SIZE, "The maximum block size for insertion, if we control the creation of blocks for insertion.", 0) \ M(UInt64, min_insert_block_size_rows, DEFAULT_INSERT_BLOCK_SIZE, "Squash blocks passed to INSERT query to specified size in rows, if blocks are not big enough.", 0) \ M(UInt64, min_insert_block_size_bytes, (DEFAULT_INSERT_BLOCK_SIZE * 256), "Squash blocks passed to INSERT query to specified size in bytes, if blocks are not big enough.", 0) \ @@ -125,6 +128,9 @@ class IColumn; M(Bool, s3_ignore_file_doesnt_exist, false, "Return 0 rows when the requested files don't exist, instead of throwing an exception in S3 table engine", 0) \ M(Bool, hdfs_ignore_file_doesnt_exist, false, "Return 0 rows when the requested files don't exist, instead of throwing an exception in HDFS table engine", 0) \ M(Bool, azure_ignore_file_doesnt_exist, false, "Return 0 rows when the requested files don't exist, instead of throwing an exception in AzureBlobStorage table engine", 0) \ + M(UInt64, azure_sdk_max_retries, 10, "Maximum number of retries in azure sdk", 0) \ + M(UInt64, azure_sdk_retry_initial_backoff_ms, 10, "Minimal backoff between retries in azure sdk", 0) \ + M(UInt64, azure_sdk_retry_max_backoff_ms, 1000, "Maximal backoff between retries in azure sdk", 0) \ M(Bool, s3_validate_request_settings, true, "Validate S3 request settings", 0) \ M(Bool, s3_disable_checksum, S3::DEFAULT_DISABLE_CHECKSUM, "Do not calculate a checksum when sending a file to S3. This speeds up writes by avoiding excessive processing passes on a file. It is mostly safe as the data of MergeTree tables is checksummed by ClickHouse anyway, and when S3 is accessed with HTTPS, the TLS layer already provides integrity while transferring through the network. While additional checksums on S3 give defense in depth.", 0) \ M(UInt64, s3_retry_attempts, S3::DEFAULT_RETRY_ATTEMPTS, "Setting for Aws::Client::RetryStrategy, Aws::Client does retries itself, 0 means no retries", 0) \ @@ -148,7 +154,7 @@ class IColumn; M(UInt64, max_local_write_bandwidth, 0, "The maximum speed of local writes in bytes per second.", 0) \ M(Bool, stream_like_engine_allow_direct_select, false, "Allow direct SELECT query for Kafka, RabbitMQ, FileLog, Redis Streams, and NATS engines. In case there are attached materialized views, SELECT query is not allowed even if this setting is enabled.", 0) \ M(String, stream_like_engine_insert_queue, "", "When stream like engine reads from multiple queues, user will need to select one queue to insert into when writing. Used by Redis Streams and NATS.", 0) \ - \ + M(Bool, dictionary_validate_primary_key_type, false, "Validate primary key type for dictionaries. By default id type for simple layouts will be implicitly converted to UInt64.", 0) \ M(Bool, distributed_insert_skip_read_only_replicas, false, "If true, INSERT into Distributed will skip read-only replicas.", 0) \ M(Bool, distributed_foreground_insert, false, "If setting is enabled, insert query into distributed waits until data are sent to all nodes in a cluster. \n\nEnables or disables synchronous data insertion into a `Distributed` table.\n\nBy default, when inserting data into a Distributed table, the ClickHouse server sends data to cluster nodes in the background. When `distributed_foreground_insert` = 1, the data is processed synchronously, and the `INSERT` operation succeeds only after all the data is saved on all shards (at least one replica for each shard if `internal_replication` is true).", 0) ALIAS(insert_distributed_sync) \ M(UInt64, distributed_background_insert_timeout, 0, "Timeout for insert query into distributed. Setting is used only with insert_distributed_sync enabled. Zero value means no timeout.", 0) ALIAS(insert_distributed_timeout) \ @@ -164,9 +170,6 @@ class IColumn; M(Bool, enable_multiple_prewhere_read_steps, true, "Move more conditions from WHERE to PREWHERE and do reads from disk and filtering in multiple steps if there are multiple conditions combined with AND", 0) \ M(Bool, move_primary_key_columns_to_end_of_prewhere, true, "Move PREWHERE conditions containing primary key columns to the end of AND chain. It is likely that these conditions are taken into account during primary key analysis and thus will not contribute a lot to PREWHERE filtering.", 0) \ \ - M(Bool, allow_statistics_optimize, false, "Allows using statistics to optimize queries", 0) ALIAS(allow_statistic_optimize) \ - M(Bool, allow_experimental_statistics, false, "Allows using statistics", 0) ALIAS(allow_experimental_statistic) \ - \ M(UInt64, alter_sync, 1, "Wait for actions to manipulate the partitions. 0 - do not wait, 1 - wait for execution only of itself, 2 - wait for everyone.", 0) ALIAS(replication_alter_partitions_sync) \ M(Int64, replication_wait_for_inactive_replica_timeout, 120, "Wait for inactive replica to execute ALTER/OPTIMIZE. Time in seconds, 0 - do not wait, negative - wait for unlimited time.", 0) \ M(Bool, alter_move_to_space_execute_async, false, "Execute ALTER TABLE MOVE ... TO [DISK|VOLUME] asynchronously", 0) \ @@ -202,21 +205,6 @@ class IColumn; M(Bool, group_by_use_nulls, false, "Treat columns mentioned in ROLLUP, CUBE or GROUPING SETS as Nullable", 0) \ \ M(NonZeroUInt64, max_parallel_replicas, 1, "The maximum number of replicas of each shard used when the query is executed. For consistency (to get different parts of the same partition), this option only works for the specified sampling key. The lag of the replicas is not controlled. Should be always greater than 0", 0) \ - M(UInt64, parallel_replicas_count, 0, "This is internal setting that should not be used directly and represents an implementation detail of the 'parallel replicas' mode. This setting will be automatically set up by the initiator server for distributed queries to the number of parallel replicas participating in query processing.", 0) \ - M(UInt64, parallel_replica_offset, 0, "This is internal setting that should not be used directly and represents an implementation detail of the 'parallel replicas' mode. This setting will be automatically set up by the initiator server for distributed queries to the index of the replica participating in query processing among parallel replicas.", 0) \ - M(String, parallel_replicas_custom_key, "", "Custom key assigning work to replicas when parallel replicas are used.", 0) \ - M(ParallelReplicasCustomKeyFilterType, parallel_replicas_custom_key_filter_type, ParallelReplicasCustomKeyFilterType::DEFAULT, "Type of filter to use with custom key for parallel replicas. default - use modulo operation on the custom key, range - use range filter on custom key using all possible values for the value type of custom key.", 0) \ - M(UInt64, parallel_replicas_custom_key_range_lower, 0, "Lower bound for the universe that the parallel replicas custom range filter is calculated over", 0) \ - M(UInt64, parallel_replicas_custom_key_range_upper, 0, "Upper bound for the universe that the parallel replicas custom range filter is calculated over. A value of 0 disables the upper bound, setting it to the max value of the custom key expression", 0) \ - \ - M(String, cluster_for_parallel_replicas, "", "Cluster for a shard in which current server is located", 0) \ - M(UInt64, allow_experimental_parallel_reading_from_replicas, 0, "Use all the replicas from a shard for SELECT query execution. Reading is parallelized and coordinated dynamically. 0 - disabled, 1 - enabled, silently disable them in case of failure, 2 - enabled, throw an exception in case of failure", 0) \ - M(Bool, parallel_replicas_allow_in_with_subquery, true, "If true, subquery for IN will be executed on every follower replica.", 0) \ - M(Float, parallel_replicas_single_task_marks_count_multiplier, 2, "A multiplier which will be added during calculation for minimal number of marks to retrieve from coordinator. This will be applied only for remote replicas.", 0) \ - M(Bool, parallel_replicas_for_non_replicated_merge_tree, false, "If true, ClickHouse will use parallel replicas algorithm also for non-replicated MergeTree tables", 0) \ - M(UInt64, parallel_replicas_min_number_of_rows_per_replica, 0, "Limit the number of replicas used in a query to (estimated rows to read / min_number_of_rows_per_replica). The max is still limited by 'max_parallel_replicas'", 0) \ - M(Bool, parallel_replicas_prefer_local_join, true, "If true, and JOIN can be executed with parallel replicas algorithm, and all storages of right JOIN part are *MergeTree, local JOIN will be used instead of GLOBAL JOIN.", 0) \ - M(UInt64, parallel_replicas_mark_segment_size, 128, "Parts virtually divided into segments to be distributed between replicas for parallel reading. This setting controls the size of these segments. Not recommended to change until you're absolutely sure in what you're doing", 0) \ \ M(Bool, skip_unavailable_shards, false, "If true, ClickHouse silently skips unavailable shards. Shard is marked as unavailable when: 1) The shard cannot be reached due to a connection failure. 2) Shard is unresolvable through DNS. 3) Table does not exist on the shard.", 0) \ \ @@ -248,8 +236,6 @@ class IColumn; M(Bool, do_not_merge_across_partitions_select_final, false, "Merge parts only in one partition in select final", 0) \ M(Bool, split_parts_ranges_into_intersecting_and_non_intersecting_final, true, "Split parts ranges into intersecting and non intersecting during FINAL optimization", 0) \ M(Bool, split_intersecting_parts_ranges_into_layers_final, true, "Split intersecting parts ranges into layers during FINAL optimization", 0) \ - M(Bool, allow_experimental_inverted_index, false, "If it is set to true, allow to use experimental inverted index.", 0) \ - M(Bool, allow_experimental_full_text_index, false, "If it is set to true, allow to use experimental full-text index.", 0) \ \ M(UInt64, mysql_max_rows_to_insert, 65536, "The maximum number of rows in MySQL batch insertion of the MySQL storage engine", 0) \ M(Bool, mysql_map_string_to_text_in_show_columns, true, "If enabled, String type will be mapped to TEXT in SHOW [FULL] COLUMNS, BLOB otherwise. Has an effect only when the connection is made through the MySQL wire protocol.", 0) \ @@ -338,7 +324,6 @@ class IColumn; M(Bool, fsync_metadata, true, "Do fsync after changing metadata for tables and databases (.sql files). Could be disabled in case of poor latency on server with high load of DDL queries and high load of disk subsystem.", 0) \ \ M(Bool, join_use_nulls, false, "Use NULLs for non-joined rows of outer JOINs for types that can be inside Nullable. If false, use default value of corresponding columns data type.", IMPORTANT) \ - M(Bool, allow_experimental_join_condition, false, "Support join with inequal conditions which involve columns from both left and right table. e.g. t1.y < t2.y.", 0) \ \ M(JoinStrictness, join_default_strictness, JoinStrictness::All, "Set default strictness in JOIN query. Possible values: empty string, 'ANY', 'ALL'. If empty, query without strictness will throw exception.", 0) \ M(Bool, any_join_distinct_right_table_keys, false, "Enable old ANY JOIN logic with many-to-one left-to-right table keys mapping for all ANY JOINs. It leads to confusing not equal results for 't1 ANY LEFT JOIN t2' and 't2 ANY RIGHT JOIN t1'. ANY RIGHT JOIN needs one-to-many keys mapping to be consistent with LEFT one.", IMPORTANT) \ @@ -364,6 +349,7 @@ class IColumn; \ M(Bool, ignore_on_cluster_for_replicated_udf_queries, false, "Ignore ON CLUSTER clause for replicated UDF management queries.", 0) \ M(Bool, ignore_on_cluster_for_replicated_access_entities_queries, false, "Ignore ON CLUSTER clause for replicated access entities management queries.", 0) \ + M(Bool, ignore_on_cluster_for_replicated_named_collections_queries, false, "Ignore ON CLUSTER clause for replicated named collections management queries.", 0) \ /** Settings for testing hedged requests */ \ M(Milliseconds, sleep_in_send_tables_status_ms, 0, "Time to sleep in sending tables status response in TCPHandler", 0) \ M(Milliseconds, sleep_in_send_data_ms, 0, "Time to sleep in sending data in TCPHandler", 0) \ @@ -378,7 +364,7 @@ class IColumn; M(UInt64, http_max_fields, 1000000, "Maximum number of fields in HTTP header", 0) \ M(UInt64, http_max_field_name_size, 128 * 1024, "Maximum length of field name in HTTP header", 0) \ M(UInt64, http_max_field_value_size, 128 * 1024, "Maximum length of field value in HTTP header", 0) \ - M(Bool, http_skip_not_found_url_for_globs, true, "Skip url's for globs with HTTP_NOT_FOUND error", 0) \ + M(Bool, http_skip_not_found_url_for_globs, true, "Skip URLs for globs with HTTP_NOT_FOUND error", 0) \ M(Bool, http_make_head_request, true, "Allows the execution of a `HEAD` request while reading data from HTTP to retrieve information about the file to be read, such as its size", 0) \ M(Bool, optimize_throw_if_noop, false, "If setting is enabled and OPTIMIZE query didn't actually assign a merge then an explanatory exception is thrown", 0) \ M(Bool, use_index_for_in_with_subqueries, true, "Try using an index if there is a subquery or a table expression on the right side of the IN operator.", 0) \ @@ -389,7 +375,6 @@ class IColumn; M(Bool, empty_result_for_aggregation_by_constant_keys_on_empty_set, true, "Return empty result when aggregating by constant keys on empty set.", 0) \ M(Bool, allow_distributed_ddl, true, "If it is set to true, then a user is allowed to executed distributed DDL queries.", 0) \ M(Bool, allow_suspicious_codecs, false, "If it is set to true, allow to specify meaningless compression codecs.", 0) \ - M(Bool, allow_experimental_codecs, false, "If it is set to true, allow to specify experimental compression codecs (but we don't have those yet and this option does nothing).", 0) \ M(Bool, enable_deflate_qpl_codec, false, "Enable/disable the DEFLATE_QPL codec.", 0) \ M(Bool, enable_zstd_qat_codec, false, "Enable/disable the ZSTD_QAT codec.", 0) \ M(UInt64, query_profiler_real_time_period_ns, QUERY_PROFILER_DEFAULT_SAMPLE_RATE_NS, "Period for real clock timer of query profiler (in nanoseconds). Set 0 value to turn off the real clock query profiler. Recommended value is at least 10000000 (100 times a second) for single queries or 1000000000 (once a second) for cluster-wide profiling.", 0) \ @@ -399,8 +384,7 @@ class IColumn; M(Float, opentelemetry_start_trace_probability, 0., "Probability to start an OpenTelemetry trace for an incoming query.", 0) \ M(Bool, opentelemetry_trace_processors, false, "Collect OpenTelemetry spans for processors.", 0) \ M(Bool, prefer_column_name_to_alias, false, "Prefer using column names instead of aliases if possible.", 0) \ - M(Bool, allow_experimental_analyzer, true, "Allow experimental analyzer.", 0) \ - M(Bool, analyzer_compatibility_join_using_top_level_identifier, false, "Force to resolve identifier in JOIN USING from projection (for example, in `SELECT a + 1 AS b FROM t1 JOIN t2 USING (b)` join will be performed by `t1.a + 1 = t2.b`, rather then `t1.b = t2.b`).", 0) \ + \ M(Bool, prefer_global_in_and_join, false, "If enabled, all IN/JOIN operators will be rewritten as GLOBAL IN/JOIN. It's useful when the to-be-joined tables are only available on the initiator and we need to always scatter their data on-the-fly during distributed processing with the GLOBAL keyword. It's also useful to reduce the need to access the external sources joining external tables.", 0) \ M(Bool, enable_vertical_final, true, "If enable, remove duplicated rows during FINAL by marking rows as deleted and filtering them later instead of merging rows", 0) \ \ @@ -470,7 +454,7 @@ class IColumn; M(UInt64, max_rows_in_join, 0, "Maximum size of the hash table for JOIN (in number of rows).", 0) \ M(UInt64, max_bytes_in_join, 0, "Maximum size of the hash table for JOIN (in number of bytes in memory).", 0) \ M(OverflowMode, join_overflow_mode, OverflowMode::THROW, "What to do when the limit is exceeded.", 0) \ - M(Bool, join_any_take_last_row, false, "When disabled (default) ANY JOIN will take the first found row for a key. When enabled, it will take the last row seen if there are multiple rows for the same key.", IMPORTANT) \ + M(Bool, join_any_take_last_row, false, "When disabled (default) ANY JOIN will take the first found row for a key. When enabled, it will take the last row seen if there are multiple rows for the same key. Can be applied only to hash join and storage join.", IMPORTANT) \ M(JoinAlgorithm, join_algorithm, JoinAlgorithm::DEFAULT, "Specify join algorithm.", 0) \ M(UInt64, cross_join_min_rows_to_compress, 10000000, "Minimal count of rows to compress block in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached.", 0) \ M(UInt64, cross_join_min_bytes_to_compress, 1_GiB, "Minimal size of block to compress in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached.", 0) \ @@ -521,6 +505,7 @@ class IColumn; M(UInt64, backup_restore_keeper_value_max_size, 1048576, "Maximum size of data of a [Zoo]Keeper's node during backup", 0) \ M(UInt64, backup_restore_batch_size_for_keeper_multiread, 10000, "Maximum size of batch for multiread request to [Zoo]Keeper during backup or restore", 0) \ M(UInt64, backup_restore_batch_size_for_keeper_multi, 1000, "Maximum size of batch for multi request to [Zoo]Keeper during backup or restore", 0) \ + M(UInt64, backup_restore_s3_retry_attempts, 1000, "Setting for Aws::Client::RetryStrategy, Aws::Client does retries itself, 0 means no retries. It takes place only for backup/restore.", 0) \ M(UInt64, max_backup_bandwidth, 0, "The maximum read speed in bytes per second for particular backup on server. Zero means unlimited.", 0) \ \ M(Bool, log_profile_events, true, "Log query performance statistics into the query_log, query_thread_log and query_views_log.", 0) \ @@ -550,6 +535,7 @@ class IColumn; M(Bool, optimize_read_in_order, true, "Enable ORDER BY optimization for reading data in corresponding order in MergeTree tables.", 0) \ M(Bool, optimize_read_in_window_order, true, "Enable ORDER BY optimization in window clause for reading data in corresponding order in MergeTree tables.", 0) \ M(Bool, optimize_aggregation_in_order, false, "Enable GROUP BY optimization for aggregating data in corresponding order in MergeTree tables.", 0) \ + M(Bool, read_in_order_use_buffering, true, "Use buffering before merging while reading in order of primary key. It increases the parallelism of query execution", 0) \ M(UInt64, aggregation_in_order_max_block_bytes, 50000000, "Maximal size of block in bytes accumulated during aggregation in order of primary key. Lower block size allows to parallelize more final merge stage of aggregation.", 0) \ M(UInt64, read_in_order_two_level_merge_threshold, 100, "Minimal number of parts to read to run preliminary merge step during multithread reading in order of primary key.", 0) \ M(Bool, low_cardinality_allow_in_native_format, true, "Use LowCardinality type in Native format. Otherwise, convert LowCardinality columns to ordinary for select query, and convert ordinary columns to required LowCardinality for insert query.", 0) \ @@ -580,7 +566,9 @@ class IColumn; M(UInt64, max_partition_size_to_drop, 50000000000lu, "Same as max_table_size_to_drop, but for the partitions.", 0) \ \ M(UInt64, postgresql_connection_pool_size, 16, "Connection pool size for PostgreSQL table engine and database engine.", 0) \ + M(UInt64, postgresql_connection_attempt_timeout, 2, "Connection timeout to PostgreSQL table engine and database engine in seconds.", 0) \ M(UInt64, postgresql_connection_pool_wait_timeout, 5000, "Connection pool push/pop timeout on empty pool for PostgreSQL table engine and database engine. By default it will block on empty pool.", 0) \ + M(UInt64, postgresql_connection_pool_retries, 2, "Connection pool push/pop retries number for PostgreSQL table engine and database engine.", 0) \ M(Bool, postgresql_connection_pool_auto_close_connection, false, "Close connection before returning connection to the pool.", 0) \ M(UInt64, glob_expansion_max_elements, 1000, "Maximum number of allowed addresses (For external storages, table functions, etc).", 0) \ M(UInt64, odbc_bridge_connection_pool_size, 16, "Connection pool size for each connection settings string in ODBC bridge.", 0) \ @@ -590,13 +578,6 @@ class IColumn; M(UInt64, distributed_replica_error_cap, DBMS_CONNECTION_POOL_WITH_FAILOVER_MAX_ERROR_COUNT, "Max number of errors per replica, prevents piling up an incredible amount of errors if replica was offline for some time and allows it to be reconsidered in a shorter amount of time.", 0) \ M(UInt64, distributed_replica_max_ignored_errors, 0, "Number of errors that will be ignored while choosing replicas", 0) \ \ - M(Bool, allow_experimental_live_view, false, "Enable LIVE VIEW. Not mature enough.", 0) \ - M(Seconds, live_view_heartbeat_interval, 15, "The heartbeat interval in seconds to indicate live query is alive.", 0) \ - M(UInt64, max_live_view_insert_blocks_before_refresh, 64, "Limit maximum number of inserted blocks after which mergeable blocks are dropped and query is re-executed.", 0) \ - M(Bool, allow_experimental_window_view, false, "Enable WINDOW VIEW. Not mature enough.", 0) \ - M(Seconds, window_view_clean_interval, 60, "The clean interval of window view in seconds to free outdated data.", 0) \ - M(Seconds, window_view_heartbeat_interval, 15, "The heartbeat interval in seconds to indicate watch query is alive.", 0) \ - M(Seconds, wait_for_window_view_fire_signal_timeout, 10, "Timeout for waiting for window view fire signal in event time processing", 0) \ M(UInt64, min_free_disk_space_for_temporary_data, 0, "The minimum disk space to keep while writing temporary data used in external sorting and aggregation.", 0) \ \ M(DefaultTableEngine, default_temporary_table_engine, DefaultTableEngine::Memory, "Default table engine used when ENGINE is not set in CREATE TEMPORARY statement.",0) \ @@ -612,6 +593,7 @@ class IColumn; M(UInt64, mutations_sync, 0, "Wait for synchronous execution of ALTER TABLE UPDATE/DELETE queries (mutations). 0 - execute asynchronously. 1 - wait current server. 2 - wait all replicas if they exist.", 0) \ M(Bool, enable_lightweight_delete, true, "Enable lightweight DELETE mutations for mergetree tables.", 0) ALIAS(allow_experimental_lightweight_delete) \ M(UInt64, lightweight_deletes_sync, 2, "The same as 'mutation_sync', but controls only execution of lightweight deletes", 0) \ + M(LightweightMutationProjectionMode, lightweight_mutation_projection_mode, LightweightMutationProjectionMode::THROW, "When lightweight delete happens on a table with projection(s), the possible operations include throw the exception as projection exists, or drop all projection related to this table then do lightweight delete.", 0) \ M(Bool, apply_deleted_mask, true, "Enables filtering out rows deleted with lightweight DELETE. If disabled, a query will be able to read those rows. This is useful for debugging and \"undelete\" scenarios", 0) \ M(Bool, optimize_normalize_count_variants, true, "Rewrite aggregate functions that semantically equals to count() as count().", 0) \ M(Bool, optimize_injective_functions_inside_uniq, true, "Delete injective functions of one argument inside uniq*() functions.", 0) \ @@ -630,13 +612,10 @@ class IColumn; M(Bool, optimize_time_filter_with_preimage, true, "Optimize Date and DateTime predicates by converting functions into equivalent comparisons without conversions (e.g. toYear(col) = 2023 -> col >= '2023-01-01' AND col <= '2023-12-31')", 0) \ M(Bool, normalize_function_names, true, "Normalize function names to their canonical names", 0) \ M(Bool, enable_early_constant_folding, true, "Enable query optimization where we analyze function and subqueries results and rewrite query if there are constants there", 0) \ - M(Bool, deduplicate_blocks_in_dependent_materialized_views, false, "Should deduplicate blocks for materialized views if the block is not a duplicate for the table. Use true to always deduplicate in dependent tables.", 0) \ + M(Bool, deduplicate_blocks_in_dependent_materialized_views, false, "Should deduplicate blocks for materialized views. Use true to always deduplicate in dependent tables.", 0) \ M(Bool, throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert, true, "Throw exception on INSERT query when the setting `deduplicate_blocks_in_dependent_materialized_views` is enabled along with `async_insert`. It guarantees correctness, because these features can't work together.", 0) \ - M(Bool, update_insert_deduplication_token_in_dependent_materialized_views, false, "Should update insert deduplication token with table identifier during insert in dependent materialized views.", 0) \ M(Bool, materialized_views_ignore_errors, false, "Allows to ignore errors for MATERIALIZED VIEW, and deliver original block to the table regardless of MVs", 0) \ M(Bool, ignore_materialized_views_with_dropped_target_table, false, "Ignore MVs with dropped target table during pushing to views", 0) \ - M(Bool, allow_experimental_refreshable_materialized_view, false, "Allow refreshable materialized views (CREATE MATERIALIZED VIEW REFRESH ...).", 0) \ - M(Bool, stop_refreshable_materialized_views_on_startup, false, "On server startup, prevent scheduling of refreshable materialized views, as if with SYSTEM STOP VIEWS. You can manually start them with SYSTEM START VIEWS or SYSTEM START VIEW afterwards. Also applies to newly created views. Has no effect on non-refreshable materialized views.", 0) \ M(Bool, use_compact_format_in_distributed_parts_names, true, "Changes format of directories names for distributed table insert parts.", 0) \ M(Bool, validate_polygons, true, "Throw exception if polygon is invalid in function pointInPolygon (e.g. self-tangent, self-intersecting). If the setting is false, the function will accept invalid polygons but may silently return wrong result.", 0) \ M(UInt64, max_parser_depth, DBMS_DEFAULT_MAX_PARSER_DEPTH, "Maximum parser depth (recursion depth of recursive descend parser).", 0) \ @@ -653,11 +632,9 @@ class IColumn; M(Bool, cast_keep_nullable, false, "CAST operator keep Nullable for result data type", 0) \ M(Bool, cast_ipv4_ipv6_default_on_conversion_error, false, "CAST operator into IPv4, CAST operator into IPV6 type, toIPv4, toIPv6 functions will return default value instead of throwing exception on conversion error.", 0) \ M(Bool, alter_partition_verbose_result, false, "Output information about affected parts. Currently works only for FREEZE and ATTACH commands.", 0) \ - M(Bool, allow_experimental_database_materialized_mysql, false, "Allow to create database with Engine=MaterializedMySQL(...).", 0) \ - M(Bool, allow_experimental_database_materialized_postgresql, false, "Allow to create database with Engine=MaterializedPostgreSQL(...).", 0) \ M(Bool, system_events_show_zero_values, false, "When querying system.events or system.metrics tables, include all metrics, even with zero values.", 0) \ M(MySQLDataTypesSupport, mysql_datatypes_support_level, MySQLDataTypesSupportList{}, "Defines how MySQL types are converted to corresponding ClickHouse types. A comma separated list in any combination of 'decimal', 'datetime64', 'date2Date32' or 'date2String'. decimal: convert NUMERIC and DECIMAL types to Decimal when precision allows it. datetime64: convert DATETIME and TIMESTAMP types to DateTime64 instead of DateTime when precision is not 0. date2Date32: convert DATE to Date32 instead of Date. Takes precedence over date2String. date2String: convert DATE to String instead of Date. Overridden by datetime64.", 0) \ - M(Bool, optimize_trivial_insert_select, true, "Optimize trivial 'INSERT INTO table SELECT ... FROM TABLES' query", 0) \ + M(Bool, optimize_trivial_insert_select, false, "Optimize trivial 'INSERT INTO table SELECT ... FROM TABLES' query", 0) \ M(Bool, allow_non_metadata_alters, true, "Allow to execute alters which affects not only tables metadata, but also data on disk", 0) \ M(Bool, enable_global_with_statement, true, "Propagate WITH statements to UNION queries and all subqueries", 0) \ M(Bool, aggregate_functions_null_for_empty, false, "Rewrite all aggregate functions in a query, adding -OrNull suffix to them", 0) \ @@ -707,31 +684,31 @@ class IColumn; M(UInt64, insert_shard_id, 0, "If non zero, when insert into a distributed table, the data will be inserted into the shard `insert_shard_id` synchronously. Possible values range from 1 to `shards_number` of corresponding distributed table", 0) \ \ M(Bool, collect_hash_table_stats_during_aggregation, true, "Enable collecting hash table statistics to optimize memory allocation", 0) \ - M(UInt64, max_entries_for_hash_table_stats, 10'000, "How many entries hash table statistics collected during aggregation is allowed to have", 0) \ M(UInt64, max_size_to_preallocate_for_aggregation, 100'000'000, "For how many elements it is allowed to preallocate space in all hash tables in total before aggregation", 0) \ \ + M(Bool, collect_hash_table_stats_during_joins, true, "Enable collecting hash table statistics to optimize memory allocation", 0) \ + M(UInt64, max_size_to_preallocate_for_joins, 100'000'000, "For how many elements it is allowed to preallocate space in all hash tables in total before join", 0) \ + \ M(Bool, kafka_disable_num_consumers_limit, false, "Disable limit on kafka_num_consumers that depends on the number of available CPU cores", 0) \ M(Bool, enable_software_prefetch_in_aggregation, true, "Enable use of software prefetch in aggregation", 0) \ M(Bool, allow_aggregate_partitions_independently, false, "Enable independent aggregation of partitions on separate threads when partition key suits group by key. Beneficial when number of partitions close to number of cores and partitions have roughly the same size", 0) \ M(Bool, force_aggregate_partitions_independently, false, "Force the use of optimization when it is applicable, but heuristics decided not to use it", 0) \ M(UInt64, max_number_of_partitions_for_independent_aggregation, 128, "Maximal number of partitions in table to apply optimization", 0) \ M(Float, min_hit_rate_to_use_consecutive_keys_optimization, 0.5, "Minimal hit rate of a cache which is used for consecutive keys optimization in aggregation to keep it enabled", 0) \ - /** Experimental feature for moving data between shards. */ \ - \ - M(Bool, allow_experimental_query_deduplication, false, "Experimental data deduplication for SELECT queries based on part UUIDs", 0) \ \ M(Bool, engine_file_empty_if_not_exists, false, "Allows to select data from a file engine table without file", 0) \ M(Bool, engine_file_truncate_on_insert, false, "Enables or disables truncate before insert in file engine tables", 0) \ M(Bool, engine_file_allow_create_multiple_files, false, "Enables or disables creating a new file on each insert in file engine tables if format has suffix.", 0) \ M(Bool, engine_file_skip_empty_files, false, "Allows to skip empty files in file table engine", 0) \ - M(Bool, engine_url_skip_empty_files, false, "Allows to skip empty files in url table engine", 0) \ - M(Bool, enable_url_encoding, true, " Allows to enable/disable decoding/encoding path in uri in URL table engine", 0) \ + M(Bool, engine_url_skip_empty_files, false, "Allows to skip empty files in the URL table engine", 0) \ + M(Bool, enable_url_encoding, true, " Allows to enable/disable decoding/encoding path in URI in the URL table engine", 0) \ M(UInt64, database_replicated_initial_query_timeout_sec, 300, "How long initial DDL query should wait for Replicated database to precess previous DDL queue entries", 0) \ M(Bool, database_replicated_enforce_synchronous_settings, false, "Enforces synchronous waiting for some queries (see also database_atomic_wait_for_drop_and_detach_synchronously, mutation_sync, alter_sync). Not recommended to enable these settings.", 0) \ M(UInt64, max_distributed_depth, 5, "Maximum distributed query depth", 0) \ M(Bool, database_replicated_always_detach_permanently, false, "Execute DETACH TABLE as DETACH TABLE PERMANENTLY if database engine is Replicated", 0) \ M(Bool, database_replicated_allow_only_replicated_engine, false, "Allow to create only Replicated tables in database with engine Replicated", 0) \ M(Bool, database_replicated_allow_replicated_engine_arguments, true, "Allow to create only Replicated tables in database with engine Replicated with explicit arguments", 0) \ + M(Bool, database_replicated_allow_heavy_create, false, "Allow long-running DDL queries (CREATE AS SELECT and POPULATE) in Replicated database engine. Note that it can block DDL queue for a long time.", 0) \ M(Bool, cloud_mode, false, "Only available in ClickHouse Cloud", 0) \ M(UInt64, cloud_mode_engine, 1, "Only available in ClickHouse Cloud", 0) \ M(DistributedDDLOutputMode, distributed_ddl_output_mode, DistributedDDLOutputMode::THROW, "Format of distributed DDL query result, one of: 'none', 'throw', 'null_status_on_timeout', 'never_throw', 'none_only_active', 'throw_only_active', 'null_status_on_timeout_only_active'", 0) \ @@ -750,6 +727,7 @@ class IColumn; M(Bool, optimize_group_by_function_keys, true, "Eliminates functions of other keys in GROUP BY section", 0) \ M(Bool, optimize_group_by_constant_keys, true, "Optimize GROUP BY when all keys in block are constant", 0) \ M(Bool, legacy_column_name_of_tuple_literal, false, "List all names of element of large tuple literals in their column names instead of hash. This settings exists only for compatibility reasons. It makes sense to set to 'true', while doing rolling update of cluster from version lower than 21.7 to higher.", 0) \ + M(Bool, enable_named_columns_in_function_tuple, true, "Generate named tuples in function tuple() when all names are unique and can be treated as unquoted identifiers.", 0) \ \ M(Bool, query_plan_enable_optimizations, true, "Globally enable/disable query optimization at the query plan level", 0) \ M(UInt64, query_plan_max_optimizations_to_apply, 10000, "Limit the total number of optimizations applied to query plan. If zero, ignored. If limit reached, throw exception", 0) \ @@ -757,6 +735,7 @@ class IColumn; M(Bool, query_plan_push_down_limit, true, "Allow to move LIMITs down in the query plan", 0) \ M(Bool, query_plan_split_filter, true, "Allow to split filters in the query plan", 0) \ M(Bool, query_plan_merge_expressions, true, "Allow to merge expressions in the query plan", 0) \ + M(Bool, query_plan_merge_filters, false, "Allow to merge filters in the query plan", 0) \ M(Bool, query_plan_filter_push_down, true, "Allow to push down filter by predicate query plan step", 0) \ M(Bool, query_plan_convert_outer_join_to_inner_join, true, "Allow to convert OUTER JOIN to INNER JOIN if filter after JOIN always filters default values", 0) \ M(Bool, query_plan_optimize_prewhere, true, "Allow to push down filter to PREWHERE expression for supported storages", 0) \ @@ -787,7 +766,7 @@ class IColumn; M(UInt64, merge_tree_min_rows_for_concurrent_read_for_remote_filesystem, (20 * 8192), "If at least as many lines are read from one file, the reading can be parallelized, when reading from remote filesystem.", 0) \ M(UInt64, merge_tree_min_bytes_for_concurrent_read_for_remote_filesystem, (24 * 10 * 1024 * 1024), "If at least as many bytes are read from one file, the reading can be parallelized, when reading from remote filesystem.", 0) \ M(UInt64, remote_read_min_bytes_for_seek, 4 * DBMS_DEFAULT_BUFFER_SIZE, "Min bytes required for remote read (url, s3) to do seek, instead of read with ignore.", 0) \ - M(UInt64, merge_tree_min_bytes_per_task_for_remote_reading, 4 * DBMS_DEFAULT_BUFFER_SIZE, "Min bytes to read per task.", 0) \ + M(UInt64, merge_tree_min_bytes_per_task_for_remote_reading, 2 * DBMS_DEFAULT_BUFFER_SIZE, "Min bytes to read per task.", 0) ALIAS(filesystem_prefetch_min_bytes_for_single_read_task) \ M(Bool, merge_tree_use_const_size_tasks_for_remote_reading, true, "Whether to use constant size tasks for reading from a remote table.", 0) \ M(Bool, merge_tree_determine_task_size_by_prewhere_columns, true, "Whether to use only prewhere columns size to determine reading task size.", 0) \ M(UInt64, merge_tree_compact_parts_min_granules_to_multibuffer_read, 16, "Only available in ClickHouse Cloud", 0) \ @@ -829,7 +808,6 @@ class IColumn; M(UInt64, prefetch_buffer_size, DBMS_DEFAULT_BUFFER_SIZE, "The maximum size of the prefetch buffer to read from the filesystem.", 0) \ M(UInt64, filesystem_prefetch_step_bytes, 0, "Prefetch step in bytes. Zero means `auto` - approximately the best prefetch step will be auto deduced, but might not be 100% the best. The actual value might be different because of setting filesystem_prefetch_min_bytes_for_single_read_task", 0) \ M(UInt64, filesystem_prefetch_step_marks, 0, "Prefetch step in marks. Zero means `auto` - approximately the best prefetch step will be auto deduced, but might not be 100% the best. The actual value might be different because of setting filesystem_prefetch_min_bytes_for_single_read_task", 0) \ - M(UInt64, filesystem_prefetch_min_bytes_for_single_read_task, "2Mi", "Do not parallelize within one file read less than this amount of bytes. E.g. one reader will not receive a read task of size less than this amount. This setting is recommended to avoid spikes of time for aws getObject requests to aws", 0) \ M(UInt64, filesystem_prefetch_max_memory_usage, "1Gi", "Maximum memory usage for prefetches.", 0) \ M(UInt64, filesystem_prefetches_limit, 200, "Maximum number of prefetches. Zero means unlimited. A setting `filesystem_prefetches_max_memory_usage` is more recommended if you want to limit the number of prefetches", 0) \ \ @@ -857,7 +835,7 @@ class IColumn; M(Bool, schema_inference_use_cache_for_azure, true, "Use cache in schema inference while using azure table function", 0) \ M(Bool, schema_inference_use_cache_for_hdfs, true, "Use cache in schema inference while using hdfs table function", 0) \ M(Bool, schema_inference_use_cache_for_url, true, "Use cache in schema inference while using url table function", 0) \ - M(Bool, schema_inference_cache_require_modification_time_for_url, true, "Use schema from cache for URL with last modification time validation (for urls with Last-Modified header)", 0) \ + M(Bool, schema_inference_cache_require_modification_time_for_url, true, "Use schema from cache for URL with last modification time validation (for URLs with Last-Modified header)", 0) \ \ M(String, compatibility, "", "Changes other settings according to provided ClickHouse version. If we know that we changed some behaviour in ClickHouse by changing some settings in some version, this compatibility setting will control these settings", 0) \ \ @@ -898,8 +876,29 @@ class IColumn; M(Bool, allow_get_client_http_header, false, "Allow to use the function `getClientHTTPHeader` which lets to obtain a value of an the current HTTP request's header. It is not enabled by default for security reasons, because some headers, such as `Cookie`, could contain sensitive info. Note that the `X-ClickHouse-*` and `Authentication` headers are always restricted and cannot be obtained with this function.", 0) \ M(Bool, cast_string_to_dynamic_use_inference, false, "Use types inference during String to Dynamic conversion", 0) \ M(Bool, enable_blob_storage_log, true, "Write information about blob storage operations to system.blob_storage_log table", 0) \ + M(Bool, allow_create_index_without_type, false, "Allow CREATE INDEX query without TYPE. Query will be ignored. Made for SQL compatibility tests.", 0) \ + M(Bool, create_index_ignore_unique, false, "Ignore UNIQUE keyword in CREATE UNIQUE INDEX. Made for SQL compatibility tests.", 0) \ + M(Bool, print_pretty_type_names, true, "Print pretty type names in DESCRIBE query and toTypeName() function", 0) \ + M(Bool, create_table_empty_primary_key_by_default, false, "Allow to create *MergeTree tables with empty primary key when ORDER BY and PRIMARY KEY not specified", 0) \ + M(Bool, allow_named_collection_override_by_default, true, "Allow named collections' fields override by default.", 0) \ + M(SQLSecurityType, default_normal_view_sql_security, SQLSecurityType::INVOKER, "Allows to set a default value for SQL SECURITY option when creating a normal view.", 0) \ + M(SQLSecurityType, default_materialized_view_sql_security, SQLSecurityType::DEFINER, "Allows to set a default value for SQL SECURITY option when creating a materialized view.", 0) \ + M(String, default_view_definer, "CURRENT_USER", "Allows to set a default value for DEFINER option when creating view.", 0) \ + M(UInt64, cache_warmer_threads, 4, "Only available in ClickHouse Cloud. Number of background threads for speculatively downloading new data parts into file cache, when cache_populated_by_fetch is enabled. Zero to disable.", 0) \ + M(Int64, ignore_cold_parts_seconds, 0, "Only available in ClickHouse Cloud. Exclude new data parts from SELECT queries until they're either pre-warmed (see cache_populated_by_fetch) or this many seconds old. Only for Replicated-/SharedMergeTree.", 0) \ + M(Int64, prefer_warmed_unmerged_parts_seconds, 0, "Only available in ClickHouse Cloud. If a merged part is less than this many seconds old and is not pre-warmed (see cache_populated_by_fetch), but all its source parts are available and pre-warmed, SELECT queries will read from those parts instead. Only for ReplicatedMergeTree. Note that this only checks whether CacheWarmer processed the part; if the part was fetched into cache by something else, it'll still be considered cold until CacheWarmer gets to it; if it was warmed, then evicted from cache, it'll still be considered warm.", 0) \ + M(Bool, iceberg_engine_ignore_schema_evolution, false, "Ignore schema evolution in Iceberg table engine and read all data using latest schema saved on table creation. Note that it can lead to incorrect result", 0) \ + M(Bool, allow_deprecated_error_prone_window_functions, false, "Allow usage of deprecated error prone window functions (neighbor, runningAccumulate, runningDifferenceStartingWithFirstValue, runningDifference)", 0) \ + M(Bool, allow_deprecated_snowflake_conversion_functions, false, "Enables deprecated functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake.", 0) \ + M(Bool, optimize_distinct_in_order, true, "Enable DISTINCT optimization if some columns in DISTINCT form a prefix of sorting. For example, prefix of sorting key in merge tree or ORDER BY statement", 0) \ + M(Bool, keeper_map_strict_mode, false, "Enforce additional checks during operations on KeeperMap. E.g. throw an exception on an insert for already existing key", 0) \ + M(UInt64, extract_key_value_pairs_max_pairs_per_row, 1000, "Max number of pairs that can be produced by the `extractKeyValuePairs` function. Used as a safeguard against consuming too much memory.", 0) ALIAS(extract_kvp_max_pairs_per_row) \ + M(Bool, mongodb_fail_on_query_build_error, true, "If enabled, MongoDB tables will return an error when a MongoDB query can't be built. Not applied for the legacy implementation, or when 'allow_experimental_analyzer=0`.", 0) \ \ - /** Experimental functions */ \ + \ + /* ###################################### */ \ + /* ######## EXPERIMENTAL FEATURES ####### */ \ + /* ###################################### */ \ M(Bool, allow_experimental_materialized_postgresql_table, false, "Allows to use the MaterializedPostgreSQL table engine. Disabled by default, because this feature is experimental", 0) \ M(Bool, allow_experimental_funnel_functions, false, "Enable experimental functions for funnel analysis.", 0) \ M(Bool, allow_experimental_nlp_functions, false, "Enable experimental functions for natural language processing.", 0) \ @@ -909,6 +908,7 @@ class IColumn; M(Bool, allow_experimental_dynamic_type, false, "Allow Dynamic data type", 0) \ M(Bool, allow_experimental_annoy_index, false, "Allows to use Annoy index. Disabled by default because this feature is experimental", 0) \ M(Bool, allow_experimental_usearch_index, false, "Allows to use USearch index. Disabled by default because this feature is experimental", 0) \ + M(Bool, allow_experimental_codecs, false, "If it is set to true, allow to specify experimental compression codecs (but we don't have those yet and this option does nothing).", 0) \ M(UInt64, max_limit_for_ann_queries, 1'000'000, "SELECT queries with LIMIT bigger than this setting cannot use ANN indexes. Helps to prevent memory overflows in ANN search indexes.", 0) \ M(UInt64, max_threads_for_annoy_index_creation, 4, "Number of threads used to build Annoy indexes (0 means all cores, not recommended)", 0) \ M(Int64, annoy_index_search_k_nodes, -1, "SELECT queries search up to this many nodes in Annoy indexes.", 0) \ @@ -917,26 +917,55 @@ class IColumn; M(Bool, implicit_transaction, false, "If enabled and not already inside a transaction, wraps the query inside a full transaction (begin + commit or rollback)", 0) \ M(UInt64, grace_hash_join_initial_buckets, 1, "Initial number of grace hash join buckets", 0) \ M(UInt64, grace_hash_join_max_buckets, 1024, "Limit on the number of grace hash join buckets", 0) \ - M(Bool, optimize_distinct_in_order, true, "Enable DISTINCT optimization if some columns in DISTINCT form a prefix of sorting. For example, prefix of sorting key in merge tree or ORDER BY statement", 0) \ - M(Bool, keeper_map_strict_mode, false, "Enforce additional checks during operations on KeeperMap. E.g. throw an exception on an insert for already existing key", 0) \ - M(UInt64, extract_key_value_pairs_max_pairs_per_row, 1000, "Max number of pairs that can be produced by the `extractKeyValuePairs` function. Used as a safeguard against consuming too much memory.", 0) ALIAS(extract_kvp_max_pairs_per_row) \ M(Timezone, session_timezone, "", "This setting can be removed in the future due to potential caveats. It is experimental and is not suitable for production usage. The default timezone for current session or query. The server default timezone if empty.", 0) \ - M(Bool, allow_create_index_without_type, false, "Allow CREATE INDEX query without TYPE. Query will be ignored. Made for SQL compatibility tests.", 0) \ - M(Bool, create_index_ignore_unique, false, "Ignore UNIQUE keyword in CREATE UNIQUE INDEX. Made for SQL compatibility tests.", 0) \ - M(Bool, print_pretty_type_names, true, "Print pretty type names in DESCRIBE query and toTypeName() function", 0) \ - M(Bool, create_table_empty_primary_key_by_default, false, "Allow to create *MergeTree tables with empty primary key when ORDER BY and PRIMARY KEY not specified", 0) \ - M(Bool, allow_named_collection_override_by_default, true, "Allow named collections' fields override by default.", 0)\ - M(SQLSecurityType, default_normal_view_sql_security, SQLSecurityType::INVOKER, "Allows to set a default value for SQL SECURITY option when creating a normal view.", 0) \ - M(SQLSecurityType, default_materialized_view_sql_security, SQLSecurityType::DEFINER, "Allows to set a default value for SQL SECURITY option when creating a materialized view.", 0) \ - M(String, default_view_definer, "CURRENT_USER", "Allows to set a default value for DEFINER option when creating view.", 0) \ - M(UInt64, cache_warmer_threads, 4, "Only available in ClickHouse Cloud. Number of background threads for speculatively downloading new data parts into file cache, when cache_populated_by_fetch is enabled. Zero to disable.", 0) \ - M(Int64, ignore_cold_parts_seconds, 0, "Only available in ClickHouse Cloud. Exclude new data parts from SELECT queries until they're either pre-warmed (see cache_populated_by_fetch) or this many seconds old. Only for Replicated-/SharedMergeTree.", 0) \ - M(Int64, prefer_warmed_unmerged_parts_seconds, 0, "Only available in ClickHouse Cloud. If a merged part is less than this many seconds old and is not pre-warmed (see cache_populated_by_fetch), but all its source parts are available and pre-warmed, SELECT queries will read from those parts instead. Only for ReplicatedMergeTree. Note that this only checks whether CacheWarmer processed the part; if the part was fetched into cache by something else, it'll still be considered cold until CacheWarmer gets to it; if it was warmed, then evicted from cache, it'll still be considered warm.", 0) \ - M(Bool, iceberg_engine_ignore_schema_evolution, false, "Ignore schema evolution in Iceberg table engine and read all data using latest schema saved on table creation. Note that it can lead to incorrect result", 0) \ - M(Bool, allow_deprecated_error_prone_window_functions, false, "Allow usage of deprecated error prone window functions (neighbor, runningAccumulate, runningDifferenceStartingWithFirstValue, runningDifference)", 0) \ - M(Bool, uniform_snowflake_conversion_functions, true, "Enables functions snowflakeIDToDateTime[64] and dateTime[64]ToSnowflakeID while disabling functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake.", 0) \ - M(Bool, allow_deprecated_snowflake_conversion_functions, false, "Enables deprecated functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake.", 0) \ - M(Bool, mongodb_fail_on_query_build_error, true, "If enabled, MongoDB tables will return an error when a MongoDB query can't be built. Not applied for the legacy implementation, or when 'allow_experimental_analyzer=0`.", 0) + \ + M(Bool, allow_statistics_optimize, false, "Allows using statistics to optimize queries", 0) ALIAS(allow_statistic_optimize) \ + M(Bool, allow_experimental_statistics, false, "Allows using statistics", 0) ALIAS(allow_experimental_statistic) \ + \ + /* Parallel replicas */ \ + M(UInt64, parallel_replicas_count, 0, "This is internal setting that should not be used directly and represents an implementation detail of the 'parallel replicas' mode. This setting will be automatically set up by the initiator server for distributed queries to the number of parallel replicas participating in query processing.", 0) \ + M(UInt64, parallel_replica_offset, 0, "This is internal setting that should not be used directly and represents an implementation detail of the 'parallel replicas' mode. This setting will be automatically set up by the initiator server for distributed queries to the index of the replica participating in query processing among parallel replicas.", 0) \ + M(String, parallel_replicas_custom_key, "", "Custom key assigning work to replicas when parallel replicas are used.", 0) \ + M(ParallelReplicasCustomKeyFilterType, parallel_replicas_custom_key_filter_type, ParallelReplicasCustomKeyFilterType::DEFAULT, "Type of filter to use with custom key for parallel replicas. default - use modulo operation on the custom key, range - use range filter on custom key using all possible values for the value type of custom key.", 0) \ + M(UInt64, parallel_replicas_custom_key_range_lower, 0, "Lower bound for the universe that the parallel replicas custom range filter is calculated over", 0) \ + M(UInt64, parallel_replicas_custom_key_range_upper, 0, "Upper bound for the universe that the parallel replicas custom range filter is calculated over. A value of 0 disables the upper bound, setting it to the max value of the custom key expression", 0) \ + M(String, cluster_for_parallel_replicas, "", "Cluster for a shard in which current server is located", 0) \ + M(UInt64, allow_experimental_parallel_reading_from_replicas, 0, "Use all the replicas from a shard for SELECT query execution. Reading is parallelized and coordinated dynamically. 0 - disabled, 1 - enabled, silently disable them in case of failure, 2 - enabled, throw an exception in case of failure", 0) \ + M(Bool, parallel_replicas_allow_in_with_subquery, true, "If true, subquery for IN will be executed on every follower replica.", 0) \ + M(Float, parallel_replicas_single_task_marks_count_multiplier, 2, "A multiplier which will be added during calculation for minimal number of marks to retrieve from coordinator. This will be applied only for remote replicas.", 0) \ + M(Bool, parallel_replicas_for_non_replicated_merge_tree, false, "If true, ClickHouse will use parallel replicas algorithm also for non-replicated MergeTree tables", 0) \ + M(UInt64, parallel_replicas_min_number_of_rows_per_replica, 0, "Limit the number of replicas used in a query to (estimated rows to read / min_number_of_rows_per_replica). The max is still limited by 'max_parallel_replicas'", 0) \ + M(Bool, parallel_replicas_prefer_local_join, true, "If true, and JOIN can be executed with parallel replicas algorithm, and all storages of right JOIN part are *MergeTree, local JOIN will be used instead of GLOBAL JOIN.", 0) \ + M(UInt64, parallel_replicas_mark_segment_size, 128, "Parts virtually divided into segments to be distributed between replicas for parallel reading. This setting controls the size of these segments. Not recommended to change until you're absolutely sure in what you're doing", 0) \ + \ + M(Bool, allow_experimental_inverted_index, false, "If it is set to true, allow to use experimental inverted index.", 0) \ + M(Bool, allow_experimental_full_text_index, false, "If it is set to true, allow to use experimental full-text index.", 0) \ + \ + M(Bool, allow_experimental_join_condition, false, "Support join with inequal conditions which involve columns from both left and right table. e.g. t1.y < t2.y.", 0) \ + \ + /* Analyzer: It's not experimental anymore (WIP) */ \ + M(Bool, allow_experimental_analyzer, true, "Allow new query analyzer.", IMPORTANT) \ + M(Bool, analyzer_compatibility_join_using_top_level_identifier, false, "Force to resolve identifier in JOIN USING from projection (for example, in `SELECT a + 1 AS b FROM t1 JOIN t2 USING (b)` join will be performed by `t1.a + 1 = t2.b`, rather then `t1.b = t2.b`).", 0) \ + \ + M(Bool, allow_experimental_live_view, false, "Enable LIVE VIEW. Not mature enough.", 0) \ + M(Seconds, live_view_heartbeat_interval, 15, "The heartbeat interval in seconds to indicate live query is alive.", 0) \ + M(UInt64, max_live_view_insert_blocks_before_refresh, 64, "Limit maximum number of inserted blocks after which mergeable blocks are dropped and query is re-executed.", 0) \ + \ + M(Bool, allow_experimental_window_view, false, "Enable WINDOW VIEW. Not mature enough.", 0) \ + M(Seconds, window_view_clean_interval, 60, "The clean interval of window view in seconds to free outdated data.", 0) \ + M(Seconds, window_view_heartbeat_interval, 15, "The heartbeat interval in seconds to indicate watch query is alive.", 0) \ + M(Seconds, wait_for_window_view_fire_signal_timeout, 10, "Timeout for waiting for window view fire signal in event time processing", 0) \ + \ + M(Bool, allow_experimental_refreshable_materialized_view, false, "Allow refreshable materialized views (CREATE MATERIALIZED VIEW REFRESH ...).", 0) \ + M(Bool, stop_refreshable_materialized_views_on_startup, false, "On server startup, prevent scheduling of refreshable materialized views, as if with SYSTEM STOP VIEWS. You can manually start them with SYSTEM START VIEWS or SYSTEM START VIEW afterwards. Also applies to newly created views. Has no effect on non-refreshable materialized views.", 0) \ + \ + M(Bool, allow_experimental_database_materialized_mysql, false, "Allow to create database with Engine=MaterializedMySQL(...).", 0) \ + M(Bool, allow_experimental_database_materialized_postgresql, false, "Allow to create database with Engine=MaterializedPostgreSQL(...).", 0) \ + \ + /** Experimental feature for moving data between shards. */ \ + M(Bool, allow_experimental_query_deduplication, false, "Experimental data deduplication for SELECT queries based on part UUIDs", 0) \ + + /** End of experimental features */ // End of COMMON_SETTINGS // Please add settings related to formats into the FORMAT_FACTORY_SETTINGS, move obsolete settings to OBSOLETE_SETTINGS and obsolete format settings to OBSOLETE_FORMAT_SETTINGS. @@ -950,6 +979,7 @@ class IColumn; #define OBSOLETE_SETTINGS(M, ALIAS) \ /** Obsolete settings that do nothing but left for compatibility reasons. Remove each one after half a year of obsolescence. */ \ + MAKE_OBSOLETE(M, Bool, update_insert_deduplication_token_in_dependent_materialized_views, 0) \ MAKE_OBSOLETE(M, UInt64, max_memory_usage_for_all_queries, 0) \ MAKE_OBSOLETE(M, UInt64, multiple_joins_rewriter_version, 0) \ MAKE_OBSOLETE(M, Bool, enable_debug_queries, false) \ @@ -989,6 +1019,7 @@ class IColumn; MAKE_DEPRECATED_BY_SERVER_CONFIG(M, UInt64, async_insert_threads, 16) \ MAKE_DEPRECATED_BY_SERVER_CONFIG(M, UInt64, max_replicated_fetches_network_bandwidth_for_server, 0) \ MAKE_DEPRECATED_BY_SERVER_CONFIG(M, UInt64, max_replicated_sends_network_bandwidth_for_server, 0) \ + MAKE_DEPRECATED_BY_SERVER_CONFIG(M, UInt64, max_entries_for_hash_table_stats, 10'000) \ /* ---- */ \ MAKE_OBSOLETE(M, DefaultDatabaseEngine, default_database_engine, DefaultDatabaseEngine::Atomic) \ MAKE_OBSOLETE(M, UInt64, max_pipeline_depth, 0) \ @@ -1043,6 +1074,7 @@ class IColumn; M(Bool, input_format_orc_allow_missing_columns, true, "Allow missing columns while reading ORC input formats", 0) \ M(Bool, input_format_orc_use_fast_decoder, true, "Use a faster ORC decoder implementation.", 0) \ M(Bool, input_format_orc_filter_push_down, true, "When reading ORC files, skip whole stripes or row groups based on the WHERE/PREWHERE expressions, min/max statistics or bloom filter in the ORC metadata.", 0) \ + M(Bool, input_format_orc_read_use_writer_time_zone, false, "Whether use the writer's time zone in ORC stripe for ORC row reader, the default ORC row reader's time zone is GMT.", 0) \ M(Bool, input_format_parquet_allow_missing_columns, true, "Allow missing columns while reading Parquet input formats", 0) \ M(UInt64, input_format_parquet_local_file_min_bytes_for_seek, 8192, "Min bytes required for local read (file) to do seek, instead of read with ignore in Parquet input format", 0) \ M(Bool, input_format_arrow_allow_missing_columns, true, "Allow missing columns while reading Arrow input formats", 0) \ @@ -1094,6 +1126,7 @@ class IColumn; M(Bool, input_format_json_defaults_for_missing_elements_in_named_tuple, true, "Insert default value in named tuple element if it's missing in json object", 0) \ M(Bool, input_format_json_throw_on_bad_escape_sequence, true, "Throw an exception if JSON string contains bad escape sequence in JSON input formats. If disabled, bad escape sequences will remain as is in the data", 0) \ M(Bool, input_format_json_ignore_unnecessary_fields, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields", 0) \ + M(Bool, input_format_json_case_insensitive_column_matching, false, "Ignore case when matching JSON keys with CH columns", 0) \ M(Bool, input_format_try_infer_integers, true, "Try to infer integers instead of floats while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_dates, true, "Try to infer dates from string fields while schema inference in text formats", 0) \ M(Bool, input_format_try_infer_datetimes, true, "Try to infer datetimes from string fields while schema inference in text formats", 0) \ @@ -1109,6 +1142,8 @@ class IColumn; M(Bool, input_format_tsv_crlf_end_of_line, false, "If it is set true, file function will read TSV format with \\r\\n instead of \\n.", 0) \ \ M(Bool, input_format_native_allow_types_conversion, true, "Allow data types conversion in Native input format", 0) \ + M(Bool, input_format_native_decode_types_in_binary_format, false, "Read data types in binary format instead of type names in Native input format", 0) \ + M(Bool, output_format_native_encode_types_in_binary_format, false, "Write data types in binary format instead of type names in Native output format", 0) \ \ M(DateTimeInputFormat, date_time_input_format, FormatSettings::DateTimeInputFormat::Basic, "Method to read DateTime from text input formats. Possible values: 'basic', 'best_effort' and 'best_effort_us'.", 0) \ M(DateTimeOutputFormat, date_time_output_format, FormatSettings::DateTimeOutputFormat::Simple, "Method to write DateTime to text output. Possible values: 'simple', 'iso', 'unix_timestamp'.", 0) \ @@ -1128,6 +1163,8 @@ class IColumn; M(Bool, input_format_avro_null_as_default, false, "For Avro/AvroConfluent format: insert default in case of null and non Nullable column", 0) \ M(UInt64, format_binary_max_string_size, 1_GiB, "The maximum allowed size for String in RowBinary format. It prevents allocating large amount of memory in case of corrupted data. 0 means there is no limit", 0) \ M(UInt64, format_binary_max_array_size, 1_GiB, "The maximum allowed size for Array in RowBinary format. It prevents allocating large amount of memory in case of corrupted data. 0 means there is no limit", 0) \ + M(Bool, input_format_binary_decode_types_in_binary_format, false, "Read data types in binary format instead of type names in RowBinaryWithNamesAndTypes input format", 0) \ + M(Bool, output_format_binary_encode_types_in_binary_format, false, "Write data types in binary format instead of type names in RowBinaryWithNamesAndTypes output format ", 0) \ M(URI, format_avro_schema_registry_url, "", "For AvroConfluent format: Confluent Schema Registry URL.", 0) \ \ M(Bool, output_format_json_quote_64bit_integers, true, "Controls quoting of 64-bit integers in JSON output format.", 0) \ @@ -1148,9 +1185,9 @@ class IColumn; M(UInt64, output_format_pretty_max_value_width, 10000, "Maximum width of value to display in Pretty formats. If greater - it will be cut.", 0) \ M(UInt64, output_format_pretty_max_value_width_apply_for_single_value, false, "Only cut values (see the `output_format_pretty_max_value_width` setting) when it is not a single value in a block. Otherwise output it entirely, which is useful for the `SHOW CREATE TABLE` query.", 0) \ M(UInt64Auto, output_format_pretty_color, "auto", "Use ANSI escape sequences in Pretty formats. 0 - disabled, 1 - enabled, 'auto' - enabled if a terminal.", 0) \ - M(String, output_format_pretty_grid_charset, "UTF-8", "Charset for printing grid borders. Available charsets: ASCII, UTF-8 (default one).", 0) \ - M(UInt64, output_format_pretty_display_footer_column_names, true, "Display column names in the footer if there are 999 or more rows.", 0) \ - M(UInt64, output_format_pretty_display_footer_column_names_min_rows, 50, "Sets the minimum threshold value of rows for which to enable displaying column names in the footer. 50 (default)", 0) \ + M(String, output_format_pretty_grid_charset, "UTF-8", "Charset for printing grid borders. Available charsets: ASCII, UTF-8 (default one).", 0) \ + M(UInt64, output_format_pretty_display_footer_column_names, true, "Display column names in the footer if there are 999 or more rows.", 0) \ + M(UInt64, output_format_pretty_display_footer_column_names_min_rows, 50, "Sets the minimum threshold value of rows for which to enable displaying column names in the footer. 50 (default)", 0) \ M(UInt64, output_format_parquet_row_group_size, 1000000, "Target row group size in rows.", 0) \ M(UInt64, output_format_parquet_row_group_size_bytes, 512 * 1024 * 1024, "Target row group size in bytes, before compression.", 0) \ M(Bool, output_format_parquet_string_as_string, true, "Use Parquet String type instead of Binary for String columns.", 0) \ @@ -1314,3 +1351,5 @@ struct FormatFactorySettings : public BaseSettings }; } + +#endif diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp new file mode 100644 index 00000000000..9faf77e9087 --- /dev/null +++ b/src/Core/SettingsChangesHistory.cpp @@ -0,0 +1,345 @@ +#include +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; + extern const int LOGICAL_ERROR; +} + +ClickHouseVersion::ClickHouseVersion(const String & version) +{ + Strings split; + boost::split(split, version, [](char c){ return c == '.'; }); + components.reserve(split.size()); + if (split.empty()) + throw Exception{ErrorCodes::BAD_ARGUMENTS, "Cannot parse ClickHouse version here: {}", version}; + + for (const auto & split_element : split) + { + size_t component; + ReadBufferFromString buf(split_element); + if (!tryReadIntText(component, buf) || !buf.eof()) + throw Exception{ErrorCodes::BAD_ARGUMENTS, "Cannot parse ClickHouse version here: {}", version}; + components.push_back(component); + } +} + +ClickHouseVersion::ClickHouseVersion(const char * version) + : ClickHouseVersion(String(version)) +{ +} + +String ClickHouseVersion::toString() const +{ + String version = std::to_string(components[0]); + for (size_t i = 1; i < components.size(); ++i) + version += "." + std::to_string(components[i]); + + return version; +} + +// clang-format off +/// History of settings changes that controls some backward incompatible changes +/// across all ClickHouse versions. It maps ClickHouse version to settings changes that were done +/// in this version. This history contains both changes to existing settings and newly added settings. +/// Settings changes is a vector of structs +/// {setting_name, previous_value, new_value, reason}. +/// For newly added setting choose the most appropriate previous_value (for example, if new setting +/// controls new feature and it's 'true' by default, use 'false' as previous_value). +/// It's used to implement `compatibility` setting (see https://github.com/ClickHouse/ClickHouse/issues/35972) +/// Note: please check if the key already exists to prevent duplicate entries. +static std::initializer_list> settings_changes_history_initializer = +{ + {"24.7", {{"output_format_parquet_write_page_index", false, true, "Add a possibility to write page index into parquet files."}, + {"output_format_binary_encode_types_in_binary_format", false, false, "Added new setting to allow to write type names in binary format in RowBinaryWithNamesAndTypes output format"}, + {"input_format_binary_decode_types_in_binary_format", false, false, "Added new setting to allow to read type names in binary format in RowBinaryWithNamesAndTypes input format"}, + {"output_format_native_encode_types_in_binary_format", false, false, "Added new setting to allow to write type names in binary format in Native output format"}, + {"input_format_native_decode_types_in_binary_format", false, false, "Added new setting to allow to read type names in binary format in Native output format"}, + {"read_in_order_use_buffering", false, true, "Use buffering before merging while reading in order of primary key"}, + {"enable_named_columns_in_function_tuple", false, true, "Generate named tuples in function tuple() when all names are unique and can be treated as unquoted identifiers."}, + {"input_format_json_case_insensitive_column_matching", false, false, "Ignore case when matching JSON keys with CH columns."}, + {"optimize_trivial_insert_select", true, false, "The optimization does not make sense in many cases."}, + {"dictionary_validate_primary_key_type", false, false, "Validate primary key type for dictionaries. By default id type for simple layouts will be implicitly converted to UInt64."}, + {"collect_hash_table_stats_during_joins", false, true, "New setting."}, + {"max_size_to_preallocate_for_joins", 0, 100'000'000, "New setting."}, + {"input_format_orc_read_use_writer_time_zone", false, false, "Whether use the writer's time zone in ORC stripe for ORC row reader, the default ORC row reader's time zone is GMT."}, + {"lightweight_mutation_projection_mode", "throw", "throw", "When lightweight delete happens on a table with projection(s), the possible operations include throw the exception as projection exists, or drop all projection related to this table then do lightweight delete."}, + {"database_replicated_allow_heavy_create", true, false, "Long-running DDL queries (CREATE AS SELECT and POPULATE) for Replicated database engine was forbidden"}, + {"query_plan_merge_filters", false, false, "Allow to merge filters in the query plan"}, + {"azure_sdk_max_retries", 10, 10, "Maximum number of retries in azure sdk"}, + {"azure_sdk_retry_initial_backoff_ms", 10, 10, "Minimal backoff between retries in azure sdk"}, + {"azure_sdk_retry_max_backoff_ms", 1000, 1000, "Maximal backoff between retries in azure sdk"}, + {"merge_tree_min_bytes_per_task_for_remote_reading", 4194304, 2097152, "Value is unified with `filesystem_prefetch_min_bytes_for_single_read_task`"}, + {"ignore_on_cluster_for_replicated_named_collections_queries", false, false, "Ignore ON CLUSTER clause for replicated named collections management queries."}, + {"backup_restore_s3_retry_attempts", 1000,1000, "Setting for Aws::Client::RetryStrategy, Aws::Client does retries itself, 0 means no retries. It takes place only for backup/restore."}, + {"postgresql_connection_attempt_timeout", 2, 2, "Allow to control 'connect_timeout' parameter of PostgreSQL connection."}, + {"postgresql_connection_pool_retries", 2, 2, "Allow to control the number of retries in PostgreSQL connection pool."} + }}, + {"24.6", {{"materialize_skip_indexes_on_insert", true, true, "Added new setting to allow to disable materialization of skip indexes on insert"}, + {"materialize_statistics_on_insert", true, true, "Added new setting to allow to disable materialization of statistics on insert"}, + {"input_format_parquet_use_native_reader", false, false, "When reading Parquet files, to use native reader instead of arrow reader."}, + {"hdfs_throw_on_zero_files_match", false, false, "Allow to throw an error when ListObjects request cannot match any files in HDFS engine instead of empty query result"}, + {"azure_throw_on_zero_files_match", false, false, "Allow to throw an error when ListObjects request cannot match any files in AzureBlobStorage engine instead of empty query result"}, + {"s3_validate_request_settings", true, true, "Allow to disable S3 request settings validation"}, + {"allow_experimental_full_text_index", false, false, "Enable experimental full-text index"}, + {"azure_skip_empty_files", false, false, "Allow to skip empty files in azure table engine"}, + {"hdfs_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in HDFS table engine"}, + {"azure_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in AzureBlobStorage table engine"}, + {"s3_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in S3 table engine"}, + {"s3_max_part_number", 10000, 10000, "Maximum part number number for s3 upload part"}, + {"s3_max_single_operation_copy_size", 32 * 1024 * 1024, 32 * 1024 * 1024, "Maximum size for a single copy operation in s3"}, + {"input_format_parquet_max_block_size", 8192, DEFAULT_BLOCK_SIZE, "Increase block size for parquet reader."}, + {"input_format_parquet_prefer_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Average block bytes output by parquet reader."}, + {"enable_blob_storage_log", true, true, "Write information about blob storage operations to system.blob_storage_log table"}, + {"allow_deprecated_snowflake_conversion_functions", true, false, "Disabled deprecated functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake."}, + {"allow_statistic_optimize", false, false, "Old setting which popped up here being renamed."}, + {"allow_experimental_statistic", false, false, "Old setting which popped up here being renamed."}, + {"allow_statistics_optimize", false, false, "The setting was renamed. The previous name is `allow_statistic_optimize`."}, + {"allow_experimental_statistics", false, false, "The setting was renamed. The previous name is `allow_experimental_statistic`."}, + {"enable_vertical_final", false, true, "Enable vertical final by default again after fixing bug"}, + {"parallel_replicas_custom_key_range_lower", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards"}, + {"parallel_replicas_custom_key_range_upper", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards. A value of 0 disables the upper limit"}, + {"output_format_pretty_display_footer_column_names", 0, 1, "Add a setting to display column names in the footer if there are many rows. Threshold value is controlled by output_format_pretty_display_footer_column_names_min_rows."}, + {"output_format_pretty_display_footer_column_names_min_rows", 0, 50, "Add a setting to control the threshold value for setting output_format_pretty_display_footer_column_names_min_rows. Default 50."}, + {"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."}, + {"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."}, + {"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."}, + }}, + {"24.5", {{"allow_deprecated_error_prone_window_functions", true, false, "Allow usage of deprecated error prone window functions (neighbor, runningAccumulate, runningDifferenceStartingWithFirstValue, runningDifference)"}, + {"allow_experimental_join_condition", false, false, "Support join with inequal conditions which involve columns from both left and right table. e.g. t1.y < t2.y."}, + {"input_format_tsv_crlf_end_of_line", false, false, "Enables reading of CRLF line endings with TSV formats"}, + {"output_format_parquet_use_custom_encoder", false, true, "Enable custom Parquet encoder."}, + {"cross_join_min_rows_to_compress", 0, 10000000, "Minimal count of rows to compress block in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached."}, + {"cross_join_min_bytes_to_compress", 0, 1_GiB, "Minimal size of block to compress in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached."}, + {"http_max_chunk_size", 0, 0, "Internal limitation"}, + {"prefer_external_sort_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Prefer maximum block bytes for external sort, reduce the memory usage during merging."}, + {"input_format_force_null_for_omitted_fields", false, false, "Disable type-defaults for omitted fields when needed"}, + {"cast_string_to_dynamic_use_inference", false, false, "Add setting to allow converting String to Dynamic through parsing"}, + {"allow_experimental_dynamic_type", false, false, "Add new experimental Dynamic type"}, + {"azure_max_blocks_in_multipart_upload", 50000, 50000, "Maximum number of blocks in multipart upload for Azure."}, + }}, + {"24.4", {{"input_format_json_throw_on_bad_escape_sequence", true, true, "Allow to save JSON strings with bad escape sequences"}, + {"max_parsing_threads", 0, 0, "Add a separate setting to control number of threads in parallel parsing from files"}, + {"ignore_drop_queries_probability", 0, 0, "Allow to ignore drop queries in server with specified probability for testing purposes"}, + {"lightweight_deletes_sync", 2, 2, "The same as 'mutation_sync', but controls only execution of lightweight deletes"}, + {"query_cache_system_table_handling", "save", "throw", "The query cache no longer caches results of queries against system tables"}, + {"input_format_json_ignore_unnecessary_fields", false, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields"}, + {"input_format_hive_text_allow_variable_number_of_columns", false, true, "Ignore extra columns in Hive Text input (if file has more columns than expected) and treat missing fields in Hive Text input as default values."}, + {"allow_experimental_database_replicated", false, true, "Database engine Replicated is now in Beta stage"}, + {"temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds", (10 * 60 * 1000), (10 * 60 * 1000), "Wait time to lock cache for sapce reservation in temporary data in filesystem cache"}, + {"optimize_rewrite_sum_if_to_count_if", false, true, "Only available for the analyzer, where it works correctly"}, + {"azure_allow_parallel_part_upload", "true", "true", "Use multiple threads for azure multipart upload."}, + {"max_recursive_cte_evaluation_depth", DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, "Maximum limit on recursive CTE evaluation depth"}, + {"query_plan_convert_outer_join_to_inner_join", false, true, "Allow to convert OUTER JOIN to INNER JOIN if filter after JOIN always filters default values"}, + }}, + {"24.3", {{"s3_connect_timeout_ms", 1000, 1000, "Introduce new dedicated setting for s3 connection timeout"}, + {"allow_experimental_shared_merge_tree", false, true, "The setting is obsolete"}, + {"use_page_cache_for_disks_without_file_cache", false, false, "Added userspace page cache"}, + {"read_from_page_cache_if_exists_otherwise_bypass_cache", false, false, "Added userspace page cache"}, + {"page_cache_inject_eviction", false, false, "Added userspace page cache"}, + {"default_table_engine", "None", "MergeTree", "Set default table engine to MergeTree for better usability"}, + {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"}, + {"traverse_shadow_remote_data_paths", false, false, "Traverse shadow directory when query system.remote_data_paths."}, + {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication in dependent materialized view cannot work together with async inserts."}, + {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"}, + {"log_processors_profiles", false, true, "Enable by default"}, + {"function_locate_has_mysql_compatible_argument_order", false, true, "Increase compatibility with MySQL's locate function."}, + {"allow_suspicious_primary_key", true, false, "Forbid suspicious PRIMARY KEY/ORDER BY for MergeTree (i.e. SimpleAggregateFunction)"}, + {"filesystem_cache_reserve_space_wait_lock_timeout_milliseconds", 1000, 1000, "Wait time to lock cache for sapce reservation in filesystem cache"}, + {"max_parser_backtracks", 0, 1000000, "Limiting the complexity of parsing"}, + {"analyzer_compatibility_join_using_top_level_identifier", false, false, "Force to resolve identifier in JOIN USING from projection"}, + {"distributed_insert_skip_read_only_replicas", false, false, "If true, INSERT into Distributed will skip read-only replicas"}, + {"keeper_max_retries", 10, 10, "Max retries for general keeper operations"}, + {"keeper_retry_initial_backoff_ms", 100, 100, "Initial backoff timeout for general keeper operations"}, + {"keeper_retry_max_backoff_ms", 5000, 5000, "Max backoff timeout for general keeper operations"}, + {"s3queue_allow_experimental_sharded_mode", false, false, "Enable experimental sharded mode of S3Queue table engine. It is experimental because it will be rewritten"}, + {"allow_experimental_analyzer", false, true, "Enable analyzer and planner by default."}, + {"merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability", 0.0, 0.0, "For testing of `PartsSplitter` - split read ranges into intersecting and non intersecting every time you read from MergeTree with the specified probability."}, + {"allow_get_client_http_header", false, false, "Introduced a new function."}, + {"output_format_pretty_row_numbers", false, true, "It is better for usability."}, + {"output_format_pretty_max_value_width_apply_for_single_value", true, false, "Single values in Pretty formats won't be cut."}, + {"output_format_parquet_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, + {"output_format_orc_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, + {"output_format_arrow_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, + {"output_format_parquet_compression_method", "lz4", "zstd", "Parquet/ORC/Arrow support many compression methods, including lz4 and zstd. ClickHouse supports each and every compression method. Some inferior tools, such as 'duckdb', lack support for the faster `lz4` compression method, that's why we set zstd by default."}, + {"output_format_orc_compression_method", "lz4", "zstd", "Parquet/ORC/Arrow support many compression methods, including lz4 and zstd. ClickHouse supports each and every compression method. Some inferior tools, such as 'duckdb', lack support for the faster `lz4` compression method, that's why we set zstd by default."}, + {"output_format_pretty_highlight_digit_groups", false, true, "If enabled and if output is a terminal, highlight every digit corresponding to the number of thousands, millions, etc. with underline."}, + {"geo_distance_returns_float64_on_float64_arguments", false, true, "Increase the default precision."}, + {"azure_max_inflight_parts_for_one_file", 20, 20, "The maximum number of a concurrent loaded parts in multipart upload request. 0 means unlimited."}, + {"azure_strict_upload_part_size", 0, 0, "The exact size of part to upload during multipart upload to Azure blob storage."}, + {"azure_min_upload_part_size", 16*1024*1024, 16*1024*1024, "The minimum size of part to upload during multipart upload to Azure blob storage."}, + {"azure_max_upload_part_size", 5ull*1024*1024*1024, 5ull*1024*1024*1024, "The maximum size of part to upload during multipart upload to Azure blob storage."}, + {"azure_upload_part_size_multiply_factor", 2, 2, "Multiply azure_min_upload_part_size by this factor each time azure_multiply_parts_count_threshold parts were uploaded from a single write to Azure blob storage."}, + {"azure_upload_part_size_multiply_parts_count_threshold", 500, 500, "Each time this number of parts was uploaded to Azure blob storage, azure_min_upload_part_size is multiplied by azure_upload_part_size_multiply_factor."}, + {"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."}, + {"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."}, + {"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."}, + }}, + {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"}, + {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"}, + {"output_format_values_escape_quote_with_quote", false, false, "If true escape ' with '', otherwise quoted with \\'"}, + {"output_format_pretty_single_large_number_tip_threshold", 0, 1'000'000, "Print a readable number tip on the right side of the table if the block consists of a single number which exceeds this value (except 0)"}, + {"input_format_try_infer_exponent_floats", true, false, "Don't infer floats in exponential notation by default"}, + {"query_plan_optimize_prewhere", true, true, "Allow to push down filter to PREWHERE expression for supported storages"}, + {"async_insert_max_data_size", 1000000, 10485760, "The previous value appeared to be too small."}, + {"async_insert_poll_timeout_ms", 10, 10, "Timeout in milliseconds for polling data from asynchronous insert queue"}, + {"async_insert_use_adaptive_busy_timeout", false, true, "Use adaptive asynchronous insert timeout"}, + {"async_insert_busy_timeout_min_ms", 50, 50, "The minimum value of the asynchronous insert timeout in milliseconds; it also serves as the initial value, which may be increased later by the adaptive algorithm"}, + {"async_insert_busy_timeout_max_ms", 200, 200, "The minimum value of the asynchronous insert timeout in milliseconds; async_insert_busy_timeout_ms is aliased to async_insert_busy_timeout_max_ms"}, + {"async_insert_busy_timeout_increase_rate", 0.2, 0.2, "The exponential growth rate at which the adaptive asynchronous insert timeout increases"}, + {"async_insert_busy_timeout_decrease_rate", 0.2, 0.2, "The exponential growth rate at which the adaptive asynchronous insert timeout decreases"}, + {"format_template_row_format", "", "", "Template row format string can be set directly in query"}, + {"format_template_resultset_format", "", "", "Template result set format string can be set in query"}, + {"split_parts_ranges_into_intersecting_and_non_intersecting_final", true, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"}, + {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}, + {"azure_max_single_part_copy_size", 256*1024*1024, 256*1024*1024, "The maximum size of object to copy using single part copy to Azure blob storage."}, + {"min_external_table_block_size_rows", DEFAULT_INSERT_BLOCK_SIZE, DEFAULT_INSERT_BLOCK_SIZE, "Squash blocks passed to external table to specified size in rows, if blocks are not big enough"}, + {"min_external_table_block_size_bytes", DEFAULT_INSERT_BLOCK_SIZE * 256, DEFAULT_INSERT_BLOCK_SIZE * 256, "Squash blocks passed to external table to specified size in bytes, if blocks are not big enough."}, + {"parallel_replicas_prefer_local_join", true, true, "If true, and JOIN can be executed with parallel replicas algorithm, and all storages of right JOIN part are *MergeTree, local JOIN will be used instead of GLOBAL JOIN."}, + {"optimize_time_filter_with_preimage", true, true, "Optimize Date and DateTime predicates by converting functions into equivalent comparisons without conversions (e.g. toYear(col) = 2023 -> col >= '2023-01-01' AND col <= '2023-12-31')"}, + {"extract_key_value_pairs_max_pairs_per_row", 0, 0, "Max number of pairs that can be produced by the `extractKeyValuePairs` function. Used as a safeguard against consuming too much memory."}, + {"default_view_definer", "CURRENT_USER", "CURRENT_USER", "Allows to set default `DEFINER` option while creating a view"}, + {"default_materialized_view_sql_security", "DEFINER", "DEFINER", "Allows to set a default value for SQL SECURITY option when creating a materialized view"}, + {"default_normal_view_sql_security", "INVOKER", "INVOKER", "Allows to set default `SQL SECURITY` option while creating a normal view"}, + {"mysql_map_string_to_text_in_show_columns", false, true, "Reduce the configuration effort to connect ClickHouse with BI tools."}, + {"mysql_map_fixed_string_to_text_in_show_columns", false, true, "Reduce the configuration effort to connect ClickHouse with BI tools."}, + }}, + {"24.1", {{"print_pretty_type_names", false, true, "Better user experience."}, + {"input_format_json_read_bools_as_strings", false, true, "Allow to read bools as strings in JSON formats by default"}, + {"output_format_arrow_use_signed_indexes_for_dictionary", false, true, "Use signed indexes type for Arrow dictionaries by default as it's recommended"}, + {"allow_experimental_variant_type", false, false, "Add new experimental Variant type"}, + {"use_variant_as_common_type", false, false, "Allow to use Variant in if/multiIf if there is no common type"}, + {"output_format_arrow_use_64_bit_indexes_for_dictionary", false, false, "Allow to use 64 bit indexes type in Arrow dictionaries"}, + {"parallel_replicas_mark_segment_size", 128, 128, "Add new setting to control segment size in new parallel replicas coordinator implementation"}, + {"ignore_materialized_views_with_dropped_target_table", false, false, "Add new setting to allow to ignore materialized views with dropped target table"}, + {"output_format_compression_level", 3, 3, "Allow to change compression level in the query output"}, + {"output_format_compression_zstd_window_log", 0, 0, "Allow to change zstd window log in the query output when zstd compression is used"}, + {"enable_zstd_qat_codec", false, false, "Add new ZSTD_QAT codec"}, + {"enable_vertical_final", false, true, "Use vertical final by default"}, + {"output_format_arrow_use_64_bit_indexes_for_dictionary", false, false, "Allow to use 64 bit indexes type in Arrow dictionaries"}, + {"max_rows_in_set_to_optimize_join", 100000, 0, "Disable join optimization as it prevents from read in order optimization"}, + {"output_format_pretty_color", true, "auto", "Setting is changed to allow also for auto value, disabling ANSI escapes if output is not a tty"}, + {"function_visible_width_behavior", 0, 1, "We changed the default behavior of `visibleWidth` to be more precise"}, + {"max_estimated_execution_time", 0, 0, "Separate max_execution_time and max_estimated_execution_time"}, + {"iceberg_engine_ignore_schema_evolution", false, false, "Allow to ignore schema evolution in Iceberg table engine"}, + {"optimize_injective_functions_in_group_by", false, true, "Replace injective functions by it's arguments in GROUP BY section in analyzer"}, + {"update_insert_deduplication_token_in_dependent_materialized_views", false, false, "Allow to update insert deduplication token with table identifier during insert in dependent materialized views"}, + {"azure_max_unexpected_write_error_retries", 4, 4, "The maximum number of retries in case of unexpected errors during Azure blob storage write"}, + {"split_parts_ranges_into_intersecting_and_non_intersecting_final", false, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"}, + {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}}}, + {"23.12", {{"allow_suspicious_ttl_expressions", true, false, "It is a new setting, and in previous versions the behavior was equivalent to allowing."}, + {"input_format_parquet_allow_missing_columns", false, true, "Allow missing columns in Parquet files by default"}, + {"input_format_orc_allow_missing_columns", false, true, "Allow missing columns in ORC files by default"}, + {"input_format_arrow_allow_missing_columns", false, true, "Allow missing columns in Arrow files by default"}}}, + {"23.11", {{"parsedatetime_parse_without_leading_zeros", false, true, "Improved compatibility with MySQL DATE_FORMAT/STR_TO_DATE"}}}, + {"23.9", {{"optimize_group_by_constant_keys", false, true, "Optimize group by constant keys by default"}, + {"input_format_json_try_infer_named_tuples_from_objects", false, true, "Try to infer named Tuples from JSON objects by default"}, + {"input_format_json_read_numbers_as_strings", false, true, "Allow to read numbers as strings in JSON formats by default"}, + {"input_format_json_read_arrays_as_strings", false, true, "Allow to read arrays as strings in JSON formats by default"}, + {"input_format_json_infer_incomplete_types_as_strings", false, true, "Allow to infer incomplete types as Strings in JSON formats by default"}, + {"input_format_json_try_infer_numbers_from_strings", true, false, "Don't infer numbers from strings in JSON formats by default to prevent possible parsing errors"}, + {"http_write_exception_in_output_format", false, true, "Output valid JSON/XML on exception in HTTP streaming."}}}, + {"23.8", {{"rewrite_count_distinct_if_with_count_distinct_implementation", false, true, "Rewrite countDistinctIf with count_distinct_implementation configuration"}}}, + {"23.7", {{"function_sleep_max_microseconds_per_block", 0, 3000000, "In previous versions, the maximum sleep time of 3 seconds was applied only for `sleep`, but not for `sleepEachRow` function. In the new version, we introduce this setting. If you set compatibility with the previous versions, we will disable the limit altogether."}}}, + {"23.6", {{"http_send_timeout", 180, 30, "3 minutes seems crazy long. Note that this is timeout for a single network write call, not for the whole upload operation."}, + {"http_receive_timeout", 180, 30, "See http_send_timeout."}}}, + {"23.5", {{"input_format_parquet_preserve_order", true, false, "Allow Parquet reader to reorder rows for better parallelism."}, + {"parallelize_output_from_storages", false, true, "Allow parallelism when executing queries that read from file/url/s3/etc. This may reorder rows."}, + {"use_with_fill_by_sorting_prefix", false, true, "Columns preceding WITH FILL columns in ORDER BY clause form sorting prefix. Rows with different values in sorting prefix are filled independently"}, + {"output_format_parquet_compliant_nested_types", false, true, "Change an internal field name in output Parquet file schema."}}}, + {"23.4", {{"allow_suspicious_indices", true, false, "If true, index can defined with identical expressions"}, + {"allow_nonconst_timezone_arguments", true, false, "Allow non-const timezone arguments in certain time-related functions like toTimeZone(), fromUnixTimestamp*(), snowflakeToDateTime*()."}, + {"connect_timeout_with_failover_ms", 50, 1000, "Increase default connect timeout because of async connect"}, + {"connect_timeout_with_failover_secure_ms", 100, 1000, "Increase default secure connect timeout because of async connect"}, + {"hedged_connection_timeout_ms", 100, 50, "Start new connection in hedged requests after 50 ms instead of 100 to correspond with previous connect timeout"}, + {"formatdatetime_f_prints_single_zero", true, false, "Improved compatibility with MySQL DATE_FORMAT()/STR_TO_DATE()"}, + {"formatdatetime_parsedatetime_m_is_month_name", false, true, "Improved compatibility with MySQL DATE_FORMAT/STR_TO_DATE"}}}, + {"23.3", {{"output_format_parquet_version", "1.0", "2.latest", "Use latest Parquet format version for output format"}, + {"input_format_json_ignore_unknown_keys_in_named_tuple", false, true, "Improve parsing JSON objects as named tuples"}, + {"input_format_native_allow_types_conversion", false, true, "Allow types conversion in Native input forma"}, + {"output_format_arrow_compression_method", "none", "lz4_frame", "Use lz4 compression in Arrow output format by default"}, + {"output_format_parquet_compression_method", "snappy", "lz4", "Use lz4 compression in Parquet output format by default"}, + {"output_format_orc_compression_method", "none", "lz4_frame", "Use lz4 compression in ORC output format by default"}, + {"async_query_sending_for_remote", false, true, "Create connections and send query async across shards"}}}, + {"23.2", {{"output_format_parquet_fixed_string_as_fixed_byte_array", false, true, "Use Parquet FIXED_LENGTH_BYTE_ARRAY type for FixedString by default"}, + {"output_format_arrow_fixed_string_as_fixed_byte_array", false, true, "Use Arrow FIXED_SIZE_BINARY type for FixedString by default"}, + {"query_plan_remove_redundant_distinct", false, true, "Remove redundant Distinct step in query plan"}, + {"optimize_duplicate_order_by_and_distinct", true, false, "Remove duplicate ORDER BY and DISTINCT if it's possible"}, + {"insert_keeper_max_retries", 0, 20, "Enable reconnections to Keeper on INSERT, improve reliability"}}}, + {"23.1", {{"input_format_json_read_objects_as_strings", 0, 1, "Enable reading nested json objects as strings while object type is experimental"}, + {"input_format_json_defaults_for_missing_elements_in_named_tuple", false, true, "Allow missing elements in JSON objects while reading named tuples by default"}, + {"input_format_csv_detect_header", false, true, "Detect header in CSV format by default"}, + {"input_format_tsv_detect_header", false, true, "Detect header in TSV format by default"}, + {"input_format_custom_detect_header", false, true, "Detect header in CustomSeparated format by default"}, + {"query_plan_remove_redundant_sorting", false, true, "Remove redundant sorting in query plan. For example, sorting steps related to ORDER BY clauses in subqueries"}}}, + {"22.12", {{"max_size_to_preallocate_for_aggregation", 10'000'000, 100'000'000, "This optimizes performance"}, + {"query_plan_aggregation_in_order", 0, 1, "Enable some refactoring around query plan"}, + {"format_binary_max_string_size", 0, 1_GiB, "Prevent allocating large amount of memory"}}}, + {"22.11", {{"use_structure_from_insertion_table_in_table_functions", 0, 2, "Improve using structure from insertion table in table functions"}}}, + {"22.9", {{"force_grouping_standard_compatibility", false, true, "Make GROUPING function output the same as in SQL standard and other DBMS"}}}, + {"22.7", {{"cross_to_inner_join_rewrite", 1, 2, "Force rewrite comma join to inner"}, + {"enable_positional_arguments", false, true, "Enable positional arguments feature by default"}, + {"format_csv_allow_single_quotes", true, false, "Most tools don't treat single quote in CSV specially, don't do it by default too"}}}, + {"22.6", {{"output_format_json_named_tuples_as_objects", false, true, "Allow to serialize named tuples as JSON objects in JSON formats by default"}, + {"input_format_skip_unknown_fields", false, true, "Optimize reading subset of columns for some input formats"}}}, + {"22.5", {{"memory_overcommit_ratio_denominator", 0, 1073741824, "Enable memory overcommit feature by default"}, + {"memory_overcommit_ratio_denominator_for_user", 0, 1073741824, "Enable memory overcommit feature by default"}}}, + {"22.4", {{"allow_settings_after_format_in_insert", true, false, "Do not allow SETTINGS after FORMAT for INSERT queries because ClickHouse interpret SETTINGS as some values, which is misleading"}}}, + {"22.3", {{"cast_ipv4_ipv6_default_on_conversion_error", true, false, "Make functions cast(value, 'IPv4') and cast(value, 'IPv6') behave same as toIPv4 and toIPv6 functions"}}}, + {"21.12", {{"stream_like_engine_allow_direct_select", true, false, "Do not allow direct select for Kafka/RabbitMQ/FileLog by default"}}}, + {"21.9", {{"output_format_decimal_trailing_zeros", true, false, "Do not output trailing zeros in text representation of Decimal types by default for better looking output"}, + {"use_hedged_requests", false, true, "Enable Hedged Requests feature by default"}}}, + {"21.7", {{"legacy_column_name_of_tuple_literal", true, false, "Add this setting only for compatibility reasons. It makes sense to set to 'true', while doing rolling update of cluster from version lower than 21.7 to higher"}}}, + {"21.5", {{"async_socket_for_remote", false, true, "Fix all problems and turn on asynchronous reads from socket for remote queries by default again"}}}, + {"21.3", {{"async_socket_for_remote", true, false, "Turn off asynchronous reads from socket for remote queries because of some problems"}, + {"optimize_normalize_count_variants", false, true, "Rewrite aggregate functions that semantically equals to count() as count() by default"}, + {"normalize_function_names", false, true, "Normalize function names to their canonical names, this was needed for projection query routing"}}}, + {"21.2", {{"enable_global_with_statement", false, true, "Propagate WITH statements to UNION queries and all subqueries by default"}}}, + {"21.1", {{"insert_quorum_parallel", false, true, "Use parallel quorum inserts by default. It is significantly more convenient to use than sequential quorum inserts"}, + {"input_format_null_as_default", false, true, "Allow to insert NULL as default for input formats by default"}, + {"optimize_on_insert", false, true, "Enable data optimization on INSERT by default for better user experience"}, + {"use_compact_format_in_distributed_parts_names", false, true, "Use compact format for async INSERT into Distributed tables by default"}}}, + {"20.10", {{"format_regexp_escaping_rule", "Escaped", "Raw", "Use Raw as default escaping rule for Regexp format to male the behaviour more like to what users expect"}}}, + {"20.7", {{"show_table_uuid_in_table_create_query_if_not_nil", true, false, "Stop showing UID of the table in its CREATE query for Engine=Atomic"}}}, + {"20.5", {{"input_format_with_names_use_header", false, true, "Enable using header with names for formats with WithNames/WithNamesAndTypes suffixes"}, + {"allow_suspicious_codecs", true, false, "Don't allow to specify meaningless compression codecs"}}}, + {"20.4", {{"validate_polygons", false, true, "Throw exception if polygon is invalid in function pointInPolygon by default instead of returning possibly wrong results"}}}, + {"19.18", {{"enable_scalar_subquery_optimization", false, true, "Prevent scalar subqueries from (de)serializing large scalar values and possibly avoid running the same subquery more than once"}}}, + {"19.14", {{"any_join_distinct_right_table_keys", true, false, "Disable ANY RIGHT and ANY FULL JOINs by default to avoid inconsistency"}}}, + {"19.12", {{"input_format_defaults_for_omitted_fields", false, true, "Enable calculation of complex default expressions for omitted fields for some input formats, because it should be the expected behaviour"}}}, + {"19.5", {{"max_partitions_per_insert_block", 0, 100, "Add a limit for the number of partitions in one block"}}}, + {"18.12.17", {{"enable_optimize_predicate_expression", 0, 1, "Optimize predicates to subqueries by default"}}}, +}; + + +const std::map & getSettingsChangesHistory() +{ + static std::map settings_changes_history; + + static std::once_flag initialized_flag; + std::call_once(initialized_flag, []() + { + for (const auto & setting_change : settings_changes_history_initializer) + { + /// Disallow duplicate keys in the settings changes history. Example: + /// {"21.2", {{"some_setting_1", false, true, "[...]"}}}, + /// [...] + /// {"21.2", {{"some_setting_2", false, true, "[...]"}}}, + /// As std::set has unique keys, one of the entries would be overwritten. + if (settings_changes_history.contains(setting_change.first)) + throw Exception{ErrorCodes::LOGICAL_ERROR, "Detected duplicate version '{}'", setting_change.first.toString()}; + + settings_changes_history[setting_change.first] = setting_change.second; + } + }); + + return settings_changes_history; +} +} diff --git a/src/Core/SettingsChangesHistory.h b/src/Core/SettingsChangesHistory.h index fba6386b9bd..b1a69c3b6d6 100644 --- a/src/Core/SettingsChangesHistory.h +++ b/src/Core/SettingsChangesHistory.h @@ -1,62 +1,25 @@ #pragma once #include -#include -#include -#include -#include #include +#include namespace DB { -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - class ClickHouseVersion { public: - ClickHouseVersion(const String & version) /// NOLINT(google-explicit-constructor) - { - Strings split; - boost::split(split, version, [](char c){ return c == '.'; }); - components.reserve(split.size()); - if (split.empty()) - throw Exception{ErrorCodes::BAD_ARGUMENTS, "Cannot parse ClickHouse version here: {}", version}; + /// NOLINTBEGIN(google-explicit-constructor) + ClickHouseVersion(const String & version); + ClickHouseVersion(const char * version); + /// NOLINTEND(google-explicit-constructor) - for (const auto & split_element : split) - { - size_t component; - ReadBufferFromString buf(split_element); - if (!tryReadIntText(component, buf) || !buf.eof()) - throw Exception{ErrorCodes::BAD_ARGUMENTS, "Cannot parse ClickHouse version here: {}", version}; - components.push_back(component); - } - } + String toString() const; - ClickHouseVersion(const char * version) : ClickHouseVersion(String(version)) {} /// NOLINT(google-explicit-constructor) - - String toString() const - { - String version = std::to_string(components[0]); - for (size_t i = 1; i < components.size(); ++i) - version += "." + std::to_string(components[i]); - - return version; - } - - bool operator<(const ClickHouseVersion & other) const - { - return components < other.components; - } - - bool operator>=(const ClickHouseVersion & other) const - { - return components >= other.components; - } + bool operator<(const ClickHouseVersion & other) const { return components < other.components; } + bool operator>=(const ClickHouseVersion & other) const { return components >= other.components; } private: std::vector components; @@ -75,253 +38,6 @@ namespace SettingsChangesHistory using SettingsChanges = std::vector; } -// clang-format off -/// History of settings changes that controls some backward incompatible changes -/// across all ClickHouse versions. It maps ClickHouse version to settings changes that were done -/// in this version. This history contains both changes to existing settings and newly added settings. -/// Settings changes is a vector of structs -/// {setting_name, previous_value, new_value, reason}. -/// For newly added setting choose the most appropriate previous_value (for example, if new setting -/// controls new feature and it's 'true' by default, use 'false' as previous_value). -/// It's used to implement `compatibility` setting (see https://github.com/ClickHouse/ClickHouse/issues/35972) -static const std::map settings_changes_history = -{ - {"24.7", {{"output_format_parquet_write_page_index", false, true, "Add a possibility to write page index into parquet files."}, - }}, - {"24.6", {{"materialize_skip_indexes_on_insert", true, true, "Added new setting to allow to disable materialization of skip indexes on insert"}, - {"materialize_statistics_on_insert", true, true, "Added new setting to allow to disable materialization of statistics on insert"}, - {"input_format_parquet_use_native_reader", false, false, "When reading Parquet files, to use native reader instead of arrow reader."}, - {"hdfs_throw_on_zero_files_match", false, false, "Allow to throw an error when ListObjects request cannot match any files in HDFS engine instead of empty query result"}, - {"azure_throw_on_zero_files_match", false, false, "Allow to throw an error when ListObjects request cannot match any files in AzureBlobStorage engine instead of empty query result"}, - {"s3_validate_request_settings", true, true, "Allow to disable S3 request settings validation"}, - {"allow_experimental_full_text_index", false, false, "Enable experimental full-text index"}, - {"azure_skip_empty_files", false, false, "Allow to skip empty files in azure table engine"}, - {"hdfs_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in HDFS table engine"}, - {"azure_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in AzureBlobStorage table engine"}, - {"s3_ignore_file_doesnt_exist", false, false, "Allow to return 0 rows when the requested files don't exist instead of throwing an exception in S3 table engine"}, - {"s3_max_part_number", 10000, 10000, "Maximum part number number for s3 upload part"}, - {"s3_max_single_operation_copy_size", 32 * 1024 * 1024, 32 * 1024 * 1024, "Maximum size for a single copy operation in s3"}, - {"input_format_parquet_max_block_size", 8192, DEFAULT_BLOCK_SIZE, "Increase block size for parquet reader."}, - {"input_format_parquet_prefer_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Average block bytes output by parquet reader."}, - {"enable_blob_storage_log", true, true, "Write information about blob storage operations to system.blob_storage_log table"}, - {"allow_deprecated_snowflake_conversion_functions", true, false, "Disabled deprecated functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake."}, - {"allow_statistic_optimize", false, false, "Old setting which popped up here being renamed."}, - {"allow_experimental_statistic", false, false, "Old setting which popped up here being renamed."}, - {"allow_statistics_optimize", false, false, "The setting was renamed. The previous name is `allow_statistic_optimize`."}, - {"allow_experimental_statistics", false, false, "The setting was renamed. The previous name is `allow_experimental_statistic`."}, - {"enable_vertical_final", false, true, "Enable vertical final by default again after fixing bug"}, - {"parallel_replicas_custom_key_range_lower", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards"}, - {"parallel_replicas_custom_key_range_upper", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards. A value of 0 disables the upper limit"}, - {"output_format_pretty_display_footer_column_names", 0, 1, "Add a setting to display column names in the footer if there are many rows. Threshold value is controlled by output_format_pretty_display_footer_column_names_min_rows."}, - {"output_format_pretty_display_footer_column_names_min_rows", 0, 50, "Add a setting to control the threshold value for setting output_format_pretty_display_footer_column_names_min_rows. Default 50."}, - {"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."}, - }}, - {"24.5", {{"allow_deprecated_error_prone_window_functions", true, false, "Allow usage of deprecated error prone window functions (neighbor, runningAccumulate, runningDifferenceStartingWithFirstValue, runningDifference)"}, - {"allow_experimental_join_condition", false, false, "Support join with inequal conditions which involve columns from both left and right table. e.g. t1.y < t2.y."}, - {"input_format_tsv_crlf_end_of_line", false, false, "Enables reading of CRLF line endings with TSV formats"}, - {"output_format_parquet_use_custom_encoder", false, true, "Enable custom Parquet encoder."}, - {"cross_join_min_rows_to_compress", 0, 10000000, "Minimal count of rows to compress block in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached."}, - {"cross_join_min_bytes_to_compress", 0, 1_GiB, "Minimal size of block to compress in CROSS JOIN. Zero value means - disable this threshold. This block is compressed when any of the two thresholds (by rows or by bytes) are reached."}, - {"http_max_chunk_size", 0, 0, "Internal limitation"}, - {"prefer_external_sort_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Prefer maximum block bytes for external sort, reduce the memory usage during merging."}, - {"input_format_force_null_for_omitted_fields", false, false, "Disable type-defaults for omitted fields when needed"}, - {"cast_string_to_dynamic_use_inference", false, false, "Add setting to allow converting String to Dynamic through parsing"}, - {"allow_experimental_dynamic_type", false, false, "Add new experimental Dynamic type"}, - {"azure_max_blocks_in_multipart_upload", 50000, 50000, "Maximum number of blocks in multipart upload for Azure."}, - }}, - {"24.4", {{"input_format_json_throw_on_bad_escape_sequence", true, true, "Allow to save JSON strings with bad escape sequences"}, - {"max_parsing_threads", 0, 0, "Add a separate setting to control number of threads in parallel parsing from files"}, - {"ignore_drop_queries_probability", 0, 0, "Allow to ignore drop queries in server with specified probability for testing purposes"}, - {"lightweight_deletes_sync", 2, 2, "The same as 'mutation_sync', but controls only execution of lightweight deletes"}, - {"query_cache_system_table_handling", "save", "throw", "The query cache no longer caches results of queries against system tables"}, - {"input_format_json_ignore_unnecessary_fields", false, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields"}, - {"input_format_hive_text_allow_variable_number_of_columns", false, true, "Ignore extra columns in Hive Text input (if file has more columns than expected) and treat missing fields in Hive Text input as default values."}, - {"allow_experimental_database_replicated", false, true, "Database engine Replicated is now in Beta stage"}, - {"temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds", (10 * 60 * 1000), (10 * 60 * 1000), "Wait time to lock cache for sapce reservation in temporary data in filesystem cache"}, - {"optimize_rewrite_sum_if_to_count_if", false, true, "Only available for the analyzer, where it works correctly"}, - {"azure_allow_parallel_part_upload", "true", "true", "Use multiple threads for azure multipart upload."}, - {"max_recursive_cte_evaluation_depth", DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, "Maximum limit on recursive CTE evaluation depth"}, - {"query_plan_convert_outer_join_to_inner_join", false, true, "Allow to convert OUTER JOIN to INNER JOIN if filter after JOIN always filters default values"}, - }}, - {"24.3", {{"s3_connect_timeout_ms", 1000, 1000, "Introduce new dedicated setting for s3 connection timeout"}, - {"allow_experimental_shared_merge_tree", false, true, "The setting is obsolete"}, - {"use_page_cache_for_disks_without_file_cache", false, false, "Added userspace page cache"}, - {"read_from_page_cache_if_exists_otherwise_bypass_cache", false, false, "Added userspace page cache"}, - {"page_cache_inject_eviction", false, false, "Added userspace page cache"}, - {"default_table_engine", "None", "MergeTree", "Set default table engine to MergeTree for better usability"}, - {"input_format_json_use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects", false, false, "Allow to use String type for ambiguous paths during named tuple inference from JSON objects"}, - {"traverse_shadow_remote_data_paths", false, false, "Traverse shadow directory when query system.remote_data_paths."}, - {"throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert", false, true, "Deduplication is dependent materialized view cannot work together with async inserts."}, - {"parallel_replicas_allow_in_with_subquery", false, true, "If true, subquery for IN will be executed on every follower replica"}, - {"log_processors_profiles", false, true, "Enable by default"}, - {"function_locate_has_mysql_compatible_argument_order", false, true, "Increase compatibility with MySQL's locate function."}, - {"allow_suspicious_primary_key", true, false, "Forbid suspicious PRIMARY KEY/ORDER BY for MergeTree (i.e. SimpleAggregateFunction)"}, - {"filesystem_cache_reserve_space_wait_lock_timeout_milliseconds", 1000, 1000, "Wait time to lock cache for sapce reservation in filesystem cache"}, - {"max_parser_backtracks", 0, 1000000, "Limiting the complexity of parsing"}, - {"analyzer_compatibility_join_using_top_level_identifier", false, false, "Force to resolve identifier in JOIN USING from projection"}, - {"distributed_insert_skip_read_only_replicas", false, false, "If true, INSERT into Distributed will skip read-only replicas"}, - {"keeper_max_retries", 10, 10, "Max retries for general keeper operations"}, - {"keeper_retry_initial_backoff_ms", 100, 100, "Initial backoff timeout for general keeper operations"}, - {"keeper_retry_max_backoff_ms", 5000, 5000, "Max backoff timeout for general keeper operations"}, - {"s3queue_allow_experimental_sharded_mode", false, false, "Enable experimental sharded mode of S3Queue table engine. It is experimental because it will be rewritten"}, - {"allow_experimental_analyzer", false, true, "Enable analyzer and planner by default."}, - {"merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability", 0.0, 0.0, "For testing of `PartsSplitter` - split read ranges into intersecting and non intersecting every time you read from MergeTree with the specified probability."}, - {"allow_get_client_http_header", false, false, "Introduced a new function."}, - {"output_format_pretty_row_numbers", false, true, "It is better for usability."}, - {"output_format_pretty_max_value_width_apply_for_single_value", true, false, "Single values in Pretty formats won't be cut."}, - {"output_format_parquet_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_orc_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_arrow_string_as_string", false, true, "ClickHouse allows arbitrary binary data in the String data type, which is typically UTF-8. Parquet/ORC/Arrow Strings only support UTF-8. That's why you can choose which Arrow's data type to use for the ClickHouse String data type - String or Binary. While Binary would be more correct and compatible, using String by default will correspond to user expectations in most cases."}, - {"output_format_parquet_compression_method", "lz4", "zstd", "Parquet/ORC/Arrow support many compression methods, including lz4 and zstd. ClickHouse supports each and every compression method. Some inferior tools, such as 'duckdb', lack support for the faster `lz4` compression method, that's why we set zstd by default."}, - {"output_format_orc_compression_method", "lz4", "zstd", "Parquet/ORC/Arrow support many compression methods, including lz4 and zstd. ClickHouse supports each and every compression method. Some inferior tools, such as 'duckdb', lack support for the faster `lz4` compression method, that's why we set zstd by default."}, - {"output_format_pretty_highlight_digit_groups", false, true, "If enabled and if output is a terminal, highlight every digit corresponding to the number of thousands, millions, etc. with underline."}, - {"geo_distance_returns_float64_on_float64_arguments", false, true, "Increase the default precision."}, - {"azure_max_inflight_parts_for_one_file", 20, 20, "The maximum number of a concurrent loaded parts in multipart upload request. 0 means unlimited."}, - {"azure_strict_upload_part_size", 0, 0, "The exact size of part to upload during multipart upload to Azure blob storage."}, - {"azure_min_upload_part_size", 16*1024*1024, 16*1024*1024, "The minimum size of part to upload during multipart upload to Azure blob storage."}, - {"azure_max_upload_part_size", 5ull*1024*1024*1024, 5ull*1024*1024*1024, "The maximum size of part to upload during multipart upload to Azure blob storage."}, - {"azure_upload_part_size_multiply_factor", 2, 2, "Multiply azure_min_upload_part_size by this factor each time azure_multiply_parts_count_threshold parts were uploaded from a single write to Azure blob storage."}, - {"azure_upload_part_size_multiply_parts_count_threshold", 500, 500, "Each time this number of parts was uploaded to Azure blob storage, azure_min_upload_part_size is multiplied by azure_upload_part_size_multiply_factor."}, - {"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."}, - {"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."}, - }}, - {"24.2", {{"allow_suspicious_variant_types", true, false, "Don't allow creating Variant type with suspicious variants by default"}, - {"validate_experimental_and_suspicious_types_inside_nested_types", false, true, "Validate usage of experimental and suspicious types inside nested types"}, - {"output_format_values_escape_quote_with_quote", false, false, "If true escape ' with '', otherwise quoted with \\'"}, - {"output_format_pretty_single_large_number_tip_threshold", 0, 1'000'000, "Print a readable number tip on the right side of the table if the block consists of a single number which exceeds this value (except 0)"}, - {"input_format_try_infer_exponent_floats", true, false, "Don't infer floats in exponential notation by default"}, - {"query_plan_optimize_prewhere", true, true, "Allow to push down filter to PREWHERE expression for supported storages"}, - {"async_insert_max_data_size", 1000000, 10485760, "The previous value appeared to be too small."}, - {"async_insert_poll_timeout_ms", 10, 10, "Timeout in milliseconds for polling data from asynchronous insert queue"}, - {"async_insert_use_adaptive_busy_timeout", false, true, "Use adaptive asynchronous insert timeout"}, - {"async_insert_busy_timeout_min_ms", 50, 50, "The minimum value of the asynchronous insert timeout in milliseconds; it also serves as the initial value, which may be increased later by the adaptive algorithm"}, - {"async_insert_busy_timeout_max_ms", 200, 200, "The minimum value of the asynchronous insert timeout in milliseconds; async_insert_busy_timeout_ms is aliased to async_insert_busy_timeout_max_ms"}, - {"async_insert_busy_timeout_increase_rate", 0.2, 0.2, "The exponential growth rate at which the adaptive asynchronous insert timeout increases"}, - {"async_insert_busy_timeout_decrease_rate", 0.2, 0.2, "The exponential growth rate at which the adaptive asynchronous insert timeout decreases"}, - {"format_template_row_format", "", "", "Template row format string can be set directly in query"}, - {"format_template_resultset_format", "", "", "Template result set format string can be set in query"}, - {"split_parts_ranges_into_intersecting_and_non_intersecting_final", true, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"}, - {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}, - {"azure_max_single_part_copy_size", 256*1024*1024, 256*1024*1024, "The maximum size of object to copy using single part copy to Azure blob storage."}, - {"min_external_table_block_size_rows", DEFAULT_INSERT_BLOCK_SIZE, DEFAULT_INSERT_BLOCK_SIZE, "Squash blocks passed to external table to specified size in rows, if blocks are not big enough"}, - {"min_external_table_block_size_bytes", DEFAULT_INSERT_BLOCK_SIZE * 256, DEFAULT_INSERT_BLOCK_SIZE * 256, "Squash blocks passed to external table to specified size in bytes, if blocks are not big enough."}, - {"parallel_replicas_prefer_local_join", true, true, "If true, and JOIN can be executed with parallel replicas algorithm, and all storages of right JOIN part are *MergeTree, local JOIN will be used instead of GLOBAL JOIN."}, - {"optimize_time_filter_with_preimage", true, true, "Optimize Date and DateTime predicates by converting functions into equivalent comparisons without conversions (e.g. toYear(col) = 2023 -> col >= '2023-01-01' AND col <= '2023-12-31')"}, - {"extract_key_value_pairs_max_pairs_per_row", 0, 0, "Max number of pairs that can be produced by the `extractKeyValuePairs` function. Used as a safeguard against consuming too much memory."}, - {"default_view_definer", "CURRENT_USER", "CURRENT_USER", "Allows to set default `DEFINER` option while creating a view"}, - {"default_materialized_view_sql_security", "DEFINER", "DEFINER", "Allows to set a default value for SQL SECURITY option when creating a materialized view"}, - {"default_normal_view_sql_security", "INVOKER", "INVOKER", "Allows to set default `SQL SECURITY` option while creating a normal view"}, - {"mysql_map_string_to_text_in_show_columns", false, true, "Reduce the configuration effort to connect ClickHouse with BI tools."}, - {"mysql_map_fixed_string_to_text_in_show_columns", false, true, "Reduce the configuration effort to connect ClickHouse with BI tools."}, - }}, - {"24.1", {{"print_pretty_type_names", false, true, "Better user experience."}, - {"input_format_json_read_bools_as_strings", false, true, "Allow to read bools as strings in JSON formats by default"}, - {"output_format_arrow_use_signed_indexes_for_dictionary", false, true, "Use signed indexes type for Arrow dictionaries by default as it's recommended"}, - {"allow_experimental_variant_type", false, false, "Add new experimental Variant type"}, - {"use_variant_as_common_type", false, false, "Allow to use Variant in if/multiIf if there is no common type"}, - {"output_format_arrow_use_64_bit_indexes_for_dictionary", false, false, "Allow to use 64 bit indexes type in Arrow dictionaries"}, - {"parallel_replicas_mark_segment_size", 128, 128, "Add new setting to control segment size in new parallel replicas coordinator implementation"}, - {"ignore_materialized_views_with_dropped_target_table", false, false, "Add new setting to allow to ignore materialized views with dropped target table"}, - {"output_format_compression_level", 3, 3, "Allow to change compression level in the query output"}, - {"output_format_compression_zstd_window_log", 0, 0, "Allow to change zstd window log in the query output when zstd compression is used"}, - {"enable_zstd_qat_codec", false, false, "Add new ZSTD_QAT codec"}, - {"enable_vertical_final", false, true, "Use vertical final by default"}, - {"output_format_arrow_use_64_bit_indexes_for_dictionary", false, false, "Allow to use 64 bit indexes type in Arrow dictionaries"}, - {"max_rows_in_set_to_optimize_join", 100000, 0, "Disable join optimization as it prevents from read in order optimization"}, - {"output_format_pretty_color", true, "auto", "Setting is changed to allow also for auto value, disabling ANSI escapes if output is not a tty"}, - {"function_visible_width_behavior", 0, 1, "We changed the default behavior of `visibleWidth` to be more precise"}, - {"max_estimated_execution_time", 0, 0, "Separate max_execution_time and max_estimated_execution_time"}, - {"iceberg_engine_ignore_schema_evolution", false, false, "Allow to ignore schema evolution in Iceberg table engine"}, - {"optimize_injective_functions_in_group_by", false, true, "Replace injective functions by it's arguments in GROUP BY section in analyzer"}, - {"update_insert_deduplication_token_in_dependent_materialized_views", false, false, "Allow to update insert deduplication token with table identifier during insert in dependent materialized views"}, - {"azure_max_unexpected_write_error_retries", 4, 4, "The maximum number of retries in case of unexpected errors during Azure blob storage write"}, - {"split_parts_ranges_into_intersecting_and_non_intersecting_final", false, true, "Allow to split parts ranges into intersecting and non intersecting during FINAL optimization"}, - {"split_intersecting_parts_ranges_into_layers_final", true, true, "Allow to split intersecting parts ranges into layers during FINAL optimization"}}}, - {"23.12", {{"allow_suspicious_ttl_expressions", true, false, "It is a new setting, and in previous versions the behavior was equivalent to allowing."}, - {"input_format_parquet_allow_missing_columns", false, true, "Allow missing columns in Parquet files by default"}, - {"input_format_orc_allow_missing_columns", false, true, "Allow missing columns in ORC files by default"}, - {"input_format_arrow_allow_missing_columns", false, true, "Allow missing columns in Arrow files by default"}}}, - {"23.9", {{"optimize_group_by_constant_keys", false, true, "Optimize group by constant keys by default"}, - {"input_format_json_try_infer_named_tuples_from_objects", false, true, "Try to infer named Tuples from JSON objects by default"}, - {"input_format_json_read_numbers_as_strings", false, true, "Allow to read numbers as strings in JSON formats by default"}, - {"input_format_json_read_arrays_as_strings", false, true, "Allow to read arrays as strings in JSON formats by default"}, - {"input_format_json_infer_incomplete_types_as_strings", false, true, "Allow to infer incomplete types as Strings in JSON formats by default"}, - {"input_format_json_try_infer_numbers_from_strings", true, false, "Don't infer numbers from strings in JSON formats by default to prevent possible parsing errors"}, - {"http_write_exception_in_output_format", false, true, "Output valid JSON/XML on exception in HTTP streaming."}}}, - {"23.8", {{"rewrite_count_distinct_if_with_count_distinct_implementation", false, true, "Rewrite countDistinctIf with count_distinct_implementation configuration"}}}, - {"23.7", {{"function_sleep_max_microseconds_per_block", 0, 3000000, "In previous versions, the maximum sleep time of 3 seconds was applied only for `sleep`, but not for `sleepEachRow` function. In the new version, we introduce this setting. If you set compatibility with the previous versions, we will disable the limit altogether."}}}, - {"23.6", {{"http_send_timeout", 180, 30, "3 minutes seems crazy long. Note that this is timeout for a single network write call, not for the whole upload operation."}, - {"http_receive_timeout", 180, 30, "See http_send_timeout."}}}, - {"23.5", {{"input_format_parquet_preserve_order", true, false, "Allow Parquet reader to reorder rows for better parallelism."}, - {"parallelize_output_from_storages", false, true, "Allow parallelism when executing queries that read from file/url/s3/etc. This may reorder rows."}, - {"use_with_fill_by_sorting_prefix", false, true, "Columns preceding WITH FILL columns in ORDER BY clause form sorting prefix. Rows with different values in sorting prefix are filled independently"}, - {"output_format_parquet_compliant_nested_types", false, true, "Change an internal field name in output Parquet file schema."}}}, - {"23.4", {{"allow_suspicious_indices", true, false, "If true, index can defined with identical expressions"}, - {"allow_nonconst_timezone_arguments", true, false, "Allow non-const timezone arguments in certain time-related functions like toTimeZone(), fromUnixTimestamp*(), snowflakeToDateTime*()."}, - {"connect_timeout_with_failover_ms", 50, 1000, "Increase default connect timeout because of async connect"}, - {"connect_timeout_with_failover_secure_ms", 100, 1000, "Increase default secure connect timeout because of async connect"}, - {"hedged_connection_timeout_ms", 100, 50, "Start new connection in hedged requests after 50 ms instead of 100 to correspond with previous connect timeout"}}}, - {"23.3", {{"output_format_parquet_version", "1.0", "2.latest", "Use latest Parquet format version for output format"}, - {"input_format_json_ignore_unknown_keys_in_named_tuple", false, true, "Improve parsing JSON objects as named tuples"}, - {"input_format_native_allow_types_conversion", false, true, "Allow types conversion in Native input forma"}, - {"output_format_arrow_compression_method", "none", "lz4_frame", "Use lz4 compression in Arrow output format by default"}, - {"output_format_parquet_compression_method", "snappy", "lz4", "Use lz4 compression in Parquet output format by default"}, - {"output_format_orc_compression_method", "none", "lz4_frame", "Use lz4 compression in ORC output format by default"}, - {"async_query_sending_for_remote", false, true, "Create connections and send query async across shards"}}}, - {"23.2", {{"output_format_parquet_fixed_string_as_fixed_byte_array", false, true, "Use Parquet FIXED_LENGTH_BYTE_ARRAY type for FixedString by default"}, - {"output_format_arrow_fixed_string_as_fixed_byte_array", false, true, "Use Arrow FIXED_SIZE_BINARY type for FixedString by default"}, - {"query_plan_remove_redundant_distinct", false, true, "Remove redundant Distinct step in query plan"}, - {"optimize_duplicate_order_by_and_distinct", true, false, "Remove duplicate ORDER BY and DISTINCT if it's possible"}, - {"insert_keeper_max_retries", 0, 20, "Enable reconnections to Keeper on INSERT, improve reliability"}}}, - {"23.1", {{"input_format_json_read_objects_as_strings", 0, 1, "Enable reading nested json objects as strings while object type is experimental"}, - {"input_format_json_defaults_for_missing_elements_in_named_tuple", false, true, "Allow missing elements in JSON objects while reading named tuples by default"}, - {"input_format_csv_detect_header", false, true, "Detect header in CSV format by default"}, - {"input_format_tsv_detect_header", false, true, "Detect header in TSV format by default"}, - {"input_format_custom_detect_header", false, true, "Detect header in CustomSeparated format by default"}, - {"query_plan_remove_redundant_sorting", false, true, "Remove redundant sorting in query plan. For example, sorting steps related to ORDER BY clauses in subqueries"}}}, - {"22.12", {{"max_size_to_preallocate_for_aggregation", 10'000'000, 100'000'000, "This optimizes performance"}, - {"query_plan_aggregation_in_order", 0, 1, "Enable some refactoring around query plan"}, - {"format_binary_max_string_size", 0, 1_GiB, "Prevent allocating large amount of memory"}}}, - {"22.11", {{"use_structure_from_insertion_table_in_table_functions", 0, 2, "Improve using structure from insertion table in table functions"}}}, - {"23.4", {{"formatdatetime_f_prints_single_zero", true, false, "Improved compatibility with MySQL DATE_FORMAT()/STR_TO_DATE()"}}}, - {"23.4", {{"formatdatetime_parsedatetime_m_is_month_name", false, true, "Improved compatibility with MySQL DATE_FORMAT/STR_TO_DATE"}}}, - {"23.11", {{"parsedatetime_parse_without_leading_zeros", false, true, "Improved compatibility with MySQL DATE_FORMAT/STR_TO_DATE"}}}, - {"22.9", {{"force_grouping_standard_compatibility", false, true, "Make GROUPING function output the same as in SQL standard and other DBMS"}}}, - {"22.7", {{"cross_to_inner_join_rewrite", 1, 2, "Force rewrite comma join to inner"}, - {"enable_positional_arguments", false, true, "Enable positional arguments feature by default"}, - {"format_csv_allow_single_quotes", true, false, "Most tools don't treat single quote in CSV specially, don't do it by default too"}}}, - {"22.6", {{"output_format_json_named_tuples_as_objects", false, true, "Allow to serialize named tuples as JSON objects in JSON formats by default"}, - {"input_format_skip_unknown_fields", false, true, "Optimize reading subset of columns for some input formats"}}}, - {"22.5", {{"memory_overcommit_ratio_denominator", 0, 1073741824, "Enable memory overcommit feature by default"}, - {"memory_overcommit_ratio_denominator_for_user", 0, 1073741824, "Enable memory overcommit feature by default"}}}, - {"22.4", {{"allow_settings_after_format_in_insert", true, false, "Do not allow SETTINGS after FORMAT for INSERT queries because ClickHouse interpret SETTINGS as some values, which is misleading"}}}, - {"22.3", {{"cast_ipv4_ipv6_default_on_conversion_error", true, false, "Make functions cast(value, 'IPv4') and cast(value, 'IPv6') behave same as toIPv4 and toIPv6 functions"}}}, - {"21.12", {{"stream_like_engine_allow_direct_select", true, false, "Do not allow direct select for Kafka/RabbitMQ/FileLog by default"}}}, - {"21.9", {{"output_format_decimal_trailing_zeros", true, false, "Do not output trailing zeros in text representation of Decimal types by default for better looking output"}, - {"use_hedged_requests", false, true, "Enable Hedged Requests feature by default"}}}, - {"21.7", {{"legacy_column_name_of_tuple_literal", true, false, "Add this setting only for compatibility reasons. It makes sense to set to 'true', while doing rolling update of cluster from version lower than 21.7 to higher"}}}, - {"21.5", {{"async_socket_for_remote", false, true, "Fix all problems and turn on asynchronous reads from socket for remote queries by default again"}}}, - {"21.3", {{"async_socket_for_remote", true, false, "Turn off asynchronous reads from socket for remote queries because of some problems"}, - {"optimize_normalize_count_variants", false, true, "Rewrite aggregate functions that semantically equals to count() as count() by default"}, - {"normalize_function_names", false, true, "Normalize function names to their canonical names, this was needed for projection query routing"}}}, - {"21.2", {{"enable_global_with_statement", false, true, "Propagate WITH statements to UNION queries and all subqueries by default"}}}, - {"21.1", {{"insert_quorum_parallel", false, true, "Use parallel quorum inserts by default. It is significantly more convenient to use than sequential quorum inserts"}, - {"input_format_null_as_default", false, true, "Allow to insert NULL as default for input formats by default"}, - {"optimize_on_insert", false, true, "Enable data optimization on INSERT by default for better user experience"}, - {"use_compact_format_in_distributed_parts_names", false, true, "Use compact format for async INSERT into Distributed tables by default"}}}, - {"20.10", {{"format_regexp_escaping_rule", "Escaped", "Raw", "Use Raw as default escaping rule for Regexp format to male the behaviour more like to what users expect"}}}, - {"20.7", {{"show_table_uuid_in_table_create_query_if_not_nil", true, false, "Stop showing UID of the table in its CREATE query for Engine=Atomic"}}}, - {"20.5", {{"input_format_with_names_use_header", false, true, "Enable using header with names for formats with WithNames/WithNamesAndTypes suffixes"}, - {"allow_suspicious_codecs", true, false, "Don't allow to specify meaningless compression codecs"}}}, - {"20.4", {{"validate_polygons", false, true, "Throw exception if polygon is invalid in function pointInPolygon by default instead of returning possibly wrong results"}}}, - {"19.18", {{"enable_scalar_subquery_optimization", false, true, "Prevent scalar subqueries from (de)serializing large scalar values and possibly avoid running the same subquery more than once"}}}, - {"19.14", {{"any_join_distinct_right_table_keys", true, false, "Disable ANY RIGHT and ANY FULL JOINs by default to avoid inconsistency"}}}, - {"19.12", {{"input_format_defaults_for_omitted_fields", false, true, "Enable calculation of complex default expressions for omitted fields for some input formats, because it should be the expected behaviour"}}}, - {"19.5", {{"max_partitions_per_insert_block", 0, 100, "Add a limit for the number of partitions in one block"}}}, - {"18.12.17", {{"enable_optimize_predicate_expression", 0, 1, "Optimize predicates to subqueries by default"}}}, -}; +const std::map & getSettingsChangesHistory(); } diff --git a/src/Core/SettingsEnums.cpp b/src/Core/SettingsEnums.cpp index 05985316566..82e7d6db410 100644 --- a/src/Core/SettingsEnums.cpp +++ b/src/Core/SettingsEnums.cpp @@ -173,6 +173,10 @@ IMPLEMENT_SETTING_ENUM(ParallelReplicasCustomKeyFilterType, ErrorCodes::BAD_ARGU {{"default", ParallelReplicasCustomKeyFilterType::DEFAULT}, {"range", ParallelReplicasCustomKeyFilterType::RANGE}}) +IMPLEMENT_SETTING_ENUM(LightweightMutationProjectionMode, ErrorCodes::BAD_ARGUMENTS, + {{"throw", LightweightMutationProjectionMode::THROW}, + {"drop", LightweightMutationProjectionMode::DROP}}) + IMPLEMENT_SETTING_AUTO_ENUM(LocalFSReadMethod, ErrorCodes::BAD_ARGUMENTS) IMPLEMENT_SETTING_ENUM(ParquetVersion, ErrorCodes::BAD_ARGUMENTS, @@ -201,13 +205,13 @@ IMPLEMENT_SETTING_ENUM(ORCCompression, ErrorCodes::BAD_ARGUMENTS, {"zlib", FormatSettings::ORCCompression::ZLIB}, {"lz4", FormatSettings::ORCCompression::LZ4}}) -IMPLEMENT_SETTING_ENUM(S3QueueMode, ErrorCodes::BAD_ARGUMENTS, - {{"ordered", S3QueueMode::ORDERED}, - {"unordered", S3QueueMode::UNORDERED}}) +IMPLEMENT_SETTING_ENUM(ObjectStorageQueueMode, ErrorCodes::BAD_ARGUMENTS, + {{"ordered", ObjectStorageQueueMode::ORDERED}, + {"unordered", ObjectStorageQueueMode::UNORDERED}}) -IMPLEMENT_SETTING_ENUM(S3QueueAction, ErrorCodes::BAD_ARGUMENTS, - {{"keep", S3QueueAction::KEEP}, - {"delete", S3QueueAction::DELETE}}) +IMPLEMENT_SETTING_ENUM(ObjectStorageQueueAction, ErrorCodes::BAD_ARGUMENTS, + {{"keep", ObjectStorageQueueAction::KEEP}, + {"delete", ObjectStorageQueueAction::DELETE}}) IMPLEMENT_SETTING_ENUM(ExternalCommandStderrReaction, ErrorCodes::BAD_ARGUMENTS, {{"none", ExternalCommandStderrReaction::NONE}, diff --git a/src/Core/SettingsEnums.h b/src/Core/SettingsEnums.h index 575cd8700c8..f6d9593ca56 100644 --- a/src/Core/SettingsEnums.h +++ b/src/Core/SettingsEnums.h @@ -1,8 +1,12 @@ #pragma once #include +#include #include +#include +#include #include +#include #include #include #include @@ -115,25 +119,6 @@ constexpr auto getEnumValues(); return getEnumValues().size();\ } -enum class LoadBalancing : uint8_t -{ - /// among replicas with a minimum number of errors selected randomly - RANDOM = 0, - /// a replica is selected among the replicas with the minimum number of errors - /// with the minimum number of distinguished characters in the replica name prefix and local hostname prefix - NEAREST_HOSTNAME, - /// just like NEAREST_HOSTNAME, but it count distinguished characters in a levenshtein distance manner - HOSTNAME_LEVENSHTEIN_DISTANCE, - // replicas with the same number of errors are accessed in the same order - // as they are specified in the configuration. - IN_ORDER, - /// if first replica one has higher number of errors, - /// pick a random one from replicas with minimum number of errors - FIRST_OR_RANDOM, - // round robin across replicas with the same number of errors. - ROUND_ROBIN, -}; - DECLARE_SETTING_ENUM(LoadBalancing) DECLARE_SETTING_ENUM(JoinStrictness) @@ -204,16 +189,6 @@ DECLARE_SETTING_ENUM_WITH_RENAME(ParquetVersion, FormatSettings::ParquetVersion) DECLARE_SETTING_ENUM(LogsLevel) - -// Make it signed for compatibility with DataTypeEnum8 -enum QueryLogElementType : int8_t -{ - QUERY_START = 1, - QUERY_FINISH = 2, - EXCEPTION_BEFORE_START = 3, - EXCEPTION_WHILE_PROCESSING = 4, -}; - DECLARE_SETTING_ENUM_WITH_RENAME(LogQueriesType, QueryLogElementType) @@ -292,13 +267,6 @@ enum class StreamingHandleErrorMode : uint8_t DECLARE_SETTING_ENUM(StreamingHandleErrorMode) -enum class ShortCircuitFunctionEvaluation : uint8_t -{ - ENABLE, // Use short-circuit function evaluation for functions that are suitable for it. - FORCE_ENABLE, // Use short-circuit function evaluation for all functions. - DISABLE, // Disable short-circuit function evaluation. -}; - DECLARE_SETTING_ENUM(ShortCircuitFunctionEvaluation) enum class TransactionsWaitCSNMode : uint8_t @@ -339,32 +307,34 @@ enum class ParallelReplicasCustomKeyFilterType : uint8_t DECLARE_SETTING_ENUM(ParallelReplicasCustomKeyFilterType) +enum class LightweightMutationProjectionMode : uint8_t +{ + THROW, + DROP, +}; + +DECLARE_SETTING_ENUM(LightweightMutationProjectionMode) + DECLARE_SETTING_ENUM(LocalFSReadMethod) -enum class S3QueueMode : uint8_t +enum class ObjectStorageQueueMode : uint8_t { ORDERED, UNORDERED, }; -DECLARE_SETTING_ENUM(S3QueueMode) +DECLARE_SETTING_ENUM(ObjectStorageQueueMode) -enum class S3QueueAction : uint8_t +enum class ObjectStorageQueueAction : uint8_t { KEEP, DELETE, }; -DECLARE_SETTING_ENUM(S3QueueAction) +DECLARE_SETTING_ENUM(ObjectStorageQueueAction) DECLARE_SETTING_ENUM(ExternalCommandStderrReaction) -enum class SchemaInferenceMode : uint8_t -{ - DEFAULT, - UNION, -}; - DECLARE_SETTING_ENUM(SchemaInferenceMode) DECLARE_SETTING_ENUM_WITH_RENAME(DateTimeOverflowBehavior, FormatSettings::DateTimeOverflowBehavior) diff --git a/src/Core/SettingsFields.cpp b/src/Core/SettingsFields.cpp index caa8b3fdffd..7d094e2a107 100644 --- a/src/Core/SettingsFields.cpp +++ b/src/Core/SettingsFields.cpp @@ -380,15 +380,6 @@ void SettingFieldString::readBinary(ReadBuffer & in) *this = std::move(str); } -/// Unbeautiful workaround for clickhouse-keeper standalone build ("-DBUILD_STANDALONE_KEEPER=1"). -/// In this build, we don't build and link library dbms (to which SettingsField.cpp belongs) but -/// only build SettingsField.cpp. Further dependencies, e.g. DataTypeString and DataTypeMap below, -/// require building of further files for clickhouse-keeper. To keep dependencies slim, we don't do -/// that. The linker does not complain only because clickhouse-keeper does not call any of below -/// functions. A cleaner alternative would be more modular libraries, e.g. one for data types, which -/// could then be linked by the server and the linker. -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD - SettingFieldMap::SettingFieldMap(const Field & f) : value(fieldToMap(f)) {} String SettingFieldMap::toString() const @@ -428,42 +419,6 @@ void SettingFieldMap::readBinary(ReadBuffer & in) *this = map; } -#else - -namespace ErrorCodes -{ - extern const int LOGICAL_ERROR; -} - -SettingFieldMap::SettingFieldMap(const Field &) : value(Map()) {} -String SettingFieldMap::toString() const -{ - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Setting of type Map not supported"); -} - - -SettingFieldMap & SettingFieldMap::operator =(const Field &) -{ - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Setting of type Map not supported"); -} - -void SettingFieldMap::parseFromString(const String &) -{ - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Setting of type Map not supported"); -} - -void SettingFieldMap::writeBinary(WriteBuffer &) const -{ - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Setting of type Map not supported"); -} - -void SettingFieldMap::readBinary(ReadBuffer &) -{ - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Setting of type Map not supported"); -} - -#endif - namespace { char stringToChar(const String & str) diff --git a/src/Core/SettingsFields.h b/src/Core/SettingsFields.h index 19809348921..266141815e3 100644 --- a/src/Core/SettingsFields.h +++ b/src/Core/SettingsFields.h @@ -247,12 +247,6 @@ struct SettingFieldString void readBinary(ReadBuffer & in); }; -#ifdef CLICKHOUSE_KEEPER_STANDALONE_BUILD -#define NORETURN [[noreturn]] -#else -#define NORETURN -#endif - struct SettingFieldMap { public: @@ -269,11 +263,11 @@ public: operator const Map &() const { return value; } /// NOLINT explicit operator Field() const { return value; } - NORETURN String toString() const; - NORETURN void parseFromString(const String & str); + String toString() const; + void parseFromString(const String & str); - NORETURN void writeBinary(WriteBuffer & out) const; - NORETURN void readBinary(ReadBuffer & in); + void writeBinary(WriteBuffer & out) const; + void readBinary(ReadBuffer & in); }; #undef NORETURN diff --git a/src/Core/ShortCircuitFunctionEvaluation.h b/src/Core/ShortCircuitFunctionEvaluation.h new file mode 100644 index 00000000000..3785efe8015 --- /dev/null +++ b/src/Core/ShortCircuitFunctionEvaluation.h @@ -0,0 +1,15 @@ +#pragma once + +#include + +namespace DB +{ + +enum class ShortCircuitFunctionEvaluation : uint8_t +{ + ENABLE, // Use short-circuit function evaluation for functions that are suitable for it. + FORCE_ENABLE, // Use short-circuit function evaluation for all functions. + DISABLE, // Disable short-circuit function evaluation. +}; + +} diff --git a/src/Core/SortDescription.h b/src/Core/SortDescription.h index 33fd6017599..5c6f3e3150a 100644 --- a/src/Core/SortDescription.h +++ b/src/Core/SortDescription.h @@ -1,15 +1,16 @@ #pragma once -#include -#include -#include -#include + #include -#include #include #include #include +#include +#include +#include +#include + namespace DB { diff --git a/src/Core/examples/CMakeLists.txt b/src/Core/examples/CMakeLists.txt index f30ee25491f..97e5e8e67e6 100644 --- a/src/Core/examples/CMakeLists.txt +++ b/src/Core/examples/CMakeLists.txt @@ -1,8 +1,8 @@ clickhouse_add_executable (string_pool string_pool.cpp) -target_link_libraries (string_pool PRIVATE clickhouse_common_io ch_contrib::sparsehash) +target_link_libraries (string_pool PRIVATE clickhouse_common_io clickhouse_common_config ch_contrib::sparsehash) clickhouse_add_executable (field field.cpp) target_link_libraries (field PRIVATE dbms) clickhouse_add_executable (string_ref_hash string_ref_hash.cpp) -target_link_libraries (string_ref_hash PRIVATE clickhouse_common_io) +target_link_libraries (string_ref_hash PRIVATE clickhouse_common_io clickhouse_common_config) diff --git a/src/Core/tests/gtest_fields_binary_enciding.cpp b/src/Core/tests/gtest_fields_binary_enciding.cpp new file mode 100644 index 00000000000..087caf746bb --- /dev/null +++ b/src/Core/tests/gtest_fields_binary_enciding.cpp @@ -0,0 +1,65 @@ +#include +#include +#include +#include + +using namespace DB; + +namespace DB::ErrorCodes +{ + extern const int UNSUPPORTED_METHOD; +} + + +void check(const Field & field) +{ +// std::cerr << "Check " << toString(field) << "\n"; + WriteBufferFromOwnString ostr; + encodeField(field, ostr); + ReadBufferFromString istr(ostr.str()); + Field decoded_field = decodeField(istr); + ASSERT_TRUE(istr.eof()); + ASSERT_EQ(field, decoded_field); +} + +GTEST_TEST(FieldBinaryEncoding, EncodeAndDecode) +{ + check(Null()); + check(POSITIVE_INFINITY); + check(NEGATIVE_INFINITY); + check(true); + check(UInt64(42)); + check(Int64(-42)); + check(UInt128(42)); + check(Int128(-42)); + check(UInt256(42)); + check(Int256(-42)); + check(UUID(42)); + check(IPv4(42)); + check(IPv6(42)); + check(Float64(42.42)); + check(String("Hello, World!")); + check(Array({Field(UInt64(42)), Field(UInt64(43))})); + check(Tuple({Field(UInt64(42)), Field(Null()), Field(UUID(42)), Field(String("Hello, World!"))})); + check(Map({Tuple{Field(UInt64(42)), Field(String("str_42"))}, Tuple{Field(UInt64(43)), Field(String("str_43"))}})); + check(Object({{String("key_1"), Field(UInt64(42))}, {String("key_2"), Field(UInt64(43))}})); + check(DecimalField(4242, 3)); + check(DecimalField(4242, 3)); + check(DecimalField(Int128(4242), 3)); + check(DecimalField(Int256(4242), 3)); + check(AggregateFunctionStateData{.name="some_name", .data="some_data"}); + try + { + check(CustomType()); + } + catch (const Exception & e) + { + ASSERT_EQ(e.code(), ErrorCodes::UNSUPPORTED_METHOD); + } + + check(Array({ + Tuple({Field(UInt64(42)), Map({Tuple{Field(UInt64(42)), Field(String("str_42"))}, Tuple{Field(UInt64(43)), Field(String("str_43"))}}), Field(UUID(42)), Field(String("Hello, World!"))}), + Tuple({Field(UInt64(43)), Map({Tuple{Field(UInt64(43)), Field(String("str_43"))}, Tuple{Field(UInt64(44)), Field(String("str_44"))}}), Field(UUID(43)), Field(String("Hello, World 2!"))}) + })); +} + diff --git a/src/Daemon/BaseDaemon.cpp b/src/Daemon/BaseDaemon.cpp index b2c425ceb79..e7ae8ea5a1d 100644 --- a/src/Daemon/BaseDaemon.cpp +++ b/src/Daemon/BaseDaemon.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include #include @@ -19,7 +20,6 @@ #endif #include #include -#include #include #include @@ -36,6 +36,7 @@ #include #include +#include #include #include #include @@ -53,7 +54,6 @@ #include #include #include -#include #include #include #include @@ -78,8 +78,6 @@ namespace DB { namespace ErrorCodes { - extern const int CANNOT_SET_SIGNAL_HANDLER; - extern const int CANNOT_SEND_SIGNAL; extern const int SYSTEM_ERROR; extern const int LOGICAL_ERROR; } @@ -87,114 +85,6 @@ namespace DB using namespace DB; -PipeFDs signal_pipe; - - -/** Reset signal handler to the default and send signal to itself. - * It's called from user signal handler to write core dump. - */ -static void call_default_signal_handler(int sig) -{ - if (SIG_ERR == signal(sig, SIG_DFL)) - throw ErrnoException(ErrorCodes::CANNOT_SET_SIGNAL_HANDLER, "Cannot set signal handler"); - - if (0 != raise(sig)) - throw ErrnoException(ErrorCodes::CANNOT_SEND_SIGNAL, "Cannot send signal"); -} - -static const size_t signal_pipe_buf_size = - sizeof(int) - + sizeof(siginfo_t) - + sizeof(ucontext_t*) - + sizeof(StackTrace) - + sizeof(UInt64) - + sizeof(UInt32) - + sizeof(void*); - -using signal_function = void(int, siginfo_t*, void*); - -static void writeSignalIDtoSignalPipe(int sig) -{ - auto saved_errno = errno; /// We must restore previous value of errno in signal handler. - - char buf[signal_pipe_buf_size]; - WriteBufferFromFileDescriptor out(signal_pipe.fds_rw[1], signal_pipe_buf_size, buf); - writeBinary(sig, out); - out.next(); - - errno = saved_errno; -} - -/** Signal handler for HUP */ -static void closeLogsSignalHandler(int sig, siginfo_t *, void *) -{ - DENY_ALLOCATIONS_IN_SCOPE; - writeSignalIDtoSignalPipe(sig); -} - -static void terminateRequestedSignalHandler(int sig, siginfo_t *, void *) -{ - DENY_ALLOCATIONS_IN_SCOPE; - writeSignalIDtoSignalPipe(sig); -} - - -static std::atomic_flag fatal_error_printed; - -/** Handler for "fault" or diagnostic signals. Send data about fault to separate thread to write into log. - */ -static void signalHandler(int sig, siginfo_t * info, void * context) -{ - if (asynchronous_stack_unwinding && sig == SIGSEGV) - siglongjmp(asynchronous_stack_unwinding_signal_jump_buffer, 1); - - DENY_ALLOCATIONS_IN_SCOPE; - auto saved_errno = errno; /// We must restore previous value of errno in signal handler. - - char buf[signal_pipe_buf_size]; - WriteBufferFromFileDescriptorDiscardOnFailure out(signal_pipe.fds_rw[1], signal_pipe_buf_size, buf); - - const ucontext_t * signal_context = reinterpret_cast(context); - const StackTrace stack_trace(*signal_context); - -#if USE_GWP_ASAN - if (const auto fault_address = reinterpret_cast(info->si_addr); - GWPAsan::isGWPAsanError(fault_address)) - GWPAsan::printReport(fault_address); -#endif - - writeBinary(sig, out); - writePODBinary(*info, out); - writePODBinary(signal_context, out); - writePODBinary(stack_trace, out); - writeVectorBinary(Exception::enable_job_stack_trace ? Exception::thread_frame_pointers : std::vector{}, out); - writeBinary(static_cast(getThreadId()), out); - writePODBinary(current_thread, out); - - out.next(); - - if (sig != SIGTSTP) /// This signal is used for debugging. - { - /// The time that is usually enough for separate thread to print info into log. - /// Under MSan full stack unwinding with DWARF info about inline functions takes 101 seconds in one case. - for (size_t i = 0; i < 300; ++i) - { - /// We will synchronize with the thread printing the messages with an atomic variable to finish earlier. - if (fatal_error_printed.test()) - break; - - /// This coarse method of synchronization is perfectly ok for fatal signals. - sleepForSeconds(1); - } - - /// Wait for all logs flush operations - sleepForSeconds(3); - call_default_signal_handler(sig); - } - - errno = saved_errno; -} - static bool getenvBool(const char * name) { @@ -206,450 +96,6 @@ static bool getenvBool(const char * name) } -/// Avoid link time dependency on DB/Interpreters - will use this function only when linked. -__attribute__((__weak__)) void collectCrashLog( - Int32 signal, UInt64 thread_id, const String & query_id, const StackTrace & stack_trace); - - -/** The thread that read info about signal or std::terminate from pipe. - * On HUP, close log files (for new files to be opened later). - * On information about std::terminate, write it to log. - * On other signals, write info to log. - */ -class SignalListener : public Poco::Runnable -{ -public: - static constexpr int StdTerminate = -1; - static constexpr int StopThread = -2; - static constexpr int SanitizerTrap = -3; - - explicit SignalListener(BaseDaemon & daemon_) - : log(getLogger("BaseDaemon")) - , daemon(daemon_) - { - } - - void run() override - { - static_assert(PIPE_BUF >= 512); - static_assert(signal_pipe_buf_size <= PIPE_BUF, "Only write of PIPE_BUF to pipe is atomic and the minimal known PIPE_BUF across supported platforms is 512"); - char buf[signal_pipe_buf_size]; - ReadBufferFromFileDescriptor in(signal_pipe.fds_rw[0], signal_pipe_buf_size, buf); - - while (!in.eof()) - { - int sig = 0; - readBinary(sig, in); - // We may log some specific signals afterwards, with different log - // levels and more info, but for completeness we log all signals - // here at trace level. - // Don't use strsignal here, because it's not thread-safe. - LOG_TRACE(log, "Received signal {}", sig); - - if (sig == StopThread) - { - LOG_INFO(log, "Stop SignalListener thread"); - break; - } - else if (sig == SIGHUP) - { - LOG_DEBUG(log, "Received signal to close logs."); - BaseDaemon::instance().closeLogs(BaseDaemon::instance().logger()); - LOG_INFO(log, "Opened new log file after received signal."); - } - else if (sig == StdTerminate) - { - UInt32 thread_num; - std::string message; - - readBinary(thread_num, in); - readBinary(message, in); - - onTerminate(message, thread_num); - } - else if (sig == SIGINT || - sig == SIGQUIT || - sig == SIGTERM) - { - daemon.handleSignal(sig); - } - else - { - siginfo_t info{}; - ucontext_t * context{}; - StackTrace stack_trace(NoCapture{}); - std::vector thread_frame_pointers; - UInt32 thread_num{}; - ThreadStatus * thread_ptr{}; - - if (sig != SanitizerTrap) - { - readPODBinary(info, in); - readPODBinary(context, in); - } - - readPODBinary(stack_trace, in); - - if (sig != SanitizerTrap) - readVectorBinary(thread_frame_pointers, in); - - readBinary(thread_num, in); - readPODBinary(thread_ptr, in); - - /// This allows to receive more signals if failure happens inside onFault function. - /// Example: segfault while symbolizing stack trace. - try - { - std::thread([=, this] { onFault(sig, info, context, stack_trace, thread_frame_pointers, thread_num, thread_ptr); }).detach(); - } - catch (...) - { - /// Likely cannot allocate thread - onFault(sig, info, context, stack_trace, thread_frame_pointers, thread_num, thread_ptr); - } - } - } - } - -private: - LoggerPtr log; - BaseDaemon & daemon; - - void onTerminate(std::string_view message, UInt32 thread_num) const - { - size_t pos = message.find('\n'); - - LOG_FATAL(log, "(version {}{}, build id: {}, git hash: {}) (from thread {}) {}", - VERSION_STRING, VERSION_OFFICIAL, daemon.build_id, daemon.git_hash, thread_num, message.substr(0, pos)); - - /// Print trace from std::terminate exception line-by-line to make it easy for grep. - while (pos != std::string_view::npos) - { - ++pos; - size_t next_pos = message.find('\n', pos); - size_t size = next_pos; - if (next_pos != std::string_view::npos) - size = next_pos - pos; - - LOG_FATAL(log, fmt::runtime(message.substr(pos, size))); - pos = next_pos; - } - } - - void onFault( - int sig, - const siginfo_t & info, - ucontext_t * context, - const StackTrace & stack_trace, - const std::vector & thread_frame_pointers, - UInt32 thread_num, - ThreadStatus * thread_ptr) const - try - { - ThreadStatus thread_status; - - /// First log those fields that are safe to access and that should not cause new fault. - /// That way we will have some duplicated info in the log but we don't loose important info - /// in case of double fault. - - LOG_FATAL(log, "########## Short fault info ############"); - LOG_FATAL(log, "(version {}{}, build id: {}, git hash: {}) (from thread {}) Received signal {}", - VERSION_STRING, VERSION_OFFICIAL, daemon.build_id, daemon.git_hash, - thread_num, sig); - - std::string signal_description = "Unknown signal"; - - /// Some of these are not really signals, but our own indications on failure reason. - if (sig == StdTerminate) - signal_description = "std::terminate"; - else if (sig == SanitizerTrap) - signal_description = "sanitizer trap"; - else if (sig >= 0) - signal_description = strsignal(sig); // NOLINT(concurrency-mt-unsafe) // it is not thread-safe but ok in this context - - LOG_FATAL(log, "Signal description: {}", signal_description); - - String error_message; - - if (sig != SanitizerTrap) - error_message = signalToErrorMessage(sig, info, *context); - else - error_message = "Sanitizer trap."; - - LOG_FATAL(log, fmt::runtime(error_message)); - - String bare_stacktrace_str; - if (stack_trace.getSize()) - { - /// Write bare stack trace (addresses) just in case if we will fail to print symbolized stack trace. - /// NOTE: This still require memory allocations and mutex lock inside logger. - /// BTW we can also print it to stderr using write syscalls. - - WriteBufferFromOwnString bare_stacktrace; - writeString("Stack trace:", bare_stacktrace); - for (size_t i = stack_trace.getOffset(); i < stack_trace.getSize(); ++i) - { - writeChar(' ', bare_stacktrace); - writePointerHex(stack_trace.getFramePointers()[i], bare_stacktrace); - } - - LOG_FATAL(log, fmt::runtime(bare_stacktrace.str())); - bare_stacktrace_str = bare_stacktrace.str(); - } - - /// Now try to access potentially unsafe data in thread_ptr. - - String query_id; - String query; - - /// Send logs from this thread to client if possible. - /// It will allow client to see failure messages directly. - if (thread_ptr) - { - query_id = thread_ptr->getQueryId(); - query = thread_ptr->getQueryForLog(); - - if (auto logs_queue = thread_ptr->getInternalTextLogsQueue()) - { - CurrentThread::attachInternalTextLogsQueue(logs_queue, LogsLevel::trace); - } - } - - LOG_FATAL(log, "########################################"); - - if (query_id.empty()) - { - LOG_FATAL(log, "(version {}{}, build id: {}, git hash: {}) (from thread {}) (no query) Received signal {} ({})", - VERSION_STRING, VERSION_OFFICIAL, daemon.build_id, daemon.git_hash, - thread_num, signal_description, sig); - } - else - { - LOG_FATAL(log, "(version {}{}, build id: {}, git hash: {}) (from thread {}) (query_id: {}) (query: {}) Received signal {} ({})", - VERSION_STRING, VERSION_OFFICIAL, daemon.build_id, daemon.git_hash, - thread_num, query_id, query, signal_description, sig); - } - - LOG_FATAL(log, fmt::runtime(error_message)); - - if (!bare_stacktrace_str.empty()) - { - LOG_FATAL(log, fmt::runtime(bare_stacktrace_str)); - } - - /// Write symbolized stack trace line by line for better grep-ability. - stack_trace.toStringEveryLine([&](std::string_view s) { LOG_FATAL(log, fmt::runtime(s)); }); - - /// In case it's a scheduled job write all previous jobs origins call stacks - std::for_each(thread_frame_pointers.rbegin(), thread_frame_pointers.rend(), - [this](const StackTrace::FramePointers & frame_pointers) - { - if (size_t size = std::ranges::find(frame_pointers, nullptr) - frame_pointers.begin()) - { - LOG_FATAL(log, "========================================"); - WriteBufferFromOwnString bare_stacktrace; - writeString("Job's origin stack trace:", bare_stacktrace); - std::for_each_n(frame_pointers.begin(), size, - [&bare_stacktrace](const void * ptr) - { - writeChar(' ', bare_stacktrace); - writePointerHex(ptr, bare_stacktrace); - } - ); - - LOG_FATAL(log, fmt::runtime(bare_stacktrace.str())); - - StackTrace::toStringEveryLine(const_cast(frame_pointers.data()), 0, size, [this](std::string_view s) { LOG_FATAL(log, fmt::runtime(s)); }); - } - } - ); - - -#if defined(OS_LINUX) - /// Write information about binary checksum. It can be difficult to calculate, so do it only after printing stack trace. - /// Please keep the below log messages in-sync with the ones in programs/server/Server.cpp - - if (daemon.stored_binary_hash.empty()) - { - LOG_FATAL(log, "Integrity check of the executable skipped because the reference checksum could not be read."); - } - else - { - String calculated_binary_hash = getHashOfLoadedBinaryHex(); - if (calculated_binary_hash == daemon.stored_binary_hash) - { - LOG_FATAL(log, "Integrity check of the executable successfully passed (checksum: {})", calculated_binary_hash); - } - else - { - LOG_FATAL( - log, - "Calculated checksum of the executable ({0}) does not correspond" - " to the reference checksum stored in the executable ({1})." - " This may indicate one of the following:" - " - the executable was changed just after startup;" - " - the executable was corrupted on disk due to faulty hardware;" - " - the loaded executable was corrupted in memory due to faulty hardware;" - " - the file was intentionally modified;" - " - a logical error in the code.", - calculated_binary_hash, - daemon.stored_binary_hash); - } - } -#endif - - /// Write crash to system.crash_log table if available. - if (collectCrashLog) - collectCrashLog(sig, thread_num, query_id, stack_trace); - -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD - Context::getGlobalContextInstance()->handleCrash(); -#endif - - /// Send crash report to developers (if configured) - if (sig != SanitizerTrap) - { - if (auto * sentry = SentryWriter::getInstance()) - sentry->onSignal(sig, error_message, stack_trace.getFramePointers(), stack_trace.getOffset(), stack_trace.getSize()); - - /// Advice the user to send it manually. - if (std::string_view(VERSION_OFFICIAL).contains("official build")) - { - const auto & date_lut = DateLUT::instance(); - - /// Approximate support period, upper bound. - if (time(nullptr) - date_lut.makeDate(2000 + VERSION_MAJOR, VERSION_MINOR, 1) < (365 + 30) * 86400) - { - LOG_FATAL(log, "Report this error to https://github.com/ClickHouse/ClickHouse/issues"); - } - else - { - LOG_FATAL(log, "ClickHouse version {} is old and should be upgraded to the latest version.", VERSION_STRING); - } - } - else - { - LOG_FATAL(log, "This ClickHouse version is not official and should be upgraded to the official build."); - } - } - - /// ClickHouse Keeper does not link to some parts of Settings. -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD - /// List changed settings. - if (!query_id.empty()) - { - ContextPtr query_context = thread_ptr->getQueryContext(); - if (query_context) - { - String changed_settings = query_context->getSettingsRef().toString(); - - if (changed_settings.empty()) - LOG_FATAL(log, "No settings were changed"); - else - LOG_FATAL(log, "Changed settings: {}", changed_settings); - } - } -#endif - - /// When everything is done, we will try to send these error messages to the client. - if (thread_ptr) - thread_ptr->onFatalError(); - - fatal_error_printed.test_and_set(); - } - catch (...) - { - /// onFault is called from the std::thread, and it should catch all exceptions; otherwise, you can get unrelated fatal errors. - PreformattedMessage message = getCurrentExceptionMessageAndPattern(true); - LOG_FATAL(getLogger(__PRETTY_FUNCTION__), message); - } -}; - - -#if defined(SANITIZER) - -template -struct ValueHolder -{ - ValueHolder(T value_) : value(value_) - {} - - T value; -}; - -extern "C" void __sanitizer_set_death_callback(void (*)()); - -/// Sanitizers may not expect some function calls from death callback. -/// Let's try to disable instrumentation to avoid possible issues. -/// However, this callback may call other functions that are still instrumented. -/// We can try [[clang::always_inline]] attribute for statements in future (available in clang-15) -/// See https://github.com/google/sanitizers/issues/1543 and https://github.com/google/sanitizers/issues/1549. -static DISABLE_SANITIZER_INSTRUMENTATION void sanitizerDeathCallback() -{ - DENY_ALLOCATIONS_IN_SCOPE; - /// Also need to send data via pipe. Otherwise it may lead to deadlocks or failures in printing diagnostic info. - - char buf[signal_pipe_buf_size]; - WriteBufferFromFileDescriptorDiscardOnFailure out(signal_pipe.fds_rw[1], signal_pipe_buf_size, buf); - - const StackTrace stack_trace; - - writeBinary(SignalListener::SanitizerTrap, out); - writePODBinary(stack_trace, out); - /// We create a dummy struct with a constructor so DISABLE_SANITIZER_INSTRUMENTATION is not applied to it - /// otherwise, Memory sanitizer can't know that values initiialized inside this function are actually initialized - /// because instrumentations are disabled leading to false positives later on - ValueHolder thread_id{static_cast(getThreadId())}; - writeBinary(thread_id.value, out); - writePODBinary(current_thread, out); - - out.next(); - - /// The time that is usually enough for separate thread to print info into log. - sleepForSeconds(20); -} -#endif - - -/** To use with std::set_terminate. - * Collects slightly more info than __gnu_cxx::__verbose_terminate_handler, - * and send it to pipe. Other thread will read this info from pipe and asynchronously write it to log. - * Look at libstdc++-v3/libsupc++/vterminate.cc for example. - */ -[[noreturn]] static void terminate_handler() -{ - static thread_local bool terminating = false; - if (terminating) - abort(); - - terminating = true; - - std::string log_message; - - if (std::current_exception()) - log_message = "Terminate called for uncaught exception:\n" + getCurrentExceptionMessage(true); - else - log_message = "Terminate called without an active exception"; - - /// POSIX.1 says that write(2)s of less than PIPE_BUF bytes must be atomic - man 7 pipe - /// And the buffer should not be too small because our exception messages can be large. - static constexpr size_t buf_size = PIPE_BUF; - - if (log_message.size() > buf_size - 16) - log_message.resize(buf_size - 16); - - char buf[buf_size]; - WriteBufferFromFileDescriptor out(signal_pipe.fds_rw[1], buf_size, buf); - - writeBinary(static_cast(SignalListener::StdTerminate), out); - writeBinary(static_cast(getThreadId()), out); - writeBinary(log_message, out); - out.next(); - - abort(); -} - - static std::string createDirectory(const std::string & file) { fs::path path = fs::path(file).parent_path(); @@ -700,25 +146,19 @@ BaseDaemon::BaseDaemon() = default; BaseDaemon::~BaseDaemon() { - writeSignalIDtoSignalPipe(SignalListener::StopThread); - signal_listener_thread.join(); - /// Reset signals to SIG_DFL to avoid trying to write to the signal_pipe that will be closed after. - for (int sig : handled_signals) - if (SIG_ERR == signal(sig, SIG_DFL)) - { - try - { - throw ErrnoException(ErrorCodes::CANNOT_SET_SIGNAL_HANDLER, "Cannot set signal handler"); - } - catch (ErrnoException &) - { - tryLogCurrentException(__PRETTY_FUNCTION__); - } - } + try + { + writeSignalIDtoSignalPipe(SignalListener::StopThread); + signal_listener_thread.join(); + HandledSignals::instance().reset(); + SentryWriter::resetInstance(); + } + catch (...) + { + tryLogCurrentException(&logger()); + } - signal_pipe.close(); - - SentryWriter::resetInstance(); + disableLogging(); } @@ -756,6 +196,8 @@ void BaseDaemon::closeFDs() #else fs::path proc_path{"/proc/self/fd"}; #endif + + const auto & signal_pipe = HandledSignals::instance().signal_pipe; if (fs::is_directory(proc_path)) /// Hooray, proc exists { /// in /proc/self/fd directory filenames are numeric file descriptors. @@ -976,56 +418,6 @@ void BaseDaemon::initialize(Application & self) } -static void addSignalHandler(const std::vector & signals, signal_function handler, std::vector * out_handled_signals) -{ - struct sigaction sa; - memset(&sa, 0, sizeof(sa)); - sa.sa_sigaction = handler; - sa.sa_flags = SA_SIGINFO; - -#if defined(OS_DARWIN) - sigemptyset(&sa.sa_mask); - for (auto signal : signals) - sigaddset(&sa.sa_mask, signal); -#else - if (sigemptyset(&sa.sa_mask)) - throw Poco::Exception("Cannot set signal handler."); - - for (auto signal : signals) - if (sigaddset(&sa.sa_mask, signal)) - throw Poco::Exception("Cannot set signal handler."); -#endif - - for (auto signal : signals) - if (sigaction(signal, &sa, nullptr)) - throw Poco::Exception("Cannot set signal handler."); - - if (out_handled_signals) - std::copy(signals.begin(), signals.end(), std::back_inserter(*out_handled_signals)); -} - - -static void blockSignals(const std::vector & signals) -{ - sigset_t sig_set; - -#if defined(OS_DARWIN) - sigemptyset(&sig_set); - for (auto signal : signals) - sigaddset(&sig_set, signal); -#else - if (sigemptyset(&sig_set)) - throw Poco::Exception("Cannot block signal."); - - for (auto signal : signals) - if (sigaddset(&sig_set, signal)) - throw Poco::Exception("Cannot block signal."); -#endif - - if (pthread_sigmask(SIG_BLOCK, &sig_set, nullptr)) - throw Poco::Exception("Cannot block signal."); -} - extern const char * GIT_HASH; void BaseDaemon::initializeTerminationAndSignalProcessing() @@ -1049,29 +441,21 @@ void BaseDaemon::initializeTerminationAndSignalProcessing() }; } } - std::set_terminate(terminate_handler); /// We want to avoid SIGPIPE when working with sockets and pipes, and just handle return value/errno instead. blockSignals({SIGPIPE}); /// Setup signal handlers. - /// SIGTSTP is added for debugging purposes. To output a stack trace of any running thread at anytime. - addSignalHandler({SIGABRT, SIGSEGV, SIGILL, SIGBUS, SIGSYS, SIGFPE, SIGPIPE, SIGTSTP, SIGTRAP}, signalHandler, &handled_signals); - addSignalHandler({SIGHUP}, closeLogsSignalHandler, &handled_signals); - addSignalHandler({SIGINT, SIGQUIT, SIGTERM}, terminateRequestedSignalHandler, &handled_signals); - -#if defined(SANITIZER) - __sanitizer_set_death_callback(sanitizerDeathCallback); -#endif + HandledSignals::instance().setupTerminateHandler(); + HandledSignals::instance().setupCommonDeadlySignalHandlers(); + HandledSignals::instance().setupCommonTerminateRequestSignalHandlers(); + HandledSignals::instance().addSignalHandler({SIGHUP}, closeLogsSignalHandler, true); /// Set up Poco ErrorHandler for Poco Threads. static KillingErrorHandler killing_error_handler; Poco::ErrorHandler::set(&killing_error_handler); - signal_pipe.setNonBlockingWrite(); - signal_pipe.tryIncreaseSize(1 << 20); - - signal_listener = std::make_unique(*this); + signal_listener = std::make_unique(this, getLogger("BaseDaemon")); signal_listener_thread.start(*signal_listener); #if defined(__ELF__) && !defined(OS_FREEBSD) @@ -1277,7 +661,7 @@ void BaseDaemon::setupWatchdog() /// Forward signals to the child process. if (forward_signals) { - addSignalHandler( + HandledSignals::instance().addSignalHandler( {SIGHUP, SIGINT, SIGQUIT, SIGTERM}, [](int sig, siginfo_t *, void *) { @@ -1293,7 +677,7 @@ void BaseDaemon::setupWatchdog() (void)res; } }, - nullptr); + false); } else { @@ -1310,6 +694,10 @@ void BaseDaemon::setupWatchdog() int status = 0; do { + // Close log files to prevent keeping descriptors of unlinked rotated files. + // On next log write files will be reopened. + closeLogs(logger()); + if (-1 != waitpid(pid, &status, WUNTRACED | WCONTINUED) || errno == ECHILD) { if (WIFSTOPPED(status)) diff --git a/src/Daemon/BaseDaemon.h b/src/Daemon/BaseDaemon.h index 3d34d404595..b15aa74fcf3 100644 --- a/src/Daemon/BaseDaemon.h +++ b/src/Daemon/BaseDaemon.h @@ -168,8 +168,6 @@ protected: String git_hash; String stored_binary_hash; - std::vector handled_signals; - bool should_setup_watchdog = false; char * argv0 = nullptr; }; diff --git a/src/Daemon/SentryWriter.cpp b/src/Daemon/SentryWriter.cpp index 9479dd65730..c51a1100639 100644 --- a/src/Daemon/SentryWriter.cpp +++ b/src/Daemon/SentryWriter.cpp @@ -19,7 +19,7 @@ #include "config.h" #include -#if USE_SENTRY && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_SENTRY # include # include diff --git a/src/DataTypes/DataTypeAggregateFunction.h b/src/DataTypes/DataTypeAggregateFunction.h index 8b4b3d6ee4c..52ed151107e 100644 --- a/src/DataTypes/DataTypeAggregateFunction.h +++ b/src/DataTypes/DataTypeAggregateFunction.h @@ -25,7 +25,6 @@ private: mutable std::optional version; String getNameImpl(bool with_version) const; - size_t getVersion() const; public: static constexpr bool is_parametric = true; @@ -39,6 +38,8 @@ public: { } + size_t getVersion() const; + String getFunctionName() const; AggregateFunctionPtr getFunction() const { return function; } diff --git a/src/DataTypes/DataTypeCustomSimpleAggregateFunction.cpp b/src/DataTypes/DataTypeCustomSimpleAggregateFunction.cpp index cae9622bcb9..d3b3adc4965 100644 --- a/src/DataTypes/DataTypeCustomSimpleAggregateFunction.cpp +++ b/src/DataTypes/DataTypeCustomSimpleAggregateFunction.cpp @@ -165,6 +165,19 @@ static std::pair create(const ASTPtr & argum return std::make_pair(storage_type, std::make_unique(std::move(custom_name), nullptr)); } +String DataTypeCustomSimpleAggregateFunction::getFunctionName() const +{ + return function->getName(); +} + +DataTypePtr createSimpleAggregateFunctionType(const AggregateFunctionPtr & function, const DataTypes & argument_types, const Array & parameters) +{ + auto custom_desc = std::make_unique( + std::make_unique(function, argument_types, parameters)); + + return DataTypeFactory::instance().getCustom(std::move(custom_desc)); +} + void registerDataTypeDomainSimpleAggregateFunction(DataTypeFactory & factory) { factory.registerDataTypeCustom("SimpleAggregateFunction", create); diff --git a/src/DataTypes/DataTypeCustomSimpleAggregateFunction.h b/src/DataTypes/DataTypeCustomSimpleAggregateFunction.h index bdabb465fe5..303da86979a 100644 --- a/src/DataTypes/DataTypeCustomSimpleAggregateFunction.h +++ b/src/DataTypes/DataTypeCustomSimpleAggregateFunction.h @@ -40,8 +40,13 @@ public: : function(function_), argument_types(argument_types_), parameters(parameters_) {} AggregateFunctionPtr getFunction() const { return function; } + String getFunctionName() const; + const DataTypes & getArgumentsDataTypes() const { return argument_types; } + const Array & getParameters() const { return parameters; } String getName() const override; static void checkSupportedFunctions(const AggregateFunctionPtr & function); }; +DataTypePtr createSimpleAggregateFunctionType(const AggregateFunctionPtr & function, const DataTypes & argument_types, const Array & parameters); + } diff --git a/src/DataTypes/DataTypeDate.cpp b/src/DataTypes/DataTypeDate.cpp index ee4b0065e59..0a7aa7deac6 100644 --- a/src/DataTypes/DataTypeDate.cpp +++ b/src/DataTypes/DataTypeDate.cpp @@ -17,7 +17,7 @@ SerializationPtr DataTypeDate::doGetDefaultSerialization() const void registerDataTypeDate(DataTypeFactory & factory) { - factory.registerSimpleDataType("Date", [] { return DataTypePtr(std::make_shared()); }, DataTypeFactory::CaseInsensitive); + factory.registerSimpleDataType("Date", [] { return DataTypePtr(std::make_shared()); }, DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/DataTypeDate32.cpp b/src/DataTypes/DataTypeDate32.cpp index 343e498d303..b2b8e7c0c1c 100644 --- a/src/DataTypes/DataTypeDate32.cpp +++ b/src/DataTypes/DataTypeDate32.cpp @@ -24,7 +24,7 @@ Field DataTypeDate32::getDefault() const void registerDataTypeDate32(DataTypeFactory & factory) { factory.registerSimpleDataType( - "Date32", [] { return DataTypePtr(std::make_shared()); }, DataTypeFactory::CaseInsensitive); + "Date32", [] { return DataTypePtr(std::make_shared()); }, DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/DataTypeDecimalBase.cpp b/src/DataTypes/DataTypeDecimalBase.cpp index 62218694924..2adfda5bb27 100644 --- a/src/DataTypes/DataTypeDecimalBase.cpp +++ b/src/DataTypes/DataTypeDecimalBase.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/DataTypes/DataTypeDomainBool.cpp b/src/DataTypes/DataTypeDomainBool.cpp index 3d19b6262d8..30dbba2d8c0 100644 --- a/src/DataTypes/DataTypeDomainBool.cpp +++ b/src/DataTypes/DataTypeDomainBool.cpp @@ -15,8 +15,8 @@ void registerDataTypeDomainBool(DataTypeFactory & factory) std::make_unique("Bool"), std::make_unique(type->getDefaultSerialization()))); }); - factory.registerAlias("bool", "Bool", DataTypeFactory::CaseInsensitive); - factory.registerAlias("boolean", "Bool", DataTypeFactory::CaseInsensitive); + factory.registerAlias("bool", "Bool", DataTypeFactory::Case::Insensitive); + factory.registerAlias("boolean", "Bool", DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/DataTypeDynamic.cpp b/src/DataTypes/DataTypeDynamic.cpp index c920e69c13b..a1b1f8325f0 100644 --- a/src/DataTypes/DataTypeDynamic.cpp +++ b/src/DataTypes/DataTypeDynamic.cpp @@ -2,9 +2,11 @@ #include #include #include +#include #include #include #include +#include #include #include #include @@ -69,7 +71,7 @@ static DataTypePtr create(const ASTPtr & arguments) auto * literal = argument->arguments->children[1]->as(); - if (!literal || literal->value.getType() != Field::Types::UInt64 || literal->value.get() == 0 || literal->value.get() > 255) + if (!literal || literal->value.getType() != Field::Types::UInt64 || literal->value.get() == 0 || literal->value.get() > ColumnVariant::MAX_NESTED_COLUMNS) throw Exception(ErrorCodes::UNEXPECTED_AST_STRUCTURE, "'max_types' argument for Dynamic type should be a positive integer between 1 and 255"); return std::make_shared(literal->value.get()); @@ -110,28 +112,58 @@ std::unique_ptr DataTypeDynamic::getDynamicSubcolumnDa } /// Extract nested subcolumn of requested dynamic subcolumn if needed. - if (!subcolumn_nested_name.empty()) + /// If requested subcolumn is null map, it's processed separately as there is no Nullable type yet. + bool is_null_map_subcolumn = subcolumn_nested_name == "null"; + if (is_null_map_subcolumn) + { + res->type = std::make_shared(); + } + else if (!subcolumn_nested_name.empty()) { res = getSubcolumnData(subcolumn_nested_name, *res, throw_if_null); if (!res) return nullptr; } - res->serialization = std::make_shared(res->serialization, subcolumn_type->getName()); - res->type = makeNullableOrLowCardinalityNullableSafe(res->type); + res->serialization = std::make_shared(res->serialization, subcolumn_type->getName(), is_null_map_subcolumn); + /// Make resulting subcolumn Nullable only if type subcolumn can be inside Nullable or can be LowCardinality(Nullable()). + bool make_subcolumn_nullable = subcolumn_type->canBeInsideNullable() || subcolumn_type->lowCardinality(); + if (!is_null_map_subcolumn && make_subcolumn_nullable) + res->type = makeNullableOrLowCardinalityNullableSafe(res->type); + if (data.column) { if (discriminator) { - /// Provided Dynamic column has subcolumn of this type, we should use VariantSubcolumnCreator to + /// Provided Dynamic column has subcolumn of this type, we should use VariantSubcolumnCreator/VariantNullMapSubcolumnCreator to /// create full subcolumn from variant according to discriminators. const auto & variant_column = assert_cast(*data.column).getVariantColumn(); - auto creator = SerializationVariantElement::VariantSubcolumnCreator(variant_column.getLocalDiscriminatorsPtr(), "", *discriminator, variant_column.localDiscriminatorByGlobal(*discriminator)); - res->column = creator.create(res->column); + std::unique_ptr creator; + if (is_null_map_subcolumn) + creator = std::make_unique( + variant_column.getLocalDiscriminatorsPtr(), + "", + *discriminator, + variant_column.localDiscriminatorByGlobal(*discriminator)); + else + creator = std::make_unique( + variant_column.getLocalDiscriminatorsPtr(), + "", + *discriminator, + variant_column.localDiscriminatorByGlobal(*discriminator), + make_subcolumn_nullable); + res->column = creator->create(res->column); + } + /// Provided Dynamic column doesn't have subcolumn of this type, just create column filled with default values. + else if (is_null_map_subcolumn) + { + /// Fill null map with 1 when there is no such Dynamic subcolumn. + auto column = ColumnUInt8::create(); + assert_cast(*column).getData().resize_fill(data.column->size(), 1); + res->column = std::move(column); } else { - /// Provided Dynamic column doesn't have subcolumn of this type, just create column filled with default values. auto column = res->type->createColumn(); column->insertManyDefaults(data.column->size()); res->column = std::move(column); diff --git a/src/DataTypes/DataTypeEnum.cpp b/src/DataTypes/DataTypeEnum.cpp index a1d5e4b39b7..08e0c0d2045 100644 --- a/src/DataTypes/DataTypeEnum.cpp +++ b/src/DataTypes/DataTypeEnum.cpp @@ -318,7 +318,7 @@ void registerDataTypeEnum(DataTypeFactory & factory) factory.registerDataType("Enum", create); /// MySQL - factory.registerAlias("ENUM", "Enum", DataTypeFactory::CaseInsensitive); + factory.registerAlias("ENUM", "Enum", DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/DataTypeFactory.cpp b/src/DataTypes/DataTypeFactory.cpp index 8c8f9999ada..eb3bc973857 100644 --- a/src/DataTypes/DataTypeFactory.cpp +++ b/src/DataTypes/DataTypeFactory.cpp @@ -10,6 +10,7 @@ #include #include #include +#include #include #include @@ -124,23 +125,6 @@ DataTypePtr DataTypeFactory::getImpl(const String & family_name_param, const AST { String family_name = getAliasToOrName(family_name_param); - if (endsWith(family_name, "WithDictionary")) - { - ASTPtr low_cardinality_params = std::make_shared(); - String param_name = family_name.substr(0, family_name.size() - strlen("WithDictionary")); - if (parameters) - { - auto func = std::make_shared(); - func->name = param_name; - func->arguments = parameters; - low_cardinality_params->children.push_back(func); - } - else - low_cardinality_params->children.push_back(std::make_shared(param_name)); - - return getImpl("LowCardinality", low_cardinality_params); - } - const auto * creator = findCreatorByName(family_name); if constexpr (nullptr_on_error) { @@ -174,7 +158,7 @@ DataTypePtr DataTypeFactory::getCustom(DataTypeCustomDescPtr customization) cons } -void DataTypeFactory::registerDataType(const String & family_name, Value creator, CaseSensitiveness case_sensitiveness) +void DataTypeFactory::registerDataType(const String & family_name, Value creator, Case case_sensitiveness) { if (creator == nullptr) throw Exception(ErrorCodes::LOGICAL_ERROR, "DataTypeFactory: the data type family {} has been provided a null constructor", family_name); @@ -188,12 +172,12 @@ void DataTypeFactory::registerDataType(const String & family_name, Value creator throw Exception(ErrorCodes::LOGICAL_ERROR, "DataTypeFactory: the data type family name '{}' is not unique", family_name); - if (case_sensitiveness == CaseInsensitive + if (case_sensitiveness == Case::Insensitive && !case_insensitive_data_types.emplace(family_name_lowercase, creator).second) throw Exception(ErrorCodes::LOGICAL_ERROR, "DataTypeFactory: the case insensitive data type family name '{}' is not unique", family_name); } -void DataTypeFactory::registerSimpleDataType(const String & name, SimpleCreator creator, CaseSensitiveness case_sensitiveness) +void DataTypeFactory::registerSimpleDataType(const String & name, SimpleCreator creator, Case case_sensitiveness) { if (creator == nullptr) throw Exception(ErrorCodes::LOGICAL_ERROR, "DataTypeFactory: the data type {} has been provided a null constructor", @@ -207,7 +191,7 @@ void DataTypeFactory::registerSimpleDataType(const String & name, SimpleCreator }, case_sensitiveness); } -void DataTypeFactory::registerDataTypeCustom(const String & family_name, CreatorWithCustom creator, CaseSensitiveness case_sensitiveness) +void DataTypeFactory::registerDataTypeCustom(const String & family_name, CreatorWithCustom creator, Case case_sensitiveness) { registerDataType(family_name, [creator](const ASTPtr & ast) { @@ -218,7 +202,7 @@ void DataTypeFactory::registerDataTypeCustom(const String & family_name, Creator }, case_sensitiveness); } -void DataTypeFactory::registerSimpleDataTypeCustom(const String & name, SimpleCreatorWithCustom creator, CaseSensitiveness case_sensitiveness) +void DataTypeFactory::registerSimpleDataTypeCustom(const String & name, SimpleCreatorWithCustom creator, Case case_sensitiveness) { registerDataTypeCustom(name, [name, creator](const ASTPtr & ast) { diff --git a/src/DataTypes/DataTypeFactory.h b/src/DataTypes/DataTypeFactory.h index 86e0203358d..edba9886d1c 100644 --- a/src/DataTypes/DataTypeFactory.h +++ b/src/DataTypes/DataTypeFactory.h @@ -41,16 +41,16 @@ public: DataTypePtr tryGet(const ASTPtr & ast) const; /// Register a type family by its name. - void registerDataType(const String & family_name, Value creator, CaseSensitiveness case_sensitiveness = CaseSensitive); + void registerDataType(const String & family_name, Value creator, Case case_sensitiveness = Case::Sensitive); /// Register a simple data type, that have no parameters. - void registerSimpleDataType(const String & name, SimpleCreator creator, CaseSensitiveness case_sensitiveness = CaseSensitive); + void registerSimpleDataType(const String & name, SimpleCreator creator, Case case_sensitiveness = Case::Sensitive); /// Register a customized type family - void registerDataTypeCustom(const String & family_name, CreatorWithCustom creator, CaseSensitiveness case_sensitiveness = CaseSensitive); + void registerDataTypeCustom(const String & family_name, CreatorWithCustom creator, Case case_sensitiveness = Case::Sensitive); /// Register a simple customized data type - void registerSimpleDataTypeCustom(const String & name, SimpleCreatorWithCustom creator, CaseSensitiveness case_sensitiveness = CaseSensitive); + void registerSimpleDataTypeCustom(const String & name, SimpleCreatorWithCustom creator, Case case_sensitiveness = Case::Sensitive); private: template diff --git a/src/DataTypes/DataTypeFixedString.cpp b/src/DataTypes/DataTypeFixedString.cpp index 85af59e852d..080ff8826a5 100644 --- a/src/DataTypes/DataTypeFixedString.cpp +++ b/src/DataTypes/DataTypeFixedString.cpp @@ -64,7 +64,7 @@ void registerDataTypeFixedString(DataTypeFactory & factory) factory.registerDataType("FixedString", create); /// Compatibility alias. - factory.registerAlias("BINARY", "FixedString", DataTypeFactory::CaseInsensitive); + factory.registerAlias("BINARY", "FixedString", DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/DataTypeIPv4andIPv6.cpp b/src/DataTypes/DataTypeIPv4andIPv6.cpp index 4c0b45f472a..de11cc50107 100644 --- a/src/DataTypes/DataTypeIPv4andIPv6.cpp +++ b/src/DataTypes/DataTypeIPv4andIPv6.cpp @@ -9,9 +9,9 @@ namespace DB void registerDataTypeIPv4andIPv6(DataTypeFactory & factory) { factory.registerSimpleDataType("IPv4", [] { return DataTypePtr(std::make_shared()); }); - factory.registerAlias("INET4", "IPv4", DataTypeFactory::CaseInsensitive); + factory.registerAlias("INET4", "IPv4", DataTypeFactory::Case::Insensitive); factory.registerSimpleDataType("IPv6", [] { return DataTypePtr(std::make_shared()); }); - factory.registerAlias("INET6", "IPv6", DataTypeFactory::CaseInsensitive); + factory.registerAlias("INET6", "IPv6", DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/DataTypeNested.h b/src/DataTypes/DataTypeNested.h index 1ad06477a6e..102e6c293cc 100644 --- a/src/DataTypes/DataTypeNested.h +++ b/src/DataTypes/DataTypeNested.h @@ -19,6 +19,8 @@ public: } String getName() const override; + const DataTypes & getElements() const { return elems; } + const Names & getNames() const { return names; } }; DataTypePtr createNested(const DataTypes & types, const Names & names); diff --git a/src/DataTypes/DataTypeObject.cpp b/src/DataTypes/DataTypeObject.cpp index 720436d0e0d..5636a46373f 100644 --- a/src/DataTypes/DataTypeObject.cpp +++ b/src/DataTypes/DataTypeObject.cpp @@ -76,7 +76,7 @@ void registerDataTypeObject(DataTypeFactory & factory) factory.registerDataType("Object", create); factory.registerSimpleDataType("JSON", [] { return std::make_shared("JSON", false); }, - DataTypeFactory::CaseInsensitive); + DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/DataTypeString.cpp b/src/DataTypes/DataTypeString.cpp index 95e49420009..ca65fb42cc8 100644 --- a/src/DataTypes/DataTypeString.cpp +++ b/src/DataTypes/DataTypeString.cpp @@ -62,38 +62,38 @@ void registerDataTypeString(DataTypeFactory & factory) /// These synonims are added for compatibility. - factory.registerAlias("CHAR", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NCHAR", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("CHARACTER", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("VARCHAR", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NVARCHAR", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("VARCHAR2", "String", DataTypeFactory::CaseInsensitive); /// Oracle - factory.registerAlias("TEXT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("TINYTEXT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("MEDIUMTEXT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("LONGTEXT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BLOB", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("CLOB", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("TINYBLOB", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("MEDIUMBLOB", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("LONGBLOB", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BYTEA", "String", DataTypeFactory::CaseInsensitive); /// PostgreSQL + factory.registerAlias("CHAR", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NCHAR", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("CHARACTER", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("VARCHAR", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NVARCHAR", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("VARCHAR2", "String", DataTypeFactory::Case::Insensitive); /// Oracle + factory.registerAlias("TEXT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("TINYTEXT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("MEDIUMTEXT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("LONGTEXT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BLOB", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("CLOB", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("TINYBLOB", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("MEDIUMBLOB", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("LONGBLOB", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BYTEA", "String", DataTypeFactory::Case::Insensitive); /// PostgreSQL - factory.registerAlias("CHARACTER LARGE OBJECT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("CHARACTER VARYING", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("CHAR LARGE OBJECT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("CHAR VARYING", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NATIONAL CHAR", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NATIONAL CHARACTER", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NATIONAL CHARACTER LARGE OBJECT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NATIONAL CHARACTER VARYING", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NATIONAL CHAR VARYING", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NCHAR VARYING", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NCHAR LARGE OBJECT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BINARY LARGE OBJECT", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BINARY VARYING", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("VARBINARY", "String", DataTypeFactory::CaseInsensitive); - factory.registerAlias("GEOMETRY", "String", DataTypeFactory::CaseInsensitive); //mysql + factory.registerAlias("CHARACTER LARGE OBJECT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("CHARACTER VARYING", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("CHAR LARGE OBJECT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("CHAR VARYING", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NATIONAL CHAR", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NATIONAL CHARACTER", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NATIONAL CHARACTER LARGE OBJECT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NATIONAL CHARACTER VARYING", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NATIONAL CHAR VARYING", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NCHAR VARYING", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NCHAR LARGE OBJECT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BINARY LARGE OBJECT", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BINARY VARYING", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("VARBINARY", "String", DataTypeFactory::Case::Insensitive); + factory.registerAlias("GEOMETRY", "String", DataTypeFactory::Case::Insensitive); //mysql } } diff --git a/src/DataTypes/DataTypeTuple.cpp b/src/DataTypes/DataTypeTuple.cpp index a4c8ed1a241..e96937d522d 100644 --- a/src/DataTypes/DataTypeTuple.cpp +++ b/src/DataTypes/DataTypeTuple.cpp @@ -32,7 +32,7 @@ namespace ErrorCodes extern const int NOT_FOUND_COLUMN_IN_BLOCK; extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; extern const int SIZES_OF_COLUMNS_IN_TUPLE_DOESNT_MATCH; - extern const int ILLEGAL_INDEX; + extern const int ARGUMENT_OUT_OF_BOUND; extern const int LOGICAL_ERROR; } @@ -286,7 +286,7 @@ std::optional DataTypeTuple::tryGetPositionByName(const String & name) c String DataTypeTuple::getNameByPosition(size_t i) const { if (i == 0 || i > names.size()) - throw Exception(ErrorCodes::ILLEGAL_INDEX, "Index of tuple element ({}) if out range ([1, {}])", i, names.size()); + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "Index of tuple element ({}) is out range ([1, {}])", i, names.size()); return names[i - 1]; } diff --git a/src/DataTypes/DataTypesBinaryEncoding.cpp b/src/DataTypes/DataTypesBinaryEncoding.cpp new file mode 100644 index 00000000000..bd994e313ba --- /dev/null +++ b/src/DataTypes/DataTypesBinaryEncoding.cpp @@ -0,0 +1,705 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int UNSUPPORTED_METHOD; + extern const int INCORRECT_DATA; +} + +namespace +{ + +enum class BinaryTypeIndex : uint8_t +{ + Nothing = 0x00, + UInt8 = 0x01, + UInt16 = 0x02, + UInt32 = 0x03, + UInt64 = 0x04, + UInt128 = 0x05, + UInt256 = 0x06, + Int8 = 0x07, + Int16 = 0x08, + Int32 = 0x09, + Int64 = 0x0A, + Int128 = 0x0B, + Int256 = 0x0C, + Float32 = 0x0D, + Float64 = 0x0E, + Date = 0x0F, + Date32 = 0x10, + DateTimeUTC = 0x11, + DateTimeWithTimezone = 0x12, + DateTime64UTC = 0x13, + DateTime64WithTimezone = 0x14, + String = 0x15, + FixedString = 0x16, + Enum8 = 0x17, + Enum16 = 0x18, + Decimal32 = 0x19, + Decimal64 = 0x1A, + Decimal128 = 0x1B, + Decimal256 = 0x1C, + UUID = 0x1D, + Array = 0x1E, + UnnamedTuple = 0x1F, + NamedTuple = 0x20, + Set = 0x21, + Interval = 0x22, + Nullable = 0x23, + Function = 0x24, + AggregateFunction = 0x25, + LowCardinality = 0x26, + Map = 0x27, + IPv4 = 0x28, + IPv6 = 0x29, + Variant = 0x2A, + Dynamic = 0x2B, + Custom = 0x2C, + Bool = 0x2D, + SimpleAggregateFunction = 0x2E, + Nested = 0x2F, +}; + +BinaryTypeIndex getBinaryTypeIndex(const DataTypePtr & type) +{ + /// By default custom types don't have their own BinaryTypeIndex. + if (type->hasCustomName()) + { + /// Some widely used custom types have separate BinaryTypeIndex for better serialization. + /// Right now it's Bool, SimpleAggregateFunction and Nested types. + /// TODO: Consider adding BinaryTypeIndex for more custom types. + + if (isBool(type)) + return BinaryTypeIndex::Bool; + + if (typeid_cast(type->getCustomName())) + return BinaryTypeIndex::SimpleAggregateFunction; + + if (isNested(type)) + return BinaryTypeIndex::Nested; + + return BinaryTypeIndex::Custom; + } + + switch (type->getTypeId()) + { + case TypeIndex::Nothing: + return BinaryTypeIndex::Nothing; + case TypeIndex::UInt8: + return BinaryTypeIndex::UInt8; + case TypeIndex::UInt16: + return BinaryTypeIndex::UInt16; + case TypeIndex::UInt32: + return BinaryTypeIndex::UInt32; + case TypeIndex::UInt64: + return BinaryTypeIndex::UInt64; + case TypeIndex::UInt128: + return BinaryTypeIndex::UInt128; + case TypeIndex::UInt256: + return BinaryTypeIndex::UInt256; + case TypeIndex::Int8: + return BinaryTypeIndex::Int8; + case TypeIndex::Int16: + return BinaryTypeIndex::Int16; + case TypeIndex::Int32: + return BinaryTypeIndex::Int32; + case TypeIndex::Int64: + return BinaryTypeIndex::Int64; + case TypeIndex::Int128: + return BinaryTypeIndex::Int128; + case TypeIndex::Int256: + return BinaryTypeIndex::Int256; + case TypeIndex::Float32: + return BinaryTypeIndex::Float32; + case TypeIndex::Float64: + return BinaryTypeIndex::Float64; + case TypeIndex::Date: + return BinaryTypeIndex::Date; + case TypeIndex::Date32: + return BinaryTypeIndex::Date32; + case TypeIndex::DateTime: + if (assert_cast(*type).hasExplicitTimeZone()) + return BinaryTypeIndex::DateTimeWithTimezone; + return BinaryTypeIndex::DateTimeUTC; + case TypeIndex::DateTime64: + if (assert_cast(*type).hasExplicitTimeZone()) + return BinaryTypeIndex::DateTime64WithTimezone; + return BinaryTypeIndex::DateTime64UTC; + case TypeIndex::String: + return BinaryTypeIndex::String; + case TypeIndex::FixedString: + return BinaryTypeIndex::FixedString; + case TypeIndex::Enum8: + return BinaryTypeIndex::Enum8; + case TypeIndex::Enum16: + return BinaryTypeIndex::Enum16; + case TypeIndex::Decimal32: + return BinaryTypeIndex::Decimal32; + case TypeIndex::Decimal64: + return BinaryTypeIndex::Decimal64; + case TypeIndex::Decimal128: + return BinaryTypeIndex::Decimal128; + case TypeIndex::Decimal256: + return BinaryTypeIndex::Decimal256; + case TypeIndex::UUID: + return BinaryTypeIndex::UUID; + case TypeIndex::Array: + return BinaryTypeIndex::Array; + case TypeIndex::Tuple: + { + const auto & tuple_type = assert_cast(*type); + if (tuple_type.haveExplicitNames()) + return BinaryTypeIndex::NamedTuple; + return BinaryTypeIndex::UnnamedTuple; + } + case TypeIndex::Set: + return BinaryTypeIndex::Set; + case TypeIndex::Interval: + return BinaryTypeIndex::Interval; + case TypeIndex::Nullable: + return BinaryTypeIndex::Nullable; + case TypeIndex::Function: + return BinaryTypeIndex::Function; + case TypeIndex::AggregateFunction: + return BinaryTypeIndex::AggregateFunction; + case TypeIndex::LowCardinality: + return BinaryTypeIndex::LowCardinality; + case TypeIndex::Map: + return BinaryTypeIndex::Map; + case TypeIndex::Object: + /// Object type will be deprecated and replaced by new implementation. No need to support it here. + throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "Binary encoding of type Object is not supported"); + case TypeIndex::IPv4: + return BinaryTypeIndex::IPv4; + case TypeIndex::IPv6: + return BinaryTypeIndex::IPv6; + case TypeIndex::Variant: + return BinaryTypeIndex::Variant; + case TypeIndex::Dynamic: + return BinaryTypeIndex::Dynamic; + /// JSONPaths is used only during schema inference and cannot be used anywhere else. + case TypeIndex::JSONPaths: + throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "Binary encoding of type JSONPaths is not supported"); + } +} + +template +void encodeEnumValues(const DataTypePtr & type, WriteBuffer & buf) +{ + const auto & enum_type = assert_cast &>(*type); + const auto & values = enum_type.getValues(); + writeVarUInt(values.size(), buf); + for (const auto & [name, value] : values) + { + writeStringBinary(name, buf); + writeBinaryLittleEndian(value, buf); + } +} + +template +DataTypePtr decodeEnum(ReadBuffer & buf) +{ + typename DataTypeEnum::Values values; + size_t size; + readVarUInt(size, buf); + for (size_t i = 0; i != size; ++i) + { + String name; + readStringBinary(name, buf); + T value; + readBinaryLittleEndian(value, buf); + values.emplace_back(name, value); + } + + return std::make_shared>(values); +} + +template +void encodeDecimal(const DataTypePtr & type, WriteBuffer & buf) +{ + const auto & decimal_type = assert_cast &>(*type); + /// Both precision and scale should be less than 76, so we can decode it in 1 byte. + writeBinary(UInt8(decimal_type.getPrecision()), buf); + writeBinary(UInt8(decimal_type.getScale()), buf); +} + +template +DataTypePtr decodeDecimal(ReadBuffer & buf) +{ + UInt8 precision; + readBinary(precision, buf); + UInt8 scale; + readBinary(scale, buf); + return std::make_shared>(precision, scale); +} + +void encodeAggregateFunction(const String & function_name, const Array & parameters, const DataTypes & arguments_types, WriteBuffer & buf) +{ + writeStringBinary(function_name, buf); + writeVarUInt(parameters.size(), buf); + for (const auto & param : parameters) + encodeField(param, buf); + writeVarUInt(arguments_types.size(), buf); + for (const auto & argument_type : arguments_types) + encodeDataType(argument_type, buf); +} + +std::tuple decodeAggregateFunction(ReadBuffer & buf) +{ + String function_name; + readStringBinary(function_name, buf); + size_t num_parameters; + readVarUInt(num_parameters, buf); + Array parameters; + parameters.reserve(num_parameters); + for (size_t i = 0; i != num_parameters; ++i) + parameters.push_back(decodeField(buf)); + size_t num_arguments; + readVarUInt(num_arguments, buf); + DataTypes arguments_types; + arguments_types.reserve(num_arguments); + for (size_t i = 0; i != num_arguments; ++i) + arguments_types.push_back(decodeDataType(buf)); + AggregateFunctionProperties properties; + auto action = NullsAction::EMPTY; + auto function = AggregateFunctionFactory::instance().get(function_name, action, arguments_types, parameters, properties); + return {function, parameters, arguments_types}; +} + +} + +void encodeDataType(const DataTypePtr & type, WriteBuffer & buf) +{ + /// First, write the BinaryTypeIndex byte. + auto binary_type_index = getBinaryTypeIndex(type); + buf.write(UInt8(binary_type_index)); + /// Then, write additional information depending on the data type. + switch (binary_type_index) + { + case BinaryTypeIndex::DateTimeWithTimezone: + { + const auto & datetime_type = assert_cast(*type); + writeStringBinary(datetime_type.getTimeZone().getTimeZone(), buf); + break; + } + case BinaryTypeIndex::DateTime64UTC: + { + const auto & datetime64_type = assert_cast(*type); + /// Maximum scale for DateTime64 is 9, so we can write it as 1 byte. + buf.write(UInt8(datetime64_type.getScale())); + break; + } + case BinaryTypeIndex::DateTime64WithTimezone: + { + const auto & datetime64_type = assert_cast(*type); + buf.write(UInt8(datetime64_type.getScale())); + writeStringBinary(datetime64_type.getTimeZone().getTimeZone(), buf); + break; + } + case BinaryTypeIndex::FixedString: + { + const auto & fixed_string_type = assert_cast(*type); + writeVarUInt(fixed_string_type.getN(), buf); + break; + } + case BinaryTypeIndex::Enum8: + { + encodeEnumValues(type, buf); + break; + } + case BinaryTypeIndex::Enum16: + { + encodeEnumValues(type, buf); + break; + } + case BinaryTypeIndex::Decimal32: + { + encodeDecimal(type, buf); + break; + } + case BinaryTypeIndex::Decimal64: + { + encodeDecimal(type, buf); + break; + } + case BinaryTypeIndex::Decimal128: + { + encodeDecimal(type, buf); + break; + } + case BinaryTypeIndex::Decimal256: + { + encodeDecimal(type, buf); + break; + } + case BinaryTypeIndex::Array: + { + const auto & array_type = assert_cast(*type); + encodeDataType(array_type.getNestedType(), buf); + break; + } + case BinaryTypeIndex::NamedTuple: + { + const auto & tuple_type = assert_cast(*type); + const auto & types = tuple_type.getElements(); + const auto & names = tuple_type.getElementNames(); + writeVarUInt(types.size(), buf); + for (size_t i = 0; i != types.size(); ++i) + { + writeStringBinary(names[i], buf); + encodeDataType(types[i], buf); + } + break; + } + case BinaryTypeIndex::UnnamedTuple: + { + const auto & tuple_type = assert_cast(*type); + const auto & element_types = tuple_type.getElements(); + writeVarUInt(element_types.size(), buf); + for (const auto & element_type : element_types) + encodeDataType(element_type, buf); + break; + } + case BinaryTypeIndex::Interval: + { + const auto & interval_type = assert_cast(*type); + writeBinary(UInt8(interval_type.getKind().kind), buf); + break; + } + case BinaryTypeIndex::Nullable: + { + const auto & nullable_type = assert_cast(*type); + encodeDataType(nullable_type.getNestedType(), buf); + break; + } + case BinaryTypeIndex::Function: + { + const auto & function_type = assert_cast(*type); + const auto & arguments_types = function_type.getArgumentTypes(); + const auto & return_type = function_type.getReturnType(); + writeVarUInt(arguments_types.size(), buf); + for (const auto & argument_type : arguments_types) + encodeDataType(argument_type, buf); + encodeDataType(return_type, buf); + break; + } + case BinaryTypeIndex::LowCardinality: + { + const auto & low_cardinality_type = assert_cast(*type); + encodeDataType(low_cardinality_type.getDictionaryType(), buf); + break; + } + case BinaryTypeIndex::Map: + { + const auto & map_type = assert_cast(*type); + encodeDataType(map_type.getKeyType(), buf); + encodeDataType(map_type.getValueType(), buf); + break; + } + case BinaryTypeIndex::Variant: + { + const auto & variant_type = assert_cast(*type); + const auto & variants = variant_type.getVariants(); + writeVarUInt(variants.size(), buf); + for (const auto & variant : variants) + encodeDataType(variant, buf); + break; + } + case BinaryTypeIndex::Dynamic: + { + const auto & dynamic_type = assert_cast(*type); + /// Maximum number of dynamic types is 255, we can write it as 1 byte. + writeBinary(UInt8(dynamic_type.getMaxDynamicTypes()), buf); + break; + } + case BinaryTypeIndex::AggregateFunction: + { + const auto & aggregate_function_type = assert_cast(*type); + writeVarUInt(aggregate_function_type.getVersion(), buf); + encodeAggregateFunction(aggregate_function_type.getFunctionName(), aggregate_function_type.getParameters(), aggregate_function_type.getArgumentsDataTypes(), buf); + break; + } + case BinaryTypeIndex::SimpleAggregateFunction: + { + const auto & simple_aggregate_function_type = assert_cast(*type->getCustomName()); + encodeAggregateFunction(simple_aggregate_function_type.getFunctionName(), simple_aggregate_function_type.getParameters(), simple_aggregate_function_type.getArgumentsDataTypes(), buf); + break; + } + case BinaryTypeIndex::Nested: + { + const auto & nested_type = assert_cast(*type->getCustomName()); + const auto & elements = nested_type.getElements(); + const auto & names = nested_type.getNames(); + writeVarUInt(elements.size(), buf); + for (size_t i = 0; i != elements.size(); ++i) + { + writeStringBinary(names[i], buf); + encodeDataType(elements[i], buf); + } + break; + } + case BinaryTypeIndex::Custom: + { + const auto & type_name = type->getName(); + writeStringBinary(type_name, buf); + break; + } + default: + break; + } +} + +String encodeDataType(const DataTypePtr & type) +{ + WriteBufferFromOwnString buf; + encodeDataType(type, buf); + return buf.str(); +} + +DataTypePtr decodeDataType(ReadBuffer & buf) +{ + UInt8 type; + readBinary(type, buf); + switch (BinaryTypeIndex(type)) + { + case BinaryTypeIndex::Nothing: + return std::make_shared(); + case BinaryTypeIndex::UInt8: + return std::make_shared(); + case BinaryTypeIndex::Bool: + return DataTypeFactory::instance().get("Bool"); + case BinaryTypeIndex::UInt16: + return std::make_shared(); + case BinaryTypeIndex::UInt32: + return std::make_shared(); + case BinaryTypeIndex::UInt64: + return std::make_shared(); + case BinaryTypeIndex::UInt128: + return std::make_shared(); + case BinaryTypeIndex::UInt256: + return std::make_shared(); + case BinaryTypeIndex::Int8: + return std::make_shared(); + case BinaryTypeIndex::Int16: + return std::make_shared(); + case BinaryTypeIndex::Int32: + return std::make_shared(); + case BinaryTypeIndex::Int64: + return std::make_shared(); + case BinaryTypeIndex::Int128: + return std::make_shared(); + case BinaryTypeIndex::Int256: + return std::make_shared(); + case BinaryTypeIndex::Float32: + return std::make_shared(); + case BinaryTypeIndex::Float64: + return std::make_shared(); + case BinaryTypeIndex::Date: + return std::make_shared(); + case BinaryTypeIndex::Date32: + return std::make_shared(); + case BinaryTypeIndex::DateTimeUTC: + return std::make_shared(); + case BinaryTypeIndex::DateTimeWithTimezone: + { + String time_zone; + readStringBinary(time_zone, buf); + return std::make_shared(time_zone); + } + case BinaryTypeIndex::DateTime64UTC: + { + UInt8 scale; + readBinary(scale, buf); + return std::make_shared(scale); + } + case BinaryTypeIndex::DateTime64WithTimezone: + { + UInt8 scale; + readBinary(scale, buf); + String time_zone; + readStringBinary(time_zone, buf); + return std::make_shared(scale, time_zone); + } + case BinaryTypeIndex::String: + return std::make_shared(); + case BinaryTypeIndex::FixedString: + { + UInt64 size; + readVarUInt(size, buf); + return std::make_shared(size); + } + case BinaryTypeIndex::Enum8: + return decodeEnum(buf); + case BinaryTypeIndex::Enum16: + return decodeEnum(buf); + case BinaryTypeIndex::Decimal32: + return decodeDecimal(buf); + case BinaryTypeIndex::Decimal64: + return decodeDecimal(buf); + case BinaryTypeIndex::Decimal128: + return decodeDecimal(buf); + case BinaryTypeIndex::Decimal256: + return decodeDecimal(buf); + case BinaryTypeIndex::UUID: + return std::make_shared(); + case BinaryTypeIndex::Array: + return std::make_shared(decodeDataType(buf)); + case BinaryTypeIndex::NamedTuple: + { + size_t size; + readVarUInt(size, buf); + DataTypes elements; + elements.reserve(size); + Names names; + names.reserve(size); + for (size_t i = 0; i != size; ++i) + { + names.emplace_back(); + readStringBinary(names.back(), buf); + elements.push_back(decodeDataType(buf)); + } + + return std::make_shared(elements, names); + } + case BinaryTypeIndex::UnnamedTuple: + { + size_t size; + readVarUInt(size, buf); + DataTypes elements; + elements.reserve(size); + for (size_t i = 0; i != size; ++i) + elements.push_back(decodeDataType(buf)); + return std::make_shared(elements); + } + case BinaryTypeIndex::Set: + return std::make_shared(); + case BinaryTypeIndex::Interval: + { + UInt8 kind; + readBinary(kind, buf); + return std::make_shared(IntervalKind(IntervalKind::Kind(kind))); + } + case BinaryTypeIndex::Nullable: + return std::make_shared(decodeDataType(buf)); + case BinaryTypeIndex::Function: + { + size_t arguments_size; + readVarUInt(arguments_size, buf); + DataTypes arguments; + arguments.reserve(arguments_size); + for (size_t i = 0; i != arguments_size; ++i) + arguments.push_back(decodeDataType(buf)); + auto return_type = decodeDataType(buf); + return std::make_shared(arguments, return_type); + } + case BinaryTypeIndex::LowCardinality: + return std::make_shared(decodeDataType(buf)); + case BinaryTypeIndex::Map: + { + auto key_type = decodeDataType(buf); + auto value_type = decodeDataType(buf); + return std::make_shared(key_type, value_type); + } + case BinaryTypeIndex::IPv4: + return std::make_shared(); + case BinaryTypeIndex::IPv6: + return std::make_shared(); + case BinaryTypeIndex::Variant: + { + size_t size; + readVarUInt(size, buf); + DataTypes variants; + variants.reserve(size); + for (size_t i = 0; i != size; ++i) + variants.push_back(decodeDataType(buf)); + return std::make_shared(variants); + } + case BinaryTypeIndex::Dynamic: + { + UInt8 max_dynamic_types; + readBinary(max_dynamic_types, buf); + return std::make_shared(max_dynamic_types); + } + case BinaryTypeIndex::AggregateFunction: + { + size_t version; + readVarUInt(version, buf); + const auto & [function, parameters, arguments_types] = decodeAggregateFunction(buf); + return std::make_shared(function, arguments_types, parameters, version); + } + case BinaryTypeIndex::SimpleAggregateFunction: + { + const auto & [function, parameters, arguments_types] = decodeAggregateFunction(buf); + return createSimpleAggregateFunctionType(function, arguments_types, parameters); + } + case BinaryTypeIndex::Nested: + { + size_t size; + readVarUInt(size, buf); + Names names; + names.reserve(size); + DataTypes elements; + elements.reserve(size); + for (size_t i = 0; i != size; ++i) + { + names.emplace_back(); + readStringBinary(names.back(), buf); + elements.push_back(decodeDataType(buf)); + } + + return createNested(elements, names); + } + case BinaryTypeIndex::Custom: + { + String type_name; + readStringBinary(type_name, buf); + return DataTypeFactory::instance().get(type_name); + } + } + + throw Exception(ErrorCodes::INCORRECT_DATA, "Unknown type code: {0:#04x}", UInt64(type)); +} + +DataTypePtr decodeDataType(const String & data) +{ + ReadBufferFromString buf(data); + return decodeDataType(buf); +} + +} diff --git a/src/DataTypes/DataTypesBinaryEncoding.h b/src/DataTypes/DataTypesBinaryEncoding.h new file mode 100644 index 00000000000..d02e7f85942 --- /dev/null +++ b/src/DataTypes/DataTypesBinaryEncoding.h @@ -0,0 +1,118 @@ +#pragma once + +#include + +namespace DB +{ + +/** + +Binary encoding for ClickHouse data types: +|------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| ClickHouse data type | Binary encoding | +|------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| +| Nothing | 0x00 | +| UInt8 | 0x01 | +| UInt16 | 0x02 | +| UInt32 | 0x03 | +| UInt64 | 0x04 | +| UInt128 | 0x05 | +| UInt256 | 0x06 | +| Int8 | 0x07 | +| Int16 | 0x08 | +| Int32 | 0x09 | +| Int64 | 0x0A | +| Int128 | 0x0B | +| Int256 | 0x0C | +| Float32 | 0x0D | +| Float64 | 0x0E | +| Date | 0x0F | +| Date32 | 0x10 | +| DateTime | 0x11 | +| DateTime(time_zone) | 0x12 | +| DateTime64(P) | 0x13 | +| DateTime64(P, time_zone) | 0x14 | +| String | 0x15 | +| FixedString(N) | 0x16 | +| Enum8 | 0x17... | +| Enum16 | 0x18...> | +| Decimal32(P, S) | 0x19 | +| Decimal64(P, S) | 0x1A | +| Decimal128(P, S) | 0x1B | +| Decimal256(P, S) | 0x1C | +| UUID | 0x1D | +| Array(T) | 0x1E | +| Tuple(T1, ..., TN) | 0x1F... | +| Tuple(name1 T1, ..., nameN TN) | 0x20... | +| Set | 0x21 | +| Interval | 0x22 | +| Nullable(T) | 0x23 | +| Function | 0x24... | +| AggregateFunction(function_name(param_1, ..., param_N), arg_T1, ..., arg_TN) | 0x25...... | +| LowCardinality(T) | 0x26 | +| Map(K, V) | 0x27 | +| IPv4 | 0x28 | +| IPv6 | 0x29 | +| Variant(T1, ..., TN) | 0x2A... | +| Dynamic(max_types=N) | 0x2B | +| Custom type (Ring, Polygon, etc) | 0x2C | +| Bool | 0x2D | +| SimpleAggregateFunction(function_name(param_1, ..., param_N), arg_T1, ..., arg_TN) | 0x2E...... | +| Nested(name1 T1, ..., nameN TN) | 0x2F... | +|------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| + +Interval kind binary encoding: +|---------------|-----------------| +| Interval kind | Binary encoding | +|---------------|-----------------| +| Nanosecond | 0x00 | +| Microsecond | 0x01 | +| Millisecond | 0x02 | +| Second | 0x03 | +| Minute | 0x04 | +| Hour | 0x05 | +| Day | 0x06 | +| Week | 0x07 | +| Month | 0x08 | +| Quarter | 0x09 | +| Year | 0x1A | +|---------------|-----------------| + +Aggregate function parameter binary encoding (binary encoding of a Field, see src/Common/FieldBinaryEncoding.h): +|------------------------|------------------------------------------------------------------------------------------------------------------------------| +| Parameter type | Binary encoding | +|------------------------|------------------------------------------------------------------------------------------------------------------------------| +| Null | 0x00 | +| UInt64 | 0x01 | +| Int64 | 0x02 | +| UInt128 | 0x03 | +| Int128 | 0x04 | +| UInt128 | 0x05 | +| Int128 | 0x06 | +| Float64 | 0x07 | +| Decimal32 | 0x08 | +| Decimal64 | 0x09 | +| Decimal128 | 0x0A | +| Decimal256 | 0x0B | +| String | 0x0C | +| Array | 0x0D... | +| Tuple | 0x0E... | +| Map | 0x0F... | +| IPv4 | 0x10 | +| IPv6 | 0x11 | +| UUID | 0x12 | +| Bool | 0x13 | +| Object | 0x14... | +| AggregateFunctionState | 0x15 | +| Negative infinity | 0xFE | +| Positive infinity | 0xFF | +|------------------------|------------------------------------------------------------------------------------------------------------------------------| +*/ + +String encodeDataType(const DataTypePtr & type); +void encodeDataType(const DataTypePtr & type, WriteBuffer & buf); + +DataTypePtr decodeDataType(const String & data); +DataTypePtr decodeDataType(ReadBuffer & buf); + +} diff --git a/src/DataTypes/DataTypesDecimal.cpp b/src/DataTypes/DataTypesDecimal.cpp index 77a7a3e7237..a427fd0717a 100644 --- a/src/DataTypes/DataTypesDecimal.cpp +++ b/src/DataTypes/DataTypesDecimal.cpp @@ -364,15 +364,15 @@ template class DataTypeDecimal; void registerDataTypeDecimal(DataTypeFactory & factory) { - factory.registerDataType("Decimal32", createExact, DataTypeFactory::CaseInsensitive); - factory.registerDataType("Decimal64", createExact, DataTypeFactory::CaseInsensitive); - factory.registerDataType("Decimal128", createExact, DataTypeFactory::CaseInsensitive); - factory.registerDataType("Decimal256", createExact, DataTypeFactory::CaseInsensitive); + factory.registerDataType("Decimal32", createExact, DataTypeFactory::Case::Insensitive); + factory.registerDataType("Decimal64", createExact, DataTypeFactory::Case::Insensitive); + factory.registerDataType("Decimal128", createExact, DataTypeFactory::Case::Insensitive); + factory.registerDataType("Decimal256", createExact, DataTypeFactory::Case::Insensitive); - factory.registerDataType("Decimal", create, DataTypeFactory::CaseInsensitive); - factory.registerAlias("DEC", "Decimal", DataTypeFactory::CaseInsensitive); - factory.registerAlias("NUMERIC", "Decimal", DataTypeFactory::CaseInsensitive); - factory.registerAlias("FIXED", "Decimal", DataTypeFactory::CaseInsensitive); + factory.registerDataType("Decimal", create, DataTypeFactory::Case::Insensitive); + factory.registerAlias("DEC", "Decimal", DataTypeFactory::Case::Insensitive); + factory.registerAlias("NUMERIC", "Decimal", DataTypeFactory::Case::Insensitive); + factory.registerAlias("FIXED", "Decimal", DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/DataTypesNumber.cpp b/src/DataTypes/DataTypesNumber.cpp index 99446d24eed..72020b0a5aa 100644 --- a/src/DataTypes/DataTypesNumber.cpp +++ b/src/DataTypes/DataTypesNumber.cpp @@ -65,41 +65,41 @@ void registerDataTypeNumbers(DataTypeFactory & factory) /// These synonyms are added for compatibility. - factory.registerAlias("TINYINT", "Int8", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INT1", "Int8", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BYTE", "Int8", DataTypeFactory::CaseInsensitive); - factory.registerAlias("TINYINT SIGNED", "Int8", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INT1 SIGNED", "Int8", DataTypeFactory::CaseInsensitive); - factory.registerAlias("SMALLINT", "Int16", DataTypeFactory::CaseInsensitive); - factory.registerAlias("SMALLINT SIGNED", "Int16", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INT", "Int32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INTEGER", "Int32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("MEDIUMINT", "Int32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("MEDIUMINT SIGNED", "Int32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INT SIGNED", "Int32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INTEGER SIGNED", "Int32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BIGINT", "Int64", DataTypeFactory::CaseInsensitive); - factory.registerAlias("SIGNED", "Int64", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BIGINT SIGNED", "Int64", DataTypeFactory::CaseInsensitive); - factory.registerAlias("TIME", "Int64", DataTypeFactory::CaseInsensitive); + factory.registerAlias("TINYINT", "Int8", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INT1", "Int8", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BYTE", "Int8", DataTypeFactory::Case::Insensitive); + factory.registerAlias("TINYINT SIGNED", "Int8", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INT1 SIGNED", "Int8", DataTypeFactory::Case::Insensitive); + factory.registerAlias("SMALLINT", "Int16", DataTypeFactory::Case::Insensitive); + factory.registerAlias("SMALLINT SIGNED", "Int16", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INT", "Int32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INTEGER", "Int32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("MEDIUMINT", "Int32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("MEDIUMINT SIGNED", "Int32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INT SIGNED", "Int32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INTEGER SIGNED", "Int32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BIGINT", "Int64", DataTypeFactory::Case::Insensitive); + factory.registerAlias("SIGNED", "Int64", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BIGINT SIGNED", "Int64", DataTypeFactory::Case::Insensitive); + factory.registerAlias("TIME", "Int64", DataTypeFactory::Case::Insensitive); - factory.registerAlias("TINYINT UNSIGNED", "UInt8", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INT1 UNSIGNED", "UInt8", DataTypeFactory::CaseInsensitive); - factory.registerAlias("SMALLINT UNSIGNED", "UInt16", DataTypeFactory::CaseInsensitive); - factory.registerAlias("YEAR", "UInt16", DataTypeFactory::CaseInsensitive); - factory.registerAlias("MEDIUMINT UNSIGNED", "UInt32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INT UNSIGNED", "UInt32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("INTEGER UNSIGNED", "UInt32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("UNSIGNED", "UInt64", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BIGINT UNSIGNED", "UInt64", DataTypeFactory::CaseInsensitive); - factory.registerAlias("BIT", "UInt64", DataTypeFactory::CaseInsensitive); - factory.registerAlias("SET", "UInt64", DataTypeFactory::CaseInsensitive); + factory.registerAlias("TINYINT UNSIGNED", "UInt8", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INT1 UNSIGNED", "UInt8", DataTypeFactory::Case::Insensitive); + factory.registerAlias("SMALLINT UNSIGNED", "UInt16", DataTypeFactory::Case::Insensitive); + factory.registerAlias("YEAR", "UInt16", DataTypeFactory::Case::Insensitive); + factory.registerAlias("MEDIUMINT UNSIGNED", "UInt32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INT UNSIGNED", "UInt32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("INTEGER UNSIGNED", "UInt32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("UNSIGNED", "UInt64", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BIGINT UNSIGNED", "UInt64", DataTypeFactory::Case::Insensitive); + factory.registerAlias("BIT", "UInt64", DataTypeFactory::Case::Insensitive); + factory.registerAlias("SET", "UInt64", DataTypeFactory::Case::Insensitive); - factory.registerAlias("FLOAT", "Float32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("REAL", "Float32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("SINGLE", "Float32", DataTypeFactory::CaseInsensitive); - factory.registerAlias("DOUBLE", "Float64", DataTypeFactory::CaseInsensitive); - factory.registerAlias("DOUBLE PRECISION", "Float64", DataTypeFactory::CaseInsensitive); + factory.registerAlias("FLOAT", "Float32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("REAL", "Float32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("SINGLE", "Float32", DataTypeFactory::Case::Insensitive); + factory.registerAlias("DOUBLE", "Float64", DataTypeFactory::Case::Insensitive); + factory.registerAlias("DOUBLE PRECISION", "Float64", DataTypeFactory::Case::Insensitive); } /// Explicit template instantiations. diff --git a/src/DataTypes/EnumValues.h b/src/DataTypes/EnumValues.h index 889878bc60f..9762ff5870f 100644 --- a/src/DataTypes/EnumValues.h +++ b/src/DataTypes/EnumValues.h @@ -7,7 +7,7 @@ namespace DB { -namespace ErrorCodesEnumValues +namespace ErrorCodes { extern const int BAD_ARGUMENTS; } diff --git a/src/DataTypes/IDataType.cpp b/src/DataTypes/IDataType.cpp index 1c9715bbf53..1cb64b65d3a 100644 --- a/src/DataTypes/IDataType.cpp +++ b/src/DataTypes/IDataType.cpp @@ -173,7 +173,7 @@ bool IDataType::hasDynamicSubcolumns() const auto data = SubstreamData(getDefaultSerialization()).withType(getPtr()); auto callback = [&](const SubstreamPath &, const String &, const SubstreamData & subcolumn_data) { - has_dynamic_subcolumns |= subcolumn_data.type->hasDynamicSubcolumnsData(); + has_dynamic_subcolumns |= subcolumn_data.type && subcolumn_data.type->hasDynamicSubcolumnsData(); }; forEachSubcolumn(callback, data); return has_dynamic_subcolumns; diff --git a/src/DataTypes/Serializations/ISerialization.cpp b/src/DataTypes/Serializations/ISerialization.cpp index bbb1d1a6cd1..7642a6619b3 100644 --- a/src/DataTypes/Serializations/ISerialization.cpp +++ b/src/DataTypes/Serializations/ISerialization.cpp @@ -64,6 +64,9 @@ String ISerialization::Substream::toString() const if (type == VariantElement) return fmt::format("VariantElement({})", variant_element_name); + if (type == VariantElementNullMap) + return fmt::format("VariantElementNullMap({}.null)", variant_element_name); + return String(magic_enum::enum_name(type)); } @@ -195,6 +198,8 @@ String getNameForSubstreamPath( stream_name += ".variant_offsets"; else if (it->type == Substream::VariantElement) stream_name += "." + it->variant_element_name; + else if (it->type == Substream::VariantElementNullMap) + stream_name += "." + it->variant_element_name + ".null"; else if (it->type == SubstreamType::DynamicStructure) stream_name += ".dynamic_structure"; } @@ -395,7 +400,8 @@ bool ISerialization::hasSubcolumnForPath(const SubstreamPath & path, size_t pref return path[last_elem].type == Substream::NullMap || path[last_elem].type == Substream::TupleElement || path[last_elem].type == Substream::ArraySizes - || path[last_elem].type == Substream::VariantElement; + || path[last_elem].type == Substream::VariantElement + || path[last_elem].type == Substream::VariantElementNullMap; } ISerialization::SubstreamData ISerialization::createFromPath(const SubstreamPath & path, size_t prefix_len) diff --git a/src/DataTypes/Serializations/ISerialization.h b/src/DataTypes/Serializations/ISerialization.h index 914ff9cf4a2..255dbbfadd2 100644 --- a/src/DataTypes/Serializations/ISerialization.h +++ b/src/DataTypes/Serializations/ISerialization.h @@ -184,6 +184,7 @@ public: VariantOffsets, VariantElements, VariantElement, + VariantElementNullMap, DynamicData, DynamicStructure, @@ -256,6 +257,11 @@ public: bool position_independent_encoding = true; + /// True if data type names should be serialized in binary encoding. + bool data_types_binary_encoding = false; + + bool use_compact_variant_discriminators_serialization = false; + enum class DynamicStatisticsMode { NONE, /// Don't write statistics. @@ -275,6 +281,9 @@ public: bool position_independent_encoding = true; + /// True if data type names should be deserialized in binary encoding. + bool data_types_binary_encoding = false; + bool native_format = false; /// If not zero, may be used to avoid reallocations while reading column of String type. @@ -434,6 +443,9 @@ protected: template State * checkAndGetState(const StatePtr & state) const; + template + static State * checkAndGetState(const StatePtr & state, const ISerialization * serialization); + [[noreturn]] void throwUnexpectedDataAfterParsedValue(IColumn & column, ReadBuffer & istr, const FormatSettings &, const String & type_name) const; }; @@ -444,10 +456,16 @@ using SubstreamType = ISerialization::Substream::Type; template State * ISerialization::checkAndGetState(const StatePtr & state) const +{ + return checkAndGetState(state, this); +} + +template +State * ISerialization::checkAndGetState(const StatePtr & state, const ISerialization * serialization) { if (!state) throw Exception(ErrorCodes::LOGICAL_ERROR, - "Got empty state for {}", demangle(typeid(*this).name())); + "Got empty state for {}", demangle(typeid(*serialization).name())); auto * state_concrete = typeid_cast(state.get()); if (!state_concrete) @@ -455,7 +473,7 @@ State * ISerialization::checkAndGetState(const StatePtr & state) const auto & state_ref = *state; throw Exception(ErrorCodes::LOGICAL_ERROR, "Invalid State for {}. Expected: {}, got {}", - demangle(typeid(*this).name()), + demangle(typeid(*serialization).name()), demangle(typeid(State).name()), demangle(typeid(state_ref).name())); } diff --git a/src/DataTypes/Serializations/SerializationArray.cpp b/src/DataTypes/Serializations/SerializationArray.cpp index ac7b8f4d084..b7d43332085 100644 --- a/src/DataTypes/Serializations/SerializationArray.cpp +++ b/src/DataTypes/Serializations/SerializationArray.cpp @@ -42,13 +42,13 @@ void SerializationArray::deserializeBinary(Field & field, ReadBuffer & istr, con { size_t size; readVarUInt(size, istr); - if (settings.max_binary_array_size && size > settings.max_binary_array_size) + if (settings.binary.max_binary_string_size && size > settings.binary.max_binary_string_size) throw Exception( ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Too large array size: {}. The maximum is: {}. To increase the maximum, use setting " "format_binary_max_array_size", size, - settings.max_binary_array_size); + settings.binary.max_binary_string_size); field = Array(); Array & arr = field.get(); @@ -82,13 +82,13 @@ void SerializationArray::deserializeBinary(IColumn & column, ReadBuffer & istr, size_t size; readVarUInt(size, istr); - if (settings.max_binary_array_size && size > settings.max_binary_array_size) + if (settings.binary.max_binary_string_size && size > settings.binary.max_binary_string_size) throw Exception( ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Too large array size: {}. The maximum is: {}. To increase the maximum, use setting " "format_binary_max_array_size", size, - settings.max_binary_array_size); + settings.binary.max_binary_string_size); IColumn & nested_column = column_array.getData(); diff --git a/src/DataTypes/Serializations/SerializationDynamic.cpp b/src/DataTypes/Serializations/SerializationDynamic.cpp index 6351ff0ca0b..7609ffc91ca 100644 --- a/src/DataTypes/Serializations/SerializationDynamic.cpp +++ b/src/DataTypes/Serializations/SerializationDynamic.cpp @@ -4,6 +4,8 @@ #include #include #include +#include +#include #include #include @@ -109,7 +111,10 @@ void SerializationDynamic::serializeBinaryBulkStatePrefix( const auto & variant_column = column_dynamic.getVariantColumn(); /// Write internal Variant type name. - writeStringBinary(dynamic_state->variant_type->getName(), *stream); + if (settings.data_types_binary_encoding) + encodeDataType(dynamic_state->variant_type, *stream); + else + writeStringBinary(dynamic_state->variant_type->getName(), *stream); /// Write statistics in prefix if needed. if (settings.dynamic_write_statistics == SerializeBinaryBulkSettings::DynamicStatisticsMode::PREFIX) @@ -178,9 +183,16 @@ ISerialization::DeserializeBinaryBulkStatePtr SerializationDynamic::deserializeD readBinaryLittleEndian(structure_version, *structure_stream); auto structure_state = std::make_shared(structure_version); /// Read internal Variant type name. - String data_type_name; - readStringBinary(data_type_name, *structure_stream); - structure_state->variant_type = DataTypeFactory::instance().get(data_type_name); + if (settings.data_types_binary_encoding) + { + structure_state->variant_type = decodeDataType(*structure_stream); + } + else + { + String data_type_name; + readStringBinary(data_type_name, *structure_stream); + structure_state->variant_type = DataTypeFactory::instance().get(data_type_name); + } const auto * variant_type = typeid_cast(structure_state->variant_type.get()); if (!variant_type) throw Exception(ErrorCodes::INCORRECT_DATA, "Incorrect type of Dynamic nested column, expected Variant, got {}", structure_state->variant_type->getName()); @@ -280,33 +292,27 @@ void SerializationDynamic::deserializeBinaryBulkWithMultipleStreams( void SerializationDynamic::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings & settings) const { - UInt8 null_bit = field.isNull(); - writeBinary(null_bit, ostr); - if (null_bit) + /// Serialize NULL as Nothing type with no value. + if (field.isNull()) + { + encodeDataType(std::make_shared(), ostr); return; + } auto field_type = applyVisitor(FieldToDataType(), field); - auto field_type_name = field_type->getName(); - writeVarUInt(field_type_name.size(), ostr); - writeString(field_type_name, ostr); + encodeDataType(field_type, ostr); field_type->getDefaultSerialization()->serializeBinary(field, ostr, settings); } void SerializationDynamic::deserializeBinary(Field & field, ReadBuffer & istr, const FormatSettings & settings) const { - UInt8 null_bit; - readBinary(null_bit, istr); - if (null_bit) + auto field_type = decodeDataType(istr); + if (isNothing(field_type)) { field = Null(); return; } - size_t field_type_name_size; - readVarUInt(field_type_name_size, istr); - String field_type_name(field_type_name_size, 0); - istr.readStrict(field_type_name.data(), field_type_name_size); - auto field_type = DataTypeFactory::instance().get(field_type_name); field_type->getDefaultSerialization()->deserializeBinary(field, istr, settings); } @@ -317,15 +323,15 @@ void SerializationDynamic::serializeBinary(const IColumn & column, size_t row_nu const auto & variant_column = dynamic_column.getVariantColumn(); auto global_discr = variant_column.globalDiscriminatorAt(row_num); - UInt8 null_bit = global_discr == ColumnVariant::NULL_DISCRIMINATOR; - writeBinary(null_bit, ostr); - if (null_bit) + /// Serialize NULL as Nothing type with no value. + if (global_discr == ColumnVariant::NULL_DISCRIMINATOR) + { + encodeDataType(std::make_shared(), ostr); return; + } const auto & variant_type = assert_cast(*variant_info.variant_type).getVariant(global_discr); - const auto & variant_type_name = variant_info.variant_names[global_discr]; - writeVarUInt(variant_type_name.size(), ostr); - writeString(variant_type_name, ostr); + encodeDataType(variant_type, ostr); variant_type->getDefaultSerialization()->serializeBinary(variant_column.getVariantByGlobalDiscriminator(global_discr), variant_column.offsetAt(row_num), ostr, settings); } @@ -346,30 +352,23 @@ static void deserializeVariant( void SerializationDynamic::deserializeBinary(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const { auto & dynamic_column = assert_cast(column); - UInt8 null_bit; - readBinary(null_bit, istr); - if (null_bit) + auto variant_type = decodeDataType(istr); + if (isNothing(variant_type)) { dynamic_column.insertDefault(); return; } - size_t variant_type_name_size; - readVarUInt(variant_type_name_size, istr); - String variant_type_name(variant_type_name_size, 0); - istr.readStrict(variant_type_name.data(), variant_type_name_size); - + auto variant_type_name = variant_type->getName(); const auto & variant_info = dynamic_column.getVariantInfo(); auto it = variant_info.variant_name_to_discriminator.find(variant_type_name); if (it != variant_info.variant_name_to_discriminator.end()) { - const auto & variant_type = assert_cast(*variant_info.variant_type).getVariant(it->second); deserializeVariant(dynamic_column.getVariantColumn(), variant_type, it->second, istr, [&settings](const ISerialization & serialization, IColumn & variant, ReadBuffer & buf){ serialization.deserializeBinary(variant, buf, settings); }); return; } /// We don't have this variant yet. Let's try to add it. - auto variant_type = DataTypeFactory::instance().get(variant_type_name); if (dynamic_column.addNewVariant(variant_type)) { auto discr = variant_info.variant_name_to_discriminator.at(variant_type_name); diff --git a/src/DataTypes/Serializations/SerializationDynamicElement.cpp b/src/DataTypes/Serializations/SerializationDynamicElement.cpp index dafd6d663b0..211f0ac9377 100644 --- a/src/DataTypes/Serializations/SerializationDynamicElement.cpp +++ b/src/DataTypes/Serializations/SerializationDynamicElement.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include @@ -77,7 +78,10 @@ void SerializationDynamicElement::deserializeBinaryBulkStatePrefix( if (auto global_discr = assert_cast(*variant_type).tryGetVariantDiscriminator(dynamic_element_name)) { settings.path.push_back(Substream::DynamicData); - dynamic_element_state->variant_serialization = std::make_shared(nested_serialization, dynamic_element_name, *global_discr); + if (is_null_map_subcolumn) + dynamic_element_state->variant_serialization = std::make_shared(dynamic_element_name, *global_discr); + else + dynamic_element_state->variant_serialization = std::make_shared(nested_serialization, dynamic_element_name, *global_discr); dynamic_element_state->variant_serialization->deserializeBinaryBulkStatePrefix(settings, dynamic_element_state->variant_element_state, cache); settings.path.pop_back(); } @@ -98,7 +102,16 @@ void SerializationDynamicElement::deserializeBinaryBulkWithMultipleStreams( SubstreamsCache * cache) const { if (!state) + { + if (is_null_map_subcolumn) + { + auto mutable_column = result_column->assumeMutable(); + auto & data = assert_cast(*mutable_column).getData(); + data.resize_fill(data.size() + limit, 1); + } + return; + } auto * dynamic_element_state = checkAndGetState(state); @@ -108,6 +121,12 @@ void SerializationDynamicElement::deserializeBinaryBulkWithMultipleStreams( dynamic_element_state->variant_serialization->deserializeBinaryBulkWithMultipleStreams(result_column, limit, settings, dynamic_element_state->variant_element_state, cache); settings.path.pop_back(); } + else if (is_null_map_subcolumn) + { + auto mutable_column = result_column->assumeMutable(); + auto & data = assert_cast(*mutable_column).getData(); + data.resize_fill(data.size() + limit, 1); + } else { auto mutable_column = result_column->assumeMutable(); diff --git a/src/DataTypes/Serializations/SerializationDynamicElement.h b/src/DataTypes/Serializations/SerializationDynamicElement.h index 2ddc3324139..127d14a55e0 100644 --- a/src/DataTypes/Serializations/SerializationDynamicElement.h +++ b/src/DataTypes/Serializations/SerializationDynamicElement.h @@ -13,11 +13,11 @@ private: /// To be able to deserialize Dynamic element as a subcolumn /// we need its type name and global discriminator. String dynamic_element_name; + bool is_null_map_subcolumn; public: - SerializationDynamicElement(const SerializationPtr & nested_, const String & dynamic_element_name_) - : SerializationWrapper(nested_) - , dynamic_element_name(dynamic_element_name_) + SerializationDynamicElement(const SerializationPtr & nested_, const String & dynamic_element_name_, bool is_null_map_subcolumn_ = false) + : SerializationWrapper(nested_), dynamic_element_name(dynamic_element_name_), is_null_map_subcolumn(is_null_map_subcolumn_) { } diff --git a/src/DataTypes/Serializations/SerializationMap.cpp b/src/DataTypes/Serializations/SerializationMap.cpp index 70fe5182ade..0bef3c7d79d 100644 --- a/src/DataTypes/Serializations/SerializationMap.cpp +++ b/src/DataTypes/Serializations/SerializationMap.cpp @@ -55,13 +55,13 @@ void SerializationMap::deserializeBinary(Field & field, ReadBuffer & istr, const { size_t size; readVarUInt(size, istr); - if (settings.max_binary_array_size && size > settings.max_binary_array_size) + if (settings.binary.max_binary_string_size && size > settings.binary.max_binary_string_size) throw Exception( ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Too large map size: {}. The maximum is: {}. To increase the maximum, use setting " "format_binary_max_array_size", size, - settings.max_binary_array_size); + settings.binary.max_binary_string_size); field = Map(); Map & map = field.get(); map.reserve(size); diff --git a/src/DataTypes/Serializations/SerializationString.cpp b/src/DataTypes/Serializations/SerializationString.cpp index 9e39ab23709..9e523d0d745 100644 --- a/src/DataTypes/Serializations/SerializationString.cpp +++ b/src/DataTypes/Serializations/SerializationString.cpp @@ -33,13 +33,13 @@ namespace ErrorCodes void SerializationString::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings & settings) const { const String & s = field.get(); - if (settings.max_binary_string_size && s.size() > settings.max_binary_string_size) + if (settings.binary.max_binary_string_size && s.size() > settings.binary.max_binary_string_size) throw Exception( ErrorCodes::TOO_LARGE_STRING_SIZE, "Too large string size: {}. The maximum is: {}. To increase the maximum, use setting " "format_binary_max_string_size", s.size(), - settings.max_binary_string_size); + settings.binary.max_binary_string_size); writeVarUInt(s.size(), ostr); writeString(s, ostr); @@ -50,13 +50,13 @@ void SerializationString::deserializeBinary(Field & field, ReadBuffer & istr, co { UInt64 size; readVarUInt(size, istr); - if (settings.max_binary_string_size && size > settings.max_binary_string_size) + if (settings.binary.max_binary_string_size && size > settings.binary.max_binary_string_size) throw Exception( ErrorCodes::TOO_LARGE_STRING_SIZE, "Too large string size: {}. The maximum is: {}. To increase the maximum, use setting " "format_binary_max_string_size", size, - settings.max_binary_string_size); + settings.binary.max_binary_string_size); field = String(); String & s = field.get(); @@ -68,13 +68,13 @@ void SerializationString::deserializeBinary(Field & field, ReadBuffer & istr, co void SerializationString::serializeBinary(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const { const StringRef & s = assert_cast(column).getDataAt(row_num); - if (settings.max_binary_string_size && s.size > settings.max_binary_string_size) + if (settings.binary.max_binary_string_size && s.size > settings.binary.max_binary_string_size) throw Exception( ErrorCodes::TOO_LARGE_STRING_SIZE, "Too large string size: {}. The maximum is: {}. To increase the maximum, use setting " "format_binary_max_string_size", s.size, - settings.max_binary_string_size); + settings.binary.max_binary_string_size); writeVarUInt(s.size, ostr); writeString(s, ostr); @@ -89,13 +89,13 @@ void SerializationString::deserializeBinary(IColumn & column, ReadBuffer & istr, UInt64 size; readVarUInt(size, istr); - if (settings.max_binary_string_size && size > settings.max_binary_string_size) + if (settings.binary.max_binary_string_size && size > settings.binary.max_binary_string_size) throw Exception( ErrorCodes::TOO_LARGE_STRING_SIZE, "Too large string size: {}. The maximum is: {}. To increase the maximum, use setting " "format_binary_max_string_size", size, - settings.max_binary_string_size); + settings.binary.max_binary_string_size); size_t old_chars_size = data.size(); size_t offset = old_chars_size + size + 1; diff --git a/src/DataTypes/Serializations/SerializationVariant.cpp b/src/DataTypes/Serializations/SerializationVariant.cpp index b386fd8ab45..e4d71e84cc7 100644 --- a/src/DataTypes/Serializations/SerializationVariant.cpp +++ b/src/DataTypes/Serializations/SerializationVariant.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include @@ -30,12 +31,18 @@ namespace ErrorCodes struct SerializeBinaryBulkStateVariant : public ISerialization::SerializeBinaryBulkState { - std::vector states; + explicit SerializeBinaryBulkStateVariant(UInt64 mode) : discriminators_mode(mode) + { + } + + SerializationVariant::DiscriminatorsSerializationMode discriminators_mode; + std::vector variant_states; }; struct DeserializeBinaryBulkStateVariant : public ISerialization::DeserializeBinaryBulkState { - std::vector states; + ISerialization::DeserializeBinaryBulkStatePtr discriminators_state; + std::vector variant_states; }; void SerializationVariant::enumerateStreams( @@ -65,13 +72,19 @@ void SerializationVariant::enumerateStreams( for (size_t i = 0; i < variants.size(); ++i) { - settings.path.back().creator = std::make_shared(local_discriminators, variant_names[i], i, column_variant ? column_variant->localDiscriminatorByGlobal(i) : i); + DataTypePtr type = type_variant ? type_variant->getVariant(i) : nullptr; + settings.path.back().creator = std::make_shared( + local_discriminators, + variant_names[i], + i, + column_variant ? column_variant->localDiscriminatorByGlobal(i) : i, + !type || type->canBeInsideNullable() || type->lowCardinality()); auto variant_data = SubstreamData(variants[i]) - .withType(type_variant ? type_variant->getVariant(i) : nullptr) + .withType(type) .withColumn(column_variant ? column_variant->getVariantPtrByGlobalDiscriminator(i) : nullptr) .withSerializationInfo(data.serialization_info) - .withDeserializeState(variant_deserialize_state ? variant_deserialize_state->states[i] : nullptr); + .withDeserializeState(variant_deserialize_state ? variant_deserialize_state->variant_states[i] : nullptr); addVariantElementToPath(settings.path, i); settings.path.back().data = variant_data; @@ -79,6 +92,24 @@ void SerializationVariant::enumerateStreams( settings.path.pop_back(); } + /// Variant subcolumns like variant.Type have type Nullable(Type), so we want to support reading null map subcolumn from it: variant.Type.null. + /// Nullable column is created during deserialization of a variant subcolumn according to the discriminators, so we don't have actual Nullable + /// serialization with null map subcolumn. To be able to read null map subcolumn from the variant subcolumn we use special serialization + /// SerializationVariantElementNullMap. + auto null_map_data = SubstreamData(std::make_shared>()) + .withType(type_variant ? std::make_shared() : nullptr) + .withColumn(column_variant ? ColumnUInt8::create() : nullptr); + + for (size_t i = 0; i < variants.size(); ++i) + { + settings.path.back().creator = std::make_shared(local_discriminators, variant_names[i], i, column_variant ? column_variant->localDiscriminatorByGlobal(i) : i); + settings.path.push_back(Substream::VariantElementNullMap); + settings.path.back().variant_element_name = variant_names[i]; + settings.path.back().data = null_map_data; + callback(settings.path); + settings.path.pop_back(); + } + settings.path.pop_back(); } @@ -87,17 +118,26 @@ void SerializationVariant::serializeBinaryBulkStatePrefix( SerializeBinaryBulkSettings & settings, SerializeBinaryBulkStatePtr & state) const { - const ColumnVariant & col = assert_cast(column); + settings.path.push_back(Substream::VariantDiscriminators); + auto * discriminators_stream = settings.getter(settings.path); + settings.path.pop_back(); - auto variant_state = std::make_shared(); - variant_state->states.resize(variants.size()); + if (!discriminators_stream) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Got empty stream for VariantDiscriminators in SerializationVariant::serializeBinaryBulkStatePrefix"); + + UInt64 mode = settings.use_compact_variant_discriminators_serialization ? DiscriminatorsSerializationMode::COMPACT : DiscriminatorsSerializationMode::BASIC; + writeBinaryLittleEndian(mode, *discriminators_stream); + + const ColumnVariant & col = assert_cast(column); + auto variant_state = std::make_shared(mode); + variant_state->variant_states.resize(variants.size()); settings.path.push_back(Substream::VariantElements); for (size_t i = 0; i < variants.size(); ++i) { addVariantElementToPath(settings.path, i); - variants[i]->serializeBinaryBulkStatePrefix(col.getVariantByGlobalDiscriminator(i), settings, variant_state->states[i]); + variants[i]->serializeBinaryBulkStatePrefix(col.getVariantByGlobalDiscriminator(i), settings, variant_state->variant_states[i]); settings.path.pop_back(); } @@ -116,7 +156,7 @@ void SerializationVariant::serializeBinaryBulkStateSuffix( for (size_t i = 0; i < variants.size(); ++i) { addVariantElementToPath(settings.path, i); - variants[i]->serializeBinaryBulkStateSuffix(settings, variant_state->states[i]); + variants[i]->serializeBinaryBulkStateSuffix(settings, variant_state->variant_states[i]); settings.path.pop_back(); } settings.path.pop_back(); @@ -128,14 +168,19 @@ void SerializationVariant::deserializeBinaryBulkStatePrefix( DeserializeBinaryBulkStatePtr & state, SubstreamsDeserializeStatesCache * cache) const { + DeserializeBinaryBulkStatePtr discriminators_state = deserializeDiscriminatorsStatePrefix(settings, cache); + if (!discriminators_state) + return; + auto variant_state = std::make_shared(); - variant_state->states.resize(variants.size()); + variant_state->discriminators_state = discriminators_state; + variant_state->variant_states.resize(variants.size()); settings.path.push_back(Substream::VariantElements); for (size_t i = 0; i < variants.size(); ++i) { addVariantElementToPath(settings.path, i); - variants[i]->deserializeBinaryBulkStatePrefix(settings, variant_state->states[i], cache); + variants[i]->deserializeBinaryBulkStatePrefix(settings, variant_state->variant_states[i], cache); settings.path.pop_back(); } @@ -143,6 +188,29 @@ void SerializationVariant::deserializeBinaryBulkStatePrefix( state = std::move(variant_state); } +ISerialization::DeserializeBinaryBulkStatePtr SerializationVariant::deserializeDiscriminatorsStatePrefix( + DeserializeBinaryBulkSettings & settings, + SubstreamsDeserializeStatesCache * cache) +{ + settings.path.push_back(Substream::VariantDiscriminators); + + DeserializeBinaryBulkStatePtr discriminators_state = nullptr; + if (auto cached_state = getFromSubstreamsDeserializeStatesCache(cache, settings.path)) + { + discriminators_state = cached_state; + } + else if (auto * discriminators_stream = settings.getter(settings.path)) + { + UInt64 mode; + readBinaryLittleEndian(mode, *discriminators_stream); + discriminators_state = std::make_shared(mode); + addToSubstreamsDeserializeStatesCache(cache, settings.path, discriminators_state); + } + + settings.path.pop_back(); + return discriminators_state; +} + void SerializationVariant::serializeBinaryBulkWithMultipleStreamsAndUpdateVariantStatistics( const IColumn & column, @@ -165,13 +233,71 @@ void SerializationVariant::serializeBinaryBulkWithMultipleStreamsAndUpdateVarian auto * variant_state = checkAndGetState(state); - /// If offset = 0 and limit == col.size() or we have only NULLs, we don't need to calculate + /// Don't write anything if column is empty. + if (limit == 0) + return; + + /// Write number of rows in this granule in compact mode. + if (variant_state->discriminators_mode.value == DiscriminatorsSerializationMode::COMPACT) + writeVarUInt(UInt64(limit), *discriminators_stream); + + /// If column has only one none empty discriminators and no NULLs we don't need to + /// calculate limits for variants and use provided offset/limit. + if (auto non_empty_local_discr = col.getLocalDiscriminatorOfOneNoneEmptyVariantNoNulls()) + { + auto non_empty_global_discr = col.globalDiscriminatorByLocal(*non_empty_local_discr); + + /// In compact mode write the format of the granule and single non-empty discriminator. + if (variant_state->discriminators_mode.value == DiscriminatorsSerializationMode::COMPACT) + { + writeBinaryLittleEndian(UInt8(CompactDiscriminatorsGranuleFormat::COMPACT), *discriminators_stream); + writeBinaryLittleEndian(non_empty_global_discr, *discriminators_stream); + } + /// For basic mode just serialize this discriminator limit times. + else + { + for (size_t i = 0; i < limit; ++i) + writeBinaryLittleEndian(non_empty_global_discr, *discriminators_stream); + } + + settings.path.push_back(Substream::VariantElements); + addVariantElementToPath(settings.path, non_empty_global_discr); + /// We can use the same offset/limit as for whole Variant column + variants[non_empty_global_discr]->serializeBinaryBulkWithMultipleStreams(col.getVariantByGlobalDiscriminator(non_empty_global_discr), offset, limit, settings, variant_state->variant_states[non_empty_global_discr]); + variants_statistics[variant_names[non_empty_global_discr]] += limit; + settings.path.pop_back(); + settings.path.pop_back(); + return; + } + /// If column has only NULLs, just serialize NULL discriminators. + else if (col.hasOnlyNulls()) + { + /// In compact mode write single NULL_DISCRIMINATOR. + if (variant_state->discriminators_mode.value == DiscriminatorsSerializationMode::COMPACT) + { + writeBinaryLittleEndian(UInt8(CompactDiscriminatorsGranuleFormat::COMPACT), *discriminators_stream); + writeBinaryLittleEndian(ColumnVariant::NULL_DISCRIMINATOR, *discriminators_stream); + } + /// In basic mode write NULL_DISCRIMINATOR limit times. + else + { + for (size_t i = 0; i < limit; ++i) + writeBinaryLittleEndian(ColumnVariant::NULL_DISCRIMINATOR, *discriminators_stream); + } + return; + } + + /// If offset = 0 and limit == col.size() we don't need to calculate /// offsets and limits for variants and need to just serialize whole columns. - if ((offset == 0 && limit == col.size()) || col.hasOnlyNulls()) + if ((offset == 0 && limit == col.size())) { /// First, serialize discriminators. - /// If we have only NULLs or local and global discriminators are the same, just serialize the column as is. - if (col.hasOnlyNulls() || col.hasGlobalVariantsOrder()) + /// Here we are sure that column contains different discriminators, use plain granule format in compact mode. + if (variant_state->discriminators_mode.value == DiscriminatorsSerializationMode::COMPACT) + writeBinaryLittleEndian(UInt8(CompactDiscriminatorsGranuleFormat::PLAIN), *discriminators_stream); + + /// If local and global discriminators are the same, just serialize the column as is. + if (col.hasGlobalVariantsOrder()) { SerializationNumber().serializeBinaryBulk(col.getLocalDiscriminatorsColumn(), *discriminators_stream, offset, limit); } @@ -188,7 +314,7 @@ void SerializationVariant::serializeBinaryBulkWithMultipleStreamsAndUpdateVarian for (size_t i = 0; i != variants.size(); ++i) { addVariantElementToPath(settings.path, i); - variants[i]->serializeBinaryBulkWithMultipleStreams(col.getVariantByGlobalDiscriminator(i), 0, 0, settings, variant_state->states[i]); + variants[i]->serializeBinaryBulkWithMultipleStreams(col.getVariantByGlobalDiscriminator(i), 0, 0, settings, variant_state->variant_states[i]); variants_statistics[variant_names[i]] += col.getVariantByGlobalDiscriminator(i).size(); settings.path.pop_back(); } @@ -196,36 +322,16 @@ void SerializationVariant::serializeBinaryBulkWithMultipleStreamsAndUpdateVarian return; } - /// If we have only one non empty variant and no NULLs, we can use the same limit offset for this variant. - if (auto non_empty_local_discr = col.getLocalDiscriminatorOfOneNoneEmptyVariantNoNulls()) - { - /// First, serialize discriminators. - /// We know that all discriminators are the same, so we just need to serialize this discriminator limit times. - auto non_empty_global_discr = col.globalDiscriminatorByLocal(*non_empty_local_discr); - for (size_t i = 0; i != limit; ++i) - writeBinaryLittleEndian(non_empty_global_discr, *discriminators_stream); - - /// Second, serialize non-empty variant (other variants are empty and we can skip their serialization). - settings.path.push_back(Substream::VariantElements); - addVariantElementToPath(settings.path, non_empty_global_discr); - /// We can use the same offset/limit as for whole Variant column - variants[non_empty_global_discr]->serializeBinaryBulkWithMultipleStreams(col.getVariantByGlobalDiscriminator(non_empty_global_discr), offset, limit, settings, variant_state->states[non_empty_global_discr]); - variants_statistics[variant_names[non_empty_global_discr]] += limit; - settings.path.pop_back(); - settings.path.pop_back(); - return; - } - /// In general case we should iterate through local discriminators in range [offset, offset + limit] to serialize global discriminators and calculate offset/limit pair for each variant. const auto & local_discriminators = col.getLocalDiscriminators(); const auto & offsets = col.getOffsets(); std::vector> variant_offsets_and_limits(variants.size(), {0, 0}); size_t end = offset + limit; + size_t num_non_empty_variants_in_range = 0; + ColumnVariant::Discriminator last_non_empty_variant_discr = 0; for (size_t i = offset; i < end; ++i) { auto global_discr = col.globalDiscriminatorByLocal(local_discriminators[i]); - writeBinaryLittleEndian(global_discr, *discriminators_stream); - if (global_discr != ColumnVariant::NULL_DISCRIMINATOR) { /// If we see this discriminator for the first time, update offset @@ -233,9 +339,38 @@ void SerializationVariant::serializeBinaryBulkWithMultipleStreamsAndUpdateVarian variant_offsets_and_limits[global_discr].first = offsets[i]; /// Update limit for this discriminator. ++variant_offsets_and_limits[global_discr].second; + ++num_non_empty_variants_in_range; + last_non_empty_variant_discr = global_discr; } } + /// In basic mode just serialize discriminators as is row by row. + if (variant_state->discriminators_mode.value == DiscriminatorsSerializationMode::BASIC) + { + for (size_t i = offset; i < end; ++i) + writeBinaryLittleEndian(col.globalDiscriminatorByLocal(local_discriminators[i]), *discriminators_stream); + } + /// In compact mode check if we have the same discriminator for all rows in this granule. + /// First, check if all values in granule are NULLs. + else if (num_non_empty_variants_in_range == 0) + { + writeBinaryLittleEndian(UInt8(CompactDiscriminatorsGranuleFormat::COMPACT), *discriminators_stream); + writeBinaryLittleEndian(ColumnVariant::NULL_DISCRIMINATOR, *discriminators_stream); + } + /// Then, check if there is only 1 variant and no NULLs in this granule. + else if (num_non_empty_variants_in_range == 1 && variant_offsets_and_limits[last_non_empty_variant_discr].second == limit) + { + writeBinaryLittleEndian(UInt8(CompactDiscriminatorsGranuleFormat::COMPACT), *discriminators_stream); + writeBinaryLittleEndian(last_non_empty_variant_discr, *discriminators_stream); + } + /// Otherwise there are different discriminators in this granule. + else + { + writeBinaryLittleEndian(UInt8(CompactDiscriminatorsGranuleFormat::PLAIN), *discriminators_stream); + for (size_t i = offset; i < end; ++i) + writeBinaryLittleEndian(col.globalDiscriminatorByLocal(local_discriminators[i]), *discriminators_stream); + } + /// Serialize variants in global order. settings.path.push_back(Substream::VariantElements); for (size_t i = 0; i != variants.size(); ++i) @@ -249,7 +384,7 @@ void SerializationVariant::serializeBinaryBulkWithMultipleStreamsAndUpdateVarian variant_offsets_and_limits[i].first, variant_offsets_and_limits[i].second, settings, - variant_state->states[i]); + variant_state->variant_states[i]); variants_statistics[variant_names[i]] += variant_offsets_and_limits[i].second; settings.path.pop_back(); } @@ -284,39 +419,68 @@ void SerializationVariant::deserializeBinaryBulkWithMultipleStreams( /// First, deserialize discriminators. settings.path.push_back(Substream::VariantDiscriminators); + + DeserializeBinaryBulkStateVariant * variant_state = nullptr; + std::vector variant_limits; if (auto cached_discriminators = getFromSubstreamsCache(cache, settings.path)) { + variant_state = checkAndGetState(state); col.getLocalDiscriminatorsPtr() = cached_discriminators; } - else + else if (auto * discriminators_stream = settings.getter(settings.path)) { - auto * discriminators_stream = settings.getter(settings.path); - if (!discriminators_stream) - return; + variant_state = checkAndGetState(state); + auto * discriminators_state = checkAndGetState(variant_state->discriminators_state); + + /// Deserialize discriminators according to serialization mode. + if (discriminators_state->mode.value == DiscriminatorsSerializationMode::BASIC) + SerializationNumber().deserializeBinaryBulk(*col.getLocalDiscriminatorsPtr()->assumeMutable(), *discriminators_stream, limit, 0); + else + variant_limits = deserializeCompactDiscriminators(col.getLocalDiscriminatorsPtr(), limit, discriminators_stream, settings.continuous_reading, *discriminators_state); - SerializationNumber().deserializeBinaryBulk(*col.getLocalDiscriminatorsPtr()->assumeMutable(), *discriminators_stream, limit, 0); addToSubstreamsCache(cache, settings.path, col.getLocalDiscriminatorsPtr()); } + /// It may happen that there is no such stream, in this case just do nothing. + else + { + settings.path.pop_back(); + return; + } + settings.path.pop_back(); - /// Second, calculate limits for each variant by iterating through new discriminators. - std::vector variant_limits(variants.size(), 0); - auto & discriminators_data = col.getLocalDiscriminators(); - size_t discriminators_offset = discriminators_data.size() - limit; - for (size_t i = discriminators_offset; i != discriminators_data.size(); ++i) + /// Second, calculate limits for each variant by iterating through new discriminators + /// if we didn't do it during discriminators deserialization. + if (variant_limits.empty()) { - ColumnVariant::Discriminator discr = discriminators_data[i]; - if (discr != ColumnVariant::NULL_DISCRIMINATOR) - ++variant_limits[discr]; + variant_limits.resize(variants.size(), 0); + auto & discriminators_data = col.getLocalDiscriminators(); + + /// We can actually read less than limit discriminators and we cannot determine the actual number of read rows + /// by discriminators column as it could be taken from the substreams cache. And we need actual number of read + /// rows to fill offsets correctly later if they are not in the cache. We can determine if offsets column is in cache + /// or not by comparing it with discriminators column size (they should be the same when offsets are in cache). + /// If offsets are not in the cache, we can use it's size to determine the actual number of read rows. + size_t num_new_discriminators = limit; + size_t offsets_size = col.getOffsetsPtr()->size(); + if (discriminators_data.size() > offsets_size) + num_new_discriminators = discriminators_data.size() - offsets_size; + size_t discriminators_offset = discriminators_data.size() - num_new_discriminators; + + for (size_t i = discriminators_offset; i != discriminators_data.size(); ++i) + { + ColumnVariant::Discriminator discr = discriminators_data[i]; + if (discr != ColumnVariant::NULL_DISCRIMINATOR) + ++variant_limits[discr]; + } } /// Now we can deserialize variants according to their limits. - auto * variant_state = checkAndGetState(state); settings.path.push_back(Substream::VariantElements); for (size_t i = 0; i != variants.size(); ++i) { addVariantElementToPath(settings.path, i); - variants[i]->deserializeBinaryBulkWithMultipleStreams(col.getVariantPtrByLocalDiscriminator(i), variant_limits[i], settings, variant_state->states[i], cache); + variants[i]->deserializeBinaryBulkWithMultipleStreams(col.getVariantPtrByLocalDiscriminator(i), variant_limits[i], settings, variant_state->variant_states[i], cache); settings.path.pop_back(); } settings.path.pop_back(); @@ -336,20 +500,49 @@ void SerializationVariant::deserializeBinaryBulkWithMultipleStreams( } else { - auto & offsets = col.getOffsets(); - offsets.reserve(offsets.size() + limit); std::vector variant_offsets; variant_offsets.reserve(variants.size()); + size_t num_non_empty_variants = 0; + ColumnVariant::Discriminator last_non_empty_discr = 0; for (size_t i = 0; i != variants.size(); ++i) - variant_offsets.push_back(col.getVariantByLocalDiscriminator(i).size() - variant_limits[i]); - - for (size_t i = discriminators_offset; i != discriminators_data.size(); ++i) { - ColumnVariant::Discriminator discr = discriminators_data[i]; - if (discr == ColumnVariant::NULL_DISCRIMINATOR) - offsets.emplace_back(); - else - offsets.push_back(variant_offsets[discr]++); + if (variant_limits[i]) + { + ++num_non_empty_variants; + last_non_empty_discr = i; + } + + variant_offsets.push_back(col.getVariantByLocalDiscriminator(i).size() - variant_limits[i]); + } + + auto & discriminators_data = col.getLocalDiscriminators(); + auto & offsets = col.getOffsets(); + size_t num_new_offsets = discriminators_data.size() - offsets.size(); + offsets.reserve(offsets.size() + num_new_offsets); + /// If there are only NULLs were read, fill offsets with 0. + if (num_non_empty_variants == 0) + { + offsets.resize_fill(discriminators_data.size(), 0); + } + /// If there is only 1 variant and no NULLs was read, fill offsets with sequential offsets of this variant. + else if (num_non_empty_variants == 1 && variant_limits[last_non_empty_discr] == num_new_offsets) + { + size_t first_offset = col.getVariantByLocalDiscriminator(last_non_empty_discr).size() - num_new_offsets; + for (size_t i = 0; i != num_new_offsets; ++i) + offsets.push_back(first_offset + i); + } + /// Otherwise iterate through discriminators and fill offsets accordingly. + else + { + size_t start = offsets.size(); + for (size_t i = start; i != discriminators_data.size(); ++i) + { + ColumnVariant::Discriminator discr = discriminators_data[i]; + if (discr == ColumnVariant::NULL_DISCRIMINATOR) + offsets.emplace_back(); + else + offsets.push_back(variant_offsets[discr]++); + } } addToSubstreamsCache(cache, settings.path, col.getOffsetsPtr()); @@ -357,6 +550,72 @@ void SerializationVariant::deserializeBinaryBulkWithMultipleStreams( settings.path.pop_back(); } +std::vector SerializationVariant::deserializeCompactDiscriminators( + DB::ColumnPtr & discriminators_column, + size_t limit, + ReadBuffer * stream, + bool continuous_reading, + DeserializeBinaryBulkStateVariantDiscriminators & state) const +{ + auto & discriminators = assert_cast(*discriminators_column->assumeMutable()); + auto & discriminators_data = discriminators.getData(); + + /// Reset state if we are reading from the start of the granule and not from the previous position in the file. + if (!continuous_reading) + state.remaining_rows_in_granule = 0; + + /// Calculate limits for variants during discriminators deserialization. + std::vector variant_limits(variants.size(), 0); + while (limit) + { + /// If we read all rows from current granule, start reading the next one. + if (state.remaining_rows_in_granule == 0) + { + if (stream->eof()) + return variant_limits; + + readDiscriminatorsGranuleStart(state, stream); + } + + size_t limit_in_granule = std::min(limit, state.remaining_rows_in_granule); + if (state.granule_format == CompactDiscriminatorsGranuleFormat::COMPACT) + { + auto & data = discriminators.getData(); + data.resize_fill(data.size() + limit_in_granule, state.compact_discr); + if (state.compact_discr != ColumnVariant::NULL_DISCRIMINATOR) + variant_limits[state.compact_discr] += limit_in_granule; + } + else + { + SerializationNumber().deserializeBinaryBulk(discriminators, *stream, limit_in_granule, 0); + size_t start = discriminators_data.size() - limit_in_granule; + for (size_t i = start; i != discriminators_data.size(); ++i) + { + ColumnVariant::Discriminator discr = discriminators_data[i]; + if (discr != ColumnVariant::NULL_DISCRIMINATOR) + ++variant_limits[discr]; + } + } + + state.remaining_rows_in_granule -= limit_in_granule; + limit -= limit_in_granule; + } + + return variant_limits; +} + +void SerializationVariant::readDiscriminatorsGranuleStart(DeserializeBinaryBulkStateVariantDiscriminators & state, DB::ReadBuffer * stream) +{ + UInt64 granule_size; + readVarUInt(granule_size, *stream); + state.remaining_rows_in_granule = granule_size; + UInt8 granule_format; + readBinaryLittleEndian(granule_format, *stream); + state.granule_format = static_cast(granule_format); + if (granule_format == CompactDiscriminatorsGranuleFormat::COMPACT) + readBinaryLittleEndian(state.compact_discr, *stream); +} + void SerializationVariant::addVariantElementToPath(DB::ISerialization::SubstreamPath & path, size_t i) const { path.push_back(Substream::VariantElement); diff --git a/src/DataTypes/Serializations/SerializationVariant.h b/src/DataTypes/Serializations/SerializationVariant.h index b6aa1534538..af89632cf81 100644 --- a/src/DataTypes/Serializations/SerializationVariant.h +++ b/src/DataTypes/Serializations/SerializationVariant.h @@ -2,10 +2,18 @@ #include #include +#include namespace DB { + +namespace ErrorCodes +{ + extern const int INCORRECT_DATA; +} + + /// Class for serializing/deserializing column with Variant type. /// It supports both text and binary bulk serializations/deserializations. /// @@ -18,6 +26,17 @@ namespace DB /// /// During binary bulk serialization it transforms local discriminators /// to global and serializes them into a separate stream VariantDiscriminators. +/// There are 2 modes of serialising discriminators: +/// Basic mode, when all discriminators are serialized as is row by row. +/// Compact mode, when we avoid writing the same discriminators in granules when there is +/// only one variant (or only NULLs) in the granule. +/// In compact mode we serialize granules in the following format: +/// +/// There are 2 different formats of granule - plain and compact. +/// Plain format is used when there are different discriminators in this granule, +/// in this format all discriminators are serialized as is row by row. +/// Compact format is used when all discriminators are the same in this granule, +/// in this case only this single discriminator is serialized. /// Each variant is serialized into a separate stream with path VariantElements/VariantElement /// (VariantElements stream is needed for correct sub-columns creation). We store and serialize /// variants in a sparse form (the size of a variant column equals to the number of its discriminator @@ -32,6 +51,25 @@ namespace DB class SerializationVariant : public ISerialization { public: + struct DiscriminatorsSerializationMode + { + enum Value + { + BASIC = 0, /// Store the whole discriminators column. + COMPACT = 1, /// Don't write discriminators in granule if all of them are the same. + }; + + static void checkMode(UInt64 mode) + { + if (mode > Value::COMPACT) + throw Exception(ErrorCodes::INCORRECT_DATA, "Invalid version for SerializationVariant discriminators column."); + } + + explicit DiscriminatorsSerializationMode(UInt64 mode) : value(static_cast(mode)) { checkMode(mode); } + + Value value; + }; + using VariantSerializations = std::vector; explicit SerializationVariant( @@ -123,8 +161,44 @@ public: static std::vector getVariantsDeserializeTextOrder(const DataTypes & variant_types); private: + friend SerializationVariantElement; + friend SerializationVariantElementNullMap; + void addVariantElementToPath(SubstreamPath & path, size_t i) const; + enum CompactDiscriminatorsGranuleFormat + { + PLAIN = 0, /// Granule has different discriminators and they are serialized as is row by row. + COMPACT = 1, /// Granule has single discriminator for all rows and it is serialized as single value. + }; + + struct DeserializeBinaryBulkStateVariantDiscriminators : public ISerialization::DeserializeBinaryBulkState + { + explicit DeserializeBinaryBulkStateVariantDiscriminators(UInt64 mode_) : mode(mode_) + { + } + + DiscriminatorsSerializationMode mode; + + /// Deserialize state of currently read granule in compact mode. + CompactDiscriminatorsGranuleFormat granule_format = CompactDiscriminatorsGranuleFormat::PLAIN; + size_t remaining_rows_in_granule = 0; + ColumnVariant::Discriminator compact_discr = 0; + }; + + static DeserializeBinaryBulkStatePtr deserializeDiscriminatorsStatePrefix( + DeserializeBinaryBulkSettings & settings, + SubstreamsDeserializeStatesCache * cache); + + std::vector deserializeCompactDiscriminators( + ColumnPtr & discriminators_column, + size_t limit, + ReadBuffer * stream, + bool continuous_reading, + DeserializeBinaryBulkStateVariantDiscriminators & state) const; + + static void readDiscriminatorsGranuleStart(DeserializeBinaryBulkStateVariantDiscriminators & state, ReadBuffer * stream); + bool tryDeserializeTextEscapedImpl(IColumn & column, const String & field, const FormatSettings & settings) const; bool tryDeserializeTextQuotedImpl(IColumn & column, const String & field, const FormatSettings & settings) const; bool tryDeserializeWholeTextImpl(IColumn & column, const String & field, const FormatSettings & settings) const; diff --git a/src/DataTypes/Serializations/SerializationVariantElement.cpp b/src/DataTypes/Serializations/SerializationVariantElement.cpp index ec0b4019c2f..03b5d9584e0 100644 --- a/src/DataTypes/Serializations/SerializationVariantElement.cpp +++ b/src/DataTypes/Serializations/SerializationVariantElement.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include @@ -10,9 +11,10 @@ namespace DB namespace ErrorCodes { extern const int NOT_IMPLEMENTED; + extern const int LOGICAL_ERROR; } -struct DeserializeBinaryBulkStateVariantElement : public ISerialization::DeserializeBinaryBulkState +struct SerializationVariantElement::DeserializeBinaryBulkStateVariantElement : public ISerialization::DeserializeBinaryBulkState { /// During deserialization discriminators and variant streams can be shared. /// For example we can read several variant elements together: "select v.UInt32, v.String from table", @@ -24,7 +26,7 @@ struct DeserializeBinaryBulkStateVariantElement : public ISerialization::Deseria /// substream cache correctly. ColumnPtr discriminators; ColumnPtr variant; - + ISerialization::DeserializeBinaryBulkStatePtr discriminators_state; ISerialization::DeserializeBinaryBulkStatePtr variant_element_state; }; @@ -65,7 +67,12 @@ void SerializationVariantElement::serializeBinaryBulkStateSuffix(SerializeBinary void SerializationVariantElement::deserializeBinaryBulkStatePrefix( DeserializeBinaryBulkSettings & settings, DeserializeBinaryBulkStatePtr & state, SubstreamsDeserializeStatesCache * cache) const { + DeserializeBinaryBulkStatePtr discriminators_state = SerializationVariant::deserializeDiscriminatorsStatePrefix(settings, cache); + if (!discriminators_state) + return; + auto variant_element_state = std::make_shared(); + variant_element_state->discriminators_state = discriminators_state; addVariantToPath(settings.path); nested_serialization->deserializeBinaryBulkStatePrefix(settings, variant_element_state->variant_element_state, cache); @@ -86,35 +93,61 @@ void SerializationVariantElement::deserializeBinaryBulkWithMultipleStreams( DeserializeBinaryBulkStatePtr & state, SubstreamsCache * cache) const { - auto * variant_element_state = checkAndGetState(state); - /// First, deserialize discriminators from Variant column. settings.path.push_back(Substream::VariantDiscriminators); + + DeserializeBinaryBulkStateVariantElement * variant_element_state = nullptr; + std::optional variant_limit; if (auto cached_discriminators = getFromSubstreamsCache(cache, settings.path)) { + variant_element_state = checkAndGetState(state); variant_element_state->discriminators = cached_discriminators; } - else + else if (auto * discriminators_stream = settings.getter(settings.path)) { - auto * discriminators_stream = settings.getter(settings.path); - if (!discriminators_stream) - return; + variant_element_state = checkAndGetState(state); + auto * discriminators_state = checkAndGetState(variant_element_state->discriminators_state); /// If we started to read a new column, reinitialize discriminators column in deserialization state. if (!variant_element_state->discriminators || result_column->empty()) variant_element_state->discriminators = ColumnVariant::ColumnDiscriminators::create(); - SerializationNumber().deserializeBinaryBulk(*variant_element_state->discriminators->assumeMutable(), *discriminators_stream, limit, 0); + /// Deserialize discriminators according to serialization mode. + if (discriminators_state->mode.value == SerializationVariant::DiscriminatorsSerializationMode::BASIC) + SerializationNumber().deserializeBinaryBulk(*variant_element_state->discriminators->assumeMutable(), *discriminators_stream, limit, 0); + else + variant_limit = deserializeCompactDiscriminators( + variant_element_state->discriminators, + variant_discriminator, + limit, + discriminators_stream, + settings.continuous_reading, + variant_element_state->discriminators_state, + this); + addToSubstreamsCache(cache, settings.path, variant_element_state->discriminators); } + else + { + settings.path.pop_back(); + return; + } + settings.path.pop_back(); - /// Iterate through new discriminators to calculate the limit for our variant. + /// We could read less than limit discriminators, but we will need actual number of read rows later. + size_t num_new_discriminators = variant_element_state->discriminators->size() - result_column->size(); + + /// Iterate through new discriminators to calculate the limit for our variant + /// if we didn't do it during discriminators deserialization. const auto & discriminators_data = assert_cast(*variant_element_state->discriminators).getData(); - size_t discriminators_offset = variant_element_state->discriminators->size() - limit; - size_t variant_limit = 0; - for (size_t i = discriminators_offset; i != discriminators_data.size(); ++i) - variant_limit += (discriminators_data[i] == variant_discriminator); + size_t discriminators_offset = variant_element_state->discriminators->size() - num_new_discriminators; + if (!variant_limit) + { + variant_limit = 0; + for (size_t i = discriminators_offset; i != discriminators_data.size(); ++i) + *variant_limit += (discriminators_data[i] == variant_discriminator); + } /// Now we know the limit for our variant and can deserialize it. @@ -125,19 +158,19 @@ void SerializationVariantElement::deserializeBinaryBulkWithMultipleStreams( auto & nullable_column = assert_cast(*mutable_column); NullMap & null_map = nullable_column.getNullMapData(); /// If we have only our discriminator in range, fill null map with 0. - if (variant_limit == limit) + if (variant_limit == num_new_discriminators) { - null_map.resize_fill(null_map.size() + limit, 0); + null_map.resize_fill(null_map.size() + num_new_discriminators, 0); } /// If no our discriminator in current range, fill null map with 1. else if (variant_limit == 0) { - null_map.resize_fill(null_map.size() + limit, 1); + null_map.resize_fill(null_map.size() + num_new_discriminators, 1); } /// Otherwise we should iterate through discriminators to fill null map. else { - null_map.reserve(null_map.size() + limit); + null_map.reserve(null_map.size() + num_new_discriminators); for (size_t i = discriminators_offset; i != discriminators_data.size(); ++i) null_map.push_back(discriminators_data[i] != variant_discriminator); } @@ -156,15 +189,8 @@ void SerializationVariantElement::deserializeBinaryBulkWithMultipleStreams( assert_cast(*variant_element_state->variant->assumeMutable()).nestedRemoveNullable(); } - /// If nothing to deserialize, just insert defaults. - if (variant_limit == 0) - { - mutable_column->insertManyDefaults(limit); - return; - } - addVariantToPath(settings.path); - nested_serialization->deserializeBinaryBulkWithMultipleStreams(variant_element_state->variant, variant_limit, settings, variant_element_state->variant_element_state, cache); + nested_serialization->deserializeBinaryBulkWithMultipleStreams(variant_element_state->variant, *variant_limit, settings, variant_element_state->variant_element_state, cache); removeVariantFromPath(settings.path); /// If nothing was deserialized when variant_limit > 0 @@ -173,16 +199,27 @@ void SerializationVariantElement::deserializeBinaryBulkWithMultipleStreams( /// In this case we should just insert default values. if (variant_element_state->variant->empty()) { - mutable_column->insertManyDefaults(limit); + mutable_column->insertManyDefaults(num_new_discriminators); return; } - size_t variant_offset = variant_element_state->variant->size() - variant_limit; + /// If there was nothing to deserialize or nothing was actually deserialized when variant_limit > 0, just insert defaults. + /// The second case means that we don't have a stream for such sub-column. It may happen during ALTER MODIFY column with Variant extension. + if (variant_limit == 0 || variant_element_state->variant->empty()) + { + mutable_column->insertManyDefaults(num_new_discriminators); + return; + } + + if (variant_element_state->variant->size() < *variant_limit) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Size of deserialized variant column less than the limit: {} < {}", variant_element_state->variant->size(), *variant_limit); + + size_t variant_offset = variant_element_state->variant->size() - *variant_limit; /// If we have only our discriminator in range, insert the whole range to result column. - if (variant_limit == limit) + if (variant_limit == num_new_discriminators) { - mutable_column->insertRangeFrom(*variant_element_state->variant, variant_offset, variant_limit); + mutable_column->insertRangeFrom(*variant_element_state->variant, variant_offset, *variant_limit); } /// Otherwise iterate through discriminators and insert value from variant or default value depending on the discriminator. else @@ -197,6 +234,59 @@ void SerializationVariantElement::deserializeBinaryBulkWithMultipleStreams( } } +size_t SerializationVariantElement::deserializeCompactDiscriminators( + DB::ColumnPtr & discriminators_column, + ColumnVariant::Discriminator variant_discriminator, + size_t limit, + DB::ReadBuffer * stream, + bool continuous_reading, + DeserializeBinaryBulkStatePtr & discriminators_state_, + const ISerialization * serialization) +{ + auto * discriminators_state = checkAndGetState(discriminators_state_, serialization); + auto & discriminators = assert_cast(*discriminators_column->assumeMutable()); + auto & discriminators_data = discriminators.getData(); + + /// Reset state if we are reading from the start of the granule and not from the previous position in the file. + if (!continuous_reading) + discriminators_state->remaining_rows_in_granule = 0; + + /// Calculate our variant limit during discriminators deserialization. + size_t variant_limit = 0; + while (limit) + { + /// If we read all rows from current granule, start reading the next one. + if (discriminators_state->remaining_rows_in_granule == 0) + { + if (stream->eof()) + return variant_limit; + + SerializationVariant::readDiscriminatorsGranuleStart(*discriminators_state, stream); + } + + size_t limit_in_granule = std::min(limit, discriminators_state->remaining_rows_in_granule); + if (discriminators_state->granule_format == SerializationVariant::CompactDiscriminatorsGranuleFormat::COMPACT) + { + auto & data = discriminators.getData(); + data.resize_fill(data.size() + limit_in_granule, discriminators_state->compact_discr); + if (discriminators_state->compact_discr == variant_discriminator) + variant_limit += limit_in_granule; + } + else + { + SerializationNumber().deserializeBinaryBulk(discriminators, *stream, limit_in_granule, 0); + size_t start = discriminators_data.size() - limit_in_granule; + for (size_t i = start; i != discriminators_data.size(); ++i) + variant_limit += (discriminators_data[i] == variant_discriminator); + } + + discriminators_state->remaining_rows_in_granule -= limit_in_granule; + limit -= limit_in_granule; + } + + return variant_limit; +} + void SerializationVariantElement::addVariantToPath(DB::ISerialization::SubstreamPath & path) const { path.push_back(Substream::VariantElements); @@ -214,17 +304,19 @@ SerializationVariantElement::VariantSubcolumnCreator::VariantSubcolumnCreator( const ColumnPtr & local_discriminators_, const String & variant_element_name_, ColumnVariant::Discriminator global_variant_discriminator_, - ColumnVariant::Discriminator local_variant_discriminator_) + ColumnVariant::Discriminator local_variant_discriminator_, + bool make_nullable_) : local_discriminators(local_discriminators_) , variant_element_name(variant_element_name_) , global_variant_discriminator(global_variant_discriminator_) , local_variant_discriminator(local_variant_discriminator_) + , make_nullable(make_nullable_) { } DataTypePtr SerializationVariantElement::VariantSubcolumnCreator::create(const DB::DataTypePtr & prev) const { - return makeNullableOrLowCardinalityNullableSafe(prev); + return make_nullable ? makeNullableOrLowCardinalityNullableSafe(prev) : prev; } SerializationPtr SerializationVariantElement::VariantSubcolumnCreator::create(const DB::SerializationPtr & prev) const @@ -237,12 +329,12 @@ ColumnPtr SerializationVariantElement::VariantSubcolumnCreator::create(const DB: /// Case when original Variant column contained only one non-empty variant and no NULLs. /// In this case just use this variant. if (prev->size() == local_discriminators->size()) - return makeNullableOrLowCardinalityNullableSafe(prev); + return make_nullable ? makeNullableOrLowCardinalityNullableSafe(prev) : prev; /// If this variant is empty, fill result column with default values. if (prev->empty()) { - auto res = makeNullableOrLowCardinalityNullableSafe(prev)->cloneEmpty(); + auto res = make_nullable ? makeNullableOrLowCardinalityNullableSafe(prev)->cloneEmpty() : prev->cloneEmpty(); res->insertManyDefaults(local_discriminators->size()); return res; } @@ -257,16 +349,16 @@ ColumnPtr SerializationVariantElement::VariantSubcolumnCreator::create(const DB: /// Now we can create new column from null-map and variant column using IColumn::expand. auto res_column = IColumn::mutate(prev); - /// Special case for LowCardinality. We want the result to be LowCardinality(Nullable), + /// Special case for LowCardinality when we want the result to be LowCardinality(Nullable), /// but we don't have a good way to apply null-mask for LowCardinality(), so, we first /// convert our column to LowCardinality(Nullable()) and then use expand which will /// fill rows with 0 in mask with default value (that is NULL). - if (prev->lowCardinality()) + if (make_nullable && prev->lowCardinality()) res_column = assert_cast(*res_column).cloneNullable(); res_column->expand(null_map, /*inverted = */ true); - if (res_column->canBeInsideNullable()) + if (make_nullable && prev->canBeInsideNullable()) { auto null_map_col = ColumnUInt8::create(); null_map_col->getData() = std::move(null_map); diff --git a/src/DataTypes/Serializations/SerializationVariantElement.h b/src/DataTypes/Serializations/SerializationVariantElement.h index 0ce0a72e250..69101aea0f5 100644 --- a/src/DataTypes/Serializations/SerializationVariantElement.h +++ b/src/DataTypes/Serializations/SerializationVariantElement.h @@ -9,6 +9,7 @@ namespace DB { class SerializationVariant; +class SerializationVariantElementNullMap; /// Serialization for Variant element when we read it as a subcolumn. class SerializationVariantElement final : public SerializationWrapper @@ -66,12 +67,14 @@ public: const String variant_element_name; const ColumnVariant::Discriminator global_variant_discriminator; const ColumnVariant::Discriminator local_variant_discriminator; + bool make_nullable; VariantSubcolumnCreator( const ColumnPtr & local_discriminators_, const String & variant_element_name_, ColumnVariant::Discriminator global_variant_discriminator_, - ColumnVariant::Discriminator local_variant_discriminator_); + ColumnVariant::Discriminator local_variant_discriminator_, + bool make_nullable_); DataTypePtr create(const DataTypePtr & prev) const override; ColumnPtr create(const ColumnPtr & prev) const override; @@ -79,6 +82,18 @@ public: }; private: friend SerializationVariant; + friend SerializationVariantElementNullMap; + + struct DeserializeBinaryBulkStateVariantElement; + + static size_t deserializeCompactDiscriminators( + ColumnPtr & discriminators_column, + ColumnVariant::Discriminator variant_discriminator, + size_t limit, + ReadBuffer * stream, + bool continuous_reading, + DeserializeBinaryBulkStatePtr & discriminators_state_, + const ISerialization * serialization); void addVariantToPath(SubstreamPath & path) const; void removeVariantFromPath(SubstreamPath & path) const; diff --git a/src/DataTypes/Serializations/SerializationVariantElementNullMap.cpp b/src/DataTypes/Serializations/SerializationVariantElementNullMap.cpp new file mode 100644 index 00000000000..f30da4fecf9 --- /dev/null +++ b/src/DataTypes/Serializations/SerializationVariantElementNullMap.cpp @@ -0,0 +1,190 @@ +#include +#include +#include +#include +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int NOT_IMPLEMENTED; +} + +struct DeserializeBinaryBulkStateVariantElementNullMap : public ISerialization::DeserializeBinaryBulkState +{ + /// During deserialization discriminators streams can be shared. + /// For example we can read several variant elements together: "select v.UInt32, v.String.null from table", + /// or we can read the whole variant and some of variant elements or their subcolumns: "select v, v.UInt32.null from table". + /// To read the same column from the same stream more than once we use substream cache, + /// but this cache stores the whole column, not only the current range. + /// During deserialization of variant elements or their subcolumns discriminators column is not stored + /// in the result column, so we need to store them inside deserialization state, so we can use + /// substream cache correctly. + ColumnPtr discriminators; + ISerialization::DeserializeBinaryBulkStatePtr discriminators_state; +}; + +void SerializationVariantElementNullMap::enumerateStreams( + DB::ISerialization::EnumerateStreamsSettings & settings, + const DB::ISerialization::StreamCallback & callback, + const DB::ISerialization::SubstreamData &) const +{ + /// We will need stream for discriminators during deserialization. + settings.path.push_back(Substream::VariantDiscriminators); + callback(settings.path); + settings.path.pop_back(); +} + +void SerializationVariantElementNullMap::serializeBinaryBulkStatePrefix( + const IColumn &, SerializeBinaryBulkSettings &, SerializeBinaryBulkStatePtr &) const +{ + throw Exception( + ErrorCodes::NOT_IMPLEMENTED, "Method serializeBinaryBulkStatePrefix is not implemented for SerializationVariantElementNullMap"); +} + +void SerializationVariantElementNullMap::serializeBinaryBulkStateSuffix(SerializeBinaryBulkSettings &, SerializeBinaryBulkStatePtr &) const +{ + throw Exception( + ErrorCodes::NOT_IMPLEMENTED, "Method serializeBinaryBulkStateSuffix is not implemented for SerializationVariantElementNullMap"); +} + +void SerializationVariantElementNullMap::deserializeBinaryBulkStatePrefix( + DeserializeBinaryBulkSettings & settings, DeserializeBinaryBulkStatePtr & state, SubstreamsDeserializeStatesCache * cache) const +{ + DeserializeBinaryBulkStatePtr discriminators_state = SerializationVariant::deserializeDiscriminatorsStatePrefix(settings, cache); + if (!discriminators_state) + return; + + auto variant_element_null_map_state = std::make_shared(); + variant_element_null_map_state->discriminators_state = std::move(discriminators_state); + state = std::move(variant_element_null_map_state); +} + +void SerializationVariantElementNullMap::serializeBinaryBulkWithMultipleStreams( + const IColumn &, size_t, size_t, SerializeBinaryBulkSettings &, SerializeBinaryBulkStatePtr &) const +{ + throw Exception( + ErrorCodes::NOT_IMPLEMENTED, + "Method serializeBinaryBulkWithMultipleStreams is not implemented for SerializationVariantElementNullMap"); +} + +void SerializationVariantElementNullMap::deserializeBinaryBulkWithMultipleStreams( + ColumnPtr & result_column, + size_t limit, + DeserializeBinaryBulkSettings & settings, + DeserializeBinaryBulkStatePtr & state, + SubstreamsCache * cache) const +{ + /// Deserialize discriminators from Variant column. + settings.path.push_back(Substream::VariantDiscriminators); + + DeserializeBinaryBulkStateVariantElementNullMap * variant_element_null_map_state = nullptr; + std::optional variant_limit; + if (auto cached_discriminators = getFromSubstreamsCache(cache, settings.path)) + { + variant_element_null_map_state = checkAndGetState(state); + variant_element_null_map_state->discriminators = cached_discriminators; + } + else if (auto * discriminators_stream = settings.getter(settings.path)) + { + variant_element_null_map_state = checkAndGetState(state); + auto * discriminators_state = checkAndGetState( + variant_element_null_map_state->discriminators_state); + + /// If we started to read a new column, reinitialize discriminators column in deserialization state. + if (!variant_element_null_map_state->discriminators || result_column->empty()) + variant_element_null_map_state->discriminators = ColumnVariant::ColumnDiscriminators::create(); + + /// Deserialize discriminators according to serialization mode. + if (discriminators_state->mode.value == SerializationVariant::DiscriminatorsSerializationMode::BASIC) + SerializationNumber().deserializeBinaryBulk( + *variant_element_null_map_state->discriminators->assumeMutable(), *discriminators_stream, limit, 0); + else + variant_limit = SerializationVariantElement::deserializeCompactDiscriminators( + variant_element_null_map_state->discriminators, + variant_discriminator, + limit, + discriminators_stream, + settings.continuous_reading, + variant_element_null_map_state->discriminators_state, + this); + + addToSubstreamsCache(cache, settings.path, variant_element_null_map_state->discriminators); + } + else + { + /// There is no such stream or cached data, it means that there is no Variant column in this part (it could happen after alter table add column). + /// In such cases columns are filled with default values, but for null-map column default value should be 1, not 0. Fill column with 1 here instead. + MutableColumnPtr mutable_column = result_column->assumeMutable(); + auto & data = assert_cast(*mutable_column).getData(); + data.resize_fill(data.size() + limit, 1); + settings.path.pop_back(); + return; + } + settings.path.pop_back(); + + MutableColumnPtr mutable_column = result_column->assumeMutable(); + auto & data = assert_cast(*mutable_column).getData(); + /// Check if there are no such variant in read range. + if (variant_limit && *variant_limit == 0) + { + data.resize_fill(data.size() + limit, 1); + } + /// Check if there is only our variant in read range. + else if (variant_limit && *variant_limit == limit) + { + data.resize_fill(data.size() + limit, 0); + } + /// Iterate through new discriminators to calculate the null map of our variant. + else + { + const auto & discriminators_data + = assert_cast(*variant_element_null_map_state->discriminators).getData(); + size_t discriminators_offset = variant_element_null_map_state->discriminators->size() - limit; + for (size_t i = discriminators_offset; i != discriminators_data.size(); ++i) + data.push_back(discriminators_data[i] != variant_discriminator); + } +} + +SerializationVariantElementNullMap::VariantNullMapSubcolumnCreator::VariantNullMapSubcolumnCreator( + const ColumnPtr & local_discriminators_, + const String & variant_element_name_, + ColumnVariant::Discriminator global_variant_discriminator_, + ColumnVariant::Discriminator local_variant_discriminator_) + : local_discriminators(local_discriminators_) + , variant_element_name(variant_element_name_) + , global_variant_discriminator(global_variant_discriminator_) + , local_variant_discriminator(local_variant_discriminator_) +{ +} + +DataTypePtr SerializationVariantElementNullMap::VariantNullMapSubcolumnCreator::create(const DB::DataTypePtr &) const +{ + return std::make_shared(); +} + +SerializationPtr SerializationVariantElementNullMap::VariantNullMapSubcolumnCreator::create(const DB::SerializationPtr &) const +{ + return std::make_shared(variant_element_name, global_variant_discriminator); +} + +ColumnPtr SerializationVariantElementNullMap::VariantNullMapSubcolumnCreator::create(const DB::ColumnPtr &) const +{ + /// Iterate through discriminators and create null-map for our variant. + auto null_map_col = ColumnUInt8::create(); + auto & null_map_data = null_map_col->getData(); + null_map_data.reserve(local_discriminators->size()); + const auto & local_discriminators_data = assert_cast(*local_discriminators).getData(); + for (auto local_discr : local_discriminators_data) + null_map_data.push_back(local_discr != local_variant_discriminator); + + return null_map_col; +} + + +} diff --git a/src/DataTypes/Serializations/SerializationVariantElementNullMap.h b/src/DataTypes/Serializations/SerializationVariantElementNullMap.h new file mode 100644 index 00000000000..cd81b445189 --- /dev/null +++ b/src/DataTypes/Serializations/SerializationVariantElementNullMap.h @@ -0,0 +1,107 @@ +#pragma once + +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int NOT_IMPLEMENTED; +} + +class SerializationVariant; +class SerializationVariantElement; + +/// Serialization for Variant element null map when we read it as a subcolumn. +/// For example, variant.UInt64.null. +/// It requires separate serialization because there is no actual Nullable column +/// and we should construct null map from variant discriminators. +/// The implementation of deserializeBinaryBulk* methods is similar to SerializationVariantElement, +/// but differs in that there is no need to read the actual data of the variant, only discriminators. +class SerializationVariantElementNullMap final : public SimpleTextSerialization +{ +public: + SerializationVariantElementNullMap(const String & variant_element_name_, ColumnVariant::Discriminator variant_discriminator_) + : variant_element_name(variant_element_name_), variant_discriminator(variant_discriminator_) + { + } + + void enumerateStreams( + EnumerateStreamsSettings & settings, + const StreamCallback & callback, + const SubstreamData & data) const override; + + void serializeBinaryBulkStatePrefix( + const IColumn & column, + SerializeBinaryBulkSettings & settings, + SerializeBinaryBulkStatePtr & state) const override; + + void serializeBinaryBulkStateSuffix( + SerializeBinaryBulkSettings & settings, + SerializeBinaryBulkStatePtr & state) const override; + + void deserializeBinaryBulkStatePrefix( + DeserializeBinaryBulkSettings & settings, + DeserializeBinaryBulkStatePtr & state, + SubstreamsDeserializeStatesCache * cache) const override; + + void serializeBinaryBulkWithMultipleStreams( + const IColumn & column, + size_t offset, + size_t limit, + SerializeBinaryBulkSettings & settings, + SerializeBinaryBulkStatePtr & state) const override; + + void deserializeBinaryBulkWithMultipleStreams( + ColumnPtr & column, + size_t limit, + DeserializeBinaryBulkSettings & settings, + DeserializeBinaryBulkStatePtr & state, + SubstreamsCache * cache) const override; + + void serializeBinary(const Field &, WriteBuffer &, const FormatSettings &) const override { throwNoSerialization(); } + void deserializeBinary(Field &, ReadBuffer &, const FormatSettings &) const override { throwNoSerialization(); } + void serializeBinary(const IColumn &, size_t, WriteBuffer &, const FormatSettings &) const override { throwNoSerialization(); } + void deserializeBinary(IColumn &, ReadBuffer &, const FormatSettings &) const override { throwNoSerialization(); } + void serializeText(const IColumn &, size_t, WriteBuffer &, const FormatSettings &) const override { throwNoSerialization(); } + void deserializeText(IColumn &, ReadBuffer &, const FormatSettings &, bool) const override { throwNoSerialization(); } + bool tryDeserializeText(IColumn &, ReadBuffer &, const FormatSettings &, bool) const override { throwNoSerialization(); } + + struct VariantNullMapSubcolumnCreator : public ISubcolumnCreator + { + const ColumnPtr local_discriminators; + const String variant_element_name; + const ColumnVariant::Discriminator global_variant_discriminator; + const ColumnVariant::Discriminator local_variant_discriminator; + + VariantNullMapSubcolumnCreator( + const ColumnPtr & local_discriminators_, + const String & variant_element_name_, + ColumnVariant::Discriminator global_variant_discriminator_, + ColumnVariant::Discriminator local_variant_discriminator_); + + DataTypePtr create(const DataTypePtr & prev) const override; + ColumnPtr create(const ColumnPtr & prev) const override; + SerializationPtr create(const SerializationPtr & prev) const override; + }; +private: + [[noreturn]] static void throwNoSerialization() + { + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Text/binary serialization is not implemented for variant element null map subcolumn"); + } + + friend SerializationVariant; + friend SerializationVariantElement; + + /// To be able to deserialize Variant element null map as a subcolumn + /// we need variant element type name and global discriminator. + String variant_element_name; + ColumnVariant::Discriminator variant_discriminator; + +}; + +} diff --git a/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp b/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp index 0ae325871fb..033a6ea8a4a 100644 --- a/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp +++ b/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp @@ -72,8 +72,8 @@ extern "C" int LLVMFuzzerTestOneInput(const uint8_t * data, size_t size) DataTypePtr type = DataTypeFactory::instance().get(data_type); FormatSettings settings; - settings.max_binary_string_size = 100; - settings.max_binary_array_size = 100; + settings.binary.max_binary_string_size = 100; + settings.binary.max_binary_string_size = 100; Field field; type->getDefaultSerialization()->deserializeBinary(field, in, settings); diff --git a/src/DataTypes/registerDataTypeDateTime.cpp b/src/DataTypes/registerDataTypeDateTime.cpp index 802356cc108..84a52d4affb 100644 --- a/src/DataTypes/registerDataTypeDateTime.cpp +++ b/src/DataTypes/registerDataTypeDateTime.cpp @@ -108,11 +108,11 @@ static DataTypePtr create64(const ASTPtr & arguments) void registerDataTypeDateTime(DataTypeFactory & factory) { - factory.registerDataType("DateTime", create, DataTypeFactory::CaseInsensitive); - factory.registerDataType("DateTime32", create32, DataTypeFactory::CaseInsensitive); - factory.registerDataType("DateTime64", create64, DataTypeFactory::CaseInsensitive); + factory.registerDataType("DateTime", create, DataTypeFactory::Case::Insensitive); + factory.registerDataType("DateTime32", create32, DataTypeFactory::Case::Insensitive); + factory.registerDataType("DateTime64", create64, DataTypeFactory::Case::Insensitive); - factory.registerAlias("TIMESTAMP", "DateTime", DataTypeFactory::CaseInsensitive); + factory.registerAlias("TIMESTAMP", "DateTime", DataTypeFactory::Case::Insensitive); } } diff --git a/src/DataTypes/tests/gtest_data_types_binary_encoding.cpp b/src/DataTypes/tests/gtest_data_types_binary_encoding.cpp new file mode 100644 index 00000000000..4d0bfc67183 --- /dev/null +++ b/src/DataTypes/tests/gtest_data_types_binary_encoding.cpp @@ -0,0 +1,129 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +using namespace DB; + +namespace DB::ErrorCodes +{ +extern const int UNSUPPORTED_METHOD; +} + + +void check(const DataTypePtr & type) +{ +// std::cerr << "Check " << type->getName() << "\n"; + WriteBufferFromOwnString ostr; + encodeDataType(type, ostr); + ReadBufferFromString istr(ostr.str()); + DataTypePtr decoded_type = decodeDataType(istr); + ASSERT_TRUE(istr.eof()); + ASSERT_EQ(type->getName(), decoded_type->getName()); + ASSERT_TRUE(type->equals(*decoded_type)); +} + +GTEST_TEST(DataTypesBinaryEncoding, EncodeAndDecode) +{ + tryRegisterAggregateFunctions(); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared()); + check(std::make_shared("EST")); + check(std::make_shared("CET")); + check(std::make_shared(3)); + check(std::make_shared(3, "EST")); + check(std::make_shared(3, "CET")); + check(std::make_shared()); + check(std::make_shared(10)); + check(DataTypeFactory::instance().get("Enum8('a' = 1, 'b' = 2, 'c' = 3, 'd' = -128)")); + check(DataTypeFactory::instance().get("Enum16('a' = 1, 'b' = 2, 'c' = 3, 'd' = -1000)")); + check(std::make_shared(3, 6)); + check(std::make_shared(3, 6)); + check(std::make_shared(3, 6)); + check(std::make_shared(3, 6)); + check(std::make_shared()); + check(DataTypeFactory::instance().get("Array(UInt32)")); + check(DataTypeFactory::instance().get("Array(Array(Array(UInt32)))")); + check(DataTypeFactory::instance().get("Tuple(UInt32, String, UUID)")); + check(DataTypeFactory::instance().get("Tuple(UInt32, String, Tuple(UUID, Date, IPv4))")); + check(DataTypeFactory::instance().get("Tuple(c1 UInt32, c2 String, c3 UUID)")); + check(DataTypeFactory::instance().get("Tuple(c1 UInt32, c2 String, c3 Tuple(c4 UUID, c5 Date, c6 IPv4))")); + check(std::make_shared()); + check(std::make_shared(IntervalKind::Kind::Nanosecond)); + check(std::make_shared(IntervalKind::Kind::Microsecond)); + check(DataTypeFactory::instance().get("Nullable(UInt32)")); + check(DataTypeFactory::instance().get("Nullable(Nothing)")); + check(DataTypeFactory::instance().get("Nullable(UUID)")); + check(std::make_shared( + DataTypes{ + std::make_shared(), + std::make_shared(), + DataTypeFactory::instance().get("Array(Array(Array(UInt32)))")}, + DataTypeFactory::instance().get("Tuple(c1 UInt32, c2 String, c3 UUID)"))); + DataTypes argument_types = {std::make_shared()}; + Array parameters = {Field(0.1), Field(0.2)}; + AggregateFunctionProperties properties; + AggregateFunctionPtr function = AggregateFunctionFactory::instance().get("quantiles", NullsAction::EMPTY, argument_types, parameters, properties); + check(std::make_shared(function, argument_types, parameters)); + check(std::make_shared(function, argument_types, parameters, 2)); + check(DataTypeFactory::instance().get("AggregateFunction(sum, UInt64)")); + check(DataTypeFactory::instance().get("AggregateFunction(quantiles(0.5, 0.9), UInt64)")); + check(DataTypeFactory::instance().get("AggregateFunction(sequenceMatch('(?1)(?2)'), Date, UInt8, UInt8)")); + check(DataTypeFactory::instance().get("AggregateFunction(sumMapFiltered([1, 4, 8]), Array(UInt64), Array(UInt64))")); + check(DataTypeFactory::instance().get("LowCardinality(UInt32)")); + check(DataTypeFactory::instance().get("LowCardinality(Nullable(String))")); + check(DataTypeFactory::instance().get("Map(String, UInt32)")); + check(DataTypeFactory::instance().get("Map(String, Map(String, Map(String, UInt32)))")); + check(std::make_shared()); + check(std::make_shared()); + check(DataTypeFactory::instance().get("Variant(String, UInt32, Date32)")); + check(std::make_shared()); + check(std::make_shared(10)); + check(std::make_shared(255)); + check(DataTypeFactory::instance().get("Bool")); + check(DataTypeFactory::instance().get("SimpleAggregateFunction(sum, UInt64)")); + check(DataTypeFactory::instance().get("SimpleAggregateFunction(maxMap, Tuple(Array(UInt32), Array(UInt32)))")); + check(DataTypeFactory::instance().get("SimpleAggregateFunction(groupArrayArray(19), Array(UInt64))")); + check(DataTypeFactory::instance().get("Nested(a UInt32, b UInt32)")); + check(DataTypeFactory::instance().get("Nested(a UInt32, b Nested(c String, d Nested(e Date)))")); + check(DataTypeFactory::instance().get("Ring")); + check(DataTypeFactory::instance().get("Point")); + check(DataTypeFactory::instance().get("Polygon")); + check(DataTypeFactory::instance().get("MultiPolygon")); + check(DataTypeFactory::instance().get("Tuple(Map(LowCardinality(String), Array(AggregateFunction(2, quantiles(0.1, 0.2), Float32))), Array(Array(Tuple(UInt32, Tuple(a Map(String, String), b Nullable(Date), c Variant(Tuple(g String, d Array(UInt32)), Date, Map(String, String)))))))")); +} diff --git a/src/Databases/DDLDependencyVisitor.cpp b/src/Databases/DDLDependencyVisitor.cpp index 75a01a6190f..d149b49d465 100644 --- a/src/Databases/DDLDependencyVisitor.cpp +++ b/src/Databases/DDLDependencyVisitor.cpp @@ -16,6 +16,7 @@ #include #include #include +#include #include @@ -30,8 +31,8 @@ namespace { friend void tryVisitNestedSelect(const String & query, DDLDependencyVisitorData & data); public: - DDLDependencyVisitorData(const ContextPtr & context_, const QualifiedTableName & table_name_, const ASTPtr & ast_) - : create_query(ast_), table_name(table_name_), current_database(context_->getCurrentDatabase()), context(context_) + DDLDependencyVisitorData(const ContextPtr & global_context_, const QualifiedTableName & table_name_, const ASTPtr & ast_, const String & current_database_) + : create_query(ast_), table_name(table_name_), default_database(global_context_->getCurrentDatabase()), current_database(current_database_), global_context(global_context_) { } @@ -71,20 +72,28 @@ namespace ASTPtr create_query; std::unordered_set skip_asts; QualifiedTableName table_name; + String default_database; String current_database; - ContextPtr context; + ContextPtr global_context; TableNamesSet dependencies; /// CREATE TABLE or CREATE DICTIONARY or CREATE VIEW or CREATE TEMPORARY TABLE or CREATE DATABASE query. void visitCreateQuery(const ASTCreateQuery & create) { - QualifiedTableName to_table{create.to_table_id.database_name, create.to_table_id.table_name}; - if (!to_table.table.empty()) + if (create.targets) { - /// TO target_table (for materialized views) - if (to_table.database.empty()) - to_table.database = current_database; - dependencies.emplace(to_table); + for (const auto & target : create.targets->targets) + { + const auto & table_id = target.table_id; + if (!table_id.table_name.empty()) + { + /// TO target_table (for materialized views) + QualifiedTableName target_name{table_id.database_name, table_id.table_name}; + if (target_name.database.empty()) + target_name.database = current_database; + dependencies.emplace(target_name); + } + } } QualifiedTableName as_table{create.as_database, create.as_table}; @@ -95,6 +104,11 @@ namespace as_table.database = current_database; dependencies.emplace(as_table); } + + /// Visit nested select query only for views, for other cases it's not + /// an actual dependency as it will be executed only once to fill the table. + if (create.select && !create.isView()) + skip_asts.insert(create.select); } /// The definition of a dictionary: SOURCE(CLICKHOUSE(...)) LAYOUT(...) LIFETIME(...) @@ -103,8 +117,8 @@ namespace if (!dictionary.source || dictionary.source->name != "clickhouse" || !dictionary.source->elements) return; - auto config = getDictionaryConfigurationFromAST(create_query->as(), context); - auto info = getInfoIfClickHouseDictionarySource(config, context); + auto config = getDictionaryConfigurationFromAST(create_query->as(), global_context); + auto info = getInfoIfClickHouseDictionarySource(config, global_context); /// We consider only dependencies on local tables. if (!info || !info->is_local) @@ -112,14 +126,21 @@ namespace if (!info->table_name.table.empty()) { + /// If database is not specified in dictionary source, use database of the dictionary itself, not the current/default database. if (info->table_name.database.empty()) - info->table_name.database = current_database; + info->table_name.database = table_name.database; dependencies.emplace(std::move(info->table_name)); } else { - /// We don't have a table name, we have a select query instead + /// We don't have a table name, we have a select query instead. + /// All tables from select query in dictionary definition won't + /// use current database, as this query is executed with global context. + /// Use default database from global context while visiting select query. + String current_database_ = current_database; + current_database = default_database; tryVisitNestedSelect(info->query, *this); + current_database = current_database_; } } @@ -176,7 +197,7 @@ namespace if (auto cluster_name = tryGetClusterNameFromArgument(table_engine, 0)) { - auto cluster = context->tryGetCluster(*cluster_name); + auto cluster = global_context->tryGetCluster(*cluster_name); if (cluster && cluster->getLocalShardCount()) has_local_replicas = true; } @@ -231,7 +252,7 @@ namespace { if (auto cluster_name = tryGetClusterNameFromArgument(function, 0)) { - if (auto cluster = context->tryGetCluster(*cluster_name)) + if (auto cluster = global_context->tryGetCluster(*cluster_name)) { if (cluster->getLocalShardCount()) has_local_replicas = true; @@ -303,7 +324,10 @@ namespace try { /// We're just searching for dependencies here, it's not safe to execute subqueries now. - auto evaluated = evaluateConstantExpressionOrIdentifierAsLiteral(arg, context); + /// Use copy of the global_context and set current database, because expressions can contain currentDatabase() function. + ContextMutablePtr global_context_copy = Context::createCopy(global_context); + global_context_copy->setCurrentDatabase(current_database); + auto evaluated = evaluateConstantExpressionOrIdentifierAsLiteral(arg, global_context_copy); const auto * literal = evaluated->as(); if (!literal || (literal->value.getType() != Field::Types::String)) return {}; @@ -444,7 +468,7 @@ namespace ParserSelectWithUnionQuery parser; String description = fmt::format("Query for ClickHouse dictionary {}", data.table_name); String fixed_query = removeWhereConditionPlaceholder(query); - const Settings & settings = data.context->getSettingsRef(); + const Settings & settings = data.global_context->getSettingsRef(); ASTPtr select = parseQuery(parser, fixed_query, description, settings.max_query_size, settings.max_parser_depth, settings.max_parser_backtracks); @@ -459,12 +483,19 @@ namespace } -TableNamesSet getDependenciesFromCreateQuery(const ContextPtr & context, const QualifiedTableName & table_name, const ASTPtr & ast) +TableNamesSet getDependenciesFromCreateQuery(const ContextPtr & global_global_context, const QualifiedTableName & table_name, const ASTPtr & ast, const String & current_database) { - DDLDependencyVisitor::Data data{context, table_name, ast}; + DDLDependencyVisitor::Data data{global_global_context, table_name, ast, current_database}; DDLDependencyVisitor::Visitor visitor{data}; visitor.visit(ast); return std::move(data).getDependencies(); } +TableNamesSet getDependenciesFromDictionaryNestedSelectQuery(const ContextPtr & global_context, const QualifiedTableName & table_name, const ASTPtr & ast, const String & select_query, const String & current_database) +{ + DDLDependencyVisitor::Data data{global_context, table_name, ast, current_database}; + tryVisitNestedSelect(select_query, data); + return std::move(data).getDependencies(); +} + } diff --git a/src/Databases/DDLDependencyVisitor.h b/src/Databases/DDLDependencyVisitor.h index 29ea6298b04..400e6b04108 100644 --- a/src/Databases/DDLDependencyVisitor.h +++ b/src/Databases/DDLDependencyVisitor.h @@ -13,6 +13,9 @@ using TableNamesSet = std::unordered_set; /// Returns a list of all tables explicitly referenced in the create query of a specified table. /// For example, a column default expression can use dictGet() and thus reference a dictionary. /// Does not validate AST, works a best-effort way. -TableNamesSet getDependenciesFromCreateQuery(const ContextPtr & context, const QualifiedTableName & table_name, const ASTPtr & ast); +TableNamesSet getDependenciesFromCreateQuery(const ContextPtr & global_context, const QualifiedTableName & table_name, const ASTPtr & ast, const String & current_database); + +/// Returns a list of all tables explicitly referenced in the select query specified as a dictionary source. +TableNamesSet getDependenciesFromDictionaryNestedSelectQuery(const ContextPtr & global_context, const QualifiedTableName & table_name, const ASTPtr & ast, const String & select_query, const String & current_database); } diff --git a/src/Databases/DDLLoadingDependencyVisitor.cpp b/src/Databases/DDLLoadingDependencyVisitor.cpp index b8690125aaa..67bce915168 100644 --- a/src/Databases/DDLLoadingDependencyVisitor.cpp +++ b/src/Databases/DDLLoadingDependencyVisitor.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -110,19 +111,30 @@ void DDLLoadingDependencyVisitor::visit(const ASTFunctionWithKeyValueArguments & auto config = getDictionaryConfigurationFromAST(data.create_query->as(), data.global_context); auto info = getInfoIfClickHouseDictionarySource(config, data.global_context); - if (!info || !info->is_local || info->table_name.table.empty()) + if (!info || !info->is_local) return; - if (info->table_name.database.empty()) - info->table_name.database = data.default_database; - data.dependencies.emplace(std::move(info->table_name)); + if (!info->table_name.table.empty()) + { + /// If database is not specified in dictionary source, use database of the dictionary itself, not the current/default database. + if (info->table_name.database.empty()) + info->table_name.database = data.table_name.database; + data.dependencies.emplace(std::move(info->table_name)); + } + else + { + /// We don't have a table name, we have a select query instead that will be executed during dictionary loading. + /// We need to find all tables used in this select query and add them to dependencies. + auto select_query_dependencies = getDependenciesFromDictionaryNestedSelectQuery(data.global_context, data.table_name, data.create_query, info->query, data.default_database); + data.dependencies.merge(select_query_dependencies); + } } void DDLLoadingDependencyVisitor::visit(const ASTStorage & storage, Data & data) { if (storage.ttl_table) { - auto ttl_dependensies = getDependenciesFromCreateQuery(data.global_context, data.table_name, storage.ttl_table->ptr()); + auto ttl_dependensies = getDependenciesFromCreateQuery(data.global_context, data.table_name, storage.ttl_table->ptr(), data.default_database); data.dependencies.merge(ttl_dependensies); } @@ -200,6 +212,13 @@ void DDLLoadingDependencyVisitor::extractTableNameFromArgument(const ASTFunction qualified_name.database = table_identifier->getDatabaseName(); qualified_name.table = table_identifier->shortName(); } + else if (arg->as()) + { + /// Allow IN subquery. + /// Do not add tables from the subquery into dependencies, + /// because CREATE will succeed anyway. + return; + } else { assert(false); diff --git a/src/Databases/DDLRenamingVisitor.cpp b/src/Databases/DDLRenamingVisitor.cpp index 6cd414635a0..38e100e2470 100644 --- a/src/Databases/DDLRenamingVisitor.cpp +++ b/src/Databases/DDLRenamingVisitor.cpp @@ -86,12 +86,19 @@ namespace create.as_table = as_table_new.table; } - QualifiedTableName to_table{create.to_table_id.database_name, create.to_table_id.table_name}; - if (!to_table.table.empty() && !to_table.database.empty()) + if (create.targets) { - auto to_table_new = data.renaming_map.getNewTableName(to_table); - if (to_table_new != to_table) - create.to_table_id = StorageID{to_table_new.database, to_table_new.table}; + for (auto & target : create.targets->targets) + { + auto & table_id = target.table_id; + if (!table_id.database_name.empty() && !table_id.table_name.empty()) + { + QualifiedTableName target_name{table_id.database_name, table_id.table_name}; + auto new_target_name = data.renaming_map.getNewTableName(target_name); + if (new_target_name != target_name) + table_id = StorageID{new_target_name.database, new_target_name.table}; + } + } } } diff --git a/src/Databases/DatabaseAtomic.cpp b/src/Databases/DatabaseAtomic.cpp index ccab72cfbae..d86e29ca915 100644 --- a/src/Databases/DatabaseAtomic.cpp +++ b/src/Databases/DatabaseAtomic.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include @@ -12,10 +13,11 @@ #include #include #include -#include "Common/logger_useful.h" +#include #include #include #include +#include namespace fs = std::filesystem; @@ -37,8 +39,10 @@ namespace ErrorCodes class AtomicDatabaseTablesSnapshotIterator final : public DatabaseTablesSnapshotIterator { public: - explicit AtomicDatabaseTablesSnapshotIterator(DatabaseTablesSnapshotIterator && base) - : DatabaseTablesSnapshotIterator(std::move(base)) {} + explicit AtomicDatabaseTablesSnapshotIterator(DatabaseTablesSnapshotIterator && base) noexcept + : DatabaseTablesSnapshotIterator(std::move(base)) + { + } UUID uuid() const override { return table()->getStorageID().uuid; } }; @@ -106,13 +110,25 @@ void DatabaseAtomic::attachTable(ContextPtr /* context_ */, const String & name, StoragePtr DatabaseAtomic::detachTable(ContextPtr /* context */, const String & name) { + // it is important to call the destructors of not_in_use without + // locked mutex to avoid potential deadlock. DetachedTables not_in_use; - std::lock_guard lock(mutex); - auto table = DatabaseOrdinary::detachTableUnlocked(name); - table_name_to_path.erase(name); - detached_tables.emplace(table->getStorageID().uuid, table); - not_in_use = cleanupDetachedTables(); - return table; + StoragePtr detached_table; + { + std::lock_guard lock(mutex); + detached_table = DatabaseOrdinary::detachTableUnlocked(name); + table_name_to_path.erase(name); + detached_tables.emplace(detached_table->getStorageID().uuid, detached_table); + not_in_use = cleanupDetachedTables(); + } + + if (!not_in_use.empty()) + { + not_in_use.clear(); + LOG_DEBUG(log, "Finished removing not used detached tables"); + } + + return detached_table; } void DatabaseAtomic::dropTable(ContextPtr local_context, const String & table_name, bool sync) @@ -397,7 +413,7 @@ DatabaseAtomic::DetachedTables DatabaseAtomic::cleanupDetachedTables() LOG_DEBUG(log, "There are {} detached tables. Start searching non used tables.", detached_tables.size()); while (it != detached_tables.end()) { - if (it->second.unique()) + if (isSharedPtrUnique(it->second)) { not_in_use.emplace(it->first, it->second); it = detached_tables.erase(it); diff --git a/src/Databases/DatabaseAtomic.h b/src/Databases/DatabaseAtomic.h index b59edd479ba..4a4ccfa2573 100644 --- a/src/Databases/DatabaseAtomic.h +++ b/src/Databases/DatabaseAtomic.h @@ -1,7 +1,8 @@ #pragma once -#include #include +#include +#include namespace DB diff --git a/src/Databases/DatabaseDictionary.cpp b/src/Databases/DatabaseDictionary.cpp index adb9a659fcd..942ffe6fd62 100644 --- a/src/Databases/DatabaseDictionary.cpp +++ b/src/Databases/DatabaseDictionary.cpp @@ -10,6 +10,7 @@ #include #include #include +#include namespace DB @@ -104,7 +105,7 @@ ASTPtr DatabaseDictionary::getCreateTableQueryImpl(const String & table_name, Co return {}; } - auto names_and_types = StorageDictionary::getNamesAndTypes(ExternalDictionariesLoader::getDictionaryStructure(*load_result.config)); + auto names_and_types = StorageDictionary::getNamesAndTypes(ExternalDictionariesLoader::getDictionaryStructure(*load_result.config), false); buffer << "CREATE TABLE " << backQuoteIfNeed(getDatabaseName()) << '.' << backQuoteIfNeed(table_name) << " ("; buffer << names_and_types.toNamesAndTypesDescription(); buffer << ") Engine = Dictionary(" << backQuoteIfNeed(table_name) << ")"; diff --git a/src/Databases/DatabaseFactory.cpp b/src/Databases/DatabaseFactory.cpp index 3a4c1423f7c..05a5e057c55 100644 --- a/src/Databases/DatabaseFactory.cpp +++ b/src/Databases/DatabaseFactory.cpp @@ -9,6 +9,7 @@ #include #include #include +#include namespace fs = std::filesystem; diff --git a/src/Databases/DatabaseFilesystem.cpp b/src/Databases/DatabaseFilesystem.cpp index b27a816a60d..31701e665a1 100644 --- a/src/Databases/DatabaseFilesystem.cpp +++ b/src/Databases/DatabaseFilesystem.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include diff --git a/src/Databases/DatabaseHDFS.cpp b/src/Databases/DatabaseHDFS.cpp index 060991d1290..eccaae5f22e 100644 --- a/src/Databases/DatabaseHDFS.cpp +++ b/src/Databases/DatabaseHDFS.cpp @@ -15,6 +15,7 @@ #include #include #include +#include #include diff --git a/src/Databases/DatabaseLazy.cpp b/src/Databases/DatabaseLazy.cpp index 233db07cd68..ca30ee6db15 100644 --- a/src/Databases/DatabaseLazy.cpp +++ b/src/Databases/DatabaseLazy.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include @@ -187,7 +188,14 @@ void DatabaseLazy::attachTable(ContextPtr /* context_ */, const String & table_n it->second.expiration_iterator = cache_expiration_queue.emplace(cache_expiration_queue.end(), current_time, table_name); - CurrentMetrics::add(CurrentMetrics::AttachedTable, 1); + LOG_DEBUG(log, "Add info for detached table {} to snapshot.", backQuote(table_name)); + if (snapshot_detached_tables.contains(table_name)) + { + LOG_DEBUG(log, "Clean info about detached table {} from snapshot.", backQuote(table_name)); + snapshot_detached_tables.erase(table_name); + } + + CurrentMetrics::add(CurrentMetrics::AttachedTable); } StoragePtr DatabaseLazy::detachTable(ContextPtr /* context */, const String & table_name) @@ -203,8 +211,17 @@ StoragePtr DatabaseLazy::detachTable(ContextPtr /* context */, const String & ta if (it->second.expiration_iterator != cache_expiration_queue.end()) cache_expiration_queue.erase(it->second.expiration_iterator); tables_cache.erase(it); + LOG_DEBUG(log, "Add info for detached table {} to snapshot.", backQuote(table_name)); + snapshot_detached_tables.emplace( + table_name, + SnapshotDetachedTable{ + .database = res->getStorageID().database_name, + .table = res->getStorageID().table_name, + .uuid = res->getStorageID().uuid, + .metadata_path = getObjectMetadataPath(table_name), + .is_permanently = false}); - CurrentMetrics::sub(CurrentMetrics::AttachedTable, 1); + CurrentMetrics::sub(CurrentMetrics::AttachedTable); } return res; } @@ -305,7 +322,7 @@ try String table_name = expired_tables.front().table_name; auto it = tables_cache.find(table_name); - if (!it->second.table || it->second.table.unique()) + if (!it->second.table || isSharedPtrUnique(it->second.table)) { LOG_DEBUG(log, "Drop table {} from cache.", backQuote(it->first)); it->second.table.reset(); diff --git a/src/Databases/DatabaseMemory.cpp b/src/Databases/DatabaseMemory.cpp index b82cf885b4a..86bf0471b8f 100644 --- a/src/Databases/DatabaseMemory.cpp +++ b/src/Databases/DatabaseMemory.cpp @@ -154,7 +154,7 @@ void DatabaseMemory::alterTable(ContextPtr local_context, const StorageID & tabl applyMetadataChangesToCreateQuery(it->second, metadata); /// The create query of the table has been just changed, we need to update dependencies too. - auto ref_dependencies = getDependenciesFromCreateQuery(local_context->getGlobalContext(), table_id.getQualifiedName(), it->second); + auto ref_dependencies = getDependenciesFromCreateQuery(local_context->getGlobalContext(), table_id.getQualifiedName(), it->second, local_context->getCurrentDatabase()); auto loading_dependencies = getLoadingDependenciesFromCreateQuery(local_context->getGlobalContext(), table_id.getQualifiedName(), it->second); DatabaseCatalog::instance().updateDependencies(table_id, ref_dependencies, loading_dependencies); } diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index b8154372116..42730cfd53d 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -23,11 +23,13 @@ #include #include #include +#include #include #include #include #include #include +#include namespace fs = std::filesystem; @@ -307,6 +309,16 @@ void DatabaseOnDisk::detachTablePermanently(ContextPtr query_context, const Stri try { FS::createFile(detached_permanently_flag); + + std::lock_guard lock(mutex); + if (const auto it = snapshot_detached_tables.find(table_name); it == snapshot_detached_tables.end()) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "Snapshot doesn't contain info about detached table={}", table_name); + } + else + { + it->second.is_permanently = true; + } } catch (Exception & e) { diff --git a/src/Databases/DatabaseOrdinary.cpp b/src/Databases/DatabaseOrdinary.cpp index 10a8e06e8f0..3ab5d3fa697 100644 --- a/src/Databases/DatabaseOrdinary.cpp +++ b/src/Databases/DatabaseOrdinary.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include @@ -30,6 +31,7 @@ #include #include #include +#include #include #include @@ -187,7 +189,7 @@ void DatabaseOrdinary::loadTablesMetadata(ContextPtr local_context, ParsedTables size_t prev_tables_count = metadata.parsed_tables.size(); size_t prev_total_dictionaries = metadata.total_dictionaries; - auto process_metadata = [&metadata, is_startup, this](const String & file_name) + auto process_metadata = [&metadata, is_startup, local_context, this](const String & file_name) { fs::path path(getMetadataPath()); fs::path file_path(file_name); @@ -195,7 +197,7 @@ void DatabaseOrdinary::loadTablesMetadata(ContextPtr local_context, ParsedTables try { - auto ast = parseQueryFromMetadata(log, getContext(), full_path.string(), /*throw_on_error*/ true, /*remove_empty*/ false); + auto ast = parseQueryFromMetadata(log, local_context, full_path.string(), /*throw_on_error*/ true, /*remove_empty*/ false); if (ast) { FunctionNameNormalizer::visit(ast.get()); @@ -224,8 +226,23 @@ void DatabaseOrdinary::loadTablesMetadata(ContextPtr local_context, ParsedTables if (fs::exists(full_path.string() + detached_suffix)) { const std::string table_name = unescapeForFileName(file_name.substr(0, file_name.size() - 4)); - permanently_detached_tables.push_back(table_name); LOG_DEBUG(log, "Skipping permanently detached table {}.", backQuote(table_name)); + + std::lock_guard lock(mutex); + permanently_detached_tables.push_back(table_name); + + const auto detached_table_name = create_query->getTable(); + + snapshot_detached_tables.emplace( + detached_table_name, + SnapshotDetachedTable{ + .database = create_query->getDatabase(), + .table = detached_table_name, + .uuid = create_query->uuid, + .metadata_path = getObjectMetadataPath(detached_table_name), + .is_permanently = true}); + + LOG_TRACE(log, "Add permanently detached table {} to system.detached_tables", detached_table_name); return; } @@ -487,6 +504,12 @@ DatabaseTablesIteratorPtr DatabaseOrdinary::getTablesIterator(ContextPtr local_c return DatabaseWithOwnTablesBase::getTablesIterator(local_context, filter_by_table_name, skip_not_loaded); } +DatabaseDetachedTablesSnapshotIteratorPtr DatabaseOrdinary::getDetachedTablesIterator( + ContextPtr local_context, const DatabaseOnDisk::FilterByNameFunction & filter_by_table_name, bool skip_not_loaded) const +{ + return DatabaseWithOwnTablesBase::getDetachedTablesIterator(local_context, filter_by_table_name, skip_not_loaded); +} + Strings DatabaseOrdinary::getAllTableNames(ContextPtr) const { std::set unique_names; @@ -539,7 +562,7 @@ void DatabaseOrdinary::alterTable(ContextPtr local_context, const StorageID & ta } /// The create query of the table has been just changed, we need to update dependencies too. - auto ref_dependencies = getDependenciesFromCreateQuery(local_context->getGlobalContext(), table_id.getQualifiedName(), ast); + auto ref_dependencies = getDependenciesFromCreateQuery(local_context->getGlobalContext(), table_id.getQualifiedName(), ast, local_context->getCurrentDatabase()); auto loading_dependencies = getLoadingDependenciesFromCreateQuery(local_context->getGlobalContext(), table_id.getQualifiedName(), ast); DatabaseCatalog::instance().updateDependencies(table_id, ref_dependencies, loading_dependencies); diff --git a/src/Databases/DatabaseOrdinary.h b/src/Databases/DatabaseOrdinary.h index ef00ac8fdfa..c2c5775e5ab 100644 --- a/src/Databases/DatabaseOrdinary.h +++ b/src/Databases/DatabaseOrdinary.h @@ -57,6 +57,9 @@ public: LoadTaskPtr startupDatabaseAsync(AsyncLoader & async_loader, LoadJobSet startup_after, LoadingStrictnessLevel mode) override; DatabaseTablesIteratorPtr getTablesIterator(ContextPtr local_context, const DatabaseOnDisk::FilterByNameFunction & filter_by_table_name, bool skip_not_loaded) const override; + DatabaseDetachedTablesSnapshotIteratorPtr getDetachedTablesIterator( + ContextPtr local_context, const DatabaseOnDisk::FilterByNameFunction & filter_by_table_name, bool skip_not_loaded) const override; + Strings getAllTableNames(ContextPtr context) const override; void alterTable( @@ -64,7 +67,11 @@ public: const StorageID & table_id, const StorageInMemoryMetadata & metadata) override; - Strings getNamesOfPermanentlyDetachedTables() const override { return permanently_detached_tables; } + Strings getNamesOfPermanentlyDetachedTables() const override + { + std::lock_guard lock(mutex); + return permanently_detached_tables; + } protected: virtual void commitAlterTable( @@ -74,7 +81,7 @@ protected: const String & statement, ContextPtr query_context); - Strings permanently_detached_tables; + Strings permanently_detached_tables TSA_GUARDED_BY(mutex); std::unordered_map load_table TSA_GUARDED_BY(mutex); std::unordered_map startup_table TSA_GUARDED_BY(mutex); diff --git a/src/Databases/DatabaseReplicated.cpp b/src/Databases/DatabaseReplicated.cpp index 8779490c8c6..21218e095bf 100644 --- a/src/Databases/DatabaseReplicated.cpp +++ b/src/Databases/DatabaseReplicated.cpp @@ -13,6 +13,8 @@ #include #include #include +#include +#include #include #include #include @@ -557,8 +559,11 @@ void DatabaseReplicated::createEmptyLogEntry(const ZooKeeperPtr & current_zookee bool DatabaseReplicated::waitForReplicaToProcessAllEntries(UInt64 timeout_ms) { - if (!ddl_worker || is_probably_dropped) - return false; + { + std::lock_guard lock{ddl_worker_mutex}; + if (!ddl_worker || is_probably_dropped) + return false; + } return ddl_worker->waitForReplicaToProcessAllEntries(timeout_ms); } @@ -639,7 +644,10 @@ LoadTaskPtr DatabaseReplicated::startupDatabaseAsync(AsyncLoader & async_loader, if (is_probably_dropped) return; - ddl_worker = std::make_unique(this, getContext()); + { + std::lock_guard lock{ddl_worker_mutex}; + ddl_worker = std::make_unique(this, getContext()); + } ddl_worker->startup(); ddl_worker_initialized = true; }); @@ -669,6 +677,178 @@ void DatabaseReplicated::stopLoading() DatabaseAtomic::stopLoading(); } +void DatabaseReplicated::dumpLocalTablesForDebugOnly(const ContextPtr & local_context) const +{ + auto table_names = getAllTableNames(context.lock()); + for (const auto & table_name : table_names) + { + auto ast_ptr = tryGetCreateTableQuery(table_name, local_context); + if (ast_ptr) + LOG_DEBUG(log, "[local] Table {} create query is {}", table_name, queryToString(ast_ptr)); + else + LOG_DEBUG(log, "[local] Table {} has no create query", table_name); + } +} + +void DatabaseReplicated::dumpTablesInZooKeeperForDebugOnly() const +{ + UInt32 max_log_ptr; + auto table_name_to_metadata = tryGetConsistentMetadataSnapshot(getZooKeeper(), max_log_ptr); + for (const auto & [table_name, create_table_query] : table_name_to_metadata) + { + auto query_ast = parseQueryFromMetadataInZooKeeper(table_name, create_table_query); + if (query_ast) + { + LOG_DEBUG(log, "[zookeeper] Table {} create query is {}", table_name, queryToString(query_ast)); + } + else + { + LOG_DEBUG(log, "[zookeeper] Table {} has no create query", table_name); + } + } +} + +void DatabaseReplicated::tryCompareLocalAndZooKeeperTablesAndDumpDiffForDebugOnly(const ContextPtr & local_context) const +{ + UInt32 max_log_ptr; + auto table_name_to_metadata_in_zk = tryGetConsistentMetadataSnapshot(getZooKeeper(), max_log_ptr); + auto table_names_local = getAllTableNames(local_context); + + if (table_name_to_metadata_in_zk.size() != table_names_local.size()) + LOG_DEBUG(log, "Amount of tables in zk {} locally {}", table_name_to_metadata_in_zk.size(), table_names_local.size()); + + std::unordered_set checked_tables; + + for (const auto & table_name : table_names_local) + { + auto local_ast_ptr = tryGetCreateTableQuery(table_name, local_context); + if (table_name_to_metadata_in_zk.contains(table_name)) + { + checked_tables.insert(table_name); + auto create_table_query_in_zk = table_name_to_metadata_in_zk[table_name]; + auto zk_ast_ptr = parseQueryFromMetadataInZooKeeper(table_name, create_table_query_in_zk); + + if (local_ast_ptr == nullptr && zk_ast_ptr == nullptr) + { + LOG_DEBUG(log, "AST for table {} is the same (nullptr) in local and ZK", table_name); + } + else if (local_ast_ptr != nullptr && zk_ast_ptr != nullptr && queryToString(local_ast_ptr) != queryToString(zk_ast_ptr)) + { + LOG_DEBUG(log, "AST differs for table {}, local {}, in zookeeper {}", table_name, queryToString(local_ast_ptr), queryToString(zk_ast_ptr)); + } + else if (local_ast_ptr == nullptr) + { + LOG_DEBUG(log, "AST differs for table {}, local nullptr, in zookeeper {}", table_name, queryToString(zk_ast_ptr)); + } + else if (zk_ast_ptr == nullptr) + { + LOG_DEBUG(log, "AST differs for table {}, local {}, in zookeeper nullptr", table_name, queryToString(local_ast_ptr)); + } + else + { + LOG_DEBUG(log, "AST for table {} is the same in local and ZK", table_name); + } + } + else + { + if (local_ast_ptr == nullptr) + LOG_DEBUG(log, "Table {} exists locally, but missing in ZK", table_name); + else + LOG_DEBUG(log, "Table {} exists locally with AST {}, but missing in ZK", table_name, queryToString(local_ast_ptr)); + } + } + for (const auto & [table_name, table_metadata] : table_name_to_metadata_in_zk) + { + if (!checked_tables.contains(table_name)) + { + auto zk_ast_ptr = parseQueryFromMetadataInZooKeeper(table_name, table_metadata); + if (zk_ast_ptr == nullptr) + LOG_DEBUG(log, "Table {} exists in ZK with AST {}, but missing locally", table_name, queryToString(zk_ast_ptr)); + else + LOG_DEBUG(log, "Table {} exists in ZK, but missing locally", table_name); + } + } +} + +void DatabaseReplicated::checkTableEngine(const ASTCreateQuery & query, ASTStorage & storage, ContextPtr query_context) const +{ + bool replicated_table = storage.engine && + (startsWith(storage.engine->name, "Replicated") || startsWith(storage.engine->name, "Shared")); + if (!replicated_table || !storage.engine->arguments) + return; + + ASTs & args_ref = storage.engine->arguments->children; + ASTs args = args_ref; + if (args.size() < 2) + return; + + /// It can be a constant expression. Try to evaluate it, ignore exception if we cannot. + bool has_expression_argument = args_ref[0]->as() || args_ref[1]->as(); + if (has_expression_argument) + { + try + { + args[0] = evaluateConstantExpressionAsLiteral(args_ref[0]->clone(), query_context); + args[1] = evaluateConstantExpressionAsLiteral(args_ref[1]->clone(), query_context); + } + catch (...) // NOLINT(bugprone-empty-catch) + { + } + } + + ASTLiteral * arg1 = args[0]->as(); + ASTLiteral * arg2 = args[1]->as(); + if (!arg1 || !arg2 || arg1->value.getType() != Field::Types::String || arg2->value.getType() != Field::Types::String) + return; + + String maybe_path = arg1->value.get(); + String maybe_replica = arg2->value.get(); + + /// Looks like it's ReplicatedMergeTree with explicit zookeeper_path and replica_name arguments. + /// Let's ensure that some macros are used. + /// NOTE: we cannot check here that substituted values will be actually different on shards and replicas. + + Macros::MacroExpansionInfo info; + info.table_id = {getDatabaseName(), query.getTable(), query.uuid}; + info.shard = getShardName(); + info.replica = getReplicaName(); + query_context->getMacros()->expand(maybe_path, info); + bool maybe_shard_macros = info.expanded_other; + info.expanded_other = false; + query_context->getMacros()->expand(maybe_replica, info); + bool maybe_replica_macros = info.expanded_other; + bool enable_functional_tests_helper = getContext()->getConfigRef().has("_functional_tests_helper_database_replicated_replace_args_macros"); + + if (!enable_functional_tests_helper) + { + if (query_context->getSettingsRef().database_replicated_allow_replicated_engine_arguments) + LOG_WARNING(log, "It's not recommended to explicitly specify zookeeper_path and replica_name in ReplicatedMergeTree arguments"); + else + throw Exception(ErrorCodes::INCORRECT_QUERY, + "It's not allowed to specify explicit zookeeper_path and replica_name " + "for ReplicatedMergeTree arguments in Replicated database. If you really want to " + "specify them explicitly, enable setting " + "database_replicated_allow_replicated_engine_arguments."); + } + + if (maybe_shard_macros && maybe_replica_macros) + return; + + if (enable_functional_tests_helper && !has_expression_argument) + { + if (maybe_path.empty() || maybe_path.back() != '/') + maybe_path += '/'; + args_ref[0]->as()->value = maybe_path + "auto_{shard}"; + args_ref[1]->as()->value = maybe_replica + "auto_{replica}"; + return; + } + + throw Exception(ErrorCodes::INCORRECT_QUERY, + "Explicit zookeeper_path and replica_name are specified in ReplicatedMergeTree arguments. " + "If you really want to specify it explicitly, then you should use some macros " + "to distinguish different shards and replicas"); +} + bool DatabaseReplicated::checkDigestValid(const ContextPtr & local_context, bool debug_check /* = true */) const { if (debug_check) @@ -694,6 +874,13 @@ bool DatabaseReplicated::checkDigestValid(const ContextPtr & local_context, bool if (local_digest != tables_metadata_digest) { LOG_ERROR(log, "Digest of local metadata ({}) is not equal to in-memory digest ({})", local_digest, tables_metadata_digest); + +#ifndef NDEBUG + dumpLocalTablesForDebugOnly(local_context); + dumpTablesInZooKeeperForDebugOnly(); + tryCompareLocalAndZooKeeperTablesAndDumpDiffForDebugOnly(local_context); +#endif + return false; } @@ -710,6 +897,11 @@ bool DatabaseReplicated::checkDigestValid(const ContextPtr & local_context, bool if (zk_digest != local_digest_str) { LOG_ERROR(log, "Digest of local metadata ({}) is not equal to digest in Keeper ({})", local_digest_str, zk_digest); +#ifndef NDEBUG + dumpLocalTablesForDebugOnly(local_context); + dumpTablesInZooKeeperForDebugOnly(); + tryCompareLocalAndZooKeeperTablesAndDumpDiffForDebugOnly(local_context); +#endif return false; } @@ -727,81 +919,14 @@ void DatabaseReplicated::checkQueryValid(const ASTPtr & query, ContextPtr query_ if (auto * create = query->as()) { - bool replicated_table = create->storage && create->storage->engine && - (startsWith(create->storage->engine->name, "Replicated") || startsWith(create->storage->engine->name, "Shared")); - if (!replicated_table || !create->storage->engine->arguments) - return; + if (create->storage) + checkTableEngine(*create, *create->storage, query_context); - ASTs & args_ref = create->storage->engine->arguments->children; - ASTs args = args_ref; - if (args.size() < 2) - return; - - /// It can be a constant expression. Try to evaluate it, ignore exception if we cannot. - bool has_expression_argument = args_ref[0]->as() || args_ref[1]->as(); - if (has_expression_argument) + if (create->targets) { - try - { - args[0] = evaluateConstantExpressionAsLiteral(args_ref[0]->clone(), query_context); - args[1] = evaluateConstantExpressionAsLiteral(args_ref[1]->clone(), query_context); - } - catch (...) // NOLINT(bugprone-empty-catch) - { - } + for (const auto & inner_table_engine : create->targets->getInnerEngines()) + checkTableEngine(*create, *inner_table_engine, query_context); } - - ASTLiteral * arg1 = args[0]->as(); - ASTLiteral * arg2 = args[1]->as(); - if (!arg1 || !arg2 || arg1->value.getType() != Field::Types::String || arg2->value.getType() != Field::Types::String) - return; - - String maybe_path = arg1->value.get(); - String maybe_replica = arg2->value.get(); - - /// Looks like it's ReplicatedMergeTree with explicit zookeeper_path and replica_name arguments. - /// Let's ensure that some macros are used. - /// NOTE: we cannot check here that substituted values will be actually different on shards and replicas. - - Macros::MacroExpansionInfo info; - info.table_id = {getDatabaseName(), create->getTable(), create->uuid}; - info.shard = getShardName(); - info.replica = getReplicaName(); - query_context->getMacros()->expand(maybe_path, info); - bool maybe_shard_macros = info.expanded_other; - info.expanded_other = false; - query_context->getMacros()->expand(maybe_replica, info); - bool maybe_replica_macros = info.expanded_other; - bool enable_functional_tests_helper = getContext()->getConfigRef().has("_functional_tests_helper_database_replicated_replace_args_macros"); - - if (!enable_functional_tests_helper) - { - if (query_context->getSettingsRef().database_replicated_allow_replicated_engine_arguments) - LOG_WARNING(log, "It's not recommended to explicitly specify zookeeper_path and replica_name in ReplicatedMergeTree arguments"); - else - throw Exception(ErrorCodes::INCORRECT_QUERY, - "It's not allowed to specify explicit zookeeper_path and replica_name " - "for ReplicatedMergeTree arguments in Replicated database. If you really want to " - "specify them explicitly, enable setting " - "database_replicated_allow_replicated_engine_arguments."); - } - - if (maybe_shard_macros && maybe_replica_macros) - return; - - if (enable_functional_tests_helper && !has_expression_argument) - { - if (maybe_path.empty() || maybe_path.back() != '/') - maybe_path += '/'; - args_ref[0]->as()->value = maybe_path + "auto_{shard}"; - args_ref[1]->as()->value = maybe_replica + "auto_{replica}"; - return; - } - - throw Exception(ErrorCodes::INCORRECT_QUERY, - "Explicit zookeeper_path and replica_name are specified in ReplicatedMergeTree arguments. " - "If you really want to specify it explicitly, then you should use some macros " - "to distinguish different shards and replicas"); } } @@ -1165,7 +1290,7 @@ void DatabaseReplicated::recoverLostReplica(const ZooKeeperPtr & current_zookeep /// And QualifiedTableName::parseFromString doesn't handle this. auto qualified_name = QualifiedTableName{.database = getDatabaseName(), .table = table_name}; auto query_ast = parseQueryFromMetadataInZooKeeper(table_name, create_table_query); - tables_dependencies.addDependencies(qualified_name, getDependenciesFromCreateQuery(getContext(), qualified_name, query_ast)); + tables_dependencies.addDependencies(qualified_name, getDependenciesFromCreateQuery(getContext()->getGlobalContext(), qualified_name, query_ast, getContext()->getCurrentDatabase())); } tables_dependencies.checkNoCyclicDependencies(); @@ -1239,7 +1364,7 @@ void DatabaseReplicated::recoverLostReplica(const ZooKeeperPtr & current_zookeep current_zookeeper->set(replica_path + "/digest", toString(tables_metadata_digest)); } -std::map DatabaseReplicated::tryGetConsistentMetadataSnapshot(const ZooKeeperPtr & zookeeper, UInt32 & max_log_ptr) +std::map DatabaseReplicated::tryGetConsistentMetadataSnapshot(const ZooKeeperPtr & zookeeper, UInt32 & max_log_ptr) const { return getConsistentMetadataSnapshotImpl(zookeeper, {}, /* max_retries= */ 10, max_log_ptr); } @@ -1300,7 +1425,7 @@ std::map DatabaseReplicated::getConsistentMetadataSnapshotImpl( return table_name_to_metadata; } -ASTPtr DatabaseReplicated::parseQueryFromMetadataInZooKeeper(const String & node_name, const String & query) +ASTPtr DatabaseReplicated::parseQueryFromMetadataInZooKeeper(const String & node_name, const String & query) const { ParserCreateQuery parser; String description = "in ZooKeeper " + zookeeper_path + "/metadata/" + node_name; @@ -1310,11 +1435,9 @@ ASTPtr DatabaseReplicated::parseQueryFromMetadataInZooKeeper(const String & node if (create.uuid == UUIDHelpers::Nil || create.getTable() != TABLE_WITH_UUID_NAME_PLACEHOLDER || create.database) throw Exception(ErrorCodes::LOGICAL_ERROR, "Got unexpected query from {}: {}", node_name, query); - bool is_materialized_view_with_inner_table = create.is_materialized_view && create.to_table_id.empty(); - create.setDatabase(getDatabaseName()); create.setTable(unescapeForFileName(node_name)); - create.attach = is_materialized_view_with_inner_table; + create.attach = create.is_materialized_view_with_inner_table(); return ast; } @@ -1399,6 +1522,7 @@ void DatabaseReplicated::renameDatabase(ContextPtr query_context, const String & void DatabaseReplicated::stopReplication() { + std::lock_guard lock{ddl_worker_mutex}; if (ddl_worker) ddl_worker->shutdown(); } diff --git a/src/Databases/DatabaseReplicated.h b/src/Databases/DatabaseReplicated.h index eab5b2ff931..0f901538c33 100644 --- a/src/Databases/DatabaseReplicated.h +++ b/src/Databases/DatabaseReplicated.h @@ -107,15 +107,17 @@ private: void fillClusterAuthInfo(String collection_name, const Poco::Util::AbstractConfiguration & config); void checkQueryValid(const ASTPtr & query, ContextPtr query_context) const; + void checkTableEngine(const ASTCreateQuery & query, ASTStorage & storage, ContextPtr query_context) const; + void recoverLostReplica(const ZooKeeperPtr & current_zookeeper, UInt32 our_log_ptr, UInt32 & max_log_ptr); - std::map tryGetConsistentMetadataSnapshot(const ZooKeeperPtr & zookeeper, UInt32 & max_log_ptr); + std::map tryGetConsistentMetadataSnapshot(const ZooKeeperPtr & zookeeper, UInt32 & max_log_ptr) const; std::map getConsistentMetadataSnapshotImpl(const ZooKeeperPtr & zookeeper, const FilterByNameFunction & filter_by_table_name, size_t max_retries, UInt32 & max_log_ptr) const; - ASTPtr parseQueryFromMetadataInZooKeeper(const String & node_name, const String & query); + ASTPtr parseQueryFromMetadataInZooKeeper(const String & node_name, const String & query) const; String readMetadataFile(const String & table_name) const; ClusterPtr getClusterImpl(bool all_groups = false) const; @@ -131,6 +133,11 @@ private: UInt64 getMetadataHash(const String & table_name) const; bool checkDigestValid(const ContextPtr & local_context, bool debug_check = true) const TSA_REQUIRES(metadata_mutex); + /// For debug purposes only, don't use in production code + void dumpLocalTablesForDebugOnly(const ContextPtr & local_context) const; + void dumpTablesInZooKeeperForDebugOnly() const; + void tryCompareLocalAndZooKeeperTablesAndDumpDiffForDebugOnly(const ContextPtr & local_context) const; + void waitDatabaseStarted() const override; void stopLoading() override; @@ -148,6 +155,7 @@ private: std::atomic_bool is_recovering = false; std::atomic_bool ddl_worker_initialized = false; std::unique_ptr ddl_worker; + std::mutex ddl_worker_mutex; UInt32 max_log_ptr_at_creation = 0; /// Usually operation with metadata are single-threaded because of the way replication works, diff --git a/src/Databases/DatabaseReplicatedWorker.cpp b/src/Databases/DatabaseReplicatedWorker.cpp index 31d6f7876a8..1ef88dc03bc 100644 --- a/src/Databases/DatabaseReplicatedWorker.cpp +++ b/src/Databases/DatabaseReplicatedWorker.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include namespace fs = std::filesystem; diff --git a/src/Databases/DatabaseS3.cpp b/src/Databases/DatabaseS3.cpp index 1589cc1c75d..2b2d978a846 100644 --- a/src/Databases/DatabaseS3.cpp +++ b/src/Databases/DatabaseS3.cpp @@ -5,6 +5,7 @@ #include #include +#include #include #include #include diff --git a/src/Databases/DatabasesCommon.cpp b/src/Databases/DatabasesCommon.cpp index 6426123bb4f..b8e9231f5c6 100644 --- a/src/Databases/DatabasesCommon.cpp +++ b/src/Databases/DatabasesCommon.cpp @@ -2,12 +2,9 @@ #include #include -#include -#include -#include -#include #include #include +#include #include #include #include @@ -16,6 +13,10 @@ #include #include #include +#include +#include +#include +#include namespace DB @@ -237,6 +238,24 @@ DatabaseTablesIteratorPtr DatabaseWithOwnTablesBase::getTablesIterator(ContextPt return std::make_unique(std::move(filtered_tables), database_name); } +DatabaseDetachedTablesSnapshotIteratorPtr DatabaseWithOwnTablesBase::getDetachedTablesIterator( + ContextPtr, const FilterByNameFunction & filter_by_table_name, bool /* skip_not_loaded */) const +{ + std::lock_guard lock(mutex); + if (!filter_by_table_name) + return std::make_unique(snapshot_detached_tables); + + SnapshotDetachedTables filtered_detached_tables; + for (const auto & [detached_table_name, snapshot] : snapshot_detached_tables) + if (filter_by_table_name(detached_table_name)) + { + filtered_detached_tables.emplace(detached_table_name, snapshot); + } + + + return std::make_unique(std::move(filtered_detached_tables)); +} + bool DatabaseWithOwnTablesBase::empty() const { std::lock_guard lock(mutex); @@ -251,27 +270,39 @@ StoragePtr DatabaseWithOwnTablesBase::detachTable(ContextPtr /* context_ */, con StoragePtr DatabaseWithOwnTablesBase::detachTableUnlocked(const String & table_name) { - StoragePtr res; - auto it = tables.find(table_name); if (it == tables.end()) throw Exception(ErrorCodes::UNKNOWN_TABLE, "Table {}.{} doesn't exist", backQuote(database_name), backQuote(table_name)); - res = it->second; + + auto table_storage = it->second; + + snapshot_detached_tables.emplace( + table_name, + SnapshotDetachedTable{ + .database = it->second->getStorageID().getDatabaseName(), + .table = table_name, + .uuid = it->second->getStorageID().uuid, + .metadata_path = getObjectMetadataPath(table_name), + .is_permanently = false}); + tables.erase(it); - res->is_detached = true; + table_storage->is_detached = true; - if (res->isSystemStorage() == false) - CurrentMetrics::sub(getAttachedCounterForStorage(res), 1); + if (!table_storage->isSystemStorage() && !DatabaseCatalog::isPredefinedDatabase(database_name)) + { + LOG_TEST(log, "Counting detached table {} to database {}", table_name, database_name); + CurrentMetrics::sub(getAttachedCounterForStorage(table_storage)); + } - auto table_id = res->getStorageID(); + auto table_id = table_storage->getStorageID(); if (table_id.hasUUID()) { assert(database_name == DatabaseCatalog::TEMPORARY_DATABASE || getUUID() != UUIDHelpers::Nil); DatabaseCatalog::instance().removeUUIDMapping(table_id.uuid); } - return res; + return table_storage; } void DatabaseWithOwnTablesBase::attachTable(ContextPtr /* context_ */, const String & table_name, const StoragePtr & table, const String &) @@ -300,12 +331,17 @@ void DatabaseWithOwnTablesBase::attachTableUnlocked(const String & table_name, c throw Exception(ErrorCodes::TABLE_ALREADY_EXISTS, "Table {} already exists.", table_id.getFullTableName()); } + snapshot_detached_tables.erase(table_name); + /// It is important to reset is_detached here since in case of RENAME in /// non-Atomic database the is_detached is set to true before RENAME. table->is_detached = false; - if (table->isSystemStorage() == false && table_id.database_name != DatabaseCatalog::SYSTEM_DATABASE) - CurrentMetrics::add(getAttachedCounterForStorage(table), 1); + if (!table->isSystemStorage() && !DatabaseCatalog::isPredefinedDatabase(database_name)) + { + LOG_TEST(log, "Counting attached table {} to database {}", table_name, database_name); + CurrentMetrics::add(getAttachedCounterForStorage(table)); + } } void DatabaseWithOwnTablesBase::shutdown() @@ -337,6 +373,7 @@ void DatabaseWithOwnTablesBase::shutdown() std::lock_guard lock(mutex); tables.clear(); + snapshot_detached_tables.clear(); } DatabaseWithOwnTablesBase::~DatabaseWithOwnTablesBase() diff --git a/src/Databases/DatabasesCommon.h b/src/Databases/DatabasesCommon.h index 2eecf8a564f..1ca49e90c23 100644 --- a/src/Databases/DatabasesCommon.h +++ b/src/Databases/DatabasesCommon.h @@ -37,6 +37,9 @@ public: DatabaseTablesIteratorPtr getTablesIterator(ContextPtr context, const FilterByNameFunction & filter_by_table_name, bool skip_not_loaded) const override; + DatabaseDetachedTablesSnapshotIteratorPtr + getDetachedTablesIterator(ContextPtr context, const FilterByNameFunction & filter_by_table_name, bool skip_not_loaded) const override; + std::vector> getTablesForBackup(const FilterByNameFunction & filter, const ContextPtr & local_context) const override; void createTableRestoredFromBackup(const ASTPtr & create_table_query, ContextMutablePtr local_context, std::shared_ptr restore_coordination, UInt64 timeout_ms) override; @@ -46,12 +49,13 @@ public: protected: Tables tables TSA_GUARDED_BY(mutex); + SnapshotDetachedTables snapshot_detached_tables TSA_GUARDED_BY(mutex); LoggerPtr log; DatabaseWithOwnTablesBase(const String & name_, const String & logger, ContextPtr context); void attachTableUnlocked(const String & table_name, const StoragePtr & table) TSA_REQUIRES(mutex); - StoragePtr detachTableUnlocked(const String & table_name) TSA_REQUIRES(mutex); + StoragePtr detachTableUnlocked(const String & table_name) TSA_REQUIRES(mutex); StoragePtr getTableUnlocked(const String & table_name) const TSA_REQUIRES(mutex); StoragePtr tryGetTableNoWait(const String & table_name) const; }; diff --git a/src/Databases/IDatabase.h b/src/Databases/IDatabase.h index b00f2fe4baf..f94326d220e 100644 --- a/src/Databases/IDatabase.h +++ b/src/Databases/IDatabase.h @@ -5,20 +5,21 @@ #include #include #include +#include #include #include -#include #include +#include #include #include -#include #include #include +#include #include #include +#include #include -#include namespace DB @@ -110,6 +111,57 @@ public: using DatabaseTablesIteratorPtr = std::unique_ptr; +struct SnapshotDetachedTable final +{ + String database; + String table; + UUID uuid = UUIDHelpers::Nil; + String metadata_path; + bool is_permanently{}; +}; + +class DatabaseDetachedTablesSnapshotIterator +{ +private: + SnapshotDetachedTables snapshot; + SnapshotDetachedTables::iterator it; + +protected: + DatabaseDetachedTablesSnapshotIterator(DatabaseDetachedTablesSnapshotIterator && other) noexcept + { + size_t idx = std::distance(other.snapshot.begin(), other.it); + std::swap(snapshot, other.snapshot); + other.it = other.snapshot.end(); + it = snapshot.begin(); + std::advance(it, idx); + } + +public: + explicit DatabaseDetachedTablesSnapshotIterator(const SnapshotDetachedTables & tables_) : snapshot(tables_), it(snapshot.begin()) + { + } + + explicit DatabaseDetachedTablesSnapshotIterator(SnapshotDetachedTables && tables_) : snapshot(std::move(tables_)), it(snapshot.begin()) + { + } + + void next() { ++it; } + + bool isValid() const { return it != snapshot.end(); } + + String database() const { return it->second.database; } + + String table() const { return it->second.table; } + + UUID uuid() const { return it->second.uuid; } + + String metadataPath() const { return it->second.metadata_path; } + + bool isPermanently() const { return it->second.is_permanently; } +}; + +using DatabaseDetachedTablesSnapshotIteratorPtr = std::unique_ptr; + /** Database engine. * It is responsible for: @@ -232,6 +284,12 @@ public: /// Wait for all tables to be loaded and started up. If `skip_not_loaded` is true, then not yet loaded or not yet started up (at the moment of iterator creation) tables are excluded. virtual DatabaseTablesIteratorPtr getTablesIterator(ContextPtr context, const FilterByNameFunction & filter_by_table_name = {}, bool skip_not_loaded = false) const = 0; /// NOLINT + virtual DatabaseDetachedTablesSnapshotIteratorPtr getDetachedTablesIterator( + ContextPtr /*context*/, const FilterByNameFunction & /*filter_by_table_name = {}*/, bool /*skip_not_loaded = false*/) const + { + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Cannot get detached tables for Database{}", getEngineName()); + } + /// Returns list of table names. virtual Strings getAllTableNames(ContextPtr context) const { diff --git a/src/Databases/MySQL/DatabaseMySQL.cpp b/src/Databases/MySQL/DatabaseMySQL.cpp index 1c82131af0d..7aa29018f4d 100644 --- a/src/Databases/MySQL/DatabaseMySQL.cpp +++ b/src/Databases/MySQL/DatabaseMySQL.cpp @@ -2,6 +2,7 @@ #if USE_MYSQL # include +# include # include # include # include @@ -29,6 +30,7 @@ # include # include # include +# include # include # include # include @@ -354,7 +356,7 @@ void DatabaseMySQL::cleanOutdatedTables() { for (auto iterator = outdated_tables.begin(); iterator != outdated_tables.end();) { - if (!iterator->unique()) + if (!isSharedPtrUnique(*iterator)) ++iterator; else { diff --git a/src/Databases/MySQL/FetchTablesColumnsList.cpp b/src/Databases/MySQL/FetchTablesColumnsList.cpp index e78f4aa2234..0828438cad2 100644 --- a/src/Databases/MySQL/FetchTablesColumnsList.cpp +++ b/src/Databases/MySQL/FetchTablesColumnsList.cpp @@ -2,6 +2,7 @@ #if USE_MYSQL #include +#include #include #include #include diff --git a/src/Databases/MySQL/FetchTablesColumnsList.h b/src/Databases/MySQL/FetchTablesColumnsList.h index 736a0ffd607..ec12cdc73b0 100644 --- a/src/Databases/MySQL/FetchTablesColumnsList.h +++ b/src/Databases/MySQL/FetchTablesColumnsList.h @@ -12,11 +12,12 @@ #include #include -#include namespace DB { +struct Settings; + std::map fetchTablesColumnsList( mysqlxx::PoolWithFailover & pool, const String & database_name, diff --git a/src/Databases/MySQL/MaterializeMetadata.h b/src/Databases/MySQL/MaterializeMetadata.h index e78b7132c8d..36d3fb569d8 100644 --- a/src/Databases/MySQL/MaterializeMetadata.h +++ b/src/Databases/MySQL/MaterializeMetadata.h @@ -6,6 +6,7 @@ #include #include +#include #include #include #include diff --git a/src/Databases/MySQL/MaterializedMySQLSyncThread.cpp b/src/Databases/MySQL/MaterializedMySQLSyncThread.cpp index 7ab4235feeb..2c342755337 100644 --- a/src/Databases/MySQL/MaterializedMySQLSyncThread.cpp +++ b/src/Databases/MySQL/MaterializedMySQLSyncThread.cpp @@ -29,6 +29,7 @@ #include #include #include +#include #include #include #include @@ -532,13 +533,17 @@ static inline void dumpDataForTables( bool MaterializedMySQLSyncThread::prepareSynchronized(MaterializeMetadata & metadata) { bool opened_transaction = false; - mysqlxx::PoolWithFailover::Entry connection; while (!isCancelled()) { try { - connection = pool.tryGet(); + mysqlxx::PoolWithFailover::Entry connection = pool.tryGet(); + SCOPE_EXIT({ + if (opened_transaction) + connection->query("ROLLBACK").execute(); + }); + if (connection.isNull()) { if (settings->max_wait_time_when_mysql_unavailable < 0) @@ -602,9 +607,6 @@ bool MaterializedMySQLSyncThread::prepareSynchronized(MaterializeMetadata & meta { tryLogCurrentException(log); - if (opened_transaction) - connection->query("ROLLBACK").execute(); - if (settings->max_wait_time_when_mysql_unavailable < 0) throw; @@ -716,6 +718,16 @@ static void writeFieldsToColumn( null_map_column->insertValue(0); } + else + { + // Column is not null but field is null. It's possible due to overrides + if (field.isNull()) + { + column_to.insertDefault(); + return false; + } + } + return true; }; @@ -791,7 +803,7 @@ static void writeFieldsToColumn( if (write_data_to_null_map(value, index)) { - const String & data = value.get(); + const String & data = value.safeGet(); casted_string_column->insertData(data.data(), data.size()); } } diff --git a/src/Databases/PostgreSQL/DatabaseMaterializedPostgreSQL.cpp b/src/Databases/PostgreSQL/DatabaseMaterializedPostgreSQL.cpp index 7ce03c74c58..6b0548b85c7 100644 --- a/src/Databases/PostgreSQL/DatabaseMaterializedPostgreSQL.cpp +++ b/src/Databases/PostgreSQL/DatabaseMaterializedPostgreSQL.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -529,7 +530,12 @@ void registerDatabaseMaterializedPostgreSQL(DatabaseFactory & factory) } auto connection_info = postgres::formatConnectionString( - configuration.database, configuration.host, configuration.port, configuration.username, configuration.password); + configuration.database, + configuration.host, + configuration.port, + configuration.username, + configuration.password, + args.context->getSettingsRef().postgresql_connection_attempt_timeout); auto postgresql_replica_settings = std::make_unique(); if (engine_define->settings) diff --git a/src/Databases/PostgreSQL/DatabasePostgreSQL.cpp b/src/Databases/PostgreSQL/DatabasePostgreSQL.cpp index 136fb7fd6d2..a846e23cd4f 100644 --- a/src/Databases/PostgreSQL/DatabasePostgreSQL.cpp +++ b/src/Databases/PostgreSQL/DatabasePostgreSQL.cpp @@ -22,6 +22,7 @@ #include #include #include +#include #include namespace fs = std::filesystem; @@ -545,8 +546,9 @@ void registerDatabasePostgreSQL(DatabaseFactory & factory) configuration, settings.postgresql_connection_pool_size, settings.postgresql_connection_pool_wait_timeout, - POSTGRESQL_POOL_WITH_FAILOVER_DEFAULT_MAX_TRIES, - settings.postgresql_connection_pool_auto_close_connection); + settings.postgresql_connection_pool_retries, + settings.postgresql_connection_pool_auto_close_connection, + settings.postgresql_connection_attempt_timeout); return std::make_shared( args.context, diff --git a/src/Databases/SQLite/DatabaseSQLite.cpp b/src/Databases/SQLite/DatabaseSQLite.cpp index e758ea35de5..471730fce29 100644 --- a/src/Databases/SQLite/DatabaseSQLite.cpp +++ b/src/Databases/SQLite/DatabaseSQLite.cpp @@ -3,6 +3,7 @@ #if USE_SQLITE #include +#include #include #include #include @@ -153,6 +154,7 @@ StoragePtr DatabaseSQLite::fetchTable(const String & table_name, ContextPtr loca table_name, ColumnsDescription{*columns}, ConstraintsDescription{}, + /* comment = */ "", local_context); return storage; diff --git a/src/Databases/TablesLoader.cpp b/src/Databases/TablesLoader.cpp index 6aa13b7b759..733e5d53981 100644 --- a/src/Databases/TablesLoader.cpp +++ b/src/Databases/TablesLoader.cpp @@ -137,7 +137,7 @@ void TablesLoader::buildDependencyGraph() { for (const auto & [table_name, table_metadata] : metadata.parsed_tables) { - auto new_ref_dependencies = getDependenciesFromCreateQuery(global_context, table_name, table_metadata.ast); + auto new_ref_dependencies = getDependenciesFromCreateQuery(global_context, table_name, table_metadata.ast, global_context->getCurrentDatabase()); auto new_loading_dependencies = getLoadingDependenciesFromCreateQuery(global_context, table_name, table_metadata.ast); if (!new_ref_dependencies.empty()) diff --git a/src/Dictionaries/DictionaryStructure.cpp b/src/Dictionaries/DictionaryStructure.cpp index c2f2f4a8532..bd9ec57af1f 100644 --- a/src/Dictionaries/DictionaryStructure.cpp +++ b/src/Dictionaries/DictionaryStructure.cpp @@ -58,15 +58,6 @@ std::optional tryGetAttributeUnderlyingType(TypeIndex i } - -DictionarySpecialAttribute::DictionarySpecialAttribute(const Poco::Util::AbstractConfiguration & config, const std::string & config_prefix) - : name{config.getString(config_prefix + ".name", "")}, expression{config.getString(config_prefix + ".expression", "")} -{ - if (name.empty() && !expression.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Element {}.name is empty", config_prefix); -} - - DictionaryStructure::DictionaryStructure(const Poco::Util::AbstractConfiguration & config, const std::string & config_prefix) { std::string structure_prefix = config_prefix + ".structure"; @@ -79,7 +70,8 @@ DictionaryStructure::DictionaryStructure(const Poco::Util::AbstractConfiguration if (has_id) { - id.emplace(config, structure_prefix + ".id"); + static constexpr auto id_default_type = "UInt64"; + id.emplace(makeDictionaryTypedSpecialAttribute(config, structure_prefix + ".id", id_default_type)); } else if (has_key) { diff --git a/src/Dictionaries/DictionaryStructure.h b/src/Dictionaries/DictionaryStructure.h index 55060b1592f..0d44b696d74 100644 --- a/src/Dictionaries/DictionaryStructure.h +++ b/src/Dictionaries/DictionaryStructure.h @@ -89,14 +89,6 @@ constexpr void callOnDictionaryAttributeType(AttributeUnderlyingType type, F && }); } -struct DictionarySpecialAttribute final -{ - const std::string name; - const std::string expression; - - DictionarySpecialAttribute(const Poco::Util::AbstractConfiguration & config, const std::string & config_prefix); -}; - struct DictionaryTypedSpecialAttribute final { const std::string name; @@ -108,7 +100,7 @@ struct DictionaryTypedSpecialAttribute final /// Name of identifier plus list of attributes struct DictionaryStructure final { - std::optional id; + std::optional id; std::optional> key; std::vector attributes; std::unordered_map attribute_name_to_index; diff --git a/src/Dictionaries/DirectDictionary.cpp b/src/Dictionaries/DirectDictionary.cpp index 2c0f7653aff..f3b32402c20 100644 --- a/src/Dictionaries/DirectDictionary.cpp +++ b/src/Dictionaries/DirectDictionary.cpp @@ -1,6 +1,7 @@ #include "DirectDictionary.h" #include +#include #include #include diff --git a/src/Dictionaries/ExecutablePoolDictionarySource.cpp b/src/Dictionaries/ExecutablePoolDictionarySource.cpp index d8111afdc19..217a67453e4 100644 --- a/src/Dictionaries/ExecutablePoolDictionarySource.cpp +++ b/src/Dictionaries/ExecutablePoolDictionarySource.cpp @@ -8,6 +8,8 @@ #include #include +#include + #include #include #include diff --git a/src/Dictionaries/HTTPDictionarySource.cpp b/src/Dictionaries/HTTPDictionarySource.cpp index dae8ec06d30..663c63dd6c6 100644 --- a/src/Dictionaries/HTTPDictionarySource.cpp +++ b/src/Dictionaries/HTTPDictionarySource.cpp @@ -1,12 +1,13 @@ #include "HTTPDictionarySource.h" +#include #include #include #include #include #include #include -#include #include +#include #include #include #include diff --git a/src/Dictionaries/HashedArrayDictionary.cpp b/src/Dictionaries/HashedArrayDictionary.cpp index 9b194acf87f..d7d50dfb0a6 100644 --- a/src/Dictionaries/HashedArrayDictionary.cpp +++ b/src/Dictionaries/HashedArrayDictionary.cpp @@ -2,6 +2,7 @@ #include #include +#include #include #include #include diff --git a/src/Dictionaries/IPAddressDictionary.cpp b/src/Dictionaries/IPAddressDictionary.cpp index 206516edab0..41fafcc162b 100644 --- a/src/Dictionaries/IPAddressDictionary.cpp +++ b/src/Dictionaries/IPAddressDictionary.cpp @@ -1,11 +1,11 @@ #include "IPAddressDictionary.h" -#include -#include + #include #include #include #include #include +#include #include #include #include @@ -23,6 +23,9 @@ #include #include +#include +#include + namespace DB { diff --git a/src/Dictionaries/PolygonDictionaryImplementations.cpp b/src/Dictionaries/PolygonDictionaryImplementations.cpp index e21fadb0e7e..84abeed2731 100644 --- a/src/Dictionaries/PolygonDictionaryImplementations.cpp +++ b/src/Dictionaries/PolygonDictionaryImplementations.cpp @@ -8,6 +8,7 @@ #include #include +#include #include diff --git a/src/Dictionaries/PostgreSQLDictionarySource.cpp b/src/Dictionaries/PostgreSQLDictionarySource.cpp index c7401386e40..f62a9a009d8 100644 --- a/src/Dictionaries/PostgreSQLDictionarySource.cpp +++ b/src/Dictionaries/PostgreSQLDictionarySource.cpp @@ -2,6 +2,7 @@ #include #include +#include #include "DictionarySourceFactory.h" #include "registerDictionaries.h" @@ -205,8 +206,9 @@ void registerDictionarySourcePostgreSQL(DictionarySourceFactory & factory) configuration.replicas_configurations, settings.postgresql_connection_pool_size, settings.postgresql_connection_pool_wait_timeout, - POSTGRESQL_POOL_WITH_FAILOVER_DEFAULT_MAX_TRIES, - settings.postgresql_connection_pool_auto_close_connection); + settings.postgresql_connection_pool_retries, + settings.postgresql_connection_pool_auto_close_connection, + settings.postgresql_connection_attempt_timeout); PostgreSQLDictionarySource::Configuration dictionary_configuration { diff --git a/src/Dictionaries/RegExpTreeDictionary.cpp b/src/Dictionaries/RegExpTreeDictionary.cpp index 830f9d09364..2522de60183 100644 --- a/src/Dictionaries/RegExpTreeDictionary.cpp +++ b/src/Dictionaries/RegExpTreeDictionary.cpp @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -1002,9 +1003,9 @@ void registerDictionaryRegExpTree(DictionaryFactory & factory) dict_struct, std::move(source_ptr), configuration, - context->getSettings().regexp_dict_allow_hyperscan, - context->getSettings().regexp_dict_flag_case_insensitive, - context->getSettings().regexp_dict_flag_dotall); + context->getSettingsRef().regexp_dict_allow_hyperscan, + context->getSettingsRef().regexp_dict_flag_case_insensitive, + context->getSettingsRef().regexp_dict_flag_dotall); }; factory.registerLayout("regexp_tree", create_layout, true); diff --git a/src/Dictionaries/XDBCDictionarySource.cpp b/src/Dictionaries/XDBCDictionarySource.cpp index 1ebfc4a29b0..590d000f16e 100644 --- a/src/Dictionaries/XDBCDictionarySource.cpp +++ b/src/Dictionaries/XDBCDictionarySource.cpp @@ -10,11 +10,13 @@ #include #include #include +#include #include "DictionarySourceFactory.h" #include "DictionaryStructure.h" #include "readInvalidateQuery.h" #include "registerDictionaries.h" #include +#include #include #include #include "config.h" @@ -239,7 +241,7 @@ void registerDictionarySourceXDBC(DictionarySourceFactory & factory) #if USE_ODBC BridgeHelperPtr bridge = std::make_shared>( global_context, - global_context->getSettings().http_receive_timeout, + global_context->getSettingsRef().http_receive_timeout, config.getString(config_prefix + ".odbc.connection_string"), config.getBool(config_prefix + ".settings.odbc_bridge_use_connection_pooling", global_context->getSettingsRef().odbc_bridge_use_connection_pooling)); diff --git a/src/Dictionaries/getDictionaryConfigurationFromAST.cpp b/src/Dictionaries/getDictionaryConfigurationFromAST.cpp index 9ee2027afc7..4ec2e1f5260 100644 --- a/src/Dictionaries/getDictionaryConfigurationFromAST.cpp +++ b/src/Dictionaries/getDictionaryConfigurationFromAST.cpp @@ -382,6 +382,15 @@ void buildPrimaryKeyConfiguration( name_element->appendChild(name); buildAttributeExpressionIfNeeded(doc, id_element, dict_attr); + + if (!dict_attr->type) + return; + + AutoPtr type_element(doc->createElement("type")); + id_element->appendChild(type_element); + + AutoPtr type(doc->createTextNode(queryToString(dict_attr->type))); + type_element->appendChild(type); } else { diff --git a/src/Dictionaries/registerCacheDictionaries.cpp b/src/Dictionaries/registerCacheDictionaries.cpp index b79261955ff..f2d027575f5 100644 --- a/src/Dictionaries/registerCacheDictionaries.cpp +++ b/src/Dictionaries/registerCacheDictionaries.cpp @@ -2,6 +2,7 @@ #include "CacheDictionaryStorage.h" #include "SSDCacheDictionaryStorage.h" #include +#include #include #include diff --git a/src/Dictionaries/registerHashedDictionary.cpp b/src/Dictionaries/registerHashedDictionary.cpp index 5fc4f5d5cb6..068b71170cc 100644 --- a/src/Dictionaries/registerHashedDictionary.cpp +++ b/src/Dictionaries/registerHashedDictionary.cpp @@ -1,8 +1,8 @@ +#include #include #include #include #include - #include namespace DB diff --git a/src/Dictionaries/registerRangeHashedDictionary.cpp b/src/Dictionaries/registerRangeHashedDictionary.cpp index 8123b811198..bc142a6bddc 100644 --- a/src/Dictionaries/registerRangeHashedDictionary.cpp +++ b/src/Dictionaries/registerRangeHashedDictionary.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include diff --git a/src/Disks/DiskFactory.cpp b/src/Disks/DiskFactory.cpp index de7ee5a74f4..4aa7f6ff564 100644 --- a/src/Disks/DiskFactory.cpp +++ b/src/Disks/DiskFactory.cpp @@ -27,7 +27,8 @@ DiskPtr DiskFactory::create( ContextPtr context, const DisksMap & map, bool attach, - bool custom_disk) const + bool custom_disk, + const std::unordered_set & skip_types) const { const auto disk_type = config.getString(config_prefix + ".type", "local"); @@ -38,6 +39,11 @@ DiskPtr DiskFactory::create( "DiskFactory: the disk '{}' has unknown disk type: {}", name, disk_type); } + if (skip_types.contains(found->first)) + { + return nullptr; + } + const auto & disk_creator = found->second; return disk_creator(name, config, config_prefix, context, map, attach, custom_disk); } diff --git a/src/Disks/DiskFactory.h b/src/Disks/DiskFactory.h index d03ffa6a40f..044ce81dbae 100644 --- a/src/Disks/DiskFactory.h +++ b/src/Disks/DiskFactory.h @@ -42,7 +42,8 @@ public: ContextPtr context, const DisksMap & map, bool attach = false, - bool custom_disk = false) const; + bool custom_disk = false, + const std::unordered_set & skip_types = {}) const; private: using DiskTypeRegistry = std::unordered_map; diff --git a/src/Disks/DiskSelector.cpp b/src/Disks/DiskSelector.cpp index a9260a249dd..f45d12618bf 100644 --- a/src/Disks/DiskSelector.cpp +++ b/src/Disks/DiskSelector.cpp @@ -7,7 +7,6 @@ #include #include -#include namespace DB { @@ -27,7 +26,8 @@ void DiskSelector::assertInitialized() const } -void DiskSelector::initialize(const Poco::Util::AbstractConfiguration & config, const String & config_prefix, ContextPtr context, DiskValidator disk_validator) +void DiskSelector::initialize( + const Poco::Util::AbstractConfiguration & config, const String & config_prefix, ContextPtr context, DiskValidator disk_validator) { Poco::Util::AbstractConfiguration::Keys keys; config.keys(config_prefix, keys); @@ -36,6 +36,8 @@ void DiskSelector::initialize(const Poco::Util::AbstractConfiguration & config, constexpr auto default_disk_name = "default"; bool has_default_disk = false; + constexpr auto local_disk_name = "local"; + bool has_local_disk = false; for (const auto & disk_name : keys) { if (!std::all_of(disk_name.begin(), disk_name.end(), isWordCharASCII)) @@ -44,21 +46,31 @@ void DiskSelector::initialize(const Poco::Util::AbstractConfiguration & config, if (disk_name == default_disk_name) has_default_disk = true; + if (disk_name == local_disk_name) + has_local_disk = true; + const auto disk_config_prefix = config_prefix + "." + disk_name; if (disk_validator && !disk_validator(config, disk_config_prefix, disk_name)) continue; - - disks.emplace(disk_name, factory.create(disk_name, config, disk_config_prefix, context, disks)); + auto created_disk + = factory.create(disk_name, config, disk_config_prefix, context, disks, /*attach*/ false, /*custom_disk*/ false, skip_types); + if (created_disk.get()) + { + disks.emplace(disk_name, std::move(created_disk)); + } } if (!has_default_disk) { disks.emplace( - default_disk_name, - std::make_shared( - default_disk_name, context->getPath(), 0, context, config, config_prefix)); + default_disk_name, std::make_shared(default_disk_name, context->getPath(), 0, context, config, config_prefix)); } + if (!has_local_disk && (context->getApplicationType() == Context::ApplicationType::DISKS)) + { + throw_away_local_on_update = true; + disks.emplace(local_disk_name, std::make_shared(local_disk_name, "/", 0, context, config, config_prefix)); + } is_initialized = true; } @@ -76,6 +88,7 @@ DiskSelectorPtr DiskSelector::updateFromConfig( std::shared_ptr result = std::make_shared(*this); constexpr auto default_disk_name = "default"; + constexpr auto local_disk_name = "local"; DisksMap old_disks_minus_new_disks(result->getDisksMap()); for (const auto & disk_name : keys) @@ -86,7 +99,12 @@ DiskSelectorPtr DiskSelector::updateFromConfig( auto disk_config_prefix = config_prefix + "." + disk_name; if (!result->getDisksMap().contains(disk_name)) { - result->addToDiskMap(disk_name, factory.create(disk_name, config, disk_config_prefix, context, result->getDisksMap())); + auto created_disk = factory.create( + disk_name, config, disk_config_prefix, context, result->getDisksMap(), /*attach*/ false, /*custom_disk*/ false, skip_types); + if (created_disk) + { + result->addToDiskMap(disk_name, created_disk); + } } else { @@ -99,6 +117,10 @@ DiskSelectorPtr DiskSelector::updateFromConfig( } old_disks_minus_new_disks.erase(default_disk_name); + if (throw_away_local_on_update) + { + old_disks_minus_new_disks.erase(local_disk_name); + } if (!old_disks_minus_new_disks.empty()) { diff --git a/src/Disks/DiskSelector.h b/src/Disks/DiskSelector.h index 6669b428158..49a1be5cf50 100644 --- a/src/Disks/DiskSelector.h +++ b/src/Disks/DiskSelector.h @@ -20,7 +20,7 @@ class DiskSelector public: static constexpr auto TMP_INTERNAL_DISK_PREFIX = "__tmp_internal_"; - DiskSelector() = default; + explicit DiskSelector(std::unordered_set skip_types_ = {}) : skip_types(skip_types_) { } DiskSelector(const DiskSelector & from) = default; using DiskValidator = std::function; @@ -48,6 +48,10 @@ private: bool is_initialized = false; void assertInitialized() const; + + const std::unordered_set skip_types; + + bool throw_away_local_on_update = false; }; } diff --git a/src/Disks/IDisk.cpp b/src/Disks/IDisk.cpp index 5f0ca850b40..4bbefad5290 100644 --- a/src/Disks/IDisk.cpp +++ b/src/Disks/IDisk.cpp @@ -186,7 +186,7 @@ void IDisk::checkAccess() DB::UUID server_uuid = DB::ServerUUID::get(); if (server_uuid == DB::UUIDHelpers::Nil) throw Exception(ErrorCodes::LOGICAL_ERROR, "Server UUID is not initialized"); - const String path = fmt::format("clickhouse_access_check_{}", DB::toString(server_uuid)); + const String path = fmt::format("clickhouse_access_check_{}", toString(server_uuid)); checkAccessImpl(path); } diff --git a/src/Disks/IDisk.h b/src/Disks/IDisk.h index 4781839cb01..412ad27e94f 100644 --- a/src/Disks/IDisk.h +++ b/src/Disks/IDisk.h @@ -427,7 +427,7 @@ public: /// Device: 10301h/66305d Inode: 3109907 Links: 1 /// Why we have always zero by default? Because normal filesystem /// manages hardlinks by itself. So you can always remove hardlink and all - /// other alive harlinks will not be removed. + /// other alive hardlinks will not be removed. virtual UInt32 getRefCount(const String &) const { return 0; } /// Revision is an incremental counter of disk operation. diff --git a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp index e9c642666d3..198f6c0ea04 100644 --- a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp +++ b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp @@ -59,7 +59,7 @@ CachedOnDiskReadBufferFromFile::CachedOnDiskReadBufferFromFile( std::optional read_until_position_, std::shared_ptr cache_log_) : ReadBufferFromFileBase(use_external_buffer_ ? 0 : settings_.remote_fs_buffer_size, nullptr, 0, file_size_) -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD , log(getLogger(fmt::format("CachedOnDiskReadBufferFromFile({})", cache_key_))) #else , log(getLogger("CachedOnDiskReadBufferFromFile")) @@ -452,7 +452,7 @@ CachedOnDiskReadBufferFromFile::getImplementationBuffer(FileSegment & file_segme { case ReadType::CACHED: { -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD size_t file_size = getFileSizeFromReadBuffer(*read_buffer_for_file_segment); if (file_size == 0 || range.left + file_size <= file_offset_of_buffer_end) throw Exception( @@ -937,7 +937,7 @@ bool CachedOnDiskReadBufferFromFile::nextImplStep() if (!result) { -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD if (read_type == ReadType::CACHED) { size_t cache_file_size = getFileSizeFromReadBuffer(*implementation_buffer); diff --git a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp index c77709c27eb..bb9761a3905 100644 --- a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp +++ b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp @@ -78,7 +78,6 @@ SeekableReadBufferPtr ReadBufferFromRemoteFSGather::createImplementationBuffer(c std::unique_ptr buf; -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD if (with_file_cache) { auto cache_key = settings.remote_fs_cache->createKeyForPath(object_path); @@ -96,7 +95,6 @@ SeekableReadBufferPtr ReadBufferFromRemoteFSGather::createImplementationBuffer(c /* read_until_position */std::nullopt, cache_log); } -#endif /// Can't wrap CachedOnDiskReadBufferFromFile in CachedInMemoryReadBufferFromFile because the /// former doesn't support seeks. diff --git a/src/Disks/IO/ReadBufferFromWebServer.cpp b/src/Disks/IO/ReadBufferFromWebServer.cpp index 7c5de8a13de..837452aa9f2 100644 --- a/src/Disks/IO/ReadBufferFromWebServer.cpp +++ b/src/Disks/IO/ReadBufferFromWebServer.cpp @@ -1,9 +1,12 @@ #include "ReadBufferFromWebServer.h" -#include +#include +#include +#include #include #include -#include +#include + #include diff --git a/src/Disks/IO/WriteBufferFromAzureBlobStorage.cpp b/src/Disks/IO/WriteBufferFromAzureBlobStorage.cpp index a2d21cf49c2..60fa2997c50 100644 --- a/src/Disks/IO/WriteBufferFromAzureBlobStorage.cpp +++ b/src/Disks/IO/WriteBufferFromAzureBlobStorage.cpp @@ -14,6 +14,15 @@ namespace ProfileEvents { extern const Event RemoteWriteThrottlerBytes; extern const Event RemoteWriteThrottlerSleepMicroseconds; + + extern const Event AzureUpload; + extern const Event AzureStageBlock; + extern const Event AzureCommitBlockList; + + extern const Event DiskAzureUpload; + extern const Event DiskAzureStageBlock; + extern const Event DiskAzureCommitBlockList; + } namespace DB @@ -30,7 +39,7 @@ struct WriteBufferFromAzureBlobStorage::PartData size_t data_size = 0; }; -BufferAllocationPolicyPtr createBufferAllocationPolicy(const AzureObjectStorageSettings & settings) +BufferAllocationPolicyPtr createBufferAllocationPolicy(const AzureBlobStorage::RequestSettings & settings) { BufferAllocationPolicy::Settings allocation_settings; allocation_settings.strict_size = settings.strict_upload_part_size; @@ -48,7 +57,7 @@ WriteBufferFromAzureBlobStorage::WriteBufferFromAzureBlobStorage( const String & blob_path_, size_t buf_size_, const WriteSettings & write_settings_, - std::shared_ptr settings_, + std::shared_ptr settings_, ThreadPoolCallbackRunnerUnsafe schedule_) : WriteBufferFromFileBase(buf_size_, nullptr, 0) , log(getLogger("WriteBufferFromAzureBlobStorage")) @@ -134,6 +143,10 @@ void WriteBufferFromAzureBlobStorage::preFinalize() /// then we use single part upload instead of multi part upload if (block_ids.empty() && detached_part_data.size() == 1 && detached_part_data.front().data_size <= max_single_part_upload_size) { + ProfileEvents::increment(ProfileEvents::AzureUpload); + if (blob_container_client->GetClickhouseOptions().IsClientForDisk) + ProfileEvents::increment(ProfileEvents::DiskAzureUpload); + auto part_data = std::move(detached_part_data.front()); auto block_blob_client = blob_container_client->GetBlockBlobClient(blob_path); Azure::Core::IO::MemoryBodyStream memory_stream(reinterpret_cast(part_data.memory.data()), part_data.data_size); @@ -164,6 +177,10 @@ void WriteBufferFromAzureBlobStorage::finalizeImpl() if (!block_ids.empty()) { auto block_blob_client = blob_container_client->GetBlockBlobClient(blob_path); + ProfileEvents::increment(ProfileEvents::AzureCommitBlockList); + if (blob_container_client->GetClickhouseOptions().IsClientForDisk) + ProfileEvents::increment(ProfileEvents::DiskAzureCommitBlockList); + execWithRetry([&](){ block_blob_client.CommitBlockList(block_ids); }, max_unexpected_write_error_retries); LOG_TRACE(log, "Committed {} blocks for blob `{}`", block_ids.size(), blob_path); } @@ -269,6 +286,10 @@ void WriteBufferFromAzureBlobStorage::writePart(WriteBufferFromAzureBlobStorage: auto & data_block_id = std::get<0>(*worker_data); auto block_blob_client = blob_container_client->GetBlockBlobClient(blob_path); + ProfileEvents::increment(ProfileEvents::AzureStageBlock); + if (blob_container_client->GetClickhouseOptions().IsClientForDisk) + ProfileEvents::increment(ProfileEvents::DiskAzureStageBlock); + Azure::Core::IO::MemoryBodyStream memory_stream(reinterpret_cast(std::get<1>(*worker_data).memory.data()), data_size); execWithRetry([&](){ block_blob_client.StageBlock(data_block_id, memory_stream); }, max_unexpected_write_error_retries, data_size); }; diff --git a/src/Disks/IO/WriteBufferFromAzureBlobStorage.h b/src/Disks/IO/WriteBufferFromAzureBlobStorage.h index 10fe871a727..3ee497c4e44 100644 --- a/src/Disks/IO/WriteBufferFromAzureBlobStorage.h +++ b/src/Disks/IO/WriteBufferFromAzureBlobStorage.h @@ -35,7 +35,7 @@ public: const String & blob_path_, size_t buf_size_, const WriteSettings & write_settings_, - std::shared_ptr settings_, + std::shared_ptr settings_, ThreadPoolCallbackRunnerUnsafe schedule_ = {}); ~WriteBufferFromAzureBlobStorage() override; diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageAuth.cpp b/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageAuth.cpp deleted file mode 100644 index 1a5388349f8..00000000000 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageAuth.cpp +++ /dev/null @@ -1,272 +0,0 @@ -#include - -#if USE_AZURE_BLOB_STORAGE - -#include -#include -#include -#include -#include -#include -#include -#include - -using namespace Azure::Storage::Blobs; - - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - - -void validateStorageAccountUrl(const String & storage_account_url) -{ - const auto * storage_account_url_pattern_str = R"(http(()|s)://[a-z0-9-.:]+(()|/)[a-z0-9]*(()|/))"; - static const RE2 storage_account_url_pattern(storage_account_url_pattern_str); - - if (!re2::RE2::FullMatch(storage_account_url, storage_account_url_pattern)) - throw Exception(ErrorCodes::BAD_ARGUMENTS, - "Blob Storage URL is not valid, should follow the format: {}, got: {}", storage_account_url_pattern_str, storage_account_url); -} - - -void validateContainerName(const String & container_name) -{ - auto len = container_name.length(); - if (len < 3 || len > 64) - throw Exception(ErrorCodes::BAD_ARGUMENTS, - "AzureBlob Storage container name is not valid, should have length between 3 and 64, but has length: {}", len); - - const auto * container_name_pattern_str = R"([a-z][a-z0-9-]+)"; - static const RE2 container_name_pattern(container_name_pattern_str); - - if (!re2::RE2::FullMatch(container_name, container_name_pattern)) - throw Exception(ErrorCodes::BAD_ARGUMENTS, - "AzureBlob Storage container name is not valid, should follow the format: {}, got: {}", - container_name_pattern_str, container_name); -} - - -AzureBlobStorageEndpoint processAzureBlobStorageEndpoint(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) -{ - String storage_url; - String account_name; - String container_name; - String prefix; - if (config.has(config_prefix + ".endpoint")) - { - String endpoint = config.getString(config_prefix + ".endpoint"); - - /// For some authentication methods account name is not present in the endpoint - /// 'endpoint_contains_account_name' bool is used to understand how to split the endpoint (default : true) - bool endpoint_contains_account_name = config.getBool(config_prefix + ".endpoint_contains_account_name", true); - - size_t pos = endpoint.find("//"); - if (pos == std::string::npos) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected '//' in endpoint"); - - if (endpoint_contains_account_name) - { - size_t acc_pos_begin = endpoint.find('/', pos+2); - if (acc_pos_begin == std::string::npos) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected account_name in endpoint"); - - storage_url = endpoint.substr(0,acc_pos_begin); - size_t acc_pos_end = endpoint.find('/',acc_pos_begin+1); - - if (acc_pos_end == std::string::npos) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected container_name in endpoint"); - - account_name = endpoint.substr(acc_pos_begin+1,(acc_pos_end-acc_pos_begin)-1); - - size_t cont_pos_end = endpoint.find('/', acc_pos_end+1); - - if (cont_pos_end != std::string::npos) - { - container_name = endpoint.substr(acc_pos_end+1,(cont_pos_end-acc_pos_end)-1); - prefix = endpoint.substr(cont_pos_end+1); - } - else - { - container_name = endpoint.substr(acc_pos_end+1); - } - } - else - { - size_t cont_pos_begin = endpoint.find('/', pos+2); - - if (cont_pos_begin == std::string::npos) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected container_name in endpoint"); - - storage_url = endpoint.substr(0,cont_pos_begin); - size_t cont_pos_end = endpoint.find('/',cont_pos_begin+1); - - if (cont_pos_end != std::string::npos) - { - container_name = endpoint.substr(cont_pos_begin+1,(cont_pos_end-cont_pos_begin)-1); - prefix = endpoint.substr(cont_pos_end+1); - } - else - { - container_name = endpoint.substr(cont_pos_begin+1); - } - } - } - else if (config.has(config_prefix + ".connection_string")) - { - storage_url = config.getString(config_prefix + ".connection_string"); - container_name = config.getString(config_prefix + ".container_name"); - } - else if (config.has(config_prefix + ".storage_account_url")) - { - storage_url = config.getString(config_prefix + ".storage_account_url"); - validateStorageAccountUrl(storage_url); - container_name = config.getString(config_prefix + ".container_name"); - } - else - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected either `storage_account_url` or `connection_string` or `endpoint` in config"); - - if (!container_name.empty()) - validateContainerName(container_name); - std::optional container_already_exists {}; - if (config.has(config_prefix + ".container_already_exists")) - container_already_exists = {config.getBool(config_prefix + ".container_already_exists")}; - return {storage_url, account_name, container_name, prefix, container_already_exists}; -} - - -template -std::unique_ptr getClientWithConnectionString(const String & connection_str, const String & container_name, const BlobClientOptions & client_options) = delete; - -template<> -std::unique_ptr getClientWithConnectionString(const String & connection_str, const String & /*container_name*/, const BlobClientOptions & client_options) -{ - return std::make_unique(BlobServiceClient::CreateFromConnectionString(connection_str, client_options)); -} - -template<> -std::unique_ptr getClientWithConnectionString(const String & connection_str, const String & container_name, const BlobClientOptions & client_options) -{ - return std::make_unique(BlobContainerClient::CreateFromConnectionString(connection_str, container_name, client_options)); -} - -template -std::unique_ptr getAzureBlobStorageClientWithAuth( - const String & url, - const String & container_name, - const Poco::Util::AbstractConfiguration & config, - const String & config_prefix, - const Azure::Storage::Blobs::BlobClientOptions & client_options) -{ - std::string connection_str; - if (config.has(config_prefix + ".connection_string")) - connection_str = config.getString(config_prefix + ".connection_string"); - - if (!connection_str.empty()) - return getClientWithConnectionString(connection_str, container_name, client_options); - - if (config.has(config_prefix + ".account_key") && config.has(config_prefix + ".account_name")) - { - auto storage_shared_key_credential = std::make_shared( - config.getString(config_prefix + ".account_name"), - config.getString(config_prefix + ".account_key") - ); - return std::make_unique(url, storage_shared_key_credential, client_options); - } - - if (config.getBool(config_prefix + ".use_workload_identity", false)) - { - auto workload_identity_credential = std::make_shared(); - return std::make_unique(url, workload_identity_credential, client_options); - } - - auto managed_identity_credential = std::make_shared(); - return std::make_unique(url, managed_identity_credential, client_options); -} - -Azure::Storage::Blobs::BlobClientOptions getAzureBlobClientOptions(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) -{ - Azure::Core::Http::Policies::RetryOptions retry_options; - retry_options.MaxRetries = config.getUInt(config_prefix + ".max_tries", 10); - retry_options.RetryDelay = std::chrono::milliseconds(config.getUInt(config_prefix + ".retry_initial_backoff_ms", 10)); - retry_options.MaxRetryDelay = std::chrono::milliseconds(config.getUInt(config_prefix + ".retry_max_backoff_ms", 1000)); - - using CurlOptions = Azure::Core::Http::CurlTransportOptions; - CurlOptions curl_options; - curl_options.NoSignal = true; - - if (config.has(config_prefix + ".curl_ip_resolve")) - { - auto value = config.getString(config_prefix + ".curl_ip_resolve"); - if (value == "ipv4") - curl_options.IPResolve = CurlOptions::CURL_IPRESOLVE_V4; - else if (value == "ipv6") - curl_options.IPResolve = CurlOptions::CURL_IPRESOLVE_V6; - else - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected value for option 'curl_ip_resolve': {}. Expected one of 'ipv4' or 'ipv6'", value); - } - - Azure::Storage::Blobs::BlobClientOptions client_options; - client_options.Retry = retry_options; - client_options.Transport.Transport = std::make_shared(curl_options); - - client_options.ClickhouseOptions = Azure::Storage::Blobs::ClickhouseClientOptions{.IsClientForDisk=true}; - - return client_options; -} - -std::unique_ptr getAzureBlobContainerClient(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) -{ - auto endpoint = processAzureBlobStorageEndpoint(config, config_prefix); - auto container_name = endpoint.container_name; - auto final_url = endpoint.getEndpoint(); - auto client_options = getAzureBlobClientOptions(config, config_prefix); - - if (endpoint.container_already_exists.value_or(false)) - return getAzureBlobStorageClientWithAuth(final_url, container_name, config, config_prefix, client_options); - - auto blob_service_client = getAzureBlobStorageClientWithAuth(endpoint.getEndpointWithoutContainer(), container_name, config, config_prefix, client_options); - - try - { - return std::make_unique(blob_service_client->CreateBlobContainer(container_name).Value); - } - catch (const Azure::Storage::StorageException & e) - { - /// If container_already_exists is not set (in config), ignore already exists error. - /// (Conflict - The specified container already exists) - if (!endpoint.container_already_exists.has_value() && e.StatusCode == Azure::Core::Http::HttpStatusCode::Conflict) - return getAzureBlobStorageClientWithAuth(final_url, container_name, config, config_prefix, client_options); - throw; - } -} - -std::unique_ptr getAzureBlobStorageSettings(const Poco::Util::AbstractConfiguration & config, const String & config_prefix, ContextPtr context) -{ - std::unique_ptr settings = std::make_unique(); - settings->max_single_part_upload_size = config.getUInt64(config_prefix + ".max_single_part_upload_size", context->getSettings().azure_max_single_part_upload_size); - settings->min_bytes_for_seek = config.getUInt64(config_prefix + ".min_bytes_for_seek", 1024 * 1024); - settings->max_single_read_retries = config.getInt(config_prefix + ".max_single_read_retries", 3); - settings->max_single_download_retries = config.getInt(config_prefix + ".max_single_download_retries", 3); - settings->list_object_keys_size = config.getInt(config_prefix + ".list_object_keys_size", 1000); - settings->min_upload_part_size = config.getUInt64(config_prefix + ".min_upload_part_size", context->getSettings().azure_min_upload_part_size); - settings->max_upload_part_size = config.getUInt64(config_prefix + ".max_upload_part_size", context->getSettings().azure_max_upload_part_size); - settings->max_single_part_copy_size = config.getUInt64(config_prefix + ".max_single_part_copy_size", context->getSettings().azure_max_single_part_copy_size); - settings->use_native_copy = config.getBool(config_prefix + ".use_native_copy", false); - settings->max_blocks_in_multipart_upload = config.getUInt64(config_prefix + ".max_blocks_in_multipart_upload", 50000); - settings->max_unexpected_write_error_retries = config.getUInt64(config_prefix + ".max_unexpected_write_error_retries", context->getSettings().azure_max_unexpected_write_error_retries); - settings->max_inflight_parts_for_one_file = config.getUInt64(config_prefix + ".max_inflight_parts_for_one_file", context->getSettings().azure_max_inflight_parts_for_one_file); - settings->strict_upload_part_size = config.getUInt64(config_prefix + ".strict_upload_part_size", context->getSettings().azure_strict_upload_part_size); - settings->upload_part_size_multiply_factor = config.getUInt64(config_prefix + ".upload_part_size_multiply_factor", context->getSettings().azure_upload_part_size_multiply_factor); - settings->upload_part_size_multiply_parts_count_threshold = config.getUInt64(config_prefix + ".upload_part_size_multiply_parts_count_threshold", context->getSettings().azure_upload_part_size_multiply_parts_count_threshold); - - return settings; -} - -} - -#endif diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageAuth.h b/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageAuth.h deleted file mode 100644 index e4775a053c1..00000000000 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageAuth.h +++ /dev/null @@ -1,58 +0,0 @@ -#pragma once - -#include "config.h" - -#if USE_AZURE_BLOB_STORAGE - -#include -#include - -namespace DB -{ - -struct AzureBlobStorageEndpoint -{ - const String storage_account_url; - const String account_name; - const String container_name; - const String prefix; - const std::optional container_already_exists; - - String getEndpoint() - { - String url = storage_account_url; - if (url.ends_with('/')) - url.pop_back(); - - if (!account_name.empty()) - url += "/" + account_name; - - if (!container_name.empty()) - url += "/" + container_name; - - if (!prefix.empty()) - url += "/" + prefix; - - return url; - } - - String getEndpointWithoutContainer() - { - String url = storage_account_url; - - if (!account_name.empty()) - url += "/" + account_name; - - return url; - } -}; - -std::unique_ptr getAzureBlobContainerClient(const Poco::Util::AbstractConfiguration & config, const String & config_prefix); - -AzureBlobStorageEndpoint processAzureBlobStorageEndpoint(const Poco::Util::AbstractConfiguration & config, const String & config_prefix); - -std::unique_ptr getAzureBlobStorageSettings(const Poco::Util::AbstractConfiguration & config, const String & config_prefix, ContextPtr context); - -} - -#endif diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.cpp b/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.cpp new file mode 100644 index 00000000000..0aa3b9c40b5 --- /dev/null +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.cpp @@ -0,0 +1,352 @@ +#include + +#if USE_AZURE_BLOB_STORAGE + +#include +#include +#include +#include +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; +} + +namespace AzureBlobStorage +{ + +static void validateStorageAccountUrl(const String & storage_account_url) +{ + const auto * storage_account_url_pattern_str = R"(http(()|s)://[a-z0-9-.:]+(()|/)[a-z0-9]*(()|/))"; + static const RE2 storage_account_url_pattern(storage_account_url_pattern_str); + + if (!re2::RE2::FullMatch(storage_account_url, storage_account_url_pattern)) + throw Exception(ErrorCodes::BAD_ARGUMENTS, + "Blob Storage URL is not valid, should follow the format: {}, got: {}", storage_account_url_pattern_str, storage_account_url); +} + +static void validateContainerName(const String & container_name) +{ + auto len = container_name.length(); + if (len < 3 || len > 64) + throw Exception(ErrorCodes::BAD_ARGUMENTS, + "AzureBlob Storage container name is not valid, should have length between 3 and 64, but has length: {}", len); + + const auto * container_name_pattern_str = R"([a-z][a-z0-9-]+)"; + static const RE2 container_name_pattern(container_name_pattern_str); + + if (!re2::RE2::FullMatch(container_name, container_name_pattern)) + throw Exception(ErrorCodes::BAD_ARGUMENTS, + "AzureBlob Storage container name is not valid, should follow the format: {}, got: {}", + container_name_pattern_str, container_name); +} + +static bool isConnectionString(const std::string & candidate) +{ + return !candidate.starts_with("http"); +} + +String ConnectionParams::getConnectionURL() const +{ + if (std::holds_alternative(auth_method)) + { + auto parsed_connection_string = Azure::Storage::_internal::ParseConnectionString(endpoint.storage_account_url); + return parsed_connection_string.BlobServiceUrl.GetAbsoluteUrl(); + } + + return endpoint.storage_account_url; +} + +std::unique_ptr ConnectionParams::createForService() const +{ + return std::visit([this](const T & auth) + { + if constexpr (std::is_same_v) + return std::make_unique(ServiceClient::CreateFromConnectionString(auth.toUnderType(), client_options)); + else + return std::make_unique(endpoint.getEndpointWithoutContainer(), auth, client_options); + }, auth_method); +} + +std::unique_ptr ConnectionParams::createForContainer() const +{ + return std::visit([this](const T & auth) + { + if constexpr (std::is_same_v) + return std::make_unique(ContainerClient::CreateFromConnectionString(auth.toUnderType(), endpoint.container_name, client_options)); + else + return std::make_unique(endpoint.getEndpoint(), auth, client_options); + }, auth_method); +} + +Endpoint processEndpoint(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) +{ + String storage_url; + String account_name; + String container_name; + String prefix; + + auto get_container_name = [&] + { + if (config.has(config_prefix + ".container_name")) + return config.getString(config_prefix + ".container_name"); + + if (config.has(config_prefix + ".container")) + return config.getString(config_prefix + ".container"); + + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected either `container` or `container_name` parameter in config"); + }; + + if (config.has(config_prefix + ".endpoint")) + { + String endpoint = config.getString(config_prefix + ".endpoint"); + + /// For some authentication methods account name is not present in the endpoint + /// 'endpoint_contains_account_name' bool is used to understand how to split the endpoint (default : true) + bool endpoint_contains_account_name = config.getBool(config_prefix + ".endpoint_contains_account_name", true); + + size_t pos = endpoint.find("//"); + if (pos == std::string::npos) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected '//' in endpoint"); + + if (endpoint_contains_account_name) + { + size_t acc_pos_begin = endpoint.find('/', pos + 2); + if (acc_pos_begin == std::string::npos) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected account_name in endpoint"); + + storage_url = endpoint.substr(0, acc_pos_begin); + size_t acc_pos_end = endpoint.find('/', acc_pos_begin + 1); + + if (acc_pos_end == std::string::npos) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected container_name in endpoint"); + + account_name = endpoint.substr(acc_pos_begin + 1, acc_pos_end - acc_pos_begin - 1); + + size_t cont_pos_end = endpoint.find('/', acc_pos_end + 1); + + if (cont_pos_end != std::string::npos) + { + container_name = endpoint.substr(acc_pos_end + 1, cont_pos_end - acc_pos_end - 1); + prefix = endpoint.substr(cont_pos_end + 1); + } + else + { + container_name = endpoint.substr(acc_pos_end + 1); + } + } + else + { + size_t cont_pos_begin = endpoint.find('/', pos + 2); + + if (cont_pos_begin == std::string::npos) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected container_name in endpoint"); + + storage_url = endpoint.substr(0, cont_pos_begin); + size_t cont_pos_end = endpoint.find('/', cont_pos_begin + 1); + + if (cont_pos_end != std::string::npos) + { + container_name = endpoint.substr(cont_pos_begin + 1,cont_pos_end - cont_pos_begin - 1); + prefix = endpoint.substr(cont_pos_end + 1); + } + else + { + container_name = endpoint.substr(cont_pos_begin + 1); + } + } + } + else if (config.has(config_prefix + ".connection_string")) + { + storage_url = config.getString(config_prefix + ".connection_string"); + container_name = get_container_name(); + } + else if (config.has(config_prefix + ".storage_account_url")) + { + storage_url = config.getString(config_prefix + ".storage_account_url"); + validateStorageAccountUrl(storage_url); + container_name = get_container_name(); + } + else + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected either `storage_account_url` or `connection_string` or `endpoint` in config"); + + if (!container_name.empty()) + validateContainerName(container_name); + + std::optional container_already_exists {}; + if (config.has(config_prefix + ".container_already_exists")) + container_already_exists = {config.getBool(config_prefix + ".container_already_exists")}; + + return {storage_url, account_name, container_name, prefix, "", container_already_exists}; +} + +void processURL(const String & url, const String & container_name, Endpoint & endpoint, AuthMethod & auth_method) +{ + endpoint.container_name = container_name; + + if (isConnectionString(url)) + { + endpoint.storage_account_url = url; + auth_method = ConnectionString{url}; + return; + } + + auto pos = url.find('?'); + + /// If conneciton_url does not have '?', then its not SAS + if (pos == std::string::npos) + { + endpoint.storage_account_url = url; + auth_method = std::make_shared(); + } + else + { + endpoint.storage_account_url = url.substr(0, pos); + endpoint.sas_auth = url.substr(pos + 1); + auth_method = std::make_shared(); + } +} + +std::unique_ptr getContainerClient(const ConnectionParams & params, bool readonly) +{ + if (params.endpoint.container_already_exists.value_or(false) || readonly) + return params.createForContainer(); + + try + { + auto service_client = params.createForService(); + return std::make_unique(service_client->CreateBlobContainer(params.endpoint.container_name).Value); + } + catch (const Azure::Storage::StorageException & e) + { + /// If container_already_exists is not set (in config), ignore already exists error. + /// (Conflict - The specified container already exists) + if (!params.endpoint.container_already_exists.has_value() && e.StatusCode == Azure::Core::Http::HttpStatusCode::Conflict) + return params.createForContainer(); + throw; + } +} + +AuthMethod getAuthMethod(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) +{ + if (config.has(config_prefix + ".account_key") && config.has(config_prefix + ".account_name")) + { + return std::make_shared( + config.getString(config_prefix + ".account_name"), + config.getString(config_prefix + ".account_key") + ); + } + + if (config.has(config_prefix + ".connection_string")) + return ConnectionString{config.getString(config_prefix + ".connection_string")}; + + if (config.getBool(config_prefix + ".use_workload_identity", false)) + return std::make_shared(); + + return std::make_shared(); +} + +BlobClientOptions getClientOptions(const RequestSettings & settings, bool for_disk) +{ + Azure::Core::Http::Policies::RetryOptions retry_options; + retry_options.MaxRetries = static_cast(settings.sdk_max_retries); + retry_options.RetryDelay = std::chrono::milliseconds(settings.sdk_retry_initial_backoff_ms); + retry_options.MaxRetryDelay = std::chrono::milliseconds(settings.sdk_retry_max_backoff_ms); + + Azure::Core::Http::CurlTransportOptions curl_options; + curl_options.NoSignal = true; + curl_options.IPResolve = settings.curl_ip_resolve; + + Azure::Storage::Blobs::BlobClientOptions client_options; + client_options.Retry = retry_options; + client_options.Transport.Transport = std::make_shared(curl_options); + client_options.ClickhouseOptions = Azure::Storage::Blobs::ClickhouseClientOptions{.IsClientForDisk=for_disk}; + + return client_options; +} + +std::unique_ptr getRequestSettings(const Settings & query_settings) +{ + auto settings = std::make_unique(); + + settings->max_single_part_upload_size = query_settings.azure_max_single_part_upload_size; + settings->max_single_read_retries = query_settings.azure_max_single_read_retries; + settings->max_single_download_retries = query_settings.azure_max_single_read_retries; + settings->list_object_keys_size = query_settings.azure_list_object_keys_size; + settings->min_upload_part_size = query_settings.azure_min_upload_part_size; + settings->max_upload_part_size = query_settings.azure_max_upload_part_size; + settings->max_single_part_copy_size = query_settings.azure_max_single_part_copy_size; + settings->max_blocks_in_multipart_upload = query_settings.azure_max_blocks_in_multipart_upload; + settings->max_unexpected_write_error_retries = query_settings.azure_max_unexpected_write_error_retries; + settings->max_inflight_parts_for_one_file = query_settings.azure_max_inflight_parts_for_one_file; + settings->strict_upload_part_size = query_settings.azure_strict_upload_part_size; + settings->upload_part_size_multiply_factor = query_settings.azure_upload_part_size_multiply_factor; + settings->upload_part_size_multiply_parts_count_threshold = query_settings.azure_upload_part_size_multiply_parts_count_threshold; + settings->sdk_max_retries = query_settings.azure_sdk_max_retries; + settings->sdk_retry_initial_backoff_ms = query_settings.azure_sdk_retry_initial_backoff_ms; + settings->sdk_retry_max_backoff_ms = query_settings.azure_sdk_retry_max_backoff_ms; + + return settings; +} + +std::unique_ptr getRequestSettingsForBackup(const Settings & query_settings, bool use_native_copy) +{ + auto settings = getRequestSettings(query_settings); + settings->use_native_copy = use_native_copy; + return settings; +} + +std::unique_ptr getRequestSettings(const Poco::Util::AbstractConfiguration & config, const String & config_prefix, ContextPtr context) +{ + auto settings = std::make_unique(); + const auto & settings_ref = context->getSettingsRef(); + + settings->min_bytes_for_seek = config.getUInt64(config_prefix + ".min_bytes_for_seek", 1024 * 1024); + settings->use_native_copy = config.getBool(config_prefix + ".use_native_copy", false); + + settings->max_single_part_upload_size = config.getUInt64(config_prefix + ".max_single_part_upload_size", settings_ref.azure_max_single_part_upload_size); + settings->max_single_read_retries = config.getUInt64(config_prefix + ".max_single_read_retries", settings_ref.azure_max_single_read_retries); + settings->max_single_download_retries = config.getUInt64(config_prefix + ".max_single_download_retries", settings_ref.azure_max_single_read_retries); + settings->list_object_keys_size = config.getUInt64(config_prefix + ".list_object_keys_size", settings_ref.azure_list_object_keys_size); + settings->min_upload_part_size = config.getUInt64(config_prefix + ".min_upload_part_size", settings_ref.azure_min_upload_part_size); + settings->max_upload_part_size = config.getUInt64(config_prefix + ".max_upload_part_size", settings_ref.azure_max_upload_part_size); + settings->max_single_part_copy_size = config.getUInt64(config_prefix + ".max_single_part_copy_size", settings_ref.azure_max_single_part_copy_size); + settings->max_blocks_in_multipart_upload = config.getUInt64(config_prefix + ".max_blocks_in_multipart_upload", settings_ref.azure_max_blocks_in_multipart_upload); + settings->max_unexpected_write_error_retries = config.getUInt64(config_prefix + ".max_unexpected_write_error_retries", settings_ref.azure_max_unexpected_write_error_retries); + settings->max_inflight_parts_for_one_file = config.getUInt64(config_prefix + ".max_inflight_parts_for_one_file", settings_ref.azure_max_inflight_parts_for_one_file); + settings->strict_upload_part_size = config.getUInt64(config_prefix + ".strict_upload_part_size", settings_ref.azure_strict_upload_part_size); + settings->upload_part_size_multiply_factor = config.getUInt64(config_prefix + ".upload_part_size_multiply_factor", settings_ref.azure_upload_part_size_multiply_factor); + settings->upload_part_size_multiply_parts_count_threshold = config.getUInt64(config_prefix + ".upload_part_size_multiply_parts_count_threshold", settings_ref.azure_upload_part_size_multiply_parts_count_threshold); + + settings->sdk_max_retries = config.getUInt64(config_prefix + ".max_tries", settings_ref.azure_sdk_max_retries); + settings->sdk_retry_initial_backoff_ms = config.getUInt64(config_prefix + ".retry_initial_backoff_ms", settings_ref.azure_sdk_retry_initial_backoff_ms); + settings->sdk_retry_max_backoff_ms = config.getUInt64(config_prefix + ".retry_max_backoff_ms", settings_ref.azure_sdk_retry_max_backoff_ms); + + if (config.has(config_prefix + ".curl_ip_resolve")) + { + using CurlOptions = Azure::Core::Http::CurlTransportOptions; + + auto value = config.getString(config_prefix + ".curl_ip_resolve"); + if (value == "ipv4") + settings->curl_ip_resolve = CurlOptions::CURL_IPRESOLVE_V4; + else if (value == "ipv6") + settings->curl_ip_resolve = CurlOptions::CURL_IPRESOLVE_V6; + else + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected value for option 'curl_ip_resolve': {}. Expected one of 'ipv4' or 'ipv6'", value); + } + + return settings; +} + +} + +} + +#endif diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.h b/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.h new file mode 100644 index 00000000000..bb2f0270924 --- /dev/null +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.h @@ -0,0 +1,139 @@ +#pragma once +#include "config.h" + +#if USE_AZURE_BLOB_STORAGE + +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +namespace DB +{ + +struct Settings; + +namespace AzureBlobStorage +{ + +using ServiceClient = Azure::Storage::Blobs::BlobServiceClient; +using ContainerClient = Azure::Storage::Blobs::BlobContainerClient; +using BlobClient = Azure::Storage::Blobs::BlobClient; +using BlobClientOptions = Azure::Storage::Blobs::BlobClientOptions; + +struct RequestSettings +{ + RequestSettings() = default; + + size_t max_single_part_upload_size = 100 * 1024 * 1024; /// NOTE: on 32-bit machines it will be at most 4GB, but size_t is also used in BufferBase for offset + size_t min_bytes_for_seek = 1024 * 1024; + size_t max_single_read_retries = 3; + size_t max_single_download_retries = 3; + size_t list_object_keys_size = 1000; + size_t min_upload_part_size = 16 * 1024 * 1024; + size_t max_upload_part_size = 5ULL * 1024 * 1024 * 1024; + size_t max_single_part_copy_size = 256 * 1024 * 1024; + size_t max_unexpected_write_error_retries = 4; + size_t max_inflight_parts_for_one_file = 20; + size_t max_blocks_in_multipart_upload = 50000; + size_t strict_upload_part_size = 0; + size_t upload_part_size_multiply_factor = 2; + size_t upload_part_size_multiply_parts_count_threshold = 500; + size_t sdk_max_retries = 10; + size_t sdk_retry_initial_backoff_ms = 10; + size_t sdk_retry_max_backoff_ms = 1000; + bool use_native_copy = false; + + using CurlOptions = Azure::Core::Http::CurlTransportOptions; + CurlOptions::CurlOptIPResolve curl_ip_resolve = CurlOptions::CURL_IPRESOLVE_WHATEVER; +}; + +struct Endpoint +{ + String storage_account_url; + String account_name; + String container_name; + String prefix; + String sas_auth; + std::optional container_already_exists; + + String getEndpoint() const + { + String url = storage_account_url; + if (url.ends_with('/')) + url.pop_back(); + + if (!account_name.empty()) + url += "/" + account_name; + + if (!container_name.empty()) + url += "/" + container_name; + + if (!prefix.empty()) + url += "/" + prefix; + + if (!sas_auth.empty()) + url += "?" + sas_auth; + + return url; + } + + String getEndpointWithoutContainer() const + { + String url = storage_account_url; + + if (!account_name.empty()) + url += "/" + account_name; + + if (!sas_auth.empty()) + url += "?" + sas_auth; + + return url; + } +}; + +using ConnectionString = StrongTypedef; + +using AuthMethod = std::variant< + ConnectionString, + std::shared_ptr, + std::shared_ptr, + std::shared_ptr>; + +struct ConnectionParams +{ + Endpoint endpoint; + AuthMethod auth_method; + BlobClientOptions client_options; + + String getContainer() const { return endpoint.container_name; } + String getConnectionURL() const; + + std::unique_ptr createForService() const; + std::unique_ptr createForContainer() const; +}; + +Endpoint processEndpoint(const Poco::Util::AbstractConfiguration & config, const String & config_prefix); +void processURL(const String & url, const String & container_name, Endpoint & endpoint, AuthMethod & auth_method); + +std::unique_ptr getContainerClient(const ConnectionParams & params, bool readonly); + +BlobClientOptions getClientOptions(const RequestSettings & settings, bool for_disk); +AuthMethod getAuthMethod(const Poco::Util::AbstractConfiguration & config, const String & config_prefix); + +std::unique_ptr getRequestSettings(const Settings & query_settings); +std::unique_ptr getRequestSettingsForBackup(const Settings & query_settings, bool use_native_copy); +std::unique_ptr getRequestSettings(const Poco::Util::AbstractConfiguration & config, const String & config_prefix, ContextPtr context); + +} + +} + +#endif diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp index 86a035f3be7..bc16955143b 100644 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp @@ -9,7 +9,7 @@ #include #include -#include +#include #include #include #include @@ -60,7 +60,6 @@ public: "ListObjectAzure") , client(client_) { - options.Prefix = path_prefix; options.PageSizeHint = static_cast(max_list_size); } @@ -105,7 +104,7 @@ private: AzureObjectStorage::AzureObjectStorage( const String & name_, - AzureClientPtr && client_, + ClientPtr && client_, SettingsPtr && settings_, const String & object_namespace_, const String & description_) @@ -150,7 +149,7 @@ ObjectStorageIteratorPtr AzureObjectStorage::iterate(const std::string & path_pr auto settings_ptr = settings.get(); auto client_ptr = client.get(); - return std::make_shared(path_prefix, client_ptr, max_keys); + return std::make_shared(path_prefix, client_ptr, max_keys ? max_keys : settings_ptr->list_object_keys_size); } void AzureObjectStorage::listObjects(const std::string & path, RelativePathsWithMetadata & children, size_t max_keys) const @@ -398,24 +397,49 @@ void AzureObjectStorage::copyObject( /// NOLINT } void AzureObjectStorage::applyNewSettings( - const Poco::Util::AbstractConfiguration & config, const std::string & config_prefix, - ContextPtr context, const ApplyNewSettingsOptions &) + const Poco::Util::AbstractConfiguration & config, + const std::string & config_prefix, + ContextPtr context, + const ApplyNewSettingsOptions & options) { - auto new_settings = getAzureBlobStorageSettings(config, config_prefix, context); + auto new_settings = AzureBlobStorage::getRequestSettings(config, config_prefix, context); settings.set(std::move(new_settings)); - /// We don't update client + + if (!options.allow_client_change) + return; + + bool is_client_for_disk = client.get()->GetClickhouseOptions().IsClientForDisk; + + AzureBlobStorage::ConnectionParams params + { + .endpoint = AzureBlobStorage::processEndpoint(config, config_prefix), + .auth_method = AzureBlobStorage::getAuthMethod(config, config_prefix), + .client_options = AzureBlobStorage::getClientOptions(*settings.get(), is_client_for_disk), + }; + + auto new_client = AzureBlobStorage::getContainerClient(params, /*readonly=*/ true); + client.set(std::move(new_client)); } -std::unique_ptr AzureObjectStorage::cloneObjectStorage(const std::string &, const Poco::Util::AbstractConfiguration & config, const std::string & config_prefix, ContextPtr context) +std::unique_ptr AzureObjectStorage::cloneObjectStorage( + const std::string & new_namespace, + const Poco::Util::AbstractConfiguration & config, + const std::string & config_prefix, + ContextPtr context) { - return std::make_unique( - name, - getAzureBlobContainerClient(config, config_prefix), - getAzureBlobStorageSettings(config, config_prefix, context), - object_namespace, - description - ); + auto new_settings = AzureBlobStorage::getRequestSettings(config, config_prefix, context); + bool is_client_for_disk = client.get()->GetClickhouseOptions().IsClientForDisk; + + AzureBlobStorage::ConnectionParams params + { + .endpoint = AzureBlobStorage::processEndpoint(config, config_prefix), + .auth_method = AzureBlobStorage::getAuthMethod(config, config_prefix), + .client_options = AzureBlobStorage::getClientOptions(*new_settings, is_client_for_disk), + }; + + auto new_client = AzureBlobStorage::getContainerClient(params, /*readonly=*/ true); + return std::make_unique(name, std::move(new_client), std::move(new_settings), new_namespace, params.endpoint.getEndpointWithoutContainer()); } } diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h index c342929d656..2c7ce5e18dc 100644 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h @@ -7,6 +7,8 @@ #include #include #include +#include +#include namespace Poco { @@ -16,71 +18,15 @@ class Logger; namespace DB { -struct AzureObjectStorageSettings -{ - AzureObjectStorageSettings( - uint64_t max_single_part_upload_size_, - uint64_t min_bytes_for_seek_, - int max_single_read_retries_, - int max_single_download_retries_, - int list_object_keys_size_, - size_t min_upload_part_size_, - size_t max_upload_part_size_, - size_t max_single_part_copy_size_, - bool use_native_copy_, - size_t max_unexpected_write_error_retries_, - size_t max_inflight_parts_for_one_file_, - size_t strict_upload_part_size_, - size_t upload_part_size_multiply_factor_, - size_t upload_part_size_multiply_parts_count_threshold_) - : max_single_part_upload_size(max_single_part_upload_size_) - , min_bytes_for_seek(min_bytes_for_seek_) - , max_single_read_retries(max_single_read_retries_) - , max_single_download_retries(max_single_download_retries_) - , list_object_keys_size(list_object_keys_size_) - , min_upload_part_size(min_upload_part_size_) - , max_upload_part_size(max_upload_part_size_) - , max_single_part_copy_size(max_single_part_copy_size_) - , use_native_copy(use_native_copy_) - , max_unexpected_write_error_retries(max_unexpected_write_error_retries_) - , max_inflight_parts_for_one_file(max_inflight_parts_for_one_file_) - , strict_upload_part_size(strict_upload_part_size_) - , upload_part_size_multiply_factor(upload_part_size_multiply_factor_) - , upload_part_size_multiply_parts_count_threshold(upload_part_size_multiply_parts_count_threshold_) - { - } - - AzureObjectStorageSettings() = default; - - size_t max_single_part_upload_size = 100 * 1024 * 1024; /// NOTE: on 32-bit machines it will be at most 4GB, but size_t is also used in BufferBase for offset - uint64_t min_bytes_for_seek = 1024 * 1024; - size_t max_single_read_retries = 3; - size_t max_single_download_retries = 3; - int list_object_keys_size = 1000; - size_t min_upload_part_size = 16 * 1024 * 1024; - size_t max_upload_part_size = 5ULL * 1024 * 1024 * 1024; - size_t max_single_part_copy_size = 256 * 1024 * 1024; - bool use_native_copy = false; - size_t max_unexpected_write_error_retries = 4; - size_t max_inflight_parts_for_one_file = 20; - size_t max_blocks_in_multipart_upload = 50000; - size_t strict_upload_part_size = 0; - size_t upload_part_size_multiply_factor = 2; - size_t upload_part_size_multiply_parts_count_threshold = 500; -}; - -using AzureClient = Azure::Storage::Blobs::BlobContainerClient; -using AzureClientPtr = std::unique_ptr; - class AzureObjectStorage : public IObjectStorage { public: - - using SettingsPtr = std::unique_ptr; + using ClientPtr = std::unique_ptr; + using SettingsPtr = std::unique_ptr; AzureObjectStorage( const String & name_, - AzureClientPtr && client_, + ClientPtr && client_, SettingsPtr && settings_, const String & object_namespace_, const String & description_); @@ -159,12 +105,8 @@ public: bool isRemote() const override { return true; } - std::shared_ptr getSettings() { return settings.get(); } - - std::shared_ptr getAzureBlobStorageClient() override - { - return client.get(); - } + std::shared_ptr getSettings() const { return settings.get(); } + std::shared_ptr getAzureBlobStorageClient() const override { return client.get(); } bool supportParallelWrite() const override { return true; } @@ -174,8 +116,8 @@ private: const String name; /// client used to access the files in the Blob Storage cloud - MultiVersion client; - MultiVersion settings; + MultiVersion client; + MultiVersion settings; const String object_namespace; /// container + prefix /// We use source url without container and prefix as description, because in Azure there are no limitations for operations between different containers. diff --git a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h index 727dbeed853..93ef2659cbb 100644 --- a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h +++ b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h @@ -127,7 +127,7 @@ public: const FileCacheSettings & getCacheSettings() const { return cache_settings; } #if USE_AZURE_BLOB_STORAGE - std::shared_ptr getAzureBlobStorageClient() override + std::shared_ptr getAzureBlobStorageClient() const override { return object_storage->getAzureBlobStorageClient(); } diff --git a/src/Disks/ObjectStorages/DiskObjectStorage.h b/src/Disks/ObjectStorages/DiskObjectStorage.h index 59cc82d8c81..5c45a258806 100644 --- a/src/Disks/ObjectStorages/DiskObjectStorage.h +++ b/src/Disks/ObjectStorages/DiskObjectStorage.h @@ -195,7 +195,6 @@ public: /// DiskObjectStorage(CachedObjectStorage(CachedObjectStorage(S3ObjectStorage))) String getStructure() const { return fmt::format("DiskObjectStorage-{}({})", getName(), object_storage->getName()); } -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD /// Add a cache layer. /// Example: DiskObjectStorage(S3ObjectStorage) -> DiskObjectStorage(CachedObjectStorage(S3ObjectStorage)) /// There can be any number of cache layers: @@ -204,7 +203,6 @@ public: /// Get names of all cache layers. Name is how cache is defined in configuration file. NameSet getCacheLayersNames() const override; -#endif bool supportsStat() const override { return metadata_storage->supportsStat(); } struct stat stat(const String & path) const override; diff --git a/src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp b/src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp index 44854633d65..56d5d11ef8a 100644 --- a/src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp +++ b/src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp @@ -222,11 +222,7 @@ ObjectKeyWithMetadata DiskObjectStorageMetadata::popLastObject() bool DiskObjectStorageMetadata::getWriteFullObjectKeySetting() { -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD return Context::getGlobalContextInstance()->getServerSettings().storage_metadata_write_full_object_key; -#else - return false; -#endif } } diff --git a/src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp b/src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp index e7c85bea1c6..b5805f6d23a 100644 --- a/src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp +++ b/src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp @@ -874,7 +874,9 @@ void DiskObjectStorageTransaction::writeFileUsingBlobWritingFunction( /// Create metadata (see create_metadata_callback in DiskObjectStorageTransaction::writeFile()). if (mode == WriteMode::Rewrite) { - if (!object_storage.isWriteOnce() && metadata_storage.exists(path)) + /// Otherwise we will produce lost blobs which nobody points to + /// WriteOnce storages are not affected by the issue + if (!object_storage.isPlain() && metadata_storage.exists(path)) object_storage.removeObjectsIfExist(metadata_storage.getStorageObjects(path)); metadata_transaction->createMetadataFile(path, std::move(object_key), object_size); diff --git a/src/Disks/ObjectStorages/IObjectStorage.h b/src/Disks/ObjectStorages/IObjectStorage.h index 9f5c14fdb7c..ceea4d5a2bb 100644 --- a/src/Disks/ObjectStorages/IObjectStorage.h +++ b/src/Disks/ObjectStorages/IObjectStorage.h @@ -75,6 +75,7 @@ struct RelativePathWithMetadata virtual std::string getPath() const { return relative_path; } virtual bool isArchive() const { return false; } virtual std::string getPathToArchive() const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Not an archive"); } + virtual size_t fileSizeInArchive() const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Not an archive"); } }; struct ObjectKeyWithMetadata @@ -260,7 +261,7 @@ public: virtual void setKeysGenerator(ObjectStorageKeysGeneratorPtr) { } #if USE_AZURE_BLOB_STORAGE - virtual std::shared_ptr getAzureBlobStorageClient() + virtual std::shared_ptr getAzureBlobStorageClient() const { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "This function is only implemented for AzureBlobStorage"); } diff --git a/src/Disks/ObjectStorages/MetadataOperationsHolder.h b/src/Disks/ObjectStorages/MetadataOperationsHolder.h index 8997f40b9a2..a042f4bd8b9 100644 --- a/src/Disks/ObjectStorages/MetadataOperationsHolder.h +++ b/src/Disks/ObjectStorages/MetadataOperationsHolder.h @@ -1,6 +1,7 @@ #pragma once #include +#include #include #include #include diff --git a/src/Disks/ObjectStorages/MetadataStorageFactory.cpp b/src/Disks/ObjectStorages/MetadataStorageFactory.cpp index ab7c2069b43..a690ecd2757 100644 --- a/src/Disks/ObjectStorages/MetadataStorageFactory.cpp +++ b/src/Disks/ObjectStorages/MetadataStorageFactory.cpp @@ -2,9 +2,7 @@ #include #include #include -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD #include -#endif #include #include @@ -135,7 +133,6 @@ void registerPlainRewritableMetadataStorage(MetadataStorageFactory & factory) }); } -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD void registerMetadataStorageFromStaticFilesWebServer(MetadataStorageFactory & factory) { factory.registerMetadataStorageType("web", []( @@ -147,7 +144,6 @@ void registerMetadataStorageFromStaticFilesWebServer(MetadataStorageFactory & fa return std::make_shared(assert_cast(*object_storage)); }); } -#endif void registerMetadataStorages() { @@ -155,9 +151,7 @@ void registerMetadataStorages() registerMetadataStorageFromDisk(factory); registerPlainMetadataStorage(factory); registerPlainRewritableMetadataStorage(factory); -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD registerMetadataStorageFromStaticFilesWebServer(factory); -#endif } } diff --git a/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.h b/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.h index f290a762205..66da0f2431e 100644 --- a/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.h +++ b/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.h @@ -22,8 +22,7 @@ using UnlinkMetadataFileOperationOutcomePtr = std::shared_ptr +#include #include +#include +#include #include #include #include "CommonPathPrefixKeyGenerator.h" + namespace DB { @@ -22,34 +26,78 @@ MetadataStorageFromPlainObjectStorage::PathMap loadPathPrefixMap(const std::stri { MetadataStorageFromPlainObjectStorage::PathMap result; - RelativePathsWithMetadata files; - object_storage->listObjects(root, files, 0); - for (const auto & file : files) + ThreadPool & pool = getIOThreadPool().get(); + ThreadPoolCallbackRunnerLocal runner(pool, "PlainRWMetaLoad"); + std::mutex mutex; + + LoggerPtr log = getLogger("MetadataStorageFromPlainObjectStorage"); + + ReadSettings settings; + settings.enable_filesystem_cache = false; + settings.remote_fs_method = RemoteFSReadMethod::read; + settings.remote_fs_buffer_size = 1024; /// These files are small. + + LOG_DEBUG(log, "Loading metadata"); + size_t num_files = 0; + for (auto iterator = object_storage->iterate(root, 0); iterator->isValid(); iterator->next()) { - auto remote_path = std::filesystem::path(file->relative_path); + ++num_files; + auto file = iterator->current(); + String path = file->getPath(); + auto remote_path = std::filesystem::path(path); if (remote_path.filename() != PREFIX_PATH_FILE_NAME) continue; - StoredObject object{file->relative_path}; + runner([remote_path, path, &object_storage, &result, &mutex, &log, &settings] + { + setThreadName("PlainRWMetaLoad"); - auto read_buf = object_storage->readObject(object); - String local_path; - readStringUntilEOF(local_path, *read_buf); + StoredObject object{path}; + String local_path; - chassert(remote_path.has_parent_path()); - auto res = result.emplace(local_path, remote_path.parent_path()); + try + { + auto read_buf = object_storage->readObject(object, settings); + readStringUntilEOF(local_path, *read_buf); + } +#if USE_AWS_S3 + catch (const S3Exception & e) + { + /// It is ok if a directory was removed just now. + /// We support attaching a filesystem that is concurrently modified by someone else. + if (e.getS3ErrorCode() == Aws::S3::S3Errors::NO_SUCH_KEY) + return; + throw; + } +#endif + catch (...) + { + throw; + } - /// This can happen if table replication is enabled, then the same local path is written - /// in `prefix.path` of each replica. - /// TODO: should replicated tables (e.g., RMT) be explicitly disallowed? - if (!res.second) - LOG_WARNING( - getLogger("MetadataStorageFromPlainObjectStorage"), - "The local path '{}' is already mapped to a remote path '{}', ignoring: '{}'", - local_path, - res.first->second, - remote_path.parent_path().string()); + chassert(remote_path.has_parent_path()); + std::pair res; + { + std::lock_guard lock(mutex); + res = result.emplace(local_path, remote_path.parent_path()); + } + + /// This can happen if table replication is enabled, then the same local path is written + /// in `prefix.path` of each replica. + /// TODO: should replicated tables (e.g., RMT) be explicitly disallowed? + if (!res.second) + LOG_WARNING( + log, + "The local path '{}' is already mapped to a remote path '{}', ignoring: '{}'", + local_path, + res.first->second, + remote_path.parent_path().string()); + }); } + + runner.waitForAllToFinishAndRethrowFirstError(); + LOG_DEBUG(log, "Loaded metadata for {} files, found {} directories", num_files, result.size()); + auto metric = object_storage->getMetadataStorageMetrics().directory_map_size; CurrentMetrics::add(metric, result.size()); return result; diff --git a/src/Disks/ObjectStorages/MetadataStorageFromPlainRewritableObjectStorage.h b/src/Disks/ObjectStorages/MetadataStorageFromPlainRewritableObjectStorage.h index 661968d7044..a5394b9428d 100644 --- a/src/Disks/ObjectStorages/MetadataStorageFromPlainRewritableObjectStorage.h +++ b/src/Disks/ObjectStorages/MetadataStorageFromPlainRewritableObjectStorage.h @@ -4,6 +4,7 @@ #include + namespace DB { diff --git a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp index 1bf8250adff..d42da6dc9c4 100644 --- a/src/Disks/ObjectStorages/ObjectStorageFactory.cpp +++ b/src/Disks/ObjectStorages/ObjectStorageFactory.cpp @@ -7,25 +7,24 @@ #include #include #endif -#if USE_HDFS && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_HDFS #include #include #endif -#if USE_AZURE_BLOB_STORAGE && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_AZURE_BLOB_STORAGE #include -#include +#include #endif -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD #include #include #include -#endif #include #include #include #include #include #include +#include #include @@ -284,7 +283,7 @@ void registerS3PlainRewritableObjectStorage(ObjectStorageFactory & factory) #endif -#if USE_HDFS && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_HDFS void registerHDFSObjectStorage(ObjectStorageFactory & factory) { factory.registerObjectStorageType( @@ -309,7 +308,7 @@ void registerHDFSObjectStorage(ObjectStorageFactory & factory) } #endif -#if USE_AZURE_BLOB_STORAGE && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_AZURE_BLOB_STORAGE void registerAzureObjectStorage(ObjectStorageFactory & factory) { auto creator = []( @@ -319,21 +318,27 @@ void registerAzureObjectStorage(ObjectStorageFactory & factory) const ContextPtr & context, bool /* skip_access_check */) -> ObjectStoragePtr { - AzureBlobStorageEndpoint endpoint = processAzureBlobStorageEndpoint(config, config_prefix); + auto azure_settings = AzureBlobStorage::getRequestSettings(config, config_prefix, context); + + AzureBlobStorage::ConnectionParams params + { + .endpoint = AzureBlobStorage::processEndpoint(config, config_prefix), + .auth_method = AzureBlobStorage::getAuthMethod(config, config_prefix), + .client_options = AzureBlobStorage::getClientOptions(*azure_settings, /*for_disk=*/ true), + }; return createObjectStorage( ObjectStorageType::Azure, config, config_prefix, name, - getAzureBlobContainerClient(config, config_prefix), - getAzureBlobStorageSettings(config, config_prefix, context), - endpoint.prefix.empty() ? endpoint.container_name : endpoint.container_name + "/" + endpoint.prefix, - endpoint.getEndpointWithoutContainer()); + AzureBlobStorage::getContainerClient(params, /*readonly=*/ false), std::move(azure_settings), + params.endpoint.prefix.empty() ? params.endpoint.container_name : params.endpoint.container_name + "/" + params.endpoint.prefix, + params.endpoint.getEndpointWithoutContainer()); }; + factory.registerObjectStorageType("azure_blob_storage", creator); factory.registerObjectStorageType("azure", creator); } #endif -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD void registerWebObjectStorage(ObjectStorageFactory & factory) { factory.registerObjectStorageType("web", []( @@ -381,7 +386,6 @@ void registerLocalObjectStorage(ObjectStorageFactory & factory) factory.registerObjectStorageType("local_blob_storage", creator); factory.registerObjectStorageType("local", creator); } -#endif void registerObjectStorages() { @@ -393,18 +397,16 @@ void registerObjectStorages() registerS3PlainRewritableObjectStorage(factory); #endif -#if USE_HDFS && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_HDFS registerHDFSObjectStorage(factory); #endif -#if USE_AZURE_BLOB_STORAGE && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_AZURE_BLOB_STORAGE registerAzureObjectStorage(factory); #endif -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD registerWebObjectStorage(factory); registerLocalObjectStorage(factory); -#endif } } diff --git a/src/Disks/ObjectStorages/ObjectStorageIterator.h b/src/Disks/ObjectStorages/ObjectStorageIterator.h index 26c3c690ba5..d814514ddcc 100644 --- a/src/Disks/ObjectStorages/ObjectStorageIterator.h +++ b/src/Disks/ObjectStorages/ObjectStorageIterator.h @@ -9,15 +9,34 @@ namespace DB class IObjectStorageIterator { public: + /// Moves iterator to the next element. If the iterator not isValid, the behavior is undefined. virtual void next() = 0; - virtual void nextBatch() = 0; + + /// Check if the iterator is valid, which means the `current` method can be called. virtual bool isValid() = 0; + + /// Return the current element. virtual RelativePathWithMetadataPtr current() = 0; - virtual RelativePathsWithMetadata currentBatch() = 0; + + /// This will initiate prefetching the next batch in background, so it can be obtained faster when needed. virtual std::optional getCurrentBatchAndScheduleNext() = 0; + + /// Returns the number of elements in the batches that were fetched so far. virtual size_t getAccumulatedSize() const = 0; virtual ~IObjectStorageIterator() = default; + +private: + /// Skips all the remaining elements in the current batch (if any), + /// and moves the iterator to the first element of the next batch, + /// or, if there is no more batches, the iterator becomes invalid. + /// If the iterator not isValid, the behavior is undefined. + virtual void nextBatch() = 0; + + /// Return the current batch of elements. + /// It is unspecified how batches are formed. + /// But this method can be used for more efficient processing. + virtual RelativePathsWithMetadata currentBatch() = 0; }; using ObjectStorageIteratorPtr = std::shared_ptr; @@ -25,6 +44,7 @@ using ObjectStorageIteratorPtr = std::shared_ptr; class ObjectStorageIteratorFromList : public IObjectStorageIterator { public: + /// Everything is represented by just a single batch. explicit ObjectStorageIteratorFromList(RelativePathsWithMetadata && batch_) : batch(std::move(batch_)) , batch_iterator(batch.begin()) {} diff --git a/src/Disks/ObjectStorages/ObjectStorageIteratorAsync.cpp b/src/Disks/ObjectStorages/ObjectStorageIteratorAsync.cpp index a249789df4b..2d2e8cd2c1a 100644 --- a/src/Disks/ObjectStorages/ObjectStorageIteratorAsync.cpp +++ b/src/Disks/ObjectStorages/ObjectStorageIteratorAsync.cpp @@ -37,10 +37,11 @@ void IObjectStorageIteratorAsync::nextBatch() { std::lock_guard lock(mutex); - if (is_finished) + if (!has_next_batch) { current_batch.clear(); current_batch_iterator = current_batch.begin(); + is_finished = true; return; } @@ -58,16 +59,23 @@ void IObjectStorageIteratorAsync::nextBatch() current_batch = std::move(result.batch); current_batch_iterator = current_batch.begin(); - accumulated_size.fetch_add(current_batch.size(), std::memory_order_relaxed); - - if (result.has_next) - outcome_future = scheduleBatch(); - else + if (current_batch.empty()) + { is_finished = true; + has_next_batch = false; + } + else + { + accumulated_size.fetch_add(current_batch.size(), std::memory_order_relaxed); + + has_next_batch = result.has_next; + if (has_next_batch) + outcome_future = scheduleBatch(); + } } catch (...) { - is_finished = true; + has_next_batch = false; throw; } } @@ -76,10 +84,12 @@ void IObjectStorageIteratorAsync::next() { std::lock_guard lock(mutex); + if (is_finished) + return; + + ++current_batch_iterator; if (current_batch_iterator == current_batch.end()) nextBatch(); - else - ++current_batch_iterator; } std::future IObjectStorageIteratorAsync::scheduleBatch() @@ -99,7 +109,7 @@ bool IObjectStorageIteratorAsync::isValid() if (!is_initialized) nextBatch(); - return current_batch_iterator != current_batch.end(); + return !is_finished; } RelativePathWithMetadataPtr IObjectStorageIteratorAsync::current() diff --git a/src/Disks/ObjectStorages/ObjectStorageIteratorAsync.h b/src/Disks/ObjectStorages/ObjectStorageIteratorAsync.h index cb4818d01ae..01371415124 100644 --- a/src/Disks/ObjectStorages/ObjectStorageIteratorAsync.h +++ b/src/Disks/ObjectStorages/ObjectStorageIteratorAsync.h @@ -35,7 +35,7 @@ public: void deactivate(); protected: - + /// This method fetches the next batch, and returns true if there are more batches after it. virtual bool getBatchAndCheckNext(RelativePathsWithMetadata & batch) = 0; struct BatchAndHasNext @@ -48,6 +48,7 @@ protected: bool is_initialized{false}; bool is_finished{false}; + bool has_next_batch{true}; bool deactivated{false}; mutable std::recursive_mutex mutex; diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp index 0f7024196ea..a6672e14e10 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -293,6 +294,8 @@ std::unique_ptr S3ObjectStorage::writeObject( /// NOLIN ObjectStorageIteratorPtr S3ObjectStorage::iterate(const std::string & path_prefix, size_t max_keys) const { auto settings_ptr = s3_settings.get(); + if (!max_keys) + max_keys = settings_ptr->list_object_keys_size; return std::make_shared(uri.bucket, path_prefix, client.get(), max_keys); } diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.h b/src/Disks/ObjectStorages/S3/S3ObjectStorage.h index 4170cea22a0..cbe004bc298 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.h +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.h @@ -67,7 +67,7 @@ private: } public: - template + template explicit S3ObjectStorage(std::unique_ptr && client_, Args && ...args) : S3ObjectStorage("S3ObjectStorage", std::move(client_), std::forward(args)...) { diff --git a/src/Disks/ObjectStorages/S3/diskSettings.cpp b/src/Disks/ObjectStorages/S3/diskSettings.cpp index 62df98f51e6..de78ce4594d 100644 --- a/src/Disks/ObjectStorages/S3/diskSettings.cpp +++ b/src/Disks/ObjectStorages/S3/diskSettings.cpp @@ -1,14 +1,14 @@ #include -#include -#include #if USE_AWS_S3 +#include #include #include #include #include #include +#include #include #include #include @@ -17,6 +17,7 @@ #include #include #include +#include #include #include diff --git a/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp b/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp index e837e056acc..7f7a3fe1a62 100644 --- a/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp +++ b/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp @@ -3,6 +3,8 @@ #include #include +#include + #include #include #include diff --git a/src/Disks/ObjectStorages/createMetadataStorageMetrics.h b/src/Disks/ObjectStorages/createMetadataStorageMetrics.h index 6dddc227ade..5cf1fbef2ab 100644 --- a/src/Disks/ObjectStorages/createMetadataStorageMetrics.h +++ b/src/Disks/ObjectStorages/createMetadataStorageMetrics.h @@ -1,14 +1,14 @@ #pragma once +#include "config.h" + #if USE_AWS_S3 # include #endif -#if USE_AZURE_BLOB_STORAGE && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_AZURE_BLOB_STORAGE # include #endif -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD -# include -#endif +#include #include namespace ProfileEvents @@ -42,7 +42,7 @@ inline MetadataStorageMetrics MetadataStorageMetrics::create inline MetadataStorageMetrics MetadataStorageMetrics::create() { @@ -53,7 +53,6 @@ inline MetadataStorageMetrics MetadataStorageMetrics::create inline MetadataStorageMetrics MetadataStorageMetrics::create() { @@ -62,6 +61,5 @@ inline MetadataStorageMetrics MetadataStorageMetrics::createexists(relative_path)) { - LOG_WARNING(getLogger("TemporaryFileOnDisk"), "Temporary path '{}' does not exist in '{}'", relative_path, disk->getPath()); + if (show_warning_if_removed) + LOG_WARNING(getLogger("TemporaryFileOnDisk"), "Temporary path '{}' does not exist in '{}'", relative_path, disk->getPath()); return; } diff --git a/src/Disks/TemporaryFileOnDisk.h b/src/Disks/TemporaryFileOnDisk.h index cccfc82cf9e..d0ff44c6f03 100644 --- a/src/Disks/TemporaryFileOnDisk.h +++ b/src/Disks/TemporaryFileOnDisk.h @@ -27,12 +27,19 @@ public: /// Return relative path (without disk) const String & getRelativePath() const { return relative_path; } + /// Sets whether the destructor should show a warning if the temporary file has been already removed. + /// By default a warning is shown. + void setShowWarningIfRemoved(bool show_warning_if_removed_) { show_warning_if_removed = show_warning_if_removed_; } + private: DiskPtr disk; /// Relative path in disk to the temporary file or directory String relative_path; + /// Whether the destructor should show a warning if the temporary file has been already removed. + bool show_warning_if_removed = true; + CurrentMetrics::Increment metric_increment; /// Specified if we know what for file is used (sort/aggregate/join). diff --git a/src/Disks/registerDisks.cpp b/src/Disks/registerDisks.cpp index b8da93ff9f2..8312fac4719 100644 --- a/src/Disks/registerDisks.cpp +++ b/src/Disks/registerDisks.cpp @@ -17,8 +17,6 @@ void registerDiskCache(DiskFactory & factory, bool global_skip_access_check); void registerDiskObjectStorage(DiskFactory & factory, bool global_skip_access_check); -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD - void registerDisks(bool global_skip_access_check) { auto & factory = DiskFactory::instance(); @@ -34,17 +32,4 @@ void registerDisks(bool global_skip_access_check) registerDiskObjectStorage(factory, global_skip_access_check); } -#else - -void registerDisks(bool global_skip_access_check) -{ - auto & factory = DiskFactory::instance(); - - registerDiskLocal(factory, global_skip_access_check); - - registerDiskObjectStorage(factory, global_skip_access_check); -} - -#endif - } diff --git a/src/Formats/EscapingRuleUtils.cpp b/src/Formats/EscapingRuleUtils.cpp index 36d16d8d154..58407a810c5 100644 --- a/src/Formats/EscapingRuleUtils.cpp +++ b/src/Formats/EscapingRuleUtils.cpp @@ -439,13 +439,15 @@ String getAdditionalFormatInfoByEscapingRule(const FormatSettings & settings, Fo case FormatSettings::EscapingRule::CSV: result += fmt::format( ", use_best_effort_in_schema_inference={}, bool_true_representation={}, bool_false_representation={}," - " null_representation={}, delimiter={}, tuple_delimiter={}", + " null_representation={}, delimiter={}, tuple_delimiter={}, try_infer_numbers_from_strings={}, try_infer_strings_from_quoted_tuples={}", settings.csv.use_best_effort_in_schema_inference, settings.bool_true_representation, settings.bool_false_representation, settings.csv.null_representation, settings.csv.delimiter, - settings.csv.tuple_delimiter); + settings.csv.tuple_delimiter, + settings.csv.try_infer_numbers_from_strings, + settings.csv.try_infer_strings_from_quoted_tuples); break; case FormatSettings::EscapingRule::JSON: result += fmt::format( diff --git a/src/Formats/FormatFactory.cpp b/src/Formats/FormatFactory.cpp index 51b18b2093d..501343d2ca9 100644 --- a/src/Formats/FormatFactory.cpp +++ b/src/Formats/FormatFactory.cpp @@ -17,6 +17,7 @@ #include #include #include +#include #include @@ -149,6 +150,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.json.try_infer_objects_as_tuples = settings.input_format_json_try_infer_named_tuples_from_objects; format_settings.json.throw_on_bad_escape_sequence = settings.input_format_json_throw_on_bad_escape_sequence; format_settings.json.ignore_unnecessary_fields = settings.input_format_json_ignore_unnecessary_fields; + format_settings.json.case_insensitive_column_matching = settings.input_format_json_case_insensitive_column_matching; format_settings.null_as_default = settings.input_format_null_as_default; format_settings.force_null_for_omitted_fields = settings.input_format_force_null_for_omitted_fields; format_settings.decimal_trailing_zeros = settings.output_format_decimal_trailing_zeros; @@ -174,7 +176,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.parquet.write_page_index = settings.output_format_parquet_write_page_index; format_settings.parquet.local_read_min_bytes_for_seek = settings.input_format_parquet_local_file_min_bytes_for_seek; format_settings.pretty.charset = settings.output_format_pretty_grid_charset.toString() == "ASCII" ? FormatSettings::Pretty::Charset::ASCII : FormatSettings::Pretty::Charset::UTF8; - format_settings.pretty.color = settings.output_format_pretty_color; + format_settings.pretty.color = settings.output_format_pretty_color.valueOr(2); format_settings.pretty.max_column_pad_width = settings.output_format_pretty_max_column_pad_width; format_settings.pretty.max_rows = settings.output_format_pretty_max_rows; format_settings.pretty.max_value_width = settings.output_format_pretty_max_value_width; @@ -242,6 +244,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.orc.output_row_index_stride = settings.output_format_orc_row_index_stride; format_settings.orc.use_fast_decoder = settings.input_format_orc_use_fast_decoder; format_settings.orc.filter_push_down = settings.input_format_orc_filter_push_down; + format_settings.orc.read_use_writer_time_zone = settings.input_format_orc_read_use_writer_time_zone; format_settings.defaults_for_omitted_fields = settings.input_format_defaults_for_omitted_fields; format_settings.capn_proto.enum_comparing_mode = settings.format_capn_proto_enum_comparising_mode; format_settings.capn_proto.skip_fields_with_unsupported_types_in_schema_inference = settings.input_format_capn_proto_skip_fields_with_unsupported_types_in_schema_inference; @@ -268,9 +271,13 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.markdown.escape_special_characters = settings.output_format_markdown_escape_special_characters; format_settings.bson.output_string_as_string = settings.output_format_bson_string_as_string; format_settings.bson.skip_fields_with_unsupported_types_in_schema_inference = settings.input_format_bson_skip_fields_with_unsupported_types_in_schema_inference; - format_settings.max_binary_string_size = settings.format_binary_max_string_size; - format_settings.max_binary_array_size = settings.format_binary_max_array_size; + format_settings.binary.max_binary_string_size = settings.format_binary_max_string_size; + format_settings.binary.max_binary_array_size = settings.format_binary_max_array_size; + format_settings.binary.encode_types_in_binary_format = settings.output_format_binary_encode_types_in_binary_format; + format_settings.binary.decode_types_in_binary_format = settings.input_format_binary_decode_types_in_binary_format; format_settings.native.allow_types_conversion = settings.input_format_native_allow_types_conversion; + format_settings.native.encode_types_in_binary_format = settings.output_format_native_encode_types_in_binary_format; + format_settings.native.decode_types_in_binary_format = settings.input_format_native_decode_types_in_binary_format; format_settings.max_parser_depth = context->getSettingsRef().max_parser_depth; format_settings.client_protocol_version = context->getClientProtocolVersion(); format_settings.date_time_overflow_behavior = settings.date_time_overflow_behavior; diff --git a/src/Formats/FormatSettings.h b/src/Formats/FormatSettings.h index 466ce0fe3c9..72d1515b7e1 100644 --- a/src/Formats/FormatSettings.h +++ b/src/Formats/FormatSettings.h @@ -1,10 +1,9 @@ #pragma once -#include #include +#include #include #include -#include namespace DB { @@ -106,8 +105,6 @@ struct FormatSettings UInt64 input_allow_errors_num = 0; Float32 input_allow_errors_ratio = 0; - UInt64 max_binary_string_size = 1_GiB; - UInt64 max_binary_array_size = 1_GiB; UInt64 client_protocol_version = 0; UInt64 max_parser_depth = DBMS_DEFAULT_MAX_PARSER_DEPTH; @@ -121,6 +118,14 @@ struct FormatSettings ZSTD }; + struct + { + UInt64 max_binary_string_size = 1_GiB; + UInt64 max_binary_array_size = 1_GiB; + bool encode_types_in_binary_format = false; + bool decode_types_in_binary_format = false; + } binary{}; + struct { UInt64 row_group_size = 1000000; @@ -228,6 +233,7 @@ struct FormatSettings bool infer_incomplete_types_as_strings = true; bool throw_on_bad_escape_sequence = true; bool ignore_unnecessary_fields = true; + bool case_insensitive_column_matching = false; } json{}; struct @@ -286,7 +292,8 @@ struct FormatSettings UInt64 max_value_width = 10000; UInt64 max_value_width_apply_for_single_value = false; bool highlight_digit_groups = true; - SettingFieldUInt64Auto color{"auto"}; + /// Set to 2 for auto + UInt64 color = 2; bool output_format_pretty_row_numbers = false; UInt64 output_format_pretty_single_large_number_tip_threshold = 1'000'000; @@ -402,6 +409,7 @@ struct FormatSettings bool use_fast_decoder = true; bool filter_push_down = true; UInt64 output_row_index_stride = 10'000; + bool read_use_writer_time_zone = false; } orc{}; /// For capnProto format we should determine how to @@ -457,6 +465,8 @@ struct FormatSettings struct { bool allow_types_conversion = true; + bool encode_types_in_binary_format = false; + bool decode_types_in_binary_format = false; } native{}; struct diff --git a/src/Formats/JSONExtractTree.cpp b/src/Formats/JSONExtractTree.cpp new file mode 100644 index 00000000000..242d2dc9f80 --- /dev/null +++ b/src/Formats/JSONExtractTree.cpp @@ -0,0 +1,1660 @@ +#include +#include + +#include +#if USE_SIMDJSON +#include +#endif +#if USE_RAPIDJSON +#include +#endif + +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int ILLEGAL_TYPE_OF_ARGUMENT; +} + +template +void jsonElementToString(const typename JSONParser::Element & element, WriteBuffer & buf, const FormatSettings & format_settings) +{ + if (element.isInt64()) + { + writeIntText(element.getInt64(), buf); + return; + } + if (element.isUInt64()) + { + writeIntText(element.getUInt64(), buf); + return; + } + if (element.isDouble()) + { + writeFloatText(element.getDouble(), buf); + return; + } + if (element.isBool()) + { + if (element.getBool()) + writeCString("true", buf); + else + writeCString("false", buf); + return; + } + if (element.isString()) + { + writeJSONString(element.getString(), buf, format_settings); + return; + } + if (element.isArray()) + { + writeChar('[', buf); + bool need_comma = false; + for (auto value : element.getArray()) + { + if (std::exchange(need_comma, true)) + writeChar(',', buf); + jsonElementToString(value, buf, format_settings); + } + writeChar(']', buf); + return; + } + if (element.isObject()) + { + writeChar('{', buf); + bool need_comma = false; + for (auto [key, value] : element.getObject()) + { + if (std::exchange(need_comma, true)) + writeChar(',', buf); + writeJSONString(key, buf, format_settings); + writeChar(':', buf); + jsonElementToString(value, buf, format_settings); + } + writeChar('}', buf); + return; + } + if (element.isNull()) + { + writeCString("null", buf); + return; + } +} + +template +bool tryGetNumericValueFromJSONElement( + NumberType & value, const typename JSONParser::Element & element, bool convert_bool_to_integer, String & error) +{ + switch (element.type()) + { + case ElementType::DOUBLE: + if constexpr (std::is_floating_point_v) + { + /// We permit inaccurate conversion of double to float. + /// Example: double 0.1 from JSON is not representable in float. + /// But it will be more convenient for user to perform conversion. + value = static_cast(element.getDouble()); + } + else if (!accurate::convertNumeric(element.getDouble(), value)) + { + error = fmt::format("cannot convert double value {} to {}", element.getDouble(), TypeName); + return false; + } + break; + case ElementType::UINT64: + if (!accurate::convertNumeric(element.getUInt64(), value)) + { + error = fmt::format("cannot convert UInt64 value {} to {}", element.getUInt64(), TypeName); + return false; + } + break; + case ElementType::INT64: + if (!accurate::convertNumeric(element.getInt64(), value)) + { + error = fmt::format("cannot convert Int64 value {} to {}", element.getInt64(), TypeName); + return false; + } + break; + case ElementType::BOOL: + if constexpr (is_integer) + { + if (convert_bool_to_integer) + { + value = static_cast(element.getBool()); + break; + } + } + error = fmt::format("cannot convert bool value to {}", TypeName); + return false; + case ElementType::STRING: { + auto rb = ReadBufferFromMemory{element.getString()}; + if constexpr (std::is_floating_point_v) + { + if (!tryReadFloatText(value, rb) || !rb.eof()) + { + error = fmt::format("cannot parse {} value here: {}", TypeName, element.getString()); + return false; + } + } + else + { + if (tryReadIntText(value, rb) && rb.eof()) + break; + + /// Try to parse float and convert it to integer. + Float64 tmp_float; + rb.position() = rb.buffer().begin(); + if (!tryReadFloatText(tmp_float, rb) || !rb.eof()) + { + error = fmt::format("cannot parse {} value here: {}", TypeName, element.getString()); + return false; + } + + if (!accurate::convertNumeric(tmp_float, value)) + { + error = fmt::format("cannot parse {} value here: {}", TypeName, element.getString()); + return false; + } + } + break; + } + default: + return false; + } + + return true; +} + +namespace +{ + +template +String jsonElementToString(const typename JSONParser::Element & element, const FormatSettings & format_settings) +{ + WriteBufferFromOwnString buf; + jsonElementToString(element, buf, format_settings); + return buf.str(); +} + +template +class NumericNode : public JSONExtractTreeNode +{ +public: + explicit NumericNode(bool is_bool_type_ = false) : is_bool_type(is_bool_type_) { } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull()) + { + if (format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + error = fmt::format("cannot parse {} value from null", TypeName); + return false; + } + + NumberType value; + if (!tryGetNumericValueFromJSONElement(value, element, insert_settings.convert_bool_to_integer || is_bool_type, error)) + { + if (error.empty()) + error = fmt::format("cannot read {} value from JSON element: {}", TypeName, jsonElementToString(element, format_settings)); + return false; + } + + if (is_bool_type) + value = static_cast(value); + + auto & col_vec = assert_cast &>(column); + col_vec.insertValue(value); + return true; + } + +protected: + bool is_bool_type; +}; + +template +class LowCardinalityNumericNode : public NumericNode +{ +public: + explicit LowCardinalityNumericNode(bool is_nullable_, bool is_bool_type_ = false) + : NumericNode(is_bool_type_), is_nullable(is_nullable_) + { + } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull()) + { + if (is_nullable || format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + error = fmt::format("cannot parse {} value from null", TypeName); + return false; + } + + NumberType value; + if (!tryGetNumericValueFromJSONElement(value, element, insert_settings.convert_bool_to_integer || this->is_bool_type, error)) + { + if (error.empty()) + error = fmt::format("cannot read {} value from JSON element: {}", TypeName, jsonElementToString(element, format_settings)); + return false; + } + + if (this->is_bool_type) + value = static_cast(value); + + auto & col_lc = assert_cast(column); + col_lc.insertData(reinterpret_cast(&value), sizeof(value)); + return true; + } + +private: + bool is_nullable; +}; + +template +class StringNode : public JSONExtractTreeNode +{ +public: + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull()) + { + if (format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + error = "cannot parse String value from null"; + return false; + } + + if (!element.isString()) + { + auto & col_str = assert_cast(column); + auto & chars = col_str.getChars(); + WriteBufferFromVector buf(chars, AppendModeTag()); + jsonElementToString(element, buf, format_settings); + buf.finalize(); + chars.push_back(0); + col_str.getOffsets().push_back(chars.size()); + } + else + { + auto value = element.getString(); + auto & col_str = assert_cast(column); + col_str.insertData(value.data(), value.size()); + } + return true; + } +}; + +template +class LowCardinalityStringNode : public JSONExtractTreeNode +{ +public: + explicit LowCardinalityStringNode(bool is_nullable_) : is_nullable(is_nullable_) { } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull()) + { + if (is_nullable || format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + error = "cannot parse String value from null"; + return false; + } + + if (!element.isString()) + { + auto value = jsonElementToString(element, format_settings); + assert_cast(column).insertData(value.data(), value.size()); + } + else + { + auto value = element.getString(); + assert_cast(column).insertData(value.data(), value.size()); + } + + return true; + } + +private: + bool is_nullable; +}; + +template +class FixedStringNode : public JSONExtractTreeNode +{ +public: + explicit FixedStringNode(size_t fixed_length_) : fixed_length(fixed_length_) { } + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull()) + { + if (format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + error = "cannot parse FixedString value from null"; + return false; + } + + if (!element.isString()) + return checkValueSizeAndInsert(column, jsonElementToString(element, format_settings), error); + return checkValueSizeAndInsert(column, element.getString(), error); + } + +private: + template + bool checkValueSizeAndInsert(IColumn & column, const T & value, String & error) const + { + if (value.size() > fixed_length) + { + error = fmt::format("too large string for FixedString({}): {}", fixed_length, value); + return false; + } + assert_cast(column).insertData(value.data(), value.size()); + return true; + } + + size_t fixed_length; +}; + +template +class LowCardinalityFixedStringNode : public JSONExtractTreeNode +{ +public: + explicit LowCardinalityFixedStringNode(bool is_nullable_, size_t fixed_length_) : is_nullable(is_nullable_), fixed_length(fixed_length_) + { + } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull()) + { + if (is_nullable || format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + error = "cannot parse FixedString value from null"; + return false; + } + + if (!element.isString()) + return checkValueSizeAndInsert(column, jsonElementToString(element, format_settings), error); + return checkValueSizeAndInsert(column, element.getString(), error); + } + +private: + template + bool checkValueSizeAndInsert(IColumn & column, const T & value, String & error) const + { + if (value.size() > fixed_length) + { + error = fmt::format("too large string for FixedString({}): {}", fixed_length, value); + return false; + } + + // For the non low cardinality case of FixedString, the padding is done in the FixedString Column implementation. + // In order to avoid having to pass the data to a FixedString Column and read it back (which would slow down the execution) + // the data is padded here and written directly to the Low Cardinality Column + if (value.size() == fixed_length) + { + assert_cast(column).insertData(value.data(), value.size()); + } + else + { + String padded_value(value); + padded_value.resize(fixed_length, '\0'); + assert_cast(column).insertData(padded_value.data(), padded_value.size()); + } + return true; + } + + bool is_nullable; + size_t fixed_length; +}; + +template +class UUIDNode : public JSONExtractTreeNode +{ +public: + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + if (!element.isString()) + { + error = fmt::format("cannot read UUID value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + + auto data = element.getString(); + UUID uuid; + if (!tryParse(uuid, data)) + { + error = fmt::format("cannot parse UUID value here: {}", data); + return false; + } + + assert_cast(column).insert(uuid); + return true; + } + + + static bool tryParse(UUID & uuid, std::string_view data) + { + ReadBufferFromMemory buf(data.data(), data.size()); + return tryReadUUIDText(uuid, buf) && buf.eof(); + } +}; + +template +class LowCardinalityUUIDNode : public JSONExtractTreeNode +{ +public: + explicit LowCardinalityUUIDNode(bool is_nullable_) : is_nullable(is_nullable_) { } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && (is_nullable || format_settings.null_as_default)) + { + column.insertDefault(); + return true; + } + + if (!element.isString()) + { + error = fmt::format("cannot read UUID value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + + auto data = element.getString(); + ReadBufferFromMemory buf(data.data(), data.size()); + UUID uuid; + if (!tryReadUUIDText(uuid, buf) || !buf.eof()) + { + error = fmt::format("cannot parse UUID value here: {}", data); + return false; + } + assert_cast(column).insertData(reinterpret_cast(&uuid), sizeof(uuid)); + return true; + } + +private: + bool is_nullable; +}; + +template +class DateNode : public JSONExtractTreeNode +{ +public: + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + if (!element.isString()) + { + error = fmt::format("cannot read Date value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + + auto data = element.getString(); + ReadBufferFromMemory buf(data.data(), data.size()); + DateType date; + if (!tryReadDateText(date, buf) || !buf.eof()) + { + error = fmt::format("cannot parse Date value here: {}", data); + return false; + } + + assert_cast &>(column).insertValue(date); + return true; + } +}; + +template +class DateTimeNode : public JSONExtractTreeNode, public TimezoneMixin +{ +public: + explicit DateTimeNode(const DataTypeDateTime & datetime_type) : TimezoneMixin(datetime_type) { } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + time_t value; + if (element.isString()) + { + if (!tryParse(value, element.getString(), format_settings.date_time_input_format)) + { + error = fmt::format("cannot parse DateTime value here: {}", element.getString()); + return false; + } + } + else if (element.isUInt64()) + { + value = element.getUInt64(); + } + else + { + error = fmt::format("cannot read DateTime value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + + assert_cast(column).insert(value); + return true; + } + + bool tryParse(time_t & value, std::string_view data, FormatSettings::DateTimeInputFormat date_time_input_format) const + { + ReadBufferFromMemory buf(data.data(), data.size()); + switch (date_time_input_format) + { + case FormatSettings::DateTimeInputFormat::Basic: + if (tryReadDateTimeText(value, buf, time_zone) && buf.eof()) + return true; + break; + case FormatSettings::DateTimeInputFormat::BestEffort: + if (tryParseDateTimeBestEffort(value, buf, time_zone, utc_time_zone) && buf.eof()) + return true; + break; + case FormatSettings::DateTimeInputFormat::BestEffortUS: + if (tryParseDateTimeBestEffortUS(value, buf, time_zone, utc_time_zone) && buf.eof()) + return true; + break; + } + + return false; + } +}; + +template +class DecimalNode : public JSONExtractTreeNode +{ +public: + explicit DecimalNode(const DataTypePtr & type) : scale(assert_cast &>(*type).getScale()) { } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + DecimalType value{}; + + switch (element.type()) + { + case ElementType::DOUBLE: + value = convertToDecimal, DataTypeDecimal>(element.getDouble(), scale); + break; + case ElementType::UINT64: + value = convertToDecimal, DataTypeDecimal>(element.getUInt64(), scale); + break; + case ElementType::INT64: + value = convertToDecimal, DataTypeDecimal>(element.getInt64(), scale); + break; + case ElementType::STRING: { + auto rb = ReadBufferFromMemory{element.getString()}; + if (!SerializationDecimal::tryReadText(value, rb, DecimalUtils::max_precision, scale)) + { + error = fmt::format("cannot parse Decimal value here: {}", element.getString()); + return false; + } + break; + } + case ElementType::NULL_VALUE: { + if (!format_settings.null_as_default) + { + error = "cannot convert null to Decimal value"; + return false; + } + break; + } + default: + { + error = fmt::format("cannot read Decimal value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + } + + assert_cast &>(column).insertValue(value); + return true; + } + +private: + UInt32 scale; +}; + + +template +class DateTime64Node : public JSONExtractTreeNode, public TimezoneMixin +{ +public: + explicit DateTime64Node(const DataTypeDateTime64 & datetime64_type) : TimezoneMixin(datetime64_type), scale(datetime64_type.getScale()) + { + } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + DateTime64 value; + if (element.isString()) + { + if (!tryParse(value, element.getString(), format_settings.date_time_input_format)) + { + error = fmt::format("cannot parse DateTime64 value here: {}", element.getString()); + return false; + } + } + else + { + switch (element.type()) + { + case ElementType::DOUBLE: + value = convertToDecimal, DataTypeDecimal>(element.getDouble(), scale); + break; + case ElementType::UINT64: + value = convertToDecimal, DataTypeDecimal>(element.getUInt64(), scale); + break; + case ElementType::INT64: + value = convertToDecimal, DataTypeDecimal>(element.getInt64(), scale); + break; + default: + error = fmt::format("cannot read DateTime64 value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + } + + assert_cast(column).insert(value); + return true; + } + + bool tryParse(DateTime64 & value, std::string_view data, FormatSettings::DateTimeInputFormat date_time_input_format) const + { + ReadBufferFromMemory buf(data.data(), data.size()); + switch (date_time_input_format) + { + case FormatSettings::DateTimeInputFormat::Basic: + if (tryReadDateTime64Text(value, scale, buf, time_zone) && buf.eof()) + return true; + break; + case FormatSettings::DateTimeInputFormat::BestEffort: + if (tryParseDateTime64BestEffort(value, scale, buf, time_zone, utc_time_zone) && buf.eof()) + return true; + break; + case FormatSettings::DateTimeInputFormat::BestEffortUS: + if (tryParseDateTime64BestEffortUS(value, scale, buf, time_zone, utc_time_zone) && buf.eof()) + return true; + break; + } + + return false; + } + +private: + UInt32 scale; +}; + +template +class EnumNode : public JSONExtractTreeNode +{ +public: + explicit EnumNode(const std::vector> & name_value_pairs_) : name_value_pairs(name_value_pairs_) + { + for (const auto & name_value_pair : name_value_pairs) + { + name_to_value_map.emplace(name_value_pair.first, name_value_pair.second); + only_values.emplace(name_value_pair.second); + } + } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull()) + { + if (format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + error = "cannot convert null to Enum value"; + return false; + } + + auto & col_vec = assert_cast &>(column); + + if (element.isInt64()) + { + Type value; + if (!accurate::convertNumeric(element.getInt64(), value) || !only_values.contains(value)) + { + error = fmt::format("cannot convert value {} to enum: there is no such value in enum", element.getInt64()); + return false; + } + col_vec.insertValue(value); + return true; + } + + if (element.isUInt64()) + { + Type value; + if (!accurate::convertNumeric(element.getUInt64(), value) || !only_values.contains(value)) + { + error = fmt::format("cannot convert value {} to enum: there is no such value in enum", element.getUInt64()); + return false; + } + col_vec.insertValue(value); + return true; + } + + if (element.isString()) + { + auto value = name_to_value_map.find(element.getString()); + if (value == name_to_value_map.end()) + { + error = fmt::format("cannot convert value {} to enum: there is no such value in enum", element.getString()); + return false; + } + col_vec.insertValue(value->second); + return true; + } + + error = fmt::format("cannot read Enum value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + +private: + std::vector> name_value_pairs; + std::unordered_map name_to_value_map; + std::unordered_set only_values; +}; + +template +class IPv4Node : public JSONExtractTreeNode +{ +public: + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + if (!element.isString()) + { + error = fmt::format("cannot read IPv4 value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + + auto data = element.getString(); + IPv4 value; + if (!tryParse(value, data)) + { + error = fmt::format("cannot parse IPv4 value here: {}", data); + return false; + } + + assert_cast(column).insert(value); + return true; + } + + static bool tryParse(IPv4 & value, std::string_view data) + { + ReadBufferFromMemory buf(data.data(), data.size()); + return tryReadIPv4Text(value, buf) && buf.eof(); + } +}; + +template +class IPv6Node : public JSONExtractTreeNode +{ +public: + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings &, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + if (!element.isString()) + { + error = fmt::format("cannot read IPv6 value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + + auto data = element.getString(); + IPv6 value; + if (!tryParse(value, data)) + { + error = fmt::format("cannot parse IPv6 value here: {}", data); + return false; + } + + assert_cast(column).insert(value); + return true; + } + + + static bool tryParse(IPv6 & value, std::string_view data) + { + ReadBufferFromMemory buf(data.data(), data.size()); + return tryReadIPv6Text(value, buf) && buf.eof(); + } +}; + +template +class NullableNode : public JSONExtractTreeNode +{ +public: + explicit NullableNode(std::unique_ptr> nested_) : nested(std::move(nested_)) { } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull()) + { + column.insertDefault(); + return true; + } + + auto & col_null = assert_cast(column); + if (!nested->insertResultToColumn(col_null.getNestedColumn(), element, insert_settings, format_settings, error)) + return false; + col_null.getNullMapColumn().insertValue(0); + return true; + } + +private: + std::unique_ptr> nested; +}; + +template +class LowCardinalityNode : public JSONExtractTreeNode +{ +public: + explicit LowCardinalityNode(bool is_nullable_, std::unique_ptr> nested_) + : is_nullable(is_nullable_), nested(std::move(nested_)) + { + } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && (is_nullable || format_settings.null_as_default)) + { + column.insertDefault(); + return true; + } + + auto & col_lc = assert_cast(column); + auto tmp_nested = col_lc.getDictionary().getNestedColumn()->cloneEmpty(); + if (!nested->insertResultToColumn(*tmp_nested, element, insert_settings, format_settings, error)) + return false; + + col_lc.insertFromFullColumn(*tmp_nested, 0); + return true; + } + +private: + bool is_nullable; + std::unique_ptr> nested; +}; + +template +class ArrayNode : public JSONExtractTreeNode +{ +public: + explicit ArrayNode(std::unique_ptr> nested_) : nested(std::move(nested_)) { } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + if (element.isNull() && format_settings.null_as_default) + { + column.insertDefault(); + return true; + } + + if (!element.isArray()) + { + error = fmt::format("cannot read Array value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + + auto array = element.getArray(); + + auto & col_arr = assert_cast(column); + auto & data = col_arr.getData(); + size_t old_size = data.size(); + bool were_valid_elements = false; + + for (auto value : array) + { + if (nested->insertResultToColumn(data, value, insert_settings, format_settings, error)) + { + were_valid_elements = true; + } + else if (insert_settings.insert_default_on_invalid_elements_in_complex_types) + { + data.insertDefault(); + } + else + { + data.popBack(data.size() - old_size); + return false; + } + } + + if (!were_valid_elements) + { + data.popBack(data.size() - old_size); + return false; + } + + col_arr.getOffsets().push_back(data.size()); + return true; + } + +private: + std::unique_ptr> nested; +}; + +template +class TupleNode : public JSONExtractTreeNode +{ +public: + TupleNode(std::vector>> nested_, const std::vector & explicit_names_) + : nested(std::move(nested_)), explicit_names(explicit_names_) + { + for (size_t i = 0; i != explicit_names.size(); ++i) + name_to_index_map.emplace(explicit_names[i], i); + } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + auto & tuple = assert_cast(column); + size_t old_size = column.size(); + bool were_valid_elements = false; + + auto set_size = [&](size_t size) + { + for (size_t i = 0; i != tuple.tupleSize(); ++i) + { + auto & col = tuple.getColumn(i); + if (col.size() != size) + { + if (col.size() > size) + col.popBack(col.size() - size); + else + while (col.size() < size) + col.insertDefault(); + } + } + }; + + if (element.isArray()) + { + auto array = element.getArray(); + auto it = array.begin(); + + for (size_t index = 0; (index != nested.size()) && (it != array.end()); ++index) + { + if (nested[index]->insertResultToColumn(tuple.getColumn(index), *it++, insert_settings, format_settings, error)) + { + were_valid_elements = true; + } + else if (insert_settings.insert_default_on_invalid_elements_in_complex_types) + { + tuple.getColumn(index).insertDefault(); + } + else + { + set_size(old_size); + error += fmt::format("(during reading tuple {} element)", index); + return false; + } + } + + set_size(old_size + static_cast(were_valid_elements)); + return were_valid_elements; + } + + if (element.isObject()) + { + auto object = element.getObject(); + if (name_to_index_map.empty()) + { + auto it = object.begin(); + for (size_t index = 0; (index != nested.size()) && (it != object.end()); ++index) + { + if (nested[index]->insertResultToColumn(tuple.getColumn(index), (*it++).second, insert_settings, format_settings, error)) + { + were_valid_elements = true; + } + else if (insert_settings.insert_default_on_invalid_elements_in_complex_types) + { + tuple.getColumn(index).insertDefault(); + } + else + { + set_size(old_size); + error += fmt::format("(during reading tuple {} element)", index); + return false; + } + } + } + else + { + for (const auto & [key, value] : object) + { + auto index = name_to_index_map.find(key); + if (index != name_to_index_map.end()) + { + if (nested[index->second]->insertResultToColumn(tuple.getColumn(index->second), value, insert_settings, format_settings, error)) + { + were_valid_elements = true; + } + else if (!insert_settings.insert_default_on_invalid_elements_in_complex_types) + { + set_size(old_size); + error += fmt::format("(during reading tuple element \"{}\")", key); + return false; + } + } + } + } + + set_size(old_size + static_cast(were_valid_elements)); + return were_valid_elements; + } + + error = fmt::format("cannot read Tuple value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + +private: + std::vector>> nested; + std::vector explicit_names; + std::unordered_map name_to_index_map; +}; + +template +class MapNode : public JSONExtractTreeNode +{ +public: + explicit MapNode(std::unique_ptr> value_) : value(std::move(value_)) { } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + if (!element.isObject()) + { + error = fmt::format("cannot read Map value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + + auto & map_col = assert_cast(column); + auto & offsets = map_col.getNestedColumn().getOffsets(); + auto & tuple_col = map_col.getNestedData(); + auto & key_col = tuple_col.getColumn(0); + auto & value_col = tuple_col.getColumn(1); + size_t old_size = tuple_col.size(); + + auto object = element.getObject(); + auto it = object.begin(); + for (; it != object.end(); ++it) + { + auto pair = *it; + + /// Insert key + key_col.insertData(pair.first.data(), pair.first.size()); + + /// Insert value + if (!value->insertResultToColumn(value_col, pair.second, insert_settings, format_settings, error)) + { + if (insert_settings.insert_default_on_invalid_elements_in_complex_types) + { + value_col.insertDefault(); + } + else + { + key_col.popBack(key_col.size() - offsets.back()); + value_col.popBack(value_col.size() - offsets.back()); + error += fmt::format("(during reading value of key \"{}\")", pair.first); + return false; + } + } + } + + offsets.push_back(old_size + object.size()); + return true; + } + +private: + std::unique_ptr> value; +}; + +template +class VariantNode : public JSONExtractTreeNode +{ +public: + VariantNode(std::vector>> variant_nodes_, std::vector order_) + : variant_nodes(std::move(variant_nodes_)), order(std::move(order_)) + { + } + + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + auto & column_variant = assert_cast(column); + for (size_t i : order) + { + auto & variant = column_variant.getVariantByGlobalDiscriminator(i); + if (variant_nodes[i]->insertResultToColumn(variant, element, insert_settings, format_settings, error)) + { + column_variant.getLocalDiscriminators().push_back(column_variant.localDiscriminatorByGlobal(i)); + column_variant.getOffsets().push_back(variant.size() - 1); + return true; + } + } + + error = fmt::format("cannot read Map value from JSON element: {}", jsonElementToString(element, format_settings)); + return false; + } + +private: + std::vector>> variant_nodes; + /// Order in which we should try variants nodes. + /// For example, String should be always the last one. + std::vector order; +}; + + +template +class DynamicNode : public JSONExtractTreeNode +{ +public: + bool insertResultToColumn( + IColumn & column, + const typename JSONParser::Element & element, + const JSONExtractInsertSettings & insert_settings, + const FormatSettings & format_settings, + String & error) const override + { + auto & column_dynamic = assert_cast(column); + /// First, check if element is NULL. + if (element.isNull()) + { + column_dynamic.insertDefault(); + return true; + } + + auto & variant_column = column_dynamic.getVariantColumn(); + auto variant_info = column_dynamic.getVariantInfo(); + /// Second, infer ClickHouse type for this element and add it as a new variant. + auto element_type = elementToDataType(element, format_settings); + if (column_dynamic.addNewVariant(element_type)) + { + auto node = buildJSONExtractTree(element_type, "Dynamic inference"); + auto global_discriminator = variant_info.variant_name_to_discriminator[element_type->getName()]; + auto & variant = variant_column.getVariantByGlobalDiscriminator(global_discriminator); + if (!node->insertResultToColumn(variant, element, insert_settings, format_settings, error)) + return false; + variant_column.getLocalDiscriminators().push_back(variant_column.localDiscriminatorByGlobal(global_discriminator)); + variant_column.getOffsets().push_back(variant.size() - 1); + return true; + } + + /// We couldn't add new variant. Try to insert element into current variants. + auto variant_node = buildJSONExtractTree(variant_info.variant_type, "Dynamic inference"); + if (variant_node->insertResultToColumn(variant_column, element, insert_settings, format_settings, error)) + return true; + + /// We couldn't insert element into any existing variant, add String variant and read value as String. + column_dynamic.addStringVariant(); + auto string_global_discriminator = variant_info.variant_name_to_discriminator["String"]; + auto & string_column = variant_column.getVariantByGlobalDiscriminator(string_global_discriminator); + if (!getStringNode()->insertResultToColumn(string_column, element, insert_settings, format_settings, error)) + return false; + variant_column.getLocalDiscriminators().push_back(variant_column.localDiscriminatorByGlobal(string_global_discriminator)); + variant_column.getOffsets().push_back(string_column.size() - 1); + return true; + } + + static const std::unique_ptr> & getStringNode() + { + static const std::unique_ptr> string_node + = buildJSONExtractTree(std::make_shared(), "Dynamic inference"); + return string_node; + } + + static DataTypePtr elementToDataType(const typename JSONParser::Element & element, const FormatSettings & format_settings) + { + JSONInferenceInfo json_inference_info; + auto type = elementToDataTypeImpl(element, format_settings, json_inference_info); + transformFinalInferredJSONTypeIfNeeded(type, format_settings, &json_inference_info); + return type; + } + +private: + static DataTypePtr elementToDataTypeImpl(const typename JSONParser::Element & element, const FormatSettings & format_settings, JSONInferenceInfo & json_inference_info) + { + switch (element.type()) + { + case ElementType::NULL_VALUE: + return makeNullable(std::make_shared()); + case ElementType::BOOL: + return DataTypeFactory::instance().get("Bool"); + case ElementType::INT64: + { + auto type = std::make_shared(); + if (element.getInt64() < 0) + json_inference_info.negative_integers.insert(type.get()); + return type; + } + case ElementType::UINT64: + return std::make_shared(); + case ElementType::DOUBLE: + return std::make_shared(); + case ElementType::STRING: + { + auto data = element.getString(); + + if (auto type = tryInferDateOrDateTimeFromString(data, format_settings)) + return type; + + if (format_settings.json.try_infer_numbers_from_strings) + { + if (auto type = tryInferJSONNumberFromString(data, format_settings, &json_inference_info)) + { + json_inference_info.numbers_parsed_from_json_strings.insert(type.get()); + return type; + } + } + + return std::make_shared(); + } + case ElementType::ARRAY: + { + auto array = element.getArray(); + DataTypes types; + types.reserve(array.size()); + for (auto value : array) + types.push_back(makeNullableSafe(elementToDataTypeImpl(value, format_settings, json_inference_info))); + + if (types.empty()) + return std::make_shared(makeNullable(std::make_shared())); + + if (checkIfTypesAreEqual(types)) + return std::make_shared(types.back()); + + /// For JSON if we have not complete types, we should not try to transform them + /// and return it as a Tuple. + /// For example, if we have types [Nullable(Float64), Nullable(Nothing), Nullable(Float64)] + /// it can be Array(Nullable(Float64)) or Tuple(Nullable(Float64), , Nullable(Float64)) and + /// we can't determine which one it is right now. But we will be able to do it later + /// when we will have the final top level type. + /// For example, we can have JSON element [[42.42, null, 43.43], [44.44, "Some string", 45.45]] and we should + /// determine the type for this element as Tuple(Nullable(Float64), Nullable(String), Nullable(Float64)). + for (const auto & type : types) + { + if (!checkIfTypeIsComplete(type)) + return std::make_shared(types); + } + + auto types_copy = types; + transformInferredJSONTypesIfNeeded(types_copy, format_settings, &json_inference_info); + + if (checkIfTypesAreEqual(types_copy)) + return std::make_shared(types_copy.back()); + + return std::make_shared(types); + } + case ElementType::OBJECT: { + /// TODO: Use new JSON type here when it's ready. + return std::make_shared(std::make_shared(), makeNullable(std::make_shared())); + } + } + } +}; + +} + +template +std::unique_ptr> buildJSONExtractTree(const DataTypePtr & type, const char * source_for_exception_message) +{ + switch (type->getTypeId()) + { + case TypeIndex::UInt8: + return std::make_unique>(isBool(type)); + case TypeIndex::UInt16: + return std::make_unique>(); + case TypeIndex::UInt32: + return std::make_unique>(); + case TypeIndex::UInt64: + return std::make_unique>(); + case TypeIndex::UInt128: + return std::make_unique>(); + case TypeIndex::UInt256: + return std::make_unique>(); + case TypeIndex::Int8: + return std::make_unique>(); + case TypeIndex::Int16: + return std::make_unique>(); + case TypeIndex::Int32: + return std::make_unique>(); + case TypeIndex::Int64: + return std::make_unique>(); + case TypeIndex::Int128: + return std::make_unique>(); + case TypeIndex::Int256: + return std::make_unique>(); + case TypeIndex::Float32: + return std::make_unique>(); + case TypeIndex::Float64: + return std::make_unique>(); + case TypeIndex::String: + return std::make_unique>(); + case TypeIndex::FixedString: + return std::make_unique>(assert_cast(*type).getN()); + case TypeIndex::UUID: + return std::make_unique>(); + case TypeIndex::IPv4: + return std::make_unique>(); + case TypeIndex::IPv6: + return std::make_unique>(); + case TypeIndex::Date:; + return std::make_unique>(); + case TypeIndex::Date32: + return std::make_unique>(); + case TypeIndex::DateTime: + return std::make_unique>(assert_cast(*type)); + case TypeIndex::DateTime64: + return std::make_unique>(assert_cast(*type)); + case TypeIndex::Decimal32: + return std::make_unique>(type); + case TypeIndex::Decimal64: + return std::make_unique>(type); + case TypeIndex::Decimal128: + return std::make_unique>(type); + case TypeIndex::Decimal256: + return std::make_unique>(type); + case TypeIndex::Enum8: + return std::make_unique>(assert_cast(*type).getValues()); + case TypeIndex::Enum16: + return std::make_unique>(assert_cast(*type).getValues()); + case TypeIndex::LowCardinality: + { + /// To optimize inserting into LowCardinality we have special nodes for LowCardinality of numeric and string types. + const auto & lc_type = assert_cast(*type); + auto dictionary_type = removeNullable(lc_type.getDictionaryType()); + bool is_nullable = lc_type.isLowCardinalityNullable(); + + switch (dictionary_type->getTypeId()) + { + case TypeIndex::UInt8: + return std::make_unique>(is_nullable, isBool(type)); + case TypeIndex::UInt16: + return std::make_unique>(is_nullable); + case TypeIndex::UInt32: + return std::make_unique>(is_nullable); + case TypeIndex::UInt64: + return std::make_unique>(is_nullable); + case TypeIndex::Int8: + return std::make_unique>(is_nullable); + case TypeIndex::Int16: + return std::make_unique>(is_nullable); + case TypeIndex::Int32: + return std::make_unique>(is_nullable); + case TypeIndex::Int64: + return std::make_unique>(is_nullable); + case TypeIndex::Float32: + return std::make_unique>(is_nullable); + case TypeIndex::Float64: + return std::make_unique>(is_nullable); + case TypeIndex::String: + return std::make_unique>(is_nullable); + case TypeIndex::FixedString: + return std::make_unique>(is_nullable, assert_cast(*dictionary_type).getN()); + case TypeIndex::UUID: + return std::make_unique>(is_nullable); + default: + return std::make_unique>(is_nullable, buildJSONExtractTree(dictionary_type, source_for_exception_message)); + } + } + case TypeIndex::Nullable: + return std::make_unique>(buildJSONExtractTree(assert_cast(*type).getNestedType(), source_for_exception_message)); + case TypeIndex::Array: + return std::make_unique>(buildJSONExtractTree(assert_cast(*type).getNestedType(), source_for_exception_message)); + case TypeIndex::Tuple: + { + const auto & tuple = assert_cast(*type); + const auto & tuple_elements = tuple.getElements(); + std::vector>> elements; + elements.reserve(tuple_elements.size()); + for (const auto & tuple_element : tuple_elements) + elements.emplace_back(buildJSONExtractTree(tuple_element, source_for_exception_message)); + return std::make_unique>(std::move(elements), tuple.haveExplicitNames() ? tuple.getElementNames() : Strings{}); + } + case TypeIndex::Map: + { + const auto & map_type = assert_cast(*type); + const auto & key_type = map_type.getKeyType(); + if (!isString(removeLowCardinality(key_type))) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "{} doesn't support the return type schema: {} with key type not String", + source_for_exception_message, + type->getName()); + + const auto & value_type = map_type.getValueType(); + return std::make_unique>(buildJSONExtractTree(value_type, source_for_exception_message)); + } + case TypeIndex::Variant: + { + const auto & variant_type = assert_cast(*type); + const auto & variants = variant_type.getVariants(); + std::vector>> variant_nodes; + variant_nodes.reserve(variants.size()); + for (const auto & variant : variants) + variant_nodes.push_back(buildJSONExtractTree(variant, source_for_exception_message)); + return std::make_unique>(std::move(variant_nodes), SerializationVariant::getVariantsDeserializeTextOrder(variants)); + } + case TypeIndex::Dynamic: + return std::make_unique>(); + default: + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "{} doesn't support the return type schema: {}", + source_for_exception_message, + type->getName()); + } +} + +#if USE_SIMDJSON +template void jsonElementToString(const SimdJSONParser::Element & element, WriteBuffer & buf, const FormatSettings & format_settings); +template std::unique_ptr> buildJSONExtractTree(const DataTypePtr & type, const char * source_for_exception_message); +#endif + +#if USE_RAPIDJSON +template void jsonElementToString(const RapidJSONParser::Element & element, WriteBuffer & buf, const FormatSettings & format_settings); +template std::unique_ptr> buildJSONExtractTree(const DataTypePtr & type, const char * source_for_exception_message); +template bool tryGetNumericValueFromJSONElement(Float64 & value, const RapidJSONParser::Element & element, bool convert_bool_to_integer, String & error); +#else +template void jsonElementToString(const DummyJSONParser::Element & element, WriteBuffer & buf, const FormatSettings & format_settings); +template std::unique_ptr> buildJSONExtractTree(const DataTypePtr & type, const char * source_for_exception_message); +#endif + +} diff --git a/src/Formats/JSONExtractTree.h b/src/Formats/JSONExtractTree.h new file mode 100644 index 00000000000..b5e82506548 --- /dev/null +++ b/src/Formats/JSONExtractTree.h @@ -0,0 +1,41 @@ +#pragma once +#include +#include +#include + + +namespace DB +{ + +struct JSONExtractInsertSettings +{ + /// If false, JSON boolean values won't be inserted into columns with integer types + /// It's used in JSONExtractInt64/JSONExtractUInt64/... functions. + bool convert_bool_to_integer = true; + /// If true, when complex type like Array/Map has both valid and invalid elements, + /// the default value will be inserted on invalid elements. + /// For example, if we have [1, "hello", 2] and type Array(UInt32), + /// we will insert [1, 0, 2] in the column. Used in all JSONExtract functions. + bool insert_default_on_invalid_elements_in_complex_types = false; +}; + +template +class JSONExtractTreeNode +{ +public: + JSONExtractTreeNode() = default; + virtual ~JSONExtractTreeNode() = default; + virtual bool insertResultToColumn(IColumn &, const typename JSONParser::Element &, const JSONExtractInsertSettings & insert_setting, const FormatSettings & format_settings, String & error) const = 0; +}; + +/// Build a tree for insertion JSON element into a column with provided data type. +template +std::unique_ptr> buildJSONExtractTree(const DataTypePtr & type, const char * source_for_exception_message); + +template +void jsonElementToString(const typename JSONParser::Element & element, WriteBuffer & buf, const FormatSettings & format_settings); + +template +bool tryGetNumericValueFromJSONElement(NumberType & value, const typename JSONParser::Element & element, bool convert_bool_to_integer, String & error); + +} diff --git a/src/Formats/NativeReader.cpp b/src/Formats/NativeReader.cpp index fa5d41d6536..f9b56cf4f61 100644 --- a/src/Formats/NativeReader.cpp +++ b/src/Formats/NativeReader.cpp @@ -6,6 +6,7 @@ #include #include +#include #include #include @@ -31,8 +32,8 @@ namespace ErrorCodes } -NativeReader::NativeReader(ReadBuffer & istr_, UInt64 server_revision_) - : istr(istr_), server_revision(server_revision_) +NativeReader::NativeReader(ReadBuffer & istr_, UInt64 server_revision_, std::optional format_settings_) + : istr(istr_), server_revision(server_revision_), format_settings(format_settings_) { } @@ -40,16 +41,12 @@ NativeReader::NativeReader( ReadBuffer & istr_, const Block & header_, UInt64 server_revision_, - bool skip_unknown_columns_, - bool null_as_default_, - bool allow_types_conversion_, + std::optional format_settings_, BlockMissingValues * block_missing_values_) : istr(istr_) , header(header_) , server_revision(server_revision_) - , skip_unknown_columns(skip_unknown_columns_) - , null_as_default(null_as_default_) - , allow_types_conversion(allow_types_conversion_) + , format_settings(std::move(format_settings_)) , block_missing_values(block_missing_values_) { } @@ -83,13 +80,14 @@ void NativeReader::resetParser() use_index = false; } -void NativeReader::readData(const ISerialization & serialization, ColumnPtr & column, ReadBuffer & istr, size_t rows, double avg_value_size_hint) +static void readData(const ISerialization & serialization, ColumnPtr & column, ReadBuffer & istr, const std::optional & format_settings, size_t rows, double avg_value_size_hint) { ISerialization::DeserializeBinaryBulkSettings settings; settings.getter = [&](ISerialization::SubstreamPath) -> ReadBuffer * { return &istr; }; settings.avg_value_size_hint = avg_value_size_hint; settings.position_independent_encoding = false; settings.native_format = true; + settings.data_types_binary_encoding = format_settings && format_settings->native.decode_types_in_binary_format; ISerialization::DeserializeBinaryBulkStatePtr state; @@ -167,8 +165,16 @@ Block NativeReader::read() /// Type String type_name; - readBinary(type_name, istr); - column.type = data_type_factory.get(type_name); + if (format_settings && format_settings->native.decode_types_in_binary_format) + { + column.type = decodeDataType(istr); + type_name = column.type->getName(); + } + else + { + readBinary(type_name, istr); + column.type = data_type_factory.get(type_name); + } setVersionToAggregateFunctions(column.type, true, server_revision); @@ -203,7 +209,7 @@ Block NativeReader::read() double avg_value_size_hint = avg_value_size_hints.empty() ? 0 : avg_value_size_hints[i]; if (rows) /// If no rows, nothing to read. - readData(*serialization, read_column, istr, rows, avg_value_size_hint); + readData(*serialization, read_column, istr, format_settings, rows, avg_value_size_hint); column.column = std::move(read_column); @@ -214,12 +220,12 @@ Block NativeReader::read() { auto & header_column = header.getByName(column.name); - if (null_as_default) + if (format_settings && format_settings->null_as_default) insertNullAsDefaultIfNeeded(column, header_column, header.getPositionByName(column.name), block_missing_values); if (!header_column.type->equals(*column.type)) { - if (allow_types_conversion) + if (format_settings && format_settings->native.allow_types_conversion) { try { @@ -246,7 +252,7 @@ Block NativeReader::read() } else { - if (!skip_unknown_columns) + if (format_settings && !format_settings->skip_unknown_fields) throw Exception(ErrorCodes::INCORRECT_DATA, "Unknown column with name {} found while reading data in Native format", column.name); use_in_result = false; } diff --git a/src/Formats/NativeReader.h b/src/Formats/NativeReader.h index 3cec4afd997..97b6ea22b15 100644 --- a/src/Formats/NativeReader.h +++ b/src/Formats/NativeReader.h @@ -20,7 +20,7 @@ class NativeReader { public: /// If a non-zero server_revision is specified, additional block information may be expected and read. - NativeReader(ReadBuffer & istr_, UInt64 server_revision_); + NativeReader(ReadBuffer & istr_, UInt64 server_revision_, std::optional format_settings_ = std::nullopt); /// For cases when data structure (header) is known in advance. /// NOTE We may use header for data validation and/or type conversions. It is not implemented. @@ -28,9 +28,7 @@ public: ReadBuffer & istr_, const Block & header_, UInt64 server_revision_, - bool skip_unknown_columns_ = false, - bool null_as_default_ = false, - bool allow_types_conversion_ = false, + std::optional format_settings_ = std::nullopt, BlockMissingValues * block_missing_values_ = nullptr); /// For cases when we have an index. It allows to skip columns. Only columns specified in the index will be read. @@ -38,8 +36,6 @@ public: IndexForNativeFormat::Blocks::const_iterator index_block_it_, IndexForNativeFormat::Blocks::const_iterator index_block_end_); - static void readData(const ISerialization & serialization, ColumnPtr & column, ReadBuffer & istr, size_t rows, double avg_value_size_hint); - Block getHeader() const; void resetParser(); @@ -50,9 +46,7 @@ private: ReadBuffer & istr; Block header; UInt64 server_revision; - bool skip_unknown_columns = false; - bool null_as_default = false; - bool allow_types_conversion = false; + std::optional format_settings = std::nullopt; BlockMissingValues * block_missing_values = nullptr; bool use_index = false; diff --git a/src/Formats/NativeWriter.cpp b/src/Formats/NativeWriter.cpp index b150561a5fc..3c87e489b1c 100644 --- a/src/Formats/NativeWriter.cpp +++ b/src/Formats/NativeWriter.cpp @@ -14,6 +14,7 @@ #include #include #include +#include namespace DB { @@ -25,10 +26,20 @@ namespace ErrorCodes NativeWriter::NativeWriter( - WriteBuffer & ostr_, UInt64 client_revision_, const Block & header_, bool remove_low_cardinality_, - IndexForNativeFormat * index_, size_t initial_size_of_file_) - : ostr(ostr_), client_revision(client_revision_), header(header_), - index(index_), initial_size_of_file(initial_size_of_file_), remove_low_cardinality(remove_low_cardinality_) + WriteBuffer & ostr_, + UInt64 client_revision_, + const Block & header_, + std::optional format_settings_, + bool remove_low_cardinality_, + IndexForNativeFormat * index_, + size_t initial_size_of_file_) + : ostr(ostr_) + , client_revision(client_revision_) + , header(header_) + , index(index_) + , initial_size_of_file(initial_size_of_file_) + , remove_low_cardinality(remove_low_cardinality_) + , format_settings(std::move(format_settings_)) { if (index) { @@ -45,7 +56,7 @@ void NativeWriter::flush() } -static void writeData(const ISerialization & serialization, const ColumnPtr & column, WriteBuffer & ostr, UInt64 offset, UInt64 limit) +static void writeData(const ISerialization & serialization, const ColumnPtr & column, WriteBuffer & ostr, const std::optional & format_settings, UInt64 offset, UInt64 limit) { /** If there are columns-constants - then we materialize them. * (Since the data type does not know how to serialize / deserialize constants.) @@ -57,6 +68,7 @@ static void writeData(const ISerialization & serialization, const ColumnPtr & co settings.getter = [&ostr](ISerialization::SubstreamPath) -> WriteBuffer * { return &ostr; }; settings.position_independent_encoding = false; settings.low_cardinality_max_dictionary_size = 0; + settings.data_types_binary_encoding = format_settings && format_settings->native.encode_types_in_binary_format; ISerialization::SerializeBinaryBulkStatePtr state; serialization.serializeBinaryBulkStatePrefix(*full_column, settings, state); @@ -121,15 +133,22 @@ size_t NativeWriter::write(const Block & block) setVersionToAggregateFunctions(column.type, include_version, include_version ? std::optional(client_revision) : std::nullopt); /// Type - String type_name = column.type->getName(); + if (format_settings && format_settings->native.encode_types_in_binary_format) + { + encodeDataType(column.type, ostr); + } + else + { + String type_name = column.type->getName(); - /// For compatibility, we will not send explicit timezone parameter in DateTime data type - /// to older clients, that cannot understand it. - if (client_revision < DBMS_MIN_REVISION_WITH_TIME_ZONE_PARAMETER_IN_DATETIME_DATA_TYPE - && startsWith(type_name, "DateTime(")) - type_name = "DateTime"; + /// For compatibility, we will not send explicit timezone parameter in DateTime data type + /// to older clients, that cannot understand it. + if (client_revision < DBMS_MIN_REVISION_WITH_TIME_ZONE_PARAMETER_IN_DATETIME_DATA_TYPE + && startsWith(type_name, "DateTime(")) + type_name = "DateTime"; - writeStringBinary(type_name, ostr); + writeStringBinary(type_name, ostr); + } /// Serialization. Dynamic, if client supports it. SerializationPtr serialization; @@ -161,7 +180,7 @@ size_t NativeWriter::write(const Block & block) /// Data if (rows) /// Zero items of data is always represented as zero number of bytes. - writeData(*serialization, column.column, ostr, 0, 0); + writeData(*serialization, column.column, ostr, format_settings, 0, 0); if (index) { diff --git a/src/Formats/NativeWriter.h b/src/Formats/NativeWriter.h index 7bb377d2e4a..b4903243d45 100644 --- a/src/Formats/NativeWriter.h +++ b/src/Formats/NativeWriter.h @@ -3,6 +3,7 @@ #include #include #include +#include namespace DB { @@ -23,7 +24,7 @@ public: /** If non-zero client_revision is specified, additional block information can be written. */ NativeWriter( - WriteBuffer & ostr_, UInt64 client_revision_, const Block & header_, bool remove_low_cardinality_ = false, + WriteBuffer & ostr_, UInt64 client_revision_, const Block & header_, std::optional format_settings_ = std::nullopt, bool remove_low_cardinality_ = false, IndexForNativeFormat * index_ = nullptr, size_t initial_size_of_file_ = 0); Block getHeader() const { return header; } @@ -44,6 +45,7 @@ private: CompressedWriteBuffer * ostr_concrete = nullptr; bool remove_low_cardinality; + std::optional format_settings; }; } diff --git a/src/Formats/NumpyDataTypes.h b/src/Formats/NumpyDataTypes.h index 062f743c0ea..6ccdf65f457 100644 --- a/src/Formats/NumpyDataTypes.h +++ b/src/Formats/NumpyDataTypes.h @@ -2,6 +2,7 @@ #include #include #include +#include namespace ErrorCodes { diff --git a/src/Formats/ReadSchemaUtils.cpp b/src/Formats/ReadSchemaUtils.cpp index 735b536986d..617595c19a2 100644 --- a/src/Formats/ReadSchemaUtils.cpp +++ b/src/Formats/ReadSchemaUtils.cpp @@ -7,6 +7,7 @@ #include #include #include +#include namespace DB { diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 31faea2e13e..3c374ada9e6 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -225,19 +225,6 @@ namespace Paths paths; }; - bool checkIfTypesAreEqual(const DataTypes & types) - { - if (types.empty()) - return true; - - for (size_t i = 1; i < types.size(); ++i) - { - if (!types[0]->equals(*types[i])) - return false; - } - return true; - } - void updateTypeIndexes(DataTypes & data_types, TypeIndexesSet & type_indexes) { type_indexes.clear(); @@ -272,24 +259,31 @@ namespace type_indexes.erase(TypeIndex::Nothing); } - /// If we have both Int64 and UInt64, convert all Int64 to UInt64, + /// If we have both Int64 and UInt64, convert all not-negative Int64 to UInt64, /// because UInt64 is inferred only in case of Int64 overflow. - void transformIntegers(DataTypes & data_types, TypeIndexesSet & type_indexes) + void transformIntegers(DataTypes & data_types, TypeIndexesSet & type_indexes, JSONInferenceInfo * json_info) { if (!type_indexes.contains(TypeIndex::Int64) || !type_indexes.contains(TypeIndex::UInt64)) return; + bool have_negative_integers = false; for (auto & type : data_types) { if (WhichDataType(type).isInt64()) - type = std::make_shared(); + { + bool is_negative = json_info && json_info->negative_integers.contains(type.get()); + have_negative_integers |= is_negative; + if (!is_negative) + type = std::make_shared(); + } } - type_indexes.erase(TypeIndex::Int64); + if (!have_negative_integers) + type_indexes.erase(TypeIndex::Int64); } /// If we have both Int64 and Float64 types, convert all Int64 to Float64. - void transformIntegersAndFloatsToFloats(DataTypes & data_types, TypeIndexesSet & type_indexes) + void transformIntegersAndFloatsToFloats(DataTypes & data_types, TypeIndexesSet & type_indexes, JSONInferenceInfo * json_info) { bool have_floats = type_indexes.contains(TypeIndex::Float64); bool have_integers = type_indexes.contains(TypeIndex::Int64) || type_indexes.contains(TypeIndex::UInt64); @@ -300,7 +294,12 @@ namespace { WhichDataType which(type); if (which.isInt64() || which.isUInt64()) - type = std::make_shared(); + { + auto new_type = std::make_shared(); + if (json_info && json_info->numbers_parsed_from_json_strings.erase(type.get())) + json_info->numbers_parsed_from_json_strings.insert(new_type.get()); + type = new_type; + } } type_indexes.erase(TypeIndex::Int64); @@ -635,9 +634,9 @@ namespace if (settings.try_infer_integers) { /// Transform Int64 to UInt64 if needed. - transformIntegers(data_types, type_indexes); + transformIntegers(data_types, type_indexes, json_info); /// Transform integers to floats if needed. - transformIntegersAndFloatsToFloats(data_types, type_indexes); + transformIntegersAndFloatsToFloats(data_types, type_indexes, json_info); } /// Transform Date to DateTime or both to String if needed. @@ -887,7 +886,7 @@ namespace } template - DataTypePtr tryInferNumber(ReadBuffer & buf, const FormatSettings & settings) + DataTypePtr tryInferNumber(ReadBuffer & buf, const FormatSettings & settings, JSONInferenceInfo * json_info) { if (buf.eof()) return nullptr; @@ -911,7 +910,12 @@ namespace Int64 tmp_int; buf.position() = number_start; if (tryReadIntText(tmp_int, buf)) - return std::make_shared(); + { + auto type = std::make_shared(); + if (json_info && tmp_int < 0) + json_info->negative_integers.insert(type.get()); + return type; + } /// In case of Int64 overflow we can try to infer UInt64. UInt64 tmp_uint; @@ -934,7 +938,12 @@ namespace Int64 tmp_int; if (tryReadIntText(tmp_int, peekable_buf)) - return std::make_shared(); + { + auto type = std::make_shared(); + if (json_info && tmp_int < 0) + json_info->negative_integers.insert(type.get()); + return type; + } peekable_buf.rollbackToCheckpoint(/* drop= */ true); /// In case of Int64 overflow we can try to infer UInt64. @@ -952,7 +961,7 @@ namespace } template - DataTypePtr tryInferNumberFromStringImpl(std::string_view field, const FormatSettings & settings) + DataTypePtr tryInferNumberFromStringImpl(std::string_view field, const FormatSettings & settings, JSONInferenceInfo * json_inference_info = nullptr) { ReadBufferFromString buf(field); @@ -960,7 +969,12 @@ namespace { Int64 tmp_int; if (tryReadIntText(tmp_int, buf) && buf.eof()) - return std::make_shared(); + { + auto type = std::make_shared(); + if (json_inference_info && tmp_int < 0) + json_inference_info->negative_integers.insert(type.get()); + return type; + } /// We can safely get back to the start of buffer, because we read from a string and we didn't reach eof. buf.position() = buf.buffer().begin(); @@ -1011,7 +1025,7 @@ namespace { if (settings.json.try_infer_numbers_from_strings) { - if (auto number_type = tryInferNumberFromStringImpl(field, settings)) + if (auto number_type = tryInferNumberFromStringImpl(field, settings, json_info)) { json_info->numbers_parsed_from_json_strings.insert(number_type.get()); return number_type; @@ -1254,10 +1268,23 @@ namespace } /// Number - return tryInferNumber(buf, settings); + return tryInferNumber(buf, settings, json_info); } } +bool checkIfTypesAreEqual(const DataTypes & types) +{ + if (types.empty()) + return true; + + for (size_t i = 1; i < types.size(); ++i) + { + if (!types[0]->equals(*types[i])) + return false; + } + return true; +} + void transformInferredTypesIfNeeded(DataTypePtr & first, DataTypePtr & second, const FormatSettings & settings) { DataTypes types = {first, second}; @@ -1275,6 +1302,11 @@ void transformInferredJSONTypesIfNeeded( second = std::move(types[1]); } +void transformInferredJSONTypesIfNeeded(DataTypes & types, const FormatSettings & settings, JSONInferenceInfo * json_info) +{ + transformInferredTypesIfNeededImpl(types, settings, json_info); +} + void transformInferredJSONTypesFromDifferentFilesIfNeeded(DataTypePtr & first, DataTypePtr & second, const FormatSettings & settings) { JSONInferenceInfo json_info; @@ -1396,6 +1428,12 @@ DataTypePtr tryInferNumberFromString(std::string_view field, const FormatSetting return tryInferNumberFromStringImpl(field, settings); } +DataTypePtr tryInferJSONNumberFromString(std::string_view field, const FormatSettings & settings, JSONInferenceInfo * json_info) +{ + return tryInferNumberFromStringImpl(field, settings, json_info); + +} + DataTypePtr tryInferDateOrDateTimeFromString(std::string_view field, const FormatSettings & settings) { if (settings.try_infer_dates && tryInferDate(field)) diff --git a/src/Formats/SchemaInferenceUtils.h b/src/Formats/SchemaInferenceUtils.h index bcf3d194825..06c14c0797a 100644 --- a/src/Formats/SchemaInferenceUtils.h +++ b/src/Formats/SchemaInferenceUtils.h @@ -2,6 +2,7 @@ #include #include +#include #include @@ -18,6 +19,11 @@ struct JSONInferenceInfo /// We store numbers that were parsed from strings. /// It's used in types transformation to change such numbers back to string if needed. std::unordered_set numbers_parsed_from_json_strings; + /// Store integer types that were inferred from negative numbers. + /// It's used to determine common type for Int64 and UInt64 + /// TODO: check it not only in JSON formats. + std::unordered_set negative_integers; + /// Indicates if currently we are inferring type for Map/Object key. bool is_object_key = false; /// When we transform types for the same column from different files @@ -48,6 +54,7 @@ DataTypePtr tryInferDateOrDateTimeFromString(std::string_view field, const Forma /// Try to parse a number value from a string. By default, it tries to parse Float64, /// but if setting try_infer_integers is enabled, it also tries to parse Int64. DataTypePtr tryInferNumberFromString(std::string_view field, const FormatSettings & settings); +DataTypePtr tryInferJSONNumberFromString(std::string_view field, const FormatSettings & settings, JSONInferenceInfo * json_info); /// It takes two types inferred for the same column and tries to transform them to a common type if possible. /// It's also used when we try to infer some not ordinary types from another types. @@ -77,6 +84,7 @@ void transformInferredTypesIfNeeded(DataTypePtr & first, DataTypePtr & second, c /// Example 2: /// We merge DataTypeJSONPaths types to a single DataTypeJSONPaths type with union of all JSON paths. void transformInferredJSONTypesIfNeeded(DataTypePtr & first, DataTypePtr & second, const FormatSettings & settings, JSONInferenceInfo * json_info); +void transformInferredJSONTypesIfNeeded(DataTypes & types, const FormatSettings & settings, JSONInferenceInfo * json_info); /// Make final transform for types inferred in JSON format. It does 3 types of transformation: /// 1) Checks if type is unnamed Tuple(...), tries to transform nested types to find a common type for them and if all nested types @@ -107,4 +115,6 @@ NamesAndTypesList getNamesAndRecursivelyNullableTypes(const Block & header); /// Check if type contains Nothing, like Array(Tuple(Nullable(Nothing), String)) bool checkIfTypeIsComplete(const DataTypePtr & type); +bool checkIfTypesAreEqual(const DataTypes & types); + } diff --git a/src/Functions/CMakeLists.txt b/src/Functions/CMakeLists.txt index c52b00150ec..7c90f83569a 100644 --- a/src/Functions/CMakeLists.txt +++ b/src/Functions/CMakeLists.txt @@ -3,35 +3,9 @@ add_subdirectory(divide) include("${ClickHouse_SOURCE_DIR}/cmake/dbms_glob_sources.cmake") add_headers_and_sources(clickhouse_functions .) -set(DBMS_FUNCTIONS - IFunction.cpp - FunctionFactory.cpp - FunctionHelpers.cpp - extractTimeZoneFromFunctionArguments.cpp - FunctionsLogical.cpp - if.cpp - multiIf.cpp - multiMatchAny.cpp - checkHyperscanRegexp.cpp - array/has.cpp - CastOverloadResolver.cpp - # Provides dependency for cast - createFunctionBaseCast() - FunctionsConversion.cpp -) -extract_into_parent_list(clickhouse_functions_sources dbms_sources ${DBMS_FUNCTIONS}) -extract_into_parent_list(clickhouse_functions_headers dbms_headers - IFunction.h - FunctionFactory.h - FunctionHelpers.h - extractTimeZoneFromFunctionArguments.h - FunctionsLogical.h - CastOverloadResolver.h -) - add_library(clickhouse_functions_obj OBJECT ${clickhouse_functions_headers} ${clickhouse_functions_sources}) if (OMIT_HEAVY_DEBUG_SYMBOLS) target_compile_options(clickhouse_functions_obj PRIVATE "-g0") - set_source_files_properties(${DBMS_FUNCTIONS} DIRECTORY .. PROPERTIES COMPILE_FLAGS "-g0") endif() list (APPEND OBJECT_LIBS $) diff --git a/src/Functions/CRC.cpp b/src/Functions/CRC.cpp index 49d6dd6fa52..0ba976669a3 100644 --- a/src/Functions/CRC.cpp +++ b/src/Functions/CRC.cpp @@ -150,9 +150,9 @@ using FunctionCRC64ECMA = FunctionCRC; REGISTER_FUNCTION(CRC) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp index 0f54ff52ba2..49f63073aaf 100644 --- a/src/Functions/CastOverloadResolver.cpp +++ b/src/Functions/CastOverloadResolver.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include @@ -136,10 +137,10 @@ FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, st REGISTER_FUNCTION(CastOverloadResolvers) { - factory.registerFunction("_CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, true, {}); }, {}, FunctionFactory::CaseInsensitive); + factory.registerFunction("_CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, true, {}); }, {}, FunctionFactory::Case::Insensitive); /// Note: "internal" (not affected by null preserving setting) versions of accurate cast functions are unneeded. - factory.registerFunction("CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, false, {}); }, {}, FunctionFactory::CaseInsensitive); + factory.registerFunction("CAST", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::nonAccurate, false, {}); }, {}, FunctionFactory::Case::Insensitive); factory.registerFunction("accurateCast", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::accurate, false, {}); }, {}); factory.registerFunction("accurateCastOrNull", [](ContextPtr context){ return CastOverloadResolverImpl::create(context, CastType::accurateOrNull, false, {}); }, {}); } diff --git a/src/Functions/DateTimeTransforms.h b/src/Functions/DateTimeTransforms.h index 34c59ecab08..5f745f3ccad 100644 --- a/src/Functions/DateTimeTransforms.h +++ b/src/Functions/DateTimeTransforms.h @@ -1954,7 +1954,10 @@ struct ToRelativeSubsecondNumImpl return t.value; if (scale > scale_multiplier) return t.value / (scale / scale_multiplier); - return t.value * (scale_multiplier / scale); + return static_cast(t.value) * static_cast((scale_multiplier / scale)); + /// Casting ^^: All integers are Int64, yet if t.value is big enough the multiplication can still + /// overflow which is UB. This place is too low-level and generic to check if t.value is sane. + /// Therefore just let it overflow safely and don't bother further. } static Int64 execute(UInt32 t, const DateLUTImpl &) { diff --git a/src/Functions/FunctionBase58Conversion.h b/src/Functions/FunctionBase58Conversion.h index e519f9768cc..a34a36d8b81 100644 --- a/src/Functions/FunctionBase58Conversion.h +++ b/src/Functions/FunctionBase58Conversion.h @@ -18,6 +18,7 @@ namespace ErrorCodes extern const int ILLEGAL_COLUMN; extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int BAD_ARGUMENTS; + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; } struct Base58Encode @@ -135,7 +136,7 @@ public: DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { if (arguments.size() != 1) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Wrong number of arguments for function {}: 1 expected.", getName()); + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Wrong number of arguments for function {}: 1 expected.", getName()); if (!isString(arguments[0].type)) throw Exception( diff --git a/src/Functions/FunctionBase64Conversion.h b/src/Functions/FunctionBase64Conversion.h index 05914be3837..363b9ee3a31 100644 --- a/src/Functions/FunctionBase64Conversion.h +++ b/src/Functions/FunctionBase64Conversion.h @@ -25,10 +25,10 @@ namespace ErrorCodes enum class Base64Variant : uint8_t { Normal, - Url + URL }; -inline std::string preprocessBase64Url(std::string_view src) +inline std::string preprocessBase64URL(std::string_view src) { std::string padded_src; padded_src.reserve(src.size() + 3); @@ -70,7 +70,7 @@ inline std::string preprocessBase64Url(std::string_view src) return padded_src; } -inline size_t postprocessBase64Url(UInt8 * dst, size_t out_len) +inline size_t postprocessBase64URL(UInt8 * dst, size_t out_len) { // Do symbol substitution as described in https://datatracker.ietf.org/doc/html/rfc4648#section-5 for (size_t i = 0; i < out_len; ++i) @@ -95,7 +95,7 @@ inline size_t postprocessBase64Url(UInt8 * dst, size_t out_len) template struct Base64Encode { - static constexpr auto name = (variant == Base64Variant::Normal) ? "base64Encode" : "base64UrlEncode"; + static constexpr auto name = (variant == Base64Variant::Normal) ? "base64Encode" : "base64URLEncode"; static size_t getBufferSize(size_t string_length, size_t string_count) { @@ -111,8 +111,8 @@ struct Base64Encode /// Memory sanitizer doesn't understand if there was uninitialized memory in SIMD register but it was not used in the result of shuffle. __msan_unpoison(dst, outlen); - if constexpr (variant == Base64Variant::Url) - outlen = postprocessBase64Url(dst, outlen); + if constexpr (variant == Base64Variant::URL) + outlen = postprocessBase64URL(dst, outlen); return outlen; } @@ -121,7 +121,7 @@ struct Base64Encode template struct Base64Decode { - static constexpr auto name = (variant == Base64Variant::Normal) ? "base64Decode" : "base64UrlDecode"; + static constexpr auto name = (variant == Base64Variant::Normal) ? "base64Decode" : "base64URLDecode"; static size_t getBufferSize(size_t string_length, size_t string_count) { @@ -132,9 +132,9 @@ struct Base64Decode { int rc; size_t outlen = 0; - if constexpr (variant == Base64Variant::Url) + if constexpr (variant == Base64Variant::URL) { - std::string src_padded = preprocessBase64Url(src); + std::string src_padded = preprocessBase64URL(src); rc = base64_decode(src_padded.data(), src_padded.size(), reinterpret_cast(dst), &outlen, 0); } else @@ -156,7 +156,7 @@ struct Base64Decode template struct TryBase64Decode { - static constexpr auto name = (variant == Base64Variant::Normal) ? "tryBase64Decode" : "tryBase64UrlDecode"; + static constexpr auto name = (variant == Base64Variant::Normal) ? "tryBase64Decode" : "tryBase64URLDecode"; static size_t getBufferSize(size_t string_length, size_t string_count) { @@ -167,9 +167,9 @@ struct TryBase64Decode { int rc; size_t outlen = 0; - if constexpr (variant == Base64Variant::Url) + if constexpr (variant == Base64Variant::URL) { - std::string src_padded = preprocessBase64Url(src); + std::string src_padded = preprocessBase64URL(src); rc = base64_decode(src_padded.data(), src_padded.size(), reinterpret_cast(dst), &outlen, 0); } else @@ -202,7 +202,7 @@ public: {"value", static_cast(&isStringOrFixedString), nullptr, "String or FixedString"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_arguments); + validateFunctionArguments(*this, arguments, mandatory_arguments); return std::make_shared(); } diff --git a/src/Functions/FunctionBinaryArithmetic.h b/src/Functions/FunctionBinaryArithmetic.h index 5d19ba44d9b..ee4e28a7858 100644 --- a/src/Functions/FunctionBinaryArithmetic.h +++ b/src/Functions/FunctionBinaryArithmetic.h @@ -1703,7 +1703,7 @@ public: { if constexpr (is_division) { - if (context->getSettingsRef().decimal_check_overflow) + if (decimalCheckArithmeticOverflow(context)) { /// Check overflow by using operands scale (based on big decimal division implementation details): /// big decimal arithmetic is based on big integers, decimal operands are converted to big integers diff --git a/src/Functions/FunctionBitTestMany.h b/src/Functions/FunctionBitTestMany.h index 71e94b1e71d..950e4ab4ea8 100644 --- a/src/Functions/FunctionBitTestMany.h +++ b/src/Functions/FunctionBitTestMany.h @@ -16,6 +16,7 @@ namespace ErrorCodes { extern const int ILLEGAL_COLUMN; extern const int ILLEGAL_TYPE_OF_ARGUMENT; + extern const int PARAMETER_OUT_OF_BOUND; extern const int TOO_FEW_ARGUMENTS_FOR_FUNCTION; } @@ -146,6 +147,9 @@ private: const auto pos = pos_col_const->getUInt(0); if (pos < 8 * sizeof(ValueType)) mask = mask | (ValueType(1) << pos); + else + throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND, + "The bit position argument {} is out of bounds for number", static_cast(pos)); } else { @@ -186,13 +190,20 @@ private: for (const auto i : collections::range(0, mask.size())) if (pos[i] < 8 * sizeof(ValueType)) mask[i] = mask[i] | (ValueType(1) << pos[i]); + else + throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND, + "The bit position argument {} is out of bounds for number", static_cast(pos[i])); return true; } else if (const auto pos_col_const = checkAndGetColumnConst>(pos_col_untyped)) { const auto & pos = pos_col_const->template getValue(); - const auto new_mask = pos < 8 * sizeof(ValueType) ? ValueType(1) << pos : 0; + if (pos >= 8 * sizeof(ValueType)) + throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND, + "The bit position argument {} is out of bounds for number", static_cast(pos)); + + const auto new_mask = ValueType(1) << pos; for (const auto i : collections::range(0, mask.size())) mask[i] = mask[i] | new_mask; diff --git a/src/Functions/FunctionChar.cpp b/src/Functions/FunctionChar.cpp index 055eb08f0c7..79e346a3ea4 100644 --- a/src/Functions/FunctionChar.cpp +++ b/src/Functions/FunctionChar.cpp @@ -15,6 +15,7 @@ namespace ErrorCodes { extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int ILLEGAL_COLUMN; + extern const int TOO_FEW_ARGUMENTS_FOR_FUNCTION; } class FunctionChar : public IFunction @@ -36,7 +37,7 @@ public: DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override { if (arguments.empty()) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Number of arguments for function {} can't be {}, should be at least 1", getName(), arguments.size()); @@ -116,7 +117,7 @@ private: REGISTER_FUNCTION(Char) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/FunctionCustomWeekToDateOrDate32.h b/src/Functions/FunctionCustomWeekToDateOrDate32.h index c904bfaec30..6ae8c847b65 100644 --- a/src/Functions/FunctionCustomWeekToDateOrDate32.h +++ b/src/Functions/FunctionCustomWeekToDateOrDate32.h @@ -1,4 +1,5 @@ #pragma once +#include #include namespace DB diff --git a/src/Functions/FunctionDateOrDateTimeToDateOrDate32.h b/src/Functions/FunctionDateOrDateTimeToDateOrDate32.h index 6eb3e534b62..28caef36879 100644 --- a/src/Functions/FunctionDateOrDateTimeToDateOrDate32.h +++ b/src/Functions/FunctionDateOrDateTimeToDateOrDate32.h @@ -1,4 +1,5 @@ #pragma once +#include #include namespace DB diff --git a/src/Functions/FunctionDateOrDateTimeToDateTimeOrDateTime64.h b/src/Functions/FunctionDateOrDateTimeToDateTimeOrDateTime64.h index 9f1066fd687..d5a1f28b7cd 100644 --- a/src/Functions/FunctionDateOrDateTimeToDateTimeOrDateTime64.h +++ b/src/Functions/FunctionDateOrDateTimeToDateTimeOrDateTime64.h @@ -1,4 +1,6 @@ #pragma once + +#include #include namespace DB diff --git a/src/Functions/FunctionFQDN.cpp b/src/Functions/FunctionFQDN.cpp index 108a96216fd..8948c948265 100644 --- a/src/Functions/FunctionFQDN.cpp +++ b/src/Functions/FunctionFQDN.cpp @@ -46,7 +46,7 @@ public: REGISTER_FUNCTION(FQDN) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); factory.registerAlias("fullHostName", "FQDN"); } diff --git a/src/Functions/FunctionFactory.cpp b/src/Functions/FunctionFactory.cpp index 004ef745a93..501cf6e725c 100644 --- a/src/Functions/FunctionFactory.cpp +++ b/src/Functions/FunctionFactory.cpp @@ -4,6 +4,7 @@ #include #include +#include #include @@ -30,7 +31,7 @@ void FunctionFactory::registerFunction( const std::string & name, FunctionCreator creator, FunctionDocumentation doc, - CaseSensitiveness case_sensitiveness) + Case case_sensitiveness) { if (!functions.emplace(name, FunctionFactoryData{creator, doc}).second) throw Exception(ErrorCodes::LOGICAL_ERROR, "FunctionFactory: the function name '{}' is not unique", name); @@ -40,7 +41,7 @@ void FunctionFactory::registerFunction( throw Exception(ErrorCodes::LOGICAL_ERROR, "FunctionFactory: the function name '{}' is already registered as alias", name); - if (case_sensitiveness == CaseInsensitive) + if (case_sensitiveness == Case::Insensitive) { if (!case_insensitive_functions.emplace(function_name_lowercase, FunctionFactoryData{creator, doc}).second) throw Exception(ErrorCodes::LOGICAL_ERROR, "FunctionFactory: the case insensitive function name '{}' is not unique", @@ -53,7 +54,7 @@ void FunctionFactory::registerFunction( const std::string & name, FunctionSimpleCreator creator, FunctionDocumentation doc, - CaseSensitiveness case_sensitiveness) + Case case_sensitiveness) { registerFunction(name, [my_creator = std::move(creator)](ContextPtr context) { diff --git a/src/Functions/FunctionFactory.h b/src/Functions/FunctionFactory.h index bb43d4719b8..d05e84439be 100644 --- a/src/Functions/FunctionFactory.h +++ b/src/Functions/FunctionFactory.h @@ -30,7 +30,7 @@ public: static FunctionFactory & instance(); template - void registerFunction(FunctionDocumentation doc = {}, CaseSensitiveness case_sensitiveness = CaseSensitive) + void registerFunction(FunctionDocumentation doc = {}, Case case_sensitiveness = Case::Sensitive) { registerFunction(Function::name, std::move(doc), case_sensitiveness); } @@ -56,13 +56,13 @@ public: const std::string & name, FunctionCreator creator, FunctionDocumentation doc = {}, - CaseSensitiveness case_sensitiveness = CaseSensitive); + Case case_sensitiveness = Case::Sensitive); void registerFunction( const std::string & name, FunctionSimpleCreator creator, FunctionDocumentation doc = {}, - CaseSensitiveness case_sensitiveness = CaseSensitive); + Case case_sensitiveness = Case::Sensitive); FunctionDocumentation getDocumentation(const std::string & name) const; @@ -79,7 +79,7 @@ private: String getFactoryName() const override { return "FunctionFactory"; } template - void registerFunction(const std::string & name, FunctionDocumentation doc = {}, CaseSensitiveness case_sensitiveness = CaseSensitive) + void registerFunction(const std::string & name, FunctionDocumentation doc = {}, Case case_sensitiveness = Case::Sensitive) { registerFunction(name, &Function::create, std::move(doc), case_sensitiveness); } diff --git a/src/Functions/FunctionGenerateRandomStructure.cpp b/src/Functions/FunctionGenerateRandomStructure.cpp index 6dc68134502..2bead8737fd 100644 --- a/src/Functions/FunctionGenerateRandomStructure.cpp +++ b/src/Functions/FunctionGenerateRandomStructure.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -354,6 +355,11 @@ namespace } } + FunctionPtr FunctionGenerateRandomStructure::create(DB::ContextPtr context) + { + return std::make_shared(context->getSettingsRef().allow_suspicious_low_cardinality_types.value); + } + DataTypePtr FunctionGenerateRandomStructure::getReturnTypeImpl(const DataTypes & arguments) const { if (arguments.size() > 2) @@ -439,8 +445,7 @@ The function returns a value of type String. {"with specified seed", "SELECT generateRandomStructure(1, 42)", "c1 UInt128"}, }, .categories{"Random"} - }, - FunctionFactory::CaseSensitive); + }); } } diff --git a/src/Functions/FunctionGenerateRandomStructure.h b/src/Functions/FunctionGenerateRandomStructure.h index 894096a6e07..5901565696f 100644 --- a/src/Functions/FunctionGenerateRandomStructure.h +++ b/src/Functions/FunctionGenerateRandomStructure.h @@ -17,10 +17,7 @@ public: { } - static FunctionPtr create(ContextPtr context) - { - return std::make_shared(context->getSettingsRef().allow_suspicious_low_cardinality_types.value); - } + static FunctionPtr create(ContextPtr context); String getName() const override { return name; } diff --git a/src/Functions/FunctionHelpers.cpp b/src/Functions/FunctionHelpers.cpp index 593646240ca..c658063b66f 100644 --- a/src/Functions/FunctionHelpers.cpp +++ b/src/Functions/FunctionHelpers.cpp @@ -95,22 +95,21 @@ ColumnsWithTypeAndName createBlockWithNestedColumns(const ColumnsWithTypeAndName return res; } -void validateArgumentType(const IFunction & func, const DataTypes & arguments, - size_t argument_index, bool (* validator_func)(const IDataType &), - const char * expected_type_description) -{ - if (arguments.size() <= argument_index) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Incorrect number of arguments of function {}", - func.getName()); - - const auto & argument = arguments[argument_index]; - if (!validator_func(*argument)) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of {} argument of function {}, expected {}", - argument->getName(), std::to_string(argument_index), func.getName(), expected_type_description); -} - namespace { + +String withOrdinalEnding(size_t i) +{ + switch (i) + { + case 0: return "1st"; + case 1: return "2nd"; + case 2: return "3rd"; + default: return std::to_string(i) + "th"; + } + +} + void validateArgumentsImpl(const IFunction & func, const ColumnsWithTypeAndName & arguments, size_t argument_offset, @@ -120,20 +119,18 @@ void validateArgumentsImpl(const IFunction & func, { const auto argument_index = i + argument_offset; if (argument_index >= arguments.size()) - { break; - } const auto & arg = arguments[i + argument_offset]; const auto & descriptor = descriptors[i]; if (int error_code = descriptor.isValid(arg.type, arg.column); error_code != 0) throw Exception(error_code, - "Illegal type of argument #{}{} of function {}{}{}", - argument_offset + i + 1, // +1 is for human-friendly 1-based indexing - (descriptor.argument_name ? " '" + std::string(descriptor.argument_name) + "'" : String{}), + "A value of illegal type was provided as {} argument '{}' to function '{}'. Expected: {}, got: {}", + withOrdinalEnding(argument_offset + i), + descriptor.name, func.getName(), - (descriptor.expected_type_description ? String(", expected ") + descriptor.expected_type_description : String{}), - (arg.type ? ", got " + arg.type->getName() : String{})); + descriptor.type_name, + arg.type ? arg.type->getName() : ""); } } @@ -141,52 +138,42 @@ void validateArgumentsImpl(const IFunction & func, int FunctionArgumentDescriptor::isValid(const DataTypePtr & data_type, const ColumnPtr & column) const { - if (type_validator_func && (data_type == nullptr || !type_validator_func(*data_type))) + if (name.empty() || type_name.empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "name or type_name are not set"); + + if (type_validator && (data_type == nullptr || !type_validator(*data_type))) return ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT; - if (column_validator_func && (column == nullptr || !column_validator_func(*column))) + if (column_validator && (column == nullptr || !column_validator(*column))) return ErrorCodes::ILLEGAL_COLUMN; return 0; } -void validateFunctionArgumentTypes(const IFunction & func, - const ColumnsWithTypeAndName & arguments, - const FunctionArgumentDescriptors & mandatory_args, - const FunctionArgumentDescriptors & optional_args) +void validateFunctionArguments(const IFunction & func, + const ColumnsWithTypeAndName & arguments, + const FunctionArgumentDescriptors & mandatory_args, + const FunctionArgumentDescriptors & optional_args) { if (arguments.size() < mandatory_args.size() || arguments.size() > mandatory_args.size() + optional_args.size()) { - auto join_argument_types = [](const auto & args, const String sep = ", ") - { - String result; - for (const auto & a : args) - { - using A = std::decay_t; - if constexpr (std::is_same_v) - { - if (a.argument_name) - result += "'" + std::string(a.argument_name) + "' : "; - if (a.expected_type_description) - result += a.expected_type_description; - } - else if constexpr (std::is_same_v) - result += a.type->getName(); + auto argument_singular_or_plural = [](const auto & args) -> std::string_view { return args.size() == 1 ? "argument" : "arguments"; }; - result += sep; - } - - if (!args.empty()) - result.erase(result.end() - sep.length(), result.end()); - - return result; - }; + String expected_args_string; + if (!mandatory_args.empty() && !optional_args.empty()) + expected_args_string = fmt::format("{} mandatory {} and {} optional {}", mandatory_args.size(), argument_singular_or_plural(mandatory_args), optional_args.size(), argument_singular_or_plural(optional_args)); + else if (!mandatory_args.empty() && optional_args.empty()) + expected_args_string = fmt::format("{} {}", mandatory_args.size(), argument_singular_or_plural(mandatory_args)); /// intentionally not "_mandatory_ arguments" + else if (mandatory_args.empty() && !optional_args.empty()) + expected_args_string = fmt::format("{} optional {}", optional_args.size(), argument_singular_or_plural(optional_args)); + else + expected_args_string = "0 arguments"; throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, - "Incorrect number of arguments for function {} provided {}{}, expected {}{} ({}{})", - func.getName(), arguments.size(), (!arguments.empty() ? " (" + join_argument_types(arguments) + ")" : String{}), - mandatory_args.size(), (!optional_args.empty() ? " to " + std::to_string(mandatory_args.size() + optional_args.size()) : ""), - join_argument_types(mandatory_args), (!optional_args.empty() ? ", [" + join_argument_types(optional_args) + "]" : "")); + "An incorrect number of arguments was specified for function '{}'. Expected {}, got {}", + func.getName(), + expected_args_string, + fmt::format("{} {}", arguments.size(), argument_singular_or_plural(arguments))); } validateArgumentsImpl(func, arguments, 0, mandatory_args); diff --git a/src/Functions/FunctionHelpers.h b/src/Functions/FunctionHelpers.h index 6267d8eacc4..4f93b236bcb 100644 --- a/src/Functions/FunctionHelpers.h +++ b/src/Functions/FunctionHelpers.h @@ -115,77 +115,58 @@ ColumnWithTypeAndName columnGetNested(const ColumnWithTypeAndName & col); /// column if it is nullable. ColumnsWithTypeAndName createBlockWithNestedColumns(const ColumnsWithTypeAndName & columns); -/// Checks argument type at specified index with predicate. -/// throws if there is no argument at specified index or if predicate returns false. -void validateArgumentType(const IFunction & func, const DataTypes & arguments, - size_t argument_index, bool (* validator_func)(const IDataType &), - const char * expected_type_description); - -/** Simple validator that is used in conjunction with validateFunctionArgumentTypes() to check if function arguments are as expected - * - * Also it is used to generate function description when arguments do not match expected ones. - * Any field can be null: - * `argument_name` - if not null, reported via type check errors. - * `expected_type_description` - if not null, reported via type check errors. - * `type_validator_func` - if not null, used to validate data type of function argument. - * `column_validator_func` - if not null, used to validate column of function argument. - */ +/// Expected arguments for a function. Can be used in conjunction with validateFunctionArguments() to check that the user-provided +/// arguments match the expected arguments. struct FunctionArgumentDescriptor { - const char * argument_name; + /// The argument name, e.g. "longitude". + /// Should not be empty. + std::string_view name; + /// A function which validates the argument data type. + /// May be nullptr. using TypeValidator = bool (*)(const IDataType &); - TypeValidator type_validator_func; + TypeValidator type_validator; + + /// A function which validates the argument column. + /// May be nullptr. using ColumnValidator = bool (*)(const IColumn &); - ColumnValidator column_validator_func; + ColumnValidator column_validator; - const char * expected_type_description; + /// The expected argument type, e.g. "const String" or "UInt64". + /// Should not be empty. + std::string_view type_name; - /** Validate argument type and column. - * - * Returns non-zero error code if: - * Validator != nullptr && (Value == nullptr || Validator(*Value) == false) - * For: - * Validator is either `type_validator_func` or `column_validator_func` - * Value is either `data_type` or `column` respectively. - * ILLEGAL_TYPE_OF_ARGUMENT if type validation fails - * - */ + /// Validate argument type and column. int isValid(const DataTypePtr & data_type, const ColumnPtr & column) const; }; using FunctionArgumentDescriptors = std::vector; -/** Validate that function arguments match specification. - * - * Designed to simplify argument validation for functions with variable arguments - * (e.g. depending on result type or other trait). - * First, checks that number of arguments is as expected (including optional arguments). - * Second, checks that mandatory args present and have valid type. - * Third, checks optional arguments types, skipping ones that are missing. - * - * Please note that if you have several optional arguments, like f([a, b, c]), - * only these calls are considered valid: - * f(a) - * f(a, b) - * f(a, b, c) - * - * But NOT these: f(a, c), f(b, c) - * In other words you can't omit middle optional arguments (just like in regular C++). - * - * If any mandatory arg is missing, throw an exception, with explicit description of expected arguments. - */ -void validateFunctionArgumentTypes(const IFunction & func, const ColumnsWithTypeAndName & arguments, - const FunctionArgumentDescriptors & mandatory_args, - const FunctionArgumentDescriptors & optional_args = {}); +/// Validates that the user-provided arguments match the expected arguments. +/// +/// Checks that +/// - the number of provided arguments matches the number of mandatory/optional arguments, +/// - all mandatory arguments are present and have the right type, +/// - optional arguments - if present - have the right type. +/// +/// With multiple optional arguments, e.g. f([a, b, c]), provided arguments must match left-to-right. E.g. these calls are considered valid: +/// f(a) +/// f(a, b) +/// f(a, b, c) +/// but these are NOT: +/// f(a, c) +/// f(b, c) +void validateFunctionArguments(const IFunction & func, const ColumnsWithTypeAndName & arguments, + const FunctionArgumentDescriptors & mandatory_args, + const FunctionArgumentDescriptors & optional_args = {}); /// Checks if a list of array columns have equal offsets. Return a pair of nested columns and offsets if true, otherwise throw. std::pair, const ColumnArray::Offset *> checkAndGetNestedArrayOffset(const IColumn ** columns, size_t num_arguments); -/** Return ColumnNullable of src, with null map as OR-ed null maps of args columns. - * Or ColumnConst(ColumnNullable) if the result is always NULL or if the result is constant and always not NULL. - */ +/// Return ColumnNullable of src, with null map as OR-ed null maps of args columns. +/// Or ColumnConst(ColumnNullable) if the result is always NULL or if the result is constant and always not NULL. ColumnPtr wrapInNullable(const ColumnPtr & src, const ColumnsWithTypeAndName & args, const DataTypePtr & result_type, size_t input_rows_count); struct NullPresence diff --git a/src/Functions/FunctionJoinGet.cpp b/src/Functions/FunctionJoinGet.cpp index 085c4db3f57..a67f3a977db 100644 --- a/src/Functions/FunctionJoinGet.cpp +++ b/src/Functions/FunctionJoinGet.cpp @@ -1,11 +1,12 @@ #include #include +#include #include #include #include #include #include -#include +#include #include #include diff --git a/src/Functions/FunctionStringReplace.h b/src/Functions/FunctionStringReplace.h index aee04a5969a..b4bcfa514a8 100644 --- a/src/Functions/FunctionStringReplace.h +++ b/src/Functions/FunctionStringReplace.h @@ -40,7 +40,7 @@ public: {"replacement", static_cast(&isString), nullptr, "String"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } diff --git a/src/Functions/FunctionStringToString.h b/src/Functions/FunctionStringToString.h index 62d7b09f79e..e0e64e47b49 100644 --- a/src/Functions/FunctionStringToString.h +++ b/src/Functions/FunctionStringToString.h @@ -59,19 +59,19 @@ public: bool useDefaultImplementationForConstants() const override { return true; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const ColumnPtr column = arguments[0].column; if (const ColumnString * col = checkAndGetColumn(column.get())) { auto col_res = ColumnString::create(); - Impl::vector(col->getChars(), col->getOffsets(), col_res->getChars(), col_res->getOffsets()); + Impl::vector(col->getChars(), col->getOffsets(), col_res->getChars(), col_res->getOffsets(), input_rows_count); return col_res; } else if (const ColumnFixedString * col_fixed = checkAndGetColumn(column.get())) { auto col_res = ColumnFixedString::create(col_fixed->getN()); - Impl::vectorFixed(col_fixed->getChars(), col_fixed->getN(), col_res->getChars()); + Impl::vectorFixed(col_fixed->getChars(), col_fixed->getN(), col_res->getChars(), input_rows_count); return col_res; } else diff --git a/src/Functions/FunctionTokens.h b/src/Functions/FunctionTokens.h index d6cf6a24983..b6d8e9ee589 100644 --- a/src/Functions/FunctionTokens.h +++ b/src/Functions/FunctionTokens.h @@ -17,6 +17,7 @@ #include #include #include +#include namespace DB @@ -83,7 +84,7 @@ public: return std::make_shared(std::make_shared()); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override { Generator generator; generator.init(arguments, max_substrings_includes_remaining_string); @@ -106,18 +107,17 @@ public: const ColumnString::Chars & src_chars = col_str->getChars(); const ColumnString::Offsets & src_offsets = col_str->getOffsets(); - res_offsets.reserve(src_offsets.size()); - res_strings_offsets.reserve(src_offsets.size() * 5); /// Constant 5 - at random. + res_offsets.reserve(input_rows_count); + res_strings_offsets.reserve(input_rows_count * 5); /// Constant 5 - at random. res_strings_chars.reserve(src_chars.size()); Pos token_begin = nullptr; Pos token_end = nullptr; - size_t size = src_offsets.size(); ColumnString::Offset current_src_offset = 0; ColumnArray::Offset current_dst_offset = 0; ColumnString::Offset current_dst_strings_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { Pos pos = reinterpret_cast(&src_chars[current_src_offset]); current_src_offset = src_offsets[i]; @@ -194,7 +194,7 @@ static inline void checkArgumentsWithSeparatorAndOptionalMaxSubstrings( {"max_substrings", static_cast(&isNativeInteger), isColumnConst, "const Number"}, }; - validateFunctionArgumentTypes(func, arguments, mandatory_args, optional_args); + validateFunctionArguments(func, arguments, mandatory_args, optional_args); } static inline void checkArgumentsWithOptionalMaxSubstrings(const IFunction & func, const ColumnsWithTypeAndName & arguments) @@ -207,7 +207,7 @@ static inline void checkArgumentsWithOptionalMaxSubstrings(const IFunction & fun {"max_substrings", static_cast(&isNativeInteger), isColumnConst, "const Number"}, }; - validateFunctionArgumentTypes(func, arguments, mandatory_args, optional_args); + validateFunctionArguments(func, arguments, mandatory_args, optional_args); } } diff --git a/src/Functions/FunctionUnixTimestamp64.cpp b/src/Functions/FunctionUnixTimestamp64.cpp new file mode 100644 index 00000000000..a22c035a0ba --- /dev/null +++ b/src/Functions/FunctionUnixTimestamp64.cpp @@ -0,0 +1,13 @@ +#include +#include + +namespace DB +{ + +FunctionFromUnixTimestamp64::FunctionFromUnixTimestamp64(size_t target_scale_, const char * name_, ContextPtr context) + : target_scale(target_scale_) + , name(name_) + , allow_nonconst_timezone_arguments(context->getSettingsRef().allow_nonconst_timezone_arguments) +{} + +} diff --git a/src/Functions/FunctionUnixTimestamp64.h b/src/Functions/FunctionUnixTimestamp64.h index c418163343b..bbd7e586a67 100644 --- a/src/Functions/FunctionUnixTimestamp64.h +++ b/src/Functions/FunctionUnixTimestamp64.h @@ -47,7 +47,7 @@ public: FunctionArgumentDescriptors args{ {"value", static_cast(&isDateTime64), nullptr, "DateTime64"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } @@ -107,11 +107,7 @@ private: const bool allow_nonconst_timezone_arguments; public: - FunctionFromUnixTimestamp64(size_t target_scale_, const char * name_, ContextPtr context) - : target_scale(target_scale_) - , name(name_) - , allow_nonconst_timezone_arguments(context->getSettings().allow_nonconst_timezone_arguments) - {} + FunctionFromUnixTimestamp64(size_t target_scale_, const char * name_, ContextPtr context); String getName() const override { return name; } size_t getNumberOfArguments() const override { return 0; } diff --git a/src/Functions/FunctionsAES.h b/src/Functions/FunctionsAES.h index 524b4f82acd..7af6265eba9 100644 --- a/src/Functions/FunctionsAES.h +++ b/src/Functions/FunctionsAES.h @@ -165,7 +165,7 @@ private: }); } - validateFunctionArgumentTypes(*this, arguments, + validateFunctionArguments(*this, arguments, FunctionArgumentDescriptors{ {"mode", static_cast(&isStringOrFixedString), isColumnConst, "encryption mode string"}, {"input", static_cast(&isStringOrFixedString), {}, "plaintext"}, @@ -438,7 +438,7 @@ private: }); } - validateFunctionArgumentTypes(*this, arguments, + validateFunctionArguments(*this, arguments, FunctionArgumentDescriptors{ {"mode", static_cast(&isStringOrFixedString), isColumnConst, "decryption mode string"}, {"input", static_cast(&isStringOrFixedString), {}, "ciphertext"}, diff --git a/src/Functions/FunctionsBinaryRepresentation.cpp b/src/Functions/FunctionsBinaryRepresentation.cpp index 0f3f8be96a7..f77d2f1f350 100644 --- a/src/Functions/FunctionsBinaryRepresentation.cpp +++ b/src/Functions/FunctionsBinaryRepresentation.cpp @@ -3,14 +3,14 @@ #include #include #include -#include -#include #include #include #include #include #include #include +#include +#include namespace DB { @@ -218,10 +218,7 @@ struct UnbinImpl static constexpr auto name = "unbin"; static constexpr size_t word_size = 8; - static void decode(const char * pos, const char * end, char *& out) - { - binStringDecode(pos, end, out); - } + static void decode(const char * pos, const char * end, char *& out) { binStringDecode(pos, end, out, word_size); } }; /// Encode number or string to string with binary or hexadecimal representation @@ -651,7 +648,15 @@ public: size_t size = in_offsets.size(); out_offsets.resize(size); - out_vec.resize(in_vec.size() / word_size + size); + + size_t max_out_len = 0; + for (size_t i = 0; i < in_offsets.size(); ++i) + { + const size_t len = in_offsets[i] - (i == 0 ? 0 : in_offsets[i - 1]) + - /* trailing zero symbol that is always added in ColumnString and that is ignored while decoding */ 1; + max_out_len += (len + word_size - 1) / word_size + /* trailing zero symbol that is always added by Impl::decode */ 1; + } + out_vec.resize(max_out_len); char * begin = reinterpret_cast(out_vec.data()); char * pos = begin; @@ -661,6 +666,7 @@ public: { size_t new_offset = in_offsets[i]; + /// `new_offset - 1` because in ColumnString each string is stored with trailing zero byte Impl::decode(reinterpret_cast(&in_vec[prev_offset]), reinterpret_cast(&in_vec[new_offset - 1]), pos); out_offsets[i] = pos - begin; @@ -668,6 +674,9 @@ public: prev_offset = new_offset; } + chassert( + static_cast(pos - begin) <= out_vec.size(), + fmt::format("too small amount of memory was preallocated: needed {}, but have only {}", pos - begin, out_vec.size())); out_vec.resize(pos - begin); return col_res; @@ -680,11 +689,11 @@ public: ColumnString::Offsets & out_offsets = col_res->getOffsets(); const ColumnString::Chars & in_vec = col_fix_string->getChars(); - size_t n = col_fix_string->getN(); + const size_t n = col_fix_string->getN(); size_t size = col_fix_string->size(); out_offsets.resize(size); - out_vec.resize(in_vec.size() / word_size + size); + out_vec.resize(((n + word_size - 1) / word_size + /* trailing zero symbol that is always added by Impl::decode */ 1) * size); char * begin = reinterpret_cast(out_vec.data()); char * pos = begin; @@ -694,6 +703,7 @@ public: { size_t new_offset = prev_offset + n; + /// here we don't subtract 1 from `new_offset` because in ColumnFixedString strings are stored without trailing zero byte Impl::decode(reinterpret_cast(&in_vec[prev_offset]), reinterpret_cast(&in_vec[new_offset]), pos); out_offsets[i] = pos - begin; @@ -701,6 +711,9 @@ public: prev_offset = new_offset; } + chassert( + static_cast(pos - begin) <= out_vec.size(), + fmt::format("too small amount of memory was preallocated: needed {}, but have only {}", pos - begin, out_vec.size())); out_vec.resize(pos - begin); return col_res; @@ -715,10 +728,10 @@ public: REGISTER_FUNCTION(BinaryRepr) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/FunctionsBitToArray.cpp b/src/Functions/FunctionsBitToArray.cpp index adabda1a7f8..beaaccad6db 100644 --- a/src/Functions/FunctionsBitToArray.cpp +++ b/src/Functions/FunctionsBitToArray.cpp @@ -284,7 +284,12 @@ public: { while (x) { - result_array_values_data.push_back(std::countr_zero(x)); + /// С++20 char8_t is not an unsigned integral type anymore https://godbolt.org/z/Mqcb7qn58 + /// and thus you cannot use std::countr_zero on it. + if constexpr (std::is_same_v) + result_array_values_data.push_back(std::countr_zero(static_cast(x))); + else + result_array_values_data.push_back(std::countr_zero(x)); x &= (x - 1); } } @@ -336,4 +341,3 @@ REGISTER_FUNCTION(BitToArray) } } - diff --git a/src/Functions/FunctionsCodingIP.cpp b/src/Functions/FunctionsCodingIP.cpp index e01967274f4..0a97d029f84 100644 --- a/src/Functions/FunctionsCodingIP.cpp +++ b/src/Functions/FunctionsCodingIP.cpp @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -1168,10 +1169,10 @@ REGISTER_FUNCTION(Coding) factory.registerFunction>(); /// MySQL compatibility aliases: - factory.registerAlias("INET_ATON", FunctionIPv4StringToNum::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("INET6_NTOA", FunctionIPv6NumToString::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("INET6_ATON", FunctionIPv6StringToNum::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("INET_NTOA", NameFunctionIPv4NumToString::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("INET_ATON", FunctionIPv4StringToNum::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("INET6_NTOA", FunctionIPv6NumToString::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("INET6_ATON", FunctionIPv6StringToNum::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("INET_NTOA", NameFunctionIPv4NumToString::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/FunctionsCodingULID.cpp b/src/Functions/FunctionsCodingULID.cpp index ff040945a15..b67224a5625 100644 --- a/src/Functions/FunctionsCodingULID.cpp +++ b/src/Functions/FunctionsCodingULID.cpp @@ -180,8 +180,7 @@ An optional second argument can be passed to specify a timezone for the timestam {"ulid", "SELECT ULIDStringToDateTime(generateULID())", ""}, {"timezone", "SELECT ULIDStringToDateTime(generateULID(), 'Asia/Istanbul')", ""}}, .categories{"ULID"} - }, - FunctionFactory::CaseSensitive); + }); } } diff --git a/src/Functions/FunctionsCodingUUID.cpp b/src/Functions/FunctionsCodingUUID.cpp index 6a44f4263a8..83fdcbc4af9 100644 --- a/src/Functions/FunctionsCodingUUID.cpp +++ b/src/Functions/FunctionsCodingUUID.cpp @@ -496,8 +496,8 @@ This function accepts a UUID and returns a FixedString(16) as its binary represe │ 612f3c40-5d3b-217e-707b-6a546a3d7b29 │ a/<@];!~p{jTj={) │ @( FunctionDocumentation{ @@ -509,8 +509,7 @@ An optional second argument can be passed to specify a timezone for the timestam .examples{ {"uuid","select UUIDv7ToDateTime(generateUUIDv7())", ""}, {"uuid","select generateUUIDv7() as uuid, UUIDv7ToDateTime(uuid), UUIDv7ToDateTime(uuid, 'America/New_York')", ""}}, - .categories{"UUID"}}, - FunctionFactory::CaseSensitive); + .categories{"UUID"}}); } } diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 2a0b2f1d075..675283d011e 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -2020,7 +2021,7 @@ public: DataTypePtr getReturnTypeImplRemovedNullable(const ColumnsWithTypeAndName & arguments) const { - FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, nullptr}}; + FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, "any type"}}; FunctionArgumentDescriptors optional_args; if constexpr (to_decimal) @@ -2049,7 +2050,7 @@ public: optional_args.push_back({"timezone", static_cast(&isString), nullptr, "String"}); } - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); if constexpr (std::is_same_v) { @@ -2390,7 +2391,7 @@ public: if (isDateTime64(arguments)) { - validateFunctionArgumentTypes(*this, arguments, + validateFunctionArguments(*this, arguments, FunctionArgumentDescriptors{{"string", static_cast(&isStringOrFixedString), nullptr, "String or FixedString"}}, // optional FunctionArgumentDescriptors{ @@ -5223,7 +5224,7 @@ REGISTER_FUNCTION(Conversion) /// MySQL compatibility alias. Cannot be registered as alias, /// because we don't want it to be normalized to toDate in queries, /// otherwise CREATE DICTIONARY query breaks. - factory.registerFunction("DATE", &FunctionToDate::create, {}, FunctionFactory::CaseInsensitive); + factory.registerFunction("DATE", &FunctionToDate::create, {}, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); diff --git a/src/Functions/FunctionsHashing.h b/src/Functions/FunctionsHashing.h index 27717ea3611..95c54ac9528 100644 --- a/src/Functions/FunctionsHashing.h +++ b/src/Functions/FunctionsHashing.h @@ -1184,7 +1184,7 @@ private: if (icolumn->size() != vec_to.size()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Argument column '{}' size {} doesn't match result column size {} of function {}", - icolumn->getName(), icolumn->size(), vec_to.size(), getName()); + icolumn->getName(), icolumn->size(), vec_to.size(), getName()); if constexpr (Keyed) if (key_cols.size() != vec_to.size() && key_cols.size() != 1) @@ -1223,6 +1223,9 @@ private: else executeGeneric(key_cols, icolumn, vec_to); } + /// Return a fixed random-looking magic number when input is empty. + static constexpr auto filler = 0xe28dbde7fe22e41c; + void executeForArgument(const KeyColumnsType & key_cols, const IDataType * type, const IColumn * column, typename ColumnVector::Container & vec_to, bool & is_first) const { /// Flattening of tuples. @@ -1231,6 +1234,11 @@ private: const auto & tuple_columns = tuple->getColumns(); const DataTypes & tuple_types = typeid_cast(*type).getElements(); size_t tuple_size = tuple_columns.size(); + + if (0 == tuple_size && is_first) + for (auto & hash : vec_to) + hash = static_cast(filler); + for (size_t i = 0; i < tuple_size; ++i) executeForArgument(key_cols, tuple_types[i].get(), tuple_columns[i].get(), vec_to, is_first); } @@ -1239,6 +1247,11 @@ private: const auto & tuple_columns = tuple_const->getColumns(); const DataTypes & tuple_types = typeid_cast(*type).getElements(); size_t tuple_size = tuple_columns.size(); + + if (0 == tuple_size && is_first) + for (auto & hash : vec_to) + hash = static_cast(filler); + for (size_t i = 0; i < tuple_size; ++i) { auto tmp = ColumnConst::create(tuple_columns[i], column->size()); @@ -1300,10 +1313,7 @@ public: constexpr size_t first_data_argument = Keyed; if (arguments.size() <= first_data_argument) - { - /// Return a fixed random-looking magic number when input is empty - vec_to.assign(input_rows_count, static_cast(0xe28dbde7fe22e41c)); - } + vec_to.assign(input_rows_count, static_cast(filler)); KeyColumnsType key_cols{}; if constexpr (Keyed) diff --git a/src/Functions/FunctionsHashingMisc.cpp b/src/Functions/FunctionsHashingMisc.cpp index 38f16af0e6d..5cc29215fe3 100644 --- a/src/Functions/FunctionsHashingMisc.cpp +++ b/src/Functions/FunctionsHashingMisc.cpp @@ -41,8 +41,7 @@ REGISTER_FUNCTION(Hashing) .description="Calculates value of XXH3 64-bit hash function. Refer to https://github.com/Cyan4973/xxHash for detailed documentation.", .examples{{"hash", "SELECT xxh3('ClickHouse')", ""}}, .categories{"Hash"} - }, - FunctionFactory::CaseSensitive); + }); factory.registerFunction(); diff --git a/src/Functions/FunctionsJSON.cpp b/src/Functions/FunctionsJSON.cpp index fbd987577e9..848856c500f 100644 --- a/src/Functions/FunctionsJSON.cpp +++ b/src/Functions/FunctionsJSON.cpp @@ -1,10 +1,1065 @@ -#include +#include +#include + +#include + +#include +#include + +#include +#include + +#include +#include + +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include #include +#include +#include +#include +#include + +#include + +#include "config.h" namespace DB { +namespace ErrorCodes +{ +extern const int ILLEGAL_TYPE_OF_ARGUMENT; +extern const int ILLEGAL_COLUMN; +extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; +} + +template +concept HasIndexOperator = requires (T t) +{ + t[0]; +}; + +/// Functions to parse JSONs and extract values from it. +/// The first argument of all these functions gets a JSON, +/// after that there are any number of arguments specifying path to a desired part from the JSON's root. +/// For example, +/// select JSONExtractInt('{"a": "hello", "b": [-100, 200.0, 300]}', 'b', 1) = -100 + +class FunctionJSONHelpers +{ +public: + template typename Impl, class JSONParser> + class Executor + { + public: + static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count, const FormatSettings & format_settings) + { + MutableColumnPtr to{result_type->createColumn()}; + to->reserve(input_rows_count); + + if (arguments.empty()) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} requires at least one argument", String(Name::name)); + + const auto & first_column = arguments[0]; + if (!isString(first_column.type)) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "The first argument of function {} should be a string containing JSON, illegal type: " + "{}", String(Name::name), first_column.type->getName()); + + const ColumnPtr & arg_json = first_column.column; + const auto * col_json_const = typeid_cast(arg_json.get()); + const auto * col_json_string + = typeid_cast(col_json_const ? col_json_const->getDataColumnPtr().get() : arg_json.get()); + + if (!col_json_string) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {}", arg_json->getName()); + + const ColumnString::Chars & chars = col_json_string->getChars(); + const ColumnString::Offsets & offsets = col_json_string->getOffsets(); + + size_t num_index_arguments = Impl::getNumberOfIndexArguments(arguments); + std::vector moves = prepareMoves(Name::name, arguments, 1, num_index_arguments); + + /// Preallocate memory in parser if necessary. + JSONParser parser; + if constexpr (has_member_function_reserve::value) + { + size_t max_size = calculateMaxSize(offsets); + if (max_size) + parser.reserve(max_size); + } + + Impl impl; + + /// prepare() does Impl-specific preparation before handling each row. + if constexpr (has_member_function_prepare::*)(const char *, const ColumnsWithTypeAndName &, const DataTypePtr &)>::value) + impl.prepare(Name::name, arguments, result_type); + + using Element = typename JSONParser::Element; + + Element document; + bool document_ok = false; + if (col_json_const) + { + std::string_view json{reinterpret_cast(chars.data()), offsets[0] - 1}; + document_ok = parser.parse(json, document); + } + + String error; + for (const auto i : collections::range(0, input_rows_count)) + { + if (!col_json_const) + { + std::string_view json{reinterpret_cast(&chars[offsets[i - 1]]), offsets[i] - offsets[i - 1] - 1}; + document_ok = parser.parse(json, document); + } + + bool added_to_column = false; + if (document_ok) + { + /// Perform moves. + Element element; + std::string_view last_key; + bool moves_ok = performMoves(arguments, i, document, moves, element, last_key); + + if (moves_ok) + added_to_column = impl.insertResultToColumn(*to, element, last_key, format_settings, error); + } + + /// We add default value (=null or zero) if something goes wrong, we don't throw exceptions in these JSON functions. + if (!added_to_column) + to->insertDefault(); + } + return to; + } + }; + +private: + BOOST_TTI_HAS_MEMBER_FUNCTION(reserve) + BOOST_TTI_HAS_MEMBER_FUNCTION(prepare) + + /// Represents a move of a JSON iterator described by a single argument passed to a JSON function. + /// For example, the call JSONExtractInt('{"a": "hello", "b": [-100, 200.0, 300]}', 'b', 1) + /// contains two moves: {MoveType::ConstKey, "b"} and {MoveType::ConstIndex, 1}. + /// Keys and indices can be nonconst, in this case they are calculated for each row. + enum class MoveType : uint8_t + { + Key, + Index, + ConstKey, + ConstIndex, + }; + + struct Move + { + explicit Move(MoveType type_, size_t index_ = 0) : type(type_), index(index_) {} + Move(MoveType type_, const String & key_) : type(type_), key(key_) {} + MoveType type; + size_t index = 0; + String key; + }; + + static std::vector prepareMoves( + const char * function_name, + const ColumnsWithTypeAndName & columns, + size_t first_index_argument, + size_t num_index_arguments) + { + std::vector moves; + moves.reserve(num_index_arguments); + for (const auto i : collections::range(first_index_argument, first_index_argument + num_index_arguments)) + { + const auto & column = columns[i]; + if (!isString(column.type) && !isNativeInteger(column.type)) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "The argument {} of function {} should be a string specifying key " + "or an integer specifying index, illegal type: {}", + std::to_string(i + 1), String(function_name), column.type->getName()); + + if (column.column && isColumnConst(*column.column)) + { + const auto & column_const = assert_cast(*column.column); + if (isString(column.type)) + moves.emplace_back(MoveType::ConstKey, column_const.getValue()); + else + moves.emplace_back(MoveType::ConstIndex, column_const.getInt(0)); + } + else + { + if (isString(column.type)) + moves.emplace_back(MoveType::Key, ""); + else + moves.emplace_back(MoveType::Index, 0); + } + } + return moves; + } + + + /// Performs moves of types MoveType::Index and MoveType::ConstIndex. + template + static bool performMoves(const ColumnsWithTypeAndName & arguments, size_t row, + const typename JSONParser::Element & document, const std::vector & moves, + typename JSONParser::Element & element, std::string_view & last_key) + { + typename JSONParser::Element res_element = document; + std::string_view key; + + for (size_t j = 0; j != moves.size(); ++j) + { + switch (moves[j].type) + { + case MoveType::ConstIndex: + { + if (!moveToElementByIndex(res_element, static_cast(moves[j].index), key)) + return false; + break; + } + case MoveType::ConstKey: + { + key = moves[j].key; + if (!moveToElementByKey(res_element, key)) + return false; + break; + } + case MoveType::Index: + { + Int64 index = (*arguments[j + 1].column)[row].get(); + if (!moveToElementByIndex(res_element, static_cast(index), key)) + return false; + break; + } + case MoveType::Key: + { + key = arguments[j + 1].column->getDataAt(row).toView(); + if (!moveToElementByKey(res_element, key)) + return false; + break; + } + } + } + + element = res_element; + last_key = key; + return true; + } + + template + static bool moveToElementByIndex(typename JSONParser::Element & element, int index, std::string_view & out_key) + { + if (element.isArray()) + { + auto array = element.getArray(); + if (index >= 0) + --index; + else + index += array.size(); + + if (static_cast(index) >= array.size()) + return false; + element = array[index]; + out_key = {}; + return true; + } + + if constexpr (HasIndexOperator) + { + if (element.isObject()) + { + auto object = element.getObject(); + if (index >= 0) + --index; + else + index += object.size(); + + if (static_cast(index) >= object.size()) + return false; + std::tie(out_key, element) = object[index]; + return true; + } + } + + return {}; + } + + /// Performs moves of types MoveType::Key and MoveType::ConstKey. + template + static bool moveToElementByKey(typename JSONParser::Element & element, std::string_view key) + { + if (!element.isObject()) + return false; + auto object = element.getObject(); + return object.find(key, element); + } + + static size_t calculateMaxSize(const ColumnString::Offsets & offsets) + { + size_t max_size = 0; + for (const auto i : collections::range(0, offsets.size())) + { + size_t size = offsets[i] - offsets[i - 1]; + max_size = std::max(max_size, size); + } + if (max_size) + --max_size; + return max_size; + } + +}; + +template +class JSONExtractImpl; + +template +class JSONExtractKeysAndValuesImpl; + +/** +* Functions JSONExtract and JSONExtractKeysAndValues force the return type - it is specified in the last argument. +* For example - `SELECT JSONExtract(materialize('{"a": 131231, "b": 1234}'), 'b', 'LowCardinality(FixedString(4))')` +* But by default ClickHouse decides on its own whether the return type will be LowCardinality based on the types of +* input arguments. +* And for these specific functions we cannot rely on this mechanism, so these functions have their own implementation - +* just convert all of the LowCardinality input columns to full ones, execute and wrap the resulting column in LowCardinality +* if needed. +*/ +template typename Impl> +constexpr bool functionForcesTheReturnType() +{ + return std::is_same_v, JSONExtractImpl> || std::is_same_v, JSONExtractKeysAndValuesImpl>; +} + +template typename Impl> +class ExecutableFunctionJSON : public IExecutableFunction +{ + +public: + explicit ExecutableFunctionJSON(const NullPresence & null_presence_, bool allow_simdjson_, const DataTypePtr & json_return_type_, const FormatSettings & format_settings_) + : null_presence(null_presence_), allow_simdjson(allow_simdjson_), json_return_type(json_return_type_), format_settings(format_settings_) + { + /// Don't escape forward slashes during converting JSON elements to raw string. + format_settings.json.escape_forward_slashes = false; + /// Don't insert default values on null during traversing the JSON element. + /// We allow to insert null only to Nullable columns in JSONExtract functions. + format_settings.null_as_default = false; + } + + String getName() const override { return Name::name; } + bool useDefaultImplementationForNulls() const override { return false; } + bool useDefaultImplementationForConstants() const override { return true; } + bool useDefaultImplementationForLowCardinalityColumns() const override + { + return !functionForcesTheReturnType(); + } + + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override + { + if (null_presence.has_null_constant) + return result_type->createColumnConstWithDefaultValue(input_rows_count); + + if constexpr (functionForcesTheReturnType()) + { + ColumnsWithTypeAndName columns_without_low_cardinality = arguments; + + for (auto & column : columns_without_low_cardinality) + { + column.column = recursiveRemoveLowCardinality(column.column); + column.type = recursiveRemoveLowCardinality(column.type); + } + + ColumnsWithTypeAndName temporary_columns = null_presence.has_nullable ? createBlockWithNestedColumns(columns_without_low_cardinality) : columns_without_low_cardinality; + ColumnPtr temporary_result = chooseAndRunJSONParser(temporary_columns, json_return_type, input_rows_count); + + if (null_presence.has_nullable) + temporary_result = wrapInNullable(temporary_result, columns_without_low_cardinality, result_type, input_rows_count); + + if (result_type->lowCardinality()) + temporary_result = recursiveLowCardinalityTypeConversion(temporary_result, json_return_type, result_type); + + return temporary_result; + } + else + { + ColumnsWithTypeAndName temporary_columns = null_presence.has_nullable ? createBlockWithNestedColumns(arguments) : arguments; + ColumnPtr temporary_result = chooseAndRunJSONParser(temporary_columns, json_return_type, input_rows_count); + + if (null_presence.has_nullable) + temporary_result = wrapInNullable(temporary_result, arguments, result_type, input_rows_count); + + if (result_type->lowCardinality()) + temporary_result = recursiveLowCardinalityTypeConversion(temporary_result, json_return_type, result_type); + + return temporary_result; + } + } + +private: + + ColumnPtr + chooseAndRunJSONParser(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const + { +#if USE_SIMDJSON + if (allow_simdjson) + return FunctionJSONHelpers::Executor::run(arguments, result_type, input_rows_count, format_settings); +#endif + +#if USE_RAPIDJSON + return FunctionJSONHelpers::Executor::run(arguments, result_type, input_rows_count, format_settings); +#else + return FunctionJSONHelpers::Executor::run(arguments, result_type, input_rows_count, format_settings); +#endif + } + + NullPresence null_presence; + bool allow_simdjson; + DataTypePtr json_return_type; + FormatSettings format_settings; +}; + + +template typename Impl> +class FunctionBaseFunctionJSON : public IFunctionBase +{ +public: + explicit FunctionBaseFunctionJSON( + const NullPresence & null_presence_, + bool allow_simdjson_, + DataTypes argument_types_, + DataTypePtr return_type_, + DataTypePtr json_return_type_, + const FormatSettings & format_settings_) + : null_presence(null_presence_) + , allow_simdjson(allow_simdjson_) + , argument_types(std::move(argument_types_)) + , return_type(std::move(return_type_)) + , json_return_type(std::move(json_return_type_)) + , format_settings(format_settings_) + { + } + + String getName() const override { return Name::name; } + + const DataTypes & getArgumentTypes() const override + { + return argument_types; + } + + const DataTypePtr & getResultType() const override + { + return return_type; + } + + bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } + + ExecutableFunctionPtr prepare(const ColumnsWithTypeAndName &) const override + { + return std::make_unique>(null_presence, allow_simdjson, json_return_type, format_settings); + } + +private: + NullPresence null_presence; + bool allow_simdjson; + DataTypes argument_types; + DataTypePtr return_type; + DataTypePtr json_return_type; + FormatSettings format_settings; +}; + +/// We use IFunctionOverloadResolver instead of IFunction to handle non-default NULL processing. +/// Both NULL and JSON NULL should generate NULL value. If any argument is NULL, return NULL. +template typename Impl> +class JSONOverloadResolver : public IFunctionOverloadResolver, WithContext +{ +public: + static constexpr auto name = Name::name; + + String getName() const override { return name; } + + static FunctionOverloadResolverPtr create(ContextPtr context_) + { + return std::make_unique(context_); + } + + explicit JSONOverloadResolver(ContextPtr context_) : WithContext(context_) {} + + bool isVariadic() const override { return true; } + size_t getNumberOfArguments() const override { return 0; } + bool useDefaultImplementationForNulls() const override { return false; } + bool useDefaultImplementationForLowCardinalityColumns() const override + { + return !functionForcesTheReturnType(); + } + + FunctionBasePtr build(const ColumnsWithTypeAndName & arguments) const override + { + bool has_nothing_argument = false; + for (const auto & arg : arguments) + has_nothing_argument |= isNothing(arg.type); + + DataTypePtr json_return_type = Impl::getReturnType(Name::name, createBlockWithNestedColumns(arguments)); + NullPresence null_presence = getNullPresense(arguments); + DataTypePtr return_type; + if (has_nothing_argument) + return_type = std::make_shared(); + else if (null_presence.has_null_constant) + return_type = makeNullable(std::make_shared()); + else if (null_presence.has_nullable) + return_type = makeNullable(json_return_type); + else + return_type = json_return_type; + + DataTypes argument_types; + argument_types.reserve(arguments.size()); + for (const auto & argument : arguments) + argument_types.emplace_back(argument.type); + return std::make_unique>( + null_presence, getContext()->getSettingsRef().allow_simdjson, argument_types, return_type, json_return_type, getFormatSettings(getContext())); + } +}; + +struct NameJSONHas { static constexpr auto name{"JSONHas"}; }; +struct NameIsValidJSON { static constexpr auto name{"isValidJSON"}; }; +struct NameJSONLength { static constexpr auto name{"JSONLength"}; }; +struct NameJSONKey { static constexpr auto name{"JSONKey"}; }; +struct NameJSONType { static constexpr auto name{"JSONType"}; }; +struct NameJSONExtractInt { static constexpr auto name{"JSONExtractInt"}; }; +struct NameJSONExtractUInt { static constexpr auto name{"JSONExtractUInt"}; }; +struct NameJSONExtractFloat { static constexpr auto name{"JSONExtractFloat"}; }; +struct NameJSONExtractBool { static constexpr auto name{"JSONExtractBool"}; }; +struct NameJSONExtractString { static constexpr auto name{"JSONExtractString"}; }; +struct NameJSONExtract { static constexpr auto name{"JSONExtract"}; }; +struct NameJSONExtractKeysAndValues { static constexpr auto name{"JSONExtractKeysAndValues"}; }; +struct NameJSONExtractRaw { static constexpr auto name{"JSONExtractRaw"}; }; +struct NameJSONExtractArrayRaw { static constexpr auto name{"JSONExtractArrayRaw"}; }; +struct NameJSONExtractKeysAndValuesRaw { static constexpr auto name{"JSONExtractKeysAndValuesRaw"}; }; +struct NameJSONExtractKeys { static constexpr auto name{"JSONExtractKeys"}; }; + + +template +class JSONHasImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) { return std::make_shared(); } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static bool insertResultToColumn(IColumn & dest, const Element &, std::string_view, const FormatSettings &, String &) + { + ColumnVector & col_vec = assert_cast &>(dest); + col_vec.insertValue(1); + return true; + } +}; + + +template +class IsValidJSONImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char * function_name, const ColumnsWithTypeAndName & arguments) + { + if (arguments.size() != 1) + { + /// IsValidJSON() shouldn't get parameters other than JSON. + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} needs exactly one argument", + String(function_name)); + } + return std::make_shared(); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName &) { return 0; } + + static bool insertResultToColumn(IColumn & dest, const Element &, std::string_view, const FormatSettings &, String &) + { + /// This function is called only if JSON is valid. + /// If JSON isn't valid then `FunctionJSON::Executor::run()` adds default value (=zero) to `dest` without calling this function. + ColumnVector & col_vec = assert_cast &>(dest); + col_vec.insertValue(1); + return true; + } +}; + + +template +class JSONLengthImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + return std::make_shared(); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings &, String &) + { + size_t size; + if (element.isArray()) + size = element.getArray().size(); + else if (element.isObject()) + size = element.getObject().size(); + else + return false; + + ColumnVector & col_vec = assert_cast &>(dest); + col_vec.insertValue(size); + return true; + } +}; + + +template +class JSONKeyImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + return std::make_shared(); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static bool insertResultToColumn(IColumn & dest, const Element &, std::string_view last_key, const FormatSettings &, String &) + { + if (last_key.empty()) + return false; + ColumnString & col_str = assert_cast(dest); + col_str.insertData(last_key.data(), last_key.size()); + return true; + } +}; + + +template +class JSONTypeImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + static const std::vector> values = { + {"Array", '['}, + {"Object", '{'}, + {"String", '"'}, + {"Int64", 'i'}, + {"UInt64", 'u'}, + {"Double", 'd'}, + {"Bool", 'b'}, + {"Null", 0}, /// the default value for the column. + }; + return std::make_shared>(values); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings &, String &) + { + UInt8 type; + switch (element.type()) + { + case ElementType::INT64: + type = 'i'; + break; + case ElementType::UINT64: + type = 'u'; + break; + case ElementType::DOUBLE: + type = 'd'; + break; + case ElementType::STRING: + type = '"'; + break; + case ElementType::ARRAY: + type = '['; + break; + case ElementType::OBJECT: + type = '{'; + break; + case ElementType::BOOL: + type = 'b'; + break; + case ElementType::NULL_VALUE: + type = 0; + break; + } + + ColumnVector & col_vec = assert_cast &>(dest); + col_vec.insertValue(type); + return true; + } +}; + + +template +class JSONExtractNumericImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + return std::make_shared>(); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static const std::unique_ptr> & getInsertNode() + { + static const std::unique_ptr> node = buildJSONExtractTree(std::make_shared>()); + } + + static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings &, String & error) + { + NumberType value; + + if (!tryGetNumericValueFromJSONElement(value, element, convert_bool_to_integer, error)) + return false; + auto & col_vec = assert_cast &>(dest); + col_vec.insertValue(value); + return true; + } +}; + + +template +using JSONExtractInt64Impl = JSONExtractNumericImpl; +template +using JSONExtractUInt64Impl = JSONExtractNumericImpl; +template +using JSONExtractFloat64Impl = JSONExtractNumericImpl; + + +template +class JSONExtractBoolImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + return std::make_shared(); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings &, String &) + { + bool value; + switch (element.type()) + { + case ElementType::BOOL: + value = element.getBool(); + break; + case ElementType::INT64: + value = element.getInt64() != 0; + break; + case ElementType::UINT64: + value = element.getUInt64() != 0; + break; + default: + return false; + } + + auto & col_vec = assert_cast &>(dest); + col_vec.insertValue(static_cast(value)); + return true; + } +}; + +template +class JSONExtractRawImpl; + +template +class JSONExtractStringImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + return std::make_shared(); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings & format_settings, String & error) + { + if (element.isNull()) + return false; + + if (!element.isString()) + return JSONExtractRawImpl::insertResultToColumn(dest, element, {}, format_settings, error); + + auto str = element.getString(); + ColumnString & col_str = assert_cast(dest); + col_str.insertData(str.data(), str.size()); + return true; + } +}; + +template +class JSONExtractImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char * function_name, const ColumnsWithTypeAndName & arguments) + { + if (arguments.size() < 2) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} requires at least two arguments", String(function_name)); + + const auto & col = arguments.back(); + const auto * col_type_const = typeid_cast(col.column.get()); + if (!col_type_const || !isString(col.type)) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, + "The last argument of function {} should " + "be a constant string specifying the return data type, illegal value: {}", + String(function_name), col.name); + + return DataTypeFactory::instance().get(col_type_const->getValue()); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 2; } + + void prepare(const char * function_name, const ColumnsWithTypeAndName &, const DataTypePtr & result_type) + { + extract_tree = buildJSONExtractTree(result_type, function_name); + insert_settings.insert_default_on_invalid_elements_in_complex_types = true; + } + + bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings & format_settings, String & error) + { + return extract_tree->insertResultToColumn(dest, element, insert_settings, format_settings, error); + } + +protected: + std::unique_ptr> extract_tree; + JSONExtractInsertSettings insert_settings; +}; + + +template +class JSONExtractKeysAndValuesImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char * function_name, const ColumnsWithTypeAndName & arguments) + { + if (arguments.size() < 2) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} requires at least two arguments", String(function_name)); + + const auto & col = arguments.back(); + const auto * col_type_const = typeid_cast(col.column.get()); + if (!col_type_const || !isString(col.type)) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, + "The last argument of function {} should " + "be a constant string specifying the values' data type, illegal value: {}", + String(function_name), col.name); + + DataTypePtr key_type = std::make_unique(); + DataTypePtr value_type = DataTypeFactory::instance().get(col_type_const->getValue()); + DataTypePtr tuple_type = std::make_unique(DataTypes{key_type, value_type}); + return std::make_unique(tuple_type); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 2; } + + void prepare(const char * function_name, const ColumnsWithTypeAndName &, const DataTypePtr & result_type) + { + const auto tuple_type = typeid_cast(result_type.get())->getNestedType(); + const auto value_type = typeid_cast(tuple_type.get())->getElements()[1]; + extract_tree = buildJSONExtractTree(value_type, function_name); + insert_settings.insert_default_on_invalid_elements_in_complex_types = true; + } + + bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings & format_settings, String & error) + { + if (!element.isObject()) + return false; + + auto object = element.getObject(); + + auto & col_arr = assert_cast(dest); + auto & col_tuple = assert_cast(col_arr.getData()); + size_t old_size = col_tuple.size(); + auto & col_key = assert_cast(col_tuple.getColumn(0)); + auto & col_value = col_tuple.getColumn(1); + + for (const auto & [key, value] : object) + { + if (extract_tree->insertResultToColumn(col_value, value, insert_settings, format_settings, error)) + col_key.insertData(key.data(), key.size()); + } + + if (col_tuple.size() == old_size) + return false; + + col_arr.getOffsets().push_back(col_tuple.size()); + return true; + } + +private: + std::unique_ptr> extract_tree; + JSONExtractInsertSettings insert_settings; +}; + + +template +class JSONExtractRawImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + return std::make_shared(); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings & format_settings, String &) + { + ColumnString & col_str = assert_cast(dest); + auto & chars = col_str.getChars(); + WriteBufferFromVector buf(chars, AppendModeTag()); + jsonElementToString(element, buf, format_settings); + buf.finalize(); + chars.push_back(0); + col_str.getOffsets().push_back(chars.size()); + return true; + } +}; + + +template +class JSONExtractArrayRawImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + return std::make_shared(std::make_shared()); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings & format_settings, String & error) + { + if (!element.isArray()) + return false; + + auto array = element.getArray(); + ColumnArray & col_res = assert_cast(dest); + + for (auto value : array) + JSONExtractRawImpl::insertResultToColumn(col_res.getData(), value, {}, format_settings, error); + + col_res.getOffsets().push_back(col_res.getOffsets().back() + array.size()); + return true; + } +}; + + +template +class JSONExtractKeysAndValuesRawImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + DataTypePtr string_type = std::make_unique(); + DataTypePtr tuple_type = std::make_unique(DataTypes{string_type, string_type}); + return std::make_unique(tuple_type); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings & format_settings, String & error) + { + if (!element.isObject()) + return false; + + auto object = element.getObject(); + + auto & col_arr = assert_cast(dest); + auto & col_tuple = assert_cast(col_arr.getData()); + auto & col_key = assert_cast(col_tuple.getColumn(0)); + auto & col_value = assert_cast(col_tuple.getColumn(1)); + + for (const auto & [key, value] : object) + { + col_key.insertData(key.data(), key.size()); + JSONExtractRawImpl::insertResultToColumn(col_value, value, {}, format_settings, error); + } + + col_arr.getOffsets().push_back(col_arr.getOffsets().back() + object.size()); + return true; + } +}; + +template +class JSONExtractKeysImpl +{ +public: + using Element = typename JSONParser::Element; + + static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) + { + return std::make_unique(std::make_shared()); + } + + static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } + + bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view, const FormatSettings &, String &) + { + if (!element.isObject()) + return false; + + auto object = element.getObject(); + + ColumnArray & col_res = assert_cast(dest); + auto & col_key = assert_cast(col_res.getData()); + + for (const auto & [key, value] : object) + { + col_key.insertData(key.data(), key.size()); + } + + col_res.getOffsets().push_back(col_res.getOffsets().back() + object.size()); + return true; + } +}; + REGISTER_FUNCTION(JSON) { factory.registerFunction>(); diff --git a/src/Functions/FunctionsJSON.h b/src/Functions/FunctionsJSON.h deleted file mode 100644 index 8a2ad457d34..00000000000 --- a/src/Functions/FunctionsJSON.h +++ /dev/null @@ -1,1781 +0,0 @@ -#pragma once - -#include -#include - -#include - -#include -#include -#include - -#include -#include - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include - -#include -#include - - -#include "config.h" - - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int ILLEGAL_TYPE_OF_ARGUMENT; - extern const int ILLEGAL_COLUMN; - extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; -} - -template -concept HasIndexOperator = requires (T t) -{ - t[0]; -}; - -/// Functions to parse JSONs and extract values from it. -/// The first argument of all these functions gets a JSON, -/// after that there are any number of arguments specifying path to a desired part from the JSON's root. -/// For example, -/// select JSONExtractInt('{"a": "hello", "b": [-100, 200.0, 300]}', 'b', 1) = -100 - -class FunctionJSONHelpers -{ -public: - template typename Impl, class JSONParser> - class Executor - { - public: - static ColumnPtr run(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) - { - MutableColumnPtr to{result_type->createColumn()}; - to->reserve(input_rows_count); - - if (arguments.empty()) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} requires at least one argument", String(Name::name)); - - const auto & first_column = arguments[0]; - if (!isString(first_column.type)) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "The first argument of function {} should be a string containing JSON, illegal type: " - "{}", String(Name::name), first_column.type->getName()); - - const ColumnPtr & arg_json = first_column.column; - const auto * col_json_const = typeid_cast(arg_json.get()); - const auto * col_json_string - = typeid_cast(col_json_const ? col_json_const->getDataColumnPtr().get() : arg_json.get()); - - if (!col_json_string) - throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {}", arg_json->getName()); - - const ColumnString::Chars & chars = col_json_string->getChars(); - const ColumnString::Offsets & offsets = col_json_string->getOffsets(); - - size_t num_index_arguments = Impl::getNumberOfIndexArguments(arguments); - std::vector moves = prepareMoves(Name::name, arguments, 1, num_index_arguments); - - /// Preallocate memory in parser if necessary. - JSONParser parser; - if constexpr (has_member_function_reserve::value) - { - size_t max_size = calculateMaxSize(offsets); - if (max_size) - parser.reserve(max_size); - } - - Impl impl; - - /// prepare() does Impl-specific preparation before handling each row. - if constexpr (has_member_function_prepare::*)(const char *, const ColumnsWithTypeAndName &, const DataTypePtr &)>::value) - impl.prepare(Name::name, arguments, result_type); - - using Element = typename JSONParser::Element; - - Element document; - bool document_ok = false; - if (col_json_const) - { - std::string_view json{reinterpret_cast(chars.data()), offsets[0] - 1}; - document_ok = parser.parse(json, document); - } - - for (const auto i : collections::range(0, input_rows_count)) - { - if (!col_json_const) - { - std::string_view json{reinterpret_cast(&chars[offsets[i - 1]]), offsets[i] - offsets[i - 1] - 1}; - document_ok = parser.parse(json, document); - } - - bool added_to_column = false; - if (document_ok) - { - /// Perform moves. - Element element; - std::string_view last_key; - bool moves_ok = performMoves(arguments, i, document, moves, element, last_key); - - if (moves_ok) - added_to_column = impl.insertResultToColumn(*to, element, last_key); - } - - /// We add default value (=null or zero) if something goes wrong, we don't throw exceptions in these JSON functions. - if (!added_to_column) - to->insertDefault(); - } - return to; - } - }; - -private: - BOOST_TTI_HAS_MEMBER_FUNCTION(reserve) - BOOST_TTI_HAS_MEMBER_FUNCTION(prepare) - - /// Represents a move of a JSON iterator described by a single argument passed to a JSON function. - /// For example, the call JSONExtractInt('{"a": "hello", "b": [-100, 200.0, 300]}', 'b', 1) - /// contains two moves: {MoveType::ConstKey, "b"} and {MoveType::ConstIndex, 1}. - /// Keys and indices can be nonconst, in this case they are calculated for each row. - enum class MoveType : uint8_t - { - Key, - Index, - ConstKey, - ConstIndex, - }; - - struct Move - { - explicit Move(MoveType type_, size_t index_ = 0) : type(type_), index(index_) {} - Move(MoveType type_, const String & key_) : type(type_), key(key_) {} - MoveType type; - size_t index = 0; - String key; - }; - - static std::vector prepareMoves( - const char * function_name, - const ColumnsWithTypeAndName & columns, - size_t first_index_argument, - size_t num_index_arguments) - { - std::vector moves; - moves.reserve(num_index_arguments); - for (const auto i : collections::range(first_index_argument, first_index_argument + num_index_arguments)) - { - const auto & column = columns[i]; - if (!isString(column.type) && !isNativeInteger(column.type)) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "The argument {} of function {} should be a string specifying key " - "or an integer specifying index, illegal type: {}", - std::to_string(i + 1), String(function_name), column.type->getName()); - - if (column.column && isColumnConst(*column.column)) - { - const auto & column_const = assert_cast(*column.column); - if (isString(column.type)) - moves.emplace_back(MoveType::ConstKey, column_const.getValue()); - else - moves.emplace_back(MoveType::ConstIndex, column_const.getInt(0)); - } - else - { - if (isString(column.type)) - moves.emplace_back(MoveType::Key, ""); - else - moves.emplace_back(MoveType::Index, 0); - } - } - return moves; - } - - - /// Performs moves of types MoveType::Index and MoveType::ConstIndex. - template - static bool performMoves(const ColumnsWithTypeAndName & arguments, size_t row, - const typename JSONParser::Element & document, const std::vector & moves, - typename JSONParser::Element & element, std::string_view & last_key) - { - typename JSONParser::Element res_element = document; - std::string_view key; - - for (size_t j = 0; j != moves.size(); ++j) - { - switch (moves[j].type) - { - case MoveType::ConstIndex: - { - if (!moveToElementByIndex(res_element, static_cast(moves[j].index), key)) - return false; - break; - } - case MoveType::ConstKey: - { - key = moves[j].key; - if (!moveToElementByKey(res_element, key)) - return false; - break; - } - case MoveType::Index: - { - Int64 index = (*arguments[j + 1].column)[row].get(); - if (!moveToElementByIndex(res_element, static_cast(index), key)) - return false; - break; - } - case MoveType::Key: - { - key = arguments[j + 1].column->getDataAt(row).toView(); - if (!moveToElementByKey(res_element, key)) - return false; - break; - } - } - } - - element = res_element; - last_key = key; - return true; - } - - template - static bool moveToElementByIndex(typename JSONParser::Element & element, int index, std::string_view & out_key) - { - if (element.isArray()) - { - auto array = element.getArray(); - if (index >= 0) - --index; - else - index += array.size(); - - if (static_cast(index) >= array.size()) - return false; - element = array[index]; - out_key = {}; - return true; - } - - if constexpr (HasIndexOperator) - { - if (element.isObject()) - { - auto object = element.getObject(); - if (index >= 0) - --index; - else - index += object.size(); - - if (static_cast(index) >= object.size()) - return false; - std::tie(out_key, element) = object[index]; - return true; - } - } - - return {}; - } - - /// Performs moves of types MoveType::Key and MoveType::ConstKey. - template - static bool moveToElementByKey(typename JSONParser::Element & element, std::string_view key) - { - if (!element.isObject()) - return false; - auto object = element.getObject(); - return object.find(key, element); - } - - static size_t calculateMaxSize(const ColumnString::Offsets & offsets) - { - size_t max_size = 0; - for (const auto i : collections::range(0, offsets.size())) - { - size_t size = offsets[i] - offsets[i - 1]; - max_size = std::max(max_size, size); - } - if (max_size) - --max_size; - return max_size; - } - -}; - -template -class JSONExtractImpl; - -template -class JSONExtractKeysAndValuesImpl; - -/** -* Functions JSONExtract and JSONExtractKeysAndValues force the return type - it is specified in the last argument. -* For example - `SELECT JSONExtract(materialize('{"a": 131231, "b": 1234}'), 'b', 'LowCardinality(FixedString(4))')` -* But by default ClickHouse decides on its own whether the return type will be LowCardinality based on the types of -* input arguments. -* And for these specific functions we cannot rely on this mechanism, so these functions have their own implementation - -* just convert all of the LowCardinality input columns to full ones, execute and wrap the resulting column in LowCardinality -* if needed. -*/ -template typename Impl> -constexpr bool functionForcesTheReturnType() -{ - return std::is_same_v, JSONExtractImpl> || std::is_same_v, JSONExtractKeysAndValuesImpl>; -} - -template typename Impl> -class ExecutableFunctionJSON : public IExecutableFunction -{ - -public: - explicit ExecutableFunctionJSON(const NullPresence & null_presence_, bool allow_simdjson_, const DataTypePtr & json_return_type_) - : null_presence(null_presence_), allow_simdjson(allow_simdjson_), json_return_type(json_return_type_) - { - } - - String getName() const override { return Name::name; } - bool useDefaultImplementationForNulls() const override { return false; } - bool useDefaultImplementationForConstants() const override { return true; } - bool useDefaultImplementationForLowCardinalityColumns() const override - { - return !functionForcesTheReturnType(); - } - - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override - { - if (null_presence.has_null_constant) - return result_type->createColumnConstWithDefaultValue(input_rows_count); - - if constexpr (functionForcesTheReturnType()) - { - ColumnsWithTypeAndName columns_without_low_cardinality = arguments; - - for (auto & column : columns_without_low_cardinality) - { - column.column = recursiveRemoveLowCardinality(column.column); - column.type = recursiveRemoveLowCardinality(column.type); - } - - ColumnsWithTypeAndName temporary_columns = null_presence.has_nullable ? createBlockWithNestedColumns(columns_without_low_cardinality) : columns_without_low_cardinality; - ColumnPtr temporary_result = chooseAndRunJSONParser(temporary_columns, json_return_type, input_rows_count); - - if (null_presence.has_nullable) - temporary_result = wrapInNullable(temporary_result, columns_without_low_cardinality, result_type, input_rows_count); - - if (result_type->lowCardinality()) - temporary_result = recursiveLowCardinalityTypeConversion(temporary_result, json_return_type, result_type); - - return temporary_result; - } - else - { - ColumnsWithTypeAndName temporary_columns = null_presence.has_nullable ? createBlockWithNestedColumns(arguments) : arguments; - ColumnPtr temporary_result = chooseAndRunJSONParser(temporary_columns, json_return_type, input_rows_count); - - if (null_presence.has_nullable) - temporary_result = wrapInNullable(temporary_result, arguments, result_type, input_rows_count); - - if (result_type->lowCardinality()) - temporary_result = recursiveLowCardinalityTypeConversion(temporary_result, json_return_type, result_type); - - return temporary_result; - } - } - -private: - - ColumnPtr - chooseAndRunJSONParser(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const - { -#if USE_SIMDJSON - if (allow_simdjson) - return FunctionJSONHelpers::Executor::run(arguments, result_type, input_rows_count); -#endif - -#if USE_RAPIDJSON - return FunctionJSONHelpers::Executor::run(arguments, result_type, input_rows_count); -#else - return FunctionJSONHelpers::Executor::run(arguments, result_type, input_rows_count); -#endif - } - - NullPresence null_presence; - bool allow_simdjson; - DataTypePtr json_return_type; -}; - - -template typename Impl> -class FunctionBaseFunctionJSON : public IFunctionBase -{ -public: - explicit FunctionBaseFunctionJSON( - const NullPresence & null_presence_, - bool allow_simdjson_, - DataTypes argument_types_, - DataTypePtr return_type_, - DataTypePtr json_return_type_) - : null_presence(null_presence_) - , allow_simdjson(allow_simdjson_) - , argument_types(std::move(argument_types_)) - , return_type(std::move(return_type_)) - , json_return_type(std::move(json_return_type_)) - { - } - - String getName() const override { return Name::name; } - - const DataTypes & getArgumentTypes() const override - { - return argument_types; - } - - const DataTypePtr & getResultType() const override - { - return return_type; - } - - bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } - - ExecutableFunctionPtr prepare(const ColumnsWithTypeAndName &) const override - { - return std::make_unique>(null_presence, allow_simdjson, json_return_type); - } - -private: - NullPresence null_presence; - bool allow_simdjson; - DataTypes argument_types; - DataTypePtr return_type; - DataTypePtr json_return_type; -}; - -/// We use IFunctionOverloadResolver instead of IFunction to handle non-default NULL processing. -/// Both NULL and JSON NULL should generate NULL value. If any argument is NULL, return NULL. -template typename Impl> -class JSONOverloadResolver : public IFunctionOverloadResolver, WithContext -{ -public: - static constexpr auto name = Name::name; - - String getName() const override { return name; } - - static FunctionOverloadResolverPtr create(ContextPtr context_) - { - return std::make_unique(context_); - } - - explicit JSONOverloadResolver(ContextPtr context_) : WithContext(context_) {} - - bool isVariadic() const override { return true; } - size_t getNumberOfArguments() const override { return 0; } - bool useDefaultImplementationForNulls() const override { return false; } - bool useDefaultImplementationForLowCardinalityColumns() const override - { - return !functionForcesTheReturnType(); - } - - FunctionBasePtr build(const ColumnsWithTypeAndName & arguments) const override - { - bool has_nothing_argument = false; - for (const auto & arg : arguments) - has_nothing_argument |= isNothing(arg.type); - - DataTypePtr json_return_type = Impl::getReturnType(Name::name, createBlockWithNestedColumns(arguments)); - NullPresence null_presence = getNullPresense(arguments); - DataTypePtr return_type; - if (has_nothing_argument) - return_type = std::make_shared(); - else if (null_presence.has_null_constant) - return_type = makeNullable(std::make_shared()); - else if (null_presence.has_nullable) - return_type = makeNullable(json_return_type); - else - return_type = json_return_type; - - /// Top-level LowCardinality columns are processed outside JSON parser. - json_return_type = removeLowCardinality(json_return_type); - - DataTypes argument_types; - argument_types.reserve(arguments.size()); - for (const auto & argument : arguments) - argument_types.emplace_back(argument.type); - return std::make_unique>( - null_presence, getContext()->getSettingsRef().allow_simdjson, argument_types, return_type, json_return_type); - } -}; - -struct NameJSONHas { static constexpr auto name{"JSONHas"}; }; -struct NameIsValidJSON { static constexpr auto name{"isValidJSON"}; }; -struct NameJSONLength { static constexpr auto name{"JSONLength"}; }; -struct NameJSONKey { static constexpr auto name{"JSONKey"}; }; -struct NameJSONType { static constexpr auto name{"JSONType"}; }; -struct NameJSONExtractInt { static constexpr auto name{"JSONExtractInt"}; }; -struct NameJSONExtractUInt { static constexpr auto name{"JSONExtractUInt"}; }; -struct NameJSONExtractFloat { static constexpr auto name{"JSONExtractFloat"}; }; -struct NameJSONExtractBool { static constexpr auto name{"JSONExtractBool"}; }; -struct NameJSONExtractString { static constexpr auto name{"JSONExtractString"}; }; -struct NameJSONExtract { static constexpr auto name{"JSONExtract"}; }; -struct NameJSONExtractKeysAndValues { static constexpr auto name{"JSONExtractKeysAndValues"}; }; -struct NameJSONExtractRaw { static constexpr auto name{"JSONExtractRaw"}; }; -struct NameJSONExtractArrayRaw { static constexpr auto name{"JSONExtractArrayRaw"}; }; -struct NameJSONExtractKeysAndValuesRaw { static constexpr auto name{"JSONExtractKeysAndValuesRaw"}; }; -struct NameJSONExtractKeys { static constexpr auto name{"JSONExtractKeys"}; }; - - -template -class JSONHasImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) { return std::make_shared(); } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element &, std::string_view) - { - ColumnVector & col_vec = assert_cast &>(dest); - col_vec.insertValue(1); - return true; - } -}; - - -template -class IsValidJSONImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char * function_name, const ColumnsWithTypeAndName & arguments) - { - if (arguments.size() != 1) - { - /// IsValidJSON() shouldn't get parameters other than JSON. - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} needs exactly one argument", - String(function_name)); - } - return std::make_shared(); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName &) { return 0; } - - static bool insertResultToColumn(IColumn & dest, const Element &, std::string_view) - { - /// This function is called only if JSON is valid. - /// If JSON isn't valid then `FunctionJSON::Executor::run()` adds default value (=zero) to `dest` without calling this function. - ColumnVector & col_vec = assert_cast &>(dest); - col_vec.insertValue(1); - return true; - } -}; - - -template -class JSONLengthImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - return std::make_shared(); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - size_t size; - if (element.isArray()) - size = element.getArray().size(); - else if (element.isObject()) - size = element.getObject().size(); - else - return false; - - ColumnVector & col_vec = assert_cast &>(dest); - col_vec.insertValue(size); - return true; - } -}; - - -template -class JSONKeyImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - return std::make_shared(); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element &, std::string_view last_key) - { - if (last_key.empty()) - return false; - ColumnString & col_str = assert_cast(dest); - col_str.insertData(last_key.data(), last_key.size()); - return true; - } -}; - - -template -class JSONTypeImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - static const std::vector> values = { - {"Array", '['}, - {"Object", '{'}, - {"String", '"'}, - {"Int64", 'i'}, - {"UInt64", 'u'}, - {"Double", 'd'}, - {"Bool", 'b'}, - {"Null", 0}, /// the default value for the column. - }; - return std::make_shared>(values); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - UInt8 type; - switch (element.type()) - { - case ElementType::INT64: - type = 'i'; - break; - case ElementType::UINT64: - type = 'u'; - break; - case ElementType::DOUBLE: - type = 'd'; - break; - case ElementType::STRING: - type = '"'; - break; - case ElementType::ARRAY: - type = '['; - break; - case ElementType::OBJECT: - type = '{'; - break; - case ElementType::BOOL: - type = 'b'; - break; - case ElementType::NULL_VALUE: - type = 0; - break; - } - - ColumnVector & col_vec = assert_cast &>(dest); - col_vec.insertValue(type); - return true; - } -}; - - -template -class JSONExtractNumericImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - return std::make_shared>(); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - NumberType value; - - switch (element.type()) - { - case ElementType::DOUBLE: - if constexpr (std::is_floating_point_v) - { - /// We permit inaccurate conversion of double to float. - /// Example: double 0.1 from JSON is not representable in float. - /// But it will be more convenient for user to perform conversion. - value = static_cast(element.getDouble()); - } - else if (!accurate::convertNumeric(element.getDouble(), value)) - return false; - break; - case ElementType::UINT64: - if (!accurate::convertNumeric(element.getUInt64(), value)) - return false; - break; - case ElementType::INT64: - if (!accurate::convertNumeric(element.getInt64(), value)) - return false; - break; - case ElementType::BOOL: - if constexpr (is_integer && convert_bool_to_integer) - { - value = static_cast(element.getBool()); - break; - } - return false; - case ElementType::STRING: - { - auto rb = ReadBufferFromMemory{element.getString()}; - if constexpr (std::is_floating_point_v) - { - if (!tryReadFloatText(value, rb) || !rb.eof()) - return false; - } - else - { - if (tryReadIntText(value, rb) && rb.eof()) - break; - - /// Try to parse float and convert it to integer. - Float64 tmp_float; - rb.position() = rb.buffer().begin(); - if (!tryReadFloatText(tmp_float, rb) || !rb.eof()) - return false; - - if (!accurate::convertNumeric(tmp_float, value)) - return false; - } - break; - } - default: - return false; - } - - if (dest.getDataType() == TypeIndex::LowCardinality) - { - ColumnLowCardinality & col_low = assert_cast(dest); - col_low.insertData(reinterpret_cast(&value), sizeof(value)); - } - else - { - auto & col_vec = assert_cast &>(dest); - col_vec.insertValue(value); - } - return true; - } -}; - - -template -using JSONExtractInt64Impl = JSONExtractNumericImpl; -template -using JSONExtractUInt64Impl = JSONExtractNumericImpl; -template -using JSONExtractFloat64Impl = JSONExtractNumericImpl; - - -template -class JSONExtractBoolImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - return std::make_shared(); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - bool value; - switch (element.type()) - { - case ElementType::BOOL: - value = element.getBool(); - break; - case ElementType::INT64: - value = element.getInt64() != 0; - break; - case ElementType::UINT64: - value = element.getUInt64() != 0; - break; - default: - return false; - } - - auto & col_vec = assert_cast &>(dest); - col_vec.insertValue(static_cast(value)); - return true; - } -}; - -template -class JSONExtractRawImpl; - -template -class JSONExtractStringImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - return std::make_shared(); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - if (element.isNull()) - return false; - - if (!element.isString()) - return JSONExtractRawImpl::insertResultToColumn(dest, element, {}); - - auto str = element.getString(); - - if (dest.getDataType() == TypeIndex::LowCardinality) - { - ColumnLowCardinality & col_low = assert_cast(dest); - col_low.insertData(str.data(), str.size()); - } - else - { - ColumnString & col_str = assert_cast(dest); - col_str.insertData(str.data(), str.size()); - } - return true; - } -}; - -/// Nodes of the extract tree. We need the extract tree to extract from JSON complex values containing array, tuples or nullables. -template -struct JSONExtractTree -{ - using Element = typename JSONParser::Element; - - class Node - { - public: - Node() = default; - virtual ~Node() = default; - virtual bool insertResultToColumn(IColumn &, const Element &) = 0; - }; - - template - class NumericNode : public Node - { - public: - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - return JSONExtractNumericImpl::insertResultToColumn(dest, element, {}); - } - }; - - class LowCardinalityFixedStringNode : public Node - { - public: - explicit LowCardinalityFixedStringNode(const size_t fixed_length_) : fixed_length(fixed_length_) { } - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - // If element is an object we delegate the insertion to JSONExtractRawImpl - if (element.isObject()) - return JSONExtractRawImpl::insertResultToLowCardinalityFixedStringColumn(dest, element, fixed_length); - else if (!element.isString()) - return false; - - auto str = element.getString(); - if (str.size() > fixed_length) - return false; - - // For the non low cardinality case of FixedString, the padding is done in the FixedString Column implementation. - // In order to avoid having to pass the data to a FixedString Column and read it back (which would slow down the execution) - // the data is padded here and written directly to the Low Cardinality Column - if (str.size() == fixed_length) - { - assert_cast(dest).insertData(str.data(), str.size()); - } - else - { - String padded_str(str); - padded_str.resize(fixed_length, '\0'); - - assert_cast(dest).insertData(padded_str.data(), padded_str.size()); - } - return true; - } - - private: - const size_t fixed_length; - }; - - class UUIDNode : public Node - { - public: - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - if (!element.isString()) - return false; - - auto uuid = parseFromString(element.getString()); - if (dest.getDataType() == TypeIndex::LowCardinality) - { - ColumnLowCardinality & col_low = assert_cast(dest); - col_low.insertData(reinterpret_cast(&uuid), sizeof(uuid)); - } - else - { - assert_cast(dest).insert(uuid); - } - return true; - } - }; - - template - class DecimalNode : public Node - { - public: - explicit DecimalNode(DataTypePtr data_type_) : data_type(data_type_) {} - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - const auto * type = assert_cast *>(data_type.get()); - - DecimalType value{}; - - switch (element.type()) - { - case ElementType::DOUBLE: - value = convertToDecimal, DataTypeDecimal>( - element.getDouble(), type->getScale()); - break; - case ElementType::UINT64: - value = convertToDecimal, DataTypeDecimal>( - element.getUInt64(), type->getScale()); - break; - case ElementType::INT64: - value = convertToDecimal, DataTypeDecimal>( - element.getInt64(), type->getScale()); - break; - case ElementType::STRING: { - auto rb = ReadBufferFromMemory{element.getString()}; - if (!SerializationDecimal::tryReadText(value, rb, DecimalUtils::max_precision, type->getScale())) - return false; - break; - } - default: - return false; - } - - assert_cast &>(dest).insertValue(value); - return true; - } - - private: - DataTypePtr data_type; - }; - - class StringNode : public Node - { - public: - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - return JSONExtractStringImpl::insertResultToColumn(dest, element, {}); - } - }; - - class FixedStringNode : public Node - { - public: - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - if (element.isNull()) - return false; - - if (!element.isString()) - return JSONExtractRawImpl::insertResultToFixedStringColumn(dest, element, {}); - - auto str = element.getString(); - auto & col_str = assert_cast(dest); - if (str.size() > col_str.getN()) - return false; - col_str.insertData(str.data(), str.size()); - - return true; - } - }; - - template - class EnumNode : public Node - { - public: - explicit EnumNode(const std::vector> & name_value_pairs_) : name_value_pairs(name_value_pairs_) - { - for (const auto & name_value_pair : name_value_pairs) - { - name_to_value_map.emplace(name_value_pair.first, name_value_pair.second); - only_values.emplace(name_value_pair.second); - } - } - - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - auto & col_vec = assert_cast &>(dest); - - if (element.isInt64()) - { - Type value; - if (!accurate::convertNumeric(element.getInt64(), value) || !only_values.contains(value)) - return false; - col_vec.insertValue(value); - return true; - } - - if (element.isUInt64()) - { - Type value; - if (!accurate::convertNumeric(element.getUInt64(), value) || !only_values.contains(value)) - return false; - col_vec.insertValue(value); - return true; - } - - if (element.isString()) - { - auto value = name_to_value_map.find(element.getString()); - if (value == name_to_value_map.end()) - return false; - col_vec.insertValue(value->second); - return true; - } - - return false; - } - - private: - std::vector> name_value_pairs; - std::unordered_map name_to_value_map; - std::unordered_set only_values; - }; - - class NullableNode : public Node - { - public: - explicit NullableNode(std::unique_ptr nested_) : nested(std::move(nested_)) {} - - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - if (dest.getDataType() == TypeIndex::LowCardinality) - { - /// We do not need to handle nullability in that case - /// because nested node handles LowCardinality columns and will call proper overload of `insertData` - return nested->insertResultToColumn(dest, element); - } - - ColumnNullable & col_null = assert_cast(dest); - if (!nested->insertResultToColumn(col_null.getNestedColumn(), element)) - return false; - col_null.getNullMapColumn().insertValue(0); - return true; - } - - private: - std::unique_ptr nested; - }; - - class ArrayNode : public Node - { - public: - explicit ArrayNode(std::unique_ptr nested_) : nested(std::move(nested_)) {} - - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - if (!element.isArray()) - return false; - - auto array = element.getArray(); - - ColumnArray & col_arr = assert_cast(dest); - auto & data = col_arr.getData(); - size_t old_size = data.size(); - bool were_valid_elements = false; - - for (auto value : array) - { - if (nested->insertResultToColumn(data, value)) - were_valid_elements = true; - else - data.insertDefault(); - } - - if (!were_valid_elements) - { - data.popBack(data.size() - old_size); - return false; - } - - col_arr.getOffsets().push_back(data.size()); - return true; - } - - private: - std::unique_ptr nested; - }; - - class TupleNode : public Node - { - public: - TupleNode(std::vector> nested_, const std::vector & explicit_names_) : nested(std::move(nested_)), explicit_names(explicit_names_) - { - for (size_t i = 0; i != explicit_names.size(); ++i) - name_to_index_map.emplace(explicit_names[i], i); - } - - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - ColumnTuple & tuple = assert_cast(dest); - size_t old_size = dest.size(); - bool were_valid_elements = false; - - auto set_size = [&](size_t size) - { - for (size_t i = 0; i != tuple.tupleSize(); ++i) - { - auto & col = tuple.getColumn(i); - if (col.size() != size) - { - if (col.size() > size) - col.popBack(col.size() - size); - else - while (col.size() < size) - col.insertDefault(); - } - } - }; - - if (element.isArray()) - { - auto array = element.getArray(); - auto it = array.begin(); - - for (size_t index = 0; (index != nested.size()) && (it != array.end()); ++index) - { - if (nested[index]->insertResultToColumn(tuple.getColumn(index), *it++)) - were_valid_elements = true; - else - tuple.getColumn(index).insertDefault(); - } - - set_size(old_size + static_cast(were_valid_elements)); - return were_valid_elements; - } - - if (element.isObject()) - { - auto object = element.getObject(); - if (name_to_index_map.empty()) - { - auto it = object.begin(); - for (size_t index = 0; (index != nested.size()) && (it != object.end()); ++index) - { - if (nested[index]->insertResultToColumn(tuple.getColumn(index), (*it++).second)) - were_valid_elements = true; - else - tuple.getColumn(index).insertDefault(); - } - } - else - { - for (const auto & [key, value] : object) - { - auto index = name_to_index_map.find(key); - if (index != name_to_index_map.end()) - { - if (nested[index->second]->insertResultToColumn(tuple.getColumn(index->second), value)) - were_valid_elements = true; - } - } - } - - set_size(old_size + static_cast(were_valid_elements)); - return were_valid_elements; - } - - return false; - } - - private: - std::vector> nested; - std::vector explicit_names; - std::unordered_map name_to_index_map; - }; - - class MapNode : public Node - { - public: - MapNode(std::unique_ptr key_, std::unique_ptr value_) : key(std::move(key_)), value(std::move(value_)) { } - - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - if (!element.isObject()) - return false; - - ColumnMap & map_col = assert_cast(dest); - auto & offsets = map_col.getNestedColumn().getOffsets(); - auto & tuple_col = map_col.getNestedData(); - auto & key_col = tuple_col.getColumn(0); - auto & value_col = tuple_col.getColumn(1); - size_t old_size = tuple_col.size(); - - auto object = element.getObject(); - auto it = object.begin(); - for (; it != object.end(); ++it) - { - auto pair = *it; - - /// Insert key - key_col.insertData(pair.first.data(), pair.first.size()); - - /// Insert value - if (!value->insertResultToColumn(value_col, pair.second)) - value_col.insertDefault(); - } - - offsets.push_back(old_size + object.size()); - return true; - } - - private: - std::unique_ptr key; - std::unique_ptr value; - }; - - class VariantNode : public Node - { - public: - VariantNode(std::vector> variant_nodes_, std::vector order_) : variant_nodes(std::move(variant_nodes_)), order(std::move(order_)) { } - - bool insertResultToColumn(IColumn & dest, const Element & element) override - { - auto & column_variant = assert_cast(dest); - for (size_t i : order) - { - auto & variant = column_variant.getVariantByGlobalDiscriminator(i); - if (variant_nodes[i]->insertResultToColumn(variant, element)) - { - column_variant.getLocalDiscriminators().push_back(column_variant.localDiscriminatorByGlobal(i)); - column_variant.getOffsets().push_back(variant.size() - 1); - return true; - } - } - - return false; - } - - private: - std::vector> variant_nodes; - /// Order in which we should try variants nodes. - /// For example, String should be always the last one. - std::vector order; - }; - - static std::unique_ptr build(const char * function_name, const DataTypePtr & type) - { - switch (type->getTypeId()) - { - case TypeIndex::UInt8: return std::make_unique>(); - case TypeIndex::UInt16: return std::make_unique>(); - case TypeIndex::UInt32: return std::make_unique>(); - case TypeIndex::UInt64: return std::make_unique>(); - case TypeIndex::UInt128: return std::make_unique>(); - case TypeIndex::UInt256: return std::make_unique>(); - case TypeIndex::Int8: return std::make_unique>(); - case TypeIndex::Int16: return std::make_unique>(); - case TypeIndex::Int32: return std::make_unique>(); - case TypeIndex::Int64: return std::make_unique>(); - case TypeIndex::Int128: return std::make_unique>(); - case TypeIndex::Int256: return std::make_unique>(); - case TypeIndex::Float32: return std::make_unique>(); - case TypeIndex::Float64: return std::make_unique>(); - case TypeIndex::String: return std::make_unique(); - case TypeIndex::FixedString: return std::make_unique(); - case TypeIndex::UUID: return std::make_unique(); - case TypeIndex::LowCardinality: - { - // The low cardinality case is treated in two different ways: - // For FixedString type, an especial class is implemented for inserting the data in the destination column, - // as the string length must be passed in order to check and pad the incoming data. - // For the rest of low cardinality types, the insertion is done in their corresponding class, adapting the data - // as needed for the insertData function of the ColumnLowCardinality. - auto dictionary_type = typeid_cast(type.get())->getDictionaryType(); - if ((*dictionary_type).getTypeId() == TypeIndex::FixedString) - { - auto fixed_length = typeid_cast(dictionary_type.get())->getN(); - return std::make_unique(fixed_length); - } - return build(function_name, dictionary_type); - } - case TypeIndex::Decimal256: return std::make_unique>(type); - case TypeIndex::Decimal128: return std::make_unique>(type); - case TypeIndex::Decimal64: return std::make_unique>(type); - case TypeIndex::Decimal32: return std::make_unique>(type); - case TypeIndex::Enum8: - return std::make_unique>(static_cast(*type).getValues()); - case TypeIndex::Enum16: - return std::make_unique>(static_cast(*type).getValues()); - case TypeIndex::Nullable: - { - return std::make_unique(build(function_name, static_cast(*type).getNestedType())); - } - case TypeIndex::Array: - { - return std::make_unique(build(function_name, static_cast(*type).getNestedType())); - } - case TypeIndex::Tuple: - { - const auto & tuple = static_cast(*type); - const auto & tuple_elements = tuple.getElements(); - std::vector> elements; - elements.reserve(tuple_elements.size()); - for (const auto & tuple_element : tuple_elements) - elements.emplace_back(build(function_name, tuple_element)); - return std::make_unique(std::move(elements), tuple.haveExplicitNames() ? tuple.getElementNames() : Strings{}); - } - case TypeIndex::Map: - { - const auto & map_type = static_cast(*type); - const auto & key_type = map_type.getKeyType(); - if (!isString(removeLowCardinality(key_type))) - throw Exception( - ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Function {} doesn't support the return type schema: {} with key type not String", - String(function_name), - type->getName()); - - const auto & value_type = map_type.getValueType(); - return std::make_unique(build(function_name, key_type), build(function_name, value_type)); - } - case TypeIndex::Variant: - { - const auto & variant_type = static_cast(*type); - const auto & variants = variant_type.getVariants(); - std::vector> variant_nodes; - variant_nodes.reserve(variants.size()); - for (const auto & variant : variants) - variant_nodes.push_back(build(function_name, variant)); - return std::make_unique(std::move(variant_nodes), SerializationVariant::getVariantsDeserializeTextOrder(variants)); - } - default: - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Function {} doesn't support the return type schema: {}", - String(function_name), type->getName()); - } - } -}; - - -template -class JSONExtractImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char * function_name, const ColumnsWithTypeAndName & arguments) - { - if (arguments.size() < 2) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} requires at least two arguments", String(function_name)); - - const auto & col = arguments.back(); - const auto * col_type_const = typeid_cast(col.column.get()); - if (!col_type_const || !isString(col.type)) - throw Exception(ErrorCodes::ILLEGAL_COLUMN, - "The last argument of function {} should " - "be a constant string specifying the return data type, illegal value: {}", - String(function_name), col.name); - - return DataTypeFactory::instance().get(col_type_const->getValue()); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 2; } - - void prepare(const char * function_name, const ColumnsWithTypeAndName &, const DataTypePtr & result_type) - { - extract_tree = JSONExtractTree::build(function_name, result_type); - } - - bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - return extract_tree->insertResultToColumn(dest, element); - } - -protected: - std::unique_ptr::Node> extract_tree; -}; - - -template -class JSONExtractKeysAndValuesImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char * function_name, const ColumnsWithTypeAndName & arguments) - { - if (arguments.size() < 2) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} requires at least two arguments", String(function_name)); - - const auto & col = arguments.back(); - const auto * col_type_const = typeid_cast(col.column.get()); - if (!col_type_const || !isString(col.type)) - throw Exception(ErrorCodes::ILLEGAL_COLUMN, - "The last argument of function {} should " - "be a constant string specifying the values' data type, illegal value: {}", - String(function_name), col.name); - - DataTypePtr key_type = std::make_unique(); - DataTypePtr value_type = DataTypeFactory::instance().get(col_type_const->getValue()); - DataTypePtr tuple_type = std::make_unique(DataTypes{key_type, value_type}); - return std::make_unique(tuple_type); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 2; } - - void prepare(const char * function_name, const ColumnsWithTypeAndName &, const DataTypePtr & result_type) - { - const auto tuple_type = typeid_cast(result_type.get())->getNestedType(); - const auto value_type = typeid_cast(tuple_type.get())->getElements()[1]; - extract_tree = JSONExtractTree::build(function_name, value_type); - } - - bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - if (!element.isObject()) - return false; - - auto object = element.getObject(); - - auto & col_arr = assert_cast(dest); - auto & col_tuple = assert_cast(col_arr.getData()); - size_t old_size = col_tuple.size(); - auto & col_key = assert_cast(col_tuple.getColumn(0)); - auto & col_value = col_tuple.getColumn(1); - - for (const auto & [key, value] : object) - { - if (extract_tree->insertResultToColumn(col_value, value)) - col_key.insertData(key.data(), key.size()); - } - - if (col_tuple.size() == old_size) - return false; - - col_arr.getOffsets().push_back(col_tuple.size()); - return true; - } - -private: - std::unique_ptr::Node> extract_tree; -}; - - -template -class JSONExtractRawImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - return std::make_shared(); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - if (dest.getDataType() == TypeIndex::LowCardinality) - { - ColumnString::Chars chars; - WriteBufferFromVector buf(chars, AppendModeTag()); - traverse(element, buf); - buf.finalize(); - assert_cast(dest).insertData(reinterpret_cast(chars.data()), chars.size()); - } - else - { - ColumnString & col_str = assert_cast(dest); - auto & chars = col_str.getChars(); - WriteBufferFromVector buf(chars, AppendModeTag()); - traverse(element, buf); - buf.finalize(); - chars.push_back(0); - col_str.getOffsets().push_back(chars.size()); - } - return true; - } - - // We use insertResultToFixedStringColumn in case we are inserting raw data in a FixedString column - static bool insertResultToFixedStringColumn(IColumn & dest, const Element & element, std::string_view) - { - ColumnFixedString::Chars chars; - WriteBufferFromVector buf(chars, AppendModeTag()); - traverse(element, buf); - buf.finalize(); - - auto & col_str = assert_cast(dest); - - if (chars.size() > col_str.getN()) - return false; - - chars.resize_fill(col_str.getN()); - col_str.insertData(reinterpret_cast(chars.data()), chars.size()); - - - return true; - } - - // We use insertResultToLowCardinalityFixedStringColumn in case we are inserting raw data in a Low Cardinality FixedString column - static bool insertResultToLowCardinalityFixedStringColumn(IColumn & dest, const Element & element, size_t fixed_length) - { - if (element.getObject().size() > fixed_length) - return false; - - ColumnFixedString::Chars chars; - WriteBufferFromVector buf(chars, AppendModeTag()); - traverse(element, buf); - buf.finalize(); - - if (chars.size() > fixed_length) - return false; - chars.resize_fill(fixed_length); - assert_cast(dest).insertData(reinterpret_cast(chars.data()), chars.size()); - - return true; - } - -private: - static void traverse(const Element & element, WriteBuffer & buf) - { - if (element.isInt64()) - { - writeIntText(element.getInt64(), buf); - return; - } - if (element.isUInt64()) - { - writeIntText(element.getUInt64(), buf); - return; - } - if (element.isDouble()) - { - writeFloatText(element.getDouble(), buf); - return; - } - if (element.isBool()) - { - if (element.getBool()) - writeCString("true", buf); - else - writeCString("false", buf); - return; - } - if (element.isString()) - { - writeJSONString(element.getString(), buf, formatSettings()); - return; - } - if (element.isArray()) - { - writeChar('[', buf); - bool need_comma = false; - for (auto value : element.getArray()) - { - if (std::exchange(need_comma, true)) - writeChar(',', buf); - traverse(value, buf); - } - writeChar(']', buf); - return; - } - if (element.isObject()) - { - writeChar('{', buf); - bool need_comma = false; - for (auto [key, value] : element.getObject()) - { - if (std::exchange(need_comma, true)) - writeChar(',', buf); - writeJSONString(key, buf, formatSettings()); - writeChar(':', buf); - traverse(value, buf); - } - writeChar('}', buf); - return; - } - if (element.isNull()) - { - writeCString("null", buf); - return; - } - } - - static const FormatSettings & formatSettings() - { - static const FormatSettings the_instance = [] - { - FormatSettings settings; - settings.json.escape_forward_slashes = false; - return settings; - }(); - return the_instance; - } -}; - - -template -class JSONExtractArrayRawImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - return std::make_shared(std::make_shared()); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - static bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - if (!element.isArray()) - return false; - - auto array = element.getArray(); - ColumnArray & col_res = assert_cast(dest); - - for (auto value : array) - JSONExtractRawImpl::insertResultToColumn(col_res.getData(), value, {}); - - col_res.getOffsets().push_back(col_res.getOffsets().back() + array.size()); - return true; - } -}; - - -template -class JSONExtractKeysAndValuesRawImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - DataTypePtr string_type = std::make_unique(); - DataTypePtr tuple_type = std::make_unique(DataTypes{string_type, string_type}); - return std::make_unique(tuple_type); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - if (!element.isObject()) - return false; - - auto object = element.getObject(); - - auto & col_arr = assert_cast(dest); - auto & col_tuple = assert_cast(col_arr.getData()); - auto & col_key = assert_cast(col_tuple.getColumn(0)); - auto & col_value = assert_cast(col_tuple.getColumn(1)); - - for (const auto & [key, value] : object) - { - col_key.insertData(key.data(), key.size()); - JSONExtractRawImpl::insertResultToColumn(col_value, value, {}); - } - - col_arr.getOffsets().push_back(col_arr.getOffsets().back() + object.size()); - return true; - } -}; - -template -class JSONExtractKeysImpl -{ -public: - using Element = typename JSONParser::Element; - - static DataTypePtr getReturnType(const char *, const ColumnsWithTypeAndName &) - { - return std::make_unique(std::make_shared()); - } - - static size_t getNumberOfIndexArguments(const ColumnsWithTypeAndName & arguments) { return arguments.size() - 1; } - - bool insertResultToColumn(IColumn & dest, const Element & element, std::string_view) - { - if (!element.isObject()) - return false; - - auto object = element.getObject(); - - ColumnArray & col_res = assert_cast(dest); - auto & col_key = assert_cast(col_res.getData()); - - for (const auto & [key, value] : object) - { - col_key.insertData(key.data(), key.size()); - } - - col_res.getOffsets().push_back(col_res.getOffsets().back() + object.size()); - return true; - } -}; - -} diff --git a/src/Functions/FunctionsLogical.cpp b/src/Functions/FunctionsLogical.cpp index 2f5ce6deebf..e1814150da6 100644 --- a/src/Functions/FunctionsLogical.cpp +++ b/src/Functions/FunctionsLogical.cpp @@ -29,7 +29,7 @@ REGISTER_FUNCTION(Logical) factory.registerFunction(); factory.registerFunction(); factory.registerFunction(); - factory.registerFunction({}, FunctionFactory::CaseInsensitive); /// Operator NOT(x) can be parsed as a function. + factory.registerFunction({}, FunctionFactory::Case::Insensitive); /// Operator NOT(x) can be parsed as a function. } namespace ErrorCodes diff --git a/src/Functions/FunctionsMultiStringFuzzySearch.h b/src/Functions/FunctionsMultiStringFuzzySearch.h index 18b411e9839..a92a6570279 100644 --- a/src/Functions/FunctionsMultiStringFuzzySearch.h +++ b/src/Functions/FunctionsMultiStringFuzzySearch.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include #include diff --git a/src/Functions/FunctionsMultiStringSearch.h b/src/Functions/FunctionsMultiStringSearch.h index c0ed90aa042..03db2651fd0 100644 --- a/src/Functions/FunctionsMultiStringSearch.h +++ b/src/Functions/FunctionsMultiStringSearch.h @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Functions/FunctionsOpDate.cpp b/src/Functions/FunctionsOpDate.cpp index 7355848f73f..c4b154736e0 100644 --- a/src/Functions/FunctionsOpDate.cpp +++ b/src/Functions/FunctionsOpDate.cpp @@ -99,8 +99,8 @@ using FunctionSubDate = FunctionOpDate; REGISTER_FUNCTION(AddInterval) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/FunctionsRound.cpp b/src/Functions/FunctionsRound.cpp index 059476acb40..d87a9e7ca43 100644 --- a/src/Functions/FunctionsRound.cpp +++ b/src/Functions/FunctionsRound.cpp @@ -7,16 +7,16 @@ namespace DB REGISTER_FUNCTION(Round) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerFunction({}, FunctionFactory::CaseSensitive); - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerFunction({}, FunctionFactory::Case::Sensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); factory.registerFunction(); /// Compatibility aliases. - factory.registerAlias("ceiling", "ceil", FunctionFactory::CaseInsensitive); - factory.registerAlias("truncate", "trunc", FunctionFactory::CaseInsensitive); + factory.registerAlias("ceiling", "ceil", FunctionFactory::Case::Insensitive); + factory.registerAlias("truncate", "trunc", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/FunctionsRound.h b/src/Functions/FunctionsRound.h index 08e257de8ac..6b65a5feaec 100644 --- a/src/Functions/FunctionsRound.h +++ b/src/Functions/FunctionsRound.h @@ -518,66 +518,78 @@ struct Dispatcher template static ColumnPtr apply(const IColumn * value_col, const IColumn * scale_col = nullptr) { - const auto & value_col_typed = checkAndGetColumn>(*value_col); - auto col_res = ColumnVector::create(); - - typename ColumnVector::Container & vec_res = col_res->getData(); - vec_res.resize(value_col_typed.getData().size()); - - if (!vec_res.empty()) + // Non-const value argument: + const auto * value_col_typed = checkAndGetColumn>(value_col); + if (value_col_typed) { - if (scale_col == nullptr || isColumnConst(*scale_col)) - { - auto scale_arg = (scale_col == nullptr) ? 0 : getScaleArg(checkAndGetColumnConst>(scale_col)); - if (scale_arg == 0) - { - size_t scale = 1; - FunctionRoundingImpl::apply(value_col_typed.getData(), scale, vec_res); - } - else if (scale_arg > 0) - { - size_t scale = intExp10(scale_arg); - FunctionRoundingImpl::apply(value_col_typed.getData(), scale, vec_res); - } - else - { - size_t scale = intExp10(-scale_arg); - FunctionRoundingImpl::apply(value_col_typed.getData(), scale, vec_res); - } - } - /// Non-const scale argument: - else if (const auto * scale_col_typed = checkAndGetColumn>(scale_col)) - { - const auto & value_data = value_col_typed.getData(); - const auto & scale_data = scale_col_typed->getData(); - const size_t rows = value_data.size(); + auto col_res = ColumnVector::create(); - for (size_t i = 0; i < rows; ++i) - { - Int64 scale64 = scale_data[i]; - validateScale(scale64); - Scale raw_scale = scale64; + typename ColumnVector::Container & vec_res = col_res->getData(); + vec_res.resize(value_col_typed->getData().size()); - if (raw_scale == 0) + if (!vec_res.empty()) + { + // Const scale argument: + if (scale_col == nullptr || isColumnConst(*scale_col)) + { + auto scale_arg = (scale_col == nullptr) ? 0 : getScaleArg(checkAndGetColumnConst>(scale_col)); + if (scale_arg == 0) { size_t scale = 1; - FunctionRoundingImpl::applyOne(value_data[i], scale, vec_res[i]); + FunctionRoundingImpl::apply(value_col_typed->getData(), scale, vec_res); } - else if (raw_scale > 0) + else if (scale_arg > 0) { - size_t scale = intExp10(raw_scale); - FunctionRoundingImpl::applyOne(value_data[i], scale, vec_res[i]); + size_t scale = intExp10(scale_arg); + FunctionRoundingImpl::apply(value_col_typed->getData(), scale, vec_res); } else { - size_t scale = intExp10(-raw_scale); - FunctionRoundingImpl::applyOne(value_data[i], scale, vec_res[i]); + size_t scale = intExp10(-scale_arg); + FunctionRoundingImpl::apply(value_col_typed->getData(), scale, vec_res); + } + } + /// Non-const scale argument: + else if (const auto * scale_col_typed = checkAndGetColumn>(scale_col)) + { + const auto & value_data = value_col_typed->getData(); + const auto & scale_data = scale_col_typed->getData(); + const size_t rows = value_data.size(); + + for (size_t i = 0; i < rows; ++i) + { + Int64 scale64 = scale_data[i]; + validateScale(scale64); + Scale raw_scale = scale64; + + if (raw_scale == 0) + { + size_t scale = 1; + FunctionRoundingImpl::applyOne(value_data[i], scale, vec_res[i]); + } + else if (raw_scale > 0) + { + size_t scale = intExp10(raw_scale); + FunctionRoundingImpl::applyOne(value_data[i], scale, vec_res[i]); + } + else + { + size_t scale = intExp10(-raw_scale); + FunctionRoundingImpl::applyOne(value_data[i], scale, vec_res[i]); + } } } } + return col_res; } - - return col_res; + // Const value argument: + const auto * value_col_typed_const = checkAndGetColumnConst>(value_col); + if (value_col_typed_const) + { + auto value_col_full = value_col_typed_const->convertToFullColumn(); + return apply(value_col_full.get(), scale_col); + } + return nullptr; } }; @@ -589,38 +601,52 @@ public: template static ColumnPtr apply(const IColumn * value_col, const IColumn * scale_col = nullptr) { - const auto & value_col_typed = checkAndGetColumn>(*value_col); - const typename ColumnDecimal::Container & vec_src = value_col_typed.getData(); - - auto col_res = ColumnDecimal::create(vec_src.size(), value_col_typed.getScale()); - auto & vec_res = col_res->getData(); - - if (!vec_res.empty()) + // Non-const value argument: + const auto * value_col_typed = checkAndGetColumn>(value_col); + if (value_col_typed) { - if (scale_col == nullptr || isColumnConst(*scale_col)) - { - auto scale_arg = scale_col == nullptr ? 0 : getScaleArg(checkAndGetColumnConst>(scale_col)); - DecimalRoundingImpl::apply(value_col_typed.getData(), value_col_typed.getScale(), vec_res, scale_arg); - } - /// Non-const scale argument - else if (const auto * scale_col_typed = checkAndGetColumn>(scale_col)) - { - const auto & scale = scale_col_typed->getData(); - const size_t rows = vec_src.size(); + const typename ColumnDecimal::Container & vec_src = value_col_typed->getData(); - for (size_t i = 0; i < rows; ++i) + auto col_res = ColumnDecimal::create(vec_src.size(), value_col_typed->getScale()); + auto & vec_res = col_res->getData(); + vec_res.resize(vec_src.size()); + + if (!vec_res.empty()) + { + /// Const scale argument: + if (scale_col == nullptr || isColumnConst(*scale_col)) { - Int64 scale64 = scale[i]; - validateScale(scale64); - Scale raw_scale = scale64; + auto scale_arg = scale_col == nullptr ? 0 : getScaleArg(checkAndGetColumnConst>(scale_col)); + DecimalRoundingImpl::apply(vec_src, value_col_typed->getScale(), vec_res, scale_arg); + } + /// Non-const scale argument: + else if (const auto * scale_col_typed = checkAndGetColumn>(scale_col)) + { + const auto & scale = scale_col_typed->getData(); + const size_t rows = vec_src.size(); - DecimalRoundingImpl::applyOne(value_col_typed.getElement(i), value_col_typed.getScale(), - reinterpret_cast::NativeT&>(col_res->getElement(i)), raw_scale); + for (size_t i = 0; i < rows; ++i) + { + Int64 scale64 = scale[i]; + validateScale(scale64); + Scale raw_scale = scale64; + + DecimalRoundingImpl::applyOne(value_col_typed->getElement(i), value_col_typed->getScale(), + reinterpret_cast::NativeT&>(col_res->getElement(i)), raw_scale); + } } } - } - return col_res; + return col_res; + } + // Const value argument: + const auto * value_col_typed_const = checkAndGetColumnConst>(value_col); + if (value_col_typed_const) + { + auto value_col_full = value_col_typed_const->convertToFullColumn(); + return apply(value_col_full.get(), scale_col); + } + return nullptr; } }; @@ -647,7 +673,7 @@ public: FunctionArgumentDescriptors optional_args{ {"N", static_cast(&isNativeInteger), nullptr, "The number of decimal places to round to"}, }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return arguments[0].type; } @@ -671,9 +697,6 @@ public: using ScaleTypes = std::decay_t; using ScaleType = typename ScaleTypes::RightType; - if (isColumnConst(*value_arg.column) && !isColumnConst(*scale_column.column)) - throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Scale column must be const for const data column"); - res = Dispatcher::template apply(value_arg.column.get(), scale_column.column.get()); return true; }; diff --git a/src/Functions/FunctionsStringHashFixedString.cpp b/src/Functions/FunctionsStringHashFixedString.cpp index e3b1b82c92f..01e989a7f2c 100644 --- a/src/Functions/FunctionsStringHashFixedString.cpp +++ b/src/Functions/FunctionsStringHashFixedString.cpp @@ -428,8 +428,7 @@ REGISTER_FUNCTION(HashFixedStrings) It returns a BLAKE3 hash as a byte array with type FixedString(32). )", .examples{{"hash", "SELECT hex(BLAKE3('ABC'))", ""}}, - .categories{"Hash"}}, - FunctionFactory::CaseSensitive); + .categories{"Hash"}}); # endif } #endif diff --git a/src/Functions/FunctionsStringSearch.h b/src/Functions/FunctionsStringSearch.h index 924a66c1130..fba6336ebff 100644 --- a/src/Functions/FunctionsStringSearch.h +++ b/src/Functions/FunctionsStringSearch.h @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Functions/FunctionsTextClassification.h b/src/Functions/FunctionsTextClassification.h index 8e0f236366d..90e8af06ccc 100644 --- a/src/Functions/FunctionsTextClassification.h +++ b/src/Functions/FunctionsTextClassification.h @@ -2,6 +2,7 @@ #include #include +#include #include #include #include diff --git a/src/Functions/IFunction.h b/src/Functions/IFunction.h index a66456cabee..12931b51df2 100644 --- a/src/Functions/IFunction.h +++ b/src/Functions/IFunction.h @@ -230,6 +230,17 @@ public: virtual bool isDeterministicInScopeOfQuery() const { return true; } + /** This is a special flags for functions which return constant value for the server, + * but the result could be different for different servers in distributed query. + * + * This functions can't support constant folding on the initiator, but can on the follower. + * We can't apply some optimizations as well (e.g. can't remove constant result from GROUP BY key). + * So, it is convenient to have a special flag for them. + * + * Examples are: "__getScalar" and every function from serverConstants.cpp + */ + virtual bool isServerConstant() const { return false; } + /** Lets you know if the function is monotonic in a range of values. * This is used to work with the index in a sorted chunk of data. * And allows to use the index not only when it is written, for example `date >= const`, but also, for example, `toMonth(date) >= 11`. @@ -488,6 +499,7 @@ public: virtual bool isInjective(const ColumnsWithTypeAndName & /*sample_columns*/) const { return false; } virtual bool isDeterministic() const { return true; } virtual bool isDeterministicInScopeOfQuery() const { return true; } + virtual bool isServerConstant() const { return false; } virtual bool isStateful() const { return false; } using ShortCircuitSettings = IFunctionBase::ShortCircuitSettings; diff --git a/src/Functions/IFunctionAdaptors.h b/src/Functions/IFunctionAdaptors.h index 04bd03a776e..c9929a083c1 100644 --- a/src/Functions/IFunctionAdaptors.h +++ b/src/Functions/IFunctionAdaptors.h @@ -86,6 +86,8 @@ public: bool isDeterministicInScopeOfQuery() const override { return function->isDeterministicInScopeOfQuery(); } + bool isServerConstant() const override { return function->isServerConstant(); } + bool isShortCircuit(ShortCircuitSettings & settings, size_t number_of_arguments) const override { return function->isShortCircuit(settings, number_of_arguments); } bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & args) const override { return function->isSuitableForShortCircuitArgumentsExecution(args); } diff --git a/src/Functions/JSONArrayLength.cpp b/src/Functions/JSONArrayLength.cpp index 84e87061398..24e93440454 100644 --- a/src/Functions/JSONArrayLength.cpp +++ b/src/Functions/JSONArrayLength.cpp @@ -48,7 +48,7 @@ namespace {"json", static_cast(&isString), nullptr, "String"}, }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(std::make_shared()); } @@ -104,7 +104,7 @@ REGISTER_FUNCTION(JSONArrayLength) .description="Returns the number of elements in the outermost JSON array. The function returns NULL if input JSON string is invalid."}); /// For Spark compatibility. - factory.registerAlias("JSON_ARRAY_LENGTH", "JSONArrayLength", FunctionFactory::CaseInsensitive); + factory.registerAlias("JSON_ARRAY_LENGTH", "JSONArrayLength", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/LowerUpperImpl.h b/src/Functions/LowerUpperImpl.h index 72b3ce1ca34..d463ef96e16 100644 --- a/src/Functions/LowerUpperImpl.h +++ b/src/Functions/LowerUpperImpl.h @@ -8,17 +8,19 @@ namespace DB template struct LowerUpperImpl { - static void vector(const ColumnString::Chars & data, + static void vector( + const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t /*input_rows_count*/) { res_data.resize_exact(data.size()); res_offsets.assign(offsets); array(data.data(), data.data() + data.size(), res_data.data()); } - static void vectorFixed(const ColumnString::Chars & data, size_t /*n*/, ColumnString::Chars & res_data) + static void vectorFixed(const ColumnString::Chars & data, size_t /*n*/, ColumnString::Chars & res_data, size_t /*input_rows_count*/) { res_data.resize_exact(data.size()); array(data.data(), data.data() + data.size(), res_data.data()); diff --git a/src/Functions/LowerUpperUTF8Impl.h b/src/Functions/LowerUpperUTF8Impl.h index eebba7b9d5f..eedabca5b22 100644 --- a/src/Functions/LowerUpperUTF8Impl.h +++ b/src/Functions/LowerUpperUTF8Impl.h @@ -90,7 +90,8 @@ struct LowerUpperUTF8Impl const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { if (data.empty()) return; @@ -98,7 +99,7 @@ struct LowerUpperUTF8Impl bool all_ascii = isAllASCII(data.data(), data.size()); if (all_ascii) { - LowerUpperImpl::vector(data, offsets, res_data, res_offsets); + LowerUpperImpl::vector(data, offsets, res_data, res_offsets, input_rows_count); return; } @@ -107,7 +108,7 @@ struct LowerUpperUTF8Impl array(data.data(), data.data() + data.size(), offsets, res_data.data()); } - static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::BAD_ARGUMENTS, "Functions lowerUTF8 and upperUTF8 cannot work with FixedString argument"); } diff --git a/src/Functions/PerformanceAdaptors.h b/src/Functions/PerformanceAdaptors.h index ef2c788bf43..48daf13ba68 100644 --- a/src/Functions/PerformanceAdaptors.h +++ b/src/Functions/PerformanceAdaptors.h @@ -4,6 +4,7 @@ #include #include +#include #include #include diff --git a/src/Functions/Regexps.h b/src/Functions/Regexps.h index b6bd463212f..b317d786fab 100644 --- a/src/Functions/Regexps.h +++ b/src/Functions/Regexps.h @@ -23,7 +23,11 @@ namespace ProfileEvents { -extern const Event RegexpCreated; + extern const Event RegexpWithMultipleNeedlesCreated; + extern const Event RegexpWithMultipleNeedlesGlobalCacheHit; + extern const Event RegexpWithMultipleNeedlesGlobalCacheMiss; + extern const Event RegexpLocalCacheHit; + extern const Event RegexpLocalCacheMiss; } @@ -72,18 +76,28 @@ public: Bucket & bucket = known_regexps[hasher(pattern) % CACHE_SIZE]; if (bucket.regexp == nullptr) [[unlikely]] + { /// insert new entry + ProfileEvents::increment(ProfileEvents::RegexpLocalCacheMiss); bucket = {pattern, std::make_shared(createRegexp(pattern))}; + } else + { if (pattern != bucket.pattern) + { /// replace existing entry + ProfileEvents::increment(ProfileEvents::RegexpLocalCacheMiss); bucket = {pattern, std::make_shared(createRegexp(pattern))}; + } + else + ProfileEvents::increment(ProfileEvents::RegexpLocalCacheHit); + } return bucket.regexp; } private: - constexpr static size_t CACHE_SIZE = 100; /// collision probability + constexpr static size_t CACHE_SIZE = 1'000; /// collision probability std::hash hasher; struct Bucket @@ -244,7 +258,7 @@ inline Regexps constructRegexps(const std::vector & str_patterns, [[mayb throw Exception(ErrorCodes::BAD_ARGUMENTS, "Pattern '{}' failed with error '{}'", str_patterns[error->expression], String(error->message)); } - ProfileEvents::increment(ProfileEvents::RegexpCreated); + ProfileEvents::increment(ProfileEvents::RegexpWithMultipleNeedlesCreated); /// We allocate the scratch space only once, then copy it across multiple threads with hs_clone_scratch /// function which is faster than allocating scratch space each time in each thread. @@ -322,9 +336,11 @@ inline DeferredConstructedRegexpsPtr getOrSet(const std::vector(str_patterns, edit_distance); }); + ProfileEvents::increment(ProfileEvents::RegexpWithMultipleNeedlesGlobalCacheMiss); bucket = {std::move(str_patterns), edit_distance, deferred_constructed_regexps}; } else + { if (bucket.patterns != str_patterns || bucket.edit_distance != edit_distance) { /// replace existing entry @@ -333,8 +349,12 @@ inline DeferredConstructedRegexpsPtr getOrSet(const std::vector(str_patterns, edit_distance); }); + ProfileEvents::increment(ProfileEvents::RegexpWithMultipleNeedlesGlobalCacheMiss); bucket = {std::move(str_patterns), edit_distance, deferred_constructed_regexps}; } + else + ProfileEvents::increment(ProfileEvents::RegexpWithMultipleNeedlesGlobalCacheHit); + } return bucket.regexps; } diff --git a/src/Functions/ReplaceRegexpImpl.h b/src/Functions/ReplaceRegexpImpl.h index 24a40c45c6e..f5fb08f71d2 100644 --- a/src/Functions/ReplaceRegexpImpl.h +++ b/src/Functions/ReplaceRegexpImpl.h @@ -1,9 +1,12 @@ #pragma once -#include #include +#include #include +#include +#include #include +#include namespace DB { @@ -48,45 +51,75 @@ struct ReplaceRegexpImpl static constexpr int max_captures = 10; - static Instructions createInstructions(std::string_view replacement, int num_captures) + /// The replacement string references must not contain non-existing capturing groups. + static void checkSubstitutions(std::string_view replacement, int num_captures) { - Instructions instructions; - - String literals; for (size_t i = 0; i < replacement.size(); ++i) { if (replacement[i] == '\\' && i + 1 < replacement.size()) { - if (isNumericASCII(replacement[i + 1])) /// Substitution + if (isNumericASCII(replacement[i + 1])) /// substitution + { + int substitution_num = replacement[i + 1] - '0'; + if (substitution_num >= num_captures) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Substitution '\\{}' in replacement argument is invalid, regexp has only {} capturing groups", substitution_num, num_captures - 1); + } + } + } + } + + static Instructions createInstructions(std::string_view replacement, int num_captures) + { + checkSubstitutions(replacement, num_captures); + + Instructions instructions; + + String literals; + literals.reserve(replacement.size()); + + for (size_t i = 0; i < replacement.size(); ++i) + { + if (replacement[i] == '\\' && i + 1 < replacement.size()) + { + if (isNumericASCII(replacement[i + 1])) /// substitution { if (!literals.empty()) { instructions.emplace_back(literals); literals = ""; } - instructions.emplace_back(replacement[i + 1] - '0'); + int substitution_num = replacement[i + 1] - '0'; + instructions.emplace_back(substitution_num); } else - literals += replacement[i + 1]; /// Escaping + literals += replacement[i + 1]; /// escaping ++i; } else - literals += replacement[i]; /// Plain character + literals += replacement[i]; /// plain character } if (!literals.empty()) instructions.emplace_back(literals); - for (const auto & instr : instructions) - if (instr.substitution_num >= num_captures) - throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "Id {} in replacement string is an invalid substitution, regexp has only {} capturing groups", - instr.substitution_num, num_captures - 1); - return instructions; } + static bool canFallbackToStringReplacement(const String & needle, const String & replacement, const re2::RE2 & searcher, int num_captures) + { + if (searcher.NumberOfCapturingGroups()) + return false; + + checkSubstitutions(replacement, num_captures); + + String required_substring; + bool is_trivial; + bool required_substring_is_prefix; + std::vector alternatives; + OptimizedRegularExpression::analyze(needle, required_substring, is_trivial, required_substring_is_prefix, alternatives); + return is_trivial && required_substring_is_prefix && required_substring == needle; + } + static void processString( const char * haystack_data, size_t haystack_length, @@ -124,7 +157,7 @@ struct ReplaceRegexpImpl { std::string_view replacement; if (instr.substitution_num >= 0) - replacement = std::string_view(matches[instr.substitution_num].data(), matches[instr.substitution_num].size()); + replacement = {matches[instr.substitution_num].data(), matches[instr.substitution_num].size()}; else replacement = instr.literal; res_data.resize(res_data.size() + replacement.size()); @@ -179,19 +212,32 @@ struct ReplaceRegexpImpl res_offsets.resize(haystack_size); re2::RE2::Options regexp_options; - /// Don't write error messages to stderr. - regexp_options.set_log_errors(false); + regexp_options.set_log_errors(false); /// don't write error messages to stderr re2::RE2 searcher(needle, regexp_options); - if (!searcher.ok()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "The pattern argument is not a valid re2 pattern: {}", searcher.error()); int num_captures = std::min(searcher.NumberOfCapturingGroups() + 1, max_captures); + /// Try to use non-regexp string replacement. This shortcut is implemented only for const-needles + const-replacement as + /// pattern analysis incurs some cost too. + if (canFallbackToStringReplacement(needle, replacement, searcher, num_captures)) + { + auto convertTrait = [](ReplaceRegexpTraits::Replace first_or_all) + { + switch (first_or_all) + { + case ReplaceRegexpTraits::Replace::First: return ReplaceStringTraits::Replace::First; + case ReplaceRegexpTraits::Replace::All: return ReplaceStringTraits::Replace::All; + } + }; + ReplaceStringImpl::vectorConstantConstant(haystack_data, haystack_offsets, needle, replacement, res_data, res_offsets); + return; + } + Instructions instructions = createInstructions(replacement, num_captures); - /// Cannot perform search for whole columns. Will process each string separately. for (size_t i = 0; i < haystack_size; ++i) { size_t from = i > 0 ? haystack_offsets[i - 1] : 0; @@ -221,10 +267,8 @@ struct ReplaceRegexpImpl res_offsets.resize(haystack_size); re2::RE2::Options regexp_options; - /// Don't write error messages to stderr. - regexp_options.set_log_errors(false); + regexp_options.set_log_errors(false); /// don't write error messages to stderr - /// Cannot perform search for whole columns. Will process each string separately. for (size_t i = 0; i < haystack_size; ++i) { size_t hs_from = i > 0 ? haystack_offsets[i - 1] : 0; @@ -242,6 +286,7 @@ struct ReplaceRegexpImpl re2::RE2 searcher(needle, regexp_options); if (!searcher.ok()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "The pattern argument is not a valid re2 pattern: {}", searcher.error()); + int num_captures = std::min(searcher.NumberOfCapturingGroups() + 1, max_captures); Instructions instructions = createInstructions(replacement, num_captures); @@ -270,17 +315,14 @@ struct ReplaceRegexpImpl res_offsets.resize(haystack_size); re2::RE2::Options regexp_options; - /// Don't write error messages to stderr. - regexp_options.set_log_errors(false); + regexp_options.set_log_errors(false); /// don't write error messages to stderr re2::RE2 searcher(needle, regexp_options); - if (!searcher.ok()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "The pattern argument is not a valid re2 pattern: {}", searcher.error()); int num_captures = std::min(searcher.NumberOfCapturingGroups() + 1, max_captures); - /// Cannot perform search for whole columns. Will process each string separately. for (size_t i = 0; i < haystack_size; ++i) { size_t hs_from = i > 0 ? haystack_offsets[i - 1] : 0; @@ -290,8 +332,9 @@ struct ReplaceRegexpImpl size_t repl_from = i > 0 ? replacement_offsets[i - 1] : 0; const char * repl_data = reinterpret_cast(replacement_data.data() + repl_from); const size_t repl_length = static_cast(replacement_offsets[i] - repl_from - 1); + std::string_view replacement(repl_data, repl_length); - Instructions instructions = createInstructions(std::string_view(repl_data, repl_length), num_captures); + Instructions instructions = createInstructions(replacement, num_captures); processString(hs_data, hs_length, res_data, res_offset, searcher, num_captures, instructions); res_offsets[i] = res_offset; @@ -317,10 +360,8 @@ struct ReplaceRegexpImpl res_offsets.resize(haystack_size); re2::RE2::Options regexp_options; - /// Don't write error messages to stderr. - regexp_options.set_log_errors(false); + regexp_options.set_log_errors(false); /// don't write error messages to stderr - /// Cannot perform search for whole columns. Will process each string separately. for (size_t i = 0; i < haystack_size; ++i) { size_t hs_from = i > 0 ? haystack_offsets[i - 1] : 0; @@ -338,12 +379,14 @@ struct ReplaceRegexpImpl size_t repl_from = i > 0 ? replacement_offsets[i - 1] : 0; const char * repl_data = reinterpret_cast(replacement_data.data() + repl_from); const size_t repl_length = static_cast(replacement_offsets[i] - repl_from - 1); + std::string_view replacement(repl_data, repl_length); re2::RE2 searcher(needle, regexp_options); if (!searcher.ok()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "The pattern argument is not a valid re2 pattern: {}", searcher.error()); + int num_captures = std::min(searcher.NumberOfCapturingGroups() + 1, max_captures); - Instructions instructions = createInstructions(std::string_view(repl_data, repl_length), num_captures); + Instructions instructions = createInstructions(replacement, num_captures); processString(hs_data, hs_length, res_data, res_offset, searcher, num_captures, instructions); res_offsets[i] = res_offset; @@ -367,16 +410,13 @@ struct ReplaceRegexpImpl res_offsets.resize(haystack_size); re2::RE2::Options regexp_options; - /// Don't write error messages to stderr. - regexp_options.set_log_errors(false); + regexp_options.set_log_errors(false); /// don't write error messages to stderr re2::RE2 searcher(needle, regexp_options); - if (!searcher.ok()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "The pattern argument is not a valid re2 pattern: {}", searcher.error()); int num_captures = std::min(searcher.NumberOfCapturingGroups() + 1, max_captures); - Instructions instructions = createInstructions(replacement, num_captures); for (size_t i = 0; i < haystack_size; ++i) diff --git a/src/Functions/StringHelpers.h b/src/Functions/StringHelpers.h index 8f3a87d5d0e..24d8fe86c62 100644 --- a/src/Functions/StringHelpers.h +++ b/src/Functions/StringHelpers.h @@ -62,12 +62,13 @@ using Pos = const char *; template struct ExtractSubstringImpl { - static void vector(const ColumnString::Chars & data, const ColumnString::Offsets & offsets, - ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) + static void vector( + const ColumnString::Chars & data, const ColumnString::Offsets & offsets, + ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - size_t size = offsets.size(); - res_offsets.resize(size); - res_data.reserve(size * Extractor::getReserveLengthForElement()); + res_offsets.resize(input_rows_count); + res_data.reserve(input_rows_count * Extractor::getReserveLengthForElement()); size_t prev_offset = 0; size_t res_offset = 0; @@ -76,7 +77,7 @@ struct ExtractSubstringImpl Pos start; size_t length; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { Extractor::execute(reinterpret_cast(&data[prev_offset]), offsets[i] - prev_offset - 1, start, length); @@ -99,7 +100,7 @@ struct ExtractSubstringImpl res_data.assign(start, length); } - static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Column of type FixedString is not supported by this function"); } @@ -111,12 +112,13 @@ struct ExtractSubstringImpl template struct CutSubstringImpl { - static void vector(const ColumnString::Chars & data, const ColumnString::Offsets & offsets, - ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) + static void vector( + const ColumnString::Chars & data, const ColumnString::Offsets & offsets, + ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets, + size_t input_rows_count) { res_data.reserve(data.size()); - size_t size = offsets.size(); - res_offsets.resize(size); + res_offsets.resize(input_rows_count); size_t prev_offset = 0; size_t res_offset = 0; @@ -125,7 +127,7 @@ struct CutSubstringImpl Pos start; size_t length; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * current = reinterpret_cast(&data[prev_offset]); Extractor::execute(current, offsets[i] - prev_offset - 1, start, length); @@ -154,7 +156,7 @@ struct CutSubstringImpl res_data.append(start + length, data.data() + data.size()); } - static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Column of type FixedString is not supported by this function"); } diff --git a/src/Functions/URL/ExtractFirstSignificantSubdomain.h b/src/Functions/URL/ExtractFirstSignificantSubdomain.h index 0d1b1cac8ef..4316fd2fc3a 100644 --- a/src/Functions/URL/ExtractFirstSignificantSubdomain.h +++ b/src/Functions/URL/ExtractFirstSignificantSubdomain.h @@ -1,8 +1,8 @@ #pragma once #include -#include "domain.h" -#include "tldLookup.h" +#include +#include #include /// TLDType namespace DB diff --git a/src/Functions/URL/URLHierarchy.cpp b/src/Functions/URL/URLHierarchy.cpp index c08f41f06ee..0f565df8172 100644 --- a/src/Functions/URL/URLHierarchy.cpp +++ b/src/Functions/URL/URLHierarchy.cpp @@ -32,7 +32,7 @@ public: {"URL", static_cast(&isString), nullptr, "String"}, }; - validateFunctionArgumentTypes(func, arguments, mandatory_args); + validateFunctionArguments(func, arguments, mandatory_args); } static constexpr auto strings_argument_position = 0uz; diff --git a/src/Functions/URL/URLPathHierarchy.cpp b/src/Functions/URL/URLPathHierarchy.cpp index 7c796116b8d..2cb5995e375 100644 --- a/src/Functions/URL/URLPathHierarchy.cpp +++ b/src/Functions/URL/URLPathHierarchy.cpp @@ -30,7 +30,7 @@ public: {"URL", static_cast(&isString), nullptr, "String"}, }; - validateFunctionArgumentTypes(func, arguments, mandatory_args); + validateFunctionArguments(func, arguments, mandatory_args); } static constexpr auto strings_argument_position = 0uz; diff --git a/src/Functions/URL/cutFragment.cpp b/src/Functions/URL/cutFragment.cpp index 3b99edf61a3..b32c4410190 100644 --- a/src/Functions/URL/cutFragment.cpp +++ b/src/Functions/URL/cutFragment.cpp @@ -1,6 +1,6 @@ #include -#include "fragment.h" #include +#include namespace DB { diff --git a/src/Functions/URL/cutQueryString.cpp b/src/Functions/URL/cutQueryString.cpp index 2886adc2e36..d2fa4a004f2 100644 --- a/src/Functions/URL/cutQueryString.cpp +++ b/src/Functions/URL/cutQueryString.cpp @@ -1,6 +1,6 @@ #include -#include "queryString.h" #include +#include namespace DB { diff --git a/src/Functions/URL/cutQueryStringAndFragment.cpp b/src/Functions/URL/cutQueryStringAndFragment.cpp index 4116b352542..ff427beb277 100644 --- a/src/Functions/URL/cutQueryStringAndFragment.cpp +++ b/src/Functions/URL/cutQueryStringAndFragment.cpp @@ -1,6 +1,6 @@ #include -#include "queryStringAndFragment.h" #include +#include namespace DB { diff --git a/src/Functions/URL/cutToFirstSignificantSubdomain.cpp b/src/Functions/URL/cutToFirstSignificantSubdomain.cpp index 6e64b0b6ab8..454a241c54f 100644 --- a/src/Functions/URL/cutToFirstSignificantSubdomain.cpp +++ b/src/Functions/URL/cutToFirstSignificantSubdomain.cpp @@ -1,6 +1,6 @@ #include #include -#include "ExtractFirstSignificantSubdomain.h" +#include namespace DB diff --git a/src/Functions/URL/cutToFirstSignificantSubdomainCustom.cpp b/src/Functions/URL/cutToFirstSignificantSubdomainCustom.cpp index 77f40e465a6..7612b6ea4af 100644 --- a/src/Functions/URL/cutToFirstSignificantSubdomainCustom.cpp +++ b/src/Functions/URL/cutToFirstSignificantSubdomainCustom.cpp @@ -1,6 +1,6 @@ #include -#include "ExtractFirstSignificantSubdomain.h" -#include "FirstSignificantSubdomainCustomImpl.h" +#include +#include namespace DB { diff --git a/src/Functions/URL/cutWWW.cpp b/src/Functions/URL/cutWWW.cpp index 992d5128440..ab3fae6b094 100644 --- a/src/Functions/URL/cutWWW.cpp +++ b/src/Functions/URL/cutWWW.cpp @@ -1,6 +1,6 @@ #include #include -#include "protocol.h" +#include #include diff --git a/src/Functions/URL/decodeURLComponent.cpp b/src/Functions/URL/decodeURLComponent.cpp index 05e3fbea3fd..bf4aaa6d5e3 100644 --- a/src/Functions/URL/decodeURLComponent.cpp +++ b/src/Functions/URL/decodeURLComponent.cpp @@ -1,7 +1,7 @@ -#include #include #include #include +#include namespace DB @@ -121,8 +121,10 @@ enum URLCodeStrategy template struct CodeURLComponentImpl { - static void vector(const ColumnString::Chars & data, const ColumnString::Offsets & offsets, - ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets) + static void vector( + const ColumnString::Chars & data, const ColumnString::Offsets & offsets, + ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets, + size_t input_rows_count) { if (code_strategy == encode) { @@ -134,13 +136,12 @@ struct CodeURLComponentImpl res_data.resize(data.size()); } - size_t size = offsets.size(); - res_offsets.resize(size); + res_offsets.resize(input_rows_count); size_t prev_offset = 0; size_t res_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * src_data = reinterpret_cast(&data[prev_offset]); size_t src_size = offsets[i] - prev_offset; @@ -165,7 +166,7 @@ struct CodeURLComponentImpl res_data.resize(res_offset); } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Column of type FixedString is not supported by URL functions"); } diff --git a/src/Functions/URL/domain.cpp b/src/Functions/URL/domain.cpp index 443a3323075..68724b57a70 100644 --- a/src/Functions/URL/domain.cpp +++ b/src/Functions/URL/domain.cpp @@ -1,5 +1,4 @@ -#include "domain.h" - +#include #include #include diff --git a/src/Functions/URL/domain.h b/src/Functions/URL/domain.h index 936fb9d5f00..328df76b570 100644 --- a/src/Functions/URL/domain.h +++ b/src/Functions/URL/domain.h @@ -1,9 +1,10 @@ #pragma once -#include "protocol.h" -#include -#include #include +#include +#include + +#include namespace DB { diff --git a/src/Functions/URL/domainWithoutWWW.cpp b/src/Functions/URL/domainWithoutWWW.cpp index f6c8b5c84fc..436aad01b82 100644 --- a/src/Functions/URL/domainWithoutWWW.cpp +++ b/src/Functions/URL/domainWithoutWWW.cpp @@ -1,6 +1,6 @@ #include #include -#include "domain.h" +#include namespace DB { diff --git a/src/Functions/URL/extractURLParameterNames.cpp b/src/Functions/URL/extractURLParameterNames.cpp index 16ace36d39b..b3d51d02162 100644 --- a/src/Functions/URL/extractURLParameterNames.cpp +++ b/src/Functions/URL/extractURLParameterNames.cpp @@ -30,7 +30,7 @@ public: {"URL", static_cast(&isString), nullptr, "String"}, }; - validateFunctionArgumentTypes(func, arguments, mandatory_args); + validateFunctionArguments(func, arguments, mandatory_args); } static constexpr auto strings_argument_position = 0uz; diff --git a/src/Functions/URL/extractURLParameters.cpp b/src/Functions/URL/extractURLParameters.cpp index 43079834872..ce2aadaeede 100644 --- a/src/Functions/URL/extractURLParameters.cpp +++ b/src/Functions/URL/extractURLParameters.cpp @@ -31,7 +31,7 @@ public: {"URL", static_cast(&isString), nullptr, "String"}, }; - validateFunctionArgumentTypes(func, arguments, mandatory_args); + validateFunctionArguments(func, arguments, mandatory_args); } void init(const ColumnsWithTypeAndName & /*arguments*/, bool /*max_substrings_includes_remaining_string*/) {} diff --git a/src/Functions/URL/firstSignificantSubdomain.cpp b/src/Functions/URL/firstSignificantSubdomain.cpp index b04f6d882ef..e929aefbc27 100644 --- a/src/Functions/URL/firstSignificantSubdomain.cpp +++ b/src/Functions/URL/firstSignificantSubdomain.cpp @@ -1,6 +1,6 @@ #include #include -#include "ExtractFirstSignificantSubdomain.h" +#include namespace DB diff --git a/src/Functions/URL/firstSignificantSubdomainCustom.cpp b/src/Functions/URL/firstSignificantSubdomainCustom.cpp index c07aa2b3ac8..95e5142667b 100644 --- a/src/Functions/URL/firstSignificantSubdomainCustom.cpp +++ b/src/Functions/URL/firstSignificantSubdomainCustom.cpp @@ -1,6 +1,6 @@ #include -#include "ExtractFirstSignificantSubdomain.h" -#include "FirstSignificantSubdomainCustomImpl.h" +#include +#include namespace DB diff --git a/src/Functions/URL/fragment.cpp b/src/Functions/URL/fragment.cpp index 262c1a4c7a6..66da5529d84 100644 --- a/src/Functions/URL/fragment.cpp +++ b/src/Functions/URL/fragment.cpp @@ -1,6 +1,6 @@ #include #include -#include "fragment.h" +#include namespace DB { diff --git a/src/Functions/URL/path.cpp b/src/Functions/URL/path.cpp index 8d609f43191..6de8e4fbf95 100644 --- a/src/Functions/URL/path.cpp +++ b/src/Functions/URL/path.cpp @@ -1,7 +1,7 @@ #include #include #include -#include "path.h" +#include #include diff --git a/src/Functions/URL/pathFull.cpp b/src/Functions/URL/pathFull.cpp index 9aacee21fed..deea617eb29 100644 --- a/src/Functions/URL/pathFull.cpp +++ b/src/Functions/URL/pathFull.cpp @@ -1,7 +1,7 @@ #include #include #include -#include "path.h" +#include #include namespace DB diff --git a/src/Functions/URL/port.cpp b/src/Functions/URL/port.cpp index c8f50f10a56..fac46281604 100644 --- a/src/Functions/URL/port.cpp +++ b/src/Functions/URL/port.cpp @@ -5,7 +5,7 @@ #include #include #include -#include "domain.h" +#include namespace DB diff --git a/src/Functions/URL/protocol.cpp b/src/Functions/URL/protocol.cpp index 6bed71207f6..1ac395dc554 100644 --- a/src/Functions/URL/protocol.cpp +++ b/src/Functions/URL/protocol.cpp @@ -1,6 +1,6 @@ #include #include -#include "protocol.h" +#include namespace DB diff --git a/src/Functions/URL/queryString.cpp b/src/Functions/URL/queryString.cpp index 7069ce46b38..f6aaf40fc96 100644 --- a/src/Functions/URL/queryString.cpp +++ b/src/Functions/URL/queryString.cpp @@ -1,6 +1,6 @@ #include #include -#include "queryString.h" +#include namespace DB { diff --git a/src/Functions/URL/queryStringAndFragment.cpp b/src/Functions/URL/queryStringAndFragment.cpp index 367a95acbdc..94f1dfa41e3 100644 --- a/src/Functions/URL/queryStringAndFragment.cpp +++ b/src/Functions/URL/queryStringAndFragment.cpp @@ -1,6 +1,6 @@ #include #include -#include "queryStringAndFragment.h" +#include namespace DB { diff --git a/src/Functions/URL/topLevelDomain.cpp b/src/Functions/URL/topLevelDomain.cpp index 25e9f383f60..b3e88832350 100644 --- a/src/Functions/URL/topLevelDomain.cpp +++ b/src/Functions/URL/topLevelDomain.cpp @@ -1,6 +1,6 @@ #include #include -#include "domain.h" +#include namespace DB { diff --git a/src/Functions/UTCTimestamp.cpp b/src/Functions/UTCTimestamp.cpp index acc34b0a974..bc8e1b28431 100644 --- a/src/Functions/UTCTimestamp.cpp +++ b/src/Functions/UTCTimestamp.cpp @@ -117,8 +117,8 @@ Example: )", .examples{ {"typical", "SELECT UTCTimestamp();", ""}}, - .categories{"Dates and Times"}}, FunctionFactory::CaseInsensitive); - factory.registerAlias("UTC_timestamp", UTCTimestampOverloadResolver::name, FunctionFactory::CaseInsensitive); + .categories{"Dates and Times"}}, FunctionFactory::Case::Insensitive); + factory.registerAlias("UTC_timestamp", UTCTimestampOverloadResolver::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/UTCTimestampTransform.cpp b/src/Functions/UTCTimestampTransform.cpp index 6d301270d8e..36ec520068f 100644 --- a/src/Functions/UTCTimestampTransform.cpp +++ b/src/Functions/UTCTimestampTransform.cpp @@ -144,8 +144,8 @@ REGISTER_FUNCTION(UTCTimestampTransform) { factory.registerFunction(); factory.registerFunction(); - factory.registerAlias("to_utc_timestamp", NameToUTCTimestamp::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("from_utc_timestamp", NameFromUTCTimestamp::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("to_utc_timestamp", NameToUTCTimestamp::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("from_utc_timestamp", NameFromUTCTimestamp::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/UserDefined/ExternalUserDefinedExecutableFunctionsLoader.cpp b/src/Functions/UserDefined/ExternalUserDefinedExecutableFunctionsLoader.cpp index 2c031158c48..a20e1807487 100644 --- a/src/Functions/UserDefined/ExternalUserDefinedExecutableFunctionsLoader.cpp +++ b/src/Functions/UserDefined/ExternalUserDefinedExecutableFunctionsLoader.cpp @@ -2,6 +2,7 @@ #include #include +#include #include @@ -184,7 +185,7 @@ ExternalLoader::LoadableMutablePtr ExternalUserDefinedExecutableFunctionsLoader: pool_size = config.getUInt64(key_in_config + ".pool_size", 16); max_command_execution_time = config.getUInt64(key_in_config + ".max_command_execution_time", 10); - size_t max_execution_time_seconds = static_cast(getContext()->getSettings().max_execution_time.totalSeconds()); + size_t max_execution_time_seconds = static_cast(getContext()->getSettingsRef().max_execution_time.totalSeconds()); if (max_execution_time_seconds != 0 && max_command_execution_time > max_execution_time_seconds) max_command_execution_time = max_execution_time_seconds; } diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp index c15958f81c9..da79e17f18a 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsBackup.cpp @@ -13,6 +13,7 @@ #include #include #include +#include namespace DB diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp index b406cc8d317..4c004d2537c 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsDiskStorage.cpp @@ -9,6 +9,8 @@ #include #include +#include + #include #include #include diff --git a/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp b/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp index 766d63eafb0..01e7e3995fa 100644 --- a/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp +++ b/src/Functions/UserDefined/UserDefinedSQLObjectsZooKeeperStorage.cpp @@ -14,7 +14,7 @@ #include #include #include - +#include namespace DB { diff --git a/src/Functions/abs.cpp b/src/Functions/abs.cpp index 9ac2363f765..742d3b85619 100644 --- a/src/Functions/abs.cpp +++ b/src/Functions/abs.cpp @@ -51,7 +51,7 @@ template <> struct FunctionUnaryArithmeticMonotonicity REGISTER_FUNCTION(Abs) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/acos.cpp b/src/Functions/acos.cpp index bc300ee77fb..39895fed64a 100644 --- a/src/Functions/acos.cpp +++ b/src/Functions/acos.cpp @@ -14,7 +14,7 @@ using FunctionAcos = FunctionMathUnary>; REGISTER_FUNCTION(Acos) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/acosh.cpp b/src/Functions/acosh.cpp index 5b071da9c40..2bab84c77af 100644 --- a/src/Functions/acosh.cpp +++ b/src/Functions/acosh.cpp @@ -5,11 +5,12 @@ namespace DB { namespace { - struct AcoshName - { - static constexpr auto name = "acosh"; - }; - using FunctionAcosh = FunctionMathUnary>; + +struct AcoshName +{ + static constexpr auto name = "acosh"; +}; +using FunctionAcosh = FunctionMathUnary>; } diff --git a/src/Functions/addMicroseconds.cpp b/src/Functions/addMicroseconds.cpp index 0dcd6b4452f..8c0ae06dcd0 100644 --- a/src/Functions/addMicroseconds.cpp +++ b/src/Functions/addMicroseconds.cpp @@ -6,6 +6,7 @@ namespace DB { using FunctionAddMicroseconds = FunctionDateOrDateTimeAddInterval; + REGISTER_FUNCTION(AddMicroseconds) { factory.registerFunction(); diff --git a/src/Functions/addMilliseconds.cpp b/src/Functions/addMilliseconds.cpp index 0e2b696d367..83e1f96ec4b 100644 --- a/src/Functions/addMilliseconds.cpp +++ b/src/Functions/addMilliseconds.cpp @@ -6,6 +6,7 @@ namespace DB { using FunctionAddMilliseconds = FunctionDateOrDateTimeAddInterval; + REGISTER_FUNCTION(AddMilliseconds) { factory.registerFunction(); diff --git a/src/Functions/addNanoseconds.cpp b/src/Functions/addNanoseconds.cpp index 93eadc814d9..8f9a54752b9 100644 --- a/src/Functions/addNanoseconds.cpp +++ b/src/Functions/addNanoseconds.cpp @@ -6,6 +6,7 @@ namespace DB { using FunctionAddNanoseconds = FunctionDateOrDateTimeAddInterval; + REGISTER_FUNCTION(AddNanoseconds) { factory.registerFunction(); diff --git a/src/Functions/aes_encrypt_mysql.cpp b/src/Functions/aes_encrypt_mysql.cpp index fb120151c25..33733f92b27 100644 --- a/src/Functions/aes_encrypt_mysql.cpp +++ b/src/Functions/aes_encrypt_mysql.cpp @@ -7,7 +7,6 @@ namespace DB { - namespace { diff --git a/src/Functions/appendTrailingCharIfAbsent.cpp b/src/Functions/appendTrailingCharIfAbsent.cpp index a5554171aaa..0e57d5c55ce 100644 --- a/src/Functions/appendTrailingCharIfAbsent.cpp +++ b/src/Functions/appendTrailingCharIfAbsent.cpp @@ -57,7 +57,7 @@ private: bool useDefaultImplementationForConstants() const override { return true; } ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1}; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const auto & column = arguments[0].column; const auto & column_char = arguments[1].column; @@ -80,14 +80,13 @@ private: auto & dst_data = col_res->getChars(); auto & dst_offsets = col_res->getOffsets(); - const auto size = src_offsets.size(); - dst_data.resize(src_data.size() + size); - dst_offsets.resize(size); + dst_data.resize(src_data.size() + input_rows_count); + dst_offsets.resize(input_rows_count); ColumnString::Offset src_offset{}; ColumnString::Offset dst_offset{}; - for (const auto i : collections::range(0, size)) + for (size_t i = 0; i < input_rows_count; ++i) { const auto src_length = src_offsets[i] - src_offset; memcpySmallAllowReadWriteOverflow15(&dst_data[dst_offset], &src_data[src_offset], src_length); diff --git a/src/Functions/array/FunctionsMapMiscellaneous.cpp b/src/Functions/array/FunctionsMapMiscellaneous.cpp index 76c1ec18171..c3586a57161 100644 --- a/src/Functions/array/FunctionsMapMiscellaneous.cpp +++ b/src/Functions/array/FunctionsMapMiscellaneous.cpp @@ -51,6 +51,8 @@ public: bool isVariadic() const override { return impl.isVariadic(); } size_t getNumberOfArguments() const override { return impl.getNumberOfArguments(); } + bool useDefaultImplementationForNulls() const override { return impl.useDefaultImplementationForNulls(); } + bool useDefaultImplementationForLowCardinalityColumns() const override { return impl.useDefaultImplementationForLowCardinalityColumns(); } bool useDefaultImplementationForConstants() const override { return impl.useDefaultImplementationForConstants(); } bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo &) const override { return false; } @@ -184,7 +186,7 @@ struct MapToNestedAdapter : public MapAdapterBase struct MapToSubcolumnAdapter { - static_assert(position <= 1); + static_assert(position <= 1, "position of Map subcolumn must be 0 or 1"); static void extractNestedTypes(DataTypes & types) { @@ -357,7 +359,7 @@ struct NameMapValues { static constexpr auto name = "mapValues"; }; using FunctionMapValues = FunctionMapToArrayAdapter, NameMapValues>; struct NameMapContains { static constexpr auto name = "mapContains"; }; -using FunctionMapContains = FunctionMapToArrayAdapter, MapToSubcolumnAdapter, NameMapContains>; +using FunctionMapContains = FunctionMapToArrayAdapter, MapToSubcolumnAdapter, NameMapContains>; struct NameMapFilter { static constexpr auto name = "mapFilter"; }; using FunctionMapFilter = FunctionMapToArrayAdapter, NameMapFilter>; diff --git a/src/Functions/array/array.cpp b/src/Functions/array/array.cpp index 03b51808799..dfe589fb74f 100644 --- a/src/Functions/array/array.cpp +++ b/src/Functions/array/array.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include diff --git a/src/Functions/array/arrayAUC.cpp b/src/Functions/array/arrayAUC.cpp index 499fe4ce7b2..3e2a3bf6863 100644 --- a/src/Functions/array/arrayAUC.cpp +++ b/src/Functions/array/arrayAUC.cpp @@ -103,27 +103,40 @@ private: sorted_labels[i].label = label; } - /// Stable sort is required for for labels to apply in same order if score is equal - std::stable_sort(sorted_labels.begin(), sorted_labels.end(), [](const auto & lhs, const auto & rhs) { return lhs.score > rhs.score; }); + /// Sorting scores in descending order to traverse the ROC curve from left to right + std::sort(sorted_labels.begin(), sorted_labels.end(), [](const auto & lhs, const auto & rhs) { return lhs.score > rhs.score; }); /// We will first calculate non-normalized area. - size_t area = 0; - size_t count_positive = 0; + Float64 area = 0.0; + Float64 prev_score = sorted_labels[0].score; + size_t prev_fp = 0, prev_tp = 0; + size_t curr_fp = 0, curr_tp = 0; for (size_t i = 0; i < size; ++i) { + // Only increment the area when the score changes + if (sorted_labels[i].score != prev_score) + { + area += (curr_fp - prev_fp) * (curr_tp + prev_tp) / 2.0; // Trapezoidal area under curve (might degenerate to zero or to a rectangle) + prev_fp = curr_fp; + prev_tp = curr_tp; + prev_score = sorted_labels[i].score; + } + if (sorted_labels[i].label) - ++count_positive; /// The curve moves one step up. No area increase. + curr_tp += 1; /// The curve moves one step up. else - area += count_positive; /// The curve moves one step right. Area is increased by 1 * height = count_positive. + curr_fp += 1; /// The curve moves one step right. } - /// Then divide the area to the area of rectangle. + area += (curr_fp - prev_fp) * (curr_tp + prev_tp) / 2.0; - if (count_positive == 0 || count_positive == size) + /// Then normalize it dividing by the area to the area of rectangle. + + if (curr_tp == 0 || curr_tp == size) return std::numeric_limits::quiet_NaN(); - return static_cast(area) / count_positive / (size - count_positive); + return area / curr_tp / (size - curr_tp); } static void vector( diff --git a/src/Functions/array/arrayFlatten.cpp b/src/Functions/array/arrayFlatten.cpp index d4eb8eebeee..553ad82bd53 100644 --- a/src/Functions/array/arrayFlatten.cpp +++ b/src/Functions/array/arrayFlatten.cpp @@ -123,7 +123,7 @@ private: REGISTER_FUNCTION(ArrayFlatten) { factory.registerFunction(); - factory.registerAlias("flatten", "arrayFlatten", FunctionFactory::CaseInsensitive); + factory.registerAlias("flatten", "arrayFlatten", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/array/arrayIndex.h b/src/Functions/array/arrayIndex.h index fa9b3dc92dd..0782f109187 100644 --- a/src/Functions/array/arrayIndex.h +++ b/src/Functions/array/arrayIndex.h @@ -1,4 +1,5 @@ #pragma once +#include #include #include #include @@ -13,6 +14,8 @@ #include #include #include +#include "Common/Logger.h" +#include "Common/logger_useful.h" #include #include #include @@ -28,6 +31,7 @@ namespace ErrorCodes { extern const int ILLEGAL_COLUMN; extern const int ILLEGAL_TYPE_OF_ARGUMENT; + extern const int LOGICAL_ERROR; } using NullMap = PaddedPODArray; @@ -424,31 +428,21 @@ public: ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t /*input_rows_count*/) const override { - if constexpr (std::is_same_v) + if (auto res = executeMap(arguments, result_type)) + return res; + + if (auto res = executeArrayLowCardinality(arguments)) + return res; + + auto new_arguments = arguments; + + for (auto & argument : new_arguments) { - if (isMap(arguments[0].type)) - { - auto non_const_map_column = arguments[0].column->convertToFullColumnIfConst(); - - const auto & map_column = assert_cast(*non_const_map_column); - const auto & map_array_column = map_column.getNestedColumn(); - auto offsets = map_array_column.getOffsetsPtr(); - auto keys = map_column.getNestedData().getColumnPtr(0); - auto array_column = ColumnArray::create(keys, offsets); - - const auto & type_map = assert_cast(*arguments[0].type); - auto array_type = std::make_shared(type_map.getKeyType()); - - auto arguments_copy = arguments; - arguments_copy[0].column = std::move(array_column); - arguments_copy[0].type = std::move(array_type); - arguments_copy[0].name = arguments[0].name; - - return executeArrayImpl(arguments_copy, result_type); - } + argument.column = recursiveRemoveLowCardinality(argument.column); + argument.type = recursiveRemoveLowCardinality(argument.type); } - return executeArrayImpl(arguments, result_type); + return executeArrayImpl(new_arguments, result_type); } private: @@ -458,18 +452,6 @@ private: using NullMaps = std::pair; - struct ExecutionData - { - const IColumn& left; - const IColumn& right; - const ColumnArray::Offsets& offsets; - ColumnPtr result_column; - NullMaps maps; - ResultColumnPtr result { ResultColumnType::create() }; - - void moveResult() { result_column = std::move(result); } - }; - static bool allowArguments(const DataTypePtr & inner_type, const DataTypePtr & arg) { auto inner_type_decayed = removeNullable(removeLowCardinality(inner_type)); @@ -574,23 +556,13 @@ private: } } -#define INTEGRAL_TPL_PACK UInt8, UInt16, UInt32, UInt64, Int8, Int16, Int32, Int64, Float32, Float64 +#define INTEGRAL_PACK UInt8, UInt16, UInt32, UInt64, Int8, Int16, Int32, Int64, Float32, Float64 ColumnPtr executeOnNonNullable(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type) const { - if (const auto* const left_arr = checkAndGetColumn(arguments[0].column.get())) - { - if (checkAndGetColumn(&left_arr->getData())) - { - if (auto res = executeLowCardinality(arguments)) - return res; - - throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal internal type of first argument of function {}", getName()); - } - } - ColumnPtr res; - if (!((res = executeIntegral(arguments)) + if (!((res = executeNothing(arguments)) + || (res = executeIntegral(arguments)) || (res = executeConst(arguments, result_type)) || (res = executeString(arguments)) || (res = executeGeneric(arguments)))) @@ -599,6 +571,8 @@ private: return res; } +#undef INTEGRAL_PACK + /** * The Array's internal data type may be quite tricky (containing a Nullable type somewhere). To process the * Nullable types correctly, for each data type specialisation we provide two null maps (one for the data and one @@ -627,6 +601,14 @@ private: return {null_map_data, null_map_item}; } + struct ExecutionData + { + const IColumn & left; + const IColumn & right; + const ColumnArray::Offsets & offsets; + NullMaps null_maps; + }; + /** * Given a variadic pack #Integral, apply executeIntegralExpanded with such parameters: * Integral s = {s1, s2, ...} @@ -635,39 +617,33 @@ private: template static ColumnPtr executeIntegral(const ColumnsWithTypeAndName & arguments) { - const ColumnArray * const left = checkAndGetColumn(arguments[0].column.get()); - - if (!left) + const auto * array = checkAndGetColumn(arguments[0].column.get()); + if (!array) return nullptr; - const ColumnPtr right_converted_ptr = arguments[1].column->convertToFullColumnIfLowCardinality(); - const IColumn& right = *right_converted_ptr.get(); - - ExecutionData data = { - left->getData(), - right, - left->getOffsets(), - nullptr, - getNullMaps(arguments) + ExecutionData data + { + .left = array->getData(), + .right = *arguments[1].column, + .offsets = array->getOffsets(), + .null_maps = getNullMaps(arguments), }; - if (executeIntegral(data)) - return data.result_column; - - return nullptr; + auto result = ResultColumnType::create(); + return executeIntegral(data, *result) ? std::move(result) : nullptr; } template - static bool executeIntegral(ExecutionData& data) + static bool executeIntegral(const ExecutionData & data, ResultColumnType & result) { - return (executeIntegralExpanded(data) || ...); + return (executeIntegralExpanded(data, result) || ...); } /// Invoke executeIntegralImpl with such parameters: (A, other1), (A, other2), ... template - static bool executeIntegralExpanded(ExecutionData& data) + static bool executeIntegralExpanded(const ExecutionData & data, ResultColumnType & result) { - return (executeIntegralImpl(data) || ...); + return (executeIntegralImpl(data, result) || ...); } /** @@ -676,40 +652,31 @@ private: * so we have to check all possible variants for #Initial and #Resulting types. */ template - static bool executeIntegralImpl(ExecutionData& data) + static bool executeIntegralImpl(const ExecutionData & data, ResultColumnType & result) { - const ColumnVector * col_nested = checkAndGetColumn>(&data.left); - - if (!col_nested) + const auto * left_typed = checkAndGetColumn>(&data.left); + if (!left_typed) return false; - const auto [null_map_data, null_map_item] = data.maps; - - if (data.right.onlyNull()) - Impl::Null::process( - data.offsets, - data.result->getData(), - null_map_data); - else if (const auto item_arg_const = checkAndGetColumnConst>(&data.right)) + if (const auto * item_arg_const = checkAndGetColumnConst>(&data.right)) Impl::Main::vector( - col_nested->getData(), + left_typed->getData(), data.offsets, item_arg_const->template getValue(), - data.result->getData(), - null_map_data, + result.getData(), + data.null_maps.first, nullptr); - else if (const auto item_arg_vector = checkAndGetColumn>(&data.right)) + else if (const auto * item_arg_vector = checkAndGetColumn>(&data.right)) Impl::Main::vector( - col_nested->getData(), + left_typed->getData(), data.offsets, item_arg_vector->getData(), - data.result->getData(), - null_map_data, - null_map_item); + result.getData(), + data.null_maps.first, + data.null_maps.second); else return false; - data.moveResult(); return true; } @@ -724,227 +691,161 @@ private: * * Tips and tricks tried can be found at https://github.com/ClickHouse/ClickHouse/pull/12550 . */ - static ColumnPtr executeLowCardinality(const ColumnsWithTypeAndName & arguments) + static ColumnPtr executeArrayLowCardinality(const ColumnsWithTypeAndName & arguments) { - const ColumnArray * const col_array = checkAndGetColumn(arguments[0].column.get()); + const auto * col_array = checkAndGetColumn(arguments[0].column.get()); + const auto * col_array_const = checkAndGetColumnConstData(arguments[0].column.get()); - if (!col_array) + if (!col_array && !col_array_const) return nullptr; - const ColumnLowCardinality * const col_lc = checkAndGetColumn(&col_array->getData()); + if (col_array_const) + col_array = col_array_const; - if (!col_lc) + const auto * left_lc = checkAndGetColumn(&col_array->getData()); + if (!left_lc) return nullptr; - const auto [null_map_data, null_map_item] = getNullMaps(arguments); + const auto * right_const = checkAndGetColumn(arguments[1].column.get()); + if (!right_const) + return nullptr; - if (const ColumnConst * col_arg_const = checkAndGetColumn(&*arguments[1].column)) + const auto & array_type = assert_cast(*arguments[0].type); + const auto target_type = recursiveRemoveLowCardinality(array_type.getNestedType()); + auto right = recursiveRemoveLowCardinality(right_const->getDataColumnPtr()); + + UInt64 index = 0; + UInt64 left_size = arguments[0].column->size(); + ResultColumnPtr col_result = ResultColumnType::create(); + + if (!right->isNullAt(0)) { - const IColumnUnique & col_lc_dict = col_lc->getDictionary(); + auto right_type = recursiveRemoveLowCardinality(arguments[1].type); + right = castColumn({right, right_type, ""}, target_type); - const DataTypeArray * const array_type = checkAndGetDataType(arguments[0].type.get()); - const DataTypePtr target_type_ptr = recursiveRemoveLowCardinality(array_type->getNestedType()); + if (right->isNullable()) + right = checkAndGetColumn(*right).getNestedColumnPtr(); - ColumnPtr col_arg_cloned = castColumn( - {col_arg_const->getDataColumnPtr(), arguments[1].type, arguments[1].name}, target_type_ptr); + StringRef elem = right->getDataAt(0); + const auto & left_dict = left_lc->getDictionary(); - ResultColumnPtr col_result = ResultColumnType::create(); - UInt64 index = 0; - - if (!col_arg_cloned->isNullAt(0)) + if (std::optional maybe_index = left_dict.getOrFindValueIndex(elem); maybe_index) { - if (col_arg_cloned->isNullable()) - col_arg_cloned = checkAndGetColumn(*col_arg_cloned).getNestedColumnPtr(); - - StringRef elem = col_arg_cloned->getDataAt(0); - - if (std::optional maybe_index = col_lc_dict.getOrFindValueIndex(elem); maybe_index) - { - index = *maybe_index; - } - else - { - const size_t offsets_size = col_array->getOffsets().size(); - auto & data = col_result->getData(); - - data.resize_fill(offsets_size); - - return col_result; - } + index = *maybe_index; } - - Impl::Main::vector( - col_lc->getIndexes(), - col_array->getOffsets(), - index, /** Assuming LowCardinality has index of NULL always as zero. */ - col_result->getData(), - null_map_data, - null_map_item); - - return col_result; - } - else if (col_lc->nestedIsNullable()) // LowCardinality(Nullable(T)) and U - { - const ColumnPtr left_casted = col_lc->convertToFullColumnIfLowCardinality(); // Nullable(T) - const ColumnNullable & left_nullable = checkAndGetColumn(*left_casted); - - const NullMap * const null_map_left_casted = &left_nullable.getNullMapColumn().getData(); - - const IColumn & left_ptr = left_nullable.getNestedColumn(); - - const ColumnPtr right_casted = arguments[1].column->convertToFullColumnIfLowCardinality(); - const ColumnNullable * const right_nullable = checkAndGetColumn(right_casted.get()); - - const NullMap * const null_map_right_casted = right_nullable - ? &right_nullable->getNullMapColumn().getData() - : null_map_item; - - const IColumn& right_ptr = right_nullable - ? right_nullable->getNestedColumn() - : *right_casted.get(); - - ExecutionData data = + else { - left_ptr, right_ptr, - col_array->getOffsets(), - nullptr, - {null_map_left_casted, null_map_right_casted}}; + col_result->getData().resize_fill(col_array->size()); - if (dispatchConvertedLowCardinalityColumns(data)) - return data.result_column; + if (col_array_const) + return ColumnConst::create(std::move(col_result), left_size); + + return col_result; + } } - else // LowCardinality(T) and U, T not Nullable - { - if (arguments[1].column->isNullable()) - return nullptr; - - if (const auto* const arg_lc = checkAndGetColumn(arguments[1].column.get()); - arg_lc && arg_lc->isNullable()) - return nullptr; - - // LowCardinality(T) and U (possibly LowCardinality(V)) - - const ColumnPtr left_casted = col_lc->convertToFullColumnIfLowCardinality(); - const ColumnPtr right_casted = arguments[1].column->convertToFullColumnIfLowCardinality(); - - ExecutionData data = - { - *left_casted.get(), *right_casted.get(), col_array->getOffsets(), - nullptr, {null_map_data, null_map_item} - }; - - if (dispatchConvertedLowCardinalityColumns(data)) - return data.result_column; - } - - return nullptr; - } - - static bool dispatchConvertedLowCardinalityColumns(ExecutionData & data) - { - if (data.left.isNumeric() && data.right.isNumeric()) // ColumnArrays - return executeIntegral(data); - - if (checkAndGetColumn(&data.left)) - return executeStringImpl(data); Impl::Main::vector( - data.left, - data.offsets, data.right, - data.result->getData(), - data.maps.first, data.maps.second); + left_lc->getIndexes(), + col_array->getOffsets(), + index, /** Assuming LowCardinality has index of NULL always as zero. */ + col_result->getData(), + nullptr, + nullptr); - data.moveResult(); - return true; + if (col_array_const) + return ColumnConst::create(std::move(col_result), left_size); + + return col_result; } -#undef INTEGRAL_TPL_PACK + ColumnPtr executeMap(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type) const + { + if constexpr (!std::is_same_v) + return nullptr; + + if (!isMap(arguments[0].type)) + return nullptr; + + auto non_const_map_column = arguments[0].column->convertToFullColumnIfConst(); + + const auto & map_column = assert_cast(*non_const_map_column); + const auto & map_array_column = map_column.getNestedColumn(); + auto offsets = map_array_column.getOffsetsPtr(); + auto keys = map_column.getNestedData().getColumnPtr(0); + auto array_column = ColumnArray::create(keys, offsets); + + const auto & type_map = assert_cast(*arguments[0].type); + auto array_type = std::make_shared(type_map.getKeyType()); + + auto arguments_copy = arguments; + arguments_copy[0].column = std::move(array_column); + arguments_copy[0].type = std::move(array_type); + arguments_copy[0].name = arguments[0].name; + + return executeArrayImpl(arguments_copy, result_type); + } static ColumnPtr executeString(const ColumnsWithTypeAndName & arguments) { - const ColumnArray * array = checkAndGetColumn(arguments[0].column.get()); - + const auto * array = checkAndGetColumn(arguments[0].column.get()); if (!array) return nullptr; - const ColumnString * left = checkAndGetColumn(&array->getData()); - + const auto * left = checkAndGetColumn(&array->getData()); if (!left) return nullptr; - const ColumnPtr right_ptr = arguments[1].column->convertToFullColumnIfLowCardinality(); - const IColumn & right = *right_ptr.get(); + const auto & right = *arguments[1].column; + const auto [null_map_data, null_map_item] = getNullMaps(arguments); - ExecutionData data = { - *left, right, array->getOffsets(), - nullptr, getNullMaps(arguments), - std::move(ResultColumnType::create()) - }; + auto result = ResultColumnType::create(); - if (executeStringImpl(data)) - return data.result_column; - - return nullptr; - } - - static bool executeStringImpl(ExecutionData& data) - { - const auto [null_map_data, null_map_item] = data.maps; - const ColumnString& left = *typeid_cast(&data.left); - - if (data.right.onlyNull()) - Impl::Null::process( - data.offsets, - data.result->getData(), - null_map_data); - else if (const auto *const item_arg_const = checkAndGetColumnConstStringOrFixedString(&data.right)) + if (const auto * item_arg_const = checkAndGetColumnConstStringOrFixedString(&right)) { - const ColumnString * item_const_string = - checkAndGetColumn(&item_arg_const->getDataColumn()); - - const ColumnFixedString * item_const_fixedstring = - checkAndGetColumn(&item_arg_const->getDataColumn()); + const auto * item_const_string = checkAndGetColumn(&item_arg_const->getDataColumn()); + const auto * item_const_fixedstring = checkAndGetColumn(&item_arg_const->getDataColumn()); if (item_const_string) Impl::String::process( - left.getChars(), - data.offsets, - left.getOffsets(), + left->getChars(), + array->getOffsets(), + left->getOffsets(), item_const_string->getChars(), item_const_string->getDataAt(0).size, - data.result->getData(), + result->getData(), null_map_data, null_map_item); else if (item_const_fixedstring) Impl::String::process( - left.getChars(), - data.offsets, - left.getOffsets(), + left->getChars(), + array->getOffsets(), + left->getOffsets(), item_const_fixedstring->getChars(), item_const_fixedstring->getN(), - data.result->getData(), + result->getData(), null_map_data, null_map_item); else - throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Logical error: ColumnConst contains not String nor FixedString column"); + throw Exception(ErrorCodes::LOGICAL_ERROR, "ColumnConst contains not String nor FixedString column"); } - else if (const auto *const item_arg_vector = checkAndGetColumn(&data.right)) + else if (const auto * item_arg_vector = checkAndGetColumn(&right)) { Impl::String::process( - left.getChars(), - data.offsets, - left.getOffsets(), + left->getChars(), + array->getOffsets(), + left->getOffsets(), item_arg_vector->getChars(), item_arg_vector->getOffsets(), - data.result->getData(), + result->getData(), null_map_data, null_map_item); } else - return false; + { + return nullptr; + } - data.moveResult(); - return true; + return result; } static ColumnPtr executeConst(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type) @@ -955,9 +856,7 @@ private: return nullptr; Array arr = col_array->getValue(); - - const ColumnPtr right_ptr = arguments[1].column->convertToFullColumnIfLowCardinality(); - const IColumn * item_arg = right_ptr.get(); + const IColumn * item_arg = arguments[1].column.get(); if (isColumnConst(*item_arg)) { @@ -1026,48 +925,59 @@ private: } } + static ColumnPtr executeNothing(const ColumnsWithTypeAndName & arguments) + { + const auto * array = checkAndGetColumn(arguments[0].column.get()); + if (!array) + return nullptr; + + if (arguments[1].column->onlyNull()) + { + auto result = ResultColumnType::create(); + Impl::Null::process(array->getOffsets(), result->getData(), getNullMaps(arguments).first); + return result; + } + + return nullptr; + } + static ColumnPtr executeGeneric(const ColumnsWithTypeAndName & arguments) { - const ColumnArray * col = checkAndGetColumn(arguments[0].column.get()); - - if (!col) + const auto * col_array = checkAndGetColumn(arguments[0].column.get()); + if (!col_array) return nullptr; DataTypePtr array_elements_type = assert_cast(*arguments[0].type).getNestedType(); const DataTypePtr & index_type = arguments[1].type; - DataTypePtr common_type = getLeastSupertype(DataTypes{array_elements_type, index_type}); - - ColumnPtr col_nested = castColumn({ col->getDataPtr(), array_elements_type, "" }, common_type); - - const ColumnPtr right_ptr = arguments[1].column->convertToFullColumnIfLowCardinality(); - ColumnPtr item_arg = castColumn({ right_ptr, removeLowCardinality(index_type), "" }, common_type); + DataTypePtr common_type = getLeastSupertype(DataTypes{array_elements_type, arguments[1].type}); + ColumnPtr col_nested = castColumn({ col_array->getDataPtr(), array_elements_type, "" }, common_type); + ColumnPtr item_arg = castColumn({ arguments[1].column, removeLowCardinality(index_type), "" }, common_type); auto col_res = ResultColumnType::create(); auto [null_map_data, null_map_item] = getNullMaps(arguments); - if (item_arg->onlyNull()) - Impl::Null::process( - col->getOffsets(), - col_res->getData(), - null_map_data); - else if (isColumnConst(*item_arg)) + if (const auto * item_arg_const = checkAndGetColumn(item_arg.get())) + { Impl::Main::vector( *col_nested, - col->getOffsets(), - typeid_cast(*item_arg).getDataColumn(), + col_array->getOffsets(), + item_arg_const->getDataColumn(), col_res->getData(), /// TODO This is wrong. null_map_data, nullptr); + } else + { Impl::Main::vector( *col_nested, - col->getOffsets(), + col_array->getOffsets(), *item_arg, col_res->getData(), null_map_data, null_map_item); + } return col_res; } diff --git a/src/Functions/array/arrayJaccardIndex.cpp b/src/Functions/array/arrayJaccardIndex.cpp index 87f3390ac73..7db20667888 100644 --- a/src/Functions/array/arrayJaccardIndex.cpp +++ b/src/Functions/array/arrayJaccardIndex.cpp @@ -87,7 +87,7 @@ public: {"array_1", static_cast(&isArray), nullptr, "Array"}, {"array_2", static_cast(&isArray), nullptr, "Array"}, }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared>(); } diff --git a/src/Functions/array/arrayRandomSample.cpp b/src/Functions/array/arrayRandomSample.cpp index b08a73b93f3..6e176b6e33d 100644 --- a/src/Functions/array/arrayRandomSample.cpp +++ b/src/Functions/array/arrayRandomSample.cpp @@ -39,7 +39,7 @@ public: {"array", static_cast(&isArray), nullptr, "Array"}, {"samples", static_cast(&isUInt), isColumnConst, "const UInt*"}, }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); // Return an array with the same nested type as the input array const DataTypePtr & array_type = arguments[0].type; diff --git a/src/Functions/array/arrayShingles.cpp b/src/Functions/array/arrayShingles.cpp index 8932482c69c..7c97d8136fb 100644 --- a/src/Functions/array/arrayShingles.cpp +++ b/src/Functions/array/arrayShingles.cpp @@ -31,7 +31,7 @@ public: {"array", static_cast(&isArray), nullptr, "Array"}, {"length", static_cast(&isInteger), nullptr, "Integer"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); const DataTypeArray * array_type = checkAndGetDataType(arguments[0].type.get()); return std::make_shared(std::make_shared(array_type->getNestedType())); diff --git a/src/Functions/array/arrayShuffle.cpp b/src/Functions/array/arrayShuffle.cpp index 10cb51d27d2..fa17aa46464 100644 --- a/src/Functions/array/arrayShuffle.cpp +++ b/src/Functions/array/arrayShuffle.cpp @@ -196,7 +196,7 @@ It is possible to override the seed to produce stable results: {"explicit_seed", "SELECT arrayShuffle([1, 2, 3, 4], 41)", ""}, {"materialize", "SELECT arrayShuffle(materialize([1, 2, 3]), 42), arrayShuffle([1, 2, 3], 42) FROM numbers(10)", ""}}, .categories{"Array"}}, - FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); factory.registerFunction>( FunctionDocumentation{ @@ -224,7 +224,7 @@ It is possible to override the seed to produce stable results: {"materialize", "SELECT arrayPartialShuffle(materialize([1, 2, 3, 4]), 2, 42), arrayPartialShuffle([1, 2, 3], 2, 42) FROM numbers(10)", ""}}, .categories{"Array"}}, - FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/array/length.cpp b/src/Functions/array/length.cpp index 91a5e5fdec2..d81c071b55e 100644 --- a/src/Functions/array/length.cpp +++ b/src/Functions/array/length.cpp @@ -100,8 +100,8 @@ It is ok to have ASCII NUL bytes in strings, and they will be counted as well. }, .categories{"String", "Array"} }, - FunctionFactory::CaseInsensitive); - factory.registerAlias("OCTET_LENGTH", "length", FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); + factory.registerAlias("OCTET_LENGTH", "length", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/array/range.cpp b/src/Functions/array/range.cpp index 1ac698a2745..b8b03ab5e74 100644 --- a/src/Functions/array/range.cpp +++ b/src/Functions/array/range.cpp @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Functions/arrayStringConcat.cpp b/src/Functions/arrayStringConcat.cpp index 421408c01f2..12bab410fec 100644 --- a/src/Functions/arrayStringConcat.cpp +++ b/src/Functions/arrayStringConcat.cpp @@ -159,7 +159,7 @@ public: {"separator", static_cast(&isString), isColumnConst, "const String"}, }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(); } diff --git a/src/Functions/ascii.cpp b/src/Functions/ascii.cpp index b43c3221391..0d50e5d203b 100644 --- a/src/Functions/ascii.cpp +++ b/src/Functions/ascii.cpp @@ -45,9 +45,7 @@ struct AsciiImpl size_t size = data.size() / n; for (size_t i = 0; i < size; ++i) - { res[i] = doAscii(data, i * n, n); - } } [[noreturn]] static void array(const ColumnString::Offsets & /*offsets*/, PaddedPODArray & /*res*/) @@ -90,7 +88,7 @@ If s is empty, the result is 0. If the first character is not an ASCII character )", .examples{{"ascii", "SELECT ascii('234')", ""}}, .categories{"String"} - }, FunctionFactory::CaseInsensitive); + }, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/asin.cpp b/src/Functions/asin.cpp index 3049b025d5e..85faf8c275d 100644 --- a/src/Functions/asin.cpp +++ b/src/Functions/asin.cpp @@ -41,7 +41,7 @@ For more details, see [https://en.wikipedia.org/wiki/Inverse_trigonometric_funct {"nan", "SELECT asin(1.1), asin(-2), asin(inf), asin(nan)", ""}}, .categories{"Mathematical", "Trigonometric"} }, - FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/asinh.cpp b/src/Functions/asinh.cpp index 6af832ae07c..b5e3626148f 100644 --- a/src/Functions/asinh.cpp +++ b/src/Functions/asinh.cpp @@ -5,11 +5,12 @@ namespace DB { namespace { - struct AsinhName - { - static constexpr auto name = "asinh"; - }; - using FunctionAsinh = FunctionMathUnary>; + +struct AsinhName +{ + static constexpr auto name = "asinh"; +}; +using FunctionAsinh = FunctionMathUnary>; } diff --git a/src/Functions/atan.cpp b/src/Functions/atan.cpp index 32a0f06db8a..3f74c510487 100644 --- a/src/Functions/atan.cpp +++ b/src/Functions/atan.cpp @@ -14,7 +14,7 @@ using FunctionAtan = FunctionMathUnary>; REGISTER_FUNCTION(Atan) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/atan2.cpp b/src/Functions/atan2.cpp index 7be177f6dfb..218f4c5406f 100644 --- a/src/Functions/atan2.cpp +++ b/src/Functions/atan2.cpp @@ -5,17 +5,18 @@ namespace DB { namespace { - struct Atan2Name - { - static constexpr auto name = "atan2"; - }; - using FunctionAtan2 = FunctionMathBinaryFloat64>; + +struct Atan2Name +{ + static constexpr auto name = "atan2"; +}; +using FunctionAtan2 = FunctionMathBinaryFloat64>; } REGISTER_FUNCTION(Atan2) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/atanh.cpp b/src/Functions/atanh.cpp index fab25414725..a36f5bcbcf0 100644 --- a/src/Functions/atanh.cpp +++ b/src/Functions/atanh.cpp @@ -5,11 +5,12 @@ namespace DB { namespace { - struct AtanhName - { - static constexpr auto name = "atanh"; - }; - using FunctionAtanh = FunctionMathUnary>; + +struct AtanhName +{ + static constexpr auto name = "atanh"; +}; +using FunctionAtanh = FunctionMathUnary>; } diff --git a/src/Functions/base58Encode.cpp b/src/Functions/base58Encode.cpp index cf790ebddab..3ae2fb12c5e 100644 --- a/src/Functions/base58Encode.cpp +++ b/src/Functions/base58Encode.cpp @@ -3,8 +3,10 @@ namespace DB { + REGISTER_FUNCTION(Base58Encode) { factory.registerFunction>(); } + } diff --git a/src/Functions/base64Decode.cpp b/src/Functions/base64Decode.cpp index 50278c4b0b2..349475af3f0 100644 --- a/src/Functions/base64Decode.cpp +++ b/src/Functions/base64Decode.cpp @@ -5,6 +5,7 @@ namespace DB { + REGISTER_FUNCTION(Base64Decode) { FunctionDocumentation::Description description = R"(Accepts a String and decodes it from base64, according to RFC 4648 (https://datatracker.ietf.org/doc/html/rfc4648#section-4). Throws an exception in case of an error. Alias: FROM_BASE64.)"; @@ -17,8 +18,9 @@ REGISTER_FUNCTION(Base64Decode) factory.registerFunction>>({description, syntax, arguments, returned_value, examples, categories}); /// MySQL compatibility alias. - factory.registerAlias("FROM_BASE64", "base64Decode", FunctionFactory::CaseInsensitive); + factory.registerAlias("FROM_BASE64", "base64Decode", FunctionFactory::Case::Insensitive); } + } #endif diff --git a/src/Functions/base64Encode.cpp b/src/Functions/base64Encode.cpp index d6e63c38a4c..fe0fa642599 100644 --- a/src/Functions/base64Encode.cpp +++ b/src/Functions/base64Encode.cpp @@ -5,6 +5,7 @@ namespace DB { + REGISTER_FUNCTION(Base64Encode) { FunctionDocumentation::Description description = R"(Encodes a String as base64, according to RFC 4648 (https://datatracker.ietf.org/doc/html/rfc4648#section-4). Alias: TO_BASE64.)"; @@ -17,8 +18,9 @@ REGISTER_FUNCTION(Base64Encode) factory.registerFunction>>({description, syntax, arguments, returned_value, examples, categories}); /// MySQL compatibility alias. - factory.registerAlias("TO_BASE64", "base64Encode", FunctionFactory::CaseInsensitive); + factory.registerAlias("TO_BASE64", "base64Encode", FunctionFactory::Case::Insensitive); } + } #endif diff --git a/src/Functions/base64UrlDecode.cpp b/src/Functions/base64URLDecode.cpp similarity index 73% rename from src/Functions/base64UrlDecode.cpp rename to src/Functions/base64URLDecode.cpp index 59975d8f9d1..f256e111619 100644 --- a/src/Functions/base64UrlDecode.cpp +++ b/src/Functions/base64URLDecode.cpp @@ -5,17 +5,19 @@ namespace DB { -REGISTER_FUNCTION(Base64UrlDecode) + +REGISTER_FUNCTION(Base64URLDecode) { FunctionDocumentation::Description description = R"(Accepts a base64-encoded URL and decodes it from base64 with URL-specific modifications, according to RFC 4648 (https://datatracker.ietf.org/doc/html/rfc4648#section-5).)"; - FunctionDocumentation::Syntax syntax = "base64UrlDecode(encodedUrl)"; - FunctionDocumentation::Arguments arguments = {{"encodedUrl", "String column or constant. If the string is not a valid Base64-encoded value, an exception is thrown."}}; + FunctionDocumentation::Syntax syntax = "base64URLDecode(encodedURL)"; + FunctionDocumentation::Arguments arguments = {{"encodedURL", "String column or constant. If the string is not a valid Base64-encoded value, an exception is thrown."}}; FunctionDocumentation::ReturnedValue returned_value = "A string containing the decoded value of the argument."; - FunctionDocumentation::Examples examples = {{"Example", "SELECT base64UrlDecode('aHR0cDovL2NsaWNraG91c2UuY29t')", "https://clickhouse.com"}}; + FunctionDocumentation::Examples examples = {{"Example", "SELECT base64URLDecode('aHR0cDovL2NsaWNraG91c2UuY29t')", "https://clickhouse.com"}}; FunctionDocumentation::Categories categories = {"String encoding"}; - factory.registerFunction>>({description, syntax, arguments, returned_value, examples, categories}); + factory.registerFunction>>({description, syntax, arguments, returned_value, examples, categories}); } + } #endif diff --git a/src/Functions/base64UrlEncode.cpp b/src/Functions/base64URLEncode.cpp similarity index 78% rename from src/Functions/base64UrlEncode.cpp rename to src/Functions/base64URLEncode.cpp index 05d50170c14..215712f7586 100644 --- a/src/Functions/base64UrlEncode.cpp +++ b/src/Functions/base64URLEncode.cpp @@ -5,17 +5,19 @@ namespace DB { -REGISTER_FUNCTION(Base64UrlEncode) + +REGISTER_FUNCTION(Base64URLEncode) { FunctionDocumentation::Description description = R"(Encodes an URL (String or FixedString) as base64 with URL-specific modifications, according to RFC 4648 (https://datatracker.ietf.org/doc/html/rfc4648#section-5).)"; - FunctionDocumentation::Syntax syntax = "base64UrlEncode(url)"; + FunctionDocumentation::Syntax syntax = "base64URLEncode(url)"; FunctionDocumentation::Arguments arguments = {{"url", "String column or constant."}}; FunctionDocumentation::ReturnedValue returned_value = "A string containing the encoded value of the argument."; - FunctionDocumentation::Examples examples = {{"Example", "SELECT base64UrlEncode('https://clickhouse.com')", "aHR0cHM6Ly9jbGlja2hvdXNlLmNvbQ"}}; + FunctionDocumentation::Examples examples = {{"Example", "SELECT base64URLEncode('https://clickhouse.com')", "aHR0cHM6Ly9jbGlja2hvdXNlLmNvbQ"}}; FunctionDocumentation::Categories categories = {"String encoding"}; - factory.registerFunction>>({description, syntax, arguments, returned_value, examples, categories}); + factory.registerFunction>>({description, syntax, arguments, returned_value, examples, categories}); } + } #endif diff --git a/src/Functions/bitShiftLeft.cpp b/src/Functions/bitShiftLeft.cpp index c366a1ecb44..d561430d51f 100644 --- a/src/Functions/bitShiftLeft.cpp +++ b/src/Functions/bitShiftLeft.cpp @@ -5,6 +5,7 @@ namespace DB { namespace ErrorCodes { + extern const int ARGUMENT_OUT_OF_BOUND; extern const int NOT_IMPLEMENTED; extern const int LOGICAL_ERROR; } @@ -24,6 +25,8 @@ struct BitShiftLeftImpl { if constexpr (is_big_int_v) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "BitShiftLeft is not implemented for big integers as second argument"); + else if (b < 0 || static_cast(b) > 8 * sizeof(A)) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); else if constexpr (is_big_int_v) return static_cast(a) << static_cast(b); else @@ -37,9 +40,12 @@ struct BitShiftLeftImpl throw Exception(ErrorCodes::NOT_IMPLEMENTED, "BitShiftLeft is not implemented for big integers as second argument"); else { - UInt8 word_size = 8; - /// To prevent overflow - if (static_cast(b) >= (static_cast(end - pos) * word_size) || b < 0) + const UInt8 word_size = 8 * sizeof(*pos); + size_t n = end - pos; + const UInt128 bit_limit = static_cast(word_size) * n; + if (b < 0 || static_cast(b) > bit_limit) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); + else if (b == bit_limit) { // insert default value out_vec.push_back(0); @@ -102,10 +108,12 @@ struct BitShiftLeftImpl throw Exception(ErrorCodes::NOT_IMPLEMENTED, "BitShiftLeft is not implemented for big integers as second argument"); else { - UInt8 word_size = 8; + const UInt8 word_size = 8; size_t n = end - pos; - /// To prevent overflow - if (static_cast(b) >= (static_cast(n) * word_size) || b < 0) + const UInt128 bit_limit = static_cast(word_size) * n; + if (b < 0 || static_cast(b) > bit_limit) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); + else if (b == bit_limit) { // insert default value out_vec.resize_fill(out_vec.size() + n); diff --git a/src/Functions/bitShiftRight.cpp b/src/Functions/bitShiftRight.cpp index 1c37cd3bf4c..05b8581c792 100644 --- a/src/Functions/bitShiftRight.cpp +++ b/src/Functions/bitShiftRight.cpp @@ -6,6 +6,7 @@ namespace DB { namespace ErrorCodes { + extern const int ARGUMENT_OUT_OF_BOUND; extern const int NOT_IMPLEMENTED; extern const int LOGICAL_ERROR; } @@ -25,6 +26,8 @@ struct BitShiftRightImpl { if constexpr (is_big_int_v) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "BitShiftRight is not implemented for big integers as second argument"); + else if (b < 0 || static_cast(b) > 8 * sizeof(A)) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); else if constexpr (is_big_int_v) return static_cast(a) >> static_cast(b); else @@ -53,9 +56,12 @@ struct BitShiftRightImpl throw Exception(ErrorCodes::NOT_IMPLEMENTED, "BitShiftRight is not implemented for big integers as second argument"); else { - UInt8 word_size = 8; - /// To prevent overflow - if (static_cast(b) >= (static_cast(end - pos) * word_size) || b < 0) + const UInt8 word_size = 8; + size_t n = end - pos; + const UInt128 bit_limit = static_cast(word_size) * n; + if (b < 0 || static_cast(b) > bit_limit) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); + else if (b == bit_limit) { /// insert default value out_vec.push_back(0); @@ -90,10 +96,12 @@ struct BitShiftRightImpl throw Exception(ErrorCodes::NOT_IMPLEMENTED, "BitShiftRight is not implemented for big integers as second argument"); else { - UInt8 word_size = 8; + const UInt8 word_size = 8; size_t n = end - pos; - /// To prevent overflow - if (static_cast(b) >= (static_cast(n) * word_size) || b < 0) + const UInt128 bit_limit = static_cast(word_size) * n; + if (b < 0 || static_cast(b) > bit_limit) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); + else if (b == bit_limit) { // insert default value out_vec.resize_fill(out_vec.size() + n); diff --git a/src/Functions/bitTest.cpp b/src/Functions/bitTest.cpp index 78ec9c8b773..cb6b83c1cf1 100644 --- a/src/Functions/bitTest.cpp +++ b/src/Functions/bitTest.cpp @@ -8,6 +8,7 @@ namespace DB namespace ErrorCodes { extern const int NOT_IMPLEMENTED; + extern const int PARAMETER_OUT_OF_BOUND; } namespace @@ -21,12 +22,21 @@ struct BitTestImpl static const constexpr bool allow_string_integer = false; template - NO_SANITIZE_UNDEFINED static Result apply(A a [[maybe_unused]], B b [[maybe_unused]]) + static Result apply(A a [[maybe_unused]], B b [[maybe_unused]]) { if constexpr (is_big_int_v || is_big_int_v) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "bitTest is not implemented for big integers as second argument"); else - return (typename NumberTraits::ToInteger::Type(a) >> typename NumberTraits::ToInteger::Type(b)) & 1; + { + typename NumberTraits::ToInteger::Type a_int = a; + typename NumberTraits::ToInteger::Type b_int = b; + const auto max_position = static_cast((8 * sizeof(a)) - 1); + if (b_int > max_position || b_int < 0) + throw Exception(ErrorCodes::PARAMETER_OUT_OF_BOUND, + "The bit position argument needs to a positive value and less or equal to {} for integer {}", + std::to_string(max_position), std::to_string(a_int)); + return (a_int >> b_int) & 1; + } } #if USE_EMBEDDED_COMPILER diff --git a/src/Functions/byteSize.cpp b/src/Functions/byteSize.cpp index 93a3a86641a..d366a1b2e12 100644 --- a/src/Functions/byteSize.cpp +++ b/src/Functions/byteSize.cpp @@ -67,11 +67,11 @@ public: const IColumn * column = arguments[arg_num].column.get(); if (arg_num == 0) - for (size_t row_num = 0; row_num < input_rows_count; ++row_num) - vec_res[row_num] = column->byteSizeAt(row_num); + for (size_t row = 0; row < input_rows_count; ++row) + vec_res[row] = column->byteSizeAt(row); else - for (size_t row_num = 0; row_num < input_rows_count; ++row_num) - vec_res[row_num] += column->byteSizeAt(row_num); + for (size_t row = 0; row < input_rows_count; ++row) + vec_res[row] += column->byteSizeAt(row); } return result_col; diff --git a/src/Functions/byteSwap.cpp b/src/Functions/byteSwap.cpp index 2a343a07720..2094ec4fa1a 100644 --- a/src/Functions/byteSwap.cpp +++ b/src/Functions/byteSwap.cpp @@ -10,6 +10,7 @@ extern const int NOT_IMPLEMENTED; namespace { + template requires std::is_integral_v T byteSwap(T x) @@ -100,7 +101,7 @@ One use-case of this function is reversing IPv4s: {"64-bit", "SELECT byteSwap(123294967295)", "18439412204227788800"}, }, .categories{"Mathematical", "Arithmetic"}}, - FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/caseWithExpression.cpp b/src/Functions/caseWithExpression.cpp index 71fccc8436e..f0a620489ef 100644 --- a/src/Functions/caseWithExpression.cpp +++ b/src/Functions/caseWithExpression.cpp @@ -98,8 +98,7 @@ public: /// Execute transform. ColumnsWithTypeAndName transform_args{args.front(), src_array_col, dst_array_col, args.back()}; - return FunctionFactory::instance().get("transform", context)->build(transform_args) - ->execute(transform_args, result_type, input_rows_count); + return FunctionFactory::instance().get("transform", context)->build(transform_args)->execute(transform_args, result_type, input_rows_count); } private: diff --git a/src/Functions/castOrDefault.cpp b/src/Functions/castOrDefault.cpp index 44b39811882..1ece17ca1e3 100644 --- a/src/Functions/castOrDefault.cpp +++ b/src/Functions/castOrDefault.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -203,7 +204,7 @@ private: DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, nullptr}}; + FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, "any type"}}; FunctionArgumentDescriptors optional_args; if (isDecimal(type) || isDateTime64(type)) @@ -212,9 +213,9 @@ private: if (isDateTimeOrDateTime64(type)) optional_args.push_back({"timezone", static_cast(&isString), isColumnConst, "const String"}); - optional_args.push_back({"default_value", nullptr, nullptr, nullptr}); + optional_args.push_back({"default_value", nullptr, nullptr, "any type"}); - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); size_t additional_argument_index = 1; diff --git a/src/Functions/changeDate.cpp b/src/Functions/changeDate.cpp new file mode 100644 index 00000000000..19e4c165ee3 --- /dev/null +++ b/src/Functions/changeDate.cpp @@ -0,0 +1,399 @@ +#include "Common/DateLUTImpl.h" +#include "Common/Exception.h" +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +namespace +{ + +enum class Component +{ + Year, + Month, + Day, + Hour, + Minute, + Second +}; + +} + +template +class FunctionChangeDate : public IFunction +{ +public: + static constexpr auto name = Traits::name; + + static FunctionPtr create(ContextPtr) { return std::make_shared(); } + String getName() const override { return Traits::name; } + bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } + size_t getNumberOfArguments() const override { return 2; } + + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override + { + FunctionArgumentDescriptors args{ + {"date_or_datetime", static_cast(&isDateOrDate32OrDateTimeOrDateTime64), nullptr, "Date or date with time"}, + {"value", static_cast(&isNativeInteger), nullptr, "Integer"} + }; + validateFunctionArguments(*this, arguments, args); + + const auto & input_type = arguments[0].type; + + if constexpr (Traits::component == Component::Hour || Traits::component == Component::Minute || Traits::component == Component::Second) + { + if (isDate(input_type)) + return std::make_shared(); + if (isDate32(input_type)) + return std::make_shared(DataTypeDateTime64::default_scale); + } + + return input_type; + } + + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override + { + const auto & input_type = arguments[0].type; + if (isDate(input_type)) + { + if constexpr (Traits::component == Component::Hour || Traits::component == Component::Minute || Traits::component == Component::Second) + return execute(arguments, input_type, result_type, input_rows_count); + return execute(arguments, input_type, result_type, input_rows_count); + } + if (isDate32(input_type)) + { + if constexpr (Traits::component == Component::Hour || Traits::component == Component::Minute || Traits::component == Component::Second) + return execute(arguments, input_type, result_type, input_rows_count); + return execute(arguments, input_type, result_type, input_rows_count); + } + if (isDateTime(input_type)) + return execute(arguments, input_type, result_type, input_rows_count); + if (isDateTime64(input_type)) + return execute(arguments, input_type, result_type, input_rows_count); + + throw Exception(ErrorCodes::LOGICAL_ERROR, "Invalid input type"); + } + + template + ColumnPtr execute(const ColumnsWithTypeAndName & arguments, const DataTypePtr & input_type, const DataTypePtr & result_type, size_t input_rows_count) const + { + typename ResultDataType::ColumnType::MutablePtr result_col; + if constexpr (std::is_same_v) + { + auto scale = DataTypeDateTime64::default_scale; + if constexpr (std::is_same_v) + scale = typeid_cast(*result_type).getScale(); + result_col = ResultDataType::ColumnType::create(input_rows_count, scale); + } + else + result_col = ResultDataType::ColumnType::create(input_rows_count); + + auto date_time_col = arguments[0].column->convertToFullIfNeeded(); + const auto & date_time_col_data = typeid_cast(*date_time_col).getData(); + + auto value_col = castColumn(arguments[1], std::make_shared()); + value_col = value_col->convertToFullIfNeeded(); + const auto & value_col_data = typeid_cast(*value_col).getData(); + + auto & result_col_data = result_col->getData(); + + if constexpr (std::is_same_v) + { + const auto scale = typeid_cast(*result_type).getScale(); + const auto & date_lut = typeid_cast(*result_type).getTimeZone(); + + Int64 deg = 1; + for (size_t j = 0; j < scale; ++j) + deg *= 10; + + for (size_t i = 0; i < input_rows_count; ++i) + { + Int64 time = date_lut.toNumYYYYMMDDhhmmss(date_time_col_data[i] / deg); + Int64 fraction = date_time_col_data[i] % deg; + + result_col_data[i] = getChangedDate(time, value_col_data[i], result_type, date_lut, scale, fraction); + } + } + else if constexpr (std::is_same_v && std::is_same_v) + { + const auto & date_lut = typeid_cast(*result_type).getTimeZone(); + for (size_t i = 0; i < input_rows_count; ++i) + { + Int64 time = static_cast(date_lut.toNumYYYYMMDD(ExtendedDayNum(date_time_col_data[i]))) * 1'000'000; + result_col_data[i] = getChangedDate(time, value_col_data[i], result_type, date_lut, 3, 0); + } + } + else if constexpr (std::is_same_v && std::is_same_v) + { + const auto & date_lut = typeid_cast(*result_type).getTimeZone(); + for (size_t i = 0; i < input_rows_count; ++i) + { + Int64 time = static_cast(date_lut.toNumYYYYMMDD(ExtendedDayNum(date_time_col_data[i]))) * 1'000'000; + result_col_data[i] = static_cast(getChangedDate(time, value_col_data[i], result_type, date_lut)); + } + } + else if constexpr (std::is_same_v) + { + const auto & date_lut = typeid_cast(*result_type).getTimeZone(); + for (size_t i = 0; i < input_rows_count; ++i) + { + Int64 time = date_lut.toNumYYYYMMDDhhmmss(date_time_col_data[i]); + result_col_data[i] = static_cast(getChangedDate(time, value_col_data[i], result_type, date_lut)); + } + } + else + { + const auto & date_lut = DateLUT::instance(); + for (size_t i = 0; i < input_rows_count; ++i) + { + Int64 time; + if (isDate(input_type)) + time = static_cast(date_lut.toNumYYYYMMDD(DayNum(date_time_col_data[i]))) * 1'000'000; + else + time = static_cast(date_lut.toNumYYYYMMDD(ExtendedDayNum(date_time_col_data[i]))) * 1'000'000; + + if (isDate(result_type)) + result_col_data[i] = static_cast(getChangedDate(time, value_col_data[i], result_type, date_lut)); + else + result_col_data[i] = static_cast(getChangedDate(time, value_col_data[i], result_type, date_lut)); + } + } + + return result_col; + } + + Int64 getChangedDate(Int64 time, Float64 new_value, const DataTypePtr & result_type, const DateLUTImpl & date_lut, Int64 scale = 0, Int64 fraction = 0) const + { + auto year = time / 10'000'000'000; + auto month = (time % 10'000'000'000) / 100'000'000; + auto day = (time % 100'000'000) / 1'000'000; + auto hours = (time % 1'000'000) / 10'000; + auto minutes = (time % 10'000) / 100; + auto seconds = time % 100; + + Int64 min_date = 0, max_date = 0; + Int16 min_year, max_year; + if (isDate(result_type)) + { + min_date = date_lut.makeDayNum(1970, 1, 1); + max_date = date_lut.makeDayNum(2149, 6, 6); + min_year = 1970; + max_year = 2149; + } + else if (isDate32(result_type)) + { + min_date = date_lut.makeDayNum(1900, 1, 1); + max_date = date_lut.makeDayNum(2299, 12, 31); + min_year = 1900; + max_year = 2299; + } + else if (isDateTime(result_type)) + { + min_date = 0; + max_date = 0x0FFFFFFFFLL; + min_year = 1970; + max_year = 2106; + } + else + { + min_date = DecimalUtils::decimalFromComponents( + date_lut.makeDateTime(1900, 1, 1, 0, 0, 0), + static_cast(0), + static_cast(scale)); + Int64 deg = 1; + for (Int64 j = 0; j < scale; ++j) + deg *= 10; + max_date = DecimalUtils::decimalFromComponents( + date_lut.makeDateTime(2299, 12, 31, 23, 59, 59), + static_cast(deg - 1), + static_cast(scale)); + min_year = 1900; + max_year = 2299; + } + + switch (Traits::component) + { + case Component::Year: + if (new_value < min_year) + return min_date; + else if (new_value > max_year) + return max_date; + year = static_cast(new_value); + break; + case Component::Month: + if (new_value < 1 || new_value > 12) + return min_date; + month = static_cast(new_value); + break; + case Component::Day: + if (new_value < 1 || new_value > 31) + return min_date; + day = static_cast(new_value); + break; + case Component::Hour: + if (new_value < 0 || new_value > 23) + return min_date; + hours = static_cast(new_value); + break; + case Component::Minute: + if (new_value < 0 || new_value > 59) + return min_date; + minutes = static_cast(new_value); + break; + case Component::Second: + if (new_value < 0 || new_value > 59) + return min_date; + seconds = static_cast(new_value); + break; + } + + Int64 result; + if (isDate(result_type) || isDate32(result_type)) + result = date_lut.makeDayNum(year, month, day); + else if (isDateTime(result_type)) + result = date_lut.makeDateTime(year, month, day, hours, minutes, seconds); + else +#ifndef __clang_analyzer__ + /// ^^ This looks funny. It is the least terrible suppression of a false positive reported by clang-analyzer (a sub-class + /// of clang-tidy checks) deep down in 'decimalFromComponents'. Usual suppressions of the form NOLINT* don't work here (they + /// would only affect code in _this_ file), and suppressing the issue in 'decimalFromComponents' may suppress true positives. + result = DecimalUtils::decimalFromComponents( + date_lut.makeDateTime(year, month, day, hours, minutes, seconds), + fraction, + static_cast(scale)); +#else + { + UNUSED(fraction); + result = 0; + } +#endif + + if (result < min_date) + return min_date; + + if (result > max_date) + return max_date; + + return result; + } +}; + + +struct ChangeYearTraits +{ + static constexpr auto name = "changeYear"; + static constexpr auto component = Component::Year; +}; + +struct ChangeMonthTraits +{ + static constexpr auto name = "changeMonth"; + static constexpr auto component = Component::Month; +}; + +struct ChangeDayTraits +{ + static constexpr auto name = "changeDay"; + static constexpr auto component = Component::Day; +}; + +struct ChangeHourTraits +{ + static constexpr auto name = "changeHour"; + static constexpr auto component = Component::Hour; +}; + +struct ChangeMinuteTraits +{ + static constexpr auto name = "changeMinute"; + static constexpr auto component = Component::Minute; +}; + +struct ChangeSecondTraits +{ + static constexpr auto name = "changeSecond"; + static constexpr auto component = Component::Second; +}; + +REGISTER_FUNCTION(ChangeDate) +{ + { + FunctionDocumentation::Description description = "Changes the year component of a date or date time."; + FunctionDocumentation::Syntax syntax = "changeYear(date_or_datetime, value);"; + FunctionDocumentation::Arguments arguments = {{"date_or_datetime", "The value to change. Type: Date, Date32, DateTime, or DateTime64"}, {"value", "The new value. Type: [U]Int*"}}; + FunctionDocumentation::ReturnedValue returned_value = "The same type as date_or_datetime."; + FunctionDocumentation::Categories categories = {"Dates and Times"}; + FunctionDocumentation function_documentation = {.description = description, .syntax = syntax, .arguments = arguments, .returned_value = returned_value, .categories = categories}; + factory.registerFunction>(function_documentation); + } + { + FunctionDocumentation::Description description = "Changes the month component of a date or date time."; + FunctionDocumentation::Syntax syntax = "changeMonth(date_or_datetime, value);"; + FunctionDocumentation::Arguments arguments = {{"date_or_datetime", "The value to change. Type: Date, Date32, DateTime, or DateTime64"}, {"value", "The new value. Type: [U]Int*"}}; + FunctionDocumentation::ReturnedValue returned_value = "The same type as date_or_datetime."; + FunctionDocumentation::Categories categories = {"Dates and Times"}; + FunctionDocumentation function_documentation = {.description = description, .syntax = syntax, .arguments = arguments, .returned_value = returned_value, .categories = categories}; + factory.registerFunction>(function_documentation); + } + { + FunctionDocumentation::Description description = "Changes the day component of a date or date time."; + FunctionDocumentation::Syntax syntax = "changeDay(date_or_datetime, value);"; + FunctionDocumentation::Arguments arguments = {{"date_or_datetime", "The value to change. Type: Date, Date32, DateTime, or DateTime64"}, {"value", "The new value. Type: [U]Int*"}}; + FunctionDocumentation::ReturnedValue returned_value = "The same type as date_or_datetime."; + FunctionDocumentation::Categories categories = {"Dates and Times"}; + FunctionDocumentation function_documentation = {.description = description, .syntax = syntax, .arguments = arguments, .returned_value = returned_value, .categories = categories}; + factory.registerFunction>(function_documentation); + } + { + FunctionDocumentation::Description description = "Changes the hour component of a date or date time."; + FunctionDocumentation::Syntax syntax = "changeHour(date_or_datetime, value);"; + FunctionDocumentation::Arguments arguments = {{"date_or_datetime", "The value to change. Type: Date, Date32, DateTime, or DateTime64"}, {"value", "The new value. Type: [U]Int*"}}; + FunctionDocumentation::ReturnedValue returned_value = "The same type as date_or_datetime. If the input is a Date, return DateTime. If the input is a Date32, return DateTime64."; + FunctionDocumentation::Categories categories = {"Dates and Times"}; + FunctionDocumentation function_documentation = {.description = description, .syntax = syntax, .arguments = arguments, .returned_value = returned_value, .categories = categories}; + factory.registerFunction>(function_documentation); + } + { + FunctionDocumentation::Description description = "Changes the minute component of a date or date time."; + FunctionDocumentation::Syntax syntax = "changeMinute(date_or_datetime, value);"; + FunctionDocumentation::Arguments arguments = {{"date_or_datetime", "The value to change. Type: Date, Date32, DateTime, or DateTime64"}, {"value", "The new value. Type: [U]Int*"}}; + FunctionDocumentation::ReturnedValue returned_value = "The same type as date_or_datetime. If the input is a Date, return DateTime. If the input is a Date32, return DateTime64."; + FunctionDocumentation::Categories categories = {"Dates and Times"}; + FunctionDocumentation function_documentation = {.description = description, .syntax = syntax, .arguments = arguments, .returned_value = returned_value, .categories = categories}; + factory.registerFunction>(function_documentation); + } + { + FunctionDocumentation::Description description = "Changes the second component of a date or date time."; + FunctionDocumentation::Syntax syntax = "changeSecond(date_or_datetime, value);"; + FunctionDocumentation::Arguments arguments = {{"date_or_datetime", "The value to change. Type: Date, Date32, DateTime, or DateTime64"}, {"value", "The new value. Type: [U]Int*"}}; + FunctionDocumentation::ReturnedValue returned_value = "The same type as date_or_datetime. If the input is a Date, return DateTime. If the input is a Date32, return DateTime64."; + FunctionDocumentation::Categories categories = {"Dates and Times"}; + FunctionDocumentation function_documentation = {.description = description, .syntax = syntax, .arguments = arguments, .returned_value = returned_value, .categories = categories}; + factory.registerFunction>(function_documentation); + } +} + +} diff --git a/src/Functions/coalesce.cpp b/src/Functions/coalesce.cpp index 722f32af523..19da6a85b38 100644 --- a/src/Functions/coalesce.cpp +++ b/src/Functions/coalesce.cpp @@ -180,7 +180,7 @@ private: REGISTER_FUNCTION(Coalesce) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/concat.cpp b/src/Functions/concat.cpp index 68cfcdb8d90..5c5e089e740 100644 --- a/src/Functions/concat.cpp +++ b/src/Functions/concat.cpp @@ -46,25 +46,30 @@ public: DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override { - if (arguments.size() < 2) - throw Exception( - ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, - "Number of arguments for function {} doesn't match: passed {}, should be at least 2", - getName(), - arguments.size()); + if (arguments.size() == 1) + throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Number of arguments for function {} should not be 1", getName()); return std::make_shared(); } ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { + if (arguments.empty()) + { + auto res_data = ColumnString::create(); + res_data->insertDefault(); + return ColumnConst::create(std::move(res_data), input_rows_count); + } + else if (arguments.size() == 1) + return arguments[0].column; /// Format function is not proven to be faster for two arguments. /// Actually there is overhead of 2 to 5 extra instructions for each string for checking empty strings in FormatImpl. /// Though, benchmarks are really close, for most examples we saw executeBinary is slightly faster (0-3%). /// For 3 and more arguments FormatStringImpl is much faster (up to 50-60%). - if (arguments.size() == 2) + else if (arguments.size() == 2) return executeBinary(arguments, input_rows_count); - return executeFormatImpl(arguments, input_rows_count); + else + return executeFormatImpl(arguments, input_rows_count); } private: @@ -209,11 +214,11 @@ public: { if (arguments.size() == 1) return FunctionFactory::instance().getImpl("toString", context)->build(arguments); - if (std::ranges::all_of(arguments, [](const auto & elem) { return isArray(elem.type); })) + if (!arguments.empty() && std::ranges::all_of(arguments, [](const auto & elem) { return isArray(elem.type); })) return FunctionFactory::instance().getImpl("arrayConcat", context)->build(arguments); - if (std::ranges::all_of(arguments, [](const auto & elem) { return isMap(elem.type); })) + if (!arguments.empty() && std::ranges::all_of(arguments, [](const auto & elem) { return isMap(elem.type); })) return FunctionFactory::instance().getImpl("mapConcat", context)->build(arguments); - if (std::ranges::all_of(arguments, [](const auto & elem) { return isTuple(elem.type); })) + if (!arguments.empty() && std::ranges::all_of(arguments, [](const auto & elem) { return isTuple(elem.type); })) return FunctionFactory::instance().getImpl("tupleConcat", context)->build(arguments); return std::make_unique( FunctionConcat::create(context), @@ -221,15 +226,8 @@ public: return_type); } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const DataTypes &) const override { - if (arguments.empty()) - throw Exception( - ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, - "Number of arguments for function {} doesn't match: passed {}, should be at least 1.", - getName(), - arguments.size()); - /// We always return Strings from concat, even if arguments were fixed strings. return std::make_shared(); } @@ -242,7 +240,7 @@ private: REGISTER_FUNCTION(Concat) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); factory.registerFunction(); } diff --git a/src/Functions/concatWithSeparator.cpp b/src/Functions/concatWithSeparator.cpp index ed02f331192..1d38ef87558 100644 --- a/src/Functions/concatWithSeparator.cpp +++ b/src/Functions/concatWithSeparator.cpp @@ -193,7 +193,7 @@ The function is named “injective” if it always returns different result for .categories{"String"}}); /// Compatibility with Spark and MySQL: - factory.registerAlias("concat_ws", "concatWithSeparator", FunctionFactory::CaseInsensitive); + factory.registerAlias("concat_ws", "concatWithSeparator", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/connectionId.cpp b/src/Functions/connectionId.cpp index 9c53482482b..c1036b2ddbe 100644 --- a/src/Functions/connectionId.cpp +++ b/src/Functions/connectionId.cpp @@ -33,8 +33,8 @@ public: REGISTER_FUNCTION(ConnectionId) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerAlias("connection_id", "connectionID", FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerAlias("connection_id", "connectionID", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/convertCharset.cpp b/src/Functions/convertCharset.cpp index b3b7394acb9..d998e88e7c2 100644 --- a/src/Functions/convertCharset.cpp +++ b/src/Functions/convertCharset.cpp @@ -88,7 +88,8 @@ private: static void convert(const String & from_charset, const String & to_charset, const ColumnString::Chars & from_chars, const ColumnString::Offsets & from_offsets, - ColumnString::Chars & to_chars, ColumnString::Offsets & to_offsets) + ColumnString::Chars & to_chars, ColumnString::Offsets & to_offsets, + size_t input_rows_count) { auto converter_from = getConverter(from_charset); auto converter_to = getConverter(to_charset); @@ -96,12 +97,11 @@ private: ColumnString::Offset current_from_offset = 0; ColumnString::Offset current_to_offset = 0; - size_t size = from_offsets.size(); - to_offsets.resize(size); + to_offsets.resize(input_rows_count); PODArray uchars; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t from_string_size = from_offsets[i] - current_from_offset - 1; @@ -184,7 +184,7 @@ public: bool useDefaultImplementationForConstants() const override { return true; } ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {1, 2}; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const ColumnWithTypeAndName & arg_from = arguments[0]; const ColumnWithTypeAndName & arg_charset_from = arguments[1]; @@ -204,7 +204,7 @@ public: if (const ColumnString * col_from = checkAndGetColumn(arg_from.column.get())) { auto col_to = ColumnString::create(); - convert(charset_from, charset_to, col_from->getChars(), col_from->getOffsets(), col_to->getChars(), col_to->getOffsets()); + convert(charset_from, charset_to, col_from->getChars(), col_from->getOffsets(), col_to->getChars(), col_to->getOffsets(), input_rows_count); return col_to; } else diff --git a/src/Functions/cos.cpp b/src/Functions/cos.cpp index 3496373a9d5..40fdede0e1c 100644 --- a/src/Functions/cos.cpp +++ b/src/Functions/cos.cpp @@ -13,7 +13,7 @@ using FunctionCos = FunctionMathUnary>; REGISTER_FUNCTION(Cos) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/cosh.cpp b/src/Functions/cosh.cpp index 54b52051aab..f4302292303 100644 --- a/src/Functions/cosh.cpp +++ b/src/Functions/cosh.cpp @@ -5,11 +5,12 @@ namespace DB { namespace { - struct CoshName - { - static constexpr auto name = "cosh"; - }; - using FunctionCosh = FunctionMathUnary>; + +struct CoshName +{ + static constexpr auto name = "cosh"; +}; +using FunctionCosh = FunctionMathUnary>; } diff --git a/src/Functions/countMatches.cpp b/src/Functions/countMatches.cpp index a8620080012..4db48b1305f 100644 --- a/src/Functions/countMatches.cpp +++ b/src/Functions/countMatches.cpp @@ -22,8 +22,8 @@ namespace DB REGISTER_FUNCTION(CountMatches) { - factory.registerFunction>({}, FunctionFactory::CaseSensitive); - factory.registerFunction>({}, FunctionFactory::CaseSensitive); + factory.registerFunction>(); + factory.registerFunction>(); } } diff --git a/src/Functions/countMatches.h b/src/Functions/countMatches.h index fbbb9d017ee..5f07b936e26 100644 --- a/src/Functions/countMatches.h +++ b/src/Functions/countMatches.h @@ -38,7 +38,7 @@ public: {"haystack", static_cast(&isStringOrFixedString), nullptr, "String or FixedString"}, {"pattern", static_cast(&isString), isColumnConst, "constant String"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } diff --git a/src/Functions/countSubstrings.cpp b/src/Functions/countSubstrings.cpp index 843b81437f5..137edb179b2 100644 --- a/src/Functions/countSubstrings.cpp +++ b/src/Functions/countSubstrings.cpp @@ -19,6 +19,6 @@ using FunctionCountSubstrings = FunctionsStringSearch({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/countSubstringsCaseInsensitiveUTF8.cpp b/src/Functions/countSubstringsCaseInsensitiveUTF8.cpp index 3f71bca63d2..99ae4f1927e 100644 --- a/src/Functions/countSubstringsCaseInsensitiveUTF8.cpp +++ b/src/Functions/countSubstringsCaseInsensitiveUTF8.cpp @@ -13,8 +13,7 @@ struct NameCountSubstringsCaseInsensitiveUTF8 static constexpr auto name = "countSubstringsCaseInsensitiveUTF8"; }; -using FunctionCountSubstringsCaseInsensitiveUTF8 = FunctionsStringSearch< - CountSubstringsImpl>; +using FunctionCountSubstringsCaseInsensitiveUTF8 = FunctionsStringSearch>; } diff --git a/src/Functions/currentDatabase.cpp b/src/Functions/currentDatabase.cpp index 954899c3c2b..16cb43ebb04 100644 --- a/src/Functions/currentDatabase.cpp +++ b/src/Functions/currentDatabase.cpp @@ -54,9 +54,9 @@ public: REGISTER_FUNCTION(CurrentDatabase) { factory.registerFunction(); - factory.registerAlias("DATABASE", FunctionCurrentDatabase::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("SCHEMA", FunctionCurrentDatabase::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("current_database", FunctionCurrentDatabase::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("DATABASE", FunctionCurrentDatabase::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("SCHEMA", FunctionCurrentDatabase::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("current_database", FunctionCurrentDatabase::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/currentSchemas.cpp b/src/Functions/currentSchemas.cpp index 322e719eb17..0a128d0e908 100644 --- a/src/Functions/currentSchemas.cpp +++ b/src/Functions/currentSchemas.cpp @@ -80,8 +80,8 @@ Requires a boolean parameter, but it is ignored actually. It is required just fo {"common", "SELECT current_schemas(true);", "['default']"} } }, - FunctionFactory::CaseInsensitive); - factory.registerAlias("current_schemas", FunctionCurrentSchemas::name, FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); + factory.registerAlias("current_schemas", FunctionCurrentSchemas::name, FunctionFactory::Case::Insensitive); } diff --git a/src/Functions/currentUser.cpp b/src/Functions/currentUser.cpp index 1679c56a929..9f48f15ffb3 100644 --- a/src/Functions/currentUser.cpp +++ b/src/Functions/currentUser.cpp @@ -54,8 +54,8 @@ public: REGISTER_FUNCTION(CurrentUser) { factory.registerFunction(); - factory.registerAlias("user", FunctionCurrentUser::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("current_user", FunctionCurrentUser::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("user", FunctionCurrentUser::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("current_user", FunctionCurrentUser::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/dateDiff.cpp b/src/Functions/dateDiff.cpp index 8e8865db7ed..157068a7c8d 100644 --- a/src/Functions/dateDiff.cpp +++ b/src/Functions/dateDiff.cpp @@ -26,8 +26,6 @@ namespace DB namespace ErrorCodes { - extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; - extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int ILLEGAL_COLUMN; extern const int BAD_ARGUMENTS; } @@ -45,84 +43,82 @@ public: template void dispatchForColumns( - const IColumn & x, const IColumn & y, + const IColumn & col_x, const IColumn & col_y, const DateLUTImpl & timezone_x, const DateLUTImpl & timezone_y, + size_t input_rows_count, ColumnInt64::Container & result) const { - if (const auto * x_vec_16 = checkAndGetColumn(&x)) - dispatchForSecondColumn(*x_vec_16, y, timezone_x, timezone_y, result); - else if (const auto * x_vec_32 = checkAndGetColumn(&x)) - dispatchForSecondColumn(*x_vec_32, y, timezone_x, timezone_y, result); - else if (const auto * x_vec_32_s = checkAndGetColumn(&x)) - dispatchForSecondColumn(*x_vec_32_s, y, timezone_x, timezone_y, result); - else if (const auto * x_vec_64 = checkAndGetColumn(&x)) - dispatchForSecondColumn(*x_vec_64, y, timezone_x, timezone_y, result); - else if (const auto * x_const_16 = checkAndGetColumnConst(&x)) - dispatchConstForSecondColumn(x_const_16->getValue(), y, timezone_x, timezone_y, result); - else if (const auto * x_const_32 = checkAndGetColumnConst(&x)) - dispatchConstForSecondColumn(x_const_32->getValue(), y, timezone_x, timezone_y, result); - else if (const auto * x_const_32_s = checkAndGetColumnConst(&x)) - dispatchConstForSecondColumn(x_const_32_s->getValue(), y, timezone_x, timezone_y, result); - else if (const auto * x_const_64 = checkAndGetColumnConst(&x)) - dispatchConstForSecondColumn(x_const_64->getValue>(), y, timezone_x, timezone_y, result); + if (const auto * x_vec_16 = checkAndGetColumn(&col_x)) + dispatchForSecondColumn(*x_vec_16, col_y, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * x_vec_32 = checkAndGetColumn(&col_x)) + dispatchForSecondColumn(*x_vec_32, col_y, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * x_vec_32_s = checkAndGetColumn(&col_x)) + dispatchForSecondColumn(*x_vec_32_s, col_y, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * x_vec_64 = checkAndGetColumn(&col_x)) + dispatchForSecondColumn(*x_vec_64, col_y, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * x_const_16 = checkAndGetColumnConst(&col_x)) + dispatchConstForSecondColumn(x_const_16->getValue(), col_y, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * x_const_32 = checkAndGetColumnConst(&col_x)) + dispatchConstForSecondColumn(x_const_32->getValue(), col_y, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * x_const_32_s = checkAndGetColumnConst(&col_x)) + dispatchConstForSecondColumn(x_const_32_s->getValue(), col_y, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * x_const_64 = checkAndGetColumnConst(&col_x)) + dispatchConstForSecondColumn(x_const_64->getValue>(), col_y, timezone_x, timezone_y, input_rows_count, result); else - throw Exception(ErrorCodes::ILLEGAL_COLUMN, - "Illegal column for first argument of function {}, must be Date, Date32, DateTime or DateTime64", - name); + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column for first argument of function {}, must be Date, Date32, DateTime or DateTime64", name); } template void dispatchForSecondColumn( - const LeftColumnType & x, const IColumn & y, + const LeftColumnType & x, const IColumn & col_y, const DateLUTImpl & timezone_x, const DateLUTImpl & timezone_y, + size_t input_rows_count, ColumnInt64::Container & result) const { - if (const auto * y_vec_16 = checkAndGetColumn(&y)) - vectorVector(x, *y_vec_16, timezone_x, timezone_y, result); - else if (const auto * y_vec_32 = checkAndGetColumn(&y)) - vectorVector(x, *y_vec_32, timezone_x, timezone_y, result); - else if (const auto * y_vec_32_s = checkAndGetColumn(&y)) - vectorVector(x, *y_vec_32_s, timezone_x, timezone_y, result); - else if (const auto * y_vec_64 = checkAndGetColumn(&y)) - vectorVector(x, *y_vec_64, timezone_x, timezone_y, result); - else if (const auto * y_const_16 = checkAndGetColumnConst(&y)) - vectorConstant(x, y_const_16->getValue(), timezone_x, timezone_y, result); - else if (const auto * y_const_32 = checkAndGetColumnConst(&y)) - vectorConstant(x, y_const_32->getValue(), timezone_x, timezone_y, result); - else if (const auto * y_const_32_s = checkAndGetColumnConst(&y)) - vectorConstant(x, y_const_32_s->getValue(), timezone_x, timezone_y, result); - else if (const auto * y_const_64 = checkAndGetColumnConst(&y)) - vectorConstant(x, y_const_64->getValue>(), timezone_x, timezone_y, result); + if (const auto * y_vec_16 = checkAndGetColumn(&col_y)) + vectorVector(x, *y_vec_16, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_vec_32 = checkAndGetColumn(&col_y)) + vectorVector(x, *y_vec_32, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_vec_32_s = checkAndGetColumn(&col_y)) + vectorVector(x, *y_vec_32_s, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_vec_64 = checkAndGetColumn(&col_y)) + vectorVector(x, *y_vec_64, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_const_16 = checkAndGetColumnConst(&col_y)) + vectorConstant(x, y_const_16->getValue(), timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_const_32 = checkAndGetColumnConst(&col_y)) + vectorConstant(x, y_const_32->getValue(), timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_const_32_s = checkAndGetColumnConst(&col_y)) + vectorConstant(x, y_const_32_s->getValue(), timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_const_64 = checkAndGetColumnConst(&col_y)) + vectorConstant(x, y_const_64->getValue>(), timezone_x, timezone_y, input_rows_count, result); else - throw Exception(ErrorCodes::ILLEGAL_COLUMN, - "Illegal column for second argument of function {}, must be Date, Date32, DateTime or DateTime64", - name); + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column for second argument of function {}, must be Date, Date32, DateTime or DateTime64", name); } template void dispatchConstForSecondColumn( - T1 x, const IColumn & y, + T1 x, const IColumn & col_y, const DateLUTImpl & timezone_x, const DateLUTImpl & timezone_y, + size_t input_rows_count, ColumnInt64::Container & result) const { - if (const auto * y_vec_16 = checkAndGetColumn(&y)) - constantVector(x, *y_vec_16, timezone_x, timezone_y, result); - else if (const auto * y_vec_32 = checkAndGetColumn(&y)) - constantVector(x, *y_vec_32, timezone_x, timezone_y, result); - else if (const auto * y_vec_32_s = checkAndGetColumn(&y)) - constantVector(x, *y_vec_32_s, timezone_x, timezone_y, result); - else if (const auto * y_vec_64 = checkAndGetColumn(&y)) - constantVector(x, *y_vec_64, timezone_x, timezone_y, result); + if (const auto * y_vec_16 = checkAndGetColumn(&col_y)) + constantVector(x, *y_vec_16, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_vec_32 = checkAndGetColumn(&col_y)) + constantVector(x, *y_vec_32, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_vec_32_s = checkAndGetColumn(&col_y)) + constantVector(x, *y_vec_32_s, timezone_x, timezone_y, input_rows_count, result); + else if (const auto * y_vec_64 = checkAndGetColumn(&col_y)) + constantVector(x, *y_vec_64, timezone_x, timezone_y, input_rows_count, result); else - throw Exception(ErrorCodes::ILLEGAL_COLUMN, - "Illegal column for second argument of function {}, must be Date, Date32, DateTime or DateTime64", - name); + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column for second argument of function {}, must be Date, Date32, DateTime or DateTime64", name); } template void vectorVector( const LeftColumnType & x, const RightColumnType & y, const DateLUTImpl & timezone_x, const DateLUTImpl & timezone_y, + size_t input_rows_count, ColumnInt64::Container & result) const { const auto & x_data = x.getData(); @@ -130,14 +126,15 @@ public: const auto transform_x = TransformDateTime64(getScale(x)); const auto transform_y = TransformDateTime64(getScale(y)); - for (size_t i = 0, size = x.size(); i < size; ++i) - result[i] = calculate(transform_x, transform_y, x_data[i], y_data[i], timezone_x, timezone_y); + for (size_t i = 0; i < input_rows_count; ++i) + result[i] = calculate(transform_x, transform_y, x_data[i], y_data[i], timezone_x, timezone_y); } template void vectorConstant( const LeftColumnType & x, T2 y, const DateLUTImpl & timezone_x, const DateLUTImpl & timezone_y, + size_t input_rows_count, ColumnInt64::Container & result) const { const auto & x_data = x.getData(); @@ -145,7 +142,7 @@ public: const auto transform_y = TransformDateTime64(getScale(y)); const auto y_value = stripDecimalFieldValue(y); - for (size_t i = 0, size = x.size(); i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) result[i] = calculate(transform_x, transform_y, x_data[i], y_value, timezone_x, timezone_y); } @@ -153,6 +150,7 @@ public: void constantVector( T1 x, const RightColumnType & y, const DateLUTImpl & timezone_x, const DateLUTImpl & timezone_y, + size_t input_rows_count, ColumnInt64::Container & result) const { const auto & y_data = y.getData(); @@ -160,20 +158,22 @@ public: const auto transform_y = TransformDateTime64(getScale(y)); const auto x_value = stripDecimalFieldValue(x); - for (size_t i = 0, size = y.size(); i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) result[i] = calculate(transform_x, transform_y, x_value, y_data[i], timezone_x, timezone_y); } template Int64 calculate(const TransformX & transform_x, const TransformY & transform_y, T1 x, T2 y, const DateLUTImpl & timezone_x, const DateLUTImpl & timezone_y) const { + auto res = static_cast(transform_y.execute(y, timezone_y)) - static_cast(transform_x.execute(x, timezone_x)); + if constexpr (is_diff) - return static_cast(transform_y.execute(y, timezone_y)) - - static_cast(transform_x.execute(x, timezone_x)); + { + return res; + } else { - auto res = static_cast(transform_y.execute(y, timezone_y)) - - static_cast(transform_x.execute(x, timezone_x)); + /// Adjust res: DateTimeComponentsWithFractionalPart a_comp; DateTimeComponentsWithFractionalPart b_comp; Int64 adjust_value; @@ -332,95 +332,73 @@ public: static constexpr auto name = is_relative ? "dateDiff" : "age"; static FunctionPtr create(ContextPtr) { return std::make_shared(); } - String getName() const override - { - return name; - } + String getName() const override { return name; } bool isVariadic() const override { return true; } bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } size_t getNumberOfArguments() const override { return 0; } + bool useDefaultImplementationForConstants() const override { return true; } + ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {0, 3}; } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - if (arguments.size() != 3 && arguments.size() != 4) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, - "Number of arguments for function {} doesn't match: passed {}, should be 3 or 4", - getName(), arguments.size()); + FunctionArgumentDescriptors mandatory_args{ + {"unit", static_cast(&isString), nullptr, "String"}, + {"startdate", static_cast(&isDateOrDate32OrDateTimeOrDateTime64), nullptr, "Date[32] or DateTime[64]"}, + {"enddate", static_cast(&isDateOrDate32OrDateTimeOrDateTime64), nullptr, "Date[32] or DateTime[64]"}, + }; - if (!isString(arguments[0])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "First argument for function {} (unit) must be String", - getName()); + FunctionArgumentDescriptors optional_args{ + {"timezone", static_cast(&isString), nullptr, "String"}, + }; - if (!isDate(arguments[1]) && !isDate32(arguments[1]) && !isDateTime(arguments[1]) && !isDateTime64(arguments[1])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Second argument for function {} must be Date, Date32, DateTime or DateTime64", - getName()); - - if (!isDate(arguments[2]) && !isDate32(arguments[2]) && !isDateTime(arguments[2]) && !isDateTime64(arguments[2])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Third argument for function {} must be Date, Date32, DateTime or DateTime64", - getName() - ); - - if (arguments.size() == 4 && !isString(arguments[3])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Fourth argument for function {} (timezone) must be String", - getName()); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(); } - bool useDefaultImplementationForConstants() const override { return true; } - ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {0, 3}; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { - const auto * unit_column = checkAndGetColumnConst(arguments[0].column.get()); - if (!unit_column) - throw Exception(ErrorCodes::ILLEGAL_COLUMN, - "First argument for function {} must be constant String", - getName()); + const auto * col_unit = checkAndGetColumnConst(arguments[0].column.get()); + if (!col_unit) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "First argument for function {} must be constant String", getName()); - String unit = Poco::toLower(unit_column->getValue()); + String unit = Poco::toLower(col_unit->getValue()); - const IColumn & x = *arguments[1].column; - const IColumn & y = *arguments[2].column; + const IColumn & col_x = *arguments[1].column; + const IColumn & col_y = *arguments[2].column; - size_t rows = input_rows_count; - auto res = ColumnInt64::create(rows); + auto col_res = ColumnInt64::create(input_rows_count); const auto & timezone_x = extractTimeZoneFromFunctionArguments(arguments, 3, 1); const auto & timezone_y = extractTimeZoneFromFunctionArguments(arguments, 3, 2); if (unit == "year" || unit == "years" || unit == "yy" || unit == "yyyy") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "quarter" || unit == "quarters" || unit == "qq" || unit == "q") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "month" || unit == "months" || unit == "mm" || unit == "m") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "week" || unit == "weeks" || unit == "wk" || unit == "ww") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "day" || unit == "days" || unit == "dd" || unit == "d") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "hour" || unit == "hours" || unit == "hh" || unit == "h") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "minute" || unit == "minutes" || unit == "mi" || unit == "n") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "second" || unit == "seconds" || unit == "ss" || unit == "s") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "millisecond" || unit == "milliseconds" || unit == "ms") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "microsecond" || unit == "microseconds" || unit == "us" || unit == "u") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else if (unit == "nanosecond" || unit == "nanoseconds" || unit == "ns") - impl.template dispatchForColumns>(x, y, timezone_x, timezone_y, res->getData()); + impl.template dispatchForColumns>(col_x, col_y, timezone_x, timezone_y, input_rows_count, col_res->getData()); else - throw Exception(ErrorCodes::BAD_ARGUMENTS, - "Function {} does not support '{}' unit", getName(), unit); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Function {} does not support '{}' unit", getName(), unit); - return res; + return col_res; } private: DateDiffImpl impl{name}; @@ -437,50 +415,35 @@ public: static constexpr auto name = "timeDiff"; static FunctionPtr create(ContextPtr) { return std::make_shared(); } - String getName() const override - { - return name; - } - + String getName() const override { return name; } + bool useDefaultImplementationForConstants() const override { return true; } + ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {}; } bool isVariadic() const override { return false; } bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } size_t getNumberOfArguments() const override { return 2; } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - if (arguments.size() != 2) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, - "Number of arguments for function {} doesn't match: passed {}, should be 2", - getName(), arguments.size()); + FunctionArgumentDescriptors args{ + {"first_datetime", static_cast(&isDateOrDate32OrDateTimeOrDateTime64), nullptr, "Date[32] or DateTime[64]"}, + {"second_datetime", static_cast(&isDateOrDate32OrDateTimeOrDateTime64), nullptr, "Date[32] or DateTime[64]"}, + }; - if (!isDate(arguments[0]) && !isDate32(arguments[0]) && !isDateTime(arguments[0]) && !isDateTime64(arguments[0])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "First argument for function {} must be Date, Date32, DateTime or DateTime64", - getName()); - - if (!isDate(arguments[1]) && !isDate32(arguments[1]) && !isDateTime(arguments[1]) && !isDateTime64(arguments[1])) - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, - "Second argument for function {} must be Date, Date32, DateTime or DateTime64", - getName() - ); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } - bool useDefaultImplementationForConstants() const override { return true; } - ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {}; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { - const IColumn & x = *arguments[0].column; - const IColumn & y = *arguments[1].column; + const IColumn & col_x = *arguments[0].column; + const IColumn & col_y = *arguments[1].column; - size_t rows = input_rows_count; - auto res = ColumnInt64::create(rows); + auto col_res = ColumnInt64::create(input_rows_count); - impl.dispatchForColumns>(x, y, DateLUT::instance(), DateLUT::instance(), res->getData()); + impl.dispatchForColumns>(col_x, col_y, DateLUT::instance(), DateLUT::instance(), input_rows_count, col_res->getData()); - return res; + return col_res; } private: DateDiffImpl impl{name}; @@ -490,7 +453,7 @@ private: REGISTER_FUNCTION(DateDiff) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); factory.registerAlias("date_diff", FunctionDateDiff::name); factory.registerAlias("DATE_DIFF", FunctionDateDiff::name); factory.registerAlias("timestampDiff", FunctionDateDiff::name); @@ -509,12 +472,12 @@ Example: )", .examples{ {"typical", "SELECT timeDiff(UTCTimestamp(), now());", ""}}, - .categories{"Dates and Times"}}, FunctionFactory::CaseInsensitive); + .categories{"Dates and Times"}}, FunctionFactory::Case::Insensitive); } REGISTER_FUNCTION(Age) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/dateName.cpp b/src/Functions/dateName.cpp index c06dfe15dc4..846cb87f1ee 100644 --- a/src/Functions/dateName.cpp +++ b/src/Functions/dateName.cpp @@ -109,14 +109,14 @@ public: ColumnPtr executeImpl( const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, - [[maybe_unused]] size_t input_rows_count) const override + size_t input_rows_count) const override { ColumnPtr res; - if (!((res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)))) + if (!((res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)))) throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of function {}, must be Date or DateTime.", @@ -127,7 +127,7 @@ public: } template - ColumnPtr executeType(const ColumnsWithTypeAndName & arguments, const DataTypePtr &) const + ColumnPtr executeType(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const { auto * times = checkAndGetColumn(arguments[1].column.get()); if (!times) @@ -144,7 +144,7 @@ public: String date_part = date_part_column->getValue(); const DateLUTImpl * time_zone_tmp; - if (std::is_same_v || std::is_same_v) + if constexpr (std::is_same_v || std::is_same_v) time_zone_tmp = &extractTimeZoneFromFunctionArguments(arguments, 2, 1); else time_zone_tmp = &DateLUT::instance(); @@ -175,7 +175,7 @@ public: using TimeType = DateTypeToTimeType; callOnDatePartWriter(date_part, [&](const auto & writer) { - for (size_t i = 0; i < times_data.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { if constexpr (std::is_same_v) { @@ -354,7 +354,7 @@ private: REGISTER_FUNCTION(DateName) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/dateTimeToSnowflakeID.cpp b/src/Functions/dateTimeToSnowflakeID.cpp index 968a7628ca5..c48f8c13152 100644 --- a/src/Functions/dateTimeToSnowflakeID.cpp +++ b/src/Functions/dateTimeToSnowflakeID.cpp @@ -43,7 +43,7 @@ public: FunctionArgumentDescriptors optional_args{ {"epoch", static_cast(&isNativeUInt), isColumnConst, "const UInt*"} }; - validateFunctionArgumentTypes(*this, arguments, args, optional_args); + validateFunctionArguments(*this, arguments, args, optional_args); return std::make_shared(); } @@ -91,7 +91,7 @@ public: FunctionArgumentDescriptors optional_args{ {"epoch", static_cast(&isNativeUInt), isColumnConst, "const UInt*"} }; - validateFunctionArgumentTypes(*this, arguments, args, optional_args); + validateFunctionArguments(*this, arguments, args, optional_args); return std::make_shared(); } diff --git a/src/Functions/date_trunc.cpp b/src/Functions/date_trunc.cpp index b8c60dd164e..dd3ea0b877b 100644 --- a/src/Functions/date_trunc.cpp +++ b/src/Functions/date_trunc.cpp @@ -178,7 +178,7 @@ REGISTER_FUNCTION(DateTrunc) factory.registerFunction(); /// Compatibility alias. - factory.registerAlias("DATE_TRUNC", "dateTrunc", FunctionFactory::CaseInsensitive); + factory.registerAlias("DATE_TRUNC", "dateTrunc", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/decodeHTMLComponent.cpp b/src/Functions/decodeHTMLComponent.cpp index 00a601b77a6..d36bee534a8 100644 --- a/src/Functions/decodeHTMLComponent.cpp +++ b/src/Functions/decodeHTMLComponent.cpp @@ -28,20 +28,20 @@ namespace const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { /// The size of result is always not more than the size of source. /// Because entities decodes to the shorter byte sequence. /// Example: &#xx... &#xx... will decode to UTF-8 byte sequence not longer than 4 bytes. res_data.resize(data.size()); - size_t size = offsets.size(); - res_offsets.resize(size); + res_offsets.resize(input_rows_count); size_t prev_offset = 0; size_t res_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * src_data = reinterpret_cast(&data[prev_offset]); size_t src_size = offsets[i] - prev_offset; @@ -55,7 +55,7 @@ namespace res_data.resize(res_offset); } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Function decodeHTMLComponent cannot work with FixedString argument"); } @@ -64,7 +64,6 @@ namespace static const int max_legal_unicode_value = 0x10FFFF; static const int max_decimal_length_of_unicode_point = 7; /// 1114111 - static size_t execute(const char * src, size_t src_size, char * dst) { const char * src_pos = src; diff --git a/src/Functions/decodeXMLComponent.cpp b/src/Functions/decodeXMLComponent.cpp index cbbe46fcb8c..8743aa4759d 100644 --- a/src/Functions/decodeXMLComponent.cpp +++ b/src/Functions/decodeXMLComponent.cpp @@ -27,20 +27,20 @@ namespace const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { /// The size of result is always not more than the size of source. /// Because entities decodes to the shorter byte sequence. /// Example: &#xx... &#xx... will decode to UTF-8 byte sequence not longer than 4 bytes. res_data.resize(data.size()); - size_t size = offsets.size(); - res_offsets.resize(size); + res_offsets.resize(input_rows_count); size_t prev_offset = 0; size_t res_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * src_data = reinterpret_cast(&data[prev_offset]); size_t src_size = offsets[i] - prev_offset; @@ -54,7 +54,7 @@ namespace res_data.resize(res_offset); } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Function decodeXMLComponent cannot work with FixedString argument"); } diff --git a/src/Functions/degrees.cpp b/src/Functions/degrees.cpp index 3aa20a77a0d..94b5ce3682c 100644 --- a/src/Functions/degrees.cpp +++ b/src/Functions/degrees.cpp @@ -7,23 +7,25 @@ namespace DB { namespace { - struct DegreesName - { - static constexpr auto name = "degrees"; - }; - Float64 degrees(Float64 r) - { - Float64 degrees = r * (180 / M_PI); - return degrees; - } +struct DegreesName +{ + static constexpr auto name = "degrees"; +}; + +Float64 degrees(Float64 r) +{ + Float64 degrees = r * (180 / M_PI); + return degrees; +} + +using FunctionDegrees = FunctionMathUnary>; - using FunctionDegrees = FunctionMathUnary>; } REGISTER_FUNCTION(Degrees) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/encodeXMLComponent.cpp b/src/Functions/encodeXMLComponent.cpp index 64d85ecaeb8..a22917838b7 100644 --- a/src/Functions/encodeXMLComponent.cpp +++ b/src/Functions/encodeXMLComponent.cpp @@ -25,17 +25,17 @@ namespace const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { /// 6 is the maximum size amplification (the maximum length of encoded entity: ") res_data.resize(data.size() * 6); - size_t size = offsets.size(); - res_offsets.resize(size); + res_offsets.resize(input_rows_count); size_t prev_offset = 0; size_t res_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * src_data = reinterpret_cast(&data[prev_offset]); size_t src_size = offsets[i] - prev_offset; @@ -49,7 +49,7 @@ namespace res_data.resize(res_offset); } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Function encodeXML cannot work with FixedString argument"); } diff --git a/src/Functions/exp.cpp b/src/Functions/exp.cpp index d352cda7460..e67cbd6d819 100644 --- a/src/Functions/exp.cpp +++ b/src/Functions/exp.cpp @@ -36,7 +36,7 @@ using FunctionExp = FunctionMathUnary>; REGISTER_FUNCTION(Exp) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/extractAll.cpp b/src/Functions/extractAll.cpp index 5801a7b8f4f..4a3eb32474c 100644 --- a/src/Functions/extractAll.cpp +++ b/src/Functions/extractAll.cpp @@ -59,7 +59,7 @@ public: {"pattern", static_cast(&isString), isColumnConst, "const String"} }; - validateFunctionArgumentTypes(func, arguments, mandatory_args); + validateFunctionArguments(func, arguments, mandatory_args); } static constexpr auto strings_argument_position = 0uz; diff --git a/src/Functions/extractAllGroups.h b/src/Functions/extractAllGroups.h index dfcd0e31715..7732855b211 100644 --- a/src/Functions/extractAllGroups.h +++ b/src/Functions/extractAllGroups.h @@ -74,7 +74,7 @@ public: {"haystack", static_cast(&isStringOrFixedString), nullptr, "const String or const FixedString"}, {"needle", static_cast(&isStringOrFixedString), isColumnConst, "const String or const FixedString"}, }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); /// Two-dimensional array of strings, each `row` of top array represents matching groups. return std::make_shared(std::make_shared(std::make_shared())); diff --git a/src/Functions/extractAllGroupsVertical.cpp b/src/Functions/extractAllGroupsVertical.cpp index 87a0b4cf7bc..6a968d89354 100644 --- a/src/Functions/extractAllGroupsVertical.cpp +++ b/src/Functions/extractAllGroupsVertical.cpp @@ -18,7 +18,7 @@ namespace DB REGISTER_FUNCTION(ExtractAllGroupsVertical) { factory.registerFunction>(); - factory.registerAlias("extractAllGroups", VerticalImpl::Name, FunctionFactory::CaseSensitive); + factory.registerAlias("extractAllGroups", VerticalImpl::Name); } } diff --git a/src/Functions/extractGroups.cpp b/src/Functions/extractGroups.cpp index f62352af0bd..ac6266a2e82 100644 --- a/src/Functions/extractGroups.cpp +++ b/src/Functions/extractGroups.cpp @@ -48,7 +48,7 @@ public: {"haystack", static_cast(&isStringOrFixedString), nullptr, "const String or const FixedString"}, {"needle", static_cast(&isStringOrFixedString), isColumnConst, "const String or const FixedString"}, }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(std::make_shared()); } diff --git a/src/Functions/factorial.cpp b/src/Functions/factorial.cpp index 7ff9126c004..9b319caad63 100644 --- a/src/Functions/factorial.cpp +++ b/src/Functions/factorial.cpp @@ -106,7 +106,7 @@ The factorial of 0 is 1. Likewise, the factorial() function returns 1 for any ne )", .examples{{"factorial", "SELECT factorial(10)", ""}}, .categories{"Mathematical"}}, - FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/filesystem.cpp b/src/Functions/filesystem.cpp index 9fbf9b0cbe7..9b168f3f088 100644 --- a/src/Functions/filesystem.cpp +++ b/src/Functions/filesystem.cpp @@ -91,7 +91,7 @@ public: auto col_res = ColumnVector::create(col_str->size()); auto & data = col_res->getData(); - for (size_t i = 0; i < col_str->size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { auto disk_name = col_str->getDataAt(i).toString(); if (auto it = disk_map.find(disk_name); it != disk_map.end()) diff --git a/src/Functions/formatDateTime.cpp b/src/Functions/formatDateTime.cpp index 1c5dc05c5aa..f33b7849a43 100644 --- a/src/Functions/formatDateTime.cpp +++ b/src/Functions/formatDateTime.cpp @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -782,9 +783,9 @@ public: static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } explicit FunctionFormatDateTimeImpl(ContextPtr context) - : mysql_M_is_month_name(context->getSettings().formatdatetime_parsedatetime_m_is_month_name) - , mysql_f_prints_single_zero(context->getSettings().formatdatetime_f_prints_single_zero) - , mysql_format_ckl_without_leading_zeros(context->getSettings().formatdatetime_format_without_leading_zeros) + : mysql_M_is_month_name(context->getSettingsRef().formatdatetime_parsedatetime_m_is_month_name) + , mysql_f_prints_single_zero(context->getSettingsRef().formatdatetime_f_prints_single_zero) + , mysql_format_ckl_without_leading_zeros(context->getSettingsRef().formatdatetime_format_without_leading_zeros) { } @@ -847,7 +848,7 @@ public: return std::make_shared(); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, [[maybe_unused]] size_t input_rows_count) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override { ColumnPtr res; if constexpr (support_integer == SupportInteger::Yes) @@ -861,17 +862,17 @@ public: if (!castType(arguments[0].type.get(), [&](const auto & type) { using FromDataType = std::decay_t; - if (!(res = executeType(arguments, result_type))) + if (!(res = executeType(arguments, result_type, input_rows_count))) throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of function {}, must be Integer, Date, Date32, DateTime or DateTime64.", arguments[0].column->getName(), getName()); return true; })) { - if (!((res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)))) + if (!((res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)))) throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of function {}, must be Integer or DateTime.", arguments[0].column->getName(), getName()); @@ -880,10 +881,10 @@ public: } else { - if (!((res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)) - || (res = executeType(arguments, result_type)))) + if (!((res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)) + || (res = executeType(arguments, result_type, input_rows_count)))) throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of function {}, must be Date or DateTime.", arguments[0].column->getName(), getName()); @@ -893,7 +894,7 @@ public: } template - ColumnPtr executeType(const ColumnsWithTypeAndName & arguments, const DataTypePtr &) const + ColumnPtr executeType(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const { auto non_const_datetime = arguments[0].column->convertToFullColumnIfConst(); auto * times = checkAndGetColumn(non_const_datetime.get()); @@ -954,13 +955,11 @@ public: else time_zone = &DateLUT::instance(); - const auto & vec = times->getData(); - auto col_res = ColumnString::create(); auto & res_data = col_res->getChars(); auto & res_offsets = col_res->getOffsets(); - res_data.resize(vec.size() * (out_template_size + 1)); - res_offsets.resize(vec.size()); + res_data.resize(input_rows_count * (out_template_size + 1)); + res_offsets.resize(input_rows_count); if constexpr (format_syntax == FormatSyntax::MySQL) { @@ -989,9 +988,11 @@ public: } } + const auto & vec = times->getData(); + auto * begin = reinterpret_cast(res_data.data()); auto * pos = begin; - for (size_t i = 0; i < vec.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { if (!const_time_zone_column && arguments.size() > 2) { @@ -1833,10 +1834,10 @@ using FunctionFromUnixTimestampInJodaSyntax = FunctionFormatDateTimeImpl(); - factory.registerAlias("DATE_FORMAT", FunctionFormatDateTime::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("DATE_FORMAT", FunctionFormatDateTime::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); - factory.registerAlias("FROM_UNIXTIME", FunctionFromUnixTimestamp::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("FROM_UNIXTIME", FunctionFromUnixTimestamp::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); diff --git a/src/Functions/formatQuery.cpp b/src/Functions/formatQuery.cpp index 3b632147864..4e3f302ce36 100644 --- a/src/Functions/formatQuery.cpp +++ b/src/Functions/formatQuery.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include @@ -54,7 +55,7 @@ public: FunctionArgumentDescriptors args{ {"query", static_cast(&isString), nullptr, "String"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); DataTypePtr string_type = std::make_shared(); if (error_handling == ErrorHandling::Null) @@ -74,7 +75,7 @@ public: if (const ColumnString * col_query_string = checkAndGetColumn(col_query.get())) { auto col_res = ColumnString::create(); - formatVector(col_query_string->getChars(), col_query_string->getOffsets(), col_res->getChars(), col_res->getOffsets(), col_null_map); + formatVector(col_query_string->getChars(), col_query_string->getOffsets(), col_res->getChars(), col_res->getOffsets(), col_null_map, input_rows_count); if (error_handling == ErrorHandling::Null) return ColumnNullable::create(std::move(col_res), std::move(col_null_map)); @@ -91,16 +92,16 @@ private: const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets, - ColumnUInt8::MutablePtr & res_null_map) const + ColumnUInt8::MutablePtr & res_null_map, + size_t input_rows_count) const { - const size_t size = offsets.size(); - res_offsets.resize(size); + res_offsets.resize(input_rows_count); res_data.resize(data.size()); size_t prev_offset = 0; size_t res_data_size = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * begin = reinterpret_cast(&data[prev_offset]); const char * end = begin + offsets[i] - prev_offset - 1; diff --git a/src/Functions/formatReadable.h b/src/Functions/formatReadable.h index 487ec9d79d0..9161ab43e28 100644 --- a/src/Functions/formatReadable.h +++ b/src/Functions/formatReadable.h @@ -55,19 +55,19 @@ public: bool useDefaultImplementationForConstants() const override { return true; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { ColumnPtr res; - if (!((res = executeType(arguments)) - || (res = executeType(arguments)) - || (res = executeType(arguments)) - || (res = executeType(arguments)) - || (res = executeType(arguments)) - || (res = executeType(arguments)) - || (res = executeType(arguments)) - || (res = executeType(arguments)) - || (res = executeType(arguments)) - || (res = executeType(arguments)))) + if (!((res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)) + || (res = executeType(arguments, input_rows_count)))) throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of argument of function {}", arguments[0].column->getName(), getName()); @@ -76,7 +76,7 @@ public: private: template - ColumnPtr executeType(const ColumnsWithTypeAndName & arguments) const + ColumnPtr executeType(const ColumnsWithTypeAndName & arguments, size_t input_rows_count) const { if (const ColumnVector * col_from = checkAndGetColumn>(arguments[0].column.get())) { @@ -85,13 +85,12 @@ private: const typename ColumnVector::Container & vec_from = col_from->getData(); ColumnString::Chars & data_to = col_to->getChars(); ColumnString::Offsets & offsets_to = col_to->getOffsets(); - size_t size = vec_from.size(); - data_to.resize(size * 2); - offsets_to.resize(size); + data_to.resize(input_rows_count * 2); + offsets_to.resize(input_rows_count); WriteBufferFromVector buf_to(data_to); - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { Impl::format(static_cast(vec_from[i]), buf_to); writeChar(0, buf_to); diff --git a/src/Functions/formatReadableDecimalSize.cpp b/src/Functions/formatReadableDecimalSize.cpp index 1aa5abc526e..9298360aebc 100644 --- a/src/Functions/formatReadableDecimalSize.cpp +++ b/src/Functions/formatReadableDecimalSize.cpp @@ -29,8 +29,7 @@ Accepts the size (number of bytes). Returns a rounded size with a suffix (KB, MB .examples{ {"formatReadableDecimalSize", "SELECT formatReadableDecimalSize(1000)", ""}}, .categories{"OtherFunctions"} - }, - FunctionFactory::CaseSensitive); + }); } } diff --git a/src/Functions/formatReadableSize.cpp b/src/Functions/formatReadableSize.cpp index 5c11603e9d7..ee66a0396df 100644 --- a/src/Functions/formatReadableSize.cpp +++ b/src/Functions/formatReadableSize.cpp @@ -22,7 +22,7 @@ namespace REGISTER_FUNCTION(FormatReadableSize) { factory.registerFunction>(); - factory.registerAlias("FORMAT_BYTES", Impl::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("FORMAT_BYTES", Impl::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/fromDaysSinceYearZero.cpp b/src/Functions/fromDaysSinceYearZero.cpp index b98c587d172..e1ba9ea533e 100644 --- a/src/Functions/fromDaysSinceYearZero.cpp +++ b/src/Functions/fromDaysSinceYearZero.cpp @@ -54,7 +54,7 @@ public: { FunctionArgumentDescriptors args{{"days", static_cast(&isNativeInteger), nullptr, "Integer"}}; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } @@ -125,7 +125,7 @@ The calculation is the same as in MySQL's FROM_DAYS() function. .examples{{"typical", "SELECT fromDaysSinceYearZero32(713569)", "2023-09-08"}}, .categories{"Dates and Times"}}); - factory.registerAlias("FROM_DAYS", FunctionFromDaysSinceYearZero::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("FROM_DAYS", FunctionFromDaysSinceYearZero::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/generateSnowflakeID.cpp b/src/Functions/generateSnowflakeID.cpp index 8ac010deafc..c95e3edf4ca 100644 --- a/src/Functions/generateSnowflakeID.cpp +++ b/src/Functions/generateSnowflakeID.cpp @@ -4,10 +4,10 @@ #include #include #include +#include #include #include "base/types.h" - namespace DB { @@ -96,10 +96,11 @@ struct SnowflakeIdRange /// 1. calculate Snowflake ID by current timestamp (`now`) /// 2. `begin = max(available, now)` /// 3. Calculate `end = begin + input_rows_count` handling `machine_seq_num` overflow -SnowflakeIdRange getRangeOfAvailableIds(const SnowflakeId & available, size_t input_rows_count) +SnowflakeIdRange getRangeOfAvailableIds(const SnowflakeId & available, uint64_t machine_id, size_t input_rows_count) + { /// 1. `now` - SnowflakeId begin = {.timestamp = getTimestamp(), .machine_id = getMachineId(), .machine_seq_num = 0}; + SnowflakeId begin = {.timestamp = getTimestamp(), .machine_id = machine_id, .machine_seq_num = 0}; /// 2. `begin` if (begin.timestamp <= available.timestamp) @@ -128,13 +129,13 @@ struct Data /// Guarantee counter monotonicity within one timestamp across all threads generating Snowflake IDs simultaneously. static inline std::atomic lowest_available_snowflake_id = 0; - SnowflakeId reserveRange(size_t input_rows_count) + SnowflakeId reserveRange(uint64_t machine_id, size_t input_rows_count) { uint64_t available_snowflake_id = lowest_available_snowflake_id.load(); SnowflakeIdRange range; do { - range = getRangeOfAvailableIds(toSnowflakeId(available_snowflake_id), input_rows_count); + range = getRangeOfAvailableIds(toSnowflakeId(available_snowflake_id), machine_id, input_rows_count); } while (!lowest_available_snowflake_id.compare_exchange_weak(available_snowflake_id, fromSnowflakeId(range.end))); /// CAS failed --> another thread updated `lowest_available_snowflake_id` and we re-try @@ -165,24 +166,32 @@ public: { FunctionArgumentDescriptors mandatory_args; FunctionArgumentDescriptors optional_args{ - {"expr", nullptr, nullptr, "Arbitrary expression"} + {"expr", nullptr, nullptr, "Arbitrary expression"}, + {"machine_id", static_cast(&isNativeUInt), static_cast(&isColumnConst), "const UInt*"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & /*arguments*/, const DataTypePtr &, size_t input_rows_count) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { auto col_res = ColumnVector::create(); typename ColumnVector::Container & vec_to = col_res->getData(); - if (input_rows_count != 0) + if (input_rows_count > 0) { vec_to.resize(input_rows_count); + uint64_t machine_id = getMachineId(); + if (arguments.size() == 2) + { + machine_id = arguments[1].column->getUInt(0); + machine_id &= (1ull << machine_id_bits_count) - 1; + } + Data data; - SnowflakeId snowflake_id = data.reserveRange(input_rows_count); /// returns begin of available snowflake ids range + SnowflakeId snowflake_id = data.reserveRange(machine_id, input_rows_count); for (UInt64 & to_row : vec_to) { @@ -208,10 +217,13 @@ public: REGISTER_FUNCTION(GenerateSnowflakeID) { FunctionDocumentation::Description description = R"(Generates a Snowflake ID. The generated Snowflake ID contains the current Unix timestamp in milliseconds (41 + 1 top zero bits), followed by a machine id (10 bits), and a counter (12 bits) to distinguish IDs within a millisecond. For any given timestamp (unix_ts_ms), the counter starts at 0 and is incremented by 1 for each new Snowflake ID until the timestamp changes. In case the counter overflows, the timestamp field is incremented by 1 and the counter is reset to 0. Function generateSnowflakeID guarantees that the counter field within a timestamp increments monotonically across all function invocations in concurrently running threads and queries.)"; - FunctionDocumentation::Syntax syntax = "generateSnowflakeID([expression])"; - FunctionDocumentation::Arguments arguments = {{"expression", "The expression is used to bypass common subexpression elimination if the function is called multiple times in a query but otherwise ignored. Optional."}}; + FunctionDocumentation::Syntax syntax = "generateSnowflakeID([expression, [machine_id]])"; + FunctionDocumentation::Arguments arguments = { + {"expression", "The expression is used to bypass common subexpression elimination if the function is called multiple times in a query but otherwise ignored. Optional."}, + {"machine_id", "A machine ID, the lowest 10 bits are used. Optional."} + }; FunctionDocumentation::ReturnedValue returned_value = "A value of type UInt64"; - FunctionDocumentation::Examples examples = {{"single", "SELECT generateSnowflakeID()", "7201148511606784000"}, {"multiple", "SELECT generateSnowflakeID(1), generateSnowflakeID(2)", ""}}; + FunctionDocumentation::Examples examples = {{"no_arguments", "SELECT generateSnowflakeID()", "7201148511606784000"}, {"with_machine_id", "SELECT generateSnowflakeID(1)", "7201148511606784001"}, {"with_expression_and_machine_id", "SELECT generateSnowflakeID('some_expression', 1)", "7201148511606784002"}}; FunctionDocumentation::Categories categories = {"Snowflake ID"}; factory.registerFunction({description, syntax, arguments, returned_value, examples, categories}); diff --git a/src/Functions/generateULID.cpp b/src/Functions/generateULID.cpp index f2f2d8ae3b9..933618ccec3 100644 --- a/src/Functions/generateULID.cpp +++ b/src/Functions/generateULID.cpp @@ -85,8 +85,7 @@ The function returns a value of type FixedString(26). {"ulid", "SELECT generateULID()", ""}, {"multiple", "SELECT generateULID(1), generateULID(2)", ""}}, .categories{"ULID"} - }, - FunctionFactory::CaseSensitive); + }); } } diff --git a/src/Functions/generateUUIDv4.cpp b/src/Functions/generateUUIDv4.cpp index b0fec43fe94..a928f9009c8 100644 --- a/src/Functions/generateUUIDv4.cpp +++ b/src/Functions/generateUUIDv4.cpp @@ -30,9 +30,9 @@ public: { FunctionArgumentDescriptors mandatory_args; FunctionArgumentDescriptors optional_args{ - {"expr", nullptr, nullptr, "Arbitrary Expression"} + {"expr", nullptr, nullptr, "any type"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(); } diff --git a/src/Functions/generateUUIDv7.cpp b/src/Functions/generateUUIDv7.cpp index b226c0840f4..5dc6f1cde32 100644 --- a/src/Functions/generateUUIDv7.cpp +++ b/src/Functions/generateUUIDv7.cpp @@ -11,20 +11,6 @@ namespace /* Bit layouts of UUIDv7 -without counter: - 0 1 2 3 - 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 -├─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┤ -| unix_ts_ms | -├─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┤ -| unix_ts_ms | ver | rand_a | -├─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┤ -|var| rand_b | -├─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┤ -| rand_b | -└─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┴─┘ - -with counter: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 ├─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┼─┤ @@ -163,7 +149,7 @@ public: FunctionArgumentDescriptors optional_args{ {"expr", nullptr, nullptr, "Arbitrary expression"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(); } diff --git a/src/Functions/geohashDecode.cpp b/src/Functions/geohashDecode.cpp index b2454f5dffc..cace6c09fec 100644 --- a/src/Functions/geohashDecode.cpp +++ b/src/Functions/geohashDecode.cpp @@ -38,9 +38,12 @@ public: bool useDefaultImplementationForConstants() const override { return true; } bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - validateArgumentType(*this, arguments, 0, isStringOrFixedString, "string or fixed string"); + FunctionArgumentDescriptors args{ + {"encoded", static_cast(&isStringOrFixedString), nullptr, "String or FixedString"} + }; + validateFunctionArguments(*this, arguments, args); return std::make_shared( DataTypes{std::make_shared(), std::make_shared()}, @@ -48,21 +51,19 @@ public: } template - bool tryExecute(const IColumn * encoded_column, ColumnPtr & result_column) const + bool tryExecute(const IColumn * encoded_column, ColumnPtr & result_column, size_t input_rows_count) const { const auto * encoded = checkAndGetColumn(encoded_column); if (!encoded) return false; - const size_t count = encoded->size(); - - auto latitude = ColumnFloat64::create(count); - auto longitude = ColumnFloat64::create(count); + auto latitude = ColumnFloat64::create(input_rows_count); + auto longitude = ColumnFloat64::create(input_rows_count); ColumnFloat64::Container & lon_data = longitude->getData(); ColumnFloat64::Container & lat_data = latitude->getData(); - for (size_t i = 0; i < count; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { std::string_view encoded_string = encoded->getDataAt(i).toView(); geohashDecode(encoded_string.data(), encoded_string.size(), &lon_data[i], &lat_data[i]); @@ -76,13 +77,13 @@ public: return true; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const IColumn * encoded = arguments[0].column.get(); ColumnPtr res_column; - if (tryExecute(encoded, res_column) || - tryExecute(encoded, res_column)) + if (tryExecute(encoded, res_column, input_rows_count) || + tryExecute(encoded, res_column, input_rows_count)) return res_column; throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Unsupported argument type:{} of argument of function {}", diff --git a/src/Functions/geohashEncode.cpp b/src/Functions/geohashEncode.cpp index 7c353b822aa..c49acddd81f 100644 --- a/src/Functions/geohashEncode.cpp +++ b/src/Functions/geohashEncode.cpp @@ -17,7 +17,6 @@ namespace DB namespace ErrorCodes { extern const int LOGICAL_ERROR; - extern const int TOO_MANY_ARGUMENTS_FOR_FUNCTION; } namespace @@ -40,24 +39,21 @@ public: bool useDefaultImplementationForConstants() const override { return true; } bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - validateArgumentType(*this, arguments, 0, isFloat, "float"); - validateArgumentType(*this, arguments, 1, isFloat, "float"); - if (arguments.size() == 3) - { - validateArgumentType(*this, arguments, 2, isInteger, "integer"); - } - if (arguments.size() > 3) - { - throw Exception(ErrorCodes::TOO_MANY_ARGUMENTS_FOR_FUNCTION, "Too many arguments for function {} expected at most 3", - getName()); - } + FunctionArgumentDescriptors mandatory_args{ + {"longitude", static_cast(&isFloat), nullptr, "Float*"}, + {"latitude", static_cast(&isFloat), nullptr, "Float*"} + }; + FunctionArgumentDescriptors optional_args{ + {"precision", static_cast(&isInteger), nullptr, "(U)Int*"} + }; + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const IColumn * longitude = arguments[0].column.get(); const IColumn * latitude = arguments[1].column.get(); @@ -69,26 +65,24 @@ public: precision = arguments[2].column; ColumnPtr res_column; - vector(longitude, latitude, precision.get(), res_column); + vector(longitude, latitude, precision.get(), res_column, input_rows_count); return res_column; } private: - void vector(const IColumn * lon_column, const IColumn * lat_column, const IColumn * precision_column, ColumnPtr & result) const + void vector(const IColumn * lon_column, const IColumn * lat_column, const IColumn * precision_column, ColumnPtr & result, size_t input_rows_count) const { auto col_str = ColumnString::create(); ColumnString::Chars & out_vec = col_str->getChars(); ColumnString::Offsets & out_offsets = col_str->getOffsets(); - const size_t size = lat_column->size(); - - out_offsets.resize(size); - out_vec.resize(size * (GEOHASH_MAX_TEXT_LENGTH + 1)); + out_offsets.resize(input_rows_count); + out_vec.resize(input_rows_count * (GEOHASH_MAX_TEXT_LENGTH + 1)); char * begin = reinterpret_cast(out_vec.data()); char * pos = begin; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const Float64 longitude_value = lon_column->getFloat64(i); const Float64 latitude_value = lat_column->getFloat64(i); diff --git a/src/Functions/geohashesInBox.cpp b/src/Functions/geohashesInBox.cpp index ac8d4a6ad8f..9429903dda7 100644 --- a/src/Functions/geohashesInBox.cpp +++ b/src/Functions/geohashesInBox.cpp @@ -35,22 +35,25 @@ public: size_t getNumberOfArguments() const override { return 5; } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - validateArgumentType(*this, arguments, 0, isFloat, "float"); - validateArgumentType(*this, arguments, 1, isFloat, "float"); - validateArgumentType(*this, arguments, 2, isFloat, "float"); - validateArgumentType(*this, arguments, 3, isFloat, "float"); - validateArgumentType(*this, arguments, 4, isUInt8, "integer"); + FunctionArgumentDescriptors args{ + {"longitute_min", static_cast(&isFloat), nullptr, "Float*"}, + {"latitude_min", static_cast(&isFloat), nullptr, "Float*"}, + {"longitute_max", static_cast(&isFloat), nullptr, "Float*"}, + {"latitude_max", static_cast(&isFloat), nullptr, "Float*"}, + {"precision", static_cast(&isUInt8), nullptr, "UInt8"} + }; + validateFunctionArguments(*this, arguments, args); - if (!(arguments[0]->equals(*arguments[1]) && - arguments[0]->equals(*arguments[2]) && - arguments[0]->equals(*arguments[3]))) + if (!(arguments[0].type->equals(*arguments[1].type) && + arguments[0].type->equals(*arguments[2].type) && + arguments[0].type->equals(*arguments[3].type))) { throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type of argument of {} all coordinate arguments must have the same type, " - "instead they are:{}, {}, {}, {}.", getName(), arguments[0]->getName(), - arguments[1]->getName(), arguments[2]->getName(), arguments[3]->getName()); + "instead they are:{}, {}, {}, {}.", getName(), arguments[0].type->getName(), + arguments[1].type->getName(), arguments[2].type->getName(), arguments[3].type->getName()); } return std::make_shared(std::make_shared()); diff --git a/src/Functions/getClientHTTPHeader.cpp b/src/Functions/getClientHTTPHeader.cpp index ebd070a90b9..140f39d03b8 100644 --- a/src/Functions/getClientHTTPHeader.cpp +++ b/src/Functions/getClientHTTPHeader.cpp @@ -5,6 +5,7 @@ #include #include #include +#include namespace DB diff --git a/src/Functions/getMacro.cpp b/src/Functions/getMacro.cpp index 8172fc8ba2e..b7f0b34d652 100644 --- a/src/Functions/getMacro.cpp +++ b/src/Functions/getMacro.cpp @@ -53,6 +53,8 @@ public: /// getMacro may return different values on different shards/replicas, so it's not constant for distributed query bool isSuitableForConstantFolding() const override { return !is_distributed; } + bool isServerConstant() const override { return true; } + size_t getNumberOfArguments() const override { return 1; diff --git a/src/Functions/getScalar.cpp b/src/Functions/getScalar.cpp index 7196cbc0a36..5131dca962e 100644 --- a/src/Functions/getScalar.cpp +++ b/src/Functions/getScalar.cpp @@ -49,6 +49,8 @@ public: bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return false; } + bool isServerConstant() const override { return true; } + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { if (arguments.size() != 1 || !isString(arguments[0].type) || !arguments[0].column || !isColumnConst(*arguments[0].column)) @@ -105,6 +107,8 @@ public: bool isDeterministic() const override { return false; } + bool isServerConstant() const override { return true; } + bool isSuitableForConstantFolding() const override { return !is_distributed; } bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return false; } diff --git a/src/Functions/getSetting.cpp b/src/Functions/getSetting.cpp index d6b8946e760..aed6b2119e4 100644 --- a/src/Functions/getSetting.cpp +++ b/src/Functions/getSetting.cpp @@ -5,6 +5,7 @@ #include #include #include +#include namespace DB diff --git a/src/Functions/greatest.cpp b/src/Functions/greatest.cpp index 87a48c887b4..88539bda4a5 100644 --- a/src/Functions/greatest.cpp +++ b/src/Functions/greatest.cpp @@ -65,7 +65,7 @@ using FunctionGreatest = FunctionBinaryArithmetic; REGISTER_FUNCTION(Greatest) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/hasColumnInTable.cpp b/src/Functions/hasColumnInTable.cpp index 8ea16f688ee..00714997b4a 100644 --- a/src/Functions/hasColumnInTable.cpp +++ b/src/Functions/hasColumnInTable.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/Functions/hasSubsequence.cpp b/src/Functions/hasSubsequence.cpp index 4bcce53b4db..1426e8cb7a9 100644 --- a/src/Functions/hasSubsequence.cpp +++ b/src/Functions/hasSubsequence.cpp @@ -24,7 +24,7 @@ using FunctionHasSubsequence = HasSubsequenceImpl({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/hasSubsequenceCaseInsensitive.cpp b/src/Functions/hasSubsequenceCaseInsensitive.cpp index c93bbead58c..8e5751066a9 100644 --- a/src/Functions/hasSubsequenceCaseInsensitive.cpp +++ b/src/Functions/hasSubsequenceCaseInsensitive.cpp @@ -23,7 +23,7 @@ using FunctionHasSubsequenceCaseInsensitive = HasSubsequenceImpl({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/hasSubsequenceCaseInsensitiveUTF8.cpp b/src/Functions/hasSubsequenceCaseInsensitiveUTF8.cpp index 18438bc8b16..039af061bf5 100644 --- a/src/Functions/hasSubsequenceCaseInsensitiveUTF8.cpp +++ b/src/Functions/hasSubsequenceCaseInsensitiveUTF8.cpp @@ -25,7 +25,7 @@ using FunctionHasSubsequenceCaseInsensitiveUTF8 = HasSubsequenceImpl({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/hasSubsequenceUTF8.cpp b/src/Functions/hasSubsequenceUTF8.cpp index 7a22211eb8c..636fbfab85f 100644 --- a/src/Functions/hasSubsequenceUTF8.cpp +++ b/src/Functions/hasSubsequenceUTF8.cpp @@ -24,7 +24,7 @@ using FunctionHasSubsequenceUTF8 = HasSubsequenceImpl({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/hasToken.cpp b/src/Functions/hasToken.cpp index fa41abf2641..299a8a16b35 100644 --- a/src/Functions/hasToken.cpp +++ b/src/Functions/hasToken.cpp @@ -25,10 +25,10 @@ using FunctionHasTokenOrNull REGISTER_FUNCTION(HasToken) { factory.registerFunction(FunctionDocumentation - {.description="Performs lookup of needle in haystack using tokenbf_v1 index."}, FunctionFactory::CaseSensitive); + {.description="Performs lookup of needle in haystack using tokenbf_v1 index."}); factory.registerFunction(FunctionDocumentation - {.description="Performs lookup of needle in haystack using tokenbf_v1 index. Returns null if needle is ill-formed."}, FunctionFactory::CaseSensitive); + {.description="Performs lookup of needle in haystack using tokenbf_v1 index. Returns null if needle is ill-formed."}); } } diff --git a/src/Functions/hasTokenCaseInsensitive.cpp b/src/Functions/hasTokenCaseInsensitive.cpp index 32675b9384d..6ff134194e3 100644 --- a/src/Functions/hasTokenCaseInsensitive.cpp +++ b/src/Functions/hasTokenCaseInsensitive.cpp @@ -26,11 +26,11 @@ REGISTER_FUNCTION(HasTokenCaseInsensitive) { factory.registerFunction( FunctionDocumentation{.description="Performs case insensitive lookup of needle in haystack using tokenbf_v1 index."}, - DB::FunctionFactory::CaseInsensitive); + DB::FunctionFactory::Case::Insensitive); factory.registerFunction( FunctionDocumentation{.description="Performs case insensitive lookup of needle in haystack using tokenbf_v1 index. Returns null if needle is ill-formed."}, - DB::FunctionFactory::CaseInsensitive); + DB::FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/hilbertEncode.cpp b/src/Functions/hilbertEncode.cpp index 13512d0d36c..a4e3cc59b76 100644 --- a/src/Functions/hilbertEncode.cpp +++ b/src/Functions/hilbertEncode.cpp @@ -11,8 +11,8 @@ namespace DB namespace ErrorCodes { - extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int ARGUMENT_OUT_OF_BOUND; + extern const int TOO_MANY_ARGUMENTS_FOR_FUNCTION; } @@ -87,7 +87,7 @@ public: return col_res; } - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + throw Exception(ErrorCodes::TOO_MANY_ARGUMENTS_FOR_FUNCTION, "Illegal number of UInt arguments of function {}: should be not more than 2 dimensions", getName()); } diff --git a/src/Functions/hypot.cpp b/src/Functions/hypot.cpp index 465471cb09b..8845d1fa8ae 100644 --- a/src/Functions/hypot.cpp +++ b/src/Functions/hypot.cpp @@ -15,7 +15,7 @@ namespace REGISTER_FUNCTION(Hypot) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/idna.cpp b/src/Functions/idna.cpp index 5a7ae3485ba..cf9e855c912 100644 --- a/src/Functions/idna.cpp +++ b/src/Functions/idna.cpp @@ -44,15 +44,15 @@ struct IdnaEncode const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - const size_t rows = offsets.size(); res_data.reserve(data.size()); /// just a guess, assuming the input is all-ASCII - res_offsets.reserve(rows); + res_offsets.reserve(input_rows_count); size_t prev_offset = 0; std::string ascii; - for (size_t row = 0; row < rows; ++row) + for (size_t row = 0; row < input_rows_count; ++row) { const char * value = reinterpret_cast(&data[prev_offset]); const size_t value_length = offsets[row] - prev_offset - 1; @@ -85,7 +85,7 @@ struct IdnaEncode } } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Arguments of type FixedString are not allowed"); } @@ -99,15 +99,15 @@ struct IdnaDecode const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - const size_t rows = offsets.size(); res_data.reserve(data.size()); /// just a guess, assuming the input is all-ASCII - res_offsets.reserve(rows); + res_offsets.reserve(input_rows_count); size_t prev_offset = 0; std::string unicode; - for (size_t row = 0; row < rows; ++row) + for (size_t row = 0; row < input_rows_count; ++row) { const char * ascii = reinterpret_cast(&data[prev_offset]); const size_t ascii_length = offsets[row] - prev_offset - 1; @@ -124,7 +124,7 @@ struct IdnaDecode } } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Arguments of type FixedString are not allowed"); } diff --git a/src/Functions/if.cpp b/src/Functions/if.cpp index 4c08cd3b931..07dbee27a9d 100644 --- a/src/Functions/if.cpp +++ b/src/Functions/if.cpp @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -1308,7 +1309,7 @@ public: REGISTER_FUNCTION(If) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } FunctionOverloadResolverPtr createInternalFunctionIfOverloadResolver(bool allow_experimental_variant_type, bool use_variant_as_common_type) diff --git a/src/Functions/ifNull.cpp b/src/Functions/ifNull.cpp index 1093f3f817f..358a52c8394 100644 --- a/src/Functions/ifNull.cpp +++ b/src/Functions/ifNull.cpp @@ -91,7 +91,7 @@ private: REGISTER_FUNCTION(IfNull) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/indexHint.h b/src/Functions/indexHint.h index 3b71c7a5585..a71e1555c78 100644 --- a/src/Functions/indexHint.h +++ b/src/Functions/indexHint.h @@ -42,6 +42,10 @@ public: bool useDefaultImplementationForNulls() const override { return false; } + bool useDefaultImplementationForLowCardinalityColumns() const override { return false; } + + bool useDefaultImplementationForSparseColumns() const override { return false; } + bool isSuitableForConstantFolding() const override { return false; } bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return false; } diff --git a/src/Functions/initcap.cpp b/src/Functions/initcap.cpp index 6b2958227bc..00414b22344 100644 --- a/src/Functions/initcap.cpp +++ b/src/Functions/initcap.cpp @@ -9,10 +9,12 @@ namespace struct InitcapImpl { - static void vector(const ColumnString::Chars & data, + static void vector( + const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t /*input_rows_count*/) { if (data.empty()) return; @@ -21,7 +23,7 @@ struct InitcapImpl array(data.data(), data.data() + data.size(), res_data.data()); } - static void vectorFixed(const ColumnString::Chars & data, size_t /*n*/, ColumnString::Chars & res_data) + static void vectorFixed(const ColumnString::Chars & data, size_t /*n*/, ColumnString::Chars & res_data, size_t) { res_data.resize(data.size()); array(data.data(), data.data() + data.size(), res_data.data()); @@ -60,7 +62,7 @@ using FunctionInitcap = FunctionStringToString; REGISTER_FUNCTION(Initcap) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/initcapUTF8.cpp b/src/Functions/initcapUTF8.cpp index 076dcff6622..282d846094e 100644 --- a/src/Functions/initcapUTF8.cpp +++ b/src/Functions/initcapUTF8.cpp @@ -22,7 +22,8 @@ struct InitcapUTF8Impl const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t /*input_rows_count*/) { if (data.empty()) return; @@ -31,7 +32,7 @@ struct InitcapUTF8Impl array(data.data(), data.data() + data.size(), offsets, res_data.data()); } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::BAD_ARGUMENTS, "Function initcapUTF8 cannot work with FixedString argument"); } diff --git a/src/Functions/initialQueryID.cpp b/src/Functions/initialQueryID.cpp index 9c9390d4e50..f32f92a2f46 100644 --- a/src/Functions/initialQueryID.cpp +++ b/src/Functions/initialQueryID.cpp @@ -41,6 +41,6 @@ public: REGISTER_FUNCTION(InitialQueryID) { factory.registerFunction(); - factory.registerAlias("initial_query_id", FunctionInitialQueryID::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("initial_query_id", FunctionInitialQueryID::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/isNotNull.cpp b/src/Functions/isNotNull.cpp index a10e7ebd40c..a48ace4243f 100644 --- a/src/Functions/isNotNull.cpp +++ b/src/Functions/isNotNull.cpp @@ -22,7 +22,9 @@ class FunctionIsNotNull : public IFunction public: static constexpr auto name = "isNotNull"; - static FunctionPtr create(ContextPtr) { return std::make_shared(); } + static FunctionPtr create(ContextPtr context) { return std::make_shared(context->getSettingsRef().allow_experimental_analyzer); } + + explicit FunctionIsNotNull(bool use_analyzer_) : use_analyzer(use_analyzer_) {} std::string getName() const override { @@ -31,6 +33,10 @@ public: ColumnPtr getConstantResultForNonConstArguments(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type) const override { + /// (column IS NULL) triggers a bug in old analyzer when it is replaced to constant. + if (!use_analyzer) + return nullptr; + const ColumnWithTypeAndName & elem = arguments[0]; if (elem.type->onlyNull()) return result_type->createColumnConst(1, UInt8(0)); @@ -123,6 +129,8 @@ private: #endif vectorImpl(null_map, res); } + + bool use_analyzer; }; } diff --git a/src/Functions/isNull.cpp b/src/Functions/isNull.cpp index 95d659b103b..6d799a4ee4e 100644 --- a/src/Functions/isNull.cpp +++ b/src/Functions/isNull.cpp @@ -7,6 +7,8 @@ #include #include #include +#include +#include namespace DB @@ -21,11 +23,13 @@ class FunctionIsNull : public IFunction public: static constexpr auto name = "isNull"; - static FunctionPtr create(ContextPtr) + static FunctionPtr create(ContextPtr context) { - return std::make_shared(); + return std::make_shared(context->getSettingsRef().allow_experimental_analyzer); } + explicit FunctionIsNull(bool use_analyzer_) : use_analyzer(use_analyzer_) {} + std::string getName() const override { return name; @@ -33,6 +37,10 @@ public: ColumnPtr getConstantResultForNonConstArguments(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type) const override { + /// (column IS NULL) triggers a bug in old analyzer when it is replaced to constant. + if (!use_analyzer) + return nullptr; + const ColumnWithTypeAndName & elem = arguments[0]; if (elem.type->onlyNull()) return result_type->createColumnConst(1, UInt8(1)); @@ -95,13 +103,16 @@ public: return DataTypeUInt8().createColumnConst(elem.column->size(), 0u); } } + +private: + bool use_analyzer; }; } REGISTER_FUNCTION(IsNull) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/isNullable.cpp b/src/Functions/isNullable.cpp index b24ee4f5e73..d21ac9cf07c 100644 --- a/src/Functions/isNullable.cpp +++ b/src/Functions/isNullable.cpp @@ -3,6 +3,8 @@ #include #include #include +#include +#include namespace DB { @@ -14,11 +16,13 @@ class FunctionIsNullable : public IFunction { public: static constexpr auto name = "isNullable"; - static FunctionPtr create(ContextPtr) + static FunctionPtr create(ContextPtr context) { - return std::make_shared(); + return std::make_shared(context->getSettingsRef().allow_experimental_analyzer); } + explicit FunctionIsNullable(bool use_analyzer_) : use_analyzer(use_analyzer_) {} + String getName() const override { return name; @@ -26,6 +30,10 @@ public: ColumnPtr getConstantResultForNonConstArguments(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type) const override { + /// isNullable(column) triggers a bug in old analyzer when it is replaced to constant. + if (!use_analyzer) + return nullptr; + const ColumnWithTypeAndName & elem = arguments[0]; if (elem.type->onlyNull() || canContainNull(*elem.type)) return result_type->createColumnConst(1, UInt8(1)); @@ -60,6 +68,9 @@ public: const auto & elem = arguments[0]; return ColumnUInt8::create(input_rows_count, isColumnNullable(*elem.column) || elem.type->isLowCardinalityNullable()); } + +private: + bool use_analyzer; }; } diff --git a/src/Functions/keyvaluepair/extractKeyValuePairs.cpp b/src/Functions/keyvaluepair/extractKeyValuePairs.cpp index 94f02861af0..1c5164e132d 100644 --- a/src/Functions/keyvaluepair/extractKeyValuePairs.cpp +++ b/src/Functions/keyvaluepair/extractKeyValuePairs.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include @@ -240,7 +241,7 @@ REGISTER_FUNCTION(ExtractKeyValuePairs) └──────────────────┘ ```)"} ); - factory.registerAlias("str_to_map", NameExtractKeyValuePairs::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("str_to_map", NameExtractKeyValuePairs::name, FunctionFactory::Case::Insensitive); factory.registerAlias("mapFromString", NameExtractKeyValuePairs::name); } diff --git a/src/Functions/least.cpp b/src/Functions/least.cpp index babb8378d80..091a868e8e2 100644 --- a/src/Functions/least.cpp +++ b/src/Functions/least.cpp @@ -65,7 +65,7 @@ using FunctionLeast = FunctionBinaryArithmetic; REGISTER_FUNCTION(Least) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/left.cpp b/src/Functions/left.cpp index 006706c8f21..c9f62a0f8f1 100644 --- a/src/Functions/left.cpp +++ b/src/Functions/left.cpp @@ -6,8 +6,8 @@ namespace DB REGISTER_FUNCTION(Left) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); - factory.registerFunction>({}, FunctionFactory::CaseSensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); + factory.registerFunction>({}, FunctionFactory::Case::Sensitive); } } diff --git a/src/Functions/lemmatize.cpp b/src/Functions/lemmatize.cpp index e1168e077bb..76e5ed7432e 100644 --- a/src/Functions/lemmatize.cpp +++ b/src/Functions/lemmatize.cpp @@ -3,6 +3,7 @@ #if USE_NLP #include +#include #include #include #include diff --git a/src/Functions/lengthUTF8.cpp b/src/Functions/lengthUTF8.cpp index 5a4af4934df..59a0d532602 100644 --- a/src/Functions/lengthUTF8.cpp +++ b/src/Functions/lengthUTF8.cpp @@ -83,8 +83,8 @@ REGISTER_FUNCTION(LengthUTF8) factory.registerFunction(); /// Compatibility aliases. - factory.registerAlias("CHAR_LENGTH", "lengthUTF8", FunctionFactory::CaseInsensitive); - factory.registerAlias("CHARACTER_LENGTH", "lengthUTF8", FunctionFactory::CaseInsensitive); + factory.registerAlias("CHAR_LENGTH", "lengthUTF8", FunctionFactory::Case::Insensitive); + factory.registerAlias("CHARACTER_LENGTH", "lengthUTF8", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/locate.cpp b/src/Functions/locate.cpp index d9a727ab3ef..076aa1bdc6d 100644 --- a/src/Functions/locate.cpp +++ b/src/Functions/locate.cpp @@ -29,6 +29,6 @@ REGISTER_FUNCTION(Locate) FunctionDocumentation::Categories doc_categories = {"String search"}; - factory.registerFunction({doc_description, doc_syntax, doc_arguments, doc_returned_value, doc_examples, doc_categories}, FunctionFactory::CaseInsensitive); + factory.registerFunction({doc_description, doc_syntax, doc_arguments, doc_returned_value, doc_examples, doc_categories}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/log.cpp b/src/Functions/log.cpp index 9096b8c6f22..8bebdb8d7bd 100644 --- a/src/Functions/log.cpp +++ b/src/Functions/log.cpp @@ -34,8 +34,8 @@ using FunctionLog = FunctionMathUnary>; REGISTER_FUNCTION(Log) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerAlias("ln", "log", FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerAlias("ln", "log", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/log10.cpp b/src/Functions/log10.cpp index 5dfe4ac9357..6241df3e092 100644 --- a/src/Functions/log10.cpp +++ b/src/Functions/log10.cpp @@ -13,7 +13,7 @@ using FunctionLog10 = FunctionMathUnary({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/log2.cpp b/src/Functions/log2.cpp index 9457ac64bc6..52b3ab52ea7 100644 --- a/src/Functions/log2.cpp +++ b/src/Functions/log2.cpp @@ -13,7 +13,7 @@ using FunctionLog2 = FunctionMathUnary>; REGISTER_FUNCTION(Log2) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/lower.cpp b/src/Functions/lower.cpp index 38ae5a8a7f0..5210a20b026 100644 --- a/src/Functions/lower.cpp +++ b/src/Functions/lower.cpp @@ -19,8 +19,8 @@ using FunctionLower = FunctionStringToString, NameLower REGISTER_FUNCTION(Lower) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerAlias("lcase", NameLower::name, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerAlias("lcase", NameLower::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/makeDate.cpp b/src/Functions/makeDate.cpp index 3d8b8617472..21d466d7708 100644 --- a/src/Functions/makeDate.cpp +++ b/src/Functions/makeDate.cpp @@ -87,7 +87,7 @@ public: {mandatory_argument_names_year_month_day[1], static_cast(&isNumber), nullptr, "Number"}, {mandatory_argument_names_year_month_day[2], static_cast(&isNumber), nullptr, "Number"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); } else { @@ -95,7 +95,7 @@ public: {mandatory_argument_names_year_dayofyear[0], static_cast(&isNumber), nullptr, "Number"}, {mandatory_argument_names_year_dayofyear[1], static_cast(&isNumber), nullptr, "Number"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); } return std::make_shared(); @@ -193,7 +193,7 @@ public: {mandatory_argument_names[0], static_cast(&isNumber), nullptr, "Number"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } @@ -357,7 +357,7 @@ public: {optional_argument_names[0], static_cast(&isString), isColumnConst, "const String"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); /// Optional timezone argument std::string timezone; @@ -440,7 +440,7 @@ public: {optional_argument_names[2], static_cast(&isString), isColumnConst, "const String"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); if (arguments.size() >= mandatory_argument_names.size() + 1) { @@ -572,7 +572,7 @@ public: {optional_argument_names[0], static_cast(&isString), isColumnConst, "const String"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); /// Optional timezone argument std::string timezone; @@ -652,7 +652,7 @@ public: {optional_argument_names[0], static_cast(&isString), isColumnConst, "const String"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); /// Optional precision argument auto precision = DEFAULT_PRECISION; @@ -724,7 +724,7 @@ public: REGISTER_FUNCTION(MakeDate) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); factory.registerFunction>(); factory.registerFunction(); factory.registerFunction(); diff --git a/src/Functions/map.cpp b/src/Functions/map.cpp index 66cd10a3f0b..0d42ce2abe1 100644 --- a/src/Functions/map.cpp +++ b/src/Functions/map.cpp @@ -10,6 +10,7 @@ #include #include #include +#include namespace DB diff --git a/src/Functions/match.cpp b/src/Functions/match.cpp index c719cc6dd82..6cd65597032 100644 --- a/src/Functions/match.cpp +++ b/src/Functions/match.cpp @@ -20,7 +20,7 @@ using FunctionMatch = FunctionsStringSearch(); - factory.registerAlias("REGEXP_MATCHES", NameMatch::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("REGEXP_MATCHES", NameMatch::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/mathConstants.cpp b/src/Functions/mathConstants.cpp index 2b199a30616..37ababbc0e5 100644 --- a/src/Functions/mathConstants.cpp +++ b/src/Functions/mathConstants.cpp @@ -44,7 +44,7 @@ REGISTER_FUNCTION(E) REGISTER_FUNCTION(Pi) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/max2.cpp b/src/Functions/max2.cpp index 928e6f22918..88b5c7c08c0 100644 --- a/src/Functions/max2.cpp +++ b/src/Functions/max2.cpp @@ -21,6 +21,6 @@ namespace REGISTER_FUNCTION(Max2) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/min2.cpp b/src/Functions/min2.cpp index f031530edf5..8ab56dbe90d 100644 --- a/src/Functions/min2.cpp +++ b/src/Functions/min2.cpp @@ -22,6 +22,6 @@ namespace REGISTER_FUNCTION(Min2) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/modulo.cpp b/src/Functions/modulo.cpp index ebc1c4f5275..76a07aeda2e 100644 --- a/src/Functions/modulo.cpp +++ b/src/Functions/modulo.cpp @@ -155,7 +155,7 @@ using FunctionModulo = BinaryArithmeticOverloadResolver(); - factory.registerAlias("mod", "modulo", FunctionFactory::CaseInsensitive); + factory.registerAlias("mod", "modulo", FunctionFactory::Case::Insensitive); } struct NameModuloLegacy { static constexpr auto name = "moduloLegacy"; }; @@ -183,11 +183,11 @@ In other words, the function returning the modulus (modulo) in the terms of Modu )", .examples{{"positiveModulo", "SELECT positiveModulo(-1, 10);", ""}}, .categories{"Arithmetic"}}, - FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); - factory.registerAlias("positive_modulo", "positiveModulo", FunctionFactory::CaseInsensitive); + factory.registerAlias("positive_modulo", "positiveModulo", FunctionFactory::Case::Insensitive); /// Compatibility with Spark: - factory.registerAlias("pmod", "positiveModulo", FunctionFactory::CaseInsensitive); + factory.registerAlias("pmod", "positiveModulo", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/monthName.cpp b/src/Functions/monthName.cpp index f49f77bd6e7..ae444460170 100644 --- a/src/Functions/monthName.cpp +++ b/src/Functions/monthName.cpp @@ -74,7 +74,7 @@ private: REGISTER_FUNCTION(MonthName) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/mortonEncode.cpp b/src/Functions/mortonEncode.cpp index 0c19c7c3134..8cf0f18dbfe 100644 --- a/src/Functions/mortonEncode.cpp +++ b/src/Functions/mortonEncode.cpp @@ -16,8 +16,8 @@ namespace DB namespace ErrorCodes { - extern const int ILLEGAL_TYPE_OF_ARGUMENT; extern const int ARGUMENT_OUT_OF_BOUND; + extern const int TOO_MANY_ARGUMENTS_FOR_FUNCTION; } #define EXTRACT_VECTOR(INDEX) \ @@ -130,7 +130,7 @@ namespace ErrorCodes MASK(8, 6, col6->getUInt(i)), \ MASK(8, 7, col7->getUInt(i))) \ \ - throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, \ + throw Exception(ErrorCodes::TOO_MANY_ARGUMENTS_FOR_FUNCTION, \ "Illegal number of UInt arguments of function {}, max: 8", \ getName()); \ diff --git a/src/Functions/multiIf.cpp b/src/Functions/multiIf.cpp index d3bf5618f66..c4b675fcf6c 100644 --- a/src/Functions/multiIf.cpp +++ b/src/Functions/multiIf.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Functions/neighbor.cpp b/src/Functions/neighbor.cpp index 62f129109f9..49b8980e7e2 100644 --- a/src/Functions/neighbor.cpp +++ b/src/Functions/neighbor.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/Functions/normalizeQuery.cpp b/src/Functions/normalizeQuery.cpp index ad9a8903733..1a78bce7d29 100644 --- a/src/Functions/normalizeQuery.cpp +++ b/src/Functions/normalizeQuery.cpp @@ -19,17 +19,19 @@ template struct Impl { static constexpr auto name = keep_names ? "normalizeQueryKeepNames" : "normalizeQuery"; - static void vector(const ColumnString::Chars & data, + + static void vector( + const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - size_t size = offsets.size(); - res_offsets.resize(size); + res_offsets.resize(input_rows_count); res_data.reserve(data.size()); ColumnString::Offset prev_src_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { ColumnString::Offset curr_src_offset = offsets[i]; @@ -43,7 +45,7 @@ struct Impl } } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Cannot apply function normalizeQuery to fixed string."); } diff --git a/src/Functions/normalizeString.cpp b/src/Functions/normalizeString.cpp index 92411490eaa..c56508ced0e 100644 --- a/src/Functions/normalizeString.cpp +++ b/src/Functions/normalizeString.cpp @@ -84,7 +84,8 @@ struct NormalizeUTF8Impl static void vector(const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { UErrorCode err = U_ZERO_ERROR; @@ -92,8 +93,7 @@ struct NormalizeUTF8Impl if (U_FAILURE(err)) throw Exception(ErrorCodes::CANNOT_NORMALIZE_STRING, "Normalization failed (getNormalizer): {}", u_errorName(err)); - size_t size = offsets.size(); - res_offsets.resize(size); + res_offsets.resize(input_rows_count); res_data.reserve(data.size() * 2); @@ -103,7 +103,7 @@ struct NormalizeUTF8Impl PODArray from_uchars; PODArray to_uchars; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t from_size = offsets[i] - current_from_offset - 1; @@ -157,7 +157,7 @@ struct NormalizeUTF8Impl res_data.resize(current_to_offset); } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Cannot apply function normalizeUTF8 to fixed string."); } diff --git a/src/Functions/normalizedQueryHash.cpp b/src/Functions/normalizedQueryHash.cpp index 63218f28af5..3dbe3ff9847 100644 --- a/src/Functions/normalizedQueryHash.cpp +++ b/src/Functions/normalizedQueryHash.cpp @@ -27,13 +27,13 @@ struct Impl static void vector( const ColumnString::Chars & data, const ColumnString::Offsets & offsets, - PaddedPODArray & res_data) + PaddedPODArray & res_data, + size_t input_rows_count) { - size_t size = offsets.size(); - res_data.resize(size); + res_data.resize(input_rows_count); ColumnString::Offset prev_src_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { ColumnString::Offset curr_src_offset = offsets[i]; res_data[i] = normalizedQueryHash( @@ -77,15 +77,15 @@ public: bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const ColumnPtr column = arguments[0].column; if (const ColumnString * col = checkAndGetColumn(column.get())) { auto col_res = ColumnUInt64::create(); typename ColumnUInt64::Container & vec_res = col_res->getData(); - vec_res.resize(col->size()); - Impl::vector(col->getChars(), col->getOffsets(), vec_res); + vec_res.resize(input_rows_count); + Impl::vector(col->getChars(), col->getOffsets(), vec_res, input_rows_count); return col_res; } else diff --git a/src/Functions/now.cpp b/src/Functions/now.cpp index 827b800a243..7b2150e3534 100644 --- a/src/Functions/now.cpp +++ b/src/Functions/now.cpp @@ -1,6 +1,7 @@ #include #include #include +#include #include #include #include @@ -90,7 +91,7 @@ public: size_t getNumberOfArguments() const override { return 0; } static FunctionOverloadResolverPtr create(ContextPtr context) { return std::make_unique(context); } explicit NowOverloadResolver(ContextPtr context) - : allow_nonconst_timezone_arguments(context->getSettings().allow_nonconst_timezone_arguments) + : allow_nonconst_timezone_arguments(context->getSettingsRef().allow_nonconst_timezone_arguments) {} DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override @@ -137,8 +138,8 @@ private: REGISTER_FUNCTION(Now) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerAlias("current_timestamp", NowOverloadResolver::name, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerAlias("current_timestamp", NowOverloadResolver::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/now64.cpp b/src/Functions/now64.cpp index d6f8474c984..9786a0c9f39 100644 --- a/src/Functions/now64.cpp +++ b/src/Functions/now64.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include @@ -118,7 +119,7 @@ public: size_t getNumberOfArguments() const override { return 0; } static FunctionOverloadResolverPtr create(ContextPtr context) { return std::make_unique(context); } explicit Now64OverloadResolver(ContextPtr context) - : allow_nonconst_timezone_arguments(context->getSettings().allow_nonconst_timezone_arguments) + : allow_nonconst_timezone_arguments(context->getSettingsRef().allow_nonconst_timezone_arguments) {} DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override @@ -169,7 +170,7 @@ private: REGISTER_FUNCTION(Now64) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/nowInBlock.cpp b/src/Functions/nowInBlock.cpp index 74f420986c8..291e43bdce5 100644 --- a/src/Functions/nowInBlock.cpp +++ b/src/Functions/nowInBlock.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -31,7 +32,7 @@ public: return std::make_shared(context); } explicit FunctionNowInBlock(ContextPtr context) - : allow_nonconst_timezone_arguments(context->getSettings().allow_nonconst_timezone_arguments) + : allow_nonconst_timezone_arguments(context->getSettingsRef().allow_nonconst_timezone_arguments) {} String getName() const override diff --git a/src/Functions/nullIf.cpp b/src/Functions/nullIf.cpp index 392cc20cfcf..550287885a1 100644 --- a/src/Functions/nullIf.cpp +++ b/src/Functions/nullIf.cpp @@ -69,7 +69,7 @@ public: REGISTER_FUNCTION(NullIf) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/padString.cpp b/src/Functions/padString.cpp index 8670c837e21..23554c3fbbc 100644 --- a/src/Functions/padString.cpp +++ b/src/Functions/padString.cpp @@ -335,8 +335,8 @@ REGISTER_FUNCTION(PadString) factory.registerFunction>(); /// rightPad factory.registerFunction>(); /// rightPadUTF8 - factory.registerAlias("lpad", "leftPad", FunctionFactory::CaseInsensitive); - factory.registerAlias("rpad", "rightPad", FunctionFactory::CaseInsensitive); + factory.registerAlias("lpad", "leftPad", FunctionFactory::Case::Insensitive); + factory.registerAlias("rpad", "rightPad", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/parseDateTime.cpp b/src/Functions/parseDateTime.cpp index 11e210d2cc2..bdca0151bba 100644 --- a/src/Functions/parseDateTime.cpp +++ b/src/Functions/parseDateTime.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include @@ -564,8 +565,8 @@ namespace static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } explicit FunctionParseDateTimeImpl(ContextPtr context) - : mysql_M_is_month_name(context->getSettings().formatdatetime_parsedatetime_m_is_month_name) - , mysql_parse_ckl_without_leading_zeros(context->getSettings().parsedatetime_parse_without_leading_zeros) + : mysql_M_is_month_name(context->getSettingsRef().formatdatetime_parsedatetime_m_is_month_name) + , mysql_parse_ckl_without_leading_zeros(context->getSettingsRef().parsedatetime_parse_without_leading_zeros) { } @@ -589,7 +590,7 @@ namespace {"timezone", static_cast(&isString), &isColumnConst, "const String"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); String time_zone_name = getTimeZone(arguments).getTimeZone(); DataTypePtr date_type = std::make_shared(time_zone_name); @@ -978,8 +979,7 @@ namespace [[nodiscard]] static PosOrError mysqlAmericanDate(Pos cur, Pos end, const String & fragment, DateTime & date) { - if (auto status = checkSpace(cur, end, 8, "mysqlAmericanDate requires size >= 8", fragment)) - return tl::unexpected(status.error()); + RETURN_ERROR_IF_FAILED(checkSpace(cur, end, 8, "mysqlAmericanDate requires size >= 8", fragment)) Int32 month; ASSIGN_RESULT_OR_RETURN_ERROR(cur, (readNumber2(cur, end, fragment, month))) @@ -993,7 +993,7 @@ namespace Int32 year; ASSIGN_RESULT_OR_RETURN_ERROR(cur, (readNumber2(cur, end, fragment, year))) - RETURN_ERROR_IF_FAILED(date.setYear(year)) + RETURN_ERROR_IF_FAILED(date.setYear(year + 2000)) return cur; } @@ -1015,8 +1015,7 @@ namespace [[nodiscard]] static PosOrError mysqlISO8601Date(Pos cur, Pos end, const String & fragment, DateTime & date) { - if (auto status = checkSpace(cur, end, 10, "mysqlISO8601Date requires size >= 10", fragment)) - return tl::unexpected(status.error()); + RETURN_ERROR_IF_FAILED(checkSpace(cur, end, 10, "mysqlISO8601Date requires size >= 10", fragment)) Int32 year; Int32 month; @@ -1462,8 +1461,7 @@ namespace [[nodiscard]] static PosOrError jodaDayOfWeekText(size_t /*min_represent_digits*/, Pos cur, Pos end, const String & fragment, DateTime & date) { - if (auto result= checkSpace(cur, end, 3, "jodaDayOfWeekText requires size >= 3", fragment); !result.has_value()) - return tl::unexpected(result.error()); + RETURN_ERROR_IF_FAILED(checkSpace(cur, end, 3, "jodaDayOfWeekText requires size >= 3", fragment)) String text1(cur, 3); boost::to_lower(text1); @@ -1556,8 +1554,8 @@ namespace Int32 day_of_month; ASSIGN_RESULT_OR_RETURN_ERROR(cur, (readNumberWithVariableLength( cur, end, false, false, false, repetitions, std::max(repetitions, 2uz), fragment, day_of_month))) - if (auto res = date.setDayOfMonth(day_of_month); !res.has_value()) - return tl::unexpected(res.error()); + RETURN_ERROR_IF_FAILED(date.setDayOfMonth(day_of_month)) + return cur; } @@ -2100,10 +2098,10 @@ namespace REGISTER_FUNCTION(ParseDateTime) { factory.registerFunction(); - factory.registerAlias("TO_UNIXTIME", FunctionParseDateTime::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("TO_UNIXTIME", FunctionParseDateTime::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); - factory.registerAlias("str_to_date", FunctionParseDateTimeOrNull::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("str_to_date", FunctionParseDateTimeOrNull::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); diff --git a/src/Functions/parseReadableSize.cpp b/src/Functions/parseReadableSize.cpp index f5c2c53439b..1abcf7f164f 100644 --- a/src/Functions/parseReadableSize.cpp +++ b/src/Functions/parseReadableSize.cpp @@ -68,7 +68,7 @@ public: { {"readable_size", static_cast(&isString), nullptr, "String"}, }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); DataTypePtr return_type = std::make_shared(); if constexpr (error_handling == ErrorHandling::Null) return std::make_shared(return_type); diff --git a/src/Functions/partitionId.cpp b/src/Functions/partitionId.cpp index e2e9038cd8b..dab56f39adc 100644 --- a/src/Functions/partitionId.cpp +++ b/src/Functions/partitionId.cpp @@ -66,6 +66,7 @@ public: REGISTER_FUNCTION(PartitionId) { factory.registerFunction(); + factory.registerAlias("partitionID", "partitionId"); } } diff --git a/src/Functions/pointInEllipses.cpp b/src/Functions/pointInEllipses.cpp index 2147428cee3..ac7a8cc4204 100644 --- a/src/Functions/pointInEllipses.cpp +++ b/src/Functions/pointInEllipses.cpp @@ -91,8 +91,6 @@ private: ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { - const auto size = input_rows_count; - /// Prepare array of ellipses. size_t ellipses_count = (arguments.size() - 2) / 4; std::vector ellipses(ellipses_count); @@ -141,13 +139,11 @@ private: auto dst = ColumnVector::create(); auto & dst_data = dst->getData(); - dst_data.resize(size); + dst_data.resize(input_rows_count); size_t start_index = 0; - for (const auto row : collections::range(0, size)) - { + for (size_t row = 0; row < input_rows_count; ++row) dst_data[row] = isPointInEllipses(col_vec_x->getData()[row], col_vec_y->getData()[row], ellipses.data(), ellipses_count, start_index); - } return dst; } @@ -157,7 +153,7 @@ private: const auto * col_const_y = assert_cast (col_y); size_t start_index = 0; UInt8 res = isPointInEllipses(col_const_x->getValue(), col_const_y->getValue(), ellipses.data(), ellipses_count, start_index); - return DataTypeUInt8().createColumnConst(size, res); + return DataTypeUInt8().createColumnConst(input_rows_count, res); } else { diff --git a/src/Functions/pointInPolygon.cpp b/src/Functions/pointInPolygon.cpp index 6b413829bd1..32d5905a513 100644 --- a/src/Functions/pointInPolygon.cpp +++ b/src/Functions/pointInPolygon.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Functions/position.cpp b/src/Functions/position.cpp index 29a5db2eb24..aad47cc5b3f 100644 --- a/src/Functions/position.cpp +++ b/src/Functions/position.cpp @@ -19,6 +19,6 @@ using FunctionPosition = FunctionsStringSearch({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/positionCaseInsensitive.cpp b/src/Functions/positionCaseInsensitive.cpp index f71ce0078cc..7c59ffa83cd 100644 --- a/src/Functions/positionCaseInsensitive.cpp +++ b/src/Functions/positionCaseInsensitive.cpp @@ -20,6 +20,6 @@ using FunctionPositionCaseInsensitive = FunctionsStringSearch(); - factory.registerAlias("instr", NamePositionCaseInsensitive::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("instr", NamePositionCaseInsensitive::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/pow.cpp b/src/Functions/pow.cpp index 9b383da97e7..f2976b4812e 100644 --- a/src/Functions/pow.cpp +++ b/src/Functions/pow.cpp @@ -13,8 +13,8 @@ using FunctionPow = FunctionMathBinaryFloat64({}, FunctionFactory::CaseInsensitive); - factory.registerAlias("power", "pow", FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerAlias("power", "pow", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/punycode.cpp b/src/Functions/punycode.cpp index 8004e3731b5..ec1fcfd0a70 100644 --- a/src/Functions/punycode.cpp +++ b/src/Functions/punycode.cpp @@ -6,11 +6,11 @@ #include #include -# pragma clang diagnostic push -# pragma clang diagnostic ignored "-Wnewline-eof" -# include -# include -# pragma clang diagnostic pop +#pragma clang diagnostic push +#pragma clang diagnostic ignored "-Wnewline-eof" +#include +#include +#pragma clang diagnostic pop namespace DB { @@ -38,16 +38,16 @@ struct PunycodeEncode const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - const size_t rows = offsets.size(); res_data.reserve(data.size()); /// just a guess, assuming the input is all-ASCII - res_offsets.reserve(rows); + res_offsets.reserve(input_rows_count); size_t prev_offset = 0; std::u32string value_utf32; std::string value_puny; - for (size_t row = 0; row < rows; ++row) + for (size_t row = 0; row < input_rows_count; ++row) { const char * value = reinterpret_cast(&data[prev_offset]); const size_t value_length = offsets[row] - prev_offset - 1; @@ -72,7 +72,7 @@ struct PunycodeEncode } } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Arguments of type FixedString are not allowed"); } @@ -86,16 +86,16 @@ struct PunycodeDecode const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - const size_t rows = offsets.size(); res_data.reserve(data.size()); /// just a guess, assuming the input is all-ASCII - res_offsets.reserve(rows); + res_offsets.reserve(input_rows_count); size_t prev_offset = 0; std::u32string value_utf32; std::string value_utf8; - for (size_t row = 0; row < rows; ++row) + for (size_t row = 0; row < input_rows_count; ++row) { const char * value = reinterpret_cast(&data[prev_offset]); const size_t value_length = offsets[row] - prev_offset - 1; @@ -129,7 +129,7 @@ struct PunycodeDecode } } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Arguments of type FixedString are not allowed"); } diff --git a/src/Functions/queryID.cpp b/src/Functions/queryID.cpp index 5d0ac719797..7299714e464 100644 --- a/src/Functions/queryID.cpp +++ b/src/Functions/queryID.cpp @@ -41,6 +41,6 @@ public: REGISTER_FUNCTION(QueryID) { factory.registerFunction(); - factory.registerAlias("query_id", FunctionQueryID::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("query_id", FunctionQueryID::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/radians.cpp b/src/Functions/radians.cpp index 2c2c2743532..9185340be15 100644 --- a/src/Functions/radians.cpp +++ b/src/Functions/radians.cpp @@ -23,7 +23,7 @@ namespace REGISTER_FUNCTION(Radians) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/rand.cpp b/src/Functions/rand.cpp index ea30922d731..35b325e59fd 100644 --- a/src/Functions/rand.cpp +++ b/src/Functions/rand.cpp @@ -13,7 +13,7 @@ using FunctionRand = FunctionRandom; REGISTER_FUNCTION(Rand) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); factory.registerAlias("rand32", NameRand::name); } diff --git a/src/Functions/randDistribution.cpp b/src/Functions/randDistribution.cpp index 4e616ada697..6a3dac748c1 100644 --- a/src/Functions/randDistribution.cpp +++ b/src/Functions/randDistribution.cpp @@ -24,6 +24,7 @@ namespace ErrorCodes extern const int ILLEGAL_COLUMN; extern const int BAD_ARGUMENTS; extern const int LOGICAL_ERROR; + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; } namespace @@ -246,7 +247,7 @@ public: { auto desired = Distribution::getNumberOfArguments(); if (arguments.size() != desired && arguments.size() != desired + 1) - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Wrong number of arguments for function {}. Should be {} or {}", getName(), desired, desired + 1); diff --git a/src/Functions/regexpExtract.cpp b/src/Functions/regexpExtract.cpp index cfb42580cb0..6bedac54e39 100644 --- a/src/Functions/regexpExtract.cpp +++ b/src/Functions/regexpExtract.cpp @@ -54,7 +54,7 @@ public: if (arguments.size() == 3) args.emplace_back(FunctionArgumentDescriptor{"index", static_cast(&isInteger), nullptr, "Integer"}); - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } @@ -253,7 +253,7 @@ REGISTER_FUNCTION(RegexpExtract) FunctionDocumentation{.description="Extracts the first string in haystack that matches the regexp pattern and corresponds to the regex group index."}); /// For Spark compatibility. - factory.registerAlias("REGEXP_EXTRACT", "regexpExtract", FunctionFactory::CaseInsensitive); + factory.registerAlias("REGEXP_EXTRACT", "regexpExtract", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/repeat.cpp b/src/Functions/repeat.cpp index 7f2fe646062..c001959b465 100644 --- a/src/Functions/repeat.cpp +++ b/src/Functions/repeat.cpp @@ -201,7 +201,7 @@ public: {"n", static_cast(&isInteger), nullptr, "Integer"}, }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } @@ -278,7 +278,7 @@ public: REGISTER_FUNCTION(Repeat) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/replaceAll.cpp b/src/Functions/replaceAll.cpp index 6c06f5984b3..9ce525390bf 100644 --- a/src/Functions/replaceAll.cpp +++ b/src/Functions/replaceAll.cpp @@ -20,7 +20,7 @@ using FunctionReplaceAll = FunctionStringReplace(); - factory.registerAlias("replace", NameReplaceAll::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("replace", NameReplaceAll::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/replaceRegexpAll.cpp b/src/Functions/replaceRegexpAll.cpp index f5f56fb0f35..77f21b6efee 100644 --- a/src/Functions/replaceRegexpAll.cpp +++ b/src/Functions/replaceRegexpAll.cpp @@ -20,7 +20,7 @@ using FunctionReplaceRegexpAll = FunctionStringReplace(); - factory.registerAlias("REGEXP_REPLACE", NameReplaceRegexpAll::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("REGEXP_REPLACE", NameReplaceRegexpAll::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/reverse.cpp b/src/Functions/reverse.cpp index 39608b77997..94b6634ffbd 100644 --- a/src/Functions/reverse.cpp +++ b/src/Functions/reverse.cpp @@ -55,19 +55,19 @@ public: bool useDefaultImplementationForConstants() const override { return true; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const ColumnPtr column = arguments[0].column; if (const ColumnString * col = checkAndGetColumn(column.get())) { auto col_res = ColumnString::create(); - ReverseImpl::vector(col->getChars(), col->getOffsets(), col_res->getChars(), col_res->getOffsets()); + ReverseImpl::vector(col->getChars(), col->getOffsets(), col_res->getChars(), col_res->getOffsets(), input_rows_count); return col_res; } else if (const ColumnFixedString * col_fixed = checkAndGetColumn(column.get())) { auto col_res = ColumnFixedString::create(col_fixed->getN()); - ReverseImpl::vectorFixed(col_fixed->getChars(), col_fixed->getN(), col_res->getChars()); + ReverseImpl::vectorFixed(col_fixed->getChars(), col_fixed->getN(), col_res->getChars(), input_rows_count); return col_res; } else @@ -113,7 +113,7 @@ private: REGISTER_FUNCTION(Reverse) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/reverse.h b/src/Functions/reverse.h index 5f999af4297..7c18d72769a 100644 --- a/src/Functions/reverse.h +++ b/src/Functions/reverse.h @@ -9,17 +9,18 @@ namespace DB */ struct ReverseImpl { - static void vector(const ColumnString::Chars & data, + static void vector( + const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { res_data.resize_exact(data.size()); res_offsets.assign(offsets); - size_t size = offsets.size(); ColumnString::Offset prev_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { for (size_t j = prev_offset; j < offsets[i] - 1; ++j) res_data[j] = data[offsets[i] + prev_offset - 2 - j]; @@ -28,12 +29,15 @@ struct ReverseImpl } } - static void vectorFixed(const ColumnString::Chars & data, size_t n, ColumnString::Chars & res_data) + static void vectorFixed( + const ColumnString::Chars & data, + size_t n, + ColumnString::Chars & res_data, + size_t input_rows_count) { res_data.resize_exact(data.size()); - size_t size = data.size() / n; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) for (size_t j = i * n; j < (i + 1) * n; ++j) res_data[j] = data[(i * 2 + 1) * n - j - 1]; } diff --git a/src/Functions/reverseUTF8.cpp b/src/Functions/reverseUTF8.cpp index 1aee349fa8d..ca57d946b19 100644 --- a/src/Functions/reverseUTF8.cpp +++ b/src/Functions/reverseUTF8.cpp @@ -23,25 +23,25 @@ namespace */ struct ReverseUTF8Impl { - static void vector(const ColumnString::Chars & data, + static void vector( + const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { bool all_ascii = isAllASCII(data.data(), data.size()); if (all_ascii) { - ReverseImpl::vector(data, offsets, res_data, res_offsets); + ReverseImpl::vector(data, offsets, res_data, res_offsets, input_rows_count); return; } res_data.resize(data.size()); res_offsets.assign(offsets); - size_t size = offsets.size(); - ColumnString::Offset prev_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { ColumnString::Offset j = prev_offset; while (j < offsets[i] - 1) @@ -73,7 +73,7 @@ struct ReverseUTF8Impl } } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Cannot apply function reverseUTF8 to fixed string."); } diff --git a/src/Functions/right.cpp b/src/Functions/right.cpp index a8ab4bf9685..ef3303ab968 100644 --- a/src/Functions/right.cpp +++ b/src/Functions/right.cpp @@ -6,8 +6,8 @@ namespace DB REGISTER_FUNCTION(Right) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); - factory.registerFunction>({}, FunctionFactory::CaseSensitive); + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); + factory.registerFunction>({}, FunctionFactory::Case::Sensitive); } } diff --git a/src/Functions/runningAccumulate.cpp b/src/Functions/runningAccumulate.cpp index d585affd91b..c94cbe67216 100644 --- a/src/Functions/runningAccumulate.cpp +++ b/src/Functions/runningAccumulate.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include diff --git a/src/Functions/runningDifference.h b/src/Functions/runningDifference.h index fe477d13744..d51ea805f3c 100644 --- a/src/Functions/runningDifference.h +++ b/src/Functions/runningDifference.h @@ -1,6 +1,7 @@ #pragma once #include #include +#include #include #include #include diff --git a/src/Functions/seriesDecomposeSTL.cpp b/src/Functions/seriesDecomposeSTL.cpp index 618808b64ed..1e1c41cafad 100644 --- a/src/Functions/seriesDecomposeSTL.cpp +++ b/src/Functions/seriesDecomposeSTL.cpp @@ -45,12 +45,12 @@ public: {"time_series", static_cast(&isArray), nullptr, "Array"}, {"period", static_cast(&isNativeUInt), nullptr, "Unsigned Integer"}, }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(std::make_shared(std::make_shared())); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { ColumnPtr array_ptr = arguments[0].column; const ColumnArray * array = checkAndGetColumn(array_ptr.get()); @@ -79,7 +79,7 @@ public: ColumnArray::Offset prev_src_offset = 0; - for (size_t i = 0; i < src_offsets.size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { UInt64 period; auto period_ptr = arguments[1].column->convertToFullColumnIfConst(); diff --git a/src/Functions/seriesOutliersDetectTukey.cpp b/src/Functions/seriesOutliersDetectTukey.cpp index 81fc904e16e..4063d0ab85b 100644 --- a/src/Functions/seriesOutliersDetectTukey.cpp +++ b/src/Functions/seriesOutliersDetectTukey.cpp @@ -51,7 +51,7 @@ public: {"max_percentile", static_cast(&isFloat), isColumnConst, "Number"}, {"k", static_cast(&isNativeNumber), isColumnConst, "Number"}}; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(std::make_shared()); } diff --git a/src/Functions/seriesPeriodDetectFFT.cpp b/src/Functions/seriesPeriodDetectFFT.cpp index e85b3a97c67..471354235d5 100644 --- a/src/Functions/seriesPeriodDetectFFT.cpp +++ b/src/Functions/seriesPeriodDetectFFT.cpp @@ -53,7 +53,7 @@ public: DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { FunctionArgumentDescriptors args{{"time_series", static_cast(&isArray), nullptr, "Array"}}; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } diff --git a/src/Functions/serverConstants.cpp b/src/Functions/serverConstants.cpp index e7e423058f1..fe999d66701 100644 --- a/src/Functions/serverConstants.cpp +++ b/src/Functions/serverConstants.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -20,117 +21,125 @@ namespace DB namespace { + template + class FunctionServerConstantBase : public FunctionConstantBase + { + public: + using FunctionConstantBase::FunctionConstantBase; + bool isServerConstant() const override { return true; } + }; + #if defined(__ELF__) && !defined(OS_FREEBSD) /// buildId() - returns the compiler build id of the running binary. - class FunctionBuildId : public FunctionConstantBase + class FunctionBuildId : public FunctionServerConstantBase { public: static constexpr auto name = "buildId"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionBuildId(ContextPtr context) : FunctionConstantBase(SymbolIndex::instance().getBuildIDHex(), context->isDistributed()) {} + explicit FunctionBuildId(ContextPtr context) : FunctionServerConstantBase(SymbolIndex::instance().getBuildIDHex(), context->isDistributed()) {} }; #endif /// Get the host name. It is constant on single server, but is not constant in distributed queries. - class FunctionHostName : public FunctionConstantBase + class FunctionHostName : public FunctionServerConstantBase { public: static constexpr auto name = "hostName"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionHostName(ContextPtr context) : FunctionConstantBase(DNSResolver::instance().getHostName(), context->isDistributed()) {} + explicit FunctionHostName(ContextPtr context) : FunctionServerConstantBase(DNSResolver::instance().getHostName(), context->isDistributed()) {} }; - class FunctionServerUUID : public FunctionConstantBase + class FunctionServerUUID : public FunctionServerConstantBase { public: static constexpr auto name = "serverUUID"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionServerUUID(ContextPtr context) : FunctionConstantBase(ServerUUID::get(), context->isDistributed()) {} + explicit FunctionServerUUID(ContextPtr context) : FunctionServerConstantBase(ServerUUID::get(), context->isDistributed()) {} }; - class FunctionTCPPort : public FunctionConstantBase + class FunctionTCPPort : public FunctionServerConstantBase { public: static constexpr auto name = "tcpPort"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionTCPPort(ContextPtr context) : FunctionConstantBase(context->getTCPPort(), context->isDistributed()) {} + explicit FunctionTCPPort(ContextPtr context) : FunctionServerConstantBase(context->getTCPPort(), context->isDistributed()) {} }; /// Returns timezone for current session. - class FunctionTimezone : public FunctionConstantBase + class FunctionTimezone : public FunctionServerConstantBase { public: static constexpr auto name = "timezone"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionTimezone(ContextPtr context) : FunctionConstantBase(DateLUT::instance().getTimeZone(), context->isDistributed()) {} + explicit FunctionTimezone(ContextPtr context) : FunctionServerConstantBase(DateLUT::instance().getTimeZone(), context->isDistributed()) {} }; /// Returns the server time zone (timezone in which server runs). - class FunctionServerTimezone : public FunctionConstantBase + class FunctionServerTimezone : public FunctionServerConstantBase { public: static constexpr auto name = "serverTimezone"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionServerTimezone(ContextPtr context) : FunctionConstantBase(DateLUT::serverTimezoneInstance().getTimeZone(), context->isDistributed()) {} + explicit FunctionServerTimezone(ContextPtr context) : FunctionServerConstantBase(DateLUT::serverTimezoneInstance().getTimeZone(), context->isDistributed()) {} }; /// Returns server uptime in seconds. - class FunctionUptime : public FunctionConstantBase + class FunctionUptime : public FunctionServerConstantBase { public: static constexpr auto name = "uptime"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionUptime(ContextPtr context) : FunctionConstantBase(context->getUptimeSeconds(), context->isDistributed()) {} + explicit FunctionUptime(ContextPtr context) : FunctionServerConstantBase(context->getUptimeSeconds(), context->isDistributed()) {} }; /// version() - returns the current version as a string. - class FunctionVersion : public FunctionConstantBase + class FunctionVersion : public FunctionServerConstantBase { public: static constexpr auto name = "version"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionVersion(ContextPtr context) : FunctionConstantBase(VERSION_STRING, context->isDistributed()) {} + explicit FunctionVersion(ContextPtr context) : FunctionServerConstantBase(VERSION_STRING, context->isDistributed()) {} }; /// revision() - returns the current revision. - class FunctionRevision : public FunctionConstantBase + class FunctionRevision : public FunctionServerConstantBase { public: static constexpr auto name = "revision"; static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } - explicit FunctionRevision(ContextPtr context) : FunctionConstantBase(ClickHouseRevision::getVersionRevision(), context->isDistributed()) {} + explicit FunctionRevision(ContextPtr context) : FunctionServerConstantBase(ClickHouseRevision::getVersionRevision(), context->isDistributed()) {} }; - class FunctionZooKeeperSessionUptime : public FunctionConstantBase + class FunctionZooKeeperSessionUptime : public FunctionServerConstantBase { public: static constexpr auto name = "zookeeperSessionUptime"; explicit FunctionZooKeeperSessionUptime(ContextPtr context) - : FunctionConstantBase(context->getZooKeeperSessionUptime(), context->isDistributed()) + : FunctionServerConstantBase(context->getZooKeeperSessionUptime(), context->isDistributed()) { } static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } }; - class FunctionGetOSKernelVersion : public FunctionConstantBase + class FunctionGetOSKernelVersion : public FunctionServerConstantBase { public: static constexpr auto name = "getOSKernelVersion"; - explicit FunctionGetOSKernelVersion(ContextPtr context) : FunctionConstantBase(Poco::Environment::osName() + " " + Poco::Environment::osVersion(), context->isDistributed()) {} + explicit FunctionGetOSKernelVersion(ContextPtr context) : FunctionServerConstantBase(Poco::Environment::osName() + " " + Poco::Environment::osVersion(), context->isDistributed()) {} static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } }; - class FunctionDisplayName : public FunctionConstantBase + class FunctionDisplayName : public FunctionServerConstantBase { public: static constexpr auto name = "displayName"; - explicit FunctionDisplayName(ContextPtr context) : FunctionConstantBase(context->getConfigRef().getString("display_name", getFQDNOrHostName()), context->isDistributed()) {} + explicit FunctionDisplayName(ContextPtr context) : FunctionServerConstantBase(context->getConfigRef().getString("display_name", getFQDNOrHostName()), context->isDistributed()) {} static FunctionPtr create(ContextPtr context) {return std::make_shared(context); } }; } @@ -197,12 +206,12 @@ REGISTER_FUNCTION(Uptime) REGISTER_FUNCTION(Version) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } REGISTER_FUNCTION(Revision) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } REGISTER_FUNCTION(ZooKeeperSessionUptime) @@ -228,8 +237,7 @@ Returns the value of `display_name` from config or server FQDN if not set. )", .examples{{"displayName", "SELECT displayName();", ""}}, .categories{"Constant", "Miscellaneous"} - }, - FunctionFactory::CaseSensitive); + }); } diff --git a/src/Functions/sign.cpp b/src/Functions/sign.cpp index 3dd2ac8e3aa..914e1ad9e1f 100644 --- a/src/Functions/sign.cpp +++ b/src/Functions/sign.cpp @@ -44,7 +44,7 @@ struct FunctionUnaryArithmeticMonotonicity REGISTER_FUNCTION(Sign) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/sin.cpp b/src/Functions/sin.cpp index 914f431adb4..945486b26a8 100644 --- a/src/Functions/sin.cpp +++ b/src/Functions/sin.cpp @@ -21,7 +21,7 @@ REGISTER_FUNCTION(Sin) .returned_value = "The sine of x.", .examples = {{.name = "simple", .query = "SELECT sin(1.23)", .result = "0.9424888019316975"}}, .categories{"Mathematical", "Trigonometric"}}, - FunctionFactory::CaseInsensitive); + FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/sleep.h b/src/Functions/sleep.h index 22748e86888..62ee19fa904 100644 --- a/src/Functions/sleep.h +++ b/src/Functions/sleep.h @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Functions/snowflake.cpp b/src/Functions/snowflake.cpp index 5ff8a636058..0d4d65ee4b0 100644 --- a/src/Functions/snowflake.cpp +++ b/src/Functions/snowflake.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include @@ -64,7 +65,7 @@ public: FunctionArgumentDescriptors args{ {"value", static_cast(&isDateTime), nullptr, "DateTime"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } @@ -121,7 +122,7 @@ public: FunctionArgumentDescriptors optional_args{ {"time_zone", static_cast(&isString), nullptr, "String"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); String timezone; if (arguments.size() == 2) @@ -190,7 +191,7 @@ public: FunctionArgumentDescriptors args{ {"value", static_cast(&isDateTime64), nullptr, "DateTime64"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } @@ -255,7 +256,7 @@ public: FunctionArgumentDescriptors optional_args{ {"time_zone", static_cast(&isString), nullptr, "String"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); String timezone; if (arguments.size() == 2) diff --git a/src/Functions/snowflakeIDToDateTime.cpp b/src/Functions/snowflakeIDToDateTime.cpp index b799792a56f..6c57c72fff1 100644 --- a/src/Functions/snowflakeIDToDateTime.cpp +++ b/src/Functions/snowflakeIDToDateTime.cpp @@ -9,6 +9,7 @@ #include #include #include +#include #include @@ -38,7 +39,7 @@ public: static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } explicit FunctionSnowflakeIDToDateTime(ContextPtr context) - : allow_nonconst_timezone_arguments(context->getSettings().allow_nonconst_timezone_arguments) + : allow_nonconst_timezone_arguments(context->getSettingsRef().allow_nonconst_timezone_arguments) {} String getName() const override { return name; } @@ -56,7 +57,7 @@ public: {"epoch", static_cast(&isNativeUInt), isColumnConst, "const UInt*"}, {"time_zone", static_cast(&isString), nullptr, "String"} }; - validateFunctionArgumentTypes(*this, arguments, args, optional_args); + validateFunctionArguments(*this, arguments, args, optional_args); String timezone; if (arguments.size() == 3) @@ -109,7 +110,7 @@ public: static FunctionPtr create(ContextPtr context) { return std::make_shared(context); } explicit FunctionSnowflakeIDToDateTime64(ContextPtr context) - : allow_nonconst_timezone_arguments(context->getSettings().allow_nonconst_timezone_arguments) + : allow_nonconst_timezone_arguments(context->getSettingsRef().allow_nonconst_timezone_arguments) {} String getName() const override { return name; } @@ -127,7 +128,7 @@ public: {"epoch", static_cast(&isNativeUInt), isColumnConst, "const UInt*"}, {"time_zone", static_cast(&isString), nullptr, "String"} }; - validateFunctionArgumentTypes(*this, arguments, args, optional_args); + validateFunctionArguments(*this, arguments, args, optional_args); String timezone; if (arguments.size() == 3) diff --git a/src/Functions/soundex.cpp b/src/Functions/soundex.cpp index 77ddb14a6ec..e8bd1b664e3 100644 --- a/src/Functions/soundex.cpp +++ b/src/Functions/soundex.cpp @@ -79,14 +79,14 @@ struct SoundexImpl const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - const size_t size = offsets.size(); - res_data.resize(size * (length + 1)); - res_offsets.resize(size); + res_data.resize(input_rows_count * (length + 1)); + res_offsets.resize(input_rows_count); size_t prev_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * value = reinterpret_cast(&data[prev_offset]); const size_t value_length = offsets[i] - prev_offset - 1; @@ -98,7 +98,7 @@ struct SoundexImpl } } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Column of type FixedString is not supported by soundex function"); } @@ -112,7 +112,7 @@ struct NameSoundex REGISTER_FUNCTION(Soundex) { factory.registerFunction>( - FunctionDocumentation{.description="Returns Soundex code of a string."}, FunctionFactory::CaseInsensitive); + FunctionDocumentation{.description="Returns Soundex code of a string."}, FunctionFactory::Case::Insensitive); } diff --git a/src/Functions/space.cpp b/src/Functions/space.cpp index 83183c991bc..cf1634e0319 100644 --- a/src/Functions/space.cpp +++ b/src/Functions/space.cpp @@ -48,14 +48,14 @@ public: {"n", static_cast(&isInteger), nullptr, "Integer"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(); } template - bool executeConstant(ColumnPtr col_times, ColumnString::Offsets & res_offsets, ColumnString::Chars & res_chars) const + bool executeConstant(ColumnPtr col_times, ColumnString::Offsets & res_offsets, ColumnString::Chars & res_chars, size_t input_rows_count) const { const ColumnConst & col_times_const = checkAndGetColumn(*col_times); @@ -71,12 +71,12 @@ public: checkRepeatTime(times); - res_offsets.resize(col_times->size()); - res_chars.resize(col_times->size() * (times + 1)); + res_offsets.resize(input_rows_count); + res_chars.resize(input_rows_count * (times + 1)); size_t pos = 0; - for (size_t i = 0; i < col_times->size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { memset(res_chars.begin() + pos, space, times); pos += times; @@ -92,20 +92,20 @@ public: template - bool executeVector(ColumnPtr col_times_, ColumnString::Offsets & res_offsets, ColumnString::Chars & res_chars) const + bool executeVector(ColumnPtr col_times_, ColumnString::Offsets & res_offsets, ColumnString::Chars & res_chars, size_t input_rows_count) const { auto * col_times = checkAndGetColumn(col_times_.get()); if (!col_times) return false; - res_offsets.resize(col_times->size()); - res_chars.resize(col_times->size() * 10); /// heuristic + res_offsets.resize(input_rows_count); + res_chars.resize(input_rows_count * 10); /// heuristic const PaddedPODArray & times_data = col_times->getData(); size_t pos = 0; - for (size_t i = 0; i < col_times->size(); ++i) + for (size_t i = 0; i < input_rows_count; ++i) { typename DataType::FieldType times = times_data[i]; @@ -132,7 +132,7 @@ public: } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const auto & col_num = arguments[0].column; @@ -143,26 +143,26 @@ public: if (const ColumnConst * col_num_const = checkAndGetColumn(col_num.get())) { - if ((executeConstant(col_num, res_offsets, res_chars)) - || (executeConstant(col_num, res_offsets, res_chars)) - || (executeConstant(col_num, res_offsets, res_chars)) - || (executeConstant(col_num, res_offsets, res_chars)) - || (executeConstant(col_num, res_offsets, res_chars)) - || (executeConstant(col_num, res_offsets, res_chars)) - || (executeConstant(col_num, res_offsets, res_chars)) - || (executeConstant(col_num, res_offsets, res_chars))) + if ((executeConstant(col_num, res_offsets, res_chars, input_rows_count)) + || (executeConstant(col_num, res_offsets, res_chars, input_rows_count)) + || (executeConstant(col_num, res_offsets, res_chars, input_rows_count)) + || (executeConstant(col_num, res_offsets, res_chars, input_rows_count)) + || (executeConstant(col_num, res_offsets, res_chars, input_rows_count)) + || (executeConstant(col_num, res_offsets, res_chars, input_rows_count)) + || (executeConstant(col_num, res_offsets, res_chars, input_rows_count)) + || (executeConstant(col_num, res_offsets, res_chars, input_rows_count))) return col_res; } else { - if ((executeVector(col_num, res_offsets, res_chars)) - || (executeVector(col_num, res_offsets, res_chars)) - || (executeVector(col_num, res_offsets, res_chars)) - || (executeVector(col_num, res_offsets, res_chars)) - || (executeVector(col_num, res_offsets, res_chars)) - || (executeVector(col_num, res_offsets, res_chars)) - || (executeVector(col_num, res_offsets, res_chars)) - || (executeVector(col_num, res_offsets, res_chars))) + if ((executeVector(col_num, res_offsets, res_chars, input_rows_count)) + || (executeVector(col_num, res_offsets, res_chars, input_rows_count)) + || (executeVector(col_num, res_offsets, res_chars, input_rows_count)) + || (executeVector(col_num, res_offsets, res_chars, input_rows_count)) + || (executeVector(col_num, res_offsets, res_chars, input_rows_count)) + || (executeVector(col_num, res_offsets, res_chars, input_rows_count)) + || (executeVector(col_num, res_offsets, res_chars, input_rows_count)) + || (executeVector(col_num, res_offsets, res_chars, input_rows_count))) return col_res; } @@ -173,7 +173,7 @@ public: REGISTER_FUNCTION(Space) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/sqid.cpp b/src/Functions/sqid.cpp index 6679646fef4..0e133590b84 100644 --- a/src/Functions/sqid.cpp +++ b/src/Functions/sqid.cpp @@ -100,7 +100,7 @@ public: FunctionArgumentDescriptors args{ {"sqid", static_cast(&isString), nullptr, "String"} }; - validateFunctionArgumentTypes(*this, arguments, args); + validateFunctionArguments(*this, arguments, args); return std::make_shared(std::make_shared()); } diff --git a/src/Functions/sqrt.cpp b/src/Functions/sqrt.cpp index 3c50f994391..a6e2dee71d9 100644 --- a/src/Functions/sqrt.cpp +++ b/src/Functions/sqrt.cpp @@ -13,7 +13,7 @@ using FunctionSqrt = FunctionMathUnary>; REGISTER_FUNCTION(Sqrt) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/stem.cpp b/src/Functions/stem.cpp index d326d17a7a8..b3be40f4022 100644 --- a/src/Functions/stem.cpp +++ b/src/Functions/stem.cpp @@ -3,6 +3,7 @@ #if USE_NLP #include +#include #include #include #include @@ -31,7 +32,8 @@ struct StemImpl const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, ColumnString::Offsets & res_offsets, - const String & language) + const String & language, + size_t input_rows_count) { sb_stemmer * stemmer = sb_stemmer_new(language.data(), "UTF_8"); @@ -44,7 +46,7 @@ struct StemImpl res_offsets.assign(offsets); UInt64 data_size = 0; - for (UInt64 i = 0; i < offsets.size(); ++i) + for (UInt64 i = 0; i < input_rows_count; ++i) { /// Note that accessing -1th element is valid for PaddedPODArray. size_t original_size = offsets[i] - offsets[i - 1]; @@ -100,7 +102,7 @@ public: ColumnNumbers getArgumentsThatAreAlwaysConstant() const override { return {0}; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const auto & langcolumn = arguments[0].column; const auto & strcolumn = arguments[1].column; @@ -118,7 +120,7 @@ public: String language = lang_col->getValue(); auto col_res = ColumnString::create(); - StemImpl::vector(words_col->getChars(), words_col->getOffsets(), col_res->getChars(), col_res->getOffsets(), language); + StemImpl::vector(words_col->getChars(), words_col->getOffsets(), col_res->getChars(), col_res->getOffsets(), language, input_rows_count); return col_res; } }; diff --git a/src/Functions/stringCutToZero.cpp b/src/Functions/stringCutToZero.cpp index b9f742cd8bc..16e57d741fa 100644 --- a/src/Functions/stringCutToZero.cpp +++ b/src/Functions/stringCutToZero.cpp @@ -40,7 +40,7 @@ public: bool useDefaultImplementationForConstants() const override { return true; } - static bool tryExecuteString(const IColumn * col, ColumnPtr & col_res) + static bool tryExecuteString(const IColumn * col, ColumnPtr & col_res, size_t input_rows_count) { const ColumnString * col_str_in = checkAndGetColumn(col); @@ -53,8 +53,7 @@ public: const ColumnString::Chars & in_vec = col_str_in->getChars(); const ColumnString::Offsets & in_offsets = col_str_in->getOffsets(); - size_t size = in_offsets.size(); - out_offsets.resize(size); + out_offsets.resize(input_rows_count); out_vec.resize(in_vec.size()); char * begin = reinterpret_cast(out_vec.data()); @@ -62,7 +61,7 @@ public: ColumnString::Offset current_in_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * pos_in = reinterpret_cast(&in_vec[current_in_offset]); size_t current_size = strlen(pos_in); @@ -87,7 +86,7 @@ public: } } - static bool tryExecuteFixedString(const IColumn * col, ColumnPtr & col_res) + static bool tryExecuteFixedString(const IColumn * col, ColumnPtr & col_res, size_t input_rows_count) { const ColumnFixedString * col_fstr_in = checkAndGetColumn(col); @@ -99,10 +98,8 @@ public: const ColumnString::Chars & in_vec = col_fstr_in->getChars(); - size_t size = col_fstr_in->size(); - - out_offsets.resize(size); - out_vec.resize(in_vec.size() + size); + out_offsets.resize(input_rows_count); + out_vec.resize(in_vec.size() + input_rows_count); char * begin = reinterpret_cast(out_vec.data()); char * pos = begin; @@ -110,7 +107,7 @@ public: size_t n = col_fstr_in->getN(); - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { size_t current_size = strnlen(pos_in, n); memcpySmallAllowReadWriteOverflow15(pos, pos_in, current_size); @@ -133,12 +130,12 @@ public: } } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const IColumn * column = arguments[0].column.get(); ColumnPtr res_column; - if (tryExecuteFixedString(column, res_column) || tryExecuteString(column, res_column)) + if (tryExecuteFixedString(column, res_column, input_rows_count) || tryExecuteString(column, res_column, input_rows_count)) return res_column; throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of argument of function {}", diff --git a/src/Functions/structureToFormatSchema.cpp b/src/Functions/structureToFormatSchema.cpp index 406da372c04..4fc2bf707a4 100644 --- a/src/Functions/structureToFormatSchema.cpp +++ b/src/Functions/structureToFormatSchema.cpp @@ -116,8 +116,7 @@ Function that converts ClickHouse table structure to CapnProto format schema "}"}, }, .categories{"Other"} - }, - FunctionFactory::CaseSensitive); + }); } @@ -138,8 +137,7 @@ Function that converts ClickHouse table structure to Protobuf format schema "}"}, }, .categories{"Other"} - }, - FunctionFactory::CaseSensitive); + }); } } diff --git a/src/Functions/substring.cpp b/src/Functions/substring.cpp index f1dea7db018..51980eb6b9c 100644 --- a/src/Functions/substring.cpp +++ b/src/Functions/substring.cpp @@ -201,12 +201,12 @@ public: REGISTER_FUNCTION(Substring) { - factory.registerFunction>({}, FunctionFactory::CaseInsensitive); - factory.registerAlias("substr", "substring", FunctionFactory::CaseInsensitive); // MySQL alias - factory.registerAlias("mid", "substring", FunctionFactory::CaseInsensitive); /// MySQL alias - factory.registerAlias("byteSlice", "substring", FunctionFactory::CaseInsensitive); /// resembles PostgreSQL's get_byte function, similar to ClickHouse's bitSlice + factory.registerFunction>({}, FunctionFactory::Case::Insensitive); + factory.registerAlias("substr", "substring", FunctionFactory::Case::Insensitive); // MySQL alias + factory.registerAlias("mid", "substring", FunctionFactory::Case::Insensitive); /// MySQL alias + factory.registerAlias("byteSlice", "substring", FunctionFactory::Case::Insensitive); /// resembles PostgreSQL's get_byte function, similar to ClickHouse's bitSlice - factory.registerFunction>({}, FunctionFactory::CaseSensitive); + factory.registerFunction>(); } } diff --git a/src/Functions/substringIndex.cpp b/src/Functions/substringIndex.cpp index 15a321bd5b0..dc12ae193ff 100644 --- a/src/Functions/substringIndex.cpp +++ b/src/Functions/substringIndex.cpp @@ -68,7 +68,7 @@ namespace return std::make_shared(); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { ColumnPtr column_string = arguments[0].column; ColumnPtr column_delim = arguments[1].column; @@ -110,10 +110,10 @@ namespace if (is_count_const) { Int64 count = column_count->getInt(0); - vectorConstant(col_str, delim, count, vec_res, offsets_res); + vectorConstant(col_str, delim, count, vec_res, offsets_res, input_rows_count); } else - vectorVector(col_str, delim, column_count.get(), vec_res, offsets_res); + vectorVector(col_str, delim, column_count.get(), vec_res, offsets_res, input_rows_count); } return column_res; } @@ -124,18 +124,18 @@ namespace const String & delim, const IColumn * count_column, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - size_t rows = str_column->size(); res_data.reserve(str_column->getChars().size() / 2); - res_offsets.reserve(rows); + res_offsets.reserve(input_rows_count); bool all_ascii = isAllASCII(str_column->getChars().data(), str_column->getChars().size()) && isAllASCII(reinterpret_cast(delim.data()), delim.size()); std::unique_ptr searcher = !is_utf8 || all_ascii ? nullptr : std::make_unique(delim.data(), delim.size()); - for (size_t i = 0; i < rows; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { StringRef str_ref = str_column->getDataAt(i); Int64 count = count_column->getInt(i); @@ -157,18 +157,18 @@ namespace const String & delim, Int64 count, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - size_t rows = str_column->size(); res_data.reserve(str_column->getChars().size() / 2); - res_offsets.reserve(rows); + res_offsets.reserve(input_rows_count); bool all_ascii = isAllASCII(str_column->getChars().data(), str_column->getChars().size()) && isAllASCII(reinterpret_cast(delim.data()), delim.size()); std::unique_ptr searcher = !is_utf8 || all_ascii ? nullptr : std::make_unique(delim.data(), delim.size()); - for (size_t i = 0; i < rows; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { StringRef str_ref = str_column->getDataAt(i); @@ -314,7 +314,7 @@ REGISTER_FUNCTION(SubstringIndex) factory.registerFunction>(); /// substringIndex factory.registerFunction>(); /// substringIndexUTF8 - factory.registerAlias("SUBSTRING_INDEX", "substringIndex", FunctionFactory::CaseInsensitive); + factory.registerAlias("SUBSTRING_INDEX", "substringIndex", FunctionFactory::Case::Insensitive); } diff --git a/src/Functions/subtractNanoseconds.cpp b/src/Functions/subtractNanoseconds.cpp index fffb4eae37a..360c5ecd9cb 100644 --- a/src/Functions/subtractNanoseconds.cpp +++ b/src/Functions/subtractNanoseconds.cpp @@ -6,6 +6,7 @@ namespace DB { using FunctionSubtractNanoseconds = FunctionDateOrDateTimeAddInterval; + REGISTER_FUNCTION(SubtractNanoseconds) { factory.registerFunction(); diff --git a/src/Functions/synonyms.cpp b/src/Functions/synonyms.cpp index 6b3e205785c..18c1557115f 100644 --- a/src/Functions/synonyms.cpp +++ b/src/Functions/synonyms.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -120,7 +121,7 @@ public: REGISTER_FUNCTION(Synonyms) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/tan.cpp b/src/Functions/tan.cpp index e39f8598419..51cf0bbcceb 100644 --- a/src/Functions/tan.cpp +++ b/src/Functions/tan.cpp @@ -13,7 +13,7 @@ using FunctionTan = FunctionMathUnary>; REGISTER_FUNCTION(Tan) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/tanh.cpp b/src/Functions/tanh.cpp index bdefa5263d7..62755737f70 100644 --- a/src/Functions/tanh.cpp +++ b/src/Functions/tanh.cpp @@ -39,7 +39,7 @@ using FunctionTanh = FunctionMathUnary>; REGISTER_FUNCTION(Tanh) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/tests/gtest_ternary_logic.cpp b/src/Functions/tests/gtest_ternary_logic.cpp deleted file mode 100644 index 5ecafabb361..00000000000 --- a/src/Functions/tests/gtest_ternary_logic.cpp +++ /dev/null @@ -1,354 +0,0 @@ -#include -#include -#include -#include -#include -#include -#include -#include -#include - -// I know that inclusion of .cpp is not good at all -#include // NOLINT - -using namespace DB; -using TernaryValues = std::vector; - -struct LinearCongruentialGenerator -{ - /// Constants from `man lrand48_r`. - static constexpr UInt64 a = 0x5DEECE66D; - static constexpr UInt64 c = 0xB; - - /// And this is from `head -c8 /dev/urandom | xxd -p` - UInt64 current = 0x09826f4a081cee35ULL; - - UInt32 next() - { - current = current * a + c; - return static_cast(current >> 16); - } -}; - -void generateRandomTernaryValue(LinearCongruentialGenerator & gen, Ternary::ResultType * output, size_t size, double false_ratio, double null_ratio) -{ - /// The LinearCongruentialGenerator generates nonnegative integers uniformly distributed over the interval [0, 2^32). - /// See https://linux.die.net/man/3/nrand48 - - double false_percentile = false_ratio; - double null_percentile = false_ratio + null_ratio; - - false_percentile = false_percentile > 1 ? 1 : false_percentile; - null_percentile = null_percentile > 1 ? 1 : null_percentile; - - UInt32 false_threshold = static_cast(static_cast(std::numeric_limits::max()) * false_percentile); - UInt32 null_threshold = static_cast(static_cast(std::numeric_limits::max()) * null_percentile); - - for (Ternary::ResultType * end = output + size; output != end; ++output) - { - UInt32 val = gen.next(); - *output = val < false_threshold ? Ternary::False : (val < null_threshold ? Ternary::Null : Ternary::True); - } -} - -template -ColumnPtr createColumnNullable(const Ternary::ResultType * ternary_values, size_t size) -{ - auto nested_column = ColumnVector::create(size); - auto null_map = ColumnUInt8::create(size); - auto & nested_column_data = nested_column->getData(); - auto & null_map_data = null_map->getData(); - - for (size_t i = 0; i < size; ++i) - { - if (ternary_values[i] == Ternary::Null) - { - null_map_data[i] = 1; - nested_column_data[i] = 0; - } - else if (ternary_values[i] == Ternary::True) - { - null_map_data[i] = 0; - nested_column_data[i] = 100; - } - else - { - null_map_data[i] = 0; - nested_column_data[i] = 0; - } - } - - return ColumnNullable::create(std::move(nested_column), std::move(null_map)); -} - -template -ColumnPtr createColumnVector(const Ternary::ResultType * ternary_values, size_t size) -{ - auto column = ColumnVector::create(size); - auto & column_data = column->getData(); - - for (size_t i = 0; i < size; ++i) - { - if (ternary_values[i] == Ternary::True) - { - column_data[i] = 100; - } - else - { - column_data[i] = 0; - } - } - - return column; -} - -template -ColumnPtr createRandomColumn(LinearCongruentialGenerator & gen, TernaryValues & ternary_values) -{ - size_t size = ternary_values.size(); - Ternary::ResultType * ternary_data = ternary_values.data(); - - if constexpr (std::is_same_v) - { - generateRandomTernaryValue(gen, ternary_data, size, 0.3, 0.7); - return createColumnNullable(ternary_data, size); - } - else if constexpr (std::is_same_v>) - { - generateRandomTernaryValue(gen, ternary_data, size, 0.5, 0); - return createColumnVector(ternary_data, size); - } - else - { - auto nested_col = ColumnNothing::create(size); - auto null_map = ColumnUInt8::create(size); - - memset(ternary_data, Ternary::Null, size); - - return ColumnNullable::create(std::move(nested_col), std::move(null_map)); - } -} - -/* The truth table of ternary And and Or operations: - * +-------+-------+---------+--------+ - * | a | b | a And b | a Or b | - * +-------+-------+---------+--------+ - * | False | False | False | False | - * | False | Null | False | Null | - * | False | True | False | True | - * | Null | False | False | Null | - * | Null | Null | Null | Null | - * | Null | True | Null | True | - * | True | False | False | True | - * | True | Null | Null | True | - * | True | True | True | True | - * +-------+-------+---------+--------+ - * - * https://en.wikibooks.org/wiki/Structured_Query_Language/NULLs_and_the_Three_Valued_Logic - */ -template -bool testTernaryLogicTruthTable() -{ - constexpr size_t size = 9; - - Ternary::ResultType col_a_ternary[] = {Ternary::False, Ternary::False, Ternary::False, Ternary::Null, Ternary::Null, Ternary::Null, Ternary::True, Ternary::True, Ternary::True}; - Ternary::ResultType col_b_ternary[] = {Ternary::False, Ternary::Null, Ternary::True, Ternary::False, Ternary::Null, Ternary::True,Ternary::False, Ternary::Null, Ternary::True}; - Ternary::ResultType and_expected_ternary[] = {Ternary::False, Ternary::False, Ternary::False, Ternary::False, Ternary::Null, Ternary::Null,Ternary::False, Ternary::Null, Ternary::True}; - Ternary::ResultType or_expected_ternary[] = {Ternary::False, Ternary::Null, Ternary::True, Ternary::Null, Ternary::Null, Ternary::True,Ternary::True, Ternary::True, Ternary::True}; - Ternary::ResultType * expected_ternary; - - - if constexpr (std::is_same_v) - { - expected_ternary = and_expected_ternary; - } - else - { - expected_ternary = or_expected_ternary; - } - - auto col_a = createColumnNullable(col_a_ternary, size); - auto col_b = createColumnNullable(col_b_ternary, size); - ColumnRawPtrs arguments = {col_a.get(), col_b.get()}; - - auto col_res = ColumnUInt8::create(size); - auto & col_res_data = col_res->getData(); - - OperationApplier::apply(arguments, col_res->getData(), false); - - for (size_t i = 0; i < size; ++i) - { - if (col_res_data[i] != expected_ternary[i]) return false; - } - - return true; -} - -template -bool testTernaryLogicOfTwoColumns(size_t size) -{ - LinearCongruentialGenerator gen; - - TernaryValues left_column_ternary(size); - TernaryValues right_column_ternary(size); - TernaryValues expected_ternary(size); - - ColumnPtr left = createRandomColumn(gen, left_column_ternary); - ColumnPtr right = createRandomColumn(gen, right_column_ternary); - - for (size_t i = 0; i < size; ++i) - { - /// Given that False is less than Null and Null is less than True, the And operation can be implemented - /// with std::min, and the Or operation can be implemented with std::max. - if constexpr (std::is_same_v) - { - expected_ternary[i] = std::min(left_column_ternary[i], right_column_ternary[i]); - } - else - { - expected_ternary[i] = std::max(left_column_ternary[i], right_column_ternary[i]); - } - } - - ColumnRawPtrs arguments = {left.get(), right.get()}; - - auto col_res = ColumnUInt8::create(size); - auto & col_res_data = col_res->getData(); - - OperationApplier::apply(arguments, col_res->getData(), false); - - for (size_t i = 0; i < size; ++i) - { - if (col_res_data[i] != expected_ternary[i]) return false; - } - - return true; -} - -TEST(TernaryLogicTruthTable, NestedUInt8) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedUInt16) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedUInt32) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedUInt64) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedInt8) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedInt16) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedInt32) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedInt64) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedFloat32) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTruthTable, NestedFloat64) -{ - bool test_1 = testTernaryLogicTruthTable(); - bool test_2 = testTernaryLogicTruthTable(); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTwoColumns, TwoNullable) -{ - bool test_1 = testTernaryLogicOfTwoColumns(100 /*size*/); - bool test_2 = testTernaryLogicOfTwoColumns(100 /*size*/); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTwoColumns, TwoVector) -{ - bool test_1 = testTernaryLogicOfTwoColumns(100 /*size*/); - bool test_2 = testTernaryLogicOfTwoColumns(100 /*size*/); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTwoColumns, TwoNothing) -{ - bool test_1 = testTernaryLogicOfTwoColumns(100 /*size*/); - bool test_2 = testTernaryLogicOfTwoColumns(100 /*size*/); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTwoColumns, NullableVector) -{ - bool test_1 = testTernaryLogicOfTwoColumns(100 /*size*/); - bool test_2 = testTernaryLogicOfTwoColumns(100 /*size*/); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTwoColumns, NullableNothing) -{ - bool test_1 = testTernaryLogicOfTwoColumns(100 /*size*/); - bool test_2 = testTernaryLogicOfTwoColumns(100 /*size*/); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} - -TEST(TernaryLogicTwoColumns, VectorNothing) -{ - bool test_1 = testTernaryLogicOfTwoColumns(100 /*size*/); - bool test_2 = testTernaryLogicOfTwoColumns(100 /*size*/); - ASSERT_EQ(test_1, true); - ASSERT_EQ(test_2, true); -} diff --git a/src/Functions/throwIf.cpp b/src/Functions/throwIf.cpp index c6ff5b9151f..e317c65c622 100644 --- a/src/Functions/throwIf.cpp +++ b/src/Functions/throwIf.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include #include @@ -151,7 +152,7 @@ private: return nullptr; } - bool allow_custom_error_code_argument; + const bool allow_custom_error_code_argument; }; } diff --git a/src/Functions/timeSlots.cpp b/src/Functions/timeSlots.cpp index 040495ab023..b62bb20c64e 100644 --- a/src/Functions/timeSlots.cpp +++ b/src/Functions/timeSlots.cpp @@ -41,18 +41,17 @@ struct TimeSlotsImpl /// The following three methods process DateTime type static void vectorVector( const PaddedPODArray & starts, const PaddedPODArray & durations, UInt32 time_slot_size, - PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets) + PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, + size_t input_rows_count) { if (time_slot_size == 0) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Time slot size cannot be zero"); - size_t size = starts.size(); - - result_offsets.resize(size); - result_values.reserve(size); + result_offsets.resize(input_rows_count); + result_values.reserve(input_rows_count); ColumnArray::Offset current_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { for (UInt32 value = starts[i] / time_slot_size, end = (starts[i] + durations[i]) / time_slot_size; value <= end; ++value) { @@ -66,18 +65,17 @@ struct TimeSlotsImpl static void vectorConstant( const PaddedPODArray & starts, UInt32 duration, UInt32 time_slot_size, - PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets) + PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, + size_t input_rows_count) { if (time_slot_size == 0) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Time slot size cannot be zero"); - size_t size = starts.size(); - - result_offsets.resize(size); - result_values.reserve(size); + result_offsets.resize(input_rows_count); + result_values.reserve(input_rows_count); ColumnArray::Offset current_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { for (UInt32 value = starts[i] / time_slot_size, end = (starts[i] + duration) / time_slot_size; value <= end; ++value) { @@ -91,18 +89,17 @@ struct TimeSlotsImpl static void constantVector( UInt32 start, const PaddedPODArray & durations, UInt32 time_slot_size, - PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets) + PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, + size_t input_rows_count) { if (time_slot_size == 0) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Time slot size cannot be zero"); - size_t size = durations.size(); - - result_offsets.resize(size); - result_values.reserve(size); + result_offsets.resize(input_rows_count); + result_values.reserve(input_rows_count); ColumnArray::Offset current_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { for (UInt32 value = start / time_slot_size, end = (start + durations[i]) / time_slot_size; value <= end; ++value) { @@ -120,12 +117,11 @@ struct TimeSlotsImpl */ static NO_SANITIZE_UNDEFINED void vectorVector( const PaddedPODArray & starts, const PaddedPODArray & durations, Decimal64 time_slot_size, - PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, UInt16 dt_scale, UInt16 duration_scale, UInt16 time_slot_scale) + PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, UInt16 dt_scale, UInt16 duration_scale, UInt16 time_slot_scale, + size_t input_rows_count) { - size_t size = starts.size(); - - result_offsets.resize(size); - result_values.reserve(size); + result_offsets.resize(input_rows_count); + result_values.reserve(input_rows_count); /// Modify all units to have same scale UInt16 max_scale = std::max({dt_scale, duration_scale, time_slot_scale}); @@ -139,7 +135,7 @@ struct TimeSlotsImpl if (time_slot_size == 0) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Time slot size cannot be zero"); - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { for (DateTime64 value = (starts[i] * dt_multiplier) / time_slot_size, end = (starts[i] * dt_multiplier + durations[i] * dur_multiplier) / time_slot_size; value <= end; value += 1) { @@ -152,12 +148,11 @@ struct TimeSlotsImpl static NO_SANITIZE_UNDEFINED void vectorConstant( const PaddedPODArray & starts, Decimal64 duration, Decimal64 time_slot_size, - PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, UInt16 dt_scale, UInt16 duration_scale, UInt16 time_slot_scale) + PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, UInt16 dt_scale, UInt16 duration_scale, UInt16 time_slot_scale, + size_t input_rows_count) { - size_t size = starts.size(); - - result_offsets.resize(size); - result_values.reserve(size); + result_offsets.resize(input_rows_count); + result_values.reserve(input_rows_count); /// Modify all units to have same scale UInt16 max_scale = std::max({dt_scale, duration_scale, time_slot_scale}); @@ -172,7 +167,7 @@ struct TimeSlotsImpl if (time_slot_size == 0) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Time slot size cannot be zero"); - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { for (DateTime64 value = (starts[i] * dt_multiplier) / time_slot_size, end = (starts[i] * dt_multiplier + duration) / time_slot_size; value <= end; value += 1) { @@ -185,12 +180,11 @@ struct TimeSlotsImpl static NO_SANITIZE_UNDEFINED void constantVector( DateTime64 start, const PaddedPODArray & durations, Decimal64 time_slot_size, - PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, UInt16 dt_scale, UInt16 duration_scale, UInt16 time_slot_scale) + PaddedPODArray & result_values, ColumnArray::Offsets & result_offsets, UInt16 dt_scale, UInt16 duration_scale, UInt16 time_slot_scale, + size_t input_rows_count) { - size_t size = durations.size(); - - result_offsets.resize(size); - result_values.reserve(size); + result_offsets.resize(input_rows_count); + result_values.reserve(input_rows_count); /// Modify all units to have same scale UInt16 max_scale = std::max({dt_scale, duration_scale, time_slot_scale}); @@ -205,7 +199,7 @@ struct TimeSlotsImpl if (time_slot_size == 0) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Time slot size cannot be zero"); - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { for (DateTime64 value = start / time_slot_size, end = (start + durations[i] * dur_multiplier) / time_slot_size; value <= end; value += 1) { @@ -282,7 +276,7 @@ public: } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { if (WhichDataType(arguments[0].type).isDateTime()) { @@ -308,17 +302,17 @@ public: if (dt_starts && durations) { - TimeSlotsImpl::vectorVector(dt_starts->getData(), durations->getData(), time_slot_size, res_values, res->getOffsets()); + TimeSlotsImpl::vectorVector(dt_starts->getData(), durations->getData(), time_slot_size, res_values, res->getOffsets(), input_rows_count); return res; } else if (dt_starts && const_durations) { - TimeSlotsImpl::vectorConstant(dt_starts->getData(), const_durations->getValue(), time_slot_size, res_values, res->getOffsets()); + TimeSlotsImpl::vectorConstant(dt_starts->getData(), const_durations->getValue(), time_slot_size, res_values, res->getOffsets(), input_rows_count); return res; } else if (dt_const_starts && durations) { - TimeSlotsImpl::constantVector(dt_const_starts->getValue(), durations->getData(), time_slot_size, res_values, res->getOffsets()); + TimeSlotsImpl::constantVector(dt_const_starts->getValue(), durations->getData(), time_slot_size, res_values, res->getOffsets(), input_rows_count); return res; } } @@ -353,21 +347,21 @@ public: if (starts && durations) { TimeSlotsImpl::vectorVector(starts->getData(), durations->getData(), time_slot_size, res_values, res->getOffsets(), - start_time_scale, duration_scale, time_slot_scale); + start_time_scale, duration_scale, time_slot_scale, input_rows_count); return res; } else if (starts && const_durations) { TimeSlotsImpl::vectorConstant( starts->getData(), const_durations->getValue(), time_slot_size, res_values, res->getOffsets(), - start_time_scale, duration_scale, time_slot_scale); + start_time_scale, duration_scale, time_slot_scale, input_rows_count); return res; } else if (const_starts && durations) { TimeSlotsImpl::constantVector( const_starts->getValue(), durations->getData(), time_slot_size, res_values, res->getOffsets(), - start_time_scale, duration_scale, time_slot_scale); + start_time_scale, duration_scale, time_slot_scale, input_rows_count); return res; } } diff --git a/src/Functions/timestamp.cpp b/src/Functions/timestamp.cpp index fbca08b0968..c2e10a2d220 100644 --- a/src/Functions/timestamp.cpp +++ b/src/Functions/timestamp.cpp @@ -46,7 +46,7 @@ public: FunctionArgumentDescriptors optional_args{ {"time", static_cast(&isString), nullptr, "String"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, optional_args); + validateFunctionArguments(*this, arguments, mandatory_args, optional_args); return std::make_shared(DATETIME_SCALE); } @@ -187,7 +187,7 @@ If the second argument 'expr_time' is provided, it adds the specified time to th {"timestamp", "SELECT timestamp('2013-12-31 12:00:00')", "2013-12-31 12:00:00.000000"}, {"timestamp", "SELECT timestamp('2013-12-31 12:00:00', '12:00:00.11')", "2014-01-01 00:00:00.110000"}, }, - .categories{"DateTime"}}, FunctionFactory::CaseInsensitive); + .categories{"DateTime"}}, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toCustomWeek.cpp b/src/Functions/toCustomWeek.cpp index 98e7aaf1d6b..61c0767654e 100644 --- a/src/Functions/toCustomWeek.cpp +++ b/src/Functions/toCustomWeek.cpp @@ -21,8 +21,8 @@ REGISTER_FUNCTION(ToCustomWeek) factory.registerFunction(); /// Compatibility aliases for mysql. - factory.registerAlias("week", "toWeek", FunctionFactory::CaseInsensitive); - factory.registerAlias("yearweek", "toYearWeek", FunctionFactory::CaseInsensitive); + factory.registerAlias("week", "toWeek", FunctionFactory::Case::Insensitive); + factory.registerAlias("yearweek", "toYearWeek", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toDayOfMonth.cpp b/src/Functions/toDayOfMonth.cpp index c20b0b75797..93013c3528b 100644 --- a/src/Functions/toDayOfMonth.cpp +++ b/src/Functions/toDayOfMonth.cpp @@ -14,8 +14,8 @@ REGISTER_FUNCTION(ToDayOfMonth) factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("DAY", "toDayOfMonth", FunctionFactory::CaseInsensitive); - factory.registerAlias("DAYOFMONTH", "toDayOfMonth", FunctionFactory::CaseInsensitive); + factory.registerAlias("DAY", "toDayOfMonth", FunctionFactory::Case::Insensitive); + factory.registerAlias("DAYOFMONTH", "toDayOfMonth", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toDayOfWeek.cpp b/src/Functions/toDayOfWeek.cpp index dc508d70814..d1f55bbddab 100644 --- a/src/Functions/toDayOfWeek.cpp +++ b/src/Functions/toDayOfWeek.cpp @@ -13,7 +13,7 @@ REGISTER_FUNCTION(ToDayOfWeek) factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("DAYOFWEEK", "toDayOfWeek", FunctionFactory::CaseInsensitive); + factory.registerAlias("DAYOFWEEK", "toDayOfWeek", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toDayOfYear.cpp b/src/Functions/toDayOfYear.cpp index 0cbafd6275a..9a27c41b0ed 100644 --- a/src/Functions/toDayOfYear.cpp +++ b/src/Functions/toDayOfYear.cpp @@ -14,7 +14,7 @@ REGISTER_FUNCTION(ToDayOfYear) factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("DAYOFYEAR", "toDayOfYear", FunctionFactory::CaseInsensitive); + factory.registerAlias("DAYOFYEAR", "toDayOfYear", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toDaysSinceYearZero.cpp b/src/Functions/toDaysSinceYearZero.cpp index f6239b2900b..b5c053a11b3 100644 --- a/src/Functions/toDaysSinceYearZero.cpp +++ b/src/Functions/toDaysSinceYearZero.cpp @@ -20,7 +20,7 @@ The calculation is the same as in MySQL's TO_DAYS() function. .categories{"Dates and Times"}}); /// MySQL compatibility alias. - factory.registerAlias("TO_DAYS", FunctionToDaysSinceYearZero::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("TO_DAYS", FunctionToDaysSinceYearZero::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toDecimalString.cpp b/src/Functions/toDecimalString.cpp index fc621b272de..3566ebc93ad 100644 --- a/src/Functions/toDecimalString.cpp +++ b/src/Functions/toDecimalString.cpp @@ -43,7 +43,7 @@ public: {"precision", static_cast(&isNativeInteger), &isColumnConst, "const Integer"} }; - validateFunctionArgumentTypes(*this, arguments, mandatory_args, {}); + validateFunctionArguments(*this, arguments, mandatory_args, {}); return std::make_shared(); } @@ -54,9 +54,9 @@ private: /// For operations with Integer/Float template void vectorConstant(const FromVectorType & vec_from, UInt8 precision, - ColumnString::Chars & vec_to, ColumnString::Offsets & result_offsets) const + ColumnString::Chars & vec_to, ColumnString::Offsets & result_offsets, + size_t input_rows_count) const { - size_t input_rows_count = vec_from.size(); result_offsets.resize(input_rows_count); /// Buffer is used here and in functions below because resulting size cannot be precisely anticipated, @@ -74,9 +74,9 @@ private: template void vectorVector(const FirstArgVectorType & vec_from, const ColumnVector::Container & vec_precision, - ColumnString::Chars & vec_to, ColumnString::Offsets & result_offsets) const + ColumnString::Chars & vec_to, ColumnString::Offsets & result_offsets, + size_t input_rows_count) const { - size_t input_rows_count = vec_from.size(); result_offsets.resize(input_rows_count); WriteBufferFromVector buf_to(vec_to); @@ -98,7 +98,8 @@ private: /// For operations with Decimal template void vectorConstant(const FirstArgVectorType & vec_from, UInt8 precision, - ColumnString::Chars & vec_to, ColumnString::Offsets & result_offsets, UInt8 from_scale) const + ColumnString::Chars & vec_to, ColumnString::Offsets & result_offsets, UInt8 from_scale, + size_t input_rows_count) const { /// There are no more than 77 meaning digits (as it is the max length of UInt256). So we can limit it with 77. constexpr size_t max_digits = std::numeric_limits::digits10; @@ -107,7 +108,6 @@ private: "Too many fractional digits requested for Decimal, must not be more than {}", max_digits); WriteBufferFromVector buf_to(vec_to); - size_t input_rows_count = vec_from.size(); result_offsets.resize(input_rows_count); for (size_t i = 0; i < input_rows_count; ++i) @@ -121,9 +121,9 @@ private: template void vectorVector(const FirstArgVectorType & vec_from, const ColumnVector::Container & vec_precision, - ColumnString::Chars & vec_to, ColumnString::Offsets & result_offsets, UInt8 from_scale) const + ColumnString::Chars & vec_to, ColumnString::Offsets & result_offsets, UInt8 from_scale, + size_t input_rows_count) const { - size_t input_rows_count = vec_from.size(); result_offsets.resize(input_rows_count); WriteBufferFromVector buf_to(vec_to); @@ -182,28 +182,28 @@ private: } public: - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { switch (arguments[0].type->getTypeId()) { - case TypeIndex::UInt8: return executeType(arguments); - case TypeIndex::UInt16: return executeType(arguments); - case TypeIndex::UInt32: return executeType(arguments); - case TypeIndex::UInt64: return executeType(arguments); - case TypeIndex::UInt128: return executeType(arguments); - case TypeIndex::UInt256: return executeType(arguments); - case TypeIndex::Int8: return executeType(arguments); - case TypeIndex::Int16: return executeType(arguments); - case TypeIndex::Int32: return executeType(arguments); - case TypeIndex::Int64: return executeType(arguments); - case TypeIndex::Int128: return executeType(arguments); - case TypeIndex::Int256: return executeType(arguments); - case TypeIndex::Float32: return executeType(arguments); - case TypeIndex::Float64: return executeType(arguments); - case TypeIndex::Decimal32: return executeType(arguments); - case TypeIndex::Decimal64: return executeType(arguments); - case TypeIndex::Decimal128: return executeType(arguments); - case TypeIndex::Decimal256: return executeType(arguments); + case TypeIndex::UInt8: return executeType(arguments, input_rows_count); + case TypeIndex::UInt16: return executeType(arguments, input_rows_count); + case TypeIndex::UInt32: return executeType(arguments, input_rows_count); + case TypeIndex::UInt64: return executeType(arguments, input_rows_count); + case TypeIndex::UInt128: return executeType(arguments, input_rows_count); + case TypeIndex::UInt256: return executeType(arguments, input_rows_count); + case TypeIndex::Int8: return executeType(arguments, input_rows_count); + case TypeIndex::Int16: return executeType(arguments, input_rows_count); + case TypeIndex::Int32: return executeType(arguments, input_rows_count); + case TypeIndex::Int64: return executeType(arguments, input_rows_count); + case TypeIndex::Int128: return executeType(arguments, input_rows_count); + case TypeIndex::Int256: return executeType(arguments, input_rows_count); + case TypeIndex::Float32: return executeType(arguments, input_rows_count); + case TypeIndex::Float64: return executeType(arguments, input_rows_count); + case TypeIndex::Decimal32: return executeType(arguments, input_rows_count); + case TypeIndex::Decimal64: return executeType(arguments, input_rows_count); + case TypeIndex::Decimal128: return executeType(arguments, input_rows_count); + case TypeIndex::Decimal256: return executeType(arguments, input_rows_count); default: throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of argument of function {}", arguments[0].column->getName(), getName()); @@ -212,7 +212,7 @@ public: private: template - ColumnPtr executeType(const ColumnsWithTypeAndName & arguments) const + ColumnPtr executeType(const ColumnsWithTypeAndName & arguments, size_t input_rows_count) const { const auto * precision_col = checkAndGetColumn>(arguments[1].column.get()); const auto * precision_col_const = checkAndGetColumnConst>(arguments[1].column.get()); @@ -230,9 +230,9 @@ private: { UInt8 from_scale = from_col->getScale(); if (precision_col_const) - vectorConstant(from_col->getData(), precision_col_const->template getValue(), result_chars, result_offsets, from_scale); + vectorConstant(from_col->getData(), precision_col_const->template getValue(), result_chars, result_offsets, from_scale, input_rows_count); else if (precision_col) - vectorVector(from_col->getData(), precision_col->getData(), result_chars, result_offsets, from_scale); + vectorVector(from_col->getData(), precision_col->getData(), result_chars, result_offsets, from_scale, input_rows_count); else throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of second argument of function formatDecimal", arguments[1].column->getName()); } @@ -245,9 +245,9 @@ private: if (from_col) { if (precision_col_const) - vectorConstant(from_col->getData(), precision_col_const->template getValue(), result_chars, result_offsets); + vectorConstant(from_col->getData(), precision_col_const->template getValue(), result_chars, result_offsets, input_rows_count); else if (precision_col) - vectorVector(from_col->getData(), precision_col->getData(), result_chars, result_offsets); + vectorVector(from_col->getData(), precision_col->getData(), result_chars, result_offsets, input_rows_count); else throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of second argument of function formatDecimal", arguments[1].column->getName()); @@ -273,7 +273,7 @@ second argument is the desired number of digits in fractional part. Returns Stri )", .examples{{"toDecimalString", "SELECT toDecimalString(2.1456,2)", ""}}, .categories{"String"} - }, FunctionFactory::CaseInsensitive); + }, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toHour.cpp b/src/Functions/toHour.cpp index fc9ec657adf..bc122538661 100644 --- a/src/Functions/toHour.cpp +++ b/src/Functions/toHour.cpp @@ -14,7 +14,7 @@ REGISTER_FUNCTION(ToHour) factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("HOUR", "toHour", FunctionFactory::CaseInsensitive); + factory.registerAlias("HOUR", "toHour", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toLastDayOfMonth.cpp b/src/Functions/toLastDayOfMonth.cpp index 9365880bfb8..004ae2718e7 100644 --- a/src/Functions/toLastDayOfMonth.cpp +++ b/src/Functions/toLastDayOfMonth.cpp @@ -13,7 +13,7 @@ REGISTER_FUNCTION(ToLastDayOfMonth) factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("LAST_DAY", "toLastDayOfMonth", FunctionFactory::CaseInsensitive); + factory.registerAlias("LAST_DAY", "toLastDayOfMonth", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toMillisecond.cpp b/src/Functions/toMillisecond.cpp index aaef517c996..efa08c322a2 100644 --- a/src/Functions/toMillisecond.cpp +++ b/src/Functions/toMillisecond.cpp @@ -27,7 +27,7 @@ Returns the millisecond component (0-999) of a date with time. ); /// MySQL compatibility alias. - factory.registerAlias("MILLISECOND", "toMillisecond", FunctionFactory::CaseInsensitive); + factory.registerAlias("MILLISECOND", "toMillisecond", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toMinute.cpp b/src/Functions/toMinute.cpp index 162ecb282df..291da33d2e8 100644 --- a/src/Functions/toMinute.cpp +++ b/src/Functions/toMinute.cpp @@ -14,7 +14,7 @@ REGISTER_FUNCTION(ToMinute) factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("MINUTE", "toMinute", FunctionFactory::CaseInsensitive); + factory.registerAlias("MINUTE", "toMinute", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toMonth.cpp b/src/Functions/toMonth.cpp index 422f21e7df8..3ef73bf1be3 100644 --- a/src/Functions/toMonth.cpp +++ b/src/Functions/toMonth.cpp @@ -13,7 +13,7 @@ REGISTER_FUNCTION(ToMonth) { factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("MONTH", "toMonth", FunctionFactory::CaseInsensitive); + factory.registerAlias("MONTH", "toMonth", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toQuarter.cpp b/src/Functions/toQuarter.cpp index 3c301095ff2..2e6d4fa93de 100644 --- a/src/Functions/toQuarter.cpp +++ b/src/Functions/toQuarter.cpp @@ -13,7 +13,7 @@ REGISTER_FUNCTION(ToQuarter) { factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("QUARTER", "toQuarter", FunctionFactory::CaseInsensitive); + factory.registerAlias("QUARTER", "toQuarter", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toSecond.cpp b/src/Functions/toSecond.cpp index 372097fd488..1ad3b46fbd7 100644 --- a/src/Functions/toSecond.cpp +++ b/src/Functions/toSecond.cpp @@ -14,7 +14,7 @@ REGISTER_FUNCTION(ToSecond) factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("SECOND", "toSecond", FunctionFactory::CaseInsensitive); + factory.registerAlias("SECOND", "toSecond", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/toStartOfInterval.cpp b/src/Functions/toStartOfInterval.cpp index 50442d1b448..21b7cf895d2 100644 --- a/src/Functions/toStartOfInterval.cpp +++ b/src/Functions/toStartOfInterval.cpp @@ -147,19 +147,20 @@ public: std::unreachable(); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t /* input_rows_count */) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override { const auto & time_column = arguments[0]; const auto & interval_column = arguments[1]; const auto & time_zone = extractTimeZoneFromFunctionArguments(arguments, 2, 0); - auto result_column = dispatchForTimeColumn(time_column, interval_column, result_type, time_zone); + auto result_column = dispatchForTimeColumn(time_column, interval_column, result_type, time_zone, input_rows_count); return result_column; } private: ColumnPtr dispatchForTimeColumn( const ColumnWithTypeAndName & time_column, const ColumnWithTypeAndName & interval_column, - const DataTypePtr & result_type, const DateLUTImpl & time_zone) const + const DataTypePtr & result_type, const DateLUTImpl & time_zone, + size_t input_rows_count) const { const auto & time_column_type = *time_column.type.get(); const auto & time_column_col = *time_column.column.get(); @@ -170,19 +171,19 @@ private: auto scale = assert_cast(time_column_type).getScale(); if (time_column_vec) - return dispatchForIntervalColumn(assert_cast(time_column_type), *time_column_vec, interval_column, result_type, time_zone, scale); + return dispatchForIntervalColumn(assert_cast(time_column_type), *time_column_vec, interval_column, result_type, time_zone, input_rows_count, scale); } else if (isDateTime(time_column_type)) { const auto * time_column_vec = checkAndGetColumn(&time_column_col); if (time_column_vec) - return dispatchForIntervalColumn(assert_cast(time_column_type), *time_column_vec, interval_column, result_type, time_zone); + return dispatchForIntervalColumn(assert_cast(time_column_type), *time_column_vec, interval_column, result_type, time_zone, input_rows_count); } else if (isDate(time_column_type)) { const auto * time_column_vec = checkAndGetColumn(&time_column_col); if (time_column_vec) - return dispatchForIntervalColumn(assert_cast(time_column_type), *time_column_vec, interval_column, result_type, time_zone); + return dispatchForIntervalColumn(assert_cast(time_column_type), *time_column_vec, interval_column, result_type, time_zone, input_rows_count); } throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal column for 1st argument of function {}, expected a Date, DateTime or DateTime64", getName()); } @@ -190,7 +191,7 @@ private: template ColumnPtr dispatchForIntervalColumn( const TimeDataType & time_data_type, const TimeColumnType & time_column, const ColumnWithTypeAndName & interval_column, - const DataTypePtr & result_type, const DateLUTImpl & time_zone, UInt16 scale = 1) const + const DataTypePtr & result_type, const DateLUTImpl & time_zone, size_t input_rows_count, UInt16 scale = 1) const { const auto * interval_type = checkAndGetDataType(interval_column.type.get()); if (!interval_type) @@ -207,27 +208,27 @@ private: switch (interval_type->getKind()) // NOLINT(bugprone-switch-missing-default-case) { case IntervalKind::Kind::Nanosecond: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Microsecond: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Millisecond: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Second: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Minute: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Hour: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Day: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Week: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Month: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Quarter: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); case IntervalKind::Kind::Year: - return execute(time_data_type, time_column, num_units, result_type, time_zone, scale); + return execute(time_data_type, time_column, num_units, result_type, time_zone, input_rows_count, scale); } std::unreachable(); @@ -236,22 +237,21 @@ private: template ColumnPtr execute( const TimeDataType &, const TimeColumnType & time_column_type, Int64 num_units, - const DataTypePtr & result_type, const DateLUTImpl & time_zone, UInt16 scale) const + const DataTypePtr & result_type, const DateLUTImpl & time_zone, size_t input_rows_count, UInt16 scale) const { using ResultColumnType = typename ResultDataType::ColumnType; using ResultFieldType = typename ResultDataType::FieldType; const auto & time_data = time_column_type.getData(); - size_t size = time_data.size(); auto result_col = result_type->createColumn(); auto * col_to = assert_cast(result_col.get()); auto & result_data = col_to->getData(); - result_data.resize(size); + result_data.resize(input_rows_count); Int64 scale_multiplier = DecimalUtils::scaleMultiplier(scale); - for (size_t i = 0; i != size; ++i) + for (size_t i = 0; i != input_rows_count; ++i) result_data[i] = static_cast(ToStartOfInterval::execute(time_data[i], num_units, time_zone, scale_multiplier)); return result_col; diff --git a/src/Functions/toTimezone.cpp b/src/Functions/toTimezone.cpp index a0d90351898..cbcdf9bc772 100644 --- a/src/Functions/toTimezone.cpp +++ b/src/Functions/toTimezone.cpp @@ -9,6 +9,7 @@ #include #include +#include namespace DB { @@ -88,7 +89,7 @@ public: size_t getNumberOfArguments() const override { return 2; } static FunctionOverloadResolverPtr create(ContextPtr context) { return std::make_unique(context); } explicit ToTimeZoneOverloadResolver(ContextPtr context) - : allow_nonconst_timezone_arguments(context->getSettings().allow_nonconst_timezone_arguments) + : allow_nonconst_timezone_arguments(context->getSettingsRef().allow_nonconst_timezone_arguments) {} DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override diff --git a/src/Functions/toTypeName.cpp b/src/Functions/toTypeName.cpp index 63bcc58d65e..abe02148d4a 100644 --- a/src/Functions/toTypeName.cpp +++ b/src/Functions/toTypeName.cpp @@ -1,6 +1,7 @@ #include #include #include +#include #include #include diff --git a/src/Functions/toValidUTF8.cpp b/src/Functions/toValidUTF8.cpp index 41d29d9c494..376732256b0 100644 --- a/src/Functions/toValidUTF8.cpp +++ b/src/Functions/toValidUTF8.cpp @@ -128,16 +128,16 @@ struct ToValidUTF8Impl const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - const size_t offsets_size = offsets.size(); /// It can be larger than that, but we believe it is unlikely to happen. res_data.resize(data.size()); - res_offsets.resize(offsets_size); + res_offsets.resize(input_rows_count); size_t prev_offset = 0; WriteBufferFromVector write_buffer(res_data); - for (size_t i = 0; i < offsets_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char * haystack_data = reinterpret_cast(&data[prev_offset]); const size_t haystack_size = offsets[i] - prev_offset - 1; @@ -149,7 +149,7 @@ struct ToValidUTF8Impl write_buffer.finalize(); } - [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + [[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Column of type FixedString is not supported by toValidUTF8 function"); } diff --git a/src/Functions/toYear.cpp b/src/Functions/toYear.cpp index 75479adb82c..0d2c8136337 100644 --- a/src/Functions/toYear.cpp +++ b/src/Functions/toYear.cpp @@ -14,7 +14,7 @@ REGISTER_FUNCTION(ToYear) factory.registerFunction(); /// MySQL compatibility alias. - factory.registerAlias("YEAR", "toYear", FunctionFactory::CaseInsensitive); + factory.registerAlias("YEAR", "toYear", FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/today.cpp b/src/Functions/today.cpp index 356660fa7b5..88eddc9b60e 100644 --- a/src/Functions/today.cpp +++ b/src/Functions/today.cpp @@ -84,8 +84,8 @@ public: REGISTER_FUNCTION(Today) { factory.registerFunction(); - factory.registerAlias("current_date", TodayOverloadResolver::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("curdate", TodayOverloadResolver::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("current_date", TodayOverloadResolver::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("curdate", TodayOverloadResolver::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/tokenExtractors.cpp b/src/Functions/tokenExtractors.cpp index e7dcb5cced3..1bbf313fbae 100644 --- a/src/Functions/tokenExtractors.cpp +++ b/src/Functions/tokenExtractors.cpp @@ -73,7 +73,7 @@ public: return std::make_shared(std::make_shared()); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { auto column_offsets = ColumnArray::ColumnOffsets::create(); @@ -90,9 +90,9 @@ public: auto input_column = arguments[0].column; if (const auto * column_string = checkAndGetColumn(input_column.get())) - executeImpl(extractor, *column_string, *result_column_string, *column_offsets); + executeImpl(extractor, *column_string, *result_column_string, *column_offsets, input_rows_count); else if (const auto * column_fixed_string = checkAndGetColumn(input_column.get())) - executeImpl(extractor, *column_fixed_string, *result_column_string, *column_offsets); + executeImpl(extractor, *column_fixed_string, *result_column_string, *column_offsets, input_rows_count); return ColumnArray::create(std::move(result_column_string), std::move(column_offsets)); } @@ -105,9 +105,9 @@ public: auto input_column = arguments[0].column; if (const auto * column_string = checkAndGetColumn(input_column.get())) - executeImpl(extractor, *column_string, *result_column_string, *column_offsets); + executeImpl(extractor, *column_string, *result_column_string, *column_offsets, input_rows_count); else if (const auto * column_fixed_string = checkAndGetColumn(input_column.get())) - executeImpl(extractor, *column_fixed_string, *result_column_string, *column_offsets); + executeImpl(extractor, *column_fixed_string, *result_column_string, *column_offsets, input_rows_count); return ColumnArray::create(std::move(result_column_string), std::move(column_offsets)); } @@ -120,15 +120,15 @@ private: const ExtractorType & extractor, StringColumnType & input_data_column, ResultStringColumnType & result_data_column, - ColumnArray::ColumnOffsets & offsets_column) const + ColumnArray::ColumnOffsets & offsets_column, + size_t input_rows_count) const { size_t current_tokens_size = 0; auto & offsets_data = offsets_column.getData(); - size_t column_size = input_data_column.size(); - offsets_data.resize(column_size); + offsets_data.resize(input_rows_count); - for (size_t i = 0; i < column_size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { auto data = input_data_column.getDataAt(i); diff --git a/src/Functions/transform.cpp b/src/Functions/transform.cpp index 0dbc9946710..0dfc9197845 100644 --- a/src/Functions/transform.cpp +++ b/src/Functions/transform.cpp @@ -138,8 +138,7 @@ namespace } } - ColumnPtr executeImpl( - const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override { std::call_once(once, [&] { initialize(arguments, result_type); }); @@ -174,30 +173,30 @@ namespace } else if (cache.table_num_to_idx) { - if (!executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted) - && !executeNum>(in, *column_result, default_non_const, *in_casted)) + if (!executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count) + && !executeNum>(in, *column_result, default_non_const, *in_casted, input_rows_count)) { throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal column {} of first argument of function {}", in->getName(), getName()); } } else if (cache.table_string_to_idx) { - if (!executeString(in, *column_result, default_non_const, *in_casted)) - executeContiguous(in, *column_result, default_non_const, *in_casted); + if (!executeString(in, *column_result, default_non_const, *in_casted, input_rows_count)) + executeContiguous(in, *column_result, default_non_const, *in_casted, input_rows_count); } else if (cache.table_anything_to_idx) { - executeAnything(in, *column_result, default_non_const, *in_casted); + executeAnything(in, *column_result, default_non_const, *in_casted, input_rows_count); } else throw Exception(ErrorCodes::LOGICAL_ERROR, "State of the function `transform` is not initialized"); @@ -218,12 +217,11 @@ namespace return impl->execute(args, result_type, input_rows_count); } - void executeAnything(const IColumn * in, IColumn & column_result, const ColumnPtr default_non_const, const IColumn & in_casted) const + void executeAnything(const IColumn * in, IColumn & column_result, const ColumnPtr default_non_const, const IColumn & in_casted, size_t input_rows_count) const { - const size_t size = in->size(); const auto & table = *cache.table_anything_to_idx; - column_result.reserve(size); - for (size_t i = 0; i < size; ++i) + column_result.reserve(input_rows_count); + for (size_t i = 0; i < input_rows_count; ++i) { SipHash hash; in->updateHashWithValue(i, hash); @@ -240,12 +238,11 @@ namespace } } - void executeContiguous(const IColumn * in, IColumn & column_result, const ColumnPtr default_non_const, const IColumn & in_casted) const + void executeContiguous(const IColumn * in, IColumn & column_result, const ColumnPtr default_non_const, const IColumn & in_casted, size_t input_rows_count) const { - const size_t size = in->size(); const auto & table = *cache.table_string_to_idx; - column_result.reserve(size); - for (size_t i = 0; i < size; ++i) + column_result.reserve(input_rows_count); + for (size_t i = 0; i < input_rows_count; ++i) { const auto * it = table.find(in->getDataAt(i)); if (it) @@ -260,7 +257,7 @@ namespace } template - bool executeNum(const IColumn * in_untyped, IColumn & column_result, const ColumnPtr default_non_const, const IColumn & in_casted) const + bool executeNum(const IColumn * in_untyped, IColumn & column_result, const ColumnPtr default_non_const, const IColumn & in_casted, size_t input_rows_count) const { const auto * const in = checkAndGetColumn(in_untyped); if (!in) @@ -270,24 +267,23 @@ namespace if constexpr (std::is_same_v, T> || std::is_same_v, T>) in_scale = in->getScale(); - if (!executeNumToString(pod, column_result, default_non_const) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale) - && !executeNumToNum>(pod, column_result, default_non_const, in_scale)) + if (!executeNumToString(pod, column_result, default_non_const, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count) + && !executeNumToNum>(pod, column_result, default_non_const, in_scale, input_rows_count)) { - const size_t size = pod.size(); const auto & table = *cache.table_num_to_idx; - column_result.reserve(size); - for (size_t i = 0; i < size; ++i) + column_result.reserve(input_rows_count); + for (size_t i = 0; i < input_rows_count; ++i) { const auto * it = table.find(bit_cast(pod[i])); if (it) @@ -304,14 +300,13 @@ namespace } template - bool executeNumToString(const PaddedPODArray & pod, IColumn & column_result, const ColumnPtr default_non_const) const + bool executeNumToString(const PaddedPODArray & pod, IColumn & column_result, const ColumnPtr default_non_const, size_t input_rows_count) const { auto * out = typeid_cast(&column_result); if (!out) return false; auto & out_offs = out->getOffsets(); - const size_t size = pod.size(); - out_offs.resize(size); + out_offs.resize(input_rows_count); auto & out_chars = out->getChars(); const auto * to_col = assert_cast(cache.to_column.get()); @@ -326,14 +321,14 @@ namespace const auto & def_offs = def->getOffsets(); const auto * def_data = def_chars.data(); auto def_size = def_offs[0]; - executeNumToStringHelper(table, pod, out_chars, out_offs, to_chars, to_offs, def_data, def_size, size); + executeNumToStringHelper(table, pod, out_chars, out_offs, to_chars, to_offs, def_data, def_size, input_rows_count); } else { const auto * def = assert_cast(default_non_const.get()); const auto & def_chars = def->getChars(); const auto & def_offs = def->getOffsets(); - executeNumToStringHelper(table, pod, out_chars, out_offs, to_chars, to_offs, def_chars, def_offs, size); + executeNumToStringHelper(table, pod, out_chars, out_offs, to_chars, to_offs, def_chars, def_offs, input_rows_count); } return true; } @@ -348,10 +343,10 @@ namespace const ColumnString::Offsets & to_offsets, const DefData & def_data, const DefOffs & def_offsets, - const size_t size) const + size_t input_rows_count) const { size_t out_cur_off = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char8_t * to = nullptr; size_t to_size = 0; @@ -383,14 +378,13 @@ namespace template bool executeNumToNum( - const PaddedPODArray & pod, IColumn & column_result, const ColumnPtr default_non_const, const UInt32 in_scale) const + const PaddedPODArray & pod, IColumn & column_result, ColumnPtr default_non_const, UInt32 in_scale, size_t input_rows_count) const { auto * out = typeid_cast(&column_result); if (!out) return false; auto & out_pod = out->getData(); - const size_t size = pod.size(); - out_pod.resize(size); + out_pod.resize(input_rows_count); UInt32 out_scale = 0; if constexpr (std::is_same_v, T> || std::is_same_v, T>) out_scale = out->getScale(); @@ -400,15 +394,15 @@ namespace if (cache.default_column) { const auto const_def = assert_cast(cache.default_column.get())->getData()[0]; - executeNumToNumHelper(table, pod, out_pod, to_pod, const_def, size, out_scale, out_scale); + executeNumToNumHelper(table, pod, out_pod, to_pod, const_def, input_rows_count, out_scale, out_scale); } else if (default_non_const) { const auto & nconst_def = assert_cast(default_non_const.get())->getData(); - executeNumToNumHelper(table, pod, out_pod, to_pod, nconst_def, size, out_scale, out_scale); + executeNumToNumHelper(table, pod, out_pod, to_pod, nconst_def, input_rows_count, out_scale, out_scale); } else - executeNumToNumHelper(table, pod, out_pod, to_pod, pod, size, out_scale, in_scale); + executeNumToNumHelper(table, pod, out_pod, to_pod, pod, input_rows_count, out_scale, in_scale); return true; } @@ -419,11 +413,11 @@ namespace PaddedPODArray & out_pod, const PaddedPODArray & to_pod, const Def & def, - const size_t size, - const UInt32 out_scale, - const UInt32 def_scale) const + size_t input_rows_count, + UInt32 out_scale, + UInt32 def_scale) const { - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const auto * it = table.find(bit_cast(pod[i])); if (it) @@ -451,7 +445,7 @@ namespace } } - bool executeString(const IColumn * in_untyped, IColumn & column_result, const ColumnPtr default_non_const, const IColumn & in_casted) const + bool executeString(const IColumn * in_untyped, IColumn & column_result, const ColumnPtr default_non_const, const IColumn & in_casted, size_t input_rows_count) const { const auto * const in = checkAndGetColumn(in_untyped); if (!in) @@ -459,19 +453,19 @@ namespace const auto & data = in->getChars(); const auto & offsets = in->getOffsets(); - if (!executeStringToString(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const) - && !executeStringToNum>(data, offsets, column_result, default_non_const)) + if (!executeStringToString(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count) + && !executeStringToNum>(data, offsets, column_result, default_non_const, input_rows_count)) { const size_t size = offsets.size(); const auto & table = *cache.table_string_to_idx; @@ -498,14 +492,14 @@ namespace const ColumnString::Chars & data, const ColumnString::Offsets & offsets, IColumn & column_result, - const ColumnPtr default_non_const) const + const ColumnPtr default_non_const, + size_t input_rows_count) const { auto * out = typeid_cast(&column_result); if (!out) return false; auto & out_offs = out->getOffsets(); - const size_t size = offsets.size(); - out_offs.resize(size); + out_offs.resize(input_rows_count); auto & out_chars = out->getChars(); const auto * to_col = assert_cast(cache.to_column.get()); @@ -520,18 +514,18 @@ namespace const auto & def_offs = def->getOffsets(); const auto * def_data = def_chars.data(); auto def_size = def_offs[0]; - executeStringToStringHelper(table, data, offsets, out_chars, out_offs, to_chars, to_offs, def_data, def_size, size); + executeStringToStringHelper(table, data, offsets, out_chars, out_offs, to_chars, to_offs, def_data, def_size, input_rows_count); } else if (default_non_const) { const auto * def = assert_cast(default_non_const.get()); const auto & def_chars = def->getChars(); const auto & def_offs = def->getOffsets(); - executeStringToStringHelper(table, data, offsets, out_chars, out_offs, to_chars, to_offs, def_chars, def_offs, size); + executeStringToStringHelper(table, data, offsets, out_chars, out_offs, to_chars, to_offs, def_chars, def_offs, input_rows_count); } else { - executeStringToStringHelper(table, data, offsets, out_chars, out_offs, to_chars, to_offs, data, offsets, size); + executeStringToStringHelper(table, data, offsets, out_chars, out_offs, to_chars, to_offs, data, offsets, input_rows_count); } return true; } @@ -547,11 +541,11 @@ namespace const ColumnString::Offsets & to_offsets, const DefData & def_data, const DefOffs & def_offsets, - const size_t size) const + size_t input_rows_count) const { ColumnString::Offset current_offset = 0; size_t out_cur_off = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const char8_t * to = nullptr; size_t to_size = 0; @@ -588,26 +582,26 @@ namespace const ColumnString::Chars & data, const ColumnString::Offsets & offsets, IColumn & column_result, - const ColumnPtr default_non_const) const + const ColumnPtr default_non_const, + size_t input_rows_count) const { auto * out = typeid_cast(&column_result); if (!out) return false; auto & out_pod = out->getData(); - const size_t size = offsets.size(); - out_pod.resize(size); + out_pod.resize(input_rows_count); const auto & to_pod = assert_cast(cache.to_column.get())->getData(); const auto & table = *cache.table_string_to_idx; if (cache.default_column) { const auto const_def = assert_cast(cache.default_column.get())->getData()[0]; - executeStringToNumHelper(table, data, offsets, out_pod, to_pod, const_def, size); + executeStringToNumHelper(table, data, offsets, out_pod, to_pod, const_def, input_rows_count); } else { const auto & nconst_def = assert_cast(default_non_const.get())->getData(); - executeStringToNumHelper(table, data, offsets, out_pod, to_pod, nconst_def, size); + executeStringToNumHelper(table, data, offsets, out_pod, to_pod, nconst_def, input_rows_count); } return true; } @@ -620,10 +614,10 @@ namespace PaddedPODArray & out_pod, const PaddedPODArray & to_pod, const Def & def, - const size_t size) const + size_t input_rows_count) const { ColumnString::Offset current_offset = 0; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { const StringRef ref{&data[current_offset], offsets[i] - current_offset - 1}; current_offset = offsets[i]; diff --git a/src/Functions/translate.cpp b/src/Functions/translate.cpp index 2df08a5664e..366640d7d20 100644 --- a/src/Functions/translate.cpp +++ b/src/Functions/translate.cpp @@ -52,7 +52,8 @@ struct TranslateImpl const std::string & map_from, const std::string & map_to, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { Map map; fillMapWithValues(map, map_from, map_to); @@ -62,7 +63,7 @@ struct TranslateImpl UInt8 * dst = res_data.data(); - for (UInt64 i = 0; i < offsets.size(); ++i) + for (UInt64 i = 0; i < input_rows_count; ++i) { const UInt8 * src = data.data() + offsets[i - 1]; const UInt8 * src_end = data.data() + offsets[i] - 1; @@ -175,19 +176,20 @@ struct TranslateUTF8Impl const std::string & map_from, const std::string & map_to, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { MapASCII map_ascii; MapUTF8 map; fillMapWithValues(map_ascii, map, map_from, map_to); res_data.resize(data.size()); - res_offsets.resize(offsets.size()); + res_offsets.resize(input_rows_count); UInt8 * dst = res_data.data(); UInt64 data_size = 0; - for (UInt64 i = 0; i < offsets.size(); ++i) + for (UInt64 i = 0; i < input_rows_count; ++i) { const UInt8 * src = data.data() + offsets[i - 1]; const UInt8 * src_end = data.data() + offsets[i] - 1; @@ -311,7 +313,7 @@ public: } } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const ColumnPtr column_src = arguments[0].column; const ColumnPtr column_map_from = arguments[1].column; @@ -330,7 +332,7 @@ public: if (const ColumnString * col = checkAndGetColumn(column_src.get())) { auto col_res = ColumnString::create(); - Impl::vector(col->getChars(), col->getOffsets(), map_from, map_to, col_res->getChars(), col_res->getOffsets()); + Impl::vector(col->getChars(), col->getOffsets(), map_from, map_to, col_res->getChars(), col_res->getOffsets(), input_rows_count); return col_res; } else if (const ColumnFixedString * col_fixed = checkAndGetColumn(column_src.get())) diff --git a/src/Functions/trim.cpp b/src/Functions/trim.cpp index 1f0011b8e99..5703e871423 100644 --- a/src/Functions/trim.cpp +++ b/src/Functions/trim.cpp @@ -43,10 +43,10 @@ public: const ColumnString::Chars & data, const ColumnString::Offsets & offsets, ColumnString::Chars & res_data, - ColumnString::Offsets & res_offsets) + ColumnString::Offsets & res_offsets, + size_t input_rows_count) { - size_t size = offsets.size(); - res_offsets.resize_exact(size); + res_offsets.resize_exact(input_rows_count); res_data.reserve_exact(data.size()); size_t prev_offset = 0; @@ -55,7 +55,7 @@ public: const UInt8 * start; size_t length; - for (size_t i = 0; i < size; ++i) + for (size_t i = 0; i < input_rows_count; ++i) { execute(reinterpret_cast(&data[prev_offset]), offsets[i] - prev_offset - 1, start, length); @@ -69,7 +69,7 @@ public: } } - static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &) + static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &, size_t) { throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Functions trimLeft, trimRight and trimBoth cannot work with FixedString argument"); } diff --git a/src/Functions/tryBase64UrlDecode.cpp b/src/Functions/tryBase64URLDecode.cpp similarity index 69% rename from src/Functions/tryBase64UrlDecode.cpp rename to src/Functions/tryBase64URLDecode.cpp index b9aaf4f9273..b44bc7538ee 100644 --- a/src/Functions/tryBase64UrlDecode.cpp +++ b/src/Functions/tryBase64URLDecode.cpp @@ -5,16 +5,16 @@ namespace DB { -REGISTER_FUNCTION(TryBase64UrlDecode) +REGISTER_FUNCTION(TryBase64URLDecode) { - FunctionDocumentation::Description description = R"(Decodes an URL from base64, like base64UrlDecode but returns an empty string in case of an error.)"; - FunctionDocumentation::Syntax syntax = "tryBase64UrlDecode(encodedUrl)"; - FunctionDocumentation::Arguments arguments = {{"encodedUrl", "String column or constant. If the string is not a valid Base64-encoded value with URL-specific modifications, returns an empty string."}}; + FunctionDocumentation::Description description = R"(Decodes an URL from base64, like base64URLDecode but returns an empty string in case of an error.)"; + FunctionDocumentation::Syntax syntax = "tryBase64URLDecode(encodedUrl)"; + FunctionDocumentation::Arguments arguments = {{"encodedURL", "String column or constant. If the string is not a valid Base64-encoded value with URL-specific modifications, returns an empty string."}}; FunctionDocumentation::ReturnedValue returned_value = "A string containing the decoded value of the argument."; - FunctionDocumentation::Examples examples = {{"valid", "SELECT tryBase64UrlDecode('aHR0cHM6Ly9jbGlja2hvdXNlLmNvbQ')", "https://clickhouse.com"}, {"invalid", "SELECT tryBase64UrlDecode('aHR0cHM6Ly9jbGlja')", ""}}; + FunctionDocumentation::Examples examples = {{"valid", "SELECT tryBase64URLDecode('aHR0cHM6Ly9jbGlja2hvdXNlLmNvbQ')", "https://clickhouse.com"}, {"invalid", "SELECT tryBase64UrlDecode('aHR0cHM6Ly9jbGlja')", ""}}; FunctionDocumentation::Categories categories = {"String encoding"}; - factory.registerFunction>>({description, syntax, arguments, returned_value, examples, categories}); + factory.registerFunction>>({description, syntax, arguments, returned_value, examples, categories}); } } diff --git a/src/Functions/tuple.cpp b/src/Functions/tuple.cpp index c83195bc976..84772f79d08 100644 --- a/src/Functions/tuple.cpp +++ b/src/Functions/tuple.cpp @@ -1,11 +1,28 @@ #include +#include + namespace DB { +FunctionPtr FunctionTuple::create(DB::ContextPtr context) +{ + return std::make_shared(context->getSettingsRef().enable_named_columns_in_function_tuple); +} + REGISTER_FUNCTION(Tuple) { - factory.registerFunction(); + factory.registerFunction(FunctionDocumentation{ + .description = R"( +Returns a tuple by grouping input arguments. + +For columns C1, C2, ... with the types T1, T2, ..., it returns a named Tuple(C1 T1, C2 T2, ...) type tuple containing these columns if their names are unique and can be treated as unquoted identifiers, otherwise a Tuple(T1, T2, ...) is returned. There is no cost to execute the function. +Tuples are normally used as intermediate values for an argument of IN operators, or for creating a list of formal parameters of lambda functions. Tuples can’t be written to a table. + +The function implements the operator `(x, y, ...)`. +)", + .examples{{"typical", "SELECT tuple(1, 2)", "(1,2)"}}, + .categories{"Miscellaneous"}}); } } diff --git a/src/Functions/tuple.h b/src/Functions/tuple.h index 8b3e041f781..03ba7ec9ef2 100644 --- a/src/Functions/tuple.h +++ b/src/Functions/tuple.h @@ -6,20 +6,24 @@ #include #include #include +#include namespace DB { -/** tuple(x, y, ...) is a function that allows you to group several columns +/** tuple(x, y, ...) is a function that allows you to group several columns. * tupleElement(tuple, n) is a function that allows you to retrieve a column from tuple. */ class FunctionTuple : public IFunction { + bool enable_named_columns; + public: static constexpr auto name = "tuple"; - /// maybe_unused: false-positive - [[ maybe_unused ]] static FunctionPtr create(ContextPtr) { return std::make_shared(); } + static FunctionPtr create(ContextPtr context); + + explicit FunctionTuple(bool enable_named_columns_ = false) : enable_named_columns(enable_named_columns_) { } String getName() const override { return name; } @@ -38,9 +42,26 @@ public: bool useDefaultImplementationForConstants() const override { return true; } bool useDefaultImplementationForLowCardinalityColumns() const override { return false; } - DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override { - return std::make_shared(arguments); + if (arguments.empty()) + return std::make_shared(DataTypes{}); + + DataTypes types; + Names names; + NameSet name_set; + for (const auto & argument : arguments) + { + types.emplace_back(argument.type); + names.emplace_back(argument.name); + name_set.emplace(argument.name); + } + + if (enable_named_columns && name_set.size() == names.size() + && std::all_of(names.cbegin(), names.cend(), [](const auto & n) { return isUnquotedIdentifier(n); })) + return std::make_shared(types, names); + else + return std::make_shared(types); } ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override @@ -53,9 +74,9 @@ public: for (size_t i = 0; i < tuple_size; ++i) { /** If tuple is mixed of constant and not constant columns, - * convert all to non-constant columns, - * because many places in code expect all non-constant columns in non-constant tuple. - */ + * convert all to non-constant columns, + * because many places in code expect all non-constant columns in non-constant tuple. + */ tuple_columns[i] = arguments[i].column->convertToFullColumnIfConst(); } return ColumnTuple::create(tuple_columns); diff --git a/src/Functions/tupleNames.cpp b/src/Functions/tupleNames.cpp new file mode 100644 index 00000000000..e444478c224 --- /dev/null +++ b/src/Functions/tupleNames.cpp @@ -0,0 +1,118 @@ +#include +#include +#include +#include +#include +#include + +namespace DB +{ +namespace ErrorCodes +{ +extern const int ILLEGAL_TYPE_OF_ARGUMENT; +} + +namespace +{ + +/** Transform a named tuple into names, which is a constant array of strings. + */ +class ExecutableFunctionTupleNames : public IExecutableFunction +{ +public: + static constexpr auto name = "tupleNames"; + + explicit ExecutableFunctionTupleNames(Array name_fields_) : name_fields(std::move(name_fields_)) { } + + String getName() const override { return name; } + + ColumnPtr executeImpl(const ColumnsWithTypeAndName &, const DataTypePtr & result_type, size_t input_rows_count) const override + { + return result_type->createColumnConst(input_rows_count, name_fields); + } + +private: + Array name_fields; +}; + +class FunctionBaseTupleNames : public IFunctionBase +{ +public: + static constexpr auto name = "tupleNames"; + + explicit FunctionBaseTupleNames(DataTypePtr argument_type, DataTypePtr result_type_, Array name_fields_) + : argument_types({std::move(argument_type)}), result_type(std::move(result_type_)), name_fields(std::move(name_fields_)) + { + } + + String getName() const override { return name; } + + bool isSuitableForConstantFolding() const override { return true; } + + bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } + + const DataTypes & getArgumentTypes() const override { return argument_types; } + + const DataTypePtr & getResultType() const override { return result_type; } + + ExecutableFunctionPtr prepare(const ColumnsWithTypeAndName &) const override + { + return std::make_unique(name_fields); + } + +private: + DataTypes argument_types; + DataTypePtr result_type; + Array name_fields; +}; + +class TupleNamesOverloadResolver : public IFunctionOverloadResolver +{ +public: + static constexpr auto name = "tupleNames"; + + static FunctionOverloadResolverPtr create(ContextPtr) { return std::make_unique(); } + + String getName() const override { return name; } + + size_t getNumberOfArguments() const override { return 1; } + + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override + { + const DataTypeTuple * tuple = checkAndGetDataType(arguments[0].type.get()); + + if (!tuple) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "First argument for function {} must be a tuple", getName()); + + return std::make_shared(std::make_shared()); + } + + FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type) const override + { + const DataTypeTuple * tuple = checkAndGetDataType(arguments[0].type.get()); + + if (!tuple) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "First argument for function {} must be a tuple", getName()); + + DataTypes types = tuple->getElements(); + Array name_fields; + for (const auto & elem_name : tuple->getElementNames()) + name_fields.emplace_back(elem_name); + + return std::make_unique(arguments[0].type, result_type, std::move(name_fields)); + } +}; + +} + +REGISTER_FUNCTION(TupleNames) +{ + factory.registerFunction(FunctionDocumentation{ + .description = R"( +Converts a tuple into an array of column names. For a tuple in the form `Tuple(a T, b T, ...)`, it returns an array of strings representing the named columns of the tuple. If the tuple elements do not have explicit names, their indices will be used as the column names instead. +)", + .examples{{"typical", "SELECT tupleNames(tuple(1 as a, 2 as b))", "['a','b']"}}, + .categories{"Miscellaneous"}}); +} + +} diff --git a/src/Functions/tupleToNameValuePairs.cpp b/src/Functions/tupleToNameValuePairs.cpp index 998e0da4f0c..92734d3d1fc 100644 --- a/src/Functions/tupleToNameValuePairs.cpp +++ b/src/Functions/tupleToNameValuePairs.cpp @@ -99,16 +99,16 @@ public: return std::make_shared(item_data_type); } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t /*input_rows_count*/) const override + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { const IColumn * tuple_col = arguments[0].column.get(); const DataTypeTuple * tuple = checkAndGetDataType(arguments[0].type.get()); - const auto * tuple_col_concrete = assert_cast(tuple_col); + const auto * tuple_col_concrete = assert_cast(tuple_col); auto keys = ColumnString::create(); MutableColumnPtr values = tuple_col_concrete->getColumn(0).cloneEmpty(); auto offsets = ColumnVector::create(); - for (size_t row = 0; row < tuple_col_concrete->size(); ++row) + for (size_t row = 0; row < input_rows_count; ++row) { for (size_t col = 0; col < tuple_col_concrete->tupleSize(); ++col) { diff --git a/src/Functions/upper.cpp b/src/Functions/upper.cpp index 3e1c7b1d800..5af0f059e3f 100644 --- a/src/Functions/upper.cpp +++ b/src/Functions/upper.cpp @@ -18,8 +18,8 @@ using FunctionUpper = FunctionStringToString, NameUpper REGISTER_FUNCTION(Upper) { - factory.registerFunction({}, FunctionFactory::CaseInsensitive); - factory.registerAlias("ucase", FunctionUpper::name, FunctionFactory::CaseInsensitive); + factory.registerFunction({}, FunctionFactory::Case::Insensitive); + factory.registerAlias("ucase", FunctionUpper::name, FunctionFactory::Case::Insensitive); } } diff --git a/src/Functions/vectorFunctions.cpp b/src/Functions/vectorFunctions.cpp index 7a2598f5f4b..5e23493c86d 100644 --- a/src/Functions/vectorFunctions.cpp +++ b/src/Functions/vectorFunctions.cpp @@ -1576,9 +1576,9 @@ using TupleOrArrayFunctionCosineDistance = TupleOrArrayFunction(); - factory.registerAlias("vectorSum", FunctionTuplePlus::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("vectorSum", FunctionTuplePlus::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); - factory.registerAlias("vectorDifference", FunctionTupleMinus::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("vectorDifference", FunctionTupleMinus::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); factory.registerFunction(); @@ -1652,7 +1652,7 @@ If the types of the first interval (or the interval in the tuple) and the second factory.registerFunction(); factory.registerFunction(); - factory.registerAlias("scalarProduct", TupleOrArrayFunctionDotProduct::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("scalarProduct", TupleOrArrayFunctionDotProduct::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); @@ -1660,11 +1660,11 @@ If the types of the first interval (or the interval in the tuple) and the second factory.registerFunction(); factory.registerFunction(); - factory.registerAlias("normL1", TupleOrArrayFunctionL1Norm::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("normL2", TupleOrArrayFunctionL2Norm::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("normL2Squared", TupleOrArrayFunctionL2SquaredNorm::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("normLinf", TupleOrArrayFunctionLinfNorm::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("normLp", FunctionLpNorm::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("normL1", TupleOrArrayFunctionL1Norm::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("normL2", TupleOrArrayFunctionL2Norm::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("normL2Squared", TupleOrArrayFunctionL2SquaredNorm::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("normLinf", TupleOrArrayFunctionLinfNorm::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("normLp", FunctionLpNorm::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); @@ -1672,21 +1672,21 @@ If the types of the first interval (or the interval in the tuple) and the second factory.registerFunction(); factory.registerFunction(); - factory.registerAlias("distanceL1", FunctionL1Distance::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("distanceL2", FunctionL2Distance::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("distanceL2Squared", FunctionL2SquaredDistance::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("distanceLinf", FunctionLinfDistance::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("distanceLp", FunctionLpDistance::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("distanceL1", FunctionL1Distance::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("distanceL2", FunctionL2Distance::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("distanceL2Squared", FunctionL2SquaredDistance::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("distanceLinf", FunctionLinfDistance::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("distanceLp", FunctionLpDistance::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); factory.registerFunction(); factory.registerFunction(); factory.registerFunction(); - factory.registerAlias("normalizeL1", FunctionL1Normalize::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("normalizeL2", FunctionL2Normalize::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("normalizeLinf", FunctionLinfNormalize::name, FunctionFactory::CaseInsensitive); - factory.registerAlias("normalizeLp", FunctionLpNormalize::name, FunctionFactory::CaseInsensitive); + factory.registerAlias("normalizeL1", FunctionL1Normalize::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("normalizeL2", FunctionL2Normalize::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("normalizeLinf", FunctionLinfNormalize::name, FunctionFactory::Case::Insensitive); + factory.registerAlias("normalizeLp", FunctionLpNormalize::name, FunctionFactory::Case::Insensitive); factory.registerFunction(); } diff --git a/src/Functions/visibleWidth.cpp b/src/Functions/visibleWidth.cpp index 9a3edd9fbec..ebd4a1ff713 100644 --- a/src/Functions/visibleWidth.cpp +++ b/src/Functions/visibleWidth.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include diff --git a/src/Functions/widthBucket.cpp b/src/Functions/widthBucket.cpp index 62ed460ca9d..d007cc968f0 100644 --- a/src/Functions/widthBucket.cpp +++ b/src/Functions/widthBucket.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -15,6 +16,7 @@ #include #include #include +#include #include #include @@ -285,7 +287,7 @@ Result: .categories{"Mathematical"}, }); - factory.registerAlias("width_bucket", "widthBucket", FunctionFactory::CaseInsensitive); + factory.registerAlias("width_bucket", "widthBucket", FunctionFactory::Case::Insensitive); } } diff --git a/src/IO/Archives/hasRegisteredArchiveFileExtension.cpp b/src/IO/Archives/hasRegisteredArchiveFileExtension.cpp index 2a979f500f7..407977f1f13 100644 --- a/src/IO/Archives/hasRegisteredArchiveFileExtension.cpp +++ b/src/IO/Archives/hasRegisteredArchiveFileExtension.cpp @@ -1,5 +1,7 @@ #include +#include +#include namespace DB { diff --git a/src/IO/AzureBlobStorage/copyAzureBlobStorageFile.cpp b/src/IO/AzureBlobStorage/copyAzureBlobStorageFile.cpp index 8bd436f218c..c10a7cd017a 100644 --- a/src/IO/AzureBlobStorage/copyAzureBlobStorageFile.cpp +++ b/src/IO/AzureBlobStorage/copyAzureBlobStorageFile.cpp @@ -16,10 +16,12 @@ namespace ProfileEvents { extern const Event AzureCopyObject; - extern const Event AzureUploadPart; + extern const Event AzureStageBlock; + extern const Event AzureCommitBlockList; extern const Event DiskAzureCopyObject; - extern const Event DiskAzureUploadPart; + extern const Event DiskAzureStageBlock; + extern const Event DiskAzureCommitBlockList; } @@ -45,9 +47,9 @@ namespace size_t total_size_, const String & dest_container_for_logging_, const String & dest_blob_, - std::shared_ptr settings_, + std::shared_ptr settings_, ThreadPoolCallbackRunnerUnsafe schedule_, - const Poco::Logger * log_) + LoggerPtr log_) : create_read_buffer(create_read_buffer_) , client(client_) , offset (offset_) @@ -70,9 +72,9 @@ namespace size_t total_size; const String & dest_container_for_logging; const String & dest_blob; - std::shared_ptr settings; + std::shared_ptr settings; ThreadPoolCallbackRunnerUnsafe schedule; - const Poco::Logger * log; + const LoggerPtr log; size_t max_single_part_upload_size; struct UploadPartTask @@ -81,7 +83,6 @@ namespace size_t part_size; std::vector block_ids; bool is_finished = false; - std::exception_ptr exception; }; size_t normal_part_size; @@ -90,6 +91,7 @@ namespace std::list TSA_GUARDED_BY(bg_tasks_mutex) bg_tasks; int num_added_bg_tasks TSA_GUARDED_BY(bg_tasks_mutex) = 0; int num_finished_bg_tasks TSA_GUARDED_BY(bg_tasks_mutex) = 0; + std::exception_ptr bg_exception TSA_GUARDED_BY(bg_tasks_mutex); std::mutex bg_tasks_mutex; std::condition_variable bg_tasks_condvar; @@ -156,6 +158,10 @@ namespace void completeMultipartUpload() { auto block_blob_client = client->GetBlockBlobClient(dest_blob); + ProfileEvents::increment(ProfileEvents::AzureCommitBlockList); + if (client->GetClickhouseOptions().IsClientForDisk) + ProfileEvents::increment(ProfileEvents::DiskAzureCommitBlockList); + block_blob_client.CommitBlockList(block_ids); } @@ -180,7 +186,7 @@ namespace } catch (...) { - tryLogCurrentException(__PRETTY_FUNCTION__); + tryLogCurrentException(log, fmt::format("While performing multipart upload of blob {} in container {}", dest_blob, dest_container_for_logging)); waitForAllBackgroundTasks(); throw; } @@ -236,7 +242,12 @@ namespace } catch (...) { - task->exception = std::current_exception(); + std::lock_guard lock(bg_tasks_mutex); + if (!bg_exception) + { + tryLogCurrentException(log, "While writing part"); + bg_exception = std::current_exception(); /// The exception will be rethrown after all background tasks stop working. + } } task_finish_notify(); }, Priority{}); @@ -259,9 +270,9 @@ namespace void processUploadPartRequest(UploadPartTask & task) { - ProfileEvents::increment(ProfileEvents::AzureUploadPart); + ProfileEvents::increment(ProfileEvents::AzureStageBlock); if (client->GetClickhouseOptions().IsClientForDisk) - ProfileEvents::increment(ProfileEvents::DiskAzureUploadPart); + ProfileEvents::increment(ProfileEvents::DiskAzureStageBlock); auto block_blob_client = client->GetBlockBlobClient(dest_blob); auto read_buffer = std::make_unique(create_read_buffer(), task.part_offset, task.part_size); @@ -293,13 +304,13 @@ namespace /// Suppress warnings because bg_tasks_mutex is actually hold, but tsa annotations do not understand std::unique_lock bg_tasks_condvar.wait(lock, [this]() {return TSA_SUPPRESS_WARNING_FOR_READ(num_added_bg_tasks) == TSA_SUPPRESS_WARNING_FOR_READ(num_finished_bg_tasks); }); - auto & tasks = TSA_SUPPRESS_WARNING_FOR_WRITE(bg_tasks); - for (auto & task : tasks) - { - if (task.exception) - std::rethrow_exception(task.exception); + auto exception = TSA_SUPPRESS_WARNING_FOR_READ(bg_exception); + if (exception) + std::rethrow_exception(exception); + + const auto & tasks = TSA_SUPPRESS_WARNING_FOR_READ(bg_tasks); + for (const auto & task : tasks) block_ids.insert(block_ids.end(),task.block_ids.begin(), task.block_ids.end()); - } } }; } @@ -312,10 +323,11 @@ void copyDataToAzureBlobStorageFile( std::shared_ptr dest_client, const String & dest_container_for_logging, const String & dest_blob, - std::shared_ptr settings, + std::shared_ptr settings, ThreadPoolCallbackRunnerUnsafe schedule) { - UploadHelper helper{create_read_buffer, dest_client, offset, size, dest_container_for_logging, dest_blob, settings, schedule, &Poco::Logger::get("copyDataToAzureBlobStorageFile")}; + auto log = getLogger("copyDataToAzureBlobStorageFile"); + UploadHelper helper{create_read_buffer, dest_client, offset, size, dest_container_for_logging, dest_blob, settings, schedule, log}; helper.performCopy(); } @@ -329,14 +341,15 @@ void copyAzureBlobStorageFile( size_t size, const String & dest_container_for_logging, const String & dest_blob, - std::shared_ptr settings, + std::shared_ptr settings, const ReadSettings & read_settings, ThreadPoolCallbackRunnerUnsafe schedule) { + auto log = getLogger("copyAzureBlobStorageFile"); if (settings->use_native_copy) { - LOG_TRACE(getLogger("copyAzureBlobStorageFile"), "Copying Blob: {} from Container: {} using native copy", src_container_for_logging, src_blob); + LOG_TRACE(log, "Copying Blob: {} from Container: {} using native copy", src_container_for_logging, src_blob); ProfileEvents::increment(ProfileEvents::AzureCopyObject); if (dest_client->GetClickhouseOptions().IsClientForDisk) ProfileEvents::increment(ProfileEvents::DiskAzureCopyObject); @@ -347,7 +360,7 @@ void copyAzureBlobStorageFile( if (size < settings->max_single_part_copy_size) { - LOG_TRACE(getLogger("copyAzureBlobStorageFile"), "Copy blob sync {} -> {}", src_blob, dest_blob); + LOG_TRACE(log, "Copy blob sync {} -> {}", src_blob, dest_blob); block_blob_client_dest.CopyFromUri(source_uri); } else @@ -363,7 +376,7 @@ void copyAzureBlobStorageFile( if (copy_status.HasValue() && copy_status.Value() == Azure::Storage::Blobs::Models::CopyStatus::Success) { - LOG_TRACE(getLogger("copyAzureBlobStorageFile"), "Copy of {} to {} finished", properties_model.CopySource.Value(), dest_blob); + LOG_TRACE(log, "Copy of {} to {} finished", properties_model.CopySource.Value(), dest_blob); } else { @@ -377,14 +390,14 @@ void copyAzureBlobStorageFile( } else { - LOG_TRACE(&Poco::Logger::get("copyAzureBlobStorageFile"), "Reading from Container: {}, Blob: {}", src_container_for_logging, src_blob); + LOG_TRACE(log, "Reading from Container: {}, Blob: {}", src_container_for_logging, src_blob); auto create_read_buffer = [&] { return std::make_unique( src_client, src_blob, read_settings, settings->max_single_read_retries, settings->max_single_download_retries); }; - UploadHelper helper{create_read_buffer, dest_client, offset, size, dest_container_for_logging, dest_blob, settings, schedule, &Poco::Logger::get("copyAzureBlobStorageFile")}; + UploadHelper helper{create_read_buffer, dest_client, offset, size, dest_container_for_logging, dest_blob, settings, schedule, log}; helper.performCopy(); } } diff --git a/src/IO/AzureBlobStorage/copyAzureBlobStorageFile.h b/src/IO/AzureBlobStorage/copyAzureBlobStorageFile.h index 9c20ee4cff0..c8e48fcd372 100644 --- a/src/IO/AzureBlobStorage/copyAzureBlobStorageFile.h +++ b/src/IO/AzureBlobStorage/copyAzureBlobStorageFile.h @@ -28,7 +28,7 @@ void copyAzureBlobStorageFile( size_t src_size, const String & dest_container_for_logging, const String & dest_blob, - std::shared_ptr settings, + std::shared_ptr settings, const ReadSettings & read_settings, ThreadPoolCallbackRunnerUnsafe schedule_ = {}); @@ -45,7 +45,7 @@ void copyDataToAzureBlobStorageFile( std::shared_ptr client, const String & dest_container_for_logging, const String & dest_blob, - std::shared_ptr settings, + std::shared_ptr settings, ThreadPoolCallbackRunnerUnsafe schedule_ = {}); } diff --git a/src/IO/BufferWithOwnMemory.h b/src/IO/BufferWithOwnMemory.h index 5c9a69893df..da38bccdea1 100644 --- a/src/IO/BufferWithOwnMemory.h +++ b/src/IO/BufferWithOwnMemory.h @@ -4,12 +4,15 @@ #include #include +#include #include #include #include +#include "config.h" + namespace ProfileEvents { @@ -41,10 +44,13 @@ struct Memory : boost::noncopyable, Allocator char * m_data = nullptr; size_t alignment = 0; + [[maybe_unused]] bool allow_gwp_asan_force_sample{false}; + Memory() = default; /// If alignment != 0, then allocate memory aligned to specified value. - explicit Memory(size_t size_, size_t alignment_ = 0) : alignment(alignment_) + explicit Memory(size_t size_, size_t alignment_ = 0, bool allow_gwp_asan_force_sample_ = false) + : alignment(alignment_), allow_gwp_asan_force_sample(allow_gwp_asan_force_sample_) { alloc(size_); } @@ -127,6 +133,11 @@ private: ProfileEvents::increment(ProfileEvents::IOBufferAllocs); ProfileEvents::increment(ProfileEvents::IOBufferAllocBytes, new_capacity); +#if USE_GWP_ASAN + if (unlikely(allow_gwp_asan_force_sample && GWPAsan::shouldForceSample())) + gwp_asan::getThreadLocals()->NextSampleCounter = 1; +#endif + m_data = static_cast(Allocator::alloc(new_capacity, alignment)); m_capacity = new_capacity; m_size = new_size; @@ -154,7 +165,7 @@ protected: public: /// If non-nullptr 'existing_memory' is passed, then buffer will not create its own memory and will use existing_memory without ownership. explicit BufferWithOwnMemory(size_t size = DBMS_DEFAULT_BUFFER_SIZE, char * existing_memory = nullptr, size_t alignment = 0) - : Base(nullptr, 0), memory(existing_memory ? 0 : size, alignment) + : Base(nullptr, 0), memory(existing_memory ? 0 : size, alignment, /*allow_gwp_asan_force_sample_=*/true) { Base::set(existing_memory ? existing_memory : memory.data(), size); Base::padded = !existing_memory; diff --git a/src/IO/ConnectionTimeouts.cpp b/src/IO/ConnectionTimeouts.cpp index da6214ae477..662a2c067cf 100644 --- a/src/IO/ConnectionTimeouts.cpp +++ b/src/IO/ConnectionTimeouts.cpp @@ -1,7 +1,10 @@ +#include +#include #include -#include #include +#include + namespace DB { diff --git a/src/IO/ConnectionTimeouts.h b/src/IO/ConnectionTimeouts.h index b86ec44d21c..b84db46d656 100644 --- a/src/IO/ConnectionTimeouts.h +++ b/src/IO/ConnectionTimeouts.h @@ -1,7 +1,6 @@ #pragma once #include -#include #include #include @@ -10,6 +9,7 @@ namespace DB { +struct ServerSettings; struct Settings; #define APPLY_FOR_ALL_CONNECTION_TIMEOUT_MEMBERS(M) \ diff --git a/src/IO/ReadBufferFromS3.cpp b/src/IO/ReadBufferFromS3.cpp index 9e001232e65..bf97e959ee0 100644 --- a/src/IO/ReadBufferFromS3.cpp +++ b/src/IO/ReadBufferFromS3.cpp @@ -25,8 +25,6 @@ namespace ProfileEvents extern const Event ReadBufferFromS3InitMicroseconds; extern const Event ReadBufferFromS3Bytes; extern const Event ReadBufferFromS3RequestsErrors; - extern const Event ReadBufferFromS3ResetSessions; - extern const Event ReadBufferFromS3PreservedSessions; extern const Event ReadBufferSeekCancelConnection; extern const Event S3GetObject; extern const Event DiskS3GetObject; diff --git a/src/IO/ReadWriteBufferFromHTTP.cpp b/src/IO/ReadWriteBufferFromHTTP.cpp index 303ffb744b5..b753e66da48 100644 --- a/src/IO/ReadWriteBufferFromHTTP.cpp +++ b/src/IO/ReadWriteBufferFromHTTP.cpp @@ -713,8 +713,12 @@ ReadWriteBufferFromHTTP::HTTPFileInfo ReadWriteBufferFromHTTP::getFileInfo() /// fall back to slow whole-file reads when HEAD is actually supported; that sounds /// like a nightmare to debug.) if (e.getHTTPStatus() >= 400 && e.getHTTPStatus() <= 499 && - e.getHTTPStatus() != Poco::Net::HTTPResponse::HTTP_TOO_MANY_REQUESTS) + e.getHTTPStatus() != Poco::Net::HTTPResponse::HTTP_TOO_MANY_REQUESTS && + e.getHTTPStatus() != Poco::Net::HTTPResponse::HTTP_REQUEST_TIMEOUT && + e.getHTTPStatus() != Poco::Net::HTTPResponse::HTTP_MISDIRECTED_REQUEST) + { return HTTPFileInfo{}; + } throw; } diff --git a/src/IO/S3/BlobStorageLogWriter.cpp b/src/IO/S3/BlobStorageLogWriter.cpp index c2f0cb86928..d3b97771790 100644 --- a/src/IO/S3/BlobStorageLogWriter.cpp +++ b/src/IO/S3/BlobStorageLogWriter.cpp @@ -56,7 +56,6 @@ void BlobStorageLogWriter::addEvent( BlobStorageLogWriterPtr BlobStorageLogWriter::create(const String & disk_name) { -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD /// Keeper standalone build doesn't have a context if (auto blob_storage_log = Context::getGlobalContextInstance()->getBlobStorageLog()) { auto log_writer = std::make_shared(std::move(blob_storage_log)); @@ -67,7 +66,6 @@ BlobStorageLogWriterPtr BlobStorageLogWriter::create(const String & disk_name) return log_writer; } -#endif return {}; } diff --git a/src/IO/S3/Client.cpp b/src/IO/S3/Client.cpp index 55441cfb86b..6e659c82def 100644 --- a/src/IO/S3/Client.cpp +++ b/src/IO/S3/Client.cpp @@ -23,10 +23,11 @@ #include #include - #include #include +#include + #include @@ -824,6 +825,17 @@ void Client::updateURIForBucket(const std::string & bucket, S3::URI new_uri) con cache->uri_for_bucket_cache.emplace(bucket, std::move(new_uri)); } +ClientCache::ClientCache(const ClientCache & other) +{ + { + std::lock_guard lock(other.region_cache_mutex); + region_for_bucket_cache = other.region_for_bucket_cache; + } + { + std::lock_guard lock(other.uri_cache_mutex); + uri_for_bucket_cache = other.uri_for_bucket_cache; + } +} void ClientCache::clearCache() { diff --git a/src/IO/S3/Client.h b/src/IO/S3/Client.h index 2fd0a9cd2d1..de554e75d67 100644 --- a/src/IO/S3/Client.h +++ b/src/IO/S3/Client.h @@ -54,10 +54,7 @@ struct ClientCache { ClientCache() = default; - ClientCache(const ClientCache & other) - : region_for_bucket_cache(other.region_for_bucket_cache) - , uri_for_bucket_cache(other.uri_for_bucket_cache) - {} + ClientCache(const ClientCache & other); ClientCache(ClientCache && other) = delete; @@ -66,11 +63,11 @@ struct ClientCache void clearCache(); - std::mutex region_cache_mutex; - std::unordered_map region_for_bucket_cache; + mutable std::mutex region_cache_mutex; + std::unordered_map region_for_bucket_cache TSA_GUARDED_BY(region_cache_mutex); - std::mutex uri_cache_mutex; - std::unordered_map uri_for_bucket_cache; + mutable std::mutex uri_cache_mutex; + std::unordered_map uri_for_bucket_cache TSA_GUARDED_BY(uri_cache_mutex); }; class ClientCacheRegistry @@ -89,7 +86,7 @@ private: ClientCacheRegistry() = default; std::mutex clients_mutex; - std::unordered_map> client_caches; + std::unordered_map> client_caches TSA_GUARDED_BY(clients_mutex); }; bool isS3ExpressEndpoint(const std::string & endpoint); @@ -162,7 +159,7 @@ public: class RetryStrategy : public Aws::Client::RetryStrategy { public: - explicit RetryStrategy(uint32_t maxRetries_ = 10, uint32_t scaleFactor_ = 25, uint32_t maxDelayMs_ = 90000); + explicit RetryStrategy(uint32_t maxRetries_ = 10, uint32_t scaleFactor_ = 25, uint32_t maxDelayMs_ = 5000); /// NOLINTNEXTLINE(google-runtime-int) bool ShouldRetry(const Aws::Client::AWSError& error, long attemptedRetries) const override; diff --git a/src/IO/S3/Credentials.cpp b/src/IO/S3/Credentials.cpp index fa9d018eaa6..dfb7727fca4 100644 --- a/src/IO/S3/Credentials.cpp +++ b/src/IO/S3/Credentials.cpp @@ -9,6 +9,21 @@ namespace ErrorCodes extern const int UNSUPPORTED_METHOD; } +namespace S3 +{ + std::string tryGetRunningAvailabilityZone() + { + try + { + return getRunningAvailabilityZone(); + } + catch (...) + { + tryLogCurrentException("tryGetRunningAvailabilityZone"); + return ""; + } + } +} } #if USE_AWS_S3 diff --git a/src/IO/S3/Credentials.h b/src/IO/S3/Credentials.h index b8698d9b302..95297ab0538 100644 --- a/src/IO/S3/Credentials.h +++ b/src/IO/S3/Credentials.h @@ -24,6 +24,7 @@ static inline constexpr char GCP_METADATA_SERVICE_ENDPOINT[] = "http://metadata. /// getRunningAvailabilityZone returns the availability zone of the underlying compute resources where the current process runs. std::string getRunningAvailabilityZone(); +std::string tryGetRunningAvailabilityZone(); class AWSEC2MetadataClient : public Aws::Internal::AWSHttpResourceClient { @@ -195,6 +196,7 @@ namespace DB namespace S3 { std::string getRunningAvailabilityZone(); +std::string tryGetRunningAvailabilityZone(); } } diff --git a/src/IO/S3/copyS3File.cpp b/src/IO/S3/copyS3File.cpp index bb654c3f5c9..361b5a707fd 100644 --- a/src/IO/S3/copyS3File.cpp +++ b/src/IO/S3/copyS3File.cpp @@ -98,7 +98,6 @@ namespace size_t part_size; String tag; bool is_finished = false; - std::exception_ptr exception; }; size_t num_parts; @@ -111,6 +110,7 @@ namespace size_t num_added_bg_tasks TSA_GUARDED_BY(bg_tasks_mutex) = 0; size_t num_finished_bg_tasks TSA_GUARDED_BY(bg_tasks_mutex) = 0; size_t num_finished_parts TSA_GUARDED_BY(bg_tasks_mutex) = 0; + std::exception_ptr bg_exception TSA_GUARDED_BY(bg_tasks_mutex); std::mutex bg_tasks_mutex; std::condition_variable bg_tasks_condvar; @@ -273,7 +273,7 @@ namespace } catch (...) { - tryLogCurrentException(__PRETTY_FUNCTION__); + tryLogCurrentException(log, fmt::format("While performing multipart upload of {}", dest_key)); // Multipart upload failed because it wasn't possible to schedule all the tasks. // To avoid execution of already scheduled tasks we abort MultipartUpload. abortMultipartUpload(); @@ -290,9 +290,9 @@ namespace if (!total_size) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chosen multipart upload for an empty file. This must not happen"); - auto max_part_number = request_settings.max_part_number; - auto min_upload_part_size = request_settings.min_upload_part_size; - auto max_upload_part_size = request_settings.max_upload_part_size; + UInt64 max_part_number = request_settings.max_part_number; + UInt64 min_upload_part_size = request_settings.min_upload_part_size; + UInt64 max_upload_part_size = request_settings.max_upload_part_size; if (!max_part_number) throw Exception(ErrorCodes::INVALID_CONFIG_PARAMETER, "max_part_number must not be 0"); @@ -385,7 +385,12 @@ namespace } catch (...) { - task->exception = std::current_exception(); + std::lock_guard lock(bg_tasks_mutex); + if (!bg_exception) + { + tryLogCurrentException(log, fmt::format("While writing part #{}", task->part_number)); + bg_exception = std::current_exception(); /// The exception will be rethrown after all background tasks stop working. + } } task_finish_notify(); }, Priority{}); @@ -435,22 +440,21 @@ namespace /// Suppress warnings because bg_tasks_mutex is actually hold, but tsa annotations do not understand std::unique_lock bg_tasks_condvar.wait(lock, [this]() {return TSA_SUPPRESS_WARNING_FOR_READ(num_added_bg_tasks) == TSA_SUPPRESS_WARNING_FOR_READ(num_finished_bg_tasks); }); - auto & tasks = TSA_SUPPRESS_WARNING_FOR_WRITE(bg_tasks); - for (auto & task : tasks) + auto exception = TSA_SUPPRESS_WARNING_FOR_READ(bg_exception); + if (exception) { - if (task.exception) - { - /// abortMultipartUpload() might be called already, see processUploadPartRequest(). - /// However if there were concurrent uploads at that time, those part uploads might or might not succeed. - /// As a result, it might be necessary to abort a given multipart upload multiple times in order to completely free - /// all storage consumed by all parts. - abortMultipartUpload(); + /// abortMultipartUpload() might be called already, see processUploadPartRequest(). + /// However if there were concurrent uploads at that time, those part uploads might or might not succeed. + /// As a result, it might be necessary to abort a given multipart upload multiple times in order to completely free + /// all storage consumed by all parts. + abortMultipartUpload(); - std::rethrow_exception(task.exception); - } - - part_tags.push_back(task.tag); + std::rethrow_exception(exception); } + + const auto & tasks = TSA_SUPPRESS_WARNING_FOR_READ(bg_tasks); + for (const auto & task : tasks) + part_tags.push_back(task.tag); } }; diff --git a/src/IO/S3Common.cpp b/src/IO/S3Common.cpp index 490bf8c2d0c..a794cdbcf05 100644 --- a/src/IO/S3Common.cpp +++ b/src/IO/S3Common.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include #include "config.h" diff --git a/src/IO/S3Common.h b/src/IO/S3Common.h index 2dca08871d3..fcfd0cfffae 100644 --- a/src/IO/S3Common.h +++ b/src/IO/S3Common.h @@ -8,7 +8,7 @@ #include #include #include -#include +#include #include #include #include @@ -30,6 +30,8 @@ namespace ErrorCodes extern const int S3_ERROR; } +struct Settings; + class S3Exception : public Exception { public: diff --git a/src/IO/S3Settings.cpp b/src/IO/S3Settings.cpp index a5a50c873cb..f99e8c0105b 100644 --- a/src/IO/S3Settings.cpp +++ b/src/IO/S3Settings.cpp @@ -1,9 +1,11 @@ #include +#include #include -#include #include +#include + namespace DB { diff --git a/src/IO/examples/CMakeLists.txt b/src/IO/examples/CMakeLists.txt index 12b85c483a1..bfd171a3d00 100644 --- a/src/IO/examples/CMakeLists.txt +++ b/src/IO/examples/CMakeLists.txt @@ -1,77 +1,77 @@ clickhouse_add_executable (read_buffer read_buffer.cpp) -target_link_libraries (read_buffer PRIVATE clickhouse_common_io) +target_link_libraries (read_buffer PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (read_buffer_perf read_buffer_perf.cpp) -target_link_libraries (read_buffer_perf PRIVATE clickhouse_common_io) +target_link_libraries (read_buffer_perf PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (read_float_perf read_float_perf.cpp) -target_link_libraries (read_float_perf PRIVATE clickhouse_common_io) +target_link_libraries (read_float_perf PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (write_buffer write_buffer.cpp) -target_link_libraries (write_buffer PRIVATE clickhouse_common_io) +target_link_libraries (write_buffer PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (write_buffer_perf write_buffer_perf.cpp) -target_link_libraries (write_buffer_perf PRIVATE clickhouse_common_io) +target_link_libraries (write_buffer_perf PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (valid_utf8_perf valid_utf8_perf.cpp) -target_link_libraries (valid_utf8_perf PRIVATE clickhouse_common_io) +target_link_libraries (valid_utf8_perf PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (valid_utf8 valid_utf8.cpp) -target_link_libraries (valid_utf8 PRIVATE clickhouse_common_io) +target_link_libraries (valid_utf8 PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (var_uint var_uint.cpp) -target_link_libraries (var_uint PRIVATE clickhouse_common_io) +target_link_libraries (var_uint PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (read_escaped_string read_escaped_string.cpp) -target_link_libraries (read_escaped_string PRIVATE clickhouse_common_io) +target_link_libraries (read_escaped_string PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (parse_int_perf parse_int_perf.cpp) -target_link_libraries (parse_int_perf PRIVATE clickhouse_common_io) +target_link_libraries (parse_int_perf PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (parse_int_perf2 parse_int_perf2.cpp) -target_link_libraries (parse_int_perf2 PRIVATE clickhouse_common_io) +target_link_libraries (parse_int_perf2 PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (read_write_int read_write_int.cpp) -target_link_libraries (read_write_int PRIVATE clickhouse_common_io) +target_link_libraries (read_write_int PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (o_direct_and_dirty_pages o_direct_and_dirty_pages.cpp) -target_link_libraries (o_direct_and_dirty_pages PRIVATE clickhouse_common_io) +target_link_libraries (o_direct_and_dirty_pages PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (io_operators io_operators.cpp) -target_link_libraries (io_operators PRIVATE clickhouse_common_io) +target_link_libraries (io_operators PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (write_int write_int.cpp) -target_link_libraries (write_int PRIVATE clickhouse_common_io) +target_link_libraries (write_int PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (zlib_buffers zlib_buffers.cpp) -target_link_libraries (zlib_buffers PRIVATE clickhouse_common_io) +target_link_libraries (zlib_buffers PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (lzma_buffers lzma_buffers.cpp) -target_link_libraries (lzma_buffers PRIVATE clickhouse_common_io) +target_link_libraries (lzma_buffers PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (limit_read_buffer limit_read_buffer.cpp) -target_link_libraries (limit_read_buffer PRIVATE clickhouse_common_io) +target_link_libraries (limit_read_buffer PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (limit_read_buffer2 limit_read_buffer2.cpp) -target_link_libraries (limit_read_buffer2 PRIVATE clickhouse_common_io) +target_link_libraries (limit_read_buffer2 PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (parse_date_time_best_effort parse_date_time_best_effort.cpp) -target_link_libraries (parse_date_time_best_effort PRIVATE clickhouse_common_io) +target_link_libraries (parse_date_time_best_effort PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (zlib_ng_bug zlib_ng_bug.cpp) -target_link_libraries (zlib_ng_bug PRIVATE ch_contrib::zlib) +target_link_libraries (zlib_ng_bug PRIVATE ch_contrib::zlib clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (dragonbox_test dragonbox_test.cpp) -target_link_libraries (dragonbox_test PRIVATE ch_contrib::dragonbox_to_chars) +target_link_libraries (dragonbox_test PRIVATE ch_contrib::dragonbox_to_chars clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (zstd_buffers zstd_buffers.cpp) -target_link_libraries (zstd_buffers PRIVATE clickhouse_common_io) +target_link_libraries (zstd_buffers PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (snappy_read_buffer snappy_read_buffer.cpp) -target_link_libraries (snappy_read_buffer PRIVATE clickhouse_common_io) +target_link_libraries (snappy_read_buffer PRIVATE clickhouse_common_io clickhouse_common_config) clickhouse_add_executable (hadoop_snappy_read_buffer hadoop_snappy_read_buffer.cpp) -target_link_libraries (hadoop_snappy_read_buffer PRIVATE clickhouse_common_io) +target_link_libraries (hadoop_snappy_read_buffer PRIVATE clickhouse_common_io clickhouse_common_config) if (TARGET ch_contrib::hdfs) clickhouse_add_executable (read_buffer_from_hdfs read_buffer_from_hdfs.cpp) diff --git a/src/IO/tests/gtest_archive_reader_and_writer.cpp b/src/IO/tests/gtest_archive_reader_and_writer.cpp index 898c7017e7d..06f8f53546b 100644 --- a/src/IO/tests/gtest_archive_reader_and_writer.cpp +++ b/src/IO/tests/gtest_archive_reader_and_writer.cpp @@ -492,48 +492,6 @@ TEST_P(ArchiveReaderAndWriterTest, ManyFilesOnDisk) } } -TEST_P(ArchiveReaderAndWriterTest, LargeFile) -{ - /// Make an archive. - std::string_view contents = "The contents of a.txt\n"; - int times = 10000000; - { - auto writer = createArchiveWriter(getPathToArchive()); - { - auto out = writer->writeFile("a.txt", times * contents.size()); - for (int i = 0; i < times; i++) - writeString(contents, *out); - out->finalize(); - } - writer->finalize(); - } - - /// Read the archive. - auto reader = createArchiveReader(getPathToArchive()); - - ASSERT_TRUE(reader->fileExists("a.txt")); - - auto file_info = reader->getFileInfo("a.txt"); - EXPECT_EQ(file_info.uncompressed_size, contents.size() * times); - EXPECT_GT(file_info.compressed_size, 0); - - { - auto in = reader->readFile("a.txt", /*throw_on_not_found=*/true); - for (int i = 0; i < times; i++) - ASSERT_TRUE(checkString(String(contents), *in)); - } - - { - /// Use an enumerator. - auto enumerator = reader->firstFile(); - ASSERT_NE(enumerator, nullptr); - EXPECT_EQ(enumerator->getFileName(), "a.txt"); - EXPECT_EQ(enumerator->getFileInfo().uncompressed_size, contents.size() * times); - EXPECT_GT(enumerator->getFileInfo().compressed_size, 0); - EXPECT_FALSE(enumerator->nextFile()); - } -} - TEST(TarArchiveReaderTest, FileExists) { String archive_path = "archive.tar"; diff --git a/src/IO/tests/gtest_memory_resize.cpp b/src/IO/tests/gtest_memory_resize.cpp index d760a948075..c3b34c352b2 100644 --- a/src/IO/tests/gtest_memory_resize.cpp +++ b/src/IO/tests/gtest_memory_resize.cpp @@ -134,7 +134,7 @@ TEST(MemoryResizeTest, SmallInitAndBigResizeOverflowWhenPadding) ASSERT_EQ(memory.m_capacity, 0x8000000000000000ULL - 1); ASSERT_EQ(memory.m_size, 0x8000000000000000ULL - PADDING_FOR_SIMD); -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD EXPECT_THROW_ERROR_CODE(memory.resize(0x8000000000000000ULL - (PADDING_FOR_SIMD - 1)), Exception, ErrorCodes::LOGICAL_ERROR); ASSERT_TRUE(memory.m_data); // state is intact after exception ASSERT_EQ(memory.m_capacity, 0x8000000000000000ULL - 1); @@ -158,7 +158,7 @@ TEST(MemoryResizeTest, SmallInitAndBigResizeOverflowWhenPadding) ASSERT_EQ(memory.m_capacity, PADDING_FOR_SIMD); ASSERT_EQ(memory.m_size, 1); -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD EXPECT_THROW_ERROR_CODE(memory.resize(0x8000000000000000ULL - (PADDING_FOR_SIMD - 1)), Exception, ErrorCodes::LOGICAL_ERROR); ASSERT_TRUE(memory.m_data); // state is intact after exception ASSERT_EQ(memory.m_capacity, PADDING_FOR_SIMD); @@ -197,7 +197,7 @@ TEST(MemoryResizeTest, BigInitAndSmallResizeOverflowWhenPadding) , ErrorCodes::ARGUMENT_OUT_OF_BOUND); } -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD { EXPECT_THROW_ERROR_CODE( { diff --git a/src/IO/tests/gtest_writebuffer_s3.cpp b/src/IO/tests/gtest_writebuffer_s3.cpp index 3c1af6538ad..b53a8b58023 100644 --- a/src/IO/tests/gtest_writebuffer_s3.cpp +++ b/src/IO/tests/gtest_writebuffer_s3.cpp @@ -917,8 +917,8 @@ TEST_P(SyncAsync, ExceptionOnUploadPart) { TEST_F(WBS3Test, PrefinalizeCalledMultipleTimes) { -#ifdef ABORT_ON_LOGICAL_ERROR - GTEST_SKIP() << "this test trigger LOGICAL_ERROR, runs only if ABORT_ON_LOGICAL_ERROR is not defined"; +#ifdef DEBUG_OR_SANITIZER_BUILD + GTEST_SKIP() << "this test trigger LOGICAL_ERROR, runs only if DEBUG_OR_SANITIZER_BUILD is not defined"; #else EXPECT_THROW({ try { diff --git a/src/Interpreters/Access/InterpreterCreateUserQuery.cpp b/src/Interpreters/Access/InterpreterCreateUserQuery.cpp index 32c51b745c7..855aa36b159 100644 --- a/src/Interpreters/Access/InterpreterCreateUserQuery.cpp +++ b/src/Interpreters/Access/InterpreterCreateUserQuery.cpp @@ -114,6 +114,34 @@ namespace else if (query.grantees) user.grantees = *query.grantees; } + + time_t getValidUntilFromAST(ASTPtr valid_until, ContextPtr context) + { + if (context) + valid_until = evaluateConstantExpressionAsLiteral(valid_until, context); + + const String valid_until_str = checkAndGetLiteralArgument(valid_until, "valid_until"); + + if (valid_until_str == "infinity") + return 0; + + time_t time = 0; + ReadBufferFromString in(valid_until_str); + + if (context) + { + const auto & time_zone = DateLUT::instance(""); + const auto & utc_time_zone = DateLUT::instance("UTC"); + + parseDateTimeBestEffort(time, in, time_zone, utc_time_zone); + } + else + { + readDateTimeText(time, in); + } + + return time; + } } BlockIO InterpreterCreateUserQuery::execute() @@ -134,23 +162,7 @@ BlockIO InterpreterCreateUserQuery::execute() std::optional valid_until; if (query.valid_until) - { - const ASTPtr valid_until_literal = evaluateConstantExpressionAsLiteral(query.valid_until, getContext()); - const String valid_until_str = checkAndGetLiteralArgument(valid_until_literal, "valid_until"); - - time_t time = 0; - - if (valid_until_str != "infinity") - { - const auto & time_zone = DateLUT::instance(""); - const auto & utc_time_zone = DateLUT::instance("UTC"); - - ReadBufferFromString in(valid_until_str); - parseDateTimeBestEffort(time, in, time_zone, utc_time_zone); - } - - valid_until = time; - } + valid_until = getValidUntilFromAST(query.valid_until, getContext()); std::optional default_roles_from_query; if (query.default_roles) @@ -259,7 +271,11 @@ void InterpreterCreateUserQuery::updateUserFromQuery(User & user, const ASTCreat if (query.auth_data) auth_data = AuthenticationData::fromAST(*query.auth_data, {}, !query.attach); - updateUserFromQueryImpl(user, query, auth_data, {}, {}, {}, {}, {}, allow_no_password, allow_plaintext_password, true); + std::optional valid_until; + if (query.valid_until) + valid_until = getValidUntilFromAST(query.valid_until, {}); + + updateUserFromQueryImpl(user, query, auth_data, {}, {}, {}, {}, valid_until, allow_no_password, allow_plaintext_password, true); } void registerInterpreterCreateUserQuery(InterpreterFactory & factory) diff --git a/src/Interpreters/Access/InterpreterSetRoleQuery.cpp b/src/Interpreters/Access/InterpreterSetRoleQuery.cpp index 24467923542..99a7a73d46c 100644 --- a/src/Interpreters/Access/InterpreterSetRoleQuery.cpp +++ b/src/Interpreters/Access/InterpreterSetRoleQuery.cpp @@ -29,33 +29,12 @@ BlockIO InterpreterSetRoleQuery::execute() void InterpreterSetRoleQuery::setRole(const ASTSetRoleQuery & query) { - auto & access_control = getContext()->getAccessControl(); auto session_context = getContext()->getSessionContext(); - auto user = session_context->getUser(); if (query.kind == ASTSetRoleQuery::Kind::SET_ROLE_DEFAULT) - { session_context->setCurrentRolesDefault(); - } else - { - RolesOrUsersSet roles_from_query{*query.roles, access_control}; - std::vector new_current_roles; - if (roles_from_query.all) - { - new_current_roles = user->granted_roles.findGranted(roles_from_query); - } - else - { - for (const auto & id : roles_from_query.getMatchingIDs()) - { - if (!user->granted_roles.isGranted(id)) - throw Exception(ErrorCodes::SET_NON_GRANTED_ROLE, "Role should be granted to set current"); - new_current_roles.emplace_back(id); - } - } - session_context->setCurrentRoles(new_current_roles); - } + session_context->setCurrentRoles(RolesOrUsersSet{*query.roles, session_context->getAccessControl()}); } diff --git a/src/Interpreters/ActionsVisitor.cpp b/src/Interpreters/ActionsVisitor.cpp index c3285d73145..a874067e43c 100644 --- a/src/Interpreters/ActionsVisitor.cpp +++ b/src/Interpreters/ActionsVisitor.cpp @@ -8,6 +8,7 @@ #include #include +#include #include #include @@ -405,10 +406,6 @@ Block createBlockForSet( } -ScopeStack::Level::Level() = default; -ScopeStack::Level::~Level() = default; -ScopeStack::Level::Level(Level &&) noexcept = default; - FutureSetPtr makeExplicitSet( const ASTFunction * node, const ActionsDAG & actions, ContextPtr context, PreparedSets & prepared_sets) { @@ -446,7 +443,7 @@ FutureSetPtr makeExplicitSet( else block = createBlockForSet(left_arg_type, right_arg, set_element_types, context); - return prepared_sets.addFromTuple(set_key, block, context->getSettings()); + return prepared_sets.addFromTuple(set_key, block, context->getSettingsRef()); } class ScopeStack::Index @@ -462,6 +459,7 @@ public: for (const auto * node : index) map.emplace(node->result_name, node); } + ~Index() = default; void addNode(const ActionsDAG::Node * node) { @@ -502,6 +500,10 @@ public: } }; +ScopeStack::Level::Level() = default; +ScopeStack::Level::~Level() = default; +ScopeStack::Level::Level(Level &&) noexcept = default; + ActionsMatcher::Data::Data( ContextPtr context_, SizeLimits set_size_limit_, diff --git a/src/Interpreters/Aggregator.cpp b/src/Interpreters/Aggregator.cpp index 45b43ae2d3a..e073b7a49b6 100644 --- a/src/Interpreters/Aggregator.cpp +++ b/src/Interpreters/Aggregator.cpp @@ -26,7 +26,6 @@ #include #include #include -#include #include #include #include @@ -78,115 +77,6 @@ namespace ErrorCodes namespace { -/** Collects observed HashMap-s sizes to avoid redundant intermediate resizes. - */ -class HashTablesStatistics -{ -public: - struct Entry - { - size_t sum_of_sizes; // used to determine if it's better to convert aggregation to two-level from the beginning - size_t median_size; // roughly the size we're going to preallocate on each thread - }; - - using Cache = DB::CacheBase; - using CachePtr = std::shared_ptr; - using Params = DB::Aggregator::Params::StatsCollectingParams; - - /// Collection and use of the statistics should be enabled. - std::optional getSizeHint(const Params & params) - { - if (!params.isCollectionAndUseEnabled()) - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Collection and use of the statistics should be enabled."); - - std::lock_guard lock(mutex); - const auto cache = getHashTableStatsCache(params, lock); - if (const auto hint = cache->get(params.key)) - { - LOG_TRACE( - getLogger("Aggregator"), - "An entry for key={} found in cache: sum_of_sizes={}, median_size={}", - params.key, - hint->sum_of_sizes, - hint->median_size); - return *hint; - } - return std::nullopt; - } - - /// Collection and use of the statistics should be enabled. - void update(size_t sum_of_sizes, size_t median_size, const Params & params) - { - if (!params.isCollectionAndUseEnabled()) - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Collection and use of the statistics should be enabled."); - - std::lock_guard lock(mutex); - const auto cache = getHashTableStatsCache(params, lock); - const auto hint = cache->get(params.key); - // We'll maintain the maximum among all the observed values until the next prediction turns out to be too wrong. - if (!hint || sum_of_sizes < hint->sum_of_sizes / 2 || hint->sum_of_sizes < sum_of_sizes || median_size < hint->median_size / 2 - || hint->median_size < median_size) - { - LOG_TRACE( - getLogger("Aggregator"), - "Statistics updated for key={}: new sum_of_sizes={}, median_size={}", - params.key, - sum_of_sizes, - median_size); - cache->set(params.key, std::make_shared(Entry{.sum_of_sizes = sum_of_sizes, .median_size = median_size})); - } - } - - std::optional getCacheStats() const - { - std::lock_guard lock(mutex); - if (hash_table_stats) - { - size_t hits = 0, misses = 0; - hash_table_stats->getStats(hits, misses); - return DB::HashTablesCacheStatistics{.entries = hash_table_stats->count(), .hits = hits, .misses = misses}; - } - return std::nullopt; - } - - static size_t calculateCacheKey(const DB::ASTPtr & select_query) - { - if (!select_query) - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Query ptr cannot be null"); - - const auto & select = select_query->as(); - - // It may happen in some corner cases like `select 1 as num group by num`. - if (!select.tables()) - return 0; - - SipHash hash; - hash.update(select.tables()->getTreeHash(/*ignore_aliases=*/ true)); - if (const auto where = select.where()) - hash.update(where->getTreeHash(/*ignore_aliases=*/ true)); - if (const auto group_by = select.groupBy()) - hash.update(group_by->getTreeHash(/*ignore_aliases=*/ true)); - return hash.get64(); - } - -private: - CachePtr getHashTableStatsCache(const Params & params, const std::lock_guard &) - { - if (!hash_table_stats || hash_table_stats->maxSizeInBytes() != params.max_entries_for_hash_table_stats) - hash_table_stats = std::make_shared(params.max_entries_for_hash_table_stats); - return hash_table_stats; - } - - mutable std::mutex mutex; - CachePtr hash_table_stats; -}; - -HashTablesStatistics & getHashTablesStatistics() -{ - static HashTablesStatistics hash_tables_stats; - return hash_tables_stats; -} - bool worthConvertToTwoLevel( size_t group_by_two_level_threshold, size_t result_size, size_t group_by_two_level_threshold_bytes, auto result_size_bytes) { @@ -215,49 +105,29 @@ void initDataVariantsWithSizeHint( DB::AggregatedDataVariants & result, DB::AggregatedDataVariants::Type method_chosen, const DB::Aggregator::Params & params) { const auto & stats_collecting_params = params.stats_collecting_params; - if (stats_collecting_params.isCollectionAndUseEnabled()) + const auto max_threads = params.group_by_two_level_threshold != 0 ? std::max(params.max_threads, 1ul) : 1; + if (auto hint = getSizeHint(stats_collecting_params, /*tables_cnt=*/max_threads)) { - if (auto hint = getHashTablesStatistics().getSizeHint(stats_collecting_params)) - { - const auto max_threads = params.group_by_two_level_threshold != 0 ? std::max(params.max_threads, 1ul) : 1; - const auto lower_limit = hint->sum_of_sizes / max_threads; - const auto upper_limit = stats_collecting_params.max_size_to_preallocate_for_aggregation / max_threads; - if (hint->median_size > upper_limit) - { - /// Since we cannot afford to preallocate as much as we want, we will likely need to do resize anyway. - /// But we will also work with the big (i.e. not so cache friendly) HT from the beginning which may result in a slight slowdown. - /// So let's just do nothing. - LOG_TRACE( - getLogger("Aggregator"), - "No space were preallocated in hash tables because 'max_size_to_preallocate_for_aggregation' has too small value: {}, " - "should be at least {}", - stats_collecting_params.max_size_to_preallocate_for_aggregation, - hint->median_size * max_threads); - } - /// https://github.com/ClickHouse/ClickHouse/issues/44402#issuecomment-1359920703 - else if ((max_threads > 1 && hint->sum_of_sizes > 100'000) || hint->sum_of_sizes > 500'000) - { - const auto adjusted = std::max(lower_limit, hint->median_size); - if (worthConvertToTwoLevel( - params.group_by_two_level_threshold, - hint->sum_of_sizes, - /*group_by_two_level_threshold_bytes*/ 0, - /*result_size_bytes*/ 0)) - method_chosen = convertToTwoLevelTypeIfPossible(method_chosen); - result.init(method_chosen, adjusted); - ProfileEvents::increment(ProfileEvents::AggregationHashTablesInitializedAsTwoLevel, result.isTwoLevel()); - return; - } - } + if (worthConvertToTwoLevel( + params.group_by_two_level_threshold, + hint->sum_of_sizes, + /*group_by_two_level_threshold_bytes*/ 0, + /*result_size_bytes*/ 0)) + method_chosen = convertToTwoLevelTypeIfPossible(method_chosen); + result.init(method_chosen, hint->median_size); } - result.init(method_chosen); + else + { + result.init(method_chosen); + } + ProfileEvents::increment(ProfileEvents::AggregationHashTablesInitializedAsTwoLevel, result.isTwoLevel()); } /// Collection and use of the statistics should be enabled. -void updateStatistics(const DB::ManyAggregatedDataVariants & data_variants, const DB::Aggregator::Params::StatsCollectingParams & params) +void updateStatistics(const DB::ManyAggregatedDataVariants & data_variants, const DB::StatsCollectingParams & params) { if (!params.isCollectionAndUseEnabled()) - throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Collection and use of the statistics should be enabled."); + return; std::vector sizes(data_variants.size()); for (size_t i = 0; i < data_variants.size(); ++i) @@ -265,7 +135,7 @@ void updateStatistics(const DB::ManyAggregatedDataVariants & data_variants, cons const auto median_size = sizes.begin() + sizes.size() / 2; // not precisely though... std::nth_element(sizes.begin(), median_size, sizes.end()); const auto sum_of_sizes = std::accumulate(sizes.begin(), sizes.end(), 0ull); - getHashTablesStatistics().update(sum_of_sizes, *median_size, params); + DB::getHashTablesStatistics().update(sum_of_sizes, *median_size, params); } DB::ColumnNumbers calculateKeysPositions(const DB::Block & header, const DB::Aggregator::Params & params) @@ -300,24 +170,6 @@ size_t getMinBytesForPrefetch() namespace DB { -std::optional getHashTablesCacheStatistics() -{ - return getHashTablesStatistics().getCacheStats(); -} - -Aggregator::Params::StatsCollectingParams::StatsCollectingParams() = default; - -Aggregator::Params::StatsCollectingParams::StatsCollectingParams( - const ASTPtr & select_query_, - bool collect_hash_table_stats_during_aggregation_, - size_t max_entries_for_hash_table_stats_, - size_t max_size_to_preallocate_for_aggregation_) - : key(collect_hash_table_stats_during_aggregation_ ? HashTablesStatistics::calculateCacheKey(select_query_) : 0) - , max_entries_for_hash_table_stats(max_entries_for_hash_table_stats_) - , max_size_to_preallocate_for_aggregation(max_size_to_preallocate_for_aggregation_) -{ -} - Block Aggregator::getHeader(bool final) const { return params.getHeader(header, final); @@ -2783,8 +2635,7 @@ ManyAggregatedDataVariants Aggregator::prepareVariantsToMerge(ManyAggregatedData LOG_TRACE(log, "Merging aggregated data"); - if (params.stats_collecting_params.isCollectionAndUseEnabled()) - updateStatistics(data_variants, params.stats_collecting_params); + updateStatistics(data_variants, params.stats_collecting_params); ManyAggregatedDataVariants non_empty_data; non_empty_data.reserve(data_variants.size()); @@ -3486,4 +3337,23 @@ void Aggregator::destroyAllAggregateStates(AggregatedDataVariants & result) cons throw Exception(ErrorCodes::UNKNOWN_AGGREGATED_DATA_VARIANT, "Unknown aggregated data variant."); } +UInt64 calculateCacheKey(const DB::ASTPtr & select_query) +{ + if (!select_query) + throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Query ptr cannot be null"); + + const auto & select = select_query->as(); + + // It may happen in some corner cases like `select 1 as num group by num`. + if (!select.tables()) + return 0; + + SipHash hash; + hash.update(select.tables()->getTreeHash(/*ignore_aliases=*/true)); + if (const auto where = select.where()) + hash.update(where->getTreeHash(/*ignore_aliases=*/true)); + if (const auto group_by = select.groupBy()) + hash.update(group_by->getTreeHash(/*ignore_aliases=*/true)); + return hash.get64(); +} } diff --git a/src/Interpreters/Aggregator.h b/src/Interpreters/Aggregator.h index 406d28597cf..f4f1e9a1df3 100644 --- a/src/Interpreters/Aggregator.h +++ b/src/Interpreters/Aggregator.h @@ -39,9 +39,10 @@ #include -#include #include #include +#include +#include namespace DB { @@ -128,24 +129,6 @@ public: const double min_hit_rate_to_use_consecutive_keys_optimization; - struct StatsCollectingParams - { - StatsCollectingParams(); - - StatsCollectingParams( - const ASTPtr & select_query_, - bool collect_hash_table_stats_during_aggregation_, - size_t max_entries_for_hash_table_stats_, - size_t max_size_to_preallocate_for_aggregation_); - - bool isCollectionAndUseEnabled() const { return key != 0; } - void disable() { key = 0; } - - UInt64 key = 0; - const size_t max_entries_for_hash_table_stats = 0; - const size_t max_size_to_preallocate_for_aggregation = 0; - }; - StatsCollectingParams stats_collecting_params; Params( @@ -674,6 +657,7 @@ private: Arena * arena); }; +UInt64 calculateCacheKey(const DB::ASTPtr & select_query); /** Get the aggregation variant by its type. */ template Method & getDataVariant(AggregatedDataVariants & variants); @@ -685,13 +669,4 @@ APPLY_FOR_AGGREGATED_VARIANTS(M) #undef M - -struct HashTablesCacheStatistics -{ - size_t entries = 0; - size_t hits = 0; - size_t misses = 0; -}; - -std::optional getHashTablesCacheStatistics(); } diff --git a/src/Interpreters/ArrayJoinAction.cpp b/src/Interpreters/ArrayJoinAction.cpp index 54ff53764e7..df7a0b48057 100644 --- a/src/Interpreters/ArrayJoinAction.cpp +++ b/src/Interpreters/ArrayJoinAction.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include diff --git a/src/Interpreters/AsynchronousInsertQueue.cpp b/src/Interpreters/AsynchronousInsertQueue.cpp index d72f3d81549..62777524c2a 100644 --- a/src/Interpreters/AsynchronousInsertQueue.cpp +++ b/src/Interpreters/AsynchronousInsertQueue.cpp @@ -301,7 +301,13 @@ void AsynchronousInsertQueue::preprocessInsertQuery(const ASTPtr & query, const auto & insert_query = query->as(); insert_query.async_insert_flush = true; - InterpreterInsertQuery interpreter(query, query_context, query_context->getSettingsRef().insert_allow_materialized_columns); + InterpreterInsertQuery interpreter( + query, + query_context, + query_context->getSettingsRef().insert_allow_materialized_columns, + /* no_squash */ false, + /* no_destination */ false, + /* async_insert */ false); auto table = interpreter.getTable(insert_query); auto sample_block = InterpreterInsertQuery::getSampleBlock(insert_query, table, table->getInMemoryMetadataPtr(), query_context); @@ -726,7 +732,10 @@ try /// Access rights must be checked for the user who executed the initial INSERT query. if (key.user_id) - insert_context->setUser(*key.user_id, key.current_roles); + { + insert_context->setUser(*key.user_id); + insert_context->setCurrentRoles(key.current_roles); + } insert_context->setSettings(key.settings); @@ -781,7 +790,12 @@ try try { interpreter = std::make_unique( - key.query, insert_context, key.settings.insert_allow_materialized_columns, false, false, true); + key.query, + insert_context, + key.settings.insert_allow_materialized_columns, + false, + false, + true); pipeline = interpreter->execute().pipeline; chassert(pipeline.pushing()); @@ -1000,7 +1014,7 @@ Chunk AsynchronousInsertQueue::processEntriesWithParsing( } Chunk chunk(executor.getResultColumns(), total_rows); - chunk.setChunkInfo(std::move(chunk_info)); + chunk.getChunkInfos().add(std::move(chunk_info)); return chunk; } @@ -1052,7 +1066,7 @@ Chunk AsynchronousInsertQueue::processPreprocessedEntries( } Chunk chunk(std::move(result_columns), total_rows); - chunk.setChunkInfo(std::move(chunk_info)); + chunk.getChunkInfos().add(std::move(chunk_info)); return chunk; } diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index 0d33e39ffa3..a88c0de2cfe 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -30,6 +30,7 @@ namespace ProfileEvents extern const Event FilesystemCacheFailToReserveSpaceBecauseOfLockContention; extern const Event FilesystemCacheFreeSpaceKeepingThreadRun; extern const Event FilesystemCacheFreeSpaceKeepingThreadWorkMilliseconds; + extern const Event FilesystemCacheFailToReserveSpaceBecauseOfCacheResize; } namespace DB @@ -813,7 +814,7 @@ bool FileCache::tryReserve( /// ok compared to the number of cases this check will help. if (cache_is_being_resized.load(std::memory_order_relaxed)) { - ProfileEvents::increment(ProfileEvents::FilesystemCacheFailToReserveSpaceBecauseOfLockContention); + ProfileEvents::increment(ProfileEvents::FilesystemCacheFailToReserveSpaceBecauseOfCacheResize); return false; } @@ -1006,7 +1007,7 @@ void FileCache::freeSpaceRatioKeepingThreadFunc() limits_satisfied = main_priority->collectCandidatesForEviction( desired_size, desired_elements_num, keep_up_free_space_remove_batch, stat, eviction_candidates, lock); -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD /// Let's make sure that we correctly processed the limits. if (limits_satisfied && eviction_candidates.size() < keep_up_free_space_remove_batch) { @@ -1109,7 +1110,7 @@ void FileCache::removeAllReleasable(const UserID & user_id) { assertInitialized(); -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD assertCacheCorrectness(); #endif @@ -1225,7 +1226,7 @@ void FileCache::loadMetadataImpl() if (first_exception) std::rethrow_exception(first_exception); -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD assertCacheCorrectness(); #endif } @@ -1392,7 +1393,7 @@ void FileCache::loadMetadataForKeys(const fs::path & keys_dir) FileCache::~FileCache() { deactivateBackgroundOperations(); -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD assertCacheCorrectness(); #endif } diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index 61a356fa3c3..1664a91b694 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -67,7 +67,7 @@ FileSegment::FileSegment( , key_metadata(key_metadata_) , queue_iterator(queue_iterator_) , cache(cache_) -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD , log(getLogger(fmt::format("FileSegment({}) : {}", key_.toString(), range().toString()))) #else , log(getLogger("FileSegment")) @@ -187,13 +187,6 @@ size_t FileSegment::getDownloadedSize() const return downloaded_size; } -void FileSegment::setDownloadedSize(size_t delta) -{ - auto lk = lock(); - downloaded_size += delta; - assert(downloaded_size == std::filesystem::file_size(getPath())); -} - bool FileSegment::isDownloaded() const { auto lk = lock(); @@ -311,6 +304,11 @@ FileSegment::RemoteFileReaderPtr FileSegment::getRemoteFileReader() return remote_file_reader; } +FileSegment::LocalCacheWriterPtr FileSegment::getLocalCacheWriter() +{ + return cache_writer; +} + void FileSegment::resetRemoteFileReader() { auto lk = lock(); @@ -340,33 +338,31 @@ void FileSegment::setRemoteFileReader(RemoteFileReaderPtr remote_file_reader_) remote_file_reader = remote_file_reader_; } -void FileSegment::write(char * from, size_t size, size_t offset) +void FileSegment::write(char * from, size_t size, size_t offset_in_file) { ProfileEventTimeIncrement watch(ProfileEvents::FileSegmentWriteMicroseconds); - - if (!size) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Writing zero size is not allowed"); - + auto file_segment_path = getPath(); { - auto lk = lock(); - assertIsDownloaderUnlocked("write", lk); - assertNotDetachedUnlocked(lk); - } + if (!size) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Writing zero size is not allowed"); - const auto file_segment_path = getPath(); + { + auto lk = lock(); + assertIsDownloaderUnlocked("write", lk); + assertNotDetachedUnlocked(lk); + } - { if (download_state != State::DOWNLOADING) throw Exception( ErrorCodes::LOGICAL_ERROR, "Expected DOWNLOADING state, got {}", stateToString(download_state)); const size_t first_non_downloaded_offset = getCurrentWriteOffset(); - if (offset != first_non_downloaded_offset) + if (offset_in_file != first_non_downloaded_offset) throw Exception( ErrorCodes::LOGICAL_ERROR, "Attempt to write {} bytes to offset: {}, but current write offset is {}", - size, offset, first_non_downloaded_offset); + size, offset_in_file, first_non_downloaded_offset); const size_t current_downloaded_size = getDownloadedSize(); chassert(reserved_size >= current_downloaded_size); @@ -389,17 +385,17 @@ void FileSegment::write(char * from, size_t size, size_t offset) try { -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD /// This mutex is only needed to have a valid assertion in assertCacheCorrectness(), - /// which is only executed in debug/sanitizer builds (under ABORT_ON_LOGICAL_ERROR). + /// which is only executed in debug/sanitizer builds (under DEBUG_OR_SANITIZER_BUILD). std::lock_guard lock(write_mutex); #endif if (!cache_writer) - cache_writer = std::make_unique(file_segment_path, /* buf_size */0); + cache_writer = std::make_unique(getPath(), /* buf_size */0); /// Size is equal to offset as offset for write buffer points to data end. - cache_writer->set(from, size, /* offset */size); + cache_writer->set(from, /* size */size, /* offset */size); /// Reset the buffer when finished. SCOPE_EXIT({ cache_writer->set(nullptr, 0); }); /// Flush the buffer. @@ -435,7 +431,6 @@ void FileSegment::write(char * from, size_t size, size_t offset) } throw; - } catch (Exception & e) { @@ -445,7 +440,7 @@ void FileSegment::write(char * from, size_t size, size_t offset) throw; } - chassert(getCurrentWriteOffset() == offset + size); + chassert(getCurrentWriteOffset() == offset_in_file + size); } FileSegment::State FileSegment::wait(size_t offset) @@ -828,7 +823,7 @@ bool FileSegment::assertCorrectnessUnlocked(const FileSegmentGuard::Lock & lock) }; const auto file_path = getPath(); - if (segment_kind != FileSegmentKind::Temporary) + { std::lock_guard lk(write_mutex); if (downloaded_size == 0) diff --git a/src/Interpreters/Cache/FileSegment.h b/src/Interpreters/Cache/FileSegment.h index f28482a1ce4..d6b37b60dc1 100644 --- a/src/Interpreters/Cache/FileSegment.h +++ b/src/Interpreters/Cache/FileSegment.h @@ -48,7 +48,7 @@ friend class FileCache; /// Because of reserved_size in tryReserve(). public: using Key = FileCacheKey; using RemoteFileReaderPtr = std::shared_ptr; - using LocalCacheWriterPtr = std::unique_ptr; + using LocalCacheWriterPtr = std::shared_ptr; using Downloader = std::string; using DownloaderId = std::string; using Priority = IFileCachePriority; @@ -204,7 +204,7 @@ public: bool reserve(size_t size_to_reserve, size_t lock_wait_timeout_milliseconds, FileCacheReserveStat * reserve_stat = nullptr); /// Write data into reserved space. - void write(char * from, size_t size, size_t offset); + void write(char * from, size_t size, size_t offset_in_file); // Invariant: if state() != DOWNLOADING and remote file reader is present, the reader's // available() == 0, and getFileOffsetOfBufferEnd() == our getCurrentWriteOffset(). @@ -212,6 +212,7 @@ public: // The reader typically requires its internal_buffer to be assigned from the outside before // calling next(). RemoteFileReaderPtr getRemoteFileReader(); + LocalCacheWriterPtr getLocalCacheWriter(); RemoteFileReaderPtr extractRemoteFileReader(); @@ -219,8 +220,6 @@ public: void setRemoteFileReader(RemoteFileReaderPtr remote_file_reader_); - void setDownloadedSize(size_t delta); - void setDownloadFailed(); private: diff --git a/src/Interpreters/Cache/Metadata.cpp b/src/Interpreters/Cache/Metadata.cpp index 5ed4ccdbeca..7e4b76d3cc6 100644 --- a/src/Interpreters/Cache/Metadata.cpp +++ b/src/Interpreters/Cache/Metadata.cpp @@ -944,14 +944,7 @@ KeyMetadata::iterator LockedKey::removeFileSegmentImpl( try { const auto path = key_metadata->getFileSegmentPath(*file_segment); - if (file_segment->segment_kind == FileSegmentKind::Temporary) - { - /// FIXME: For temporary file segment the requirement is not as strong because - /// the implementation of "temporary data in cache" creates files in advance. - if (fs::exists(path)) - fs::remove(path); - } - else if (file_segment->downloaded_size == 0) + if (file_segment->downloaded_size == 0) { chassert(!fs::exists(path)); } @@ -970,7 +963,7 @@ KeyMetadata::iterator LockedKey::removeFileSegmentImpl( } else if (!can_be_broken) { -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected path {} to exist", path); #else LOG_WARNING(key_metadata->logger(), "Expected path {} to exist, while removing {}:{}", diff --git a/src/Interpreters/Cache/Metadata.h b/src/Interpreters/Cache/Metadata.h index a5c8f3c0cf4..0e85ead3265 100644 --- a/src/Interpreters/Cache/Metadata.h +++ b/src/Interpreters/Cache/Metadata.h @@ -1,11 +1,16 @@ #pragma once + + #include +#include #include #include #include #include #include #include + +#include #include namespace DB @@ -30,7 +35,7 @@ struct FileSegmentMetadata : private boost::noncopyable explicit FileSegmentMetadata(FileSegmentPtr && file_segment_); - bool releasable() const { return file_segment.unique(); } + bool releasable() const { return isSharedPtrUnique(file_segment); } size_t size() const; diff --git a/src/Interpreters/Cache/WriteBufferToFileSegment.cpp b/src/Interpreters/Cache/WriteBufferToFileSegment.cpp index a593ebfdab2..e6ebf6ad50c 100644 --- a/src/Interpreters/Cache/WriteBufferToFileSegment.cpp +++ b/src/Interpreters/Cache/WriteBufferToFileSegment.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -11,6 +12,8 @@ #include #include +#include + namespace DB { @@ -33,21 +36,20 @@ namespace } WriteBufferToFileSegment::WriteBufferToFileSegment(FileSegment * file_segment_) - : WriteBufferFromFileDecorator(std::make_unique(file_segment_->getPath())) + : WriteBufferFromFileBase(DBMS_DEFAULT_BUFFER_SIZE, nullptr, 0) , file_segment(file_segment_) , reserve_space_lock_wait_timeout_milliseconds(getCacheLockWaitTimeout()) { } WriteBufferToFileSegment::WriteBufferToFileSegment(FileSegmentsHolderPtr segment_holder_) - : WriteBufferFromFileDecorator( - segment_holder_->size() == 1 - ? std::make_unique(segment_holder_->front().getPath()) - : throw Exception(ErrorCodes::LOGICAL_ERROR, "WriteBufferToFileSegment can be created only from single segment")) + : WriteBufferFromFileBase(DBMS_DEFAULT_BUFFER_SIZE, nullptr, 0) , file_segment(&segment_holder_->front()) , segment_holder(std::move(segment_holder_)) , reserve_space_lock_wait_timeout_milliseconds(getCacheLockWaitTimeout()) { + if (segment_holder->size() != 1) + throw Exception(ErrorCodes::LOGICAL_ERROR, "WriteBufferToFileSegment can be created only from single segment"); } /// If it throws an exception, the file segment will be incomplete, so you should not use it in the future. @@ -82,9 +84,6 @@ void WriteBufferToFileSegment::nextImpl() reserve_stat_msg += fmt::format("{} hold {}, can release {}; ", toString(kind), ReadableSize(stat.non_releasable_size), ReadableSize(stat.releasable_size)); - if (std::filesystem::exists(file_segment->getPath())) - std::filesystem::remove(file_segment->getPath()); - throw Exception(ErrorCodes::NOT_ENOUGH_SPACE, "Failed to reserve {} bytes for {}: {}(segment info: {})", bytes_to_write, file_segment->getKind() == FileSegmentKind::Temporary ? "temporary file" : "the file in cache", @@ -95,17 +94,37 @@ void WriteBufferToFileSegment::nextImpl() try { - SwapHelper swap(*this, *impl); /// Write data to the underlying buffer. - impl->next(); + file_segment->write(working_buffer.begin(), bytes_to_write, written_bytes); + written_bytes += bytes_to_write; } catch (...) { LOG_WARNING(getLogger("WriteBufferToFileSegment"), "Failed to write to the underlying buffer ({})", file_segment->getInfoForLog()); throw; } +} - file_segment->setDownloadedSize(bytes_to_write); +void WriteBufferToFileSegment::finalizeImpl() +{ + next(); + auto cache_writer = file_segment->getLocalCacheWriter(); + if (cache_writer) + { + SwapHelper swap(*this, *cache_writer); + cache_writer->finalize(); + } +} + +void WriteBufferToFileSegment::sync() +{ + next(); + auto cache_writer = file_segment->getLocalCacheWriter(); + if (cache_writer) + { + SwapHelper swap(*this, *cache_writer); + cache_writer->sync(); + } } std::unique_ptr WriteBufferToFileSegment::getReadBufferImpl() @@ -114,7 +133,10 @@ std::unique_ptr WriteBufferToFileSegment::getReadBufferImpl() * because in case destructor called without `getReadBufferImpl` called, data won't be read. */ finalize(); - return std::make_unique(file_segment->getPath()); + if (file_segment->getDownloadedSize() > 0) + return std::make_unique(file_segment->getPath()); + else + return std::make_unique(); } } diff --git a/src/Interpreters/Cache/WriteBufferToFileSegment.h b/src/Interpreters/Cache/WriteBufferToFileSegment.h index c4b0491f8c0..4719dd4be89 100644 --- a/src/Interpreters/Cache/WriteBufferToFileSegment.h +++ b/src/Interpreters/Cache/WriteBufferToFileSegment.h @@ -9,7 +9,7 @@ namespace DB class FileSegment; -class WriteBufferToFileSegment : public WriteBufferFromFileDecorator, public IReadableWriteBuffer +class WriteBufferToFileSegment : public WriteBufferFromFileBase, public IReadableWriteBuffer { public: explicit WriteBufferToFileSegment(FileSegment * file_segment_); @@ -17,6 +17,13 @@ public: void nextImpl() override; + std::string getFileName() const override { return file_segment->getPath(); } + + void sync() override; + +protected: + void finalizeImpl() override; + private: std::unique_ptr getReadBufferImpl() override; @@ -29,6 +36,7 @@ private: FileSegmentsHolderPtr segment_holder; const size_t reserve_space_lock_wait_timeout_milliseconds; + size_t written_bytes = 0; }; diff --git a/src/Interpreters/ClientInfo.cpp b/src/Interpreters/ClientInfo.cpp index ce1efb61cc0..39fdef23baa 100644 --- a/src/Interpreters/ClientInfo.cpp +++ b/src/Interpreters/ClientInfo.cpp @@ -95,7 +95,7 @@ void ClientInfo::write(WriteBuffer & out, UInt64 server_protocol_revision) const if (server_protocol_revision >= DBMS_MIN_REVISION_WITH_PARALLEL_REPLICAS) { writeVarUInt(static_cast(collaborate_with_initiator), out); - writeVarUInt(count_participating_replicas, out); + writeVarUInt(obsolete_count_participating_replicas, out); writeVarUInt(number_of_current_replica, out); } } @@ -185,7 +185,7 @@ void ClientInfo::read(ReadBuffer & in, UInt64 client_protocol_revision) UInt64 value; readVarUInt(value, in); collaborate_with_initiator = static_cast(value); - readVarUInt(count_participating_replicas, in); + readVarUInt(obsolete_count_participating_replicas, in); readVarUInt(number_of_current_replica, in); } } diff --git a/src/Interpreters/ClientInfo.h b/src/Interpreters/ClientInfo.h index 3054667e264..ca32b4c5cfa 100644 --- a/src/Interpreters/ClientInfo.h +++ b/src/Interpreters/ClientInfo.h @@ -127,7 +127,7 @@ public: /// For parallel processing on replicas bool collaborate_with_initiator{false}; - UInt64 count_participating_replicas{0}; + UInt64 obsolete_count_participating_replicas{0}; UInt64 number_of_current_replica{0}; enum class BackgroundOperationType : uint8_t diff --git a/src/Interpreters/ClusterProxy/SelectStreamFactory.cpp b/src/Interpreters/ClusterProxy/SelectStreamFactory.cpp index 94a02eacdaa..e35d31d2350 100644 --- a/src/Interpreters/ClusterProxy/SelectStreamFactory.cpp +++ b/src/Interpreters/ClusterProxy/SelectStreamFactory.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/ClusterProxy/executeQuery.cpp b/src/Interpreters/ClusterProxy/executeQuery.cpp index 91c0c592f28..d04a73e384e 100644 --- a/src/Interpreters/ClusterProxy/executeQuery.cpp +++ b/src/Interpreters/ClusterProxy/executeQuery.cpp @@ -8,23 +8,27 @@ #include #include #include +#include #include -#include -#include #include +#include +#include +#include +#include +#include #include +#include #include #include -#include #include +#include #include -#include +#include #include #include -#include +#include #include -#include -#include +#include namespace DB { @@ -33,7 +37,6 @@ namespace ErrorCodes { extern const int TOO_LARGE_DISTRIBUTED_DEPTH; extern const int LOGICAL_ERROR; - extern const int CLUSTER_DOESNT_EXIST; extern const int UNEXPECTED_CLUSTER; } @@ -172,7 +175,7 @@ ContextMutablePtr updateSettingsAndClientInfoForCluster(const Cluster & cluster, /// in case of parallel replicas custom key use round robing load balancing /// so custom key partitions will be spread over nodes in round-robin fashion - if (context->canUseParallelReplicasCustomKey(cluster) && !settings.load_balancing.changed) + if (context->canUseParallelReplicasCustomKeyForCluster(cluster) && !settings.load_balancing.changed) { new_settings.load_balancing = LoadBalancing::ROUND_ROBIN; } @@ -180,6 +183,10 @@ ContextMutablePtr updateSettingsAndClientInfoForCluster(const Cluster & cluster, auto new_context = Context::createCopy(context); new_context->setSettings(new_settings); new_context->setClientInfo(new_client_info); + + if (context->canUseParallelReplicasCustomKeyForCluster(cluster)) + new_context->disableOffsetParallelReplicas(); + return new_context; } @@ -220,6 +227,35 @@ static ThrottlerPtr getThrottler(const ContextPtr & context) return throttler; } +AdditionalShardFilterGenerator +getShardFilterGeneratorForCustomKey(const Cluster & cluster, ContextPtr context, const ColumnsDescription & columns) +{ + if (!context->canUseParallelReplicasCustomKeyForCluster(cluster)) + return {}; + + const auto & settings = context->getSettingsRef(); + auto custom_key_ast = parseCustomKeyForTable(settings.parallel_replicas_custom_key, *context); + if (custom_key_ast == nullptr) + return {}; + + return [my_custom_key_ast = std::move(custom_key_ast), + column_description = columns, + custom_key_type = settings.parallel_replicas_custom_key_filter_type.value, + custom_key_range_lower = settings.parallel_replicas_custom_key_range_lower.value, + custom_key_range_upper = settings.parallel_replicas_custom_key_range_upper.value, + query_context = context, + replica_count = cluster.getShardsInfo().front().per_replica_pools.size()](uint64_t replica_num) -> ASTPtr + { + return getCustomKeyFilterForParallelReplica( + replica_count, + replica_num - 1, + my_custom_key_ast, + {custom_key_type, custom_key_range_lower, custom_key_range_upper}, + column_description, + query_context); + }; +} + void executeQuery( QueryPlan & query_plan, @@ -412,14 +448,7 @@ void executeQueryWithParallelReplicas( const auto & settings = context->getSettingsRef(); /// check cluster for parallel replicas - if (settings.cluster_for_parallel_replicas.value.empty()) - { - throw Exception( - ErrorCodes::CLUSTER_DOESNT_EXIST, - "Reading in parallel from replicas is enabled but cluster to execute query is not provided. Please set " - "'cluster_for_parallel_replicas' setting"); - } - auto not_optimized_cluster = context->getCluster(settings.cluster_for_parallel_replicas); + auto not_optimized_cluster = context->getClusterForParallelReplicas(); auto new_context = Context::createCopy(context); @@ -487,14 +516,11 @@ void executeQueryWithParallelReplicas( "`cluster_for_parallel_replicas` setting refers to cluster with several shards. Expected a cluster with one shard"); } - auto coordinator = std::make_shared( - new_cluster->getShardsInfo().begin()->getAllNodeCount(), settings.parallel_replicas_mark_segment_size); auto external_tables = new_context->getExternalTables(); auto read_from_remote = std::make_unique( query_ast, new_cluster, storage_id, - std::move(coordinator), header, processed_stage, new_context, @@ -542,6 +568,84 @@ void executeQueryWithParallelReplicas( executeQueryWithParallelReplicas(query_plan, storage_id, header, processed_stage, modified_query_ast, context, storage_limits); } +void executeQueryWithParallelReplicasCustomKey( + QueryPlan & query_plan, + const StorageID & storage_id, + const SelectQueryInfo & query_info, + const ColumnsDescription & columns, + const StorageSnapshotPtr & snapshot, + QueryProcessingStage::Enum processed_stage, + const Block & header, + ContextPtr context) +{ + /// Return directly (with correct header) if no shard to query. + if (query_info.getCluster()->getShardsInfo().empty()) + { + if (context->getSettingsRef().allow_experimental_analyzer) + return; + + Pipe pipe(std::make_shared(header)); + auto read_from_pipe = std::make_unique(std::move(pipe)); + read_from_pipe->setStepDescription("Read from NullSource (Distributed)"); + query_plan.addStep(std::move(read_from_pipe)); + return; + } + + ColumnsDescriptionByShardNum columns_object; + if (hasDynamicSubcolumns(columns)) + columns_object = getExtendedObjectsOfRemoteTables(*query_info.cluster, storage_id, columns, context); + + ClusterProxy::SelectStreamFactory select_stream_factory + = ClusterProxy::SelectStreamFactory(header, columns_object, snapshot, processed_stage); + + auto shard_filter_generator = getShardFilterGeneratorForCustomKey(*query_info.getCluster(), context, columns); + + ClusterProxy::executeQuery( + query_plan, + header, + processed_stage, + storage_id, + /*table_func_ptr=*/nullptr, + select_stream_factory, + getLogger("executeQueryWithParallelReplicasCustomKey"), + context, + query_info, + /*sharding_key_expr=*/nullptr, + /*sharding_key_column_name=*/{}, + /*distributed_settings=*/{}, + shard_filter_generator, + /*is_remote_function=*/false); +} + +void executeQueryWithParallelReplicasCustomKey( + QueryPlan & query_plan, + const StorageID & storage_id, + const SelectQueryInfo & query_info, + const ColumnsDescription & columns, + const StorageSnapshotPtr & snapshot, + QueryProcessingStage::Enum processed_stage, + const QueryTreeNodePtr & query_tree, + ContextPtr context) +{ + auto header = InterpreterSelectQueryAnalyzer::getSampleBlock(query_tree, context, SelectQueryOptions(processed_stage).analyze()); + executeQueryWithParallelReplicasCustomKey(query_plan, storage_id, query_info, columns, snapshot, processed_stage, header, context); +} + +void executeQueryWithParallelReplicasCustomKey( + QueryPlan & query_plan, + const StorageID & storage_id, + SelectQueryInfo query_info, + const ColumnsDescription & columns, + const StorageSnapshotPtr & snapshot, + QueryProcessingStage::Enum processed_stage, + const ASTPtr & query_ast, + ContextPtr context) +{ + auto header = InterpreterSelectQuery(query_ast, context, SelectQueryOptions(processed_stage).analyze()).getSampleBlock(); + query_info.query = ClusterProxy::rewriteSelectQuery( + context, query_info.query, storage_id.getDatabaseName(), storage_id.getTableName(), /*table_function_ptr=*/nullptr); + executeQueryWithParallelReplicasCustomKey(query_plan, storage_id, query_info, columns, snapshot, processed_stage, header, context); +} } } diff --git a/src/Interpreters/ClusterProxy/executeQuery.h b/src/Interpreters/ClusterProxy/executeQuery.h index 6548edf8939..c22fcd24f03 100644 --- a/src/Interpreters/ClusterProxy/executeQuery.h +++ b/src/Interpreters/ClusterProxy/executeQuery.h @@ -1,7 +1,7 @@ #pragma once -#include #include +#include #include namespace DB @@ -13,6 +13,11 @@ class Cluster; using ClusterPtr = std::shared_ptr; struct SelectQueryInfo; +class ColumnsDescription; +struct StorageSnapshot; + +using StorageSnapshotPtr = std::shared_ptr; + class Pipe; class QueryPlan; @@ -47,6 +52,9 @@ class SelectStreamFactory; ContextMutablePtr updateSettingsForCluster(const Cluster & cluster, ContextPtr context, const Settings & settings, const StorageID & main_table); using AdditionalShardFilterGenerator = std::function; +AdditionalShardFilterGenerator +getShardFilterGeneratorForCustomKey(const Cluster & cluster, ContextPtr context, const ColumnsDescription & columns); + /// Execute a distributed query, creating a query plan, from which the query pipeline can be built. /// `stream_factory` object encapsulates the logic of creating plans for a different type of query /// (currently SELECT, DESCRIBE). @@ -91,6 +99,36 @@ void executeQueryWithParallelReplicas( const PlannerContextPtr & planner_context, ContextPtr context, std::shared_ptr storage_limits); + +void executeQueryWithParallelReplicasCustomKey( + QueryPlan & query_plan, + const StorageID & storage_id, + const SelectQueryInfo & query_info, + const ColumnsDescription & columns, + const StorageSnapshotPtr & snapshot, + QueryProcessingStage::Enum processed_stage, + const Block & header, + ContextPtr context); + +void executeQueryWithParallelReplicasCustomKey( + QueryPlan & query_plan, + const StorageID & storage_id, + const SelectQueryInfo & query_info, + const ColumnsDescription & columns, + const StorageSnapshotPtr & snapshot, + QueryProcessingStage::Enum processed_stage, + const QueryTreeNodePtr & query_tree, + ContextPtr context); + +void executeQueryWithParallelReplicasCustomKey( + QueryPlan & query_plan, + const StorageID & storage_id, + SelectQueryInfo query_info, + const ColumnsDescription & columns, + const StorageSnapshotPtr & snapshot, + QueryProcessingStage::Enum processed_stage, + const ASTPtr & query_ast, + ContextPtr context); } } diff --git a/src/Interpreters/ConcurrentHashJoin.cpp b/src/Interpreters/ConcurrentHashJoin.cpp index 53987694e46..ac940c62a1a 100644 --- a/src/Interpreters/ConcurrentHashJoin.cpp +++ b/src/Interpreters/ConcurrentHashJoin.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include @@ -16,12 +17,18 @@ #include #include #include +#include #include #include #include #include #include +namespace ProfileEvents +{ +extern const Event HashJoinPreallocatedElementsInHashTables; +} + namespace CurrentMetrics { extern const Metric ConcurrentHashJoinPoolThreads; @@ -29,6 +36,25 @@ extern const Metric ConcurrentHashJoinPoolThreadsActive; extern const Metric ConcurrentHashJoinPoolThreadsScheduled; } +namespace +{ + +void updateStatistics(const auto & hash_joins, const DB::StatsCollectingParams & params) +{ + if (!params.isCollectionAndUseEnabled()) + return; + + std::vector sizes(hash_joins.size()); + for (size_t i = 0; i < hash_joins.size(); ++i) + sizes[i] = hash_joins[i]->data->getTotalRowCount(); + const auto median_size = sizes.begin() + sizes.size() / 2; // not precisely though... + std::nth_element(sizes.begin(), median_size, sizes.end()); + if (auto sum_of_sizes = std::accumulate(sizes.begin(), sizes.end(), 0ull)) + DB::getHashTablesStatistics().update(sum_of_sizes, *median_size, params); +} + +} + namespace DB { @@ -46,7 +72,12 @@ static UInt32 toPowerOfTwo(UInt32 x) } ConcurrentHashJoin::ConcurrentHashJoin( - ContextPtr context_, std::shared_ptr table_join_, size_t slots_, const Block & right_sample_block, bool any_take_last_row_) + ContextPtr context_, + std::shared_ptr table_join_, + size_t slots_, + const Block & right_sample_block, + const StatsCollectingParams & stats_collecting_params_, + bool any_take_last_row_) : context(context_) , table_join(table_join_) , slots(toPowerOfTwo(std::min(static_cast(slots_), 256))) @@ -55,6 +86,7 @@ ConcurrentHashJoin::ConcurrentHashJoin( CurrentMetrics::ConcurrentHashJoinPoolThreadsActive, CurrentMetrics::ConcurrentHashJoinPoolThreadsScheduled, slots)) + , stats_collecting_params(stats_collecting_params_) { hash_joins.resize(slots); @@ -74,9 +106,14 @@ ConcurrentHashJoin::ConcurrentHashJoin( CurrentThread::attachToGroupIfDetached(thread_group); setThreadName("ConcurrentJoin"); + size_t reserve_size = 0; + if (auto hint = getSizeHint(stats_collecting_params, slots)) + reserve_size = hint->median_size; + ProfileEvents::increment(ProfileEvents::HashJoinPreallocatedElementsInHashTables, reserve_size); + auto inner_hash_join = std::make_shared(); inner_hash_join->data = std::make_unique( - table_join_, right_sample_block, any_take_last_row_, 0, fmt::format("concurrent{}", idx)); + table_join_, right_sample_block, any_take_last_row_, reserve_size, fmt::format("concurrent{}", idx)); /// Non zero `max_joined_block_rows` allows to process block partially and return not processed part. /// TODO: It's not handled properly in ConcurrentHashJoin case, so we set it to 0 to disable this feature. inner_hash_join->data->setMaxJoinedBlockRows(0); @@ -97,6 +134,8 @@ ConcurrentHashJoin::~ConcurrentHashJoin() { try { + updateStatistics(hash_joins, stats_collecting_params); + for (size_t i = 0; i < slots; ++i) { // Hash tables destruction may be very time-consuming. @@ -271,7 +310,7 @@ IColumn::Selector ConcurrentHashJoin::selectDispatchBlock(const Strings & key_co { const auto & key_col = from_block.getByName(key_name).column->convertToFullColumnIfConst(); const auto & key_col_no_lc = recursiveRemoveLowCardinality(recursiveRemoveSparse(key_col)); - key_col_no_lc->updateWeakHash32(hash); + hash.update(key_col_no_lc->getWeakHash32()); } return hashToSelector(hash, num_shards); } @@ -300,4 +339,16 @@ Blocks ConcurrentHashJoin::dispatchBlock(const Strings & key_columns_names, cons return result; } +UInt64 calculateCacheKey(std::shared_ptr & table_join, const QueryTreeNodePtr & right_table_expression) +{ + IQueryTreeNode::HashState hash; + chassert(right_table_expression); + hash.update(right_table_expression->getTreeHash()); + chassert(table_join && table_join->oneDisjunct()); + const auto keys + = NameOrderedSet{table_join->getClauses().at(0).key_names_right.begin(), table_join->getClauses().at(0).key_names_right.end()}; + for (const auto & name : keys) + hash.update(name); + return hash.get64(); +} } diff --git a/src/Interpreters/ConcurrentHashJoin.h b/src/Interpreters/ConcurrentHashJoin.h index c797ff27ece..a911edaccc3 100644 --- a/src/Interpreters/ConcurrentHashJoin.h +++ b/src/Interpreters/ConcurrentHashJoin.h @@ -3,9 +3,11 @@ #include #include #include +#include #include #include -#include +#include +#include #include #include #include @@ -38,6 +40,7 @@ public: std::shared_ptr table_join_, size_t slots_, const Block & right_sample_block, + const StatsCollectingParams & stats_collecting_params_, bool any_take_last_row_ = false); ~ConcurrentHashJoin() override; @@ -70,6 +73,8 @@ private: std::unique_ptr pool; std::vector> hash_joins; + StatsCollectingParams stats_collecting_params; + std::mutex totals_mutex; Block totals; @@ -77,4 +82,5 @@ private: Blocks dispatchBlock(const Strings & key_columns_names, const Block & from_block); }; +UInt64 calculateCacheKey(std::shared_ptr & table_join, const QueryTreeNodePtr & right_table_expression); } diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index ff8e2cb2adc..94bcb88ed53 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include @@ -50,7 +51,6 @@ #include #include #include -#include #include #include #include @@ -58,6 +58,7 @@ #include #include #include +#include #include #include #include @@ -190,6 +191,7 @@ namespace ErrorCodes extern const int ILLEGAL_COLUMN; extern const int NUMBER_OF_COLUMNS_DOESNT_MATCH; extern const int CLUSTER_DOESNT_EXIST; + extern const int SET_NON_GRANTED_ROLE; } #define SHUTDOWN(log, desc, ptr, method) do \ @@ -822,8 +824,69 @@ void ContextSharedMutex::lockSharedImpl() ProfileEvents::increment(ProfileEvents::ContextLockWaitMicroseconds, watch.elapsedMicroseconds()); } -ContextData::ContextData() = default; -ContextData::ContextData(const ContextData &) = default; +ContextData::ContextData() +{ + settings = std::make_unique(); +} +ContextData::ContextData(const ContextData &o) : + shared(o.shared), + client_info(o.client_info), + external_tables_initializer_callback(o.external_tables_initializer_callback), + input_initializer_callback(o.input_initializer_callback), + input_blocks_reader(o.input_blocks_reader), + user_id(o.user_id), + current_roles(o.current_roles), + settings_constraints_and_current_profiles(o.settings_constraints_and_current_profiles), + access(o.access), + need_recalculate_access(o.need_recalculate_access), + current_database(o.current_database), + settings(std::make_unique(*o.settings)), + progress_callback(o.progress_callback), + file_progress_callback(o.file_progress_callback), + process_list_elem(o.process_list_elem), + has_process_list_elem(o.has_process_list_elem), + insertion_table_info(o.insertion_table_info), + is_distributed(o.is_distributed), + default_format(o.default_format), + insert_format(o.insert_format), + external_tables_mapping(o.external_tables_mapping), + scalars(o.scalars), + special_scalars(o.special_scalars), + next_task_callback(o.next_task_callback), + merge_tree_read_task_callback(o.merge_tree_read_task_callback), + merge_tree_all_ranges_callback(o.merge_tree_all_ranges_callback), + parallel_replicas_group_uuid(o.parallel_replicas_group_uuid), + client_protocol_version(o.client_protocol_version), + query_access_info(std::make_shared(*o.query_access_info)), + query_factories_info(o.query_factories_info), + query_privileges_info(o.query_privileges_info), + async_read_counters(o.async_read_counters), + view_source(o.view_source), + table_function_results(o.table_function_results), + query_context(o.query_context), + session_context(o.session_context), + global_context(o.global_context), + buffer_context(o.buffer_context), + is_internal_query(o.is_internal_query), + temp_data_on_disk(o.temp_data_on_disk), + classifier(o.classifier), + prepared_sets_cache(o.prepared_sets_cache), + offset_parallel_replicas_enabled(o.offset_parallel_replicas_enabled), + kitchen_sink(o.kitchen_sink), + part_uuids(o.part_uuids), + ignored_part_uuids(o.ignored_part_uuids), + query_parameters(o.query_parameters), + host_context(o.host_context), + metadata_transaction(o.metadata_transaction), + merge_tree_transaction(o.merge_tree_transaction), + merge_tree_transaction_holder(o.merge_tree_transaction_holder), + remote_read_query_throttler(o.remote_read_query_throttler), + remote_write_query_throttler(o.remote_write_query_throttler), + local_read_query_throttler(o.local_read_query_throttler), + local_write_query_throttler(o.local_write_query_throttler), + backups_query_throttler(o.backups_query_throttler) +{ +} Context::Context() = default; Context::Context(const Context & rhs) : ContextData(rhs), std::enable_shared_from_this(rhs) {} @@ -863,7 +926,6 @@ ContextMutablePtr Context::createCopy(const ContextPtr & other) { SharedLockGuard lock(other->mutex); auto new_context = std::shared_ptr(new Context(*other)); - new_context->query_access_info = std::make_shared(*other->query_access_info); return new_context; } @@ -969,7 +1031,7 @@ Strings Context::getWarnings() const } /// Make setting's name ordered std::set obsolete_settings; - for (const auto & setting : settings) + for (const auto & setting : *settings) { if (setting.isValueChanged() && setting.isObsolete()) obsolete_settings.emplace(setting.getName()); @@ -1303,7 +1365,7 @@ ConfigurationPtr Context::getUsersConfig() return shared->users_config; } -void Context::setUser(const UUID & user_id_, const std::optional> & current_roles_) +void Context::setUser(const UUID & user_id_) { /// Prepare lists of user's profiles, constraints, settings, roles. /// NOTE: AccessControl::read() and other AccessControl's functions may require some IO work, @@ -1312,8 +1374,8 @@ void Context::setUser(const UUID & user_id_, const std::optional(user_id_); - auto new_current_roles = current_roles_ ? user->granted_roles.findGranted(*current_roles_) : user->granted_roles.findGranted(user->default_roles); - auto enabled_roles = access_control.getEnabledRolesInfo(new_current_roles, {}); + auto default_roles = user->granted_roles.findGranted(user->default_roles); + auto enabled_roles = access_control.getEnabledRolesInfo(default_roles, {}); auto enabled_profiles = access_control.getEnabledSettingsInfo(user_id_, user->settings, enabled_roles->enabled_roles, enabled_roles->settings_from_enabled_roles); const auto & database = user->default_database; @@ -1327,7 +1389,7 @@ void Context::setUser(const UUID & user_id_, const std::optional Context::getUserID() const return user_id; } -void Context::setCurrentRolesWithLock(const std::vector & current_roles_, const std::lock_guard &) +void Context::setCurrentRolesWithLock(const std::vector & new_current_roles, const std::lock_guard &) { - if (current_roles_.empty()) + if (new_current_roles.empty()) current_roles = nullptr; else - current_roles = std::make_shared>(current_roles_); + current_roles = std::make_shared>(new_current_roles); need_recalculate_access = true; } -void Context::setCurrentRoles(const std::vector & current_roles_) +void Context::setCurrentRolesImpl(const std::vector & new_current_roles, bool throw_if_not_granted, bool skip_if_not_granted, const std::shared_ptr & user) { - std::lock_guard lock(mutex); - setCurrentRolesWithLock(current_roles_, lock); + if (skip_if_not_granted) + { + auto filtered_role_ids = user->granted_roles.findGranted(new_current_roles); + std::lock_guard lock{mutex}; + setCurrentRolesWithLock(filtered_role_ids, lock); + return; + } + if (throw_if_not_granted) + { + for (const auto & role_id : new_current_roles) + { + if (!user->granted_roles.isGranted(role_id)) + { + auto role_name = getAccessControl().tryReadName(role_id); + throw Exception(ErrorCodes::SET_NON_GRANTED_ROLE, "Role {} should be granted to set as a current", role_name.value_or(toString(role_id))); + } + } + } + std::lock_guard lock2{mutex}; + setCurrentRolesWithLock(new_current_roles, lock2); +} + +void Context::setCurrentRoles(const std::vector & new_current_roles, bool check_grants) +{ + setCurrentRolesImpl(new_current_roles, /* throw_if_not_granted= */ check_grants, /* skip_if_not_granted= */ !check_grants, getUser()); +} + +void Context::setCurrentRoles(const RolesOrUsersSet & new_current_roles, bool check_grants) +{ + if (new_current_roles.all) + { + auto user = getUser(); + setCurrentRolesImpl(user->granted_roles.findGranted(new_current_roles), /* throw_if_not_granted= */ false, /* skip_if_not_granted= */ false, user); + } + else + { + setCurrentRoles(new_current_roles.getMatchingIDs(), check_grants); + } +} + +void Context::setCurrentRoles(const Strings & new_current_roles, bool check_grants) +{ + setCurrentRoles(getAccessControl().getIDs(new_current_roles), check_grants); } void Context::setCurrentRolesDefault() { auto user = getUser(); - setCurrentRoles(user->granted_roles.findGranted(user->default_roles)); + setCurrentRolesImpl(user->granted_roles.findGranted(user->default_roles), /* throw_if_not_granted= */ false, /* skip_if_not_granted= */ false, user); } std::vector Context::getCurrentRoles() const @@ -1443,7 +1546,7 @@ std::shared_ptr Context::getAccess() const /// If setUserID() was never called then this must be the global context with the full access. bool full_access = !user_id; - return ContextAccessParams{user_id, full_access, /* use_default_roles= */ false, current_roles, settings, current_database, client_info}; + return ContextAccessParams{user_id, full_access, /* use_default_roles= */ false, current_roles, *settings, current_database, client_info}; }; /// Check if the current access rights are still valid, otherwise get parameters for recalculating access rights. @@ -1525,7 +1628,7 @@ void Context::setCurrentProfilesWithLock(const SettingsProfilesInfo & profiles_i checkSettingsConstraintsWithLock(profiles_info.settings, SettingSource::PROFILE); applySettingsChangesWithLock(profiles_info.settings, lock); settings_constraints_and_current_profiles = profiles_info.getConstraintsAndProfileIDs(settings_constraints_and_current_profiles); - contextSanityClampSettingsWithLock(*this, settings, lock); + contextSanityClampSettingsWithLock(*this, *settings, lock); } void Context::setCurrentProfile(const String & profile_name, bool check_constraints) @@ -2166,15 +2269,15 @@ bool Context::displaySecretsInShowAndSelect() const Settings Context::getSettings() const { SharedLockGuard lock(mutex); - return settings; + return *settings; } void Context::setSettings(const Settings & settings_) { std::lock_guard lock(mutex); - settings = settings_; + *settings = settings_; need_recalculate_access = true; - contextSanityClampSettings(*this, settings); + contextSanityClampSettings(*this, *settings); } void Context::setSettingWithLock(std::string_view name, const String & value, const std::lock_guard & lock) @@ -2184,10 +2287,10 @@ void Context::setSettingWithLock(std::string_view name, const String & value, co setCurrentProfileWithLock(value, true /*check_constraints*/, lock); return; } - settings.set(name, value); + settings->set(name, value); if (ContextAccessParams::dependsOnSettingName(name)) need_recalculate_access = true; - contextSanityClampSettingsWithLock(*this, settings, lock); + contextSanityClampSettingsWithLock(*this, *settings, lock); } void Context::setSettingWithLock(std::string_view name, const Field & value, const std::lock_guard & lock) @@ -2197,7 +2300,7 @@ void Context::setSettingWithLock(std::string_view name, const Field & value, con setCurrentProfileWithLock(value.safeGet(), true /*check_constraints*/, lock); return; } - settings.set(name, value); + settings->set(name, value); if (ContextAccessParams::dependsOnSettingName(name)) need_recalculate_access = true; } @@ -2207,7 +2310,7 @@ void Context::applySettingChangeWithLock(const SettingChange & change, const std try { setSettingWithLock(change.name, change.value, lock); - contextSanityClampSettingsWithLock(*this, settings, lock); + contextSanityClampSettingsWithLock(*this, *settings, lock); } catch (Exception & e) { @@ -2222,7 +2325,7 @@ void Context::applySettingsChangesWithLock(const SettingsChanges & changes, cons { for (const SettingChange & change : changes) applySettingChangeWithLock(change, lock); - applySettingsQuirks(settings); + applySettingsQuirks(*settings); } void Context::setSetting(std::string_view name, const String & value) @@ -2235,7 +2338,13 @@ void Context::setSetting(std::string_view name, const Field & value) { std::lock_guard lock(mutex); setSettingWithLock(name, value, lock); - contextSanityClampSettingsWithLock(*this, settings, lock); + contextSanityClampSettingsWithLock(*this, *settings, lock); +} + +void Context::setServerSetting(std::string_view name, const Field & value) +{ + std::lock_guard lock(mutex); + shared->server_settings.set(name, value); } void Context::applySettingChange(const SettingChange & change) @@ -2262,37 +2371,37 @@ void Context::applySettingsChanges(const SettingsChanges & changes) void Context::checkSettingsConstraintsWithLock(const SettingsProfileElements & profile_elements, SettingSource source) { - getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(settings, profile_elements, source); + getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(*settings, profile_elements, source); if (getApplicationType() == ApplicationType::LOCAL || getApplicationType() == ApplicationType::SERVER) - doSettingsSanityCheckClamp(settings, getLogger("SettingsSanity")); + doSettingsSanityCheckClamp(*settings, getLogger("SettingsSanity")); } void Context::checkSettingsConstraintsWithLock(const SettingChange & change, SettingSource source) { - getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(settings, change, source); + getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(*settings, change, source); if (getApplicationType() == ApplicationType::LOCAL || getApplicationType() == ApplicationType::SERVER) - doSettingsSanityCheckClamp(settings, getLogger("SettingsSanity")); + doSettingsSanityCheckClamp(*settings, getLogger("SettingsSanity")); } void Context::checkSettingsConstraintsWithLock(const SettingsChanges & changes, SettingSource source) { - getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(settings, changes, source); + getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(*settings, changes, source); if (getApplicationType() == ApplicationType::LOCAL || getApplicationType() == ApplicationType::SERVER) - doSettingsSanityCheckClamp(settings, getLogger("SettingsSanity")); + doSettingsSanityCheckClamp(*settings, getLogger("SettingsSanity")); } void Context::checkSettingsConstraintsWithLock(SettingsChanges & changes, SettingSource source) { - getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(settings, changes, source); + getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(*settings, changes, source); if (getApplicationType() == ApplicationType::LOCAL || getApplicationType() == ApplicationType::SERVER) - doSettingsSanityCheckClamp(settings, getLogger("SettingsSanity")); + doSettingsSanityCheckClamp(*settings, getLogger("SettingsSanity")); } void Context::clampToSettingsConstraintsWithLock(SettingsChanges & changes, SettingSource source) { - getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.clamp(settings, changes, source); + getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.clamp(*settings, changes, source); if (getApplicationType() == ApplicationType::LOCAL || getApplicationType() == ApplicationType::SERVER) - doSettingsSanityCheckClamp(settings, getLogger("SettingsSanity")); + doSettingsSanityCheckClamp(*settings, getLogger("SettingsSanity")); } void Context::checkMergeTreeSettingsConstraintsWithLock(const MergeTreeSettings & merge_tree_settings, const SettingsChanges & changes) const @@ -2315,8 +2424,8 @@ void Context::checkSettingsConstraints(const SettingChange & change, SettingSour void Context::checkSettingsConstraints(const SettingsChanges & changes, SettingSource source) { SharedLockGuard lock(mutex); - getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(settings, changes, source); - doSettingsSanityCheckClamp(settings, getLogger("SettingsSanity")); + getSettingsConstraintsAndCurrentProfilesWithLock()->constraints.check(*settings, changes, source); + doSettingsSanityCheckClamp(*settings, getLogger("SettingsSanity")); } void Context::checkSettingsConstraints(SettingsChanges & changes, SettingSource source) @@ -2341,7 +2450,7 @@ void Context::resetSettingsToDefaultValue(const std::vector & names) { std::lock_guard lock(mutex); for (const String & name: names) - settings.setDefaultValue(name); + settings->setDefaultValue(name); } std::shared_ptr Context::getSettingsConstraintsAndCurrentProfilesWithLock() const @@ -2561,14 +2670,14 @@ void Context::makeQueryContextForMerge(const MergeTreeSettings & merge_tree_sett { makeQueryContext(); classifier.reset(); // It is assumed that there are no active queries running using this classifier, otherwise this will lead to crashes - settings.workload = merge_tree_settings.merge_workload.value.empty() ? getMergeWorkload() : merge_tree_settings.merge_workload; + settings->workload = merge_tree_settings.merge_workload.value.empty() ? getMergeWorkload() : merge_tree_settings.merge_workload; } void Context::makeQueryContextForMutate(const MergeTreeSettings & merge_tree_settings) { makeQueryContext(); classifier.reset(); // It is assumed that there are no active queries running using this classifier, otherwise this will lead to crashes - settings.workload = merge_tree_settings.mutation_workload.value.empty() ? getMutationWorkload() : merge_tree_settings.mutation_workload; + settings->workload = merge_tree_settings.mutation_workload.value.empty() ? getMutationWorkload() : merge_tree_settings.mutation_workload; } void Context::makeSessionContext() @@ -3402,8 +3511,6 @@ zkutil::ZooKeeperPtr Context::getZooKeeper() const const auto & config = shared->zookeeper_config ? *shared->zookeeper_config : getConfigRef(); if (!shared->zookeeper) shared->zookeeper = zkutil::ZooKeeper::create(config, zkutil::getZooKeeperConfigName(config), getZooKeeperLog()); - else if (shared->zookeeper->hasReachedDeadline()) - shared->zookeeper->finalize("ZooKeeper session has reached its deadline"); if (shared->zookeeper->expired()) { @@ -3853,7 +3960,7 @@ void Context::reloadClusterConfig() const } const auto & config = cluster_config ? *cluster_config : getConfigRef(); - auto new_clusters = std::make_shared(config, settings, getMacros()); + auto new_clusters = std::make_shared(config, *settings, getMacros()); { std::lock_guard lock(shared->clusters_mutex); @@ -3888,7 +3995,7 @@ std::shared_ptr Context::getClustersImpl(std::lock_guard & if (!shared->clusters) { const auto & config = shared->clusters_config ? *shared->clusters_config : getConfigRef(); - shared->clusters = std::make_shared(config, settings, getMacros()); + shared->clusters = std::make_shared(config, *settings, getMacros()); } return shared->clusters; @@ -3920,9 +4027,9 @@ void Context::setClustersConfig(const ConfigurationPtr & config, bool enable_dis shared->clusters_config = config; if (!shared->clusters) - shared->clusters = std::make_shared(*shared->clusters_config, settings, getMacros(), config_name); + shared->clusters = std::make_shared(*shared->clusters_config, *settings, getMacros(), config_name); else - shared->clusters->updateClusters(*shared->clusters_config, settings, config_name, old_clusters_config); + shared->clusters->updateClusters(*shared->clusters_config, *settings, config_name, old_clusters_config); ++shared->clusters_version; } @@ -4004,7 +4111,6 @@ std::shared_ptr Context::getQueryThreadLog() const std::shared_ptr Context::getQueryViewsLog() const { SharedLockGuard lock(shared->mutex); - if (!shared->system_logs) return {}; @@ -4135,7 +4241,7 @@ std::shared_ptr Context::getFilesystemCacheLog() const return shared->system_logs->filesystem_cache_log; } -std::shared_ptr Context::getS3QueueLog() const +std::shared_ptr Context::getS3QueueLog() const { SharedLockGuard lock(shared->mutex); if (!shared->system_logs) @@ -4144,6 +4250,15 @@ std::shared_ptr Context::getS3QueueLog() const return shared->system_logs->s3_queue_log; } +std::shared_ptr Context::getAzureQueueLog() const +{ + SharedLockGuard lock(shared->mutex); + if (!shared->system_logs) + return {}; + + return shared->system_logs->azure_queue_log; +} + std::shared_ptr Context::getFilesystemReadPrefetchesLog() const { SharedLockGuard lock(shared->mutex); @@ -4175,7 +4290,7 @@ std::shared_ptr Context::getBackupLog() const std::shared_ptr Context::getBlobStorageLog() const { - bool enable_blob_storage_log = settings.enable_blob_storage_log; + bool enable_blob_storage_log = settings->enable_blob_storage_log; if (hasQueryContext()) enable_blob_storage_log = getQueryContext()->getSettingsRef().enable_blob_storage_log; @@ -4616,8 +4731,8 @@ void Context::setDefaultProfiles(const Poco::Util::AbstractConfiguration & confi shared->system_profile_name = config.getString("system_profile", shared->default_profile_name); setCurrentProfile(shared->system_profile_name); - applySettingsQuirks(settings, getLogger("SettingsQuirks")); - doSettingsSanityCheckClamp(settings, getLogger("SettingsSanity")); + applySettingsQuirks(*settings, getLogger("SettingsQuirks")); + doSettingsSanityCheckClamp(*settings, getLogger("SettingsSanity")); shared->buffer_profile_name = config.getString("buffer_profile", shared->system_profile_name); buffer_context = Context::createCopy(shared_from_this()); @@ -4886,13 +5001,6 @@ void Context::setConnectionClientVersion(UInt64 client_version_major, UInt64 cli client_info.connection_tcp_protocol_version = client_tcp_protocol_version; } -void Context::setReplicaInfo(bool collaborate_with_initiator, size_t all_replicas_count, size_t number_of_current_replica) -{ - client_info.collaborate_with_initiator = collaborate_with_initiator; - client_info.count_participating_replicas = all_replicas_count; - client_info.number_of_current_replica = number_of_current_replica; -} - void Context::increaseDistributedDepth() { ++client_info.distributed_depth; @@ -5186,11 +5294,11 @@ AsynchronousInsertQueue * Context::tryGetAsynchronousInsertQueue() const void Context::setAsynchronousInsertQueue(const std::shared_ptr & ptr) { - AsynchronousInsertQueue::validateSettings(settings, getLogger("Context")); + AsynchronousInsertQueue::validateSettings(*settings, getLogger("Context")); SharedLockGuard lock(shared->mutex); - if (std::chrono::milliseconds(settings.async_insert_poll_timeout_ms) == std::chrono::milliseconds::zero()) + if (std::chrono::milliseconds(settings->async_insert_poll_timeout_ms) == std::chrono::milliseconds::zero()) throw Exception(ErrorCodes::INVALID_SETTING_VALUE, "Setting async_insert_poll_timeout_ms can't be zero"); shared->async_insert_queue = ptr; @@ -5337,69 +5445,69 @@ ReadSettings Context::getReadSettings() const { ReadSettings res; - std::string_view read_method_str = settings.local_filesystem_read_method.value; + std::string_view read_method_str = settings->local_filesystem_read_method.value; if (auto opt_method = magic_enum::enum_cast(read_method_str)) res.local_fs_method = *opt_method; else throw Exception(ErrorCodes::UNKNOWN_READ_METHOD, "Unknown read method '{}' for local filesystem", read_method_str); - read_method_str = settings.remote_filesystem_read_method.value; + read_method_str = settings->remote_filesystem_read_method.value; if (auto opt_method = magic_enum::enum_cast(read_method_str)) res.remote_fs_method = *opt_method; else throw Exception(ErrorCodes::UNKNOWN_READ_METHOD, "Unknown read method '{}' for remote filesystem", read_method_str); - res.local_fs_prefetch = settings.local_filesystem_read_prefetch; - res.remote_fs_prefetch = settings.remote_filesystem_read_prefetch; + res.local_fs_prefetch = settings->local_filesystem_read_prefetch; + res.remote_fs_prefetch = settings->remote_filesystem_read_prefetch; - res.load_marks_asynchronously = settings.load_marks_asynchronously; + res.load_marks_asynchronously = settings->load_marks_asynchronously; - res.enable_filesystem_read_prefetches_log = settings.enable_filesystem_read_prefetches_log; + res.enable_filesystem_read_prefetches_log = settings->enable_filesystem_read_prefetches_log; - res.remote_fs_read_max_backoff_ms = settings.remote_fs_read_max_backoff_ms; - res.remote_fs_read_backoff_max_tries = settings.remote_fs_read_backoff_max_tries; - res.enable_filesystem_cache = settings.enable_filesystem_cache; - res.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = settings.read_from_filesystem_cache_if_exists_otherwise_bypass_cache; - res.enable_filesystem_cache_log = settings.enable_filesystem_cache_log; - res.filesystem_cache_segments_batch_size = settings.filesystem_cache_segments_batch_size; - res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds; + res.remote_fs_read_max_backoff_ms = settings->remote_fs_read_max_backoff_ms; + res.remote_fs_read_backoff_max_tries = settings->remote_fs_read_backoff_max_tries; + res.enable_filesystem_cache = settings->enable_filesystem_cache; + res.read_from_filesystem_cache_if_exists_otherwise_bypass_cache = settings->read_from_filesystem_cache_if_exists_otherwise_bypass_cache; + res.enable_filesystem_cache_log = settings->enable_filesystem_cache_log; + res.filesystem_cache_segments_batch_size = settings->filesystem_cache_segments_batch_size; + res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings->filesystem_cache_reserve_space_wait_lock_timeout_milliseconds; - res.filesystem_cache_max_download_size = settings.filesystem_cache_max_download_size; - res.skip_download_if_exceeds_query_cache = settings.skip_download_if_exceeds_query_cache; + res.filesystem_cache_max_download_size = settings->filesystem_cache_max_download_size; + res.skip_download_if_exceeds_query_cache = settings->skip_download_if_exceeds_query_cache; res.page_cache = getPageCache(); - res.use_page_cache_for_disks_without_file_cache = settings.use_page_cache_for_disks_without_file_cache; - res.read_from_page_cache_if_exists_otherwise_bypass_cache = settings.read_from_page_cache_if_exists_otherwise_bypass_cache; - res.page_cache_inject_eviction = settings.page_cache_inject_eviction; + res.use_page_cache_for_disks_without_file_cache = settings->use_page_cache_for_disks_without_file_cache; + res.read_from_page_cache_if_exists_otherwise_bypass_cache = settings->read_from_page_cache_if_exists_otherwise_bypass_cache; + res.page_cache_inject_eviction = settings->page_cache_inject_eviction; - res.remote_read_min_bytes_for_seek = settings.remote_read_min_bytes_for_seek; + res.remote_read_min_bytes_for_seek = settings->remote_read_min_bytes_for_seek; /// Zero read buffer will not make progress. - if (!settings.max_read_buffer_size) + if (!settings->max_read_buffer_size) { throw Exception(ErrorCodes::INVALID_SETTING_VALUE, - "Invalid value '{}' for max_read_buffer_size", settings.max_read_buffer_size); + "Invalid value '{}' for max_read_buffer_size", settings->max_read_buffer_size); } res.local_fs_buffer_size - = settings.max_read_buffer_size_local_fs ? settings.max_read_buffer_size_local_fs : settings.max_read_buffer_size; + = settings->max_read_buffer_size_local_fs ? settings->max_read_buffer_size_local_fs : settings->max_read_buffer_size; res.remote_fs_buffer_size - = settings.max_read_buffer_size_remote_fs ? settings.max_read_buffer_size_remote_fs : settings.max_read_buffer_size; - res.prefetch_buffer_size = settings.prefetch_buffer_size; - res.direct_io_threshold = settings.min_bytes_to_use_direct_io; - res.mmap_threshold = settings.min_bytes_to_use_mmap_io; - res.priority = Priority{settings.read_priority}; + = settings->max_read_buffer_size_remote_fs ? settings->max_read_buffer_size_remote_fs : settings->max_read_buffer_size; + res.prefetch_buffer_size = settings->prefetch_buffer_size; + res.direct_io_threshold = settings->min_bytes_to_use_direct_io; + res.mmap_threshold = settings->min_bytes_to_use_mmap_io; + res.priority = Priority{settings->read_priority}; res.remote_throttler = getRemoteReadThrottler(); res.local_throttler = getLocalReadThrottler(); - res.http_max_tries = settings.http_max_tries; - res.http_retry_initial_backoff_ms = settings.http_retry_initial_backoff_ms; - res.http_retry_max_backoff_ms = settings.http_retry_max_backoff_ms; - res.http_skip_not_found_url_for_globs = settings.http_skip_not_found_url_for_globs; - res.http_make_head_request = settings.http_make_head_request; + res.http_max_tries = settings->http_max_tries; + res.http_retry_initial_backoff_ms = settings->http_retry_initial_backoff_ms; + res.http_retry_max_backoff_ms = settings->http_retry_max_backoff_ms; + res.http_skip_not_found_url_for_globs = settings->http_skip_not_found_url_for_globs; + res.http_make_head_request = settings->http_make_head_request; res.mmap_cache = getMMappedFileCache().get(); @@ -5410,13 +5518,13 @@ WriteSettings Context::getWriteSettings() const { WriteSettings res; - res.enable_filesystem_cache_on_write_operations = settings.enable_filesystem_cache_on_write_operations; - res.enable_filesystem_cache_log = settings.enable_filesystem_cache_log; - res.throw_on_error_from_cache = settings.throw_on_error_from_cache_on_write_operations; - res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds; + res.enable_filesystem_cache_on_write_operations = settings->enable_filesystem_cache_on_write_operations; + res.enable_filesystem_cache_log = settings->enable_filesystem_cache_log; + res.throw_on_error_from_cache = settings->throw_on_error_from_cache_on_write_operations; + res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings->filesystem_cache_reserve_space_wait_lock_timeout_milliseconds; - res.s3_allow_parallel_part_upload = settings.s3_allow_parallel_part_upload; - res.azure_allow_parallel_part_upload = settings.azure_allow_parallel_part_upload; + res.s3_allow_parallel_part_upload = settings->s3_allow_parallel_part_upload; + res.azure_allow_parallel_part_upload = settings->azure_allow_parallel_part_upload; res.remote_throttler = getRemoteWriteThrottler(); res.local_throttler = getLocalWriteThrottler(); @@ -5462,10 +5570,37 @@ bool Context::canUseParallelReplicasOnFollower() const return canUseTaskBasedParallelReplicas() && getClientInfo().collaborate_with_initiator; } -bool Context::canUseParallelReplicasCustomKey(const Cluster & cluster) const +bool Context::canUseParallelReplicasCustomKey() const { - return settings.max_parallel_replicas > 1 && getParallelReplicasMode() == Context::ParallelReplicasMode::CUSTOM_KEY - && cluster.getShardCount() == 1 && cluster.getShardsInfo()[0].getAllNodeCount() > 1; + return settings->max_parallel_replicas > 1 && getParallelReplicasMode() == Context::ParallelReplicasMode::CUSTOM_KEY; +} + +bool Context::canUseParallelReplicasCustomKeyForCluster(const Cluster & cluster) const +{ + return canUseParallelReplicasCustomKey() && cluster.getShardCount() == 1 && cluster.getShardsInfo()[0].getAllNodeCount() > 1; +} + +bool Context::canUseOffsetParallelReplicas() const +{ + return offset_parallel_replicas_enabled && settings->max_parallel_replicas > 1 + && getParallelReplicasMode() != Context::ParallelReplicasMode::READ_TASKS; +} + +void Context::disableOffsetParallelReplicas() +{ + offset_parallel_replicas_enabled = false; +} + +ClusterPtr Context::getClusterForParallelReplicas() const +{ + /// check cluster for parallel replicas + if (settings->cluster_for_parallel_replicas.value.empty()) + throw Exception( + ErrorCodes::CLUSTER_DOESNT_EXIST, + "Reading in parallel from replicas is enabled but cluster to execute query is not provided. Please set " + "'cluster_for_parallel_replicas' setting"); + + return getCluster(settings->cluster_for_parallel_replicas); } void Context::setPreparedSetsCache(const PreparedSetsCachePtr & cache) @@ -5493,4 +5628,39 @@ const ServerSettings & Context::getServerSettings() const return shared->server_settings; } +uint64_t HTTPContext::getMaxHstsAge() const +{ + return context->getSettingsRef().hsts_max_age; +} + + uint64_t HTTPContext::getMaxUriSize() const +{ + return context->getSettingsRef().http_max_uri_size; +} + +uint64_t HTTPContext::getMaxFields() const +{ +return context->getSettingsRef().http_max_fields; +} + +uint64_t HTTPContext::getMaxFieldNameSize() const +{ +return context->getSettingsRef().http_max_field_name_size; +} + +uint64_t HTTPContext::getMaxFieldValueSize() const +{ +return context->getSettingsRef().http_max_field_value_size; +} + +Poco::Timespan HTTPContext::getReceiveTimeout() const +{ +return context->getSettingsRef().http_receive_timeout; +} + +Poco::Timespan HTTPContext::getSendTimeout() const +{ +return context->getSettingsRef().http_send_timeout; +} + } diff --git a/src/Interpreters/Context.h b/src/Interpreters/Context.h index d1874b4902f..cb553d07513 100644 --- a/src/Interpreters/Context.h +++ b/src/Interpreters/Context.h @@ -1,7 +1,5 @@ #pragma once -#ifndef CLICKHOUSE_KEEPER_STANDALONE_BUILD - #include #include #include @@ -13,9 +11,11 @@ #include #include #include -#include #include +#include #include +#include +#include #include #include #include @@ -63,6 +63,7 @@ class AccessFlags; struct AccessRightsElement; class AccessRightsElements; enum class RowPolicyFilterType : uint8_t; +struct RolesOrUsersSet; class EmbeddedDictionaries; class ExternalDictionariesLoader; class ExternalUserDefinedExecutableFunctionsLoader; @@ -107,7 +108,7 @@ class TransactionsInfoLog; class ProcessorsProfileLog; class FilesystemCacheLog; class FilesystemReadPrefetchesLog; -class S3QueueLog; +class ObjectStorageQueueLog; class AsynchronousInsertLog; class BackupLog; class BlobStorageLog; @@ -130,6 +131,7 @@ class ShellCommand; class ICompressionCodec; class AccessControl; class GSSAcceptorContext; +struct Settings; struct SettingsConstraintsAndProfileIDs; class SettingsProfileElements; class RemoteHostFilter; @@ -152,6 +154,8 @@ class AsyncLoader; struct TemporaryTableHolder; using TemporaryTablesMapping = std::map>; +using ClusterPtr = std::shared_ptr; + class LoadTask; using LoadTaskPtr = std::shared_ptr; using LoadTaskPtrs = std::vector; @@ -275,7 +279,7 @@ protected: mutable std::shared_ptr access; mutable bool need_recalculate_access = true; String current_database; - Settings settings; /// Setting for query execution. + std::unique_ptr settings{}; /// Setting for query execution. using ProgressCallback = std::function; ProgressCallback progress_callback; /// Callback for tracking progress of query execution. @@ -459,6 +463,11 @@ protected: /// mutation tasks of one mutation executed against different parts of the same table. PreparedSetsCachePtr prepared_sets_cache; + /// this is a mode of parallel replicas where we set parallel_replicas_count and parallel_replicas_offset + /// and generate specific filters on the replicas (e.g. when using parallel replicas with sample key) + /// if we already use a different mode of parallel replicas we want to disable this mode + bool offset_parallel_replicas_enabled = true; + public: /// Some counters for current query execution. /// Most of them are workarounds and should be removed in the future. @@ -602,13 +611,15 @@ public: /// Sets the current user assuming that he/she is already authenticated. /// WARNING: This function doesn't check password! - void setUser(const UUID & user_id_, const std::optional> & current_roles_ = {}); + void setUser(const UUID & user_id_); UserPtr getUser() const; std::optional getUserID() const; String getUserName() const; - void setCurrentRoles(const std::vector & current_roles_); + void setCurrentRoles(const Strings & new_current_roles, bool check_grants = true); + void setCurrentRoles(const std::vector & new_current_roles, bool check_grants = true); + void setCurrentRoles(const RolesOrUsersSet & new_current_roles, bool check_grants = true); void setCurrentRolesDefault(); std::vector getCurrentRoles() const; std::vector getEnabledRoles() const; @@ -688,7 +699,6 @@ public: void setInitialQueryStartTime(std::chrono::time_point initial_query_start_time); void setQuotaClientKey(const String & quota_key); void setConnectionClientVersion(UInt64 client_version_major, UInt64 client_version_minor, UInt64 client_version_patch, unsigned client_tcp_protocol_version); - void setReplicaInfo(bool collaborate_with_initiator, size_t all_replicas_count, size_t number_of_current_replica); void increaseDistributedDepth(); const OpenTelemetry::TracingContext & getClientTraceContext() const { return client_info.client_trace_context; } OpenTelemetry::TracingContext & getClientTraceContext() { return client_info.client_trace_context; } @@ -825,6 +835,7 @@ public: /// Set settings by name. void setSetting(std::string_view name, const String & value); void setSetting(std::string_view name, const Field & value); + void setServerSetting(std::string_view name, const Field & value); void applySettingChange(const SettingChange & change); void applySettingsChanges(const SettingsChanges & changes); @@ -943,7 +954,7 @@ public: void makeSessionContext(); void makeGlobalContext(); - const Settings & getSettingsRef() const { return settings; } + const Settings & getSettingsRef() const { return *settings; } void setProgressCallback(ProgressCallback callback); /// Used in executeQuery() to pass it to the QueryPipeline. @@ -1133,7 +1144,8 @@ public: std::shared_ptr getTransactionsInfoLog() const; std::shared_ptr getProcessorsProfileLog() const; std::shared_ptr getFilesystemCacheLog() const; - std::shared_ptr getS3QueueLog() const; + std::shared_ptr getS3QueueLog() const; + std::shared_ptr getAzureQueueLog() const; std::shared_ptr getFilesystemReadPrefetchesLog() const; std::shared_ptr getAsynchronousInsertLog() const; std::shared_ptr getBackupLog() const; @@ -1309,7 +1321,13 @@ public: bool canUseTaskBasedParallelReplicas() const; bool canUseParallelReplicasOnInitiator() const; bool canUseParallelReplicasOnFollower() const; - bool canUseParallelReplicasCustomKey(const Cluster & cluster) const; + bool canUseParallelReplicasCustomKey() const; + bool canUseParallelReplicasCustomKeyForCluster(const Cluster & cluster) const; + bool canUseOffsetParallelReplicas() const; + + void disableOffsetParallelReplicas(); + + ClusterPtr getClusterForParallelReplicas() const; enum class ParallelReplicasMode : uint8_t { @@ -1334,7 +1352,7 @@ private: void setCurrentProfilesWithLock(const SettingsProfilesInfo & profiles_info, bool check_constraints, const std::lock_guard & lock); - void setCurrentRolesWithLock(const std::vector & current_roles_, const std::lock_guard & lock); + void setCurrentRolesWithLock(const std::vector & new_current_roles, const std::lock_guard & lock); void setSettingWithLock(std::string_view name, const String & value, const std::lock_guard & lock); @@ -1367,6 +1385,7 @@ private: void initGlobal(); void setUserID(const UUID & user_id_); + void setCurrentRolesImpl(const std::vector & new_current_roles, bool throw_if_not_granted, bool skip_if_not_granted, const std::shared_ptr & user); template void checkAccessImpl(const Args &... args) const; @@ -1411,48 +1430,21 @@ struct HTTPContext : public IHTTPContext : context(Context::createCopy(context_)) {} - uint64_t getMaxHstsAge() const override - { - return context->getSettingsRef().hsts_max_age; - } + uint64_t getMaxHstsAge() const override; - uint64_t getMaxUriSize() const override - { - return context->getSettingsRef().http_max_uri_size; - } + uint64_t getMaxUriSize() const override; - uint64_t getMaxFields() const override - { - return context->getSettingsRef().http_max_fields; - } + uint64_t getMaxFields() const override; - uint64_t getMaxFieldNameSize() const override - { - return context->getSettingsRef().http_max_field_name_size; - } + uint64_t getMaxFieldNameSize() const override; - uint64_t getMaxFieldValueSize() const override - { - return context->getSettingsRef().http_max_field_value_size; - } + uint64_t getMaxFieldValueSize() const override; - Poco::Timespan getReceiveTimeout() const override - { - return context->getSettingsRef().http_receive_timeout; - } + Poco::Timespan getReceiveTimeout() const override; - Poco::Timespan getSendTimeout() const override - { - return context->getSettingsRef().http_send_timeout; - } + Poco::Timespan getSendTimeout() const override; ContextPtr context; }; } - -#else - -#include - -#endif diff --git a/src/Interpreters/DDLTask.cpp b/src/Interpreters/DDLTask.cpp index 6c346836ed8..fa197d59c13 100644 --- a/src/Interpreters/DDLTask.cpp +++ b/src/Interpreters/DDLTask.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/DDLTask.h b/src/Interpreters/DDLTask.h index 0b0460b26c8..515a35d8671 100644 --- a/src/Interpreters/DDLTask.h +++ b/src/Interpreters/DDLTask.h @@ -3,6 +3,7 @@ #include #include #include +#include #include #include diff --git a/src/Interpreters/DDLWorker.cpp b/src/Interpreters/DDLWorker.cpp index 5639eed552e..697fd0f406b 100644 --- a/src/Interpreters/DDLWorker.cpp +++ b/src/Interpreters/DDLWorker.cpp @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/DatabaseCatalog.cpp b/src/Interpreters/DatabaseCatalog.cpp index 0f4c8cc26a6..bb2dd158710 100644 --- a/src/Interpreters/DatabaseCatalog.cpp +++ b/src/Interpreters/DatabaseCatalog.cpp @@ -1,5 +1,7 @@ +#include #include #include +#include #include #include #include @@ -26,7 +28,9 @@ #include #include #include +#include +#include #include #include "config.h" @@ -63,6 +67,7 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; extern const int HAVE_DEPENDENT_OBJECTS; extern const int UNFINISHED; + extern const int INFINITE_LOOP; } class DatabaseNameHints : public IHints<> @@ -189,6 +194,7 @@ void DatabaseCatalog::initializeAndLoadTemporaryDatabase() unused_dir_rm_timeout_sec = getContext()->getConfigRef().getInt64("database_catalog_unused_dir_rm_timeout_sec", unused_dir_rm_timeout_sec); unused_dir_cleanup_period_sec = getContext()->getConfigRef().getInt64("database_catalog_unused_dir_cleanup_period_sec", unused_dir_cleanup_period_sec); drop_error_cooldown_sec = getContext()->getConfigRef().getInt64("database_catalog_drop_error_cooldown_sec", drop_error_cooldown_sec); + drop_table_concurrency = getContext()->getConfigRef().getInt64("database_catalog_drop_table_concurrency", drop_table_concurrency); auto db_for_temporary_and_external_tables = std::make_shared(TEMPORARY_DATABASE, getContext()); attachDatabase(TEMPORARY_DATABASE, db_for_temporary_and_external_tables); @@ -273,10 +279,12 @@ void DatabaseCatalog::shutdownImpl() database->shutdown(); } + TablesMarkedAsDropped tables_marked_dropped_to_destroy; { std::lock_guard lock(tables_marked_dropped_mutex); - tables_marked_dropped.clear(); + tables_marked_dropped.swap(tables_marked_dropped_to_destroy); } + tables_marked_dropped_to_destroy.clear(); std::lock_guard lock(databases_mutex); for (const auto & db : databases) @@ -1140,7 +1148,7 @@ void DatabaseCatalog::enqueueDroppedTableCleanup(StorageID table_id, StoragePtr (*drop_task)->schedule(); } -void DatabaseCatalog::dequeueDroppedTableCleanup(StorageID table_id) +void DatabaseCatalog::undropTable(StorageID table_id) { String latest_metadata_dropped_path; TableMarkedAsDropped dropped_table; @@ -1194,7 +1202,7 @@ void DatabaseCatalog::dequeueDroppedTableCleanup(StorageID table_id) /// It's unsafe to create another instance while the old one exists /// We cannot wait on shared_ptr's refcount, so it's busy wait - while (!dropped_table.table.unique()) + while (!isSharedPtrUnique(dropped_table.table)) std::this_thread::sleep_for(std::chrono::milliseconds(100)); dropped_table.table.reset(); @@ -1215,91 +1223,162 @@ void DatabaseCatalog::dequeueDroppedTableCleanup(StorageID table_id) LOG_INFO(log, "Table {} was successfully undropped.", dropped_table.table_id.getNameForLogs()); } +std::tuple DatabaseCatalog::getDroppedTablesCountAndInuseCount() +{ + std::lock_guard lock(tables_marked_dropped_mutex); + + size_t in_use_count = 0; + for (const auto & item : tables_marked_dropped) + { + bool in_use = item.table && !isSharedPtrUnique(item.table); + in_use_count += in_use; + } + return {tables_marked_dropped.size(), in_use_count}; +} + +time_t DatabaseCatalog::getMinDropTime() +{ + time_t min_drop_time = std::numeric_limits::max(); + for (const auto & item : tables_marked_dropped) + { + min_drop_time = std::min(min_drop_time, item.drop_time); + } + return min_drop_time; +} + +std::vector DatabaseCatalog::getTablesToDrop() +{ + time_t current_time = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now()); + decltype(getTablesToDrop()) result; + + std::lock_guard lock(tables_marked_dropped_mutex); + + for (auto it = tables_marked_dropped.begin(); it != tables_marked_dropped.end(); ++it) + { + bool in_use = it->table && !isSharedPtrUnique(it->table); + bool old_enough = it->drop_time <= current_time; + if (!in_use && old_enough) + result.emplace_back(it); + } + + return result; +} + +void DatabaseCatalog::rescheduleDropTableTask() +{ + std::lock_guard lock(tables_marked_dropped_mutex); + + if (tables_marked_dropped.empty()) + return; + + if (first_async_drop_in_queue != tables_marked_dropped.begin()) + { + (*drop_task)->scheduleAfter(0); + return; + } + + time_t current_time = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now()); + auto min_drop_time = getMinDropTime(); + time_t schedule_after_ms = min_drop_time > current_time ? (min_drop_time - current_time) * 1000 : 0; + + (*drop_task)->scheduleAfter(schedule_after_ms); +} + +void DatabaseCatalog::dropTablesParallel(std::vector tables_to_drop) +{ + if (tables_to_drop.empty()) + return; + + ThreadPool pool( + CurrentMetrics::DatabaseCatalogThreads, + CurrentMetrics::DatabaseCatalogThreadsActive, + CurrentMetrics::DatabaseCatalogThreadsScheduled, + /* max_threads */drop_table_concurrency, + /* max_free_threads */0, + /* queue_size */tables_to_drop.size()); + + for (const auto & item : tables_to_drop) + { + auto job = [&, table_iterator = item] () + { + try + { + dropTableFinally(*table_iterator); + + TableMarkedAsDropped table_to_delete_without_lock; + { + std::lock_guard lock(tables_marked_dropped_mutex); + + if (first_async_drop_in_queue == table_iterator) + ++first_async_drop_in_queue; + + [[maybe_unused]] auto removed = tables_marked_dropped_ids.erase(table_iterator->table_id.uuid); + chassert(removed); + + table_to_delete_without_lock = std::move(*table_iterator); + tables_marked_dropped.erase(table_iterator); + + wait_table_finally_dropped.notify_all(); + } + } + catch (...) + { + tryLogCurrentException(log, "Cannot drop table " + table_iterator->table_id.getNameForLogs() + + ". Will retry later."); + { + std::lock_guard lock(tables_marked_dropped_mutex); + + if (first_async_drop_in_queue == table_iterator) + ++first_async_drop_in_queue; + + tables_marked_dropped.splice(tables_marked_dropped.end(), tables_marked_dropped, table_iterator); + table_iterator->drop_time = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now()) + drop_error_cooldown_sec; + + if (first_async_drop_in_queue == tables_marked_dropped.end()) + --first_async_drop_in_queue; + } + } + }; + + try + { + pool.scheduleOrThrowOnError(std::move(job)); + } + catch (...) + { + tryLogCurrentException(log, "Cannot drop tables. Will retry later."); + break; + } + } + + try + { + pool.wait(); + } + catch (...) + { + tryLogCurrentException(log, "Cannot drop tables. Will retry later."); + } +} + void DatabaseCatalog::dropTableDataTask() { /// Background task that removes data of tables which were marked as dropped by Atomic databases. /// Table can be removed when it's not used by queries and drop_delay_sec elapsed since it was marked as dropped. - bool need_reschedule = true; - /// Default reschedule time for the case when we are waiting for reference count to become 1. - size_t schedule_after_ms = reschedule_time_ms; - TableMarkedAsDropped table; - try + auto [drop_tables_count, drop_tables_in_use_count] = getDroppedTablesCountAndInuseCount(); + + auto tables_to_drop = getTablesToDrop(); + + if (!tables_to_drop.empty()) { - std::lock_guard lock(tables_marked_dropped_mutex); - if (tables_marked_dropped.empty()) - return; - time_t current_time = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now()); - time_t min_drop_time = std::numeric_limits::max(); - size_t tables_in_use_count = 0; - auto it = std::find_if(tables_marked_dropped.begin(), tables_marked_dropped.end(), [&](const auto & elem) - { - bool not_in_use = !elem.table || elem.table.unique(); - bool old_enough = elem.drop_time <= current_time; - min_drop_time = std::min(min_drop_time, elem.drop_time); - tables_in_use_count += !not_in_use; - return not_in_use && old_enough; - }); - if (it != tables_marked_dropped.end()) - { - table = std::move(*it); - LOG_INFO(log, "Have {} tables in drop queue ({} of them are in use), will try drop {}", - tables_marked_dropped.size(), tables_in_use_count, table.table_id.getNameForLogs()); - if (first_async_drop_in_queue == it) - ++first_async_drop_in_queue; - tables_marked_dropped.erase(it); - /// Schedule the task as soon as possible, while there are suitable tables to drop. - schedule_after_ms = 0; - } - else if (current_time < min_drop_time) - { - /// We are waiting for drop_delay_sec to exceed, no sense to wakeup until min_drop_time. - /// If new table is added to the queue with ignore_delay flag, schedule() is called to wakeup the task earlier. - schedule_after_ms = (min_drop_time - current_time) * 1000; - LOG_TRACE(log, "Not found any suitable tables to drop, still have {} tables in drop queue ({} of them are in use). " - "Will check again after {} seconds", tables_marked_dropped.size(), tables_in_use_count, min_drop_time - current_time); - } - need_reschedule = !tables_marked_dropped.empty(); - } - catch (...) - { - tryLogCurrentException(log, __PRETTY_FUNCTION__); + LOG_INFO(log, "Have {} tables in drop queue ({} of them are in use), will try drop {} tables", + drop_tables_count, drop_tables_in_use_count, tables_to_drop.size()); + + dropTablesParallel(tables_to_drop); } - if (table.table_id) - { - try - { - dropTableFinally(table); - std::lock_guard lock(tables_marked_dropped_mutex); - [[maybe_unused]] auto removed = tables_marked_dropped_ids.erase(table.table_id.uuid); - assert(removed); - } - catch (...) - { - tryLogCurrentException(log, "Cannot drop table " + table.table_id.getNameForLogs() + - ". Will retry later."); - { - table.drop_time = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now()) + drop_error_cooldown_sec; - std::lock_guard lock(tables_marked_dropped_mutex); - tables_marked_dropped.emplace_back(std::move(table)); - if (first_async_drop_in_queue == tables_marked_dropped.end()) - --first_async_drop_in_queue; - /// If list of dropped tables was empty, schedule a task to retry deletion. - if (tables_marked_dropped.size() == 1) - { - need_reschedule = true; - schedule_after_ms = drop_error_cooldown_sec * 1000; - } - } - } - - wait_table_finally_dropped.notify_all(); - } - - /// Do not schedule a task if there is no tables to drop - if (need_reschedule) - (*drop_task)->scheduleAfter(schedule_after_ms); + rescheduleDropTableTask(); } void DatabaseCatalog::dropTableFinally(const TableMarkedAsDropped & table) @@ -1473,6 +1552,114 @@ void DatabaseCatalog::checkTableCanBeRemovedOrRenamedUnlocked( removing_table, fmt::join(from_other_databases, ", ")); } +void DatabaseCatalog::checkTableCanBeAddedWithNoCyclicDependencies( + const QualifiedTableName & table_name, + const TableNamesSet & new_referential_dependencies, + const TableNamesSet & new_loading_dependencies) +{ + std::lock_guard lock{databases_mutex}; + + StorageID table_id = StorageID{table_name}; + + auto check = [&](TablesDependencyGraph & dependencies, const TableNamesSet & new_dependencies) + { + auto old_dependencies = dependencies.removeDependencies(table_id); + dependencies.addDependencies(table_name, new_dependencies); + auto restore_dependencies = [&]() + { + dependencies.removeDependencies(table_id); + if (!old_dependencies.empty()) + dependencies.addDependencies(table_id, old_dependencies); + }; + + if (dependencies.hasCyclicDependencies()) + { + auto cyclic_dependencies_description = dependencies.describeCyclicDependencies(); + restore_dependencies(); + throw Exception( + ErrorCodes::INFINITE_LOOP, + "Cannot add dependencies for '{}', because it will lead to cyclic dependencies: {}", + table_name.getFullName(), + cyclic_dependencies_description); + } + + restore_dependencies(); + }; + + check(referential_dependencies, new_referential_dependencies); + check(loading_dependencies, new_loading_dependencies); +} + +void DatabaseCatalog::checkTableCanBeRenamedWithNoCyclicDependencies(const StorageID & from_table_id, const StorageID & to_table_id) +{ + std::lock_guard lock{databases_mutex}; + + auto check = [&](TablesDependencyGraph & dependencies) + { + auto old_dependencies = dependencies.removeDependencies(from_table_id); + dependencies.addDependencies(to_table_id, old_dependencies); + auto restore_dependencies = [&]() + { + dependencies.removeDependencies(to_table_id); + dependencies.addDependencies(from_table_id, old_dependencies); + }; + + if (dependencies.hasCyclicDependencies()) + { + auto cyclic_dependencies_description = dependencies.describeCyclicDependencies(); + restore_dependencies(); + throw Exception( + ErrorCodes::INFINITE_LOOP, + "Cannot rename '{}' to '{}', because it will lead to cyclic dependencies: {}", + from_table_id.getFullTableName(), + to_table_id.getFullTableName(), + cyclic_dependencies_description); + } + + restore_dependencies(); + }; + + check(referential_dependencies); + check(loading_dependencies); +} + +void DatabaseCatalog::checkTablesCanBeExchangedWithNoCyclicDependencies(const StorageID & table_id_1, const StorageID & table_id_2) +{ + std::lock_guard lock{databases_mutex}; + + auto check = [&](TablesDependencyGraph & dependencies) + { + auto old_dependencies_1 = dependencies.removeDependencies(table_id_1); + auto old_dependencies_2 = dependencies.removeDependencies(table_id_2); + dependencies.addDependencies(table_id_1, old_dependencies_2); + dependencies.addDependencies(table_id_2, old_dependencies_1); + auto restore_dependencies = [&]() + { + dependencies.removeDependencies(table_id_1); + dependencies.removeDependencies(table_id_2); + dependencies.addDependencies(table_id_1, old_dependencies_1); + dependencies.addDependencies(table_id_2, old_dependencies_2); + }; + + if (dependencies.hasCyclicDependencies()) + { + auto cyclic_dependencies_description = dependencies.describeCyclicDependencies(); + restore_dependencies(); + throw Exception( + ErrorCodes::INFINITE_LOOP, + "Cannot exchange '{}' and '{}', because it will lead to cyclic dependencies: {}", + table_id_1.getFullTableName(), + table_id_2.getFullTableName(), + cyclic_dependencies_description); + } + + restore_dependencies(); + }; + + check(referential_dependencies); + check(loading_dependencies); +} + void DatabaseCatalog::cleanupStoreDirectoryTask() { for (const auto & [disk_name, disk] : getContext()->getDisksMap()) diff --git a/src/Interpreters/DatabaseCatalog.h b/src/Interpreters/DatabaseCatalog.h index ac9c01e5dd7..23e38a6445e 100644 --- a/src/Interpreters/DatabaseCatalog.h +++ b/src/Interpreters/DatabaseCatalog.h @@ -225,7 +225,7 @@ public: String getPathForDroppedMetadata(const StorageID & table_id) const; String getPathForMetadata(const StorageID & table_id) const; void enqueueDroppedTableCleanup(StorageID table_id, StoragePtr table, String dropped_metadata_path, bool ignore_delay = false); - void dequeueDroppedTableCleanup(StorageID table_id); + void undropTable(StorageID table_id); void waitTableFinallyDropped(const UUID & uuid); @@ -245,6 +245,9 @@ public: void checkTableCanBeRemovedOrRenamed(const StorageID & table_id, bool check_referential_dependencies, bool check_loading_dependencies, bool is_drop_database = false) const; + void checkTableCanBeAddedWithNoCyclicDependencies(const QualifiedTableName & table_name, const TableNamesSet & new_referential_dependencies, const TableNamesSet & new_loading_dependencies); + void checkTableCanBeRenamedWithNoCyclicDependencies(const StorageID & from_table_id, const StorageID & to_table_id); + void checkTablesCanBeExchangedWithNoCyclicDependencies(const StorageID & table_id_1, const StorageID & table_id_2); struct TableMarkedAsDropped { @@ -293,6 +296,12 @@ private: void dropTableDataTask(); void dropTableFinally(const TableMarkedAsDropped & table); + time_t getMinDropTime() TSA_REQUIRES(tables_marked_dropped_mutex); + std::tuple getDroppedTablesCountAndInuseCount(); + std::vector getTablesToDrop(); + void dropTablesParallel(std::vector tables); + void rescheduleDropTableTask(); + void cleanupStoreDirectoryTask(); bool maybeRemoveDirectory(const String & disk_name, const DiskPtr & disk, const String & unused_dir); @@ -360,6 +369,9 @@ private: static constexpr time_t default_drop_error_cooldown_sec = 5; time_t drop_error_cooldown_sec = default_drop_error_cooldown_sec; + static constexpr size_t default_drop_table_concurrency = 10; + size_t drop_table_concurrency = default_drop_table_concurrency; + std::unique_ptr reload_disks_task; std::mutex reload_disks_mutex; std::set disks_to_reload; diff --git a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp index a70ff3c6c53..4bd1c47d5a0 100644 --- a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp +++ b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp @@ -2,6 +2,7 @@ #include #include +#include #include #include #include diff --git a/src/Interpreters/ExpressionActions.cpp b/src/Interpreters/ExpressionActions.cpp index 7f96c927d82..d832f568cb8 100644 --- a/src/Interpreters/ExpressionActions.cpp +++ b/src/Interpreters/ExpressionActions.cpp @@ -195,6 +195,10 @@ static void setLazyExecutionInfo( } lazy_execution_info.short_circuit_ancestors_info[parent].insert(indexes.begin(), indexes.end()); + /// After checking arguments_with_disabled_lazy_execution, if there is no relation with parent, + /// disable the current node. + if (indexes.empty()) + lazy_execution_info.can_be_lazy_executed = false; } else /// If lazy execution is disabled for one of parents, we should disable it for current node. @@ -292,9 +296,9 @@ static std::unordered_set processShortCircuitFunctions /// Firstly, find all short-circuit functions and get their settings. std::unordered_map short_circuit_nodes; - IFunctionBase::ShortCircuitSettings short_circuit_settings; for (const auto & node : nodes) { + IFunctionBase::ShortCircuitSettings short_circuit_settings; if (node.type == ActionsDAG::ActionType::FUNCTION && node.function_base->isShortCircuit(short_circuit_settings, node.children.size()) && !node.children.empty()) short_circuit_nodes[&node] = short_circuit_settings; } diff --git a/src/Interpreters/ExpressionActionsSettings.h b/src/Interpreters/ExpressionActionsSettings.h index 326de983748..384a61a54a2 100644 --- a/src/Interpreters/ExpressionActionsSettings.h +++ b/src/Interpreters/ExpressionActionsSettings.h @@ -1,7 +1,7 @@ #pragma once +#include #include -#include #include diff --git a/src/Interpreters/ExpressionAnalyzer.cpp b/src/Interpreters/ExpressionAnalyzer.cpp index 62cddd9caf7..16d0eb71278 100644 --- a/src/Interpreters/ExpressionAnalyzer.cpp +++ b/src/Interpreters/ExpressionAnalyzer.cpp @@ -24,7 +24,7 @@ #include #include #include -#include +#include #include #include #include @@ -986,7 +986,8 @@ static std::shared_ptr tryCreateJoin( const auto & settings = context->getSettings(); if (analyzed_join->allowParallelHashJoin()) - return std::make_shared(context, analyzed_join, settings.max_threads, right_sample_block); + return std::make_shared( + context, analyzed_join, settings.max_threads, right_sample_block, StatsCollectingParams{}); return std::make_shared(analyzed_join, right_sample_block); } diff --git a/src/Interpreters/ExternalDictionariesLoader.cpp b/src/Interpreters/ExternalDictionariesLoader.cpp index 1685c06d387..5f6d7e33927 100644 --- a/src/Interpreters/ExternalDictionariesLoader.cpp +++ b/src/Interpreters/ExternalDictionariesLoader.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include "config.h" diff --git a/src/Interpreters/ExternalLoader.h b/src/Interpreters/ExternalLoader.h index 49b5e68d821..6356a174a01 100644 --- a/src/Interpreters/ExternalLoader.h +++ b/src/Interpreters/ExternalLoader.h @@ -2,6 +2,7 @@ #include #include +#include #include #include #include diff --git a/src/Interpreters/FilesystemCacheLog.h b/src/Interpreters/FilesystemCacheLog.h index 27c616ff40c..6d92fbd7a43 100644 --- a/src/Interpreters/FilesystemCacheLog.h +++ b/src/Interpreters/FilesystemCacheLog.h @@ -2,7 +2,6 @@ #include #include -#include #include #include #include diff --git a/src/Interpreters/GlobalSubqueriesVisitor.h b/src/Interpreters/GlobalSubqueriesVisitor.h index 64b6eb5dce9..986134a6998 100644 --- a/src/Interpreters/GlobalSubqueriesVisitor.h +++ b/src/Interpreters/GlobalSubqueriesVisitor.h @@ -2,6 +2,7 @@ #include #include +#include #include #include #include diff --git a/src/Interpreters/GraceHashJoin.cpp b/src/Interpreters/GraceHashJoin.cpp index 4dd2f89b90a..a241d1cd258 100644 --- a/src/Interpreters/GraceHashJoin.cpp +++ b/src/Interpreters/GraceHashJoin.cpp @@ -3,13 +3,14 @@ #include #include #include -#include +#include #include #include #include #include #include #include +#include #include #include diff --git a/src/Interpreters/HashJoin.cpp b/src/Interpreters/HashJoin.cpp deleted file mode 100644 index 1c2a054b2a5..00000000000 --- a/src/Interpreters/HashJoin.cpp +++ /dev/null @@ -1,2877 +0,0 @@ -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - - -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include - -#include - -#include -#include -#include -#include -#include -#include "Core/Joins.h" -#include "Interpreters/TemporaryDataOnDisk.h" - -#include -#include - -namespace CurrentMetrics -{ - extern const Metric TemporaryFilesForJoin; -} - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int NOT_IMPLEMENTED; - extern const int NO_SUCH_COLUMN_IN_TABLE; - extern const int INCOMPATIBLE_TYPE_OF_JOIN; - extern const int UNSUPPORTED_JOIN_KEYS; - extern const int LOGICAL_ERROR; - extern const int SYNTAX_ERROR; - extern const int SET_SIZE_LIMIT_EXCEEDED; - extern const int TYPE_MISMATCH; - extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; - extern const int INVALID_JOIN_ON_EXPRESSION; -} - -namespace -{ - -struct NotProcessedCrossJoin : public ExtraBlock -{ - size_t left_position; - size_t right_block; - std::unique_ptr reader; -}; - - -Int64 getCurrentQueryMemoryUsage() -{ - /// Use query-level memory tracker - if (auto * memory_tracker_child = CurrentThread::getMemoryTracker()) - if (auto * memory_tracker = memory_tracker_child->getParent()) - return memory_tracker->get(); - return 0; -} - -} - -namespace JoinStuff -{ - /// for single disjunct - bool JoinUsedFlags::getUsedSafe(size_t i) const - { - return getUsedSafe(nullptr, i); - } - - /// for multiple disjuncts - bool JoinUsedFlags::getUsedSafe(const Block * block_ptr, size_t row_idx) const - { - if (auto it = flags.find(block_ptr); it != flags.end()) - return it->second[row_idx].load(); - return !need_flags; - } - - /// for single disjunct - template - void JoinUsedFlags::reinit(size_t size) - { - if constexpr (MapGetter::flagged) - { - assert(flags[nullptr].size() <= size); - need_flags = true; - // For one disjunct clause case, we don't need to reinit each time we call addBlockToJoin. - // and there is no value inserted in this JoinUsedFlags before addBlockToJoin finish. - // So we reinit only when the hash table is rehashed to a larger size. - if (flags.empty() || flags[nullptr].size() < size) [[unlikely]] - { - flags[nullptr] = std::vector(size); - } - } - } - - /// for multiple disjuncts - template - void JoinUsedFlags::reinit(const Block * block_ptr) - { - if constexpr (MapGetter::flagged) - { - assert(flags[block_ptr].size() <= block_ptr->rows()); - need_flags = true; - flags[block_ptr] = std::vector(block_ptr->rows()); - } - } - - template - void JoinUsedFlags::setUsed(const FindResult & f) - { - if constexpr (!use_flags) - return; - - /// Could be set simultaneously from different threads. - if constexpr (flag_per_row) - { - auto & mapped = f.getMapped(); - flags[mapped.block][mapped.row_num].store(true, std::memory_order_relaxed); - } - else - { - flags[nullptr][f.getOffset()].store(true, std::memory_order_relaxed); - } - } - - template - void JoinUsedFlags::setUsed(const Block * block, size_t row_num, size_t offset) - { - if constexpr (!use_flags) - return; - - /// Could be set simultaneously from different threads. - if constexpr (flag_per_row) - { - flags[block][row_num].store(true, std::memory_order_relaxed); - } - else - { - flags[nullptr][offset].store(true, std::memory_order_relaxed); - } - } - - template - bool JoinUsedFlags::getUsed(const FindResult & f) - { - if constexpr (!use_flags) - return true; - - if constexpr (flag_per_row) - { - auto & mapped = f.getMapped(); - return flags[mapped.block][mapped.row_num].load(); - } - else - { - return flags[nullptr][f.getOffset()].load(); - } - } - - template - bool JoinUsedFlags::setUsedOnce(const FindResult & f) - { - if constexpr (!use_flags) - return true; - - if constexpr (flag_per_row) - { - auto & mapped = f.getMapped(); - - /// fast check to prevent heavy CAS with seq_cst order - if (flags[mapped.block][mapped.row_num].load(std::memory_order_relaxed)) - return false; - - bool expected = false; - return flags[mapped.block][mapped.row_num].compare_exchange_strong(expected, true); - } - else - { - auto off = f.getOffset(); - - /// fast check to prevent heavy CAS with seq_cst order - if (flags[nullptr][off].load(std::memory_order_relaxed)) - return false; - - bool expected = false; - return flags[nullptr][off].compare_exchange_strong(expected, true); - } - } -} - -static void correctNullabilityInplace(ColumnWithTypeAndName & column, bool nullable) -{ - if (nullable) - { - JoinCommon::convertColumnToNullable(column); - } - else - { - /// We have to replace values masked by NULLs with defaults. - if (column.column) - if (const auto * nullable_column = checkAndGetColumn(&*column.column)) - column.column = JoinCommon::filterWithBlanks(column.column, nullable_column->getNullMapColumn().getData(), true); - - JoinCommon::removeColumnNullability(column); - } -} - -static void correctNullabilityInplace(ColumnWithTypeAndName & column, bool nullable, const IColumn::Filter & negative_null_map) -{ - if (nullable) - { - JoinCommon::convertColumnToNullable(column); - if (column.type->isNullable() && !negative_null_map.empty()) - { - MutableColumnPtr mutable_column = IColumn::mutate(std::move(column.column)); - assert_cast(*mutable_column).applyNegatedNullMap(negative_null_map); - column.column = std::move(mutable_column); - } - } - else - JoinCommon::removeColumnNullability(column); -} - -HashJoin::HashJoin(std::shared_ptr table_join_, const Block & right_sample_block_, - bool any_take_last_row_, size_t reserve_num_, const String & instance_id_) - : table_join(table_join_) - , kind(table_join->kind()) - , strictness(table_join->strictness()) - , any_take_last_row(any_take_last_row_) - , reserve_num(reserve_num_) - , instance_id(instance_id_) - , asof_inequality(table_join->getAsofInequality()) - , data(std::make_shared()) - , tmp_data( - table_join_->getTempDataOnDisk() - ? std::make_unique(table_join_->getTempDataOnDisk(), CurrentMetrics::TemporaryFilesForJoin) - : nullptr) - , right_sample_block(right_sample_block_) - , max_joined_block_rows(table_join->maxJoinedBlockRows()) - , instance_log_id(!instance_id_.empty() ? "(" + instance_id_ + ") " : "") - , log(getLogger("HashJoin")) -{ - LOG_TRACE(log, "{}Keys: {}, datatype: {}, kind: {}, strictness: {}, right header: {}", - instance_log_id, TableJoin::formatClauses(table_join->getClauses(), true), data->type, kind, strictness, right_sample_block.dumpStructure()); - - validateAdditionalFilterExpression(table_join->getMixedJoinExpression()); - - if (isCrossOrComma(kind)) - { - data->type = Type::CROSS; - sample_block_with_columns_to_add = right_sample_block; - } - else if (table_join->getClauses().empty()) - { - data->type = Type::EMPTY; - /// We might need to insert default values into the right columns, materialize them - sample_block_with_columns_to_add = materializeBlock(right_sample_block); - } - else if (table_join->oneDisjunct()) - { - const auto & key_names_right = table_join->getOnlyClause().key_names_right; - JoinCommon::splitAdditionalColumns(key_names_right, right_sample_block, right_table_keys, sample_block_with_columns_to_add); - required_right_keys = table_join->getRequiredRightKeys(right_table_keys, required_right_keys_sources); - } - else - { - /// required right keys concept does not work well if multiple disjuncts, we need all keys - sample_block_with_columns_to_add = right_table_keys = materializeBlock(right_sample_block); - } - - materializeBlockInplace(right_table_keys); - initRightBlockStructure(data->sample_block); - data->sample_block = prepareRightBlock(data->sample_block); - - JoinCommon::createMissedColumns(sample_block_with_columns_to_add); - - size_t disjuncts_num = table_join->getClauses().size(); - data->maps.resize(disjuncts_num); - key_sizes.reserve(disjuncts_num); - - for (const auto & clause : table_join->getClauses()) - { - const auto & key_names_right = clause.key_names_right; - ColumnRawPtrs key_columns = JoinCommon::extractKeysForJoin(right_table_keys, key_names_right); - - if (strictness == JoinStrictness::Asof) - { - assert(disjuncts_num == 1); - - /// @note ASOF JOIN is not INNER. It's better avoid use of 'INNER ASOF' combination in messages. - /// In fact INNER means 'LEFT SEMI ASOF' while LEFT means 'LEFT OUTER ASOF'. - if (!isLeft(kind) && !isInner(kind)) - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Wrong ASOF JOIN type. Only ASOF and LEFT ASOF joins are supported"); - - if (key_columns.size() <= 1) - throw Exception(ErrorCodes::SYNTAX_ERROR, "ASOF join needs at least one equi-join column"); - - size_t asof_size; - asof_type = SortedLookupVectorBase::getTypeSize(*key_columns.back(), asof_size); - key_columns.pop_back(); - - /// this is going to set up the appropriate hash table for the direct lookup part of the join - /// However, this does not depend on the size of the asof join key (as that goes into the BST) - /// Therefore, add it back in such that it can be extracted appropriately from the full stored - /// key_columns and key_sizes - auto & asof_key_sizes = key_sizes.emplace_back(); - data->type = chooseMethod(kind, key_columns, asof_key_sizes); - asof_key_sizes.push_back(asof_size); - } - else - { - /// Choose data structure to use for JOIN. - auto current_join_method = chooseMethod(kind, key_columns, key_sizes.emplace_back()); - if (data->type == Type::EMPTY) - data->type = current_join_method; - else if (data->type != current_join_method) - data->type = Type::hashed; - } - } - - for (auto & maps : data->maps) - dataMapInit(maps); -} - -HashJoin::Type HashJoin::chooseMethod(JoinKind kind, const ColumnRawPtrs & key_columns, Sizes & key_sizes) -{ - size_t keys_size = key_columns.size(); - - if (keys_size == 0) - { - if (isCrossOrComma(kind)) - return Type::CROSS; - return Type::EMPTY; - } - - bool all_fixed = true; - size_t keys_bytes = 0; - key_sizes.resize(keys_size); - for (size_t j = 0; j < keys_size; ++j) - { - if (!key_columns[j]->isFixedAndContiguous()) - { - all_fixed = false; - break; - } - key_sizes[j] = key_columns[j]->sizeOfValueIfFixed(); - keys_bytes += key_sizes[j]; - } - - /// If there is one numeric key that fits in 64 bits - if (keys_size == 1 && key_columns[0]->isNumeric()) - { - size_t size_of_field = key_columns[0]->sizeOfValueIfFixed(); - if (size_of_field == 1) - return Type::key8; - if (size_of_field == 2) - return Type::key16; - if (size_of_field == 4) - return Type::key32; - if (size_of_field == 8) - return Type::key64; - if (size_of_field == 16) - return Type::keys128; - if (size_of_field == 32) - return Type::keys256; - throw Exception(ErrorCodes::LOGICAL_ERROR, "Numeric column has sizeOfField not in 1, 2, 4, 8, 16, 32."); - } - - /// If the keys fit in N bits, we will use a hash table for N-bit-packed keys - if (all_fixed && keys_bytes <= 16) - return Type::keys128; - if (all_fixed && keys_bytes <= 32) - return Type::keys256; - - /// If there is single string key, use hash table of it's values. - if (keys_size == 1) - { - auto is_string_column = [](const IColumn * column_ptr) -> bool - { - if (const auto * lc_column_ptr = typeid_cast(column_ptr)) - return typeid_cast(lc_column_ptr->getDictionary().getNestedColumn().get()); - return typeid_cast(column_ptr); - }; - - const auto * key_column = key_columns[0]; - if (is_string_column(key_column) || - (isColumnConst(*key_column) && is_string_column(assert_cast(key_column)->getDataColumnPtr().get()))) - return Type::key_string; - } - - if (keys_size == 1 && typeid_cast(key_columns[0])) - return Type::key_fixed_string; - - /// Otherwise, will use set of cryptographic hashes of unambiguously serialized values. - return Type::hashed; -} - -template -static KeyGetter createKeyGetter(const ColumnRawPtrs & key_columns, const Sizes & key_sizes) -{ - if constexpr (is_asof_join) - { - auto key_column_copy = key_columns; - auto key_size_copy = key_sizes; - key_column_copy.pop_back(); - key_size_copy.pop_back(); - return KeyGetter(key_column_copy, key_size_copy, nullptr); - } - else - return KeyGetter(key_columns, key_sizes, nullptr); -} - -template -using FindResultImpl = ColumnsHashing::columns_hashing_impl::FindResultImpl; - -/// Dummy key getter, always find nothing, used for JOIN ON NULL -template -class KeyGetterEmpty -{ -public: - struct MappedType - { - using mapped_type = Mapped; - }; - - using FindResult = ColumnsHashing::columns_hashing_impl::FindResultImpl; - - KeyGetterEmpty() = default; - - FindResult findKey(MappedType, size_t, const Arena &) { return FindResult(); } -}; - -template -struct KeyGetterForTypeImpl; - -constexpr bool use_offset = true; - -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodOneNumber; -}; -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodOneNumber; -}; -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodOneNumber; -}; -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodOneNumber; -}; -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodString; -}; -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodFixedString; -}; -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodKeysFixed; -}; -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodKeysFixed; -}; -template struct KeyGetterForTypeImpl -{ - using Type = ColumnsHashing::HashMethodHashed; -}; - -template -struct KeyGetterForType -{ - using Value = typename Data::value_type; - using Mapped_t = typename Data::mapped_type; - using Mapped = std::conditional_t, const Mapped_t, Mapped_t>; - using Type = typename KeyGetterForTypeImpl::Type; -}; - -void HashJoin::dataMapInit(MapsVariant & map) -{ - if (kind == JoinKind::Cross) - return; - joinDispatchInit(kind, strictness, map); - joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { map_.create(data->type); }); - - if (reserve_num) - { - joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { map_.reserve(data->type, reserve_num); }); - } - - if (!data) - throw Exception(ErrorCodes::LOGICAL_ERROR, "HashJoin::dataMapInit called with empty data"); -} - -bool HashJoin::empty() const -{ - return data->type == Type::EMPTY; -} - -bool HashJoin::alwaysReturnsEmptySet() const -{ - return isInnerOrRight(getKind()) && data->empty; -} - -size_t HashJoin::getTotalRowCount() const -{ - if (!data) - return 0; - - size_t res = 0; - - if (data->type == Type::CROSS) - { - for (const auto & block : data->blocks) - res += block.rows(); - } - else - { - for (const auto & map : data->maps) - { - joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { res += map_.getTotalRowCount(data->type); }); - } - } - - return res; -} - -size_t HashJoin::getTotalByteCount() const -{ - if (!data) - return 0; - -#ifndef NDEBUG - size_t debug_blocks_allocated_size = 0; - for (const auto & block : data->blocks) - debug_blocks_allocated_size += block.allocatedBytes(); - - if (data->blocks_allocated_size != debug_blocks_allocated_size) - throw Exception(ErrorCodes::LOGICAL_ERROR, "data->blocks_allocated_size != debug_blocks_allocated_size ({} != {})", - data->blocks_allocated_size, debug_blocks_allocated_size); - - size_t debug_blocks_nullmaps_allocated_size = 0; - for (const auto & nullmap : data->blocks_nullmaps) - debug_blocks_nullmaps_allocated_size += nullmap.second->allocatedBytes(); - - if (data->blocks_nullmaps_allocated_size != debug_blocks_nullmaps_allocated_size) - throw Exception(ErrorCodes::LOGICAL_ERROR, "data->blocks_nullmaps_allocated_size != debug_blocks_nullmaps_allocated_size ({} != {})", - data->blocks_nullmaps_allocated_size, debug_blocks_nullmaps_allocated_size); -#endif - - size_t res = 0; - - res += data->blocks_allocated_size; - res += data->blocks_nullmaps_allocated_size; - res += data->pool.allocatedBytes(); - - if (data->type != Type::CROSS) - { - for (const auto & map : data->maps) - { - joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { res += map_.getTotalByteCountImpl(data->type); }); - } - } - return res; -} - -namespace -{ - /// Inserting an element into a hash table of the form `key -> reference to a string`, which will then be used by JOIN. - template - struct Inserter - { - static ALWAYS_INLINE bool insertOne(const HashJoin & join, Map & map, KeyGetter & key_getter, Block * stored_block, size_t i, - Arena & pool) - { - auto emplace_result = key_getter.emplaceKey(map, i, pool); - - if (emplace_result.isInserted() || join.anyTakeLastRow()) - { - new (&emplace_result.getMapped()) typename Map::mapped_type(stored_block, i); - return true; - } - return false; - } - - static ALWAYS_INLINE void insertAll(const HashJoin &, Map & map, KeyGetter & key_getter, Block * stored_block, size_t i, Arena & pool) - { - auto emplace_result = key_getter.emplaceKey(map, i, pool); - - if (emplace_result.isInserted()) - new (&emplace_result.getMapped()) typename Map::mapped_type(stored_block, i); - else - { - /// The first element of the list is stored in the value of the hash table, the rest in the pool. - emplace_result.getMapped().insert({stored_block, i}, pool); - } - } - - static ALWAYS_INLINE void insertAsof(HashJoin & join, Map & map, KeyGetter & key_getter, Block * stored_block, size_t i, Arena & pool, - const IColumn & asof_column) - { - auto emplace_result = key_getter.emplaceKey(map, i, pool); - typename Map::mapped_type * time_series_map = &emplace_result.getMapped(); - - TypeIndex asof_type = *join.getAsofType(); - if (emplace_result.isInserted()) - time_series_map = new (time_series_map) typename Map::mapped_type(createAsofRowRef(asof_type, join.getAsofInequality())); - (*time_series_map)->insert(asof_column, stored_block, i); - } - }; - - - template - size_t NO_INLINE insertFromBlockImplTypeCase( - HashJoin & join, Map & map, size_t rows, const ColumnRawPtrs & key_columns, - const Sizes & key_sizes, Block * stored_block, ConstNullMapPtr null_map, UInt8ColumnDataPtr join_mask, Arena & pool, bool & is_inserted) - { - [[maybe_unused]] constexpr bool mapped_one = std::is_same_v; - constexpr bool is_asof_join = STRICTNESS == JoinStrictness::Asof; - - const IColumn * asof_column [[maybe_unused]] = nullptr; - if constexpr (is_asof_join) - asof_column = key_columns.back(); - - auto key_getter = createKeyGetter(key_columns, key_sizes); - - /// For ALL and ASOF join always insert values - is_inserted = !mapped_one || is_asof_join; - - for (size_t i = 0; i < rows; ++i) - { - if (null_map && (*null_map)[i]) - { - /// nulls are not inserted into hash table, - /// keep them for RIGHT and FULL joins - is_inserted = true; - continue; - } - - /// Check condition for right table from ON section - if (join_mask && !(*join_mask)[i]) - continue; - - if constexpr (is_asof_join) - Inserter::insertAsof(join, map, key_getter, stored_block, i, pool, *asof_column); - else if constexpr (mapped_one) - is_inserted |= Inserter::insertOne(join, map, key_getter, stored_block, i, pool); - else - Inserter::insertAll(join, map, key_getter, stored_block, i, pool); - } - return map.getBufferSizeInCells(); - } - - template - size_t insertFromBlockImpl( - HashJoin & join, HashJoin::Type type, Maps & maps, size_t rows, const ColumnRawPtrs & key_columns, - const Sizes & key_sizes, Block * stored_block, ConstNullMapPtr null_map, UInt8ColumnDataPtr join_mask, Arena & pool, bool & is_inserted) - { - switch (type) - { - case HashJoin::Type::EMPTY: - [[fallthrough]]; - case HashJoin::Type::CROSS: - /// Do nothing. We will only save block, and it is enough - is_inserted = true; - return 0; - - #define M(TYPE) \ - case HashJoin::Type::TYPE: \ - return insertFromBlockImplTypeCase>::Type>(\ - join, *maps.TYPE, rows, key_columns, key_sizes, stored_block, null_map, join_mask, pool, is_inserted); \ - break; - - APPLY_FOR_JOIN_VARIANTS(M) - #undef M - } - } -} - -void HashJoin::initRightBlockStructure(Block & saved_block_sample) -{ - if (isCrossOrComma(kind)) - { - /// cross join doesn't have keys, just add all columns - saved_block_sample = sample_block_with_columns_to_add.cloneEmpty(); - return; - } - - bool multiple_disjuncts = !table_join->oneDisjunct(); - /// We could remove key columns for LEFT | INNER HashJoin but we should keep them for JoinSwitcher (if any). - bool save_key_columns = table_join->isEnabledAlgorithm(JoinAlgorithm::AUTO) || - table_join->isEnabledAlgorithm(JoinAlgorithm::GRACE_HASH) || - isRightOrFull(kind) || - multiple_disjuncts || - table_join->getMixedJoinExpression(); - if (save_key_columns) - { - saved_block_sample = right_table_keys.cloneEmpty(); - } - else if (strictness == JoinStrictness::Asof) - { - /// Save ASOF key - saved_block_sample.insert(right_table_keys.safeGetByPosition(right_table_keys.columns() - 1)); - } - - /// Save non key columns - for (auto & column : sample_block_with_columns_to_add) - { - if (auto * col = saved_block_sample.findByName(column.name)) - *col = column; - else - saved_block_sample.insert(column); - } -} - -Block HashJoin::prepareRightBlock(const Block & block, const Block & saved_block_sample_) -{ - Block structured_block; - for (const auto & sample_column : saved_block_sample_.getColumnsWithTypeAndName()) - { - ColumnWithTypeAndName column = block.getByName(sample_column.name); - - /// There's no optimization for right side const columns. Remove constness if any. - column.column = recursiveRemoveSparse(column.column->convertToFullColumnIfConst()); - - if (column.column->lowCardinality() && !sample_column.column->lowCardinality()) - { - column.column = column.column->convertToFullColumnIfLowCardinality(); - column.type = removeLowCardinality(column.type); - } - - if (sample_column.column->isNullable()) - JoinCommon::convertColumnToNullable(column); - - structured_block.insert(std::move(column)); - } - - return structured_block; -} - -Block HashJoin::prepareRightBlock(const Block & block) const -{ - return prepareRightBlock(block, savedBlockSample()); -} - -bool HashJoin::addBlockToJoin(const Block & source_block_, bool check_limits) -{ - if (!data) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Join data was released"); - - /// RowRef::SizeT is uint32_t (not size_t) for hash table Cell memory efficiency. - /// It's possible to split bigger blocks and insert them by parts here. But it would be a dead code. - if (unlikely(source_block_.rows() > std::numeric_limits::max())) - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Too many rows in right table block for HashJoin: {}", source_block_.rows()); - - /** We do not allocate memory for stored blocks inside HashJoin, only for hash table. - * In case when we have all the blocks allocated before the first `addBlockToJoin` call, will already be quite high. - * In that case memory consumed by stored blocks will be underestimated. - */ - if (!memory_usage_before_adding_blocks) - memory_usage_before_adding_blocks = getCurrentQueryMemoryUsage(); - - Block source_block = source_block_; - if (strictness == JoinStrictness::Asof) - { - chassert(kind == JoinKind::Left || kind == JoinKind::Inner); - - /// Filter out rows with NULLs in ASOF key, nulls are not joined with anything since they are not comparable - /// We support only INNER/LEFT ASOF join, so rows with NULLs never return from the right joined table. - /// So filter them out here not to handle in implementation. - const auto & asof_key_name = table_join->getOnlyClause().key_names_right.back(); - auto & asof_column = source_block.getByName(asof_key_name); - - if (asof_column.type->isNullable()) - { - /// filter rows with nulls in asof key - if (const auto * asof_const_column = typeid_cast(asof_column.column.get())) - { - if (asof_const_column->isNullAt(0)) - return false; - } - else - { - const auto & asof_column_nullable = assert_cast(*asof_column.column).getNullMapData(); - - NullMap negative_null_map(asof_column_nullable.size()); - for (size_t i = 0; i < asof_column_nullable.size(); ++i) - negative_null_map[i] = !asof_column_nullable[i]; - - for (auto & column : source_block) - column.column = column.column->filter(negative_null_map, -1); - } - } - } - - size_t rows = source_block.rows(); - - const auto & right_key_names = table_join->getAllNames(JoinTableSide::Right); - ColumnPtrMap all_key_columns(right_key_names.size()); - for (const auto & column_name : right_key_names) - { - const auto & column = source_block.getByName(column_name).column; - all_key_columns[column_name] = recursiveRemoveSparse(column->convertToFullColumnIfConst())->convertToFullColumnIfLowCardinality(); - } - - Block block_to_save = prepareRightBlock(source_block); - if (shrink_blocks) - block_to_save = block_to_save.shrinkToFit(); - - size_t max_bytes_in_join = table_join->sizeLimits().max_bytes; - size_t max_rows_in_join = table_join->sizeLimits().max_rows; - - if (kind == JoinKind::Cross && tmp_data - && (tmp_stream || (max_bytes_in_join && getTotalByteCount() + block_to_save.allocatedBytes() >= max_bytes_in_join) - || (max_rows_in_join && getTotalRowCount() + block_to_save.rows() >= max_rows_in_join))) - { - if (tmp_stream == nullptr) - { - tmp_stream = &tmp_data->createStream(right_sample_block); - } - tmp_stream->write(block_to_save); - return true; - } - - size_t total_rows = 0; - size_t total_bytes = 0; - { - if (storage_join_lock) - throw DB::Exception(ErrorCodes::LOGICAL_ERROR, "addBlockToJoin called when HashJoin locked to prevent updates"); - - assertBlocksHaveEqualStructure(data->sample_block, block_to_save, "joined block"); - - size_t min_bytes_to_compress = table_join->crossJoinMinBytesToCompress(); - size_t min_rows_to_compress = table_join->crossJoinMinRowsToCompress(); - - if (kind == JoinKind::Cross - && ((min_bytes_to_compress && getTotalByteCount() >= min_bytes_to_compress) - || (min_rows_to_compress && getTotalRowCount() >= min_rows_to_compress))) - { - block_to_save = block_to_save.compress(); - have_compressed = true; - } - - data->blocks_allocated_size += block_to_save.allocatedBytes(); - data->blocks.emplace_back(std::move(block_to_save)); - Block * stored_block = &data->blocks.back(); - - if (rows) - data->empty = false; - - bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); - const auto & onexprs = table_join->getClauses(); - for (size_t onexpr_idx = 0; onexpr_idx < onexprs.size(); ++onexpr_idx) - { - ColumnRawPtrs key_columns; - for (const auto & name : onexprs[onexpr_idx].key_names_right) - key_columns.push_back(all_key_columns[name].get()); - - /// We will insert to the map only keys, where all components are not NULL. - ConstNullMapPtr null_map{}; - ColumnPtr null_map_holder = extractNestedColumnsAndNullMap(key_columns, null_map); - - /// If RIGHT or FULL save blocks with nulls for NotJoinedBlocks - UInt8 save_nullmap = 0; - if (isRightOrFull(kind) && null_map) - { - /// Save rows with NULL keys - for (size_t i = 0; !save_nullmap && i < null_map->size(); ++i) - save_nullmap |= (*null_map)[i]; - } - - auto join_mask_col = JoinCommon::getColumnAsMask(source_block, onexprs[onexpr_idx].condColumnNames().second); - /// Save blocks that do not hold conditions in ON section - ColumnUInt8::MutablePtr not_joined_map = nullptr; - if (!flag_per_row && isRightOrFull(kind) && join_mask_col.hasData()) - { - const auto & join_mask = join_mask_col.getData(); - /// Save rows that do not hold conditions - not_joined_map = ColumnUInt8::create(rows, 0); - for (size_t i = 0, sz = join_mask->size(); i < sz; ++i) - { - /// Condition hold, do not save row - if ((*join_mask)[i]) - continue; - - /// NULL key will be saved anyway because, do not save twice - if (save_nullmap && (*null_map)[i]) - continue; - - not_joined_map->getData()[i] = 1; - } - } - - bool is_inserted = false; - if (kind != JoinKind::Cross) - { - joinDispatch(kind, strictness, data->maps[onexpr_idx], [&](auto kind_, auto strictness_, auto & map) - { - size_t size = insertFromBlockImpl( - *this, data->type, map, rows, key_columns, key_sizes[onexpr_idx], stored_block, null_map, - /// If mask is false constant, rows are added to hashmap anyway. It's not a happy-flow, so this case is not optimized - join_mask_col.getData(), - data->pool, is_inserted); - - if (flag_per_row) - used_flags.reinit(stored_block); - else if (is_inserted) - /// Number of buckets + 1 value from zero storage - used_flags.reinit(size + 1); - }); - } - - if (!flag_per_row && save_nullmap && is_inserted) - { - data->blocks_nullmaps_allocated_size += null_map_holder->allocatedBytes(); - data->blocks_nullmaps.emplace_back(stored_block, null_map_holder); - } - - if (!flag_per_row && not_joined_map && is_inserted) - { - data->blocks_nullmaps_allocated_size += not_joined_map->allocatedBytes(); - data->blocks_nullmaps.emplace_back(stored_block, std::move(not_joined_map)); - } - - if (!flag_per_row && !is_inserted) - { - LOG_TRACE(log, "Skipping inserting block with {} rows", rows); - data->blocks_allocated_size -= stored_block->allocatedBytes(); - data->blocks.pop_back(); - } - - if (!check_limits) - return true; - - /// TODO: Do not calculate them every time - total_rows = getTotalRowCount(); - total_bytes = getTotalByteCount(); - } - } - - shrinkStoredBlocksToFit(total_bytes); - - return table_join->sizeLimits().check(total_rows, total_bytes, "JOIN", ErrorCodes::SET_SIZE_LIMIT_EXCEEDED); -} - -void HashJoin::shrinkStoredBlocksToFit(size_t & total_bytes_in_join) -{ - if (shrink_blocks) - return; /// Already shrunk - - Int64 current_memory_usage = getCurrentQueryMemoryUsage(); - Int64 query_memory_usage_delta = current_memory_usage - memory_usage_before_adding_blocks; - Int64 max_total_bytes_for_query = memory_usage_before_adding_blocks ? table_join->getMaxMemoryUsage() : 0; - - auto max_total_bytes_in_join = table_join->sizeLimits().max_bytes; - - /** If accounted data size is more than half of `max_bytes_in_join` - * or query memory consumption growth from the beginning of adding blocks (estimation of memory consumed by join using memory tracker) - * is bigger than half of all memory available for query, - * then shrink stored blocks to fit. - */ - shrink_blocks = (max_total_bytes_in_join && total_bytes_in_join > max_total_bytes_in_join / 2) || - (max_total_bytes_for_query && query_memory_usage_delta > max_total_bytes_for_query / 2); - if (!shrink_blocks) - return; - - LOG_DEBUG(log, "Shrinking stored blocks, memory consumption is {} {} calculated by join, {} {} by memory tracker", - ReadableSize(total_bytes_in_join), max_total_bytes_in_join ? fmt::format("/ {}", ReadableSize(max_total_bytes_in_join)) : "", - ReadableSize(query_memory_usage_delta), max_total_bytes_for_query ? fmt::format("/ {}", ReadableSize(max_total_bytes_for_query)) : ""); - - for (auto & stored_block : data->blocks) - { - size_t old_size = stored_block.allocatedBytes(); - stored_block = stored_block.shrinkToFit(); - size_t new_size = stored_block.allocatedBytes(); - - if (old_size >= new_size) - { - if (data->blocks_allocated_size < old_size - new_size) - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Blocks allocated size value is broken: " - "blocks_allocated_size = {}, old_size = {}, new_size = {}", - data->blocks_allocated_size, old_size, new_size); - - data->blocks_allocated_size -= old_size - new_size; - } - else - /// Sometimes after clone resized block can be bigger than original - data->blocks_allocated_size += new_size - old_size; - } - - auto new_total_bytes_in_join = getTotalByteCount(); - - Int64 new_current_memory_usage = getCurrentQueryMemoryUsage(); - - LOG_DEBUG(log, "Shrunk stored blocks {} freed ({} by memory tracker), new memory consumption is {} ({} by memory tracker)", - ReadableSize(total_bytes_in_join - new_total_bytes_in_join), ReadableSize(current_memory_usage - new_current_memory_usage), - ReadableSize(new_total_bytes_in_join), ReadableSize(new_current_memory_usage)); - - total_bytes_in_join = new_total_bytes_in_join; -} - - -namespace -{ - -struct JoinOnKeyColumns -{ - Names key_names; - - Columns materialized_keys_holder; - ColumnRawPtrs key_columns; - - ConstNullMapPtr null_map; - ColumnPtr null_map_holder; - - /// Only rows where mask == true can be joined - JoinCommon::JoinMask join_mask_column; - - Sizes key_sizes; - - explicit JoinOnKeyColumns(const Block & block, const Names & key_names_, const String & cond_column_name, const Sizes & key_sizes_) - : key_names(key_names_) - , materialized_keys_holder(JoinCommon::materializeColumns(block, key_names)) /// Rare case, when keys are constant or low cardinality. To avoid code bloat, simply materialize them. - , key_columns(JoinCommon::getRawPointers(materialized_keys_holder)) - , null_map(nullptr) - , null_map_holder(extractNestedColumnsAndNullMap(key_columns, null_map)) - , join_mask_column(JoinCommon::getColumnAsMask(block, cond_column_name)) - , key_sizes(key_sizes_) - { - } - - bool isRowFiltered(size_t i) const { return join_mask_column.isRowFiltered(i); } -}; - -template -class AddedColumns -{ -public: - struct TypeAndName - { - DataTypePtr type; - String name; - String qualified_name; - - TypeAndName(DataTypePtr type_, const String & name_, const String & qualified_name_) - : type(type_), name(name_), qualified_name(qualified_name_) - { - } - }; - - struct LazyOutput - { - PaddedPODArray blocks; - PaddedPODArray row_nums; - }; - - AddedColumns( - const Block & left_block_, - const Block & block_with_columns_to_add, - const Block & saved_block_sample, - const HashJoin & join, - std::vector && join_on_keys_, - ExpressionActionsPtr additional_filter_expression_, - bool is_asof_join, - bool is_join_get_) - : left_block(left_block_) - , join_on_keys(join_on_keys_) - , additional_filter_expression(additional_filter_expression_) - , rows_to_add(left_block.rows()) - , is_join_get(is_join_get_) - { - size_t num_columns_to_add = block_with_columns_to_add.columns(); - if (is_asof_join) - ++num_columns_to_add; - - if constexpr (lazy) - { - has_columns_to_add = num_columns_to_add > 0; - lazy_output.blocks.reserve(rows_to_add); - lazy_output.row_nums.reserve(rows_to_add); - } - - columns.reserve(num_columns_to_add); - type_name.reserve(num_columns_to_add); - right_indexes.reserve(num_columns_to_add); - - for (const auto & src_column : block_with_columns_to_add) - { - /// Column names `src_column.name` and `qualified_name` can differ for StorageJoin, - /// because it uses not qualified right block column names - auto qualified_name = join.getTableJoin().renamedRightColumnName(src_column.name); - /// Don't insert column if it's in left block - if (!left_block.has(qualified_name)) - addColumn(src_column, qualified_name); - } - - if (is_asof_join) - { - assert(join_on_keys.size() == 1); - const ColumnWithTypeAndName & right_asof_column = join.rightAsofKeyColumn(); - addColumn(right_asof_column, right_asof_column.name); - left_asof_key = join_on_keys[0].key_columns.back(); - } - - for (auto & tn : type_name) - right_indexes.push_back(saved_block_sample.getPositionByName(tn.name)); - - nullable_column_ptrs.resize(right_indexes.size(), nullptr); - for (size_t j = 0; j < right_indexes.size(); ++j) - { - /** If it's joinGetOrNull, we will have nullable columns in result block - * even if right column is not nullable in storage (saved_block_sample). - */ - const auto & saved_column = saved_block_sample.getByPosition(right_indexes[j]).column; - if (columns[j]->isNullable() && !saved_column->isNullable()) - nullable_column_ptrs[j] = typeid_cast(columns[j].get()); - } - } - - size_t size() const { return columns.size(); } - - void buildOutput(); - - ColumnWithTypeAndName moveColumn(size_t i) - { - return ColumnWithTypeAndName(std::move(columns[i]), type_name[i].type, type_name[i].qualified_name); - } - - void appendFromBlock(const Block & block, size_t row_num, bool has_default); - - void appendDefaultRow(); - - void applyLazyDefaults(); - - const IColumn & leftAsofKey() const { return *left_asof_key; } - - Block left_block; - std::vector join_on_keys; - ExpressionActionsPtr additional_filter_expression; - - size_t max_joined_block_rows = 0; - size_t rows_to_add; - std::unique_ptr offsets_to_replicate; - bool need_filter = false; - IColumn::Filter filter; - - void reserve(bool need_replicate) - { - if (!max_joined_block_rows) - return; - - /// Do not allow big allocations when user set max_joined_block_rows to huge value - size_t reserve_size = std::min(max_joined_block_rows, DEFAULT_BLOCK_SIZE * 2); - - if (need_replicate) - /// Reserve 10% more space for columns, because some rows can be repeated - reserve_size = static_cast(1.1 * reserve_size); - - for (auto & column : columns) - column->reserve(reserve_size); - } - -private: - - void checkBlock(const Block & block) - { - for (size_t j = 0; j < right_indexes.size(); ++j) - { - const auto * column_from_block = block.getByPosition(right_indexes[j]).column.get(); - const auto * dest_column = columns[j].get(); - if (auto * nullable_col = nullable_column_ptrs[j]) - { - if (!is_join_get) - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Columns {} and {} can have different nullability only in joinGetOrNull", - dest_column->getName(), column_from_block->getName()); - dest_column = nullable_col->getNestedColumnPtr().get(); - } - /** Using dest_column->structureEquals(*column_from_block) will not work for low cardinality columns, - * because dictionaries can be different, while calling insertFrom on them is safe, for example: - * ColumnLowCardinality(size = 0, UInt8(size = 0), ColumnUnique(size = 1, String(size = 1))) - * and - * ColumnLowCardinality(size = 0, UInt16(size = 0), ColumnUnique(size = 1, String(size = 1))) - */ - if (typeid(*dest_column) != typeid(*column_from_block)) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Columns {} and {} have different types {} and {}", - dest_column->getName(), column_from_block->getName(), - demangle(typeid(*dest_column).name()), demangle(typeid(*column_from_block).name())); - } - } - - MutableColumns columns; - bool is_join_get; - std::vector right_indexes; - std::vector type_name; - std::vector nullable_column_ptrs; - size_t lazy_defaults_count = 0; - - /// for lazy - // The default row is represented by an empty RowRef, so that fixed-size blocks can be generated sequentially, - // default_count cannot represent the position of the row - LazyOutput lazy_output; - bool has_columns_to_add; - - /// for ASOF - const IColumn * left_asof_key = nullptr; - - - void addColumn(const ColumnWithTypeAndName & src_column, const std::string & qualified_name) - { - columns.push_back(src_column.column->cloneEmpty()); - columns.back()->reserve(src_column.column->size()); - type_name.emplace_back(src_column.type, src_column.name, qualified_name); - } -}; -template<> void AddedColumns::buildOutput() -{ -} - -template<> -void AddedColumns::buildOutput() -{ - for (size_t i = 0; i < this->size(); ++i) - { - auto& col = columns[i]; - size_t default_count = 0; - auto apply_default = [&]() - { - if (default_count > 0) - { - JoinCommon::addDefaultValues(*col, type_name[i].type, default_count); - default_count = 0; - } - }; - - for (size_t j = 0; j < lazy_output.blocks.size(); ++j) - { - if (!lazy_output.blocks[j]) - { - default_count++; - continue; - } - apply_default(); - const auto & column_from_block = reinterpret_cast(lazy_output.blocks[j])->getByPosition(right_indexes[i]); - /// If it's joinGetOrNull, we need to wrap not-nullable columns in StorageJoin. - if (is_join_get) - { - if (auto * nullable_col = typeid_cast(col.get()); - nullable_col && !column_from_block.column->isNullable()) - { - nullable_col->insertFromNotNullable(*column_from_block.column, lazy_output.row_nums[j]); - continue; - } - } - col->insertFrom(*column_from_block.column, lazy_output.row_nums[j]); - } - apply_default(); - } -} - -template<> -void AddedColumns::applyLazyDefaults() -{ - if (lazy_defaults_count) - { - for (size_t j = 0, size = right_indexes.size(); j < size; ++j) - JoinCommon::addDefaultValues(*columns[j], type_name[j].type, lazy_defaults_count); - lazy_defaults_count = 0; - } -} - -template<> -void AddedColumns::applyLazyDefaults() -{ -} - -template <> -void AddedColumns::appendFromBlock(const Block & block, size_t row_num,const bool has_defaults) -{ - if (has_defaults) - applyLazyDefaults(); - -#ifndef NDEBUG - checkBlock(block); -#endif - if (is_join_get) - { - size_t right_indexes_size = right_indexes.size(); - for (size_t j = 0; j < right_indexes_size; ++j) - { - const auto & column_from_block = block.getByPosition(right_indexes[j]); - if (auto * nullable_col = nullable_column_ptrs[j]) - nullable_col->insertFromNotNullable(*column_from_block.column, row_num); - else - columns[j]->insertFrom(*column_from_block.column, row_num); - } - } - else - { - size_t right_indexes_size = right_indexes.size(); - for (size_t j = 0; j < right_indexes_size; ++j) - { - const auto & column_from_block = block.getByPosition(right_indexes[j]); - columns[j]->insertFrom(*column_from_block.column, row_num); - } - } -} - -template <> -void AddedColumns::appendFromBlock(const Block & block, size_t row_num, bool) -{ -#ifndef NDEBUG - checkBlock(block); -#endif - if (has_columns_to_add) - { - lazy_output.blocks.emplace_back(reinterpret_cast(&block)); - lazy_output.row_nums.emplace_back(static_cast(row_num)); - } -} -template<> -void AddedColumns::appendDefaultRow() -{ - ++lazy_defaults_count; -} - -template<> -void AddedColumns::appendDefaultRow() -{ - if (has_columns_to_add) - { - lazy_output.blocks.emplace_back(0); - lazy_output.row_nums.emplace_back(0); - } -} - -template -struct JoinFeatures -{ - static constexpr bool is_any_join = STRICTNESS == JoinStrictness::Any; - static constexpr bool is_any_or_semi_join = STRICTNESS == JoinStrictness::Any || STRICTNESS == JoinStrictness::RightAny || (STRICTNESS == JoinStrictness::Semi && KIND == JoinKind::Left); - static constexpr bool is_all_join = STRICTNESS == JoinStrictness::All; - static constexpr bool is_asof_join = STRICTNESS == JoinStrictness::Asof; - static constexpr bool is_semi_join = STRICTNESS == JoinStrictness::Semi; - static constexpr bool is_anti_join = STRICTNESS == JoinStrictness::Anti; - - static constexpr bool left = KIND == JoinKind::Left; - static constexpr bool right = KIND == JoinKind::Right; - static constexpr bool inner = KIND == JoinKind::Inner; - static constexpr bool full = KIND == JoinKind::Full; - - static constexpr bool need_replication = is_all_join || (is_any_join && right) || (is_semi_join && right); - static constexpr bool need_filter = !need_replication && (inner || right || (is_semi_join && left) || (is_anti_join && left)); - static constexpr bool add_missing = (left || full) && !is_semi_join; - - static constexpr bool need_flags = MapGetter::flagged; -}; - -template -class KnownRowsHolder; - -/// Keep already joined rows to prevent duplication if many disjuncts -/// if for a particular pair of rows condition looks like TRUE or TRUE or TRUE -/// we want to have it once in resultset -template<> -class KnownRowsHolder -{ -public: - using Type = std::pair; - -private: - static const size_t MAX_LINEAR = 16; // threshold to switch from Array to Set - using ArrayHolder = std::array; - using SetHolder = std::set; - using SetHolderPtr = std::unique_ptr; - - ArrayHolder array_holder; - SetHolderPtr set_holder_ptr; - - size_t items; - -public: - KnownRowsHolder() - : items(0) - { - } - - - template - void add(InputIt from, InputIt to) - { - const size_t new_items = std::distance(from, to); - if (items + new_items <= MAX_LINEAR) - { - std::copy(from, to, &array_holder[items]); - } - else - { - if (items <= MAX_LINEAR) - { - set_holder_ptr = std::make_unique(); - set_holder_ptr->insert(std::cbegin(array_holder), std::cbegin(array_holder) + items); - } - set_holder_ptr->insert(from, to); - } - items += new_items; - } - - template - bool isKnown(const Needle & needle) - { - return items <= MAX_LINEAR - ? std::find(std::cbegin(array_holder), std::cbegin(array_holder) + items, needle) != std::cbegin(array_holder) + items - : set_holder_ptr->find(needle) != set_holder_ptr->end(); - } -}; - -template<> -class KnownRowsHolder -{ -public: - template - void add(InputIt, InputIt) - { - } - - template - static bool isKnown(const Needle &) - { - return false; - } -}; - -template -void addFoundRowAll( - const typename Map::mapped_type & mapped, - AddedColumns & added, - IColumn::Offset & current_offset, - KnownRowsHolder & known_rows [[maybe_unused]], - JoinStuff::JoinUsedFlags * used_flags [[maybe_unused]]) -{ - if constexpr (add_missing) - added.applyLazyDefaults(); - - if constexpr (flag_per_row) - { - std::unique_ptr::Type>> new_known_rows_ptr; - - for (auto it = mapped.begin(); it.ok(); ++it) - { - if (!known_rows.isKnown(std::make_pair(it->block, it->row_num))) - { - added.appendFromBlock(*it->block, it->row_num, false); - ++current_offset; - if (!new_known_rows_ptr) - { - new_known_rows_ptr = std::make_unique::Type>>(); - } - new_known_rows_ptr->push_back(std::make_pair(it->block, it->row_num)); - if (used_flags) - { - used_flags->JoinStuff::JoinUsedFlags::setUsedOnce( - FindResultImpl(*it, true, 0)); - } - } - } - - if (new_known_rows_ptr) - { - known_rows.add(std::cbegin(*new_known_rows_ptr), std::cend(*new_known_rows_ptr)); - } - } - else - { - for (auto it = mapped.begin(); it.ok(); ++it) - { - added.appendFromBlock(*it->block, it->row_num, false); - ++current_offset; - } - } -} - -template -void addNotFoundRow(AddedColumns & added [[maybe_unused]], IColumn::Offset & current_offset [[maybe_unused]]) -{ - if constexpr (add_missing) - { - added.appendDefaultRow(); - if constexpr (need_offset) - ++current_offset; - } -} - -template -void setUsed(IColumn::Filter & filter [[maybe_unused]], size_t pos [[maybe_unused]]) -{ - if constexpr (need_filter) - filter[pos] = 1; -} - -template -ColumnPtr buildAdditionalFilter( - size_t left_start_row, - const std::vector & selected_rows, - const std::vector & row_replicate_offset, - AddedColumns & added_columns) -{ - ColumnPtr result_column; - do - { - if (selected_rows.empty()) - { - result_column = ColumnUInt8::create(); - break; - } - const Block & sample_right_block = *selected_rows.begin()->block; - if (!sample_right_block || !added_columns.additional_filter_expression) - { - auto filter = ColumnUInt8::create(); - filter->insertMany(1, selected_rows.size()); - result_column = std::move(filter); - break; - } - - auto required_cols = added_columns.additional_filter_expression->getRequiredColumnsWithTypes(); - if (required_cols.empty()) - { - Block block; - added_columns.additional_filter_expression->execute(block); - result_column = block.getByPosition(0).column->cloneResized(selected_rows.size()); - break; - } - NameSet required_column_names; - for (auto & col : required_cols) - required_column_names.insert(col.name); - - Block executed_block; - size_t right_col_pos = 0; - for (const auto & col : sample_right_block.getColumnsWithTypeAndName()) - { - if (required_column_names.contains(col.name)) - { - auto new_col = col.column->cloneEmpty(); - for (const auto & selected_row : selected_rows) - { - const auto & src_col = selected_row.block->getByPosition(right_col_pos); - new_col->insertFrom(*src_col.column, selected_row.row_num); - } - executed_block.insert({std::move(new_col), col.type, col.name}); - } - right_col_pos += 1; - } - if (!executed_block) - { - result_column = ColumnUInt8::create(); - break; - } - - for (const auto & col_name : required_column_names) - { - const auto * src_col = added_columns.left_block.findByName(col_name); - if (!src_col) - continue; - auto new_col = src_col->column->cloneEmpty(); - size_t prev_left_offset = 0; - for (size_t i = 1; i < row_replicate_offset.size(); ++i) - { - const size_t & left_offset = row_replicate_offset[i]; - size_t rows = left_offset - prev_left_offset; - if (rows) - new_col->insertManyFrom(*src_col->column, left_start_row + i - 1, rows); - prev_left_offset = left_offset; - } - executed_block.insert({std::move(new_col), src_col->type, col_name}); - } - if (!executed_block) - { - throw Exception( - ErrorCodes::LOGICAL_ERROR, - "required columns: [{}], but not found any in left/right table. right table: {}, left table: {}", - required_cols.toString(), - sample_right_block.dumpNames(), - added_columns.left_block.dumpNames()); - } - - for (const auto & col : executed_block.getColumnsWithTypeAndName()) - if (!col.column || !col.type) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Illegal nullptr column in input block: {}", executed_block.dumpStructure()); - - added_columns.additional_filter_expression->execute(executed_block); - result_column = executed_block.getByPosition(0).column->convertToFullColumnIfConst(); - executed_block.clear(); - } while (false); - - result_column = result_column->convertToFullIfNeeded(); - if (result_column->isNullable()) - { - /// Convert Nullable(UInt8) to UInt8 ensuring that nulls are zeros - /// Trying to avoid copying data, since we are the only owner of the column. - ColumnPtr mask_column = assert_cast(*result_column).getNullMapColumnPtr(); - - MutableColumnPtr mutable_column; - { - ColumnPtr nested_column = assert_cast(*result_column).getNestedColumnPtr(); - result_column.reset(); - mutable_column = IColumn::mutate(std::move(nested_column)); - } - - auto & column_data = assert_cast(*mutable_column).getData(); - const auto & mask_column_data = assert_cast(*mask_column).getData(); - for (size_t i = 0; i < column_data.size(); ++i) - { - if (mask_column_data[i]) - column_data[i] = 0; - } - return mutable_column; - } - return result_column; -} - -/// Adapter class to pass into addFoundRowAll -/// In joinRightColumnsWithAdditionalFilter we don't want to add rows directly into AddedColumns, -/// because they need to be filtered by additional_filter_expression. -class PreSelectedRows : public std::vector -{ -public: - void appendFromBlock(const Block & block, size_t row_num, bool /* has_default */) { this->emplace_back(&block, row_num); } -}; - -/// First to collect all matched rows refs by join keys, then filter out rows which are not true in additional filter expression. -template < - typename KeyGetter, - typename Map, - bool need_replication, - typename AddedColumns> -NO_INLINE size_t joinRightColumnsWithAddtitionalFilter( - std::vector && key_getter_vector, - const std::vector & mapv, - AddedColumns & added_columns, - JoinStuff::JoinUsedFlags & used_flags [[maybe_unused]], - bool need_filter [[maybe_unused]], - bool need_flags [[maybe_unused]], - bool add_missing [[maybe_unused]], - bool flag_per_row [[maybe_unused]]) -{ - size_t left_block_rows = added_columns.rows_to_add; - if (need_filter) - added_columns.filter = IColumn::Filter(left_block_rows, 0); - - std::unique_ptr pool; - - if constexpr (need_replication) - added_columns.offsets_to_replicate = std::make_unique(left_block_rows); - - std::vector row_replicate_offset; - row_replicate_offset.reserve(left_block_rows); - - using FindResult = typename KeyGetter::FindResult; - size_t max_joined_block_rows = added_columns.max_joined_block_rows; - size_t left_row_iter = 0; - PreSelectedRows selected_rows; - selected_rows.reserve(left_block_rows); - std::vector find_results; - find_results.reserve(left_block_rows); - bool exceeded_max_block_rows = false; - IColumn::Offset total_added_rows = 0; - IColumn::Offset current_added_rows = 0; - - auto collect_keys_matched_rows_refs = [&]() - { - pool = std::make_unique(); - find_results.clear(); - row_replicate_offset.clear(); - row_replicate_offset.push_back(0); - current_added_rows = 0; - selected_rows.clear(); - for (; left_row_iter < left_block_rows; ++left_row_iter) - { - if constexpr (need_replication) - { - if (unlikely(total_added_rows + current_added_rows >= max_joined_block_rows)) - { - break; - } - } - KnownRowsHolder all_flag_known_rows; - KnownRowsHolder single_flag_know_rows; - for (size_t join_clause_idx = 0; join_clause_idx < added_columns.join_on_keys.size(); ++join_clause_idx) - { - const auto & join_keys = added_columns.join_on_keys[join_clause_idx]; - if (join_keys.null_map && (*join_keys.null_map)[left_row_iter]) - continue; - - bool row_acceptable = !join_keys.isRowFiltered(left_row_iter); - auto find_result = row_acceptable - ? key_getter_vector[join_clause_idx].findKey(*(mapv[join_clause_idx]), left_row_iter, *pool) - : FindResult(); - - if (find_result.isFound()) - { - auto & mapped = find_result.getMapped(); - find_results.push_back(find_result); - if (flag_per_row) - addFoundRowAll(mapped, selected_rows, current_added_rows, all_flag_known_rows, nullptr); - else - addFoundRowAll(mapped, selected_rows, current_added_rows, single_flag_know_rows, nullptr); - } - } - row_replicate_offset.push_back(current_added_rows); - } - }; - - auto copy_final_matched_rows = [&](size_t left_start_row, ColumnPtr filter_col) - { - const PaddedPODArray & filter_flags = assert_cast(*filter_col).getData(); - - size_t prev_replicated_row = 0; - auto selected_right_row_it = selected_rows.begin(); - size_t find_result_index = 0; - for (size_t i = 1, n = row_replicate_offset.size(); i < n; ++i) - { - bool any_matched = false; - /// For all right join, flag_per_row is true, we need mark used flags for each row. - if (flag_per_row) - { - for (size_t replicated_row = prev_replicated_row; replicated_row < row_replicate_offset[i]; ++replicated_row) - { - if (filter_flags[replicated_row]) - { - any_matched = true; - added_columns.appendFromBlock(*selected_right_row_it->block, selected_right_row_it->row_num, add_missing); - total_added_rows += 1; - if (need_flags) - used_flags.template setUsed(selected_right_row_it->block, selected_right_row_it->row_num, 0); - } - ++selected_right_row_it; - } - } - else - { - for (size_t replicated_row = prev_replicated_row; replicated_row < row_replicate_offset[i]; ++replicated_row) - { - if (filter_flags[replicated_row]) - { - any_matched = true; - added_columns.appendFromBlock(*selected_right_row_it->block, selected_right_row_it->row_num, add_missing); - total_added_rows += 1; - } - ++selected_right_row_it; - } - } - if (!any_matched) - { - if (add_missing) - addNotFoundRow(added_columns, total_added_rows); - else - addNotFoundRow(added_columns, total_added_rows); - } - else - { - if (!flag_per_row && need_flags) - used_flags.template setUsed(find_results[find_result_index]); - if (need_filter) - setUsed(added_columns.filter, left_start_row + i - 1); - if (add_missing) - added_columns.applyLazyDefaults(); - } - find_result_index += (prev_replicated_row != row_replicate_offset[i]); - - if constexpr (need_replication) - { - (*added_columns.offsets_to_replicate)[left_start_row + i - 1] = total_added_rows; - } - prev_replicated_row = row_replicate_offset[i]; - } - }; - - while (left_row_iter < left_block_rows && !exceeded_max_block_rows) - { - auto left_start_row = left_row_iter; - collect_keys_matched_rows_refs(); - if (selected_rows.size() != current_added_rows || row_replicate_offset.size() != left_row_iter - left_start_row + 1) - { - throw Exception( - ErrorCodes::LOGICAL_ERROR, - "Sizes are mismatched. selected_rows.size:{}, current_added_rows:{}, row_replicate_offset.size:{}, left_row_iter: {}, " - "left_start_row: {}", - selected_rows.size(), - current_added_rows, - row_replicate_offset.size(), - left_row_iter, - left_start_row); - } - auto filter_col = buildAdditionalFilter(left_start_row, selected_rows, row_replicate_offset, added_columns); - copy_final_matched_rows(left_start_row, filter_col); - - if constexpr (need_replication) - { - // Add a check for current_added_rows to avoid run the filter expression on too small size batch. - if (total_added_rows >= max_joined_block_rows || current_added_rows < 1024) - { - exceeded_max_block_rows = true; - } - } - } - - if constexpr (need_replication) - { - added_columns.offsets_to_replicate->resize_assume_reserved(left_row_iter); - added_columns.filter.resize_assume_reserved(left_row_iter); - } - added_columns.applyLazyDefaults(); - return left_row_iter; -} - -/// Joins right table columns which indexes are present in right_indexes using specified map. -/// Makes filter (1 if row presented in right table) and returns offsets to replicate (for ALL JOINS). -template -NO_INLINE size_t joinRightColumns( - std::vector && key_getter_vector, - const std::vector & mapv, - AddedColumns & added_columns, - JoinStuff::JoinUsedFlags & used_flags [[maybe_unused]]) -{ - constexpr JoinFeatures join_features; - - size_t rows = added_columns.rows_to_add; - if constexpr (need_filter) - added_columns.filter = IColumn::Filter(rows, 0); - - Arena pool; - - if constexpr (join_features.need_replication) - added_columns.offsets_to_replicate = std::make_unique(rows); - - IColumn::Offset current_offset = 0; - size_t max_joined_block_rows = added_columns.max_joined_block_rows; - size_t i = 0; - for (; i < rows; ++i) - { - if constexpr (join_features.need_replication) - { - if (unlikely(current_offset >= max_joined_block_rows)) - { - added_columns.offsets_to_replicate->resize_assume_reserved(i); - added_columns.filter.resize_assume_reserved(i); - break; - } - } - - bool right_row_found = false; - - KnownRowsHolder known_rows; - for (size_t onexpr_idx = 0; onexpr_idx < added_columns.join_on_keys.size(); ++onexpr_idx) - { - const auto & join_keys = added_columns.join_on_keys[onexpr_idx]; - if (join_keys.null_map && (*join_keys.null_map)[i]) - continue; - - bool row_acceptable = !join_keys.isRowFiltered(i); - using FindResult = typename KeyGetter::FindResult; - auto find_result = row_acceptable ? key_getter_vector[onexpr_idx].findKey(*(mapv[onexpr_idx]), i, pool) : FindResult(); - - if (find_result.isFound()) - { - right_row_found = true; - auto & mapped = find_result.getMapped(); - if constexpr (join_features.is_asof_join) - { - const IColumn & left_asof_key = added_columns.leftAsofKey(); - - auto row_ref = mapped->findAsof(left_asof_key, i); - if (row_ref.block) - { - setUsed(added_columns.filter, i); - if constexpr (flag_per_row) - used_flags.template setUsed(row_ref.block, row_ref.row_num, 0); - else - used_flags.template setUsed(find_result); - - added_columns.appendFromBlock(*row_ref.block, row_ref.row_num, join_features.add_missing); - } - else - addNotFoundRow(added_columns, current_offset); - } - else if constexpr (join_features.is_all_join) - { - setUsed(added_columns.filter, i); - used_flags.template setUsed(find_result); - auto used_flags_opt = join_features.need_flags ? &used_flags : nullptr; - addFoundRowAll(mapped, added_columns, current_offset, known_rows, used_flags_opt); - } - else if constexpr ((join_features.is_any_join || join_features.is_semi_join) && join_features.right) - { - /// Use first appeared left key + it needs left columns replication - bool used_once = used_flags.template setUsedOnce(find_result); - if (used_once) - { - auto used_flags_opt = join_features.need_flags ? &used_flags : nullptr; - setUsed(added_columns.filter, i); - addFoundRowAll(mapped, added_columns, current_offset, known_rows, used_flags_opt); - } - } - else if constexpr (join_features.is_any_join && KIND == JoinKind::Inner) - { - bool used_once = used_flags.template setUsedOnce(find_result); - - /// Use first appeared left key only - if (used_once) - { - setUsed(added_columns.filter, i); - added_columns.appendFromBlock(*mapped.block, mapped.row_num, join_features.add_missing); - } - - break; - } - else if constexpr (join_features.is_any_join && join_features.full) - { - /// TODO - } - else if constexpr (join_features.is_anti_join) - { - if constexpr (join_features.right && join_features.need_flags) - used_flags.template setUsed(find_result); - } - else /// ANY LEFT, SEMI LEFT, old ANY (RightAny) - { - setUsed(added_columns.filter, i); - used_flags.template setUsed(find_result); - added_columns.appendFromBlock(*mapped.block, mapped.row_num, join_features.add_missing); - - if (join_features.is_any_or_semi_join) - { - break; - } - } - } - } - - if (!right_row_found) - { - if constexpr (join_features.is_anti_join && join_features.left) - setUsed(added_columns.filter, i); - addNotFoundRow(added_columns, current_offset); - } - - if constexpr (join_features.need_replication) - { - (*added_columns.offsets_to_replicate)[i] = current_offset; - } - } - - added_columns.applyLazyDefaults(); - return i; -} - -template -size_t joinRightColumnsSwitchMultipleDisjuncts( - std::vector && key_getter_vector, - const std::vector & mapv, - AddedColumns & added_columns, - JoinStuff::JoinUsedFlags & used_flags [[maybe_unused]]) -{ - constexpr JoinFeatures join_features; - if constexpr (join_features.is_all_join) - { - if (added_columns.additional_filter_expression) - { - bool mark_per_row_used = join_features.right || join_features.full || mapv.size() > 1; - return joinRightColumnsWithAddtitionalFilter( - std::forward>(key_getter_vector), - mapv, - added_columns, - used_flags, - need_filter, - join_features.need_flags, - join_features.add_missing, - mark_per_row_used); - } - } - - if (added_columns.additional_filter_expression) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Additional filter expression is not supported for this JOIN"); - - return mapv.size() > 1 - ? joinRightColumns(std::forward>(key_getter_vector), mapv, added_columns, used_flags) - : joinRightColumns(std::forward>(key_getter_vector), mapv, added_columns, used_flags); -} - -template -size_t joinRightColumnsSwitchNullability( - std::vector && key_getter_vector, - const std::vector & mapv, - AddedColumns & added_columns, - JoinStuff::JoinUsedFlags & used_flags) -{ - if (added_columns.need_filter) - { - return joinRightColumnsSwitchMultipleDisjuncts(std::forward>(key_getter_vector), mapv, added_columns, used_flags); - } - else - { - return joinRightColumnsSwitchMultipleDisjuncts(std::forward>(key_getter_vector), mapv, added_columns, used_flags); - } -} - -template -size_t switchJoinRightColumns( - const std::vector & mapv, - AddedColumns & added_columns, - HashJoin::Type type, - JoinStuff::JoinUsedFlags & used_flags) -{ - constexpr bool is_asof_join = STRICTNESS == JoinStrictness::Asof; - switch (type) - { - case HashJoin::Type::EMPTY: - { - if constexpr (!is_asof_join) - { - using KeyGetter = KeyGetterEmpty; - std::vector key_getter_vector; - key_getter_vector.emplace_back(); - - using MapTypeVal = typename KeyGetter::MappedType; - std::vector a_map_type_vector; - a_map_type_vector.emplace_back(); - return joinRightColumnsSwitchNullability( - std::move(key_getter_vector), a_map_type_vector, added_columns, used_flags); - } - throw Exception(ErrorCodes::UNSUPPORTED_JOIN_KEYS, "Unsupported JOIN keys. Type: {}", type); - } - #define M(TYPE) \ - case HashJoin::Type::TYPE: \ - { \ - using MapTypeVal = const typename std::remove_reference_t::element_type; \ - using KeyGetter = typename KeyGetterForType::Type; \ - std::vector a_map_type_vector(mapv.size()); \ - std::vector key_getter_vector; \ - for (size_t d = 0; d < added_columns.join_on_keys.size(); ++d) \ - { \ - const auto & join_on_key = added_columns.join_on_keys[d]; \ - a_map_type_vector[d] = mapv[d]->TYPE.get(); \ - key_getter_vector.push_back(std::move(createKeyGetter(join_on_key.key_columns, join_on_key.key_sizes))); \ - } \ - return joinRightColumnsSwitchNullability( \ - std::move(key_getter_vector), a_map_type_vector, added_columns, used_flags); \ - } - APPLY_FOR_JOIN_VARIANTS(M) - #undef M - - default: - throw Exception(ErrorCodes::UNSUPPORTED_JOIN_KEYS, "Unsupported JOIN keys (type: {})", type); - } -} - -/** Since we do not store right key columns, - * this function is used to copy left key columns to right key columns. - * If the user requests some right columns, we just copy left key columns to right, since they are equal. - * Example: SELECT t1.key, t2.key FROM t1 FULL JOIN t2 ON t1.key = t2.key; - * In that case for matched rows in t2.key we will use values from t1.key. - * However, in some cases we might need to adjust the type of column, e.g. t1.key :: LowCardinality(String) and t2.key :: String - * Also, the nullability of the column might be different. - * Returns the right column after with necessary adjustments. - */ -ColumnWithTypeAndName copyLeftKeyColumnToRight( - const DataTypePtr & right_key_type, const String & renamed_right_column, const ColumnWithTypeAndName & left_column, const IColumn::Filter * null_map_filter = nullptr) -{ - ColumnWithTypeAndName right_column = left_column; - right_column.name = renamed_right_column; - - if (null_map_filter) - right_column.column = JoinCommon::filterWithBlanks(right_column.column, *null_map_filter); - - bool should_be_nullable = isNullableOrLowCardinalityNullable(right_key_type); - if (null_map_filter) - correctNullabilityInplace(right_column, should_be_nullable, *null_map_filter); - else - correctNullabilityInplace(right_column, should_be_nullable); - - if (!right_column.type->equals(*right_key_type)) - { - right_column.column = castColumnAccurate(right_column, right_key_type); - right_column.type = right_key_type; - } - - right_column.column = right_column.column->convertToFullColumnIfConst(); - return right_column; -} - -/// Cut first num_rows rows from block in place and returns block with remaining rows -Block sliceBlock(Block & block, size_t num_rows) -{ - size_t total_rows = block.rows(); - if (num_rows >= total_rows) - return {}; - size_t remaining_rows = total_rows - num_rows; - Block remaining_block = block.cloneEmpty(); - for (size_t i = 0; i < block.columns(); ++i) - { - auto & col = block.getByPosition(i); - remaining_block.getByPosition(i).column = col.column->cut(num_rows, remaining_rows); - col.column = col.column->cut(0, num_rows); - } - return remaining_block; -} - -} /// nameless - -template -Block HashJoin::joinBlockImpl( - Block & block, - const Block & block_with_columns_to_add, - const std::vector & maps_, - bool is_join_get) const -{ - constexpr JoinFeatures join_features; - - std::vector join_on_keys; - const auto & onexprs = table_join->getClauses(); - for (size_t i = 0; i < onexprs.size(); ++i) - { - const auto & key_names = !is_join_get ? onexprs[i].key_names_left : onexprs[i].key_names_right; - join_on_keys.emplace_back(block, key_names, onexprs[i].condColumnNames().first, key_sizes[i]); - } - size_t existing_columns = block.columns(); - - /** If you use FULL or RIGHT JOIN, then the columns from the "left" table must be materialized. - * Because if they are constants, then in the "not joined" rows, they may have different values - * - default values, which can differ from the values of these constants. - */ - if constexpr (join_features.right || join_features.full) - { - materializeBlockInplace(block); - } - - /** For LEFT/INNER JOIN, the saved blocks do not contain keys. - * For FULL/RIGHT JOIN, the saved blocks contain keys; - * but they will not be used at this stage of joining (and will be in `AdderNonJoined`), and they need to be skipped. - * For ASOF, the last column is used as the ASOF column - */ - AddedColumns added_columns( - block, - block_with_columns_to_add, - savedBlockSample(), - *this, - std::move(join_on_keys), - table_join->getMixedJoinExpression(), - join_features.is_asof_join, - is_join_get); - - bool has_required_right_keys = (required_right_keys.columns() != 0); - added_columns.need_filter = join_features.need_filter || has_required_right_keys; - added_columns.max_joined_block_rows = max_joined_block_rows; - if (!added_columns.max_joined_block_rows) - added_columns.max_joined_block_rows = std::numeric_limits::max(); - else - added_columns.reserve(join_features.need_replication); - - size_t num_joined = switchJoinRightColumns(maps_, added_columns, data->type, used_flags); - /// Do not hold memory for join_on_keys anymore - added_columns.join_on_keys.clear(); - Block remaining_block = sliceBlock(block, num_joined); - - added_columns.buildOutput(); - for (size_t i = 0; i < added_columns.size(); ++i) - block.insert(added_columns.moveColumn(i)); - - std::vector right_keys_to_replicate [[maybe_unused]]; - - if constexpr (join_features.need_filter) - { - /// If ANY INNER | RIGHT JOIN - filter all the columns except the new ones. - for (size_t i = 0; i < existing_columns; ++i) - block.safeGetByPosition(i).column = block.safeGetByPosition(i).column->filter(added_columns.filter, -1); - - /// Add join key columns from right block if needed using value from left table because of equality - for (size_t i = 0; i < required_right_keys.columns(); ++i) - { - const auto & right_key = required_right_keys.getByPosition(i); - /// asof column is already in block. - if (join_features.is_asof_join && right_key.name == table_join->getOnlyClause().key_names_right.back()) - continue; - - const auto & left_column = block.getByName(required_right_keys_sources[i]); - const auto & right_col_name = getTableJoin().renamedRightColumnName(right_key.name); - auto right_col = copyLeftKeyColumnToRight(right_key.type, right_col_name, left_column); - block.insert(std::move(right_col)); - } - } - else if (has_required_right_keys) - { - /// Add join key columns from right block if needed. - for (size_t i = 0; i < required_right_keys.columns(); ++i) - { - const auto & right_key = required_right_keys.getByPosition(i); - auto right_col_name = getTableJoin().renamedRightColumnName(right_key.name); - /// asof column is already in block. - if (join_features.is_asof_join && right_key.name == table_join->getOnlyClause().key_names_right.back()) - continue; - - const auto & left_column = block.getByName(required_right_keys_sources[i]); - auto right_col = copyLeftKeyColumnToRight(right_key.type, right_col_name, left_column, &added_columns.filter); - block.insert(std::move(right_col)); - - if constexpr (join_features.need_replication) - right_keys_to_replicate.push_back(block.getPositionByName(right_col_name)); - } - } - - if constexpr (join_features.need_replication) - { - std::unique_ptr & offsets_to_replicate = added_columns.offsets_to_replicate; - - /// If ALL ... JOIN - we replicate all the columns except the new ones. - for (size_t i = 0; i < existing_columns; ++i) - { - block.safeGetByPosition(i).column = block.safeGetByPosition(i).column->replicate(*offsets_to_replicate); - } - - /// Replicate additional right keys - for (size_t pos : right_keys_to_replicate) - { - block.safeGetByPosition(pos).column = block.safeGetByPosition(pos).column->replicate(*offsets_to_replicate); - } - } - - return remaining_block; -} - -void HashJoin::joinBlockImplCross(Block & block, ExtraBlockPtr & not_processed) const -{ - size_t start_left_row = 0; - size_t start_right_block = 0; - std::unique_ptr reader = nullptr; - if (not_processed) - { - auto & continuation = static_cast(*not_processed); - start_left_row = continuation.left_position; - start_right_block = continuation.right_block; - reader = std::move(continuation.reader); - not_processed.reset(); - } - - size_t num_existing_columns = block.columns(); - size_t num_columns_to_add = sample_block_with_columns_to_add.columns(); - - ColumnRawPtrs src_left_columns; - MutableColumns dst_columns; - - { - src_left_columns.reserve(num_existing_columns); - dst_columns.reserve(num_existing_columns + num_columns_to_add); - - for (const ColumnWithTypeAndName & left_column : block) - { - src_left_columns.push_back(left_column.column.get()); - dst_columns.emplace_back(src_left_columns.back()->cloneEmpty()); - } - - for (const ColumnWithTypeAndName & right_column : sample_block_with_columns_to_add) - dst_columns.emplace_back(right_column.column->cloneEmpty()); - - for (auto & dst : dst_columns) - dst->reserve(max_joined_block_rows); - } - - size_t rows_left = block.rows(); - size_t rows_added = 0; - for (size_t left_row = start_left_row; left_row < rows_left; ++left_row) - { - size_t block_number = 0; - - auto process_right_block = [&](const Block & block_right) - { - size_t rows_right = block_right.rows(); - rows_added += rows_right; - - for (size_t col_num = 0; col_num < num_existing_columns; ++col_num) - dst_columns[col_num]->insertManyFrom(*src_left_columns[col_num], left_row, rows_right); - - for (size_t col_num = 0; col_num < num_columns_to_add; ++col_num) - { - const IColumn & column_right = *block_right.getByPosition(col_num).column; - dst_columns[num_existing_columns + col_num]->insertRangeFrom(column_right, 0, rows_right); - } - }; - - for (const Block & block_right : data->blocks) - { - ++block_number; - if (block_number < start_right_block) - continue; - - /// The following statement cannot be substituted with `process_right_block(!have_compressed ? block_right : block_right.decompress())` - /// because it will lead to copying of `block_right` even if its branch is taken (because common type of `block_right` and `block_right.decompress()` is `Block`). - if (!have_compressed) - process_right_block(block_right); - else - process_right_block(block_right.decompress()); - - if (rows_added > max_joined_block_rows) - { - break; - } - } - - if (tmp_stream && rows_added <= max_joined_block_rows) - { - if (reader == nullptr) - { - tmp_stream->finishWritingAsyncSafe(); - reader = tmp_stream->getReadStream(); - } - while (auto block_right = reader->read()) - { - ++block_number; - process_right_block(block_right); - if (rows_added > max_joined_block_rows) - { - break; - } - } - - /// It means, that reader->read() returned {} - if (rows_added <= max_joined_block_rows) - { - reader.reset(); - } - } - - start_right_block = 0; - - if (rows_added > max_joined_block_rows) - { - not_processed = std::make_shared( - NotProcessedCrossJoin{{block.cloneEmpty()}, left_row, block_number + 1, std::move(reader)}); - not_processed->block.swap(block); - break; - } - } - - for (const ColumnWithTypeAndName & src_column : sample_block_with_columns_to_add) - block.insert(src_column); - - block = block.cloneWithColumns(std::move(dst_columns)); -} - -DataTypePtr HashJoin::joinGetCheckAndGetReturnType(const DataTypes & data_types, const String & column_name, bool or_null) const -{ - size_t num_keys = data_types.size(); - if (right_table_keys.columns() != num_keys) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, - "Number of arguments for function joinGet{} doesn't match: passed, should be equal to {}", - toString(or_null ? "OrNull" : ""), toString(num_keys)); - - for (size_t i = 0; i < num_keys; ++i) - { - const auto & left_type_origin = data_types[i]; - const auto & [c2, right_type_origin, right_name] = right_table_keys.safeGetByPosition(i); - auto left_type = removeNullable(recursiveRemoveLowCardinality(left_type_origin)); - auto right_type = removeNullable(recursiveRemoveLowCardinality(right_type_origin)); - if (!left_type->equals(*right_type)) - throw Exception(ErrorCodes::TYPE_MISMATCH, "Type mismatch in joinGet key {}: " - "found type {}, while the needed type is {}", i, left_type->getName(), right_type->getName()); - } - - if (!sample_block_with_columns_to_add.has(column_name)) - throw Exception(ErrorCodes::NO_SUCH_COLUMN_IN_TABLE, "StorageJoin doesn't contain column {}", column_name); - - auto elem = sample_block_with_columns_to_add.getByName(column_name); - if (or_null && JoinCommon::canBecomeNullable(elem.type)) - elem.type = makeNullable(elem.type); - return elem.type; -} - -/// TODO: return multiple columns as named tuple -/// TODO: return array of values when strictness == JoinStrictness::All -ColumnWithTypeAndName HashJoin::joinGet(const Block & block, const Block & block_with_columns_to_add) const -{ - bool is_valid = (strictness == JoinStrictness::Any || strictness == JoinStrictness::RightAny) - && kind == JoinKind::Left; - if (!is_valid) - throw Exception(ErrorCodes::INCOMPATIBLE_TYPE_OF_JOIN, "joinGet only supports StorageJoin of type Left Any"); - const auto & key_names_right = table_join->getOnlyClause().key_names_right; - - /// Assemble the key block with correct names. - Block keys; - for (size_t i = 0; i < block.columns(); ++i) - { - auto key = block.getByPosition(i); - key.name = key_names_right[i]; - keys.insert(std::move(key)); - } - - static_assert(!MapGetter::flagged, - "joinGet are not protected from hash table changes between block processing"); - - std::vector maps_vector; - maps_vector.push_back(&std::get(data->maps[0])); - joinBlockImpl( - keys, block_with_columns_to_add, maps_vector, /* is_join_get = */ true); - return keys.getByPosition(keys.columns() - 1); -} - -void HashJoin::checkTypesOfKeys(const Block & block) const -{ - for (const auto & onexpr : table_join->getClauses()) - { - JoinCommon::checkTypesOfKeys(block, onexpr.key_names_left, right_table_keys, onexpr.key_names_right); - } -} - -void HashJoin::joinBlock(Block & block, ExtraBlockPtr & not_processed) -{ - if (!data) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot join after data has been released"); - - for (const auto & onexpr : table_join->getClauses()) - { - auto cond_column_name = onexpr.condColumnNames(); - JoinCommon::checkTypesOfKeys( - block, onexpr.key_names_left, cond_column_name.first, - right_sample_block, onexpr.key_names_right, cond_column_name.second); - } - - if (kind == JoinKind::Cross) - { - joinBlockImplCross(block, not_processed); - return; - } - - if (kind == JoinKind::Right || kind == JoinKind::Full) - { - materializeBlockInplace(block); - } - - { - std::vectormaps[0])> * > maps_vector; - for (size_t i = 0; i < table_join->getClauses().size(); ++i) - maps_vector.push_back(&data->maps[i]); - - if (joinDispatch(kind, strictness, maps_vector, [&](auto kind_, auto strictness_, auto & maps_vector_) - { - Block remaining_block = joinBlockImpl(block, sample_block_with_columns_to_add, maps_vector_); - if (remaining_block.rows()) - not_processed = std::make_shared(ExtraBlock{std::move(remaining_block)}); - else - not_processed.reset(); - })) - { - /// Joined - } - else - throw Exception(ErrorCodes::LOGICAL_ERROR, "Wrong JOIN combination: {} {}", strictness, kind); - } -} - -HashJoin::~HashJoin() -{ - if (!data) - { - LOG_TEST(log, "{}Join data has been already released", instance_log_id); - return; - } - LOG_TEST( - log, - "{}Join data is being destroyed, {} bytes and {} rows in hash table", - instance_log_id, - getTotalByteCount(), - getTotalRowCount()); -} - -template -struct AdderNonJoined -{ - static void add(const Mapped & mapped, size_t & rows_added, MutableColumns & columns_right) - { - constexpr bool mapped_asof = std::is_same_v; - [[maybe_unused]] constexpr bool mapped_one = std::is_same_v; - - if constexpr (mapped_asof) - { - /// Do nothing - } - else if constexpr (mapped_one) - { - for (size_t j = 0; j < columns_right.size(); ++j) - { - const auto & mapped_column = mapped.block->getByPosition(j).column; - columns_right[j]->insertFrom(*mapped_column, mapped.row_num); - } - - ++rows_added; - } - else - { - for (auto it = mapped.begin(); it.ok(); ++it) - { - for (size_t j = 0; j < columns_right.size(); ++j) - { - const auto & mapped_column = it->block->getByPosition(j).column; - columns_right[j]->insertFrom(*mapped_column, it->row_num); - } - - ++rows_added; - } - } - } -}; - -/// Stream from not joined earlier rows of the right table. -/// Based on: -/// - map offsetInternal saved in used_flags for single disjuncts -/// - flags in BlockWithFlags for multiple disjuncts -class NotJoinedHash final : public NotJoinedBlocks::RightColumnsFiller -{ -public: - NotJoinedHash(const HashJoin & parent_, UInt64 max_block_size_, bool flag_per_row_) - : parent(parent_) - , max_block_size(max_block_size_) - , flag_per_row(flag_per_row_) - , current_block_start(0) - { - if (parent.data == nullptr) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot join after data has been released"); - } - - Block getEmptyBlock() override { return parent.savedBlockSample().cloneEmpty(); } - - size_t fillColumns(MutableColumns & columns_right) override - { - size_t rows_added = 0; - if (unlikely(parent.data->type == HashJoin::Type::EMPTY)) - { - rows_added = fillColumnsFromData(parent.data->blocks, columns_right); - } - else - { - auto fill_callback = [&](auto, auto, auto & map) - { - rows_added = fillColumnsFromMap(map, columns_right); - }; - - if (!joinDispatch(parent.kind, parent.strictness, parent.data->maps.front(), fill_callback)) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown JOIN strictness '{}' (must be on of: ANY, ALL, ASOF)", parent.strictness); - } - - if (!flag_per_row) - { - fillNullsFromBlocks(columns_right, rows_added); - } - - return rows_added; - } - -private: - const HashJoin & parent; - UInt64 max_block_size; - bool flag_per_row; - - size_t current_block_start; - - std::any position; - std::optional nulls_position; - std::optional used_position; - - size_t fillColumnsFromData(const BlocksList & blocks, MutableColumns & columns_right) - { - if (!position.has_value()) - position = std::make_any(blocks.begin()); - - auto & block_it = std::any_cast(position); - auto end = blocks.end(); - - size_t rows_added = 0; - for (; block_it != end; ++block_it) - { - size_t rows_from_block = std::min(max_block_size - rows_added, block_it->rows() - current_block_start); - for (size_t j = 0; j < columns_right.size(); ++j) - { - const auto & col = block_it->getByPosition(j).column; - columns_right[j]->insertRangeFrom(*col, current_block_start, rows_from_block); - } - rows_added += rows_from_block; - - if (rows_added >= max_block_size) - { - /// How many rows have been read - current_block_start += rows_from_block; - if (block_it->rows() <= current_block_start) - { - /// current block was fully read - ++block_it; - current_block_start = 0; - } - break; - } - current_block_start = 0; - } - return rows_added; - } - - template - size_t fillColumnsFromMap(const Maps & maps, MutableColumns & columns_keys_and_right) - { - switch (parent.data->type) - { - #define M(TYPE) \ - case HashJoin::Type::TYPE: \ - return fillColumns(*maps.TYPE, columns_keys_and_right); - APPLY_FOR_JOIN_VARIANTS(M) - #undef M - default: - throw Exception(ErrorCodes::UNSUPPORTED_JOIN_KEYS, "Unsupported JOIN keys (type: {})", parent.data->type); - } - } - - template - size_t fillColumns(const Map & map, MutableColumns & columns_keys_and_right) - { - size_t rows_added = 0; - - if (flag_per_row) - { - if (!used_position.has_value()) - used_position = parent.data->blocks.begin(); - - auto end = parent.data->blocks.end(); - - for (auto & it = *used_position; it != end && rows_added < max_block_size; ++it) - { - const Block & mapped_block = *it; - - for (size_t row = 0; row < mapped_block.rows(); ++row) - { - if (!parent.isUsed(&mapped_block, row)) - { - for (size_t colnum = 0; colnum < columns_keys_and_right.size(); ++colnum) - { - columns_keys_and_right[colnum]->insertFrom(*mapped_block.getByPosition(colnum).column, row); - } - - ++rows_added; - } - } - } - } - else - { - using Mapped = typename Map::mapped_type; - using Iterator = typename Map::const_iterator; - - - if (!position.has_value()) - position = std::make_any(map.begin()); - - Iterator & it = std::any_cast(position); - auto end = map.end(); - - for (; it != end; ++it) - { - const Mapped & mapped = it->getMapped(); - - size_t offset = map.offsetInternal(it.getPtr()); - if (parent.isUsed(offset)) - continue; - AdderNonJoined::add(mapped, rows_added, columns_keys_and_right); - - if (rows_added >= max_block_size) - { - ++it; - break; - } - } - } - - return rows_added; - } - - void fillNullsFromBlocks(MutableColumns & columns_keys_and_right, size_t & rows_added) - { - if (!nulls_position.has_value()) - nulls_position = parent.data->blocks_nullmaps.begin(); - - auto end = parent.data->blocks_nullmaps.end(); - - for (auto & it = *nulls_position; it != end && rows_added < max_block_size; ++it) - { - const auto * block = it->first; - ConstNullMapPtr nullmap = nullptr; - if (it->second) - nullmap = &assert_cast(*it->second).getData(); - - for (size_t row = 0; row < block->rows(); ++row) - { - if (nullmap && (*nullmap)[row]) - { - for (size_t col = 0; col < columns_keys_and_right.size(); ++col) - columns_keys_and_right[col]->insertFrom(*block->getByPosition(col).column, row); - ++rows_added; - } - } - } - } -}; - -IBlocksStreamPtr HashJoin::getNonJoinedBlocks(const Block & left_sample_block, - const Block & result_sample_block, - UInt64 max_block_size) const -{ - if (!JoinCommon::hasNonJoinedBlocks(*table_join)) - return {}; - size_t left_columns_count = left_sample_block.columns(); - - bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); - if (!flag_per_row) - { - /// With multiple disjuncts, all keys are in sample_block_with_columns_to_add, so invariant is not held - size_t expected_columns_count = left_columns_count + required_right_keys.columns() + sample_block_with_columns_to_add.columns(); - if (expected_columns_count != result_sample_block.columns()) - { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected number of columns in result sample block: {} instead of {} ({} + {} + {})", - result_sample_block.columns(), expected_columns_count, - left_columns_count, required_right_keys.columns(), sample_block_with_columns_to_add.columns()); - } - } - - auto non_joined = std::make_unique(*this, max_block_size, flag_per_row); - return std::make_unique(std::move(non_joined), result_sample_block, left_columns_count, *table_join); -} - -void HashJoin::reuseJoinedData(const HashJoin & join) -{ - data = join.data; - from_storage_join = true; - - bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); - if (flag_per_row) - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "StorageJoin with ORs is not supported"); - - for (auto & map : data->maps) - { - joinDispatch(kind, strictness, map, [this](auto kind_, auto strictness_, auto & map_) - { - used_flags.reinit(map_.getBufferSizeInCells(data->type) + 1); - }); - } -} - -BlocksList HashJoin::releaseJoinedBlocks(bool restructure) -{ - LOG_TRACE(log, "{}Join data is being released, {} bytes and {} rows in hash table", instance_log_id, getTotalByteCount(), getTotalRowCount()); - - BlocksList right_blocks = std::move(data->blocks); - if (!restructure) - { - data.reset(); - return right_blocks; - } - - data->maps.clear(); - data->blocks_nullmaps.clear(); - - BlocksList restored_blocks; - - /// names to positions optimization - std::vector positions; - std::vector is_nullable; - if (!right_blocks.empty()) - { - positions.reserve(right_sample_block.columns()); - const Block & tmp_block = *right_blocks.begin(); - for (const auto & sample_column : right_sample_block) - { - positions.emplace_back(tmp_block.getPositionByName(sample_column.name)); - is_nullable.emplace_back(isNullableOrLowCardinalityNullable(sample_column.type)); - } - } - - for (Block & saved_block : right_blocks) - { - Block restored_block; - for (size_t i = 0; i < positions.size(); ++i) - { - auto & column = saved_block.getByPosition(positions[i]); - correctNullabilityInplace(column, is_nullable[i]); - restored_block.insert(column); - } - restored_blocks.emplace_back(std::move(restored_block)); - } - - data.reset(); - return restored_blocks; -} - -const ColumnWithTypeAndName & HashJoin::rightAsofKeyColumn() const -{ - /// It should be nullable when right side is nullable - return savedBlockSample().getByName(table_join->getOnlyClause().key_names_right.back()); -} - -void HashJoin::validateAdditionalFilterExpression(ExpressionActionsPtr additional_filter_expression) -{ - if (!additional_filter_expression) - return; - - Block expression_sample_block = additional_filter_expression->getSampleBlock(); - - if (expression_sample_block.columns() != 1) - { - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Unexpected expression in JOIN ON section. Expected single column, got '{}'", - expression_sample_block.dumpStructure()); - } - - auto type = removeNullable(expression_sample_block.getByPosition(0).type); - if (!type->equals(*std::make_shared())) - { - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Unexpected expression in JOIN ON section. Expected boolean (UInt8), got '{}'. expression:\n{}", - expression_sample_block.getByPosition(0).type->getName(), - additional_filter_expression->dumpActions()); - } - - bool is_supported = (strictness == JoinStrictness::All) && (isInnerOrLeft(kind) || isRightOrFull(kind)); - if (!is_supported) - { - throw Exception(ErrorCodes::INVALID_JOIN_ON_EXPRESSION, - "Non equi condition '{}' from JOIN ON section is supported only for ALL INNER/LEFT/FULL/RIGHT JOINs", - expression_sample_block.getByPosition(0).name); - } -} - -bool HashJoin::needUsedFlagsForPerRightTableRow(std::shared_ptr table_join_) const -{ - if (!table_join_->oneDisjunct()) - return true; - /// If it'a a all right join with inequal conditions, we need to mark each row - if (table_join_->getMixedJoinExpression() && isRightOrFull(table_join_->kind())) - return true; - return false; -} - -} diff --git a/src/Interpreters/HashJoin/AddedColumns.cpp b/src/Interpreters/HashJoin/AddedColumns.cpp new file mode 100644 index 00000000000..930a352744d --- /dev/null +++ b/src/Interpreters/HashJoin/AddedColumns.cpp @@ -0,0 +1,138 @@ +#include +#include + +namespace DB +{ +JoinOnKeyColumns::JoinOnKeyColumns(const Block & block, const Names & key_names_, const String & cond_column_name, const Sizes & key_sizes_) + : key_names(key_names_) + , materialized_keys_holder(JoinCommon::materializeColumns( + block, key_names)) /// Rare case, when keys are constant or low cardinality. To avoid code bloat, simply materialize them. + , key_columns(JoinCommon::getRawPointers(materialized_keys_holder)) + , null_map(nullptr) + , null_map_holder(extractNestedColumnsAndNullMap(key_columns, null_map)) + , join_mask_column(JoinCommon::getColumnAsMask(block, cond_column_name)) + , key_sizes(key_sizes_) +{ +} + +template<> void AddedColumns::buildOutput() +{ +} + +template<> +void AddedColumns::buildOutput() +{ + for (size_t i = 0; i < this->size(); ++i) + { + auto& col = columns[i]; + size_t default_count = 0; + auto apply_default = [&]() + { + if (default_count > 0) + { + JoinCommon::addDefaultValues(*col, type_name[i].type, default_count); + default_count = 0; + } + }; + + for (size_t j = 0; j < lazy_output.blocks.size(); ++j) + { + if (!lazy_output.blocks[j]) + { + default_count++; + continue; + } + apply_default(); + const auto & column_from_block = reinterpret_cast(lazy_output.blocks[j])->getByPosition(right_indexes[i]); + /// If it's joinGetOrNull, we need to wrap not-nullable columns in StorageJoin. + if (is_join_get) + { + if (auto * nullable_col = typeid_cast(col.get()); + nullable_col && !column_from_block.column->isNullable()) + { + nullable_col->insertFromNotNullable(*column_from_block.column, lazy_output.row_nums[j]); + continue; + } + } + col->insertFrom(*column_from_block.column, lazy_output.row_nums[j]); + } + apply_default(); + } +} + +template<> +void AddedColumns::applyLazyDefaults() +{ + if (lazy_defaults_count) + { + for (size_t j = 0, size = right_indexes.size(); j < size; ++j) + JoinCommon::addDefaultValues(*columns[j], type_name[j].type, lazy_defaults_count); + lazy_defaults_count = 0; + } +} + +template<> +void AddedColumns::applyLazyDefaults() +{ +} + +template <> +void AddedColumns::appendFromBlock(const Block & block, size_t row_num,const bool has_defaults) +{ + if (has_defaults) + applyLazyDefaults(); + +#ifndef NDEBUG + checkBlock(block); +#endif + if (is_join_get) + { + size_t right_indexes_size = right_indexes.size(); + for (size_t j = 0; j < right_indexes_size; ++j) + { + const auto & column_from_block = block.getByPosition(right_indexes[j]); + if (auto * nullable_col = nullable_column_ptrs[j]) + nullable_col->insertFromNotNullable(*column_from_block.column, row_num); + else + columns[j]->insertFrom(*column_from_block.column, row_num); + } + } + else + { + size_t right_indexes_size = right_indexes.size(); + for (size_t j = 0; j < right_indexes_size; ++j) + { + const auto & column_from_block = block.getByPosition(right_indexes[j]); + columns[j]->insertFrom(*column_from_block.column, row_num); + } + } +} + +template <> +void AddedColumns::appendFromBlock(const Block & block, size_t row_num, bool) +{ +#ifndef NDEBUG + checkBlock(block); +#endif + if (has_columns_to_add) + { + lazy_output.blocks.emplace_back(reinterpret_cast(&block)); + lazy_output.row_nums.emplace_back(static_cast(row_num)); + } +} +template<> +void AddedColumns::appendDefaultRow() +{ + ++lazy_defaults_count; +} + +template<> +void AddedColumns::appendDefaultRow() +{ + if (has_columns_to_add) + { + lazy_output.blocks.emplace_back(0); + lazy_output.row_nums.emplace_back(0); + } +} +} diff --git a/src/Interpreters/HashJoin/AddedColumns.h b/src/Interpreters/HashJoin/AddedColumns.h new file mode 100644 index 00000000000..13a7df6f498 --- /dev/null +++ b/src/Interpreters/HashJoin/AddedColumns.h @@ -0,0 +1,226 @@ +#pragma once +#include +#include + +namespace DB +{ +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +class ExpressionActions; +using ExpressionActionsPtr = std::shared_ptr; + +struct JoinOnKeyColumns +{ + Names key_names; + + Columns materialized_keys_holder; + ColumnRawPtrs key_columns; + + ConstNullMapPtr null_map; + ColumnPtr null_map_holder; + + /// Only rows where mask == true can be joined + JoinCommon::JoinMask join_mask_column; + + Sizes key_sizes; + + explicit JoinOnKeyColumns(const Block & block, const Names & key_names_, const String & cond_column_name, const Sizes & key_sizes_); + + bool isRowFiltered(size_t i) const { return join_mask_column.isRowFiltered(i); } +}; + +template +class AddedColumns +{ +public: + struct TypeAndName + { + DataTypePtr type; + String name; + String qualified_name; + + TypeAndName(DataTypePtr type_, const String & name_, const String & qualified_name_) + : type(type_), name(name_), qualified_name(qualified_name_) + { + } + }; + + struct LazyOutput + { + PaddedPODArray blocks; + PaddedPODArray row_nums; + }; + + AddedColumns( + const Block & left_block_, + const Block & block_with_columns_to_add, + const Block & saved_block_sample, + const HashJoin & join, + std::vector && join_on_keys_, + ExpressionActionsPtr additional_filter_expression_, + bool is_asof_join, + bool is_join_get_) + : left_block(left_block_) + , join_on_keys(join_on_keys_) + , additional_filter_expression(additional_filter_expression_) + , rows_to_add(left_block.rows()) + , is_join_get(is_join_get_) + { + size_t num_columns_to_add = block_with_columns_to_add.columns(); + if (is_asof_join) + ++num_columns_to_add; + + if constexpr (lazy) + { + has_columns_to_add = num_columns_to_add > 0; + lazy_output.blocks.reserve(rows_to_add); + lazy_output.row_nums.reserve(rows_to_add); + } + + columns.reserve(num_columns_to_add); + type_name.reserve(num_columns_to_add); + right_indexes.reserve(num_columns_to_add); + + for (const auto & src_column : block_with_columns_to_add) + { + /// Column names `src_column.name` and `qualified_name` can differ for StorageJoin, + /// because it uses not qualified right block column names + auto qualified_name = join.getTableJoin().renamedRightColumnName(src_column.name); + /// Don't insert column if it's in left block + if (!left_block.has(qualified_name)) + addColumn(src_column, qualified_name); + } + + if (is_asof_join) + { + assert(join_on_keys.size() == 1); + const ColumnWithTypeAndName & right_asof_column = join.rightAsofKeyColumn(); + addColumn(right_asof_column, right_asof_column.name); + left_asof_key = join_on_keys[0].key_columns.back(); + } + + for (auto & tn : type_name) + right_indexes.push_back(saved_block_sample.getPositionByName(tn.name)); + + nullable_column_ptrs.resize(right_indexes.size(), nullptr); + for (size_t j = 0; j < right_indexes.size(); ++j) + { + /** If it's joinGetOrNull, we will have nullable columns in result block + * even if right column is not nullable in storage (saved_block_sample). + */ + const auto & saved_column = saved_block_sample.getByPosition(right_indexes[j]).column; + if (columns[j]->isNullable() && !saved_column->isNullable()) + nullable_column_ptrs[j] = typeid_cast(columns[j].get()); + } + } + + size_t size() const { return columns.size(); } + + void buildOutput(); + + ColumnWithTypeAndName moveColumn(size_t i) + { + return ColumnWithTypeAndName(std::move(columns[i]), type_name[i].type, type_name[i].qualified_name); + } + + void appendFromBlock(const Block & block, size_t row_num, bool has_default); + + void appendDefaultRow(); + + void applyLazyDefaults(); + + const IColumn & leftAsofKey() const { return *left_asof_key; } + + Block left_block; + std::vector join_on_keys; + ExpressionActionsPtr additional_filter_expression; + + size_t max_joined_block_rows = 0; + size_t rows_to_add; + std::unique_ptr offsets_to_replicate; + bool need_filter = false; + IColumn::Filter filter; + + void reserve(bool need_replicate) + { + if (!max_joined_block_rows) + return; + + /// Do not allow big allocations when user set max_joined_block_rows to huge value + size_t reserve_size = std::min(max_joined_block_rows, DEFAULT_BLOCK_SIZE * 2); + + if (need_replicate) + /// Reserve 10% more space for columns, because some rows can be repeated + reserve_size = static_cast(1.1 * reserve_size); + + for (auto & column : columns) + column->reserve(reserve_size); + } + +private: + + void checkBlock(const Block & block) + { + for (size_t j = 0; j < right_indexes.size(); ++j) + { + const auto * column_from_block = block.getByPosition(right_indexes[j]).column.get(); + const auto * dest_column = columns[j].get(); + if (auto * nullable_col = nullable_column_ptrs[j]) + { + if (!is_join_get) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Columns {} and {} can have different nullability only in joinGetOrNull", + dest_column->getName(), column_from_block->getName()); + dest_column = nullable_col->getNestedColumnPtr().get(); + } + /** Using dest_column->structureEquals(*column_from_block) will not work for low cardinality columns, + * because dictionaries can be different, while calling insertFrom on them is safe, for example: + * ColumnLowCardinality(size = 0, UInt8(size = 0), ColumnUnique(size = 1, String(size = 1))) + * and + * ColumnLowCardinality(size = 0, UInt16(size = 0), ColumnUnique(size = 1, String(size = 1))) + */ + if (typeid(*dest_column) != typeid(*column_from_block)) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Columns {} and {} have different types {} and {}", + dest_column->getName(), column_from_block->getName(), + demangle(typeid(*dest_column).name()), demangle(typeid(*column_from_block).name())); + } + } + + MutableColumns columns; + bool is_join_get; + std::vector right_indexes; + std::vector type_name; + std::vector nullable_column_ptrs; + size_t lazy_defaults_count = 0; + + /// for lazy + // The default row is represented by an empty RowRef, so that fixed-size blocks can be generated sequentially, + // default_count cannot represent the position of the row + LazyOutput lazy_output; + bool has_columns_to_add; + + /// for ASOF + const IColumn * left_asof_key = nullptr; + + + void addColumn(const ColumnWithTypeAndName & src_column, const std::string & qualified_name) + { + columns.push_back(src_column.column->cloneEmpty()); + columns.back()->reserve(src_column.column->size()); + type_name.emplace_back(src_column.type, src_column.name, qualified_name); + } +}; + +/// Adapter class to pass into addFoundRowAll +/// In joinRightColumnsWithAdditionalFilter we don't want to add rows directly into AddedColumns, +/// because they need to be filtered by additional_filter_expression. +class PreSelectedRows : public std::vector +{ +public: + void appendFromBlock(const Block & block, size_t row_num, bool /* has_default */) { this->emplace_back(&block, row_num); } +}; + +} diff --git a/src/Interpreters/HashJoin/FullHashJoin.cpp b/src/Interpreters/HashJoin/FullHashJoin.cpp new file mode 100644 index 00000000000..5d058d10fc2 --- /dev/null +++ b/src/Interpreters/HashJoin/FullHashJoin.cpp @@ -0,0 +1,11 @@ +#include + +namespace DB +{ +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +} diff --git a/src/Interpreters/HashJoin/HashJoin.cpp b/src/Interpreters/HashJoin/HashJoin.cpp new file mode 100644 index 00000000000..0c7cad4360d --- /dev/null +++ b/src/Interpreters/HashJoin/HashJoin.cpp @@ -0,0 +1,1333 @@ +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + + +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + + +#include +#include + +namespace CurrentMetrics +{ + extern const Metric TemporaryFilesForJoin; +} + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int NOT_IMPLEMENTED; + extern const int NO_SUCH_COLUMN_IN_TABLE; + extern const int INCOMPATIBLE_TYPE_OF_JOIN; + extern const int UNSUPPORTED_JOIN_KEYS; + extern const int LOGICAL_ERROR; + extern const int SYNTAX_ERROR; + extern const int SET_SIZE_LIMIT_EXCEEDED; + extern const int TYPE_MISMATCH; + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; + extern const int INVALID_JOIN_ON_EXPRESSION; +} + +namespace +{ + +struct NotProcessedCrossJoin : public ExtraBlock +{ + size_t left_position; + size_t right_block; + std::unique_ptr reader; +}; + + +Int64 getCurrentQueryMemoryUsage() +{ + /// Use query-level memory tracker + if (auto * memory_tracker_child = CurrentThread::getMemoryTracker()) + if (auto * memory_tracker = memory_tracker_child->getParent()) + return memory_tracker->get(); + return 0; +} + +} + +static void correctNullabilityInplace(ColumnWithTypeAndName & column, bool nullable) +{ + if (nullable) + { + JoinCommon::convertColumnToNullable(column); + } + else + { + /// We have to replace values masked by NULLs with defaults. + if (column.column) + if (const auto * nullable_column = checkAndGetColumn(&*column.column)) + column.column = JoinCommon::filterWithBlanks(column.column, nullable_column->getNullMapColumn().getData(), true); + + JoinCommon::removeColumnNullability(column); + } +} + +HashJoin::HashJoin(std::shared_ptr table_join_, const Block & right_sample_block_, + bool any_take_last_row_, size_t reserve_num_, const String & instance_id_) + : table_join(table_join_) + , kind(table_join->kind()) + , strictness(table_join->strictness()) + , any_take_last_row(any_take_last_row_) + , reserve_num(reserve_num_) + , instance_id(instance_id_) + , asof_inequality(table_join->getAsofInequality()) + , data(std::make_shared()) + , tmp_data( + table_join_->getTempDataOnDisk() + ? std::make_unique(table_join_->getTempDataOnDisk(), CurrentMetrics::TemporaryFilesForJoin) + : nullptr) + , right_sample_block(right_sample_block_) + , max_joined_block_rows(table_join->maxJoinedBlockRows()) + , instance_log_id(!instance_id_.empty() ? "(" + instance_id_ + ") " : "") + , log(getLogger("HashJoin")) +{ + LOG_TRACE(log, "{}Keys: {}, datatype: {}, kind: {}, strictness: {}, right header: {}", + instance_log_id, TableJoin::formatClauses(table_join->getClauses(), true), data->type, kind, strictness, right_sample_block.dumpStructure()); + + validateAdditionalFilterExpression(table_join->getMixedJoinExpression()); + + used_flags = std::make_unique(); + + if (isCrossOrComma(kind)) + { + data->type = Type::CROSS; + sample_block_with_columns_to_add = materializeBlock(right_sample_block); + } + else if (table_join->getClauses().empty()) + { + data->type = Type::EMPTY; + /// We might need to insert default values into the right columns, materialize them + sample_block_with_columns_to_add = materializeBlock(right_sample_block); + } + else if (table_join->oneDisjunct()) + { + const auto & key_names_right = table_join->getOnlyClause().key_names_right; + JoinCommon::splitAdditionalColumns(key_names_right, right_sample_block, right_table_keys, sample_block_with_columns_to_add); + required_right_keys = table_join->getRequiredRightKeys(right_table_keys, required_right_keys_sources); + } + else + { + /// required right keys concept does not work well if multiple disjuncts, we need all keys + sample_block_with_columns_to_add = right_table_keys = materializeBlock(right_sample_block); + } + + materializeBlockInplace(right_table_keys); + initRightBlockStructure(data->sample_block); + data->sample_block = prepareRightBlock(data->sample_block); + + JoinCommon::createMissedColumns(sample_block_with_columns_to_add); + + size_t disjuncts_num = table_join->getClauses().size(); + data->maps.resize(disjuncts_num); + key_sizes.reserve(disjuncts_num); + + for (const auto & clause : table_join->getClauses()) + { + const auto & key_names_right = clause.key_names_right; + ColumnRawPtrs key_columns = JoinCommon::extractKeysForJoin(right_table_keys, key_names_right); + + if (strictness == JoinStrictness::Asof) + { + assert(disjuncts_num == 1); + + /// @note ASOF JOIN is not INNER. It's better avoid use of 'INNER ASOF' combination in messages. + /// In fact INNER means 'LEFT SEMI ASOF' while LEFT means 'LEFT OUTER ASOF'. + if (!isLeft(kind) && !isInner(kind)) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Wrong ASOF JOIN type. Only ASOF and LEFT ASOF joins are supported"); + + if (key_columns.size() <= 1) + throw Exception(ErrorCodes::SYNTAX_ERROR, "ASOF join needs at least one equi-join column"); + + size_t asof_size; + asof_type = SortedLookupVectorBase::getTypeSize(*key_columns.back(), asof_size); + key_columns.pop_back(); + + /// this is going to set up the appropriate hash table for the direct lookup part of the join + /// However, this does not depend on the size of the asof join key (as that goes into the BST) + /// Therefore, add it back in such that it can be extracted appropriately from the full stored + /// key_columns and key_sizes + auto & asof_key_sizes = key_sizes.emplace_back(); + data->type = chooseMethod(kind, key_columns, asof_key_sizes); + asof_key_sizes.push_back(asof_size); + } + else + { + /// Choose data structure to use for JOIN. + auto current_join_method = chooseMethod(kind, key_columns, key_sizes.emplace_back()); + if (data->type == Type::EMPTY) + data->type = current_join_method; + else if (data->type != current_join_method) + data->type = Type::hashed; + } + } + + for (auto & maps : data->maps) + dataMapInit(maps); +} + +HashJoin::Type HashJoin::chooseMethod(JoinKind kind, const ColumnRawPtrs & key_columns, Sizes & key_sizes) +{ + size_t keys_size = key_columns.size(); + + if (keys_size == 0) + { + if (isCrossOrComma(kind)) + return Type::CROSS; + return Type::EMPTY; + } + + bool all_fixed = true; + size_t keys_bytes = 0; + key_sizes.resize(keys_size); + for (size_t j = 0; j < keys_size; ++j) + { + if (!key_columns[j]->isFixedAndContiguous()) + { + all_fixed = false; + break; + } + key_sizes[j] = key_columns[j]->sizeOfValueIfFixed(); + keys_bytes += key_sizes[j]; + } + + /// If there is one numeric key that fits in 64 bits + if (keys_size == 1 && key_columns[0]->isNumeric()) + { + size_t size_of_field = key_columns[0]->sizeOfValueIfFixed(); + if (size_of_field == 1) + return Type::key8; + if (size_of_field == 2) + return Type::key16; + if (size_of_field == 4) + return Type::key32; + if (size_of_field == 8) + return Type::key64; + if (size_of_field == 16) + return Type::keys128; + if (size_of_field == 32) + return Type::keys256; + throw Exception(ErrorCodes::LOGICAL_ERROR, "Numeric column has sizeOfField not in 1, 2, 4, 8, 16, 32."); + } + + /// If the keys fit in N bits, we will use a hash table for N-bit-packed keys + if (all_fixed && keys_bytes <= 16) + return Type::keys128; + if (all_fixed && keys_bytes <= 32) + return Type::keys256; + + /// If there is single string key, use hash table of it's values. + if (keys_size == 1) + { + auto is_string_column = [](const IColumn * column_ptr) -> bool + { + if (const auto * lc_column_ptr = typeid_cast(column_ptr)) + return typeid_cast(lc_column_ptr->getDictionary().getNestedColumn().get()); + return typeid_cast(column_ptr); + }; + + const auto * key_column = key_columns[0]; + if (is_string_column(key_column) || + (isColumnConst(*key_column) && is_string_column(assert_cast(key_column)->getDataColumnPtr().get()))) + return Type::key_string; + } + + if (keys_size == 1 && typeid_cast(key_columns[0])) + return Type::key_fixed_string; + + /// Otherwise, will use set of cryptographic hashes of unambiguously serialized values. + return Type::hashed; +} + +template +static KeyGetter createKeyGetter(const ColumnRawPtrs & key_columns, const Sizes & key_sizes) +{ + if constexpr (is_asof_join) + { + auto key_column_copy = key_columns; + auto key_size_copy = key_sizes; + key_column_copy.pop_back(); + key_size_copy.pop_back(); + return KeyGetter(key_column_copy, key_size_copy, nullptr); + } + else + return KeyGetter(key_columns, key_sizes, nullptr); +} + +void HashJoin::dataMapInit(MapsVariant & map) +{ + if (kind == JoinKind::Cross) + return; + joinDispatchInit(kind, strictness, map); + joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { map_.create(data->type); }); + + if (reserve_num) + { + joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { map_.reserve(data->type, reserve_num); }); + } + + if (!data) + throw Exception(ErrorCodes::LOGICAL_ERROR, "HashJoin::dataMapInit called with empty data"); +} + +bool HashJoin::empty() const +{ + return data->type == Type::EMPTY; +} + +bool HashJoin::alwaysReturnsEmptySet() const +{ + return isInnerOrRight(getKind()) && data->empty; +} + +size_t HashJoin::getTotalRowCount() const +{ + if (!data) + return 0; + + size_t res = 0; + + if (data->type == Type::CROSS) + { + for (const auto & block : data->blocks) + res += block.rows(); + } + else + { + for (const auto & map : data->maps) + { + joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { res += map_.getTotalRowCount(data->type); }); + } + } + + return res; +} + +size_t HashJoin::getTotalByteCount() const +{ + if (!data) + return 0; + +#ifndef NDEBUG + size_t debug_blocks_allocated_size = 0; + for (const auto & block : data->blocks) + debug_blocks_allocated_size += block.allocatedBytes(); + + if (data->blocks_allocated_size != debug_blocks_allocated_size) + throw Exception(ErrorCodes::LOGICAL_ERROR, "data->blocks_allocated_size != debug_blocks_allocated_size ({} != {})", + data->blocks_allocated_size, debug_blocks_allocated_size); + + size_t debug_blocks_nullmaps_allocated_size = 0; + for (const auto & nullmap : data->blocks_nullmaps) + debug_blocks_nullmaps_allocated_size += nullmap.second->allocatedBytes(); + + if (data->blocks_nullmaps_allocated_size != debug_blocks_nullmaps_allocated_size) + throw Exception(ErrorCodes::LOGICAL_ERROR, "data->blocks_nullmaps_allocated_size != debug_blocks_nullmaps_allocated_size ({} != {})", + data->blocks_nullmaps_allocated_size, debug_blocks_nullmaps_allocated_size); +#endif + + size_t res = 0; + + res += data->blocks_allocated_size; + res += data->blocks_nullmaps_allocated_size; + res += data->pool.allocatedBytes(); + + if (data->type != Type::CROSS) + { + for (const auto & map : data->maps) + { + joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { res += map_.getTotalByteCountImpl(data->type); }); + } + } + return res; +} + +void HashJoin::initRightBlockStructure(Block & saved_block_sample) +{ + if (isCrossOrComma(kind)) + { + /// cross join doesn't have keys, just add all columns + saved_block_sample = sample_block_with_columns_to_add.cloneEmpty(); + return; + } + + bool multiple_disjuncts = !table_join->oneDisjunct(); + /// We could remove key columns for LEFT | INNER HashJoin but we should keep them for JoinSwitcher (if any). + bool save_key_columns = table_join->isEnabledAlgorithm(JoinAlgorithm::AUTO) || + table_join->isEnabledAlgorithm(JoinAlgorithm::GRACE_HASH) || + isRightOrFull(kind) || + multiple_disjuncts || + table_join->getMixedJoinExpression(); + if (save_key_columns) + { + saved_block_sample = right_table_keys.cloneEmpty(); + } + else if (strictness == JoinStrictness::Asof) + { + /// Save ASOF key + saved_block_sample.insert(right_table_keys.safeGetByPosition(right_table_keys.columns() - 1)); + } + + /// Save non key columns + for (auto & column : sample_block_with_columns_to_add) + { + if (auto * col = saved_block_sample.findByName(column.name)) + *col = column; + else + saved_block_sample.insert(column); + } +} + +Block HashJoin::prepareRightBlock(const Block & block, const Block & saved_block_sample_) +{ + Block structured_block; + for (const auto & sample_column : saved_block_sample_.getColumnsWithTypeAndName()) + { + ColumnWithTypeAndName column = block.getByName(sample_column.name); + + /// There's no optimization for right side const columns. Remove constness if any. + column.column = recursiveRemoveSparse(column.column->convertToFullColumnIfConst()); + + if (column.column->lowCardinality() && !sample_column.column->lowCardinality()) + { + column.column = column.column->convertToFullColumnIfLowCardinality(); + column.type = removeLowCardinality(column.type); + } + + if (sample_column.column->isNullable()) + JoinCommon::convertColumnToNullable(column); + + structured_block.insert(std::move(column)); + } + + return structured_block; +} + +Block HashJoin::prepareRightBlock(const Block & block) const +{ + return prepareRightBlock(block, savedBlockSample()); +} + +bool HashJoin::addBlockToJoin(const Block & source_block_, bool check_limits) +{ + if (!data) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Join data was released"); + + /// RowRef::SizeT is uint32_t (not size_t) for hash table Cell memory efficiency. + /// It's possible to split bigger blocks and insert them by parts here. But it would be a dead code. + if (unlikely(source_block_.rows() > std::numeric_limits::max())) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Too many rows in right table block for HashJoin: {}", source_block_.rows()); + + /** We do not allocate memory for stored blocks inside HashJoin, only for hash table. + * In case when we have all the blocks allocated before the first `addBlockToJoin` call, will already be quite high. + * In that case memory consumed by stored blocks will be underestimated. + */ + if (!memory_usage_before_adding_blocks) + memory_usage_before_adding_blocks = getCurrentQueryMemoryUsage(); + + Block source_block = source_block_; + if (strictness == JoinStrictness::Asof) + { + chassert(kind == JoinKind::Left || kind == JoinKind::Inner); + + /// Filter out rows with NULLs in ASOF key, nulls are not joined with anything since they are not comparable + /// We support only INNER/LEFT ASOF join, so rows with NULLs never return from the right joined table. + /// So filter them out here not to handle in implementation. + const auto & asof_key_name = table_join->getOnlyClause().key_names_right.back(); + auto & asof_column = source_block.getByName(asof_key_name); + + if (asof_column.type->isNullable()) + { + /// filter rows with nulls in asof key + if (const auto * asof_const_column = typeid_cast(asof_column.column.get())) + { + if (asof_const_column->isNullAt(0)) + return false; + } + else + { + const auto & asof_column_nullable = assert_cast(*asof_column.column).getNullMapData(); + + NullMap negative_null_map(asof_column_nullable.size()); + for (size_t i = 0; i < asof_column_nullable.size(); ++i) + negative_null_map[i] = !asof_column_nullable[i]; + + for (auto & column : source_block) + column.column = column.column->filter(negative_null_map, -1); + } + } + } + + size_t rows = source_block.rows(); + + const auto & right_key_names = table_join->getAllNames(JoinTableSide::Right); + ColumnPtrMap all_key_columns(right_key_names.size()); + for (const auto & column_name : right_key_names) + { + const auto & column = source_block.getByName(column_name).column; + all_key_columns[column_name] = recursiveRemoveSparse(column->convertToFullColumnIfConst())->convertToFullColumnIfLowCardinality(); + } + + Block block_to_save = prepareRightBlock(source_block); + if (shrink_blocks) + block_to_save = block_to_save.shrinkToFit(); + + size_t max_bytes_in_join = table_join->sizeLimits().max_bytes; + size_t max_rows_in_join = table_join->sizeLimits().max_rows; + + if (kind == JoinKind::Cross && tmp_data + && (tmp_stream || (max_bytes_in_join && getTotalByteCount() + block_to_save.allocatedBytes() >= max_bytes_in_join) + || (max_rows_in_join && getTotalRowCount() + block_to_save.rows() >= max_rows_in_join))) + { + if (tmp_stream == nullptr) + { + tmp_stream = &tmp_data->createStream(right_sample_block); + } + tmp_stream->write(block_to_save); + return true; + } + + size_t total_rows = 0; + size_t total_bytes = 0; + { + if (storage_join_lock) + throw DB::Exception(ErrorCodes::LOGICAL_ERROR, "addBlockToJoin called when HashJoin locked to prevent updates"); + + assertBlocksHaveEqualStructure(data->sample_block, block_to_save, "joined block"); + + size_t min_bytes_to_compress = table_join->crossJoinMinBytesToCompress(); + size_t min_rows_to_compress = table_join->crossJoinMinRowsToCompress(); + + if (kind == JoinKind::Cross + && ((min_bytes_to_compress && getTotalByteCount() >= min_bytes_to_compress) + || (min_rows_to_compress && getTotalRowCount() >= min_rows_to_compress))) + { + block_to_save = block_to_save.compress(); + have_compressed = true; + } + + data->blocks_allocated_size += block_to_save.allocatedBytes(); + data->blocks.emplace_back(std::move(block_to_save)); + Block * stored_block = &data->blocks.back(); + + if (rows) + data->empty = false; + + bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); + const auto & onexprs = table_join->getClauses(); + for (size_t onexpr_idx = 0; onexpr_idx < onexprs.size(); ++onexpr_idx) + { + ColumnRawPtrs key_columns; + for (const auto & name : onexprs[onexpr_idx].key_names_right) + key_columns.push_back(all_key_columns[name].get()); + + /// We will insert to the map only keys, where all components are not NULL. + ConstNullMapPtr null_map{}; + ColumnPtr null_map_holder = extractNestedColumnsAndNullMap(key_columns, null_map); + + /// If RIGHT or FULL save blocks with nulls for NotJoinedBlocks + UInt8 save_nullmap = 0; + if (isRightOrFull(kind) && null_map) + { + /// Save rows with NULL keys + for (size_t i = 0; !save_nullmap && i < null_map->size(); ++i) + save_nullmap |= (*null_map)[i]; + } + + auto join_mask_col = JoinCommon::getColumnAsMask(source_block, onexprs[onexpr_idx].condColumnNames().second); + /// Save blocks that do not hold conditions in ON section + ColumnUInt8::MutablePtr not_joined_map = nullptr; + if (!flag_per_row && isRightOrFull(kind) && join_mask_col.hasData()) + { + const auto & join_mask = join_mask_col.getData(); + /// Save rows that do not hold conditions + not_joined_map = ColumnUInt8::create(rows, 0); + for (size_t i = 0, sz = join_mask->size(); i < sz; ++i) + { + /// Condition hold, do not save row + if ((*join_mask)[i]) + continue; + + /// NULL key will be saved anyway because, do not save twice + if (save_nullmap && (*null_map)[i]) + continue; + + not_joined_map->getData()[i] = 1; + } + } + + bool is_inserted = false; + if (kind != JoinKind::Cross) + { + joinDispatch(kind, strictness, data->maps[onexpr_idx], [&](auto kind_, auto strictness_, auto & map) + { + size_t size = HashJoinMethods>::insertFromBlockImpl( + *this, + data->type, + map, + rows, + key_columns, + key_sizes[onexpr_idx], + stored_block, + null_map, + join_mask_col.getData(), + data->pool, + is_inserted); + + if (flag_per_row) + used_flags->reinit(stored_block); + else if (is_inserted) + /// Number of buckets + 1 value from zero storage + used_flags->reinit(size + 1); + }); + } + + if (!flag_per_row && save_nullmap && is_inserted) + { + data->blocks_nullmaps_allocated_size += null_map_holder->allocatedBytes(); + data->blocks_nullmaps.emplace_back(stored_block, null_map_holder); + } + + if (!flag_per_row && not_joined_map && is_inserted) + { + data->blocks_nullmaps_allocated_size += not_joined_map->allocatedBytes(); + data->blocks_nullmaps.emplace_back(stored_block, std::move(not_joined_map)); + } + + if (!flag_per_row && !is_inserted) + { + LOG_TRACE(log, "Skipping inserting block with {} rows", rows); + data->blocks_allocated_size -= stored_block->allocatedBytes(); + data->blocks.pop_back(); + } + + if (!check_limits) + return true; + + /// TODO: Do not calculate them every time + total_rows = getTotalRowCount(); + total_bytes = getTotalByteCount(); + } + } + + shrinkStoredBlocksToFit(total_bytes); + + return table_join->sizeLimits().check(total_rows, total_bytes, "JOIN", ErrorCodes::SET_SIZE_LIMIT_EXCEEDED); +} + +void HashJoin::shrinkStoredBlocksToFit(size_t & total_bytes_in_join) +{ + if (shrink_blocks) + return; /// Already shrunk + + Int64 current_memory_usage = getCurrentQueryMemoryUsage(); + Int64 query_memory_usage_delta = current_memory_usage - memory_usage_before_adding_blocks; + Int64 max_total_bytes_for_query = memory_usage_before_adding_blocks ? table_join->getMaxMemoryUsage() : 0; + + auto max_total_bytes_in_join = table_join->sizeLimits().max_bytes; + + /** If accounted data size is more than half of `max_bytes_in_join` + * or query memory consumption growth from the beginning of adding blocks (estimation of memory consumed by join using memory tracker) + * is bigger than half of all memory available for query, + * then shrink stored blocks to fit. + */ + shrink_blocks = (max_total_bytes_in_join && total_bytes_in_join > max_total_bytes_in_join / 2) || + (max_total_bytes_for_query && query_memory_usage_delta > max_total_bytes_for_query / 2); + if (!shrink_blocks) + return; + + LOG_DEBUG(log, "Shrinking stored blocks, memory consumption is {} {} calculated by join, {} {} by memory tracker", + ReadableSize(total_bytes_in_join), max_total_bytes_in_join ? fmt::format("/ {}", ReadableSize(max_total_bytes_in_join)) : "", + ReadableSize(query_memory_usage_delta), max_total_bytes_for_query ? fmt::format("/ {}", ReadableSize(max_total_bytes_for_query)) : ""); + + for (auto & stored_block : data->blocks) + { + size_t old_size = stored_block.allocatedBytes(); + stored_block = stored_block.shrinkToFit(); + size_t new_size = stored_block.allocatedBytes(); + + if (old_size >= new_size) + { + if (data->blocks_allocated_size < old_size - new_size) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Blocks allocated size value is broken: " + "blocks_allocated_size = {}, old_size = {}, new_size = {}", + data->blocks_allocated_size, old_size, new_size); + + data->blocks_allocated_size -= old_size - new_size; + } + else + /// Sometimes after clone resized block can be bigger than original + data->blocks_allocated_size += new_size - old_size; + } + + auto new_total_bytes_in_join = getTotalByteCount(); + + Int64 new_current_memory_usage = getCurrentQueryMemoryUsage(); + + LOG_DEBUG(log, "Shrunk stored blocks {} freed ({} by memory tracker), new memory consumption is {} ({} by memory tracker)", + ReadableSize(total_bytes_in_join - new_total_bytes_in_join), ReadableSize(current_memory_usage - new_current_memory_usage), + ReadableSize(new_total_bytes_in_join), ReadableSize(new_current_memory_usage)); + + total_bytes_in_join = new_total_bytes_in_join; +} + +void HashJoin::joinBlockImplCross(Block & block, ExtraBlockPtr & not_processed) const +{ + size_t start_left_row = 0; + size_t start_right_block = 0; + std::unique_ptr reader = nullptr; + if (not_processed) + { + auto & continuation = static_cast(*not_processed); + start_left_row = continuation.left_position; + start_right_block = continuation.right_block; + reader = std::move(continuation.reader); + not_processed.reset(); + } + + size_t num_existing_columns = block.columns(); + size_t num_columns_to_add = sample_block_with_columns_to_add.columns(); + + ColumnRawPtrs src_left_columns; + MutableColumns dst_columns; + + { + src_left_columns.reserve(num_existing_columns); + dst_columns.reserve(num_existing_columns + num_columns_to_add); + + for (const ColumnWithTypeAndName & left_column : block) + { + src_left_columns.push_back(left_column.column.get()); + dst_columns.emplace_back(src_left_columns.back()->cloneEmpty()); + } + + for (const ColumnWithTypeAndName & right_column : sample_block_with_columns_to_add) + dst_columns.emplace_back(right_column.column->cloneEmpty()); + + for (auto & dst : dst_columns) + dst->reserve(max_joined_block_rows); + } + + size_t rows_left = block.rows(); + size_t rows_added = 0; + for (size_t left_row = start_left_row; left_row < rows_left; ++left_row) + { + size_t block_number = 0; + + auto process_right_block = [&](const Block & block_right) + { + size_t rows_right = block_right.rows(); + rows_added += rows_right; + + for (size_t col_num = 0; col_num < num_existing_columns; ++col_num) + dst_columns[col_num]->insertManyFrom(*src_left_columns[col_num], left_row, rows_right); + + for (size_t col_num = 0; col_num < num_columns_to_add; ++col_num) + { + const IColumn & column_right = *block_right.getByPosition(col_num).column; + dst_columns[num_existing_columns + col_num]->insertRangeFrom(column_right, 0, rows_right); + } + }; + + for (const Block & block_right : data->blocks) + { + ++block_number; + if (block_number < start_right_block) + continue; + /// The following statement cannot be substituted with `process_right_block(!have_compressed ? block_right : block_right.decompress())` + /// because it will lead to copying of `block_right` even if its branch is taken (because common type of `block_right` and `block_right.decompress()` is `Block`). + if (!have_compressed) + process_right_block(block_right); + else + process_right_block(block_right.decompress()); + + if (rows_added > max_joined_block_rows) + { + break; + } + } + + if (tmp_stream && rows_added <= max_joined_block_rows) + { + if (reader == nullptr) + { + tmp_stream->finishWritingAsyncSafe(); + reader = tmp_stream->getReadStream(); + } + while (auto block_right = reader->read()) + { + ++block_number; + process_right_block(block_right); + if (rows_added > max_joined_block_rows) + { + break; + } + } + + /// It means, that reader->read() returned {} + if (rows_added <= max_joined_block_rows) + { + reader.reset(); + } + } + + start_right_block = 0; + + if (rows_added > max_joined_block_rows) + { + not_processed = std::make_shared( + NotProcessedCrossJoin{{block.cloneEmpty()}, left_row, block_number + 1, std::move(reader)}); + not_processed->block.swap(block); + break; + } + } + + for (const ColumnWithTypeAndName & src_column : sample_block_with_columns_to_add) + block.insert(src_column); + + block = block.cloneWithColumns(std::move(dst_columns)); +} + +DataTypePtr HashJoin::joinGetCheckAndGetReturnType(const DataTypes & data_types, const String & column_name, bool or_null) const +{ + size_t num_keys = data_types.size(); + if (right_table_keys.columns() != num_keys) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + "Number of arguments for function joinGet{} doesn't match: passed, should be equal to {}", + toString(or_null ? "OrNull" : ""), toString(num_keys)); + + for (size_t i = 0; i < num_keys; ++i) + { + const auto & left_type_origin = data_types[i]; + const auto & [c2, right_type_origin, right_name] = right_table_keys.safeGetByPosition(i); + auto left_type = removeNullable(recursiveRemoveLowCardinality(left_type_origin)); + auto right_type = removeNullable(recursiveRemoveLowCardinality(right_type_origin)); + if (!left_type->equals(*right_type)) + throw Exception(ErrorCodes::TYPE_MISMATCH, "Type mismatch in joinGet key {}: " + "found type {}, while the needed type is {}", i, left_type->getName(), right_type->getName()); + } + + if (!sample_block_with_columns_to_add.has(column_name)) + throw Exception(ErrorCodes::NO_SUCH_COLUMN_IN_TABLE, "StorageJoin doesn't contain column {}", column_name); + + auto elem = sample_block_with_columns_to_add.getByName(column_name); + if (or_null && JoinCommon::canBecomeNullable(elem.type)) + elem.type = makeNullable(elem.type); + return elem.type; +} + +/// TODO: return multiple columns as named tuple +/// TODO: return array of values when strictness == JoinStrictness::All +ColumnWithTypeAndName HashJoin::joinGet(const Block & block, const Block & block_with_columns_to_add) const +{ + bool is_valid = (strictness == JoinStrictness::Any || strictness == JoinStrictness::RightAny) + && kind == JoinKind::Left; + if (!is_valid) + throw Exception(ErrorCodes::INCOMPATIBLE_TYPE_OF_JOIN, "joinGet only supports StorageJoin of type Left Any"); + const auto & key_names_right = table_join->getOnlyClause().key_names_right; + + /// Assemble the key block with correct names. + Block keys; + for (size_t i = 0; i < block.columns(); ++i) + { + auto key = block.getByPosition(i); + key.name = key_names_right[i]; + keys.insert(std::move(key)); + } + + static_assert(!MapGetter::flagged, + "joinGet are not protected from hash table changes between block processing"); + + std::vector maps_vector; + maps_vector.push_back(&std::get(data->maps[0])); + HashJoinMethods::joinBlockImpl(*this, keys, block_with_columns_to_add, maps_vector, /* is_join_get = */ true); + return keys.getByPosition(keys.columns() - 1); +} + +void HashJoin::checkTypesOfKeys(const Block & block) const +{ + for (const auto & onexpr : table_join->getClauses()) + { + JoinCommon::checkTypesOfKeys(block, onexpr.key_names_left, right_table_keys, onexpr.key_names_right); + } +} + +void HashJoin::joinBlock(Block & block, ExtraBlockPtr & not_processed) +{ + if (!data) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot join after data has been released"); + + for (const auto & onexpr : table_join->getClauses()) + { + auto cond_column_name = onexpr.condColumnNames(); + JoinCommon::checkTypesOfKeys( + block, onexpr.key_names_left, cond_column_name.first, + right_sample_block, onexpr.key_names_right, cond_column_name.second); + } + + if (kind == JoinKind::Cross) + { + joinBlockImplCross(block, not_processed); + return; + } + + if (kind == JoinKind::Right || kind == JoinKind::Full) + { + materializeBlockInplace(block); + } + + { + std::vectormaps[0])> * > maps_vector; + for (size_t i = 0; i < table_join->getClauses().size(); ++i) + maps_vector.push_back(&data->maps[i]); + + if (joinDispatch(kind, strictness, maps_vector, [&](auto kind_, auto strictness_, auto & maps_vector_) + { + using MapType = typename MapGetter::Map; + Block remaining_block = HashJoinMethods::joinBlockImpl( + *this, block, sample_block_with_columns_to_add, maps_vector_); + if (remaining_block.rows()) + not_processed = std::make_shared(ExtraBlock{std::move(remaining_block)}); + else + not_processed.reset(); + })) + { + /// Joined + } + else + throw Exception(ErrorCodes::LOGICAL_ERROR, "Wrong JOIN combination: {} {}", strictness, kind); + } +} + +HashJoin::~HashJoin() +{ + if (!data) + { + LOG_TEST(log, "{}Join data has been already released", instance_log_id); + return; + } + LOG_TEST( + log, + "{}Join data is being destroyed, {} bytes and {} rows in hash table", + instance_log_id, + getTotalByteCount(), + getTotalRowCount()); +} + +template +struct AdderNonJoined +{ + static void add(const Mapped & mapped, size_t & rows_added, MutableColumns & columns_right) + { + constexpr bool mapped_asof = std::is_same_v; + [[maybe_unused]] constexpr bool mapped_one = std::is_same_v; + + if constexpr (mapped_asof) + { + /// Do nothing + } + else if constexpr (mapped_one) + { + for (size_t j = 0; j < columns_right.size(); ++j) + { + const auto & mapped_column = mapped.block->getByPosition(j).column; + columns_right[j]->insertFrom(*mapped_column, mapped.row_num); + } + + ++rows_added; + } + else + { + for (auto it = mapped.begin(); it.ok(); ++it) + { + for (size_t j = 0; j < columns_right.size(); ++j) + { + const auto & mapped_column = it->block->getByPosition(j).column; + columns_right[j]->insertFrom(*mapped_column, it->row_num); + } + + ++rows_added; + } + } + } +}; + +/// Stream from not joined earlier rows of the right table. +/// Based on: +/// - map offsetInternal saved in used_flags for single disjuncts +/// - flags in BlockWithFlags for multiple disjuncts +class NotJoinedHash final : public NotJoinedBlocks::RightColumnsFiller +{ +public: + NotJoinedHash(const HashJoin & parent_, UInt64 max_block_size_, bool flag_per_row_) + : parent(parent_) + , max_block_size(max_block_size_) + , flag_per_row(flag_per_row_) + , current_block_start(0) + { + if (parent.data == nullptr) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot join after data has been released"); + } + + Block getEmptyBlock() override { return parent.savedBlockSample().cloneEmpty(); } + + size_t fillColumns(MutableColumns & columns_right) override + { + size_t rows_added = 0; + if (unlikely(parent.data->type == HashJoin::Type::EMPTY)) + { + rows_added = fillColumnsFromData(parent.data->blocks, columns_right); + } + else + { + auto fill_callback = [&](auto, auto, auto & map) + { + rows_added = fillColumnsFromMap(map, columns_right); + }; + + if (!joinDispatch(parent.kind, parent.strictness, parent.data->maps.front(), fill_callback)) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown JOIN strictness '{}' (must be on of: ANY, ALL, ASOF)", parent.strictness); + } + + if (!flag_per_row) + { + fillNullsFromBlocks(columns_right, rows_added); + } + + return rows_added; + } + +private: + const HashJoin & parent; + UInt64 max_block_size; + bool flag_per_row; + + size_t current_block_start; + + std::any position; + std::optional nulls_position; + std::optional used_position; + + size_t fillColumnsFromData(const BlocksList & blocks, MutableColumns & columns_right) + { + if (!position.has_value()) + position = std::make_any(blocks.begin()); + + auto & block_it = std::any_cast(position); + auto end = blocks.end(); + + size_t rows_added = 0; + for (; block_it != end; ++block_it) + { + size_t rows_from_block = std::min(max_block_size - rows_added, block_it->rows() - current_block_start); + for (size_t j = 0; j < columns_right.size(); ++j) + { + const auto & col = block_it->getByPosition(j).column; + columns_right[j]->insertRangeFrom(*col, current_block_start, rows_from_block); + } + rows_added += rows_from_block; + + if (rows_added >= max_block_size) + { + /// How many rows have been read + current_block_start += rows_from_block; + if (block_it->rows() <= current_block_start) + { + /// current block was fully read + ++block_it; + current_block_start = 0; + } + break; + } + current_block_start = 0; + } + return rows_added; + } + + template + size_t fillColumnsFromMap(const Maps & maps, MutableColumns & columns_keys_and_right) + { + switch (parent.data->type) + { + #define M(TYPE) \ + case HashJoin::Type::TYPE: \ + return fillColumns(*maps.TYPE, columns_keys_and_right); + APPLY_FOR_JOIN_VARIANTS(M) + #undef M + default: + throw Exception(ErrorCodes::UNSUPPORTED_JOIN_KEYS, "Unsupported JOIN keys (type: {})", parent.data->type); + } + } + + template + size_t fillColumns(const Map & map, MutableColumns & columns_keys_and_right) + { + size_t rows_added = 0; + + if (flag_per_row) + { + if (!used_position.has_value()) + used_position = parent.data->blocks.begin(); + + auto end = parent.data->blocks.end(); + + for (auto & it = *used_position; it != end && rows_added < max_block_size; ++it) + { + const Block & mapped_block = *it; + + for (size_t row = 0; row < mapped_block.rows(); ++row) + { + if (!parent.isUsed(&mapped_block, row)) + { + for (size_t colnum = 0; colnum < columns_keys_and_right.size(); ++colnum) + { + columns_keys_and_right[colnum]->insertFrom(*mapped_block.getByPosition(colnum).column, row); + } + + ++rows_added; + } + } + } + } + else + { + using Mapped = typename Map::mapped_type; + using Iterator = typename Map::const_iterator; + + + if (!position.has_value()) + position = std::make_any(map.begin()); + + Iterator & it = std::any_cast(position); + auto end = map.end(); + + for (; it != end; ++it) + { + const Mapped & mapped = it->getMapped(); + + size_t offset = map.offsetInternal(it.getPtr()); + if (parent.isUsed(offset)) + continue; + AdderNonJoined::add(mapped, rows_added, columns_keys_and_right); + + if (rows_added >= max_block_size) + { + ++it; + break; + } + } + } + + return rows_added; + } + + void fillNullsFromBlocks(MutableColumns & columns_keys_and_right, size_t & rows_added) + { + if (!nulls_position.has_value()) + nulls_position = parent.data->blocks_nullmaps.begin(); + + auto end = parent.data->blocks_nullmaps.end(); + + for (auto & it = *nulls_position; it != end && rows_added < max_block_size; ++it) + { + const auto * block = it->first; + ConstNullMapPtr nullmap = nullptr; + if (it->second) + nullmap = &assert_cast(*it->second).getData(); + + for (size_t row = 0; row < block->rows(); ++row) + { + if (nullmap && (*nullmap)[row]) + { + for (size_t col = 0; col < columns_keys_and_right.size(); ++col) + columns_keys_and_right[col]->insertFrom(*block->getByPosition(col).column, row); + ++rows_added; + } + } + } + } +}; + +IBlocksStreamPtr HashJoin::getNonJoinedBlocks(const Block & left_sample_block, + const Block & result_sample_block, + UInt64 max_block_size) const +{ + if (!JoinCommon::hasNonJoinedBlocks(*table_join)) + return {}; + size_t left_columns_count = left_sample_block.columns(); + + bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); + if (!flag_per_row) + { + /// With multiple disjuncts, all keys are in sample_block_with_columns_to_add, so invariant is not held + size_t expected_columns_count = left_columns_count + required_right_keys.columns() + sample_block_with_columns_to_add.columns(); + if (expected_columns_count != result_sample_block.columns()) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected number of columns in result sample block: {} instead of {} ({} + {} + {})", + result_sample_block.columns(), expected_columns_count, + left_columns_count, required_right_keys.columns(), sample_block_with_columns_to_add.columns()); + } + } + + auto non_joined = std::make_unique(*this, max_block_size, flag_per_row); + return std::make_unique(std::move(non_joined), result_sample_block, left_columns_count, *table_join); +} + +void HashJoin::reuseJoinedData(const HashJoin & join) +{ + data = join.data; + from_storage_join = true; + + bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); + if (flag_per_row) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "StorageJoin with ORs is not supported"); + + for (auto & map : data->maps) + { + joinDispatch(kind, strictness, map, [this](auto kind_, auto strictness_, auto & map_) + { + used_flags->reinit(map_.getBufferSizeInCells(data->type) + 1); + }); + } +} + +BlocksList HashJoin::releaseJoinedBlocks(bool restructure) +{ + LOG_TRACE(log, "{}Join data is being released, {} bytes and {} rows in hash table", instance_log_id, getTotalByteCount(), getTotalRowCount()); + + BlocksList right_blocks = std::move(data->blocks); + if (!restructure) + { + data.reset(); + return right_blocks; + } + + data->maps.clear(); + data->blocks_nullmaps.clear(); + + BlocksList restored_blocks; + + /// names to positions optimization + std::vector positions; + std::vector is_nullable; + if (!right_blocks.empty()) + { + positions.reserve(right_sample_block.columns()); + const Block & tmp_block = *right_blocks.begin(); + for (const auto & sample_column : right_sample_block) + { + positions.emplace_back(tmp_block.getPositionByName(sample_column.name)); + is_nullable.emplace_back(isNullableOrLowCardinalityNullable(sample_column.type)); + } + } + + for (Block & saved_block : right_blocks) + { + Block restored_block; + for (size_t i = 0; i < positions.size(); ++i) + { + auto & column = saved_block.getByPosition(positions[i]); + correctNullabilityInplace(column, is_nullable[i]); + restored_block.insert(column); + } + restored_blocks.emplace_back(std::move(restored_block)); + } + + data.reset(); + return restored_blocks; +} + +const ColumnWithTypeAndName & HashJoin::rightAsofKeyColumn() const +{ + /// It should be nullable when right side is nullable + return savedBlockSample().getByName(table_join->getOnlyClause().key_names_right.back()); +} + +void HashJoin::validateAdditionalFilterExpression(ExpressionActionsPtr additional_filter_expression) +{ + if (!additional_filter_expression) + return; + + Block expression_sample_block = additional_filter_expression->getSampleBlock(); + + if (expression_sample_block.columns() != 1) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Unexpected expression in JOIN ON section. Expected single column, got '{}'", + expression_sample_block.dumpStructure()); + } + + auto type = removeNullable(expression_sample_block.getByPosition(0).type); + if (!type->equals(*std::make_shared())) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Unexpected expression in JOIN ON section. Expected boolean (UInt8), got '{}'. expression:\n{}", + expression_sample_block.getByPosition(0).type->getName(), + additional_filter_expression->dumpActions()); + } + + bool is_supported = (strictness == JoinStrictness::All) && (isInnerOrLeft(kind) || isRightOrFull(kind)); + if (!is_supported) + { + throw Exception(ErrorCodes::INVALID_JOIN_ON_EXPRESSION, + "Non equi condition '{}' from JOIN ON section is supported only for ALL INNER/LEFT/FULL/RIGHT JOINs", + expression_sample_block.getByPosition(0).name); + } +} + +bool HashJoin::isUsed(size_t off) const +{ + return used_flags->getUsedSafe(off); +} + +bool HashJoin::isUsed(const Block * block_ptr, size_t row_idx) const +{ + return used_flags->getUsedSafe(block_ptr, row_idx); +} + + +bool HashJoin::needUsedFlagsForPerRightTableRow(std::shared_ptr table_join_) const +{ + if (!table_join_->oneDisjunct()) + return true; + /// If it'a a all right join with inequal conditions, we need to mark each row + if (table_join_->getMixedJoinExpression() && isRightOrFull(table_join_->kind())) + return true; + return false; +} + +} diff --git a/src/Interpreters/HashJoin.h b/src/Interpreters/HashJoin/HashJoin.h similarity index 91% rename from src/Interpreters/HashJoin.h rename to src/Interpreters/HashJoin/HashJoin.h index 56a1768a7ff..0b115b9fdbb 100644 --- a/src/Interpreters/HashJoin.h +++ b/src/Interpreters/HashJoin/HashJoin.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include #include @@ -36,47 +37,13 @@ class ExpressionActions; namespace JoinStuff { - /// Flags needed to implement RIGHT and FULL JOINs. -class JoinUsedFlags -{ - using RawBlockPtr = const Block *; - using UsedFlagsForBlock = std::vector; - - /// For multiple dijuncts each empty in hashmap stores flags for particular block - /// For single dicunct we store all flags in `nullptr` entry, index is the offset in FindResult - std::unordered_map flags; - - bool need_flags; - -public: - /// Update size for vector with flags. - /// Calling this method invalidates existing flags. - /// It can be called several times, but all of them should happen before using this structure. - template - void reinit(size_t size_); - - template - void reinit(const Block * block_ptr); - - bool getUsedSafe(size_t i) const; - bool getUsedSafe(const Block * block_ptr, size_t row_idx) const; - - template - void setUsed(const T & f); - - template - void setUsed(const Block * block, size_t row_num, size_t offset); - - template - bool getUsed(const T & f); - - template - bool setUsedOnce(const T & f); -}; - +class JoinUsedFlags; } +template +class HashJoinMethods; + /** Data structure for implementation of JOIN. * It is just a hash table: keys -> rows of joined ("right") table. * Additionally, CROSS JOIN is supported: instead of hash table, it use just set of blocks without keys. @@ -400,8 +367,8 @@ public: const Block & savedBlockSample() const { return data->sample_block; } - bool isUsed(size_t off) const { return used_flags.getUsedSafe(off); } - bool isUsed(const Block * block_ptr, size_t row_idx) const { return used_flags.getUsedSafe(block_ptr, row_idx); } + bool isUsed(size_t off) const; + bool isUsed(const Block * block_ptr, size_t row_idx) const; void debugKeys() const; @@ -414,6 +381,9 @@ private: friend class JoinSource; + template + friend class HashJoinMethods; + std::shared_ptr table_join; const JoinKind kind; const JoinStrictness strictness; @@ -433,8 +403,7 @@ private: /// Number of this flags equals to hashtable buffer size (plus one for zero value). /// Changes in hash table broke correspondence, /// so we must guarantee constantness of hash table during HashJoin lifetime (using method setLock) - mutable JoinStuff::JoinUsedFlags used_flags; - + mutable std::unique_ptr used_flags; RightTableDataPtr data; bool have_compressed = false; @@ -476,13 +445,6 @@ private: void initRightBlockStructure(Block & saved_block_sample); - template - Block joinBlockImpl( - Block & block, - const Block & block_with_columns_to_add, - const std::vector & maps_, - bool is_join_get = false) const; - void joinBlockImplCross(Block & block, ExtraBlockPtr & not_processed) const; static Type chooseMethod(JoinKind kind, const ColumnRawPtrs & key_columns, Sizes & key_sizes); diff --git a/src/Interpreters/HashJoin/HashJoinMethods.h b/src/Interpreters/HashJoin/HashJoinMethods.h new file mode 100644 index 00000000000..0dfafa94efc --- /dev/null +++ b/src/Interpreters/HashJoin/HashJoinMethods.h @@ -0,0 +1,954 @@ +#pragma once +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + + +namespace DB +{ +namespace ErrorCodes +{ + extern const int UNSUPPORTED_JOIN_KEYS; + extern const int LOGICAL_ERROR; +} + +/// Inserting an element into a hash table of the form `key -> reference to a string`, which will then be used by JOIN. +template +struct Inserter +{ + static ALWAYS_INLINE bool + insertOne(const HashJoin & join, HashMap & map, KeyGetter & key_getter, Block * stored_block, size_t i, Arena & pool) + { + auto emplace_result = key_getter.emplaceKey(map, i, pool); + + if (emplace_result.isInserted() || join.anyTakeLastRow()) + { + new (&emplace_result.getMapped()) typename HashMap::mapped_type(stored_block, i); + return true; + } + return false; + } + + static ALWAYS_INLINE void insertAll(const HashJoin &, HashMap & map, KeyGetter & key_getter, Block * stored_block, size_t i, Arena & pool) + { + auto emplace_result = key_getter.emplaceKey(map, i, pool); + + if (emplace_result.isInserted()) + new (&emplace_result.getMapped()) typename HashMap::mapped_type(stored_block, i); + else + { + /// The first element of the list is stored in the value of the hash table, the rest in the pool. + emplace_result.getMapped().insert({stored_block, i}, pool); + } + } + + static ALWAYS_INLINE void insertAsof( + HashJoin & join, HashMap & map, KeyGetter & key_getter, Block * stored_block, size_t i, Arena & pool, const IColumn & asof_column) + { + auto emplace_result = key_getter.emplaceKey(map, i, pool); + typename HashMap::mapped_type * time_series_map = &emplace_result.getMapped(); + + TypeIndex asof_type = *join.getAsofType(); + if (emplace_result.isInserted()) + time_series_map = new (time_series_map) typename HashMap::mapped_type(createAsofRowRef(asof_type, join.getAsofInequality())); + (*time_series_map)->insert(asof_column, stored_block, i); + } +}; + + +/// MapsTemplate is one of MapsOne, MapsAll and MapsAsof +template +class HashJoinMethods +{ +public: + static size_t insertFromBlockImpl( + HashJoin & join, + HashJoin::Type type, + MapsTemplate & maps, + size_t rows, + const ColumnRawPtrs & key_columns, + const Sizes & key_sizes, + Block * stored_block, + ConstNullMapPtr null_map, + UInt8ColumnDataPtr join_mask, + Arena & pool, + bool & is_inserted) + { + switch (type) + { + case HashJoin::Type::EMPTY: + [[fallthrough]]; + case HashJoin::Type::CROSS: + /// Do nothing. We will only save block, and it is enough + is_inserted = true; + return 0; + + #define M(TYPE) \ + case HashJoin::Type::TYPE: \ + return insertFromBlockImplTypeCase>::Type>(\ + join, *maps.TYPE, rows, key_columns, key_sizes, stored_block, null_map, join_mask, pool, is_inserted); \ + break; + + APPLY_FOR_JOIN_VARIANTS(M) + #undef M + } + } + + using MapsTemplateVector = std::vector; + + static Block joinBlockImpl( + const HashJoin & join, + Block & block, + const Block & block_with_columns_to_add, + const MapsTemplateVector & maps_, + bool is_join_get = false) + { + constexpr JoinFeatures join_features; + + std::vector join_on_keys; + const auto & onexprs = join.table_join->getClauses(); + for (size_t i = 0; i < onexprs.size(); ++i) + { + const auto & key_names = !is_join_get ? onexprs[i].key_names_left : onexprs[i].key_names_right; + join_on_keys.emplace_back(block, key_names, onexprs[i].condColumnNames().first, join.key_sizes[i]); + } + size_t existing_columns = block.columns(); + + /** If you use FULL or RIGHT JOIN, then the columns from the "left" table must be materialized. + * Because if they are constants, then in the "not joined" rows, they may have different values + * - default values, which can differ from the values of these constants. + */ + if constexpr (join_features.right || join_features.full) + { + materializeBlockInplace(block); + } + + /** For LEFT/INNER JOIN, the saved blocks do not contain keys. + * For FULL/RIGHT JOIN, the saved blocks contain keys; + * but they will not be used at this stage of joining (and will be in `AdderNonJoined`), and they need to be skipped. + * For ASOF, the last column is used as the ASOF column + */ + AddedColumns added_columns( + block, + block_with_columns_to_add, + join.savedBlockSample(), + join, + std::move(join_on_keys), + join.table_join->getMixedJoinExpression(), + join_features.is_asof_join, + is_join_get); + + bool has_required_right_keys = (join.required_right_keys.columns() != 0); + added_columns.need_filter = join_features.need_filter || has_required_right_keys; + added_columns.max_joined_block_rows = join.max_joined_block_rows; + if (!added_columns.max_joined_block_rows) + added_columns.max_joined_block_rows = std::numeric_limits::max(); + else + added_columns.reserve(join_features.need_replication); + + size_t num_joined = switchJoinRightColumns(maps_, added_columns, join.data->type, *join.used_flags); + /// Do not hold memory for join_on_keys anymore + added_columns.join_on_keys.clear(); + Block remaining_block = sliceBlock(block, num_joined); + + added_columns.buildOutput(); + for (size_t i = 0; i < added_columns.size(); ++i) + block.insert(added_columns.moveColumn(i)); + + std::vector right_keys_to_replicate [[maybe_unused]]; + + if constexpr (join_features.need_filter) + { + /// If ANY INNER | RIGHT JOIN - filter all the columns except the new ones. + for (size_t i = 0; i < existing_columns; ++i) + block.safeGetByPosition(i).column = block.safeGetByPosition(i).column->filter(added_columns.filter, -1); + + /// Add join key columns from right block if needed using value from left table because of equality + for (size_t i = 0; i < join.required_right_keys.columns(); ++i) + { + const auto & right_key = join.required_right_keys.getByPosition(i); + /// asof column is already in block. + if (join_features.is_asof_join && right_key.name == join.table_join->getOnlyClause().key_names_right.back()) + continue; + + const auto & left_column = block.getByName(join.required_right_keys_sources[i]); + const auto & right_col_name = join.getTableJoin().renamedRightColumnName(right_key.name); + auto right_col = copyLeftKeyColumnToRight(right_key.type, right_col_name, left_column); + block.insert(std::move(right_col)); + } + } + else if (has_required_right_keys) + { + /// Add join key columns from right block if needed. + for (size_t i = 0; i < join.required_right_keys.columns(); ++i) + { + const auto & right_key = join.required_right_keys.getByPosition(i); + auto right_col_name = join.getTableJoin().renamedRightColumnName(right_key.name); + /// asof column is already in block. + if (join_features.is_asof_join && right_key.name == join.table_join->getOnlyClause().key_names_right.back()) + continue; + + const auto & left_column = block.getByName(join.required_right_keys_sources[i]); + auto right_col = copyLeftKeyColumnToRight(right_key.type, right_col_name, left_column, &added_columns.filter); + block.insert(std::move(right_col)); + + if constexpr (join_features.need_replication) + right_keys_to_replicate.push_back(block.getPositionByName(right_col_name)); + } + } + + if constexpr (join_features.need_replication) + { + std::unique_ptr & offsets_to_replicate = added_columns.offsets_to_replicate; + + /// If ALL ... JOIN - we replicate all the columns except the new ones. + for (size_t i = 0; i < existing_columns; ++i) + { + block.safeGetByPosition(i).column = block.safeGetByPosition(i).column->replicate(*offsets_to_replicate); + } + + /// Replicate additional right keys + for (size_t pos : right_keys_to_replicate) + { + block.safeGetByPosition(pos).column = block.safeGetByPosition(pos).column->replicate(*offsets_to_replicate); + } + } + + return remaining_block; + } + +private: + template + static KeyGetter createKeyGetter(const ColumnRawPtrs & key_columns, const Sizes & key_sizes) + { + if constexpr (is_asof_join) + { + auto key_column_copy = key_columns; + auto key_size_copy = key_sizes; + key_column_copy.pop_back(); + key_size_copy.pop_back(); + return KeyGetter(key_column_copy, key_size_copy, nullptr); + } + else + return KeyGetter(key_columns, key_sizes, nullptr); + } + + template + static size_t NO_INLINE insertFromBlockImplTypeCase( + HashJoin & join, HashMap & map, size_t rows, const ColumnRawPtrs & key_columns, + const Sizes & key_sizes, Block * stored_block, ConstNullMapPtr null_map, UInt8ColumnDataPtr join_mask, Arena & pool, bool & is_inserted) + { + [[maybe_unused]] constexpr bool mapped_one = std::is_same_v; + constexpr bool is_asof_join = STRICTNESS == JoinStrictness::Asof; + + const IColumn * asof_column [[maybe_unused]] = nullptr; + if constexpr (is_asof_join) + asof_column = key_columns.back(); + + auto key_getter = createKeyGetter(key_columns, key_sizes); + + /// For ALL and ASOF join always insert values + is_inserted = !mapped_one || is_asof_join; + + for (size_t i = 0; i < rows; ++i) + { + if (null_map && (*null_map)[i]) + { + /// nulls are not inserted into hash table, + /// keep them for RIGHT and FULL joins + is_inserted = true; + continue; + } + + /// Check condition for right table from ON section + if (join_mask && !(*join_mask)[i]) + continue; + + if constexpr (is_asof_join) + Inserter::insertAsof(join, map, key_getter, stored_block, i, pool, *asof_column); + else if constexpr (mapped_one) + is_inserted |= Inserter::insertOne(join, map, key_getter, stored_block, i, pool); + else + Inserter::insertAll(join, map, key_getter, stored_block, i, pool); + } + return map.getBufferSizeInCells(); + } + + template + static size_t switchJoinRightColumns( + const std::vector & mapv, + AddedColumns & added_columns, + HashJoin::Type type, + JoinStuff::JoinUsedFlags & used_flags) + { + constexpr bool is_asof_join = STRICTNESS == JoinStrictness::Asof; + switch (type) + { + case HashJoin::Type::EMPTY: { + if constexpr (!is_asof_join) + { + using KeyGetter = KeyGetterEmpty; + std::vector key_getter_vector; + key_getter_vector.emplace_back(); + + using MapTypeVal = typename KeyGetter::MappedType; + std::vector a_map_type_vector; + a_map_type_vector.emplace_back(); + return joinRightColumnsSwitchNullability( + std::move(key_getter_vector), a_map_type_vector, added_columns, used_flags); + } + throw Exception(ErrorCodes::UNSUPPORTED_JOIN_KEYS, "Unsupported JOIN keys. Type: {}", type); + } + #define M(TYPE) \ + case HashJoin::Type::TYPE: \ + { \ + using MapTypeVal = const typename std::remove_reference_t::element_type; \ + using KeyGetter = typename KeyGetterForType::Type; \ + std::vector a_map_type_vector(mapv.size()); \ + std::vector key_getter_vector; \ + for (size_t d = 0; d < added_columns.join_on_keys.size(); ++d) \ + { \ + const auto & join_on_key = added_columns.join_on_keys[d]; \ + a_map_type_vector[d] = mapv[d]->TYPE.get(); \ + key_getter_vector.push_back(std::move(createKeyGetter(join_on_key.key_columns, join_on_key.key_sizes))); \ + } \ + return joinRightColumnsSwitchNullability( \ + std::move(key_getter_vector), a_map_type_vector, added_columns, used_flags); \ + } + APPLY_FOR_JOIN_VARIANTS(M) + #undef M + + default: + throw Exception(ErrorCodes::UNSUPPORTED_JOIN_KEYS, "Unsupported JOIN keys (type: {})", type); + } + } + + template + static size_t joinRightColumnsSwitchNullability( + std::vector && key_getter_vector, + const std::vector & mapv, + AddedColumns & added_columns, + JoinStuff::JoinUsedFlags & used_flags) + { + if (added_columns.need_filter) + { + return joinRightColumnsSwitchMultipleDisjuncts( + std::forward>(key_getter_vector), mapv, added_columns, used_flags); + } + else + { + return joinRightColumnsSwitchMultipleDisjuncts( + std::forward>(key_getter_vector), mapv, added_columns, used_flags); + } + } + + template + static size_t joinRightColumnsSwitchMultipleDisjuncts( + std::vector && key_getter_vector, + const std::vector & mapv, + AddedColumns & added_columns, + JoinStuff::JoinUsedFlags & used_flags) + { + constexpr JoinFeatures join_features; + if constexpr (join_features.is_all_join) + { + if (added_columns.additional_filter_expression) + { + bool mark_per_row_used = join_features.right || join_features.full || mapv.size() > 1; + return joinRightColumnsWithAddtitionalFilter( + std::forward>(key_getter_vector), + mapv, + added_columns, + used_flags, + need_filter, + join_features.need_flags, + join_features.add_missing, + mark_per_row_used); + } + } + + if (added_columns.additional_filter_expression) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Additional filter expression is not supported for this JOIN"); + + return mapv.size() > 1 ? joinRightColumns( + std::forward>(key_getter_vector), mapv, added_columns, used_flags) + : joinRightColumns( + std::forward>(key_getter_vector), mapv, added_columns, used_flags); + } + + /// Joins right table columns which indexes are present in right_indexes using specified map. + /// Makes filter (1 if row presented in right table) and returns offsets to replicate (for ALL JOINS). + template + static size_t joinRightColumns( + std::vector && key_getter_vector, + const std::vector & mapv, + AddedColumns & added_columns, + JoinStuff::JoinUsedFlags & used_flags) + { + constexpr JoinFeatures join_features; + + size_t rows = added_columns.rows_to_add; + if constexpr (need_filter) + added_columns.filter = IColumn::Filter(rows, 0); + + Arena pool; + + if constexpr (join_features.need_replication) + added_columns.offsets_to_replicate = std::make_unique(rows); + + IColumn::Offset current_offset = 0; + size_t max_joined_block_rows = added_columns.max_joined_block_rows; + size_t i = 0; + for (; i < rows; ++i) + { + if constexpr (join_features.need_replication) + { + if (unlikely(current_offset >= max_joined_block_rows)) + { + added_columns.offsets_to_replicate->resize_assume_reserved(i); + added_columns.filter.resize_assume_reserved(i); + break; + } + } + + bool right_row_found = false; + + KnownRowsHolder known_rows; + for (size_t onexpr_idx = 0; onexpr_idx < added_columns.join_on_keys.size(); ++onexpr_idx) + { + const auto & join_keys = added_columns.join_on_keys[onexpr_idx]; + if (join_keys.null_map && (*join_keys.null_map)[i]) + continue; + + bool row_acceptable = !join_keys.isRowFiltered(i); + using FindResult = typename KeyGetter::FindResult; + auto find_result = row_acceptable ? key_getter_vector[onexpr_idx].findKey(*(mapv[onexpr_idx]), i, pool) : FindResult(); + + if (find_result.isFound()) + { + right_row_found = true; + auto & mapped = find_result.getMapped(); + if constexpr (join_features.is_asof_join) + { + const IColumn & left_asof_key = added_columns.leftAsofKey(); + + auto row_ref = mapped->findAsof(left_asof_key, i); + if (row_ref.block) + { + setUsed(added_columns.filter, i); + if constexpr (flag_per_row) + used_flags.template setUsed(row_ref.block, row_ref.row_num, 0); + else + used_flags.template setUsed(find_result); + + added_columns.appendFromBlock(*row_ref.block, row_ref.row_num, join_features.add_missing); + } + else + addNotFoundRow(added_columns, current_offset); + } + else if constexpr (join_features.is_all_join) + { + setUsed(added_columns.filter, i); + used_flags.template setUsed(find_result); + auto used_flags_opt = join_features.need_flags ? &used_flags : nullptr; + addFoundRowAll(mapped, added_columns, current_offset, known_rows, used_flags_opt); + } + else if constexpr ((join_features.is_any_join || join_features.is_semi_join) && join_features.right) + { + /// Use first appeared left key + it needs left columns replication + bool used_once = used_flags.template setUsedOnce(find_result); + if (used_once) + { + auto used_flags_opt = join_features.need_flags ? &used_flags : nullptr; + setUsed(added_columns.filter, i); + addFoundRowAll( + mapped, added_columns, current_offset, known_rows, used_flags_opt); + } + } + else if constexpr (join_features.is_any_join && KIND == JoinKind::Inner) + { + bool used_once = used_flags.template setUsedOnce(find_result); + + /// Use first appeared left key only + if (used_once) + { + setUsed(added_columns.filter, i); + added_columns.appendFromBlock(*mapped.block, mapped.row_num, join_features.add_missing); + } + + break; + } + else if constexpr (join_features.is_any_join && join_features.full) + { + /// TODO + } + else if constexpr (join_features.is_anti_join) + { + if constexpr (join_features.right && join_features.need_flags) + used_flags.template setUsed(find_result); + } + else /// ANY LEFT, SEMI LEFT, old ANY (RightAny) + { + setUsed(added_columns.filter, i); + used_flags.template setUsed(find_result); + added_columns.appendFromBlock(*mapped.block, mapped.row_num, join_features.add_missing); + + if (join_features.is_any_or_semi_join) + { + break; + } + } + } + } + + if (!right_row_found) + { + if constexpr (join_features.is_anti_join && join_features.left) + setUsed(added_columns.filter, i); + addNotFoundRow(added_columns, current_offset); + } + + if constexpr (join_features.need_replication) + { + (*added_columns.offsets_to_replicate)[i] = current_offset; + } + } + + added_columns.applyLazyDefaults(); + return i; + } + + template + static void setUsed(IColumn::Filter & filter [[maybe_unused]], size_t pos [[maybe_unused]]) + { + if constexpr (need_filter) + filter[pos] = 1; + } + + template + static ColumnPtr buildAdditionalFilter( + size_t left_start_row, + const std::vector & selected_rows, + const std::vector & row_replicate_offset, + AddedColumns & added_columns) + { + ColumnPtr result_column; + do + { + if (selected_rows.empty()) + { + result_column = ColumnUInt8::create(); + break; + } + const Block & sample_right_block = *selected_rows.begin()->block; + if (!sample_right_block || !added_columns.additional_filter_expression) + { + auto filter = ColumnUInt8::create(); + filter->insertMany(1, selected_rows.size()); + result_column = std::move(filter); + break; + } + + auto required_cols = added_columns.additional_filter_expression->getRequiredColumnsWithTypes(); + if (required_cols.empty()) + { + Block block; + added_columns.additional_filter_expression->execute(block); + result_column = block.getByPosition(0).column->cloneResized(selected_rows.size()); + break; + } + NameSet required_column_names; + for (auto & col : required_cols) + required_column_names.insert(col.name); + + Block executed_block; + size_t right_col_pos = 0; + for (const auto & col : sample_right_block.getColumnsWithTypeAndName()) + { + if (required_column_names.contains(col.name)) + { + auto new_col = col.column->cloneEmpty(); + for (const auto & selected_row : selected_rows) + { + const auto & src_col = selected_row.block->getByPosition(right_col_pos); + new_col->insertFrom(*src_col.column, selected_row.row_num); + } + executed_block.insert({std::move(new_col), col.type, col.name}); + } + right_col_pos += 1; + } + if (!executed_block) + { + result_column = ColumnUInt8::create(); + break; + } + + for (const auto & col_name : required_column_names) + { + const auto * src_col = added_columns.left_block.findByName(col_name); + if (!src_col) + continue; + auto new_col = src_col->column->cloneEmpty(); + size_t prev_left_offset = 0; + for (size_t i = 1; i < row_replicate_offset.size(); ++i) + { + const size_t & left_offset = row_replicate_offset[i]; + size_t rows = left_offset - prev_left_offset; + if (rows) + new_col->insertManyFrom(*src_col->column, left_start_row + i - 1, rows); + prev_left_offset = left_offset; + } + executed_block.insert({std::move(new_col), src_col->type, col_name}); + } + if (!executed_block) + { + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "required columns: [{}], but not found any in left/right table. right table: {}, left table: {}", + required_cols.toString(), + sample_right_block.dumpNames(), + added_columns.left_block.dumpNames()); + } + + for (const auto & col : executed_block.getColumnsWithTypeAndName()) + if (!col.column || !col.type) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Illegal nullptr column in input block: {}", executed_block.dumpStructure()); + + added_columns.additional_filter_expression->execute(executed_block); + result_column = executed_block.getByPosition(0).column->convertToFullColumnIfConst(); + executed_block.clear(); + } while (false); + + result_column = result_column->convertToFullIfNeeded(); + if (result_column->isNullable()) + { + /// Convert Nullable(UInt8) to UInt8 ensuring that nulls are zeros + /// Trying to avoid copying data, since we are the only owner of the column. + ColumnPtr mask_column = assert_cast(*result_column).getNullMapColumnPtr(); + + MutableColumnPtr mutable_column; + { + ColumnPtr nested_column = assert_cast(*result_column).getNestedColumnPtr(); + result_column.reset(); + mutable_column = IColumn::mutate(std::move(nested_column)); + } + + auto & column_data = assert_cast(*mutable_column).getData(); + const auto & mask_column_data = assert_cast(*mask_column).getData(); + for (size_t i = 0; i < column_data.size(); ++i) + { + if (mask_column_data[i]) + column_data[i] = 0; + } + return mutable_column; + } + return result_column; + } + + /// First to collect all matched rows refs by join keys, then filter out rows which are not true in additional filter expression. + template + static size_t joinRightColumnsWithAddtitionalFilter( + std::vector && key_getter_vector, + const std::vector & mapv, + AddedColumns & added_columns, + JoinStuff::JoinUsedFlags & used_flags [[maybe_unused]], + bool need_filter [[maybe_unused]], + bool need_flags [[maybe_unused]], + bool add_missing [[maybe_unused]], + bool flag_per_row [[maybe_unused]]) + { + size_t left_block_rows = added_columns.rows_to_add; + if (need_filter) + added_columns.filter = IColumn::Filter(left_block_rows, 0); + + std::unique_ptr pool; + + if constexpr (need_replication) + added_columns.offsets_to_replicate = std::make_unique(left_block_rows); + + std::vector row_replicate_offset; + row_replicate_offset.reserve(left_block_rows); + + using FindResult = typename KeyGetter::FindResult; + size_t max_joined_block_rows = added_columns.max_joined_block_rows; + size_t left_row_iter = 0; + PreSelectedRows selected_rows; + selected_rows.reserve(left_block_rows); + std::vector find_results; + find_results.reserve(left_block_rows); + bool exceeded_max_block_rows = false; + IColumn::Offset total_added_rows = 0; + IColumn::Offset current_added_rows = 0; + + auto collect_keys_matched_rows_refs = [&]() + { + pool = std::make_unique(); + find_results.clear(); + row_replicate_offset.clear(); + row_replicate_offset.push_back(0); + current_added_rows = 0; + selected_rows.clear(); + for (; left_row_iter < left_block_rows; ++left_row_iter) + { + if constexpr (need_replication) + { + if (unlikely(total_added_rows + current_added_rows >= max_joined_block_rows)) + { + break; + } + } + KnownRowsHolder all_flag_known_rows; + KnownRowsHolder single_flag_know_rows; + for (size_t join_clause_idx = 0; join_clause_idx < added_columns.join_on_keys.size(); ++join_clause_idx) + { + const auto & join_keys = added_columns.join_on_keys[join_clause_idx]; + if (join_keys.null_map && (*join_keys.null_map)[left_row_iter]) + continue; + + bool row_acceptable = !join_keys.isRowFiltered(left_row_iter); + auto find_result = row_acceptable + ? key_getter_vector[join_clause_idx].findKey(*(mapv[join_clause_idx]), left_row_iter, *pool) + : FindResult(); + + if (find_result.isFound()) + { + auto & mapped = find_result.getMapped(); + find_results.push_back(find_result); + if (flag_per_row) + addFoundRowAll(mapped, selected_rows, current_added_rows, all_flag_known_rows, nullptr); + else + addFoundRowAll(mapped, selected_rows, current_added_rows, single_flag_know_rows, nullptr); + } + } + row_replicate_offset.push_back(current_added_rows); + } + }; + + auto copy_final_matched_rows = [&](size_t left_start_row, ColumnPtr filter_col) + { + const PaddedPODArray & filter_flags = assert_cast(*filter_col).getData(); + + size_t prev_replicated_row = 0; + auto selected_right_row_it = selected_rows.begin(); + size_t find_result_index = 0; + for (size_t i = 1, n = row_replicate_offset.size(); i < n; ++i) + { + bool any_matched = false; + /// For all right join, flag_per_row is true, we need mark used flags for each row. + if (flag_per_row) + { + for (size_t replicated_row = prev_replicated_row; replicated_row < row_replicate_offset[i]; ++replicated_row) + { + if (filter_flags[replicated_row]) + { + any_matched = true; + added_columns.appendFromBlock(*selected_right_row_it->block, selected_right_row_it->row_num, add_missing); + total_added_rows += 1; + if (need_flags) + used_flags.template setUsed(selected_right_row_it->block, selected_right_row_it->row_num, 0); + } + ++selected_right_row_it; + } + } + else + { + for (size_t replicated_row = prev_replicated_row; replicated_row < row_replicate_offset[i]; ++replicated_row) + { + if (filter_flags[replicated_row]) + { + any_matched = true; + added_columns.appendFromBlock(*selected_right_row_it->block, selected_right_row_it->row_num, add_missing); + total_added_rows += 1; + } + ++selected_right_row_it; + } + } + if (!any_matched) + { + if (add_missing) + addNotFoundRow(added_columns, total_added_rows); + else + addNotFoundRow(added_columns, total_added_rows); + } + else + { + if (!flag_per_row && need_flags) + used_flags.template setUsed(find_results[find_result_index]); + if (need_filter) + setUsed(added_columns.filter, left_start_row + i - 1); + if (add_missing) + added_columns.applyLazyDefaults(); + } + find_result_index += (prev_replicated_row != row_replicate_offset[i]); + + if constexpr (need_replication) + { + (*added_columns.offsets_to_replicate)[left_start_row + i - 1] = total_added_rows; + } + prev_replicated_row = row_replicate_offset[i]; + } + }; + + while (left_row_iter < left_block_rows && !exceeded_max_block_rows) + { + auto left_start_row = left_row_iter; + collect_keys_matched_rows_refs(); + if (selected_rows.size() != current_added_rows || row_replicate_offset.size() != left_row_iter - left_start_row + 1) + { + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "Sizes are mismatched. selected_rows.size:{}, current_added_rows:{}, row_replicate_offset.size:{}, left_row_iter: {}, " + "left_start_row: {}", + selected_rows.size(), + current_added_rows, + row_replicate_offset.size(), + left_row_iter, + left_start_row); + } + auto filter_col = buildAdditionalFilter(left_start_row, selected_rows, row_replicate_offset, added_columns); + copy_final_matched_rows(left_start_row, filter_col); + + if constexpr (need_replication) + { + // Add a check for current_added_rows to avoid run the filter expression on too small size batch. + if (total_added_rows >= max_joined_block_rows || current_added_rows < 1024) + exceeded_max_block_rows = true; + } + } + + if constexpr (need_replication) + { + added_columns.offsets_to_replicate->resize_assume_reserved(left_row_iter); + added_columns.filter.resize_assume_reserved(left_row_iter); + } + added_columns.applyLazyDefaults(); + return left_row_iter; + } + + /// Cut first num_rows rows from block in place and returns block with remaining rows + static Block sliceBlock(Block & block, size_t num_rows) + { + size_t total_rows = block.rows(); + if (num_rows >= total_rows) + return {}; + size_t remaining_rows = total_rows - num_rows; + Block remaining_block = block.cloneEmpty(); + for (size_t i = 0; i < block.columns(); ++i) + { + auto & col = block.getByPosition(i); + remaining_block.getByPosition(i).column = col.column->cut(num_rows, remaining_rows); + col.column = col.column->cut(0, num_rows); + } + return remaining_block; + } + + /** Since we do not store right key columns, + * this function is used to copy left key columns to right key columns. + * If the user requests some right columns, we just copy left key columns to right, since they are equal. + * Example: SELECT t1.key, t2.key FROM t1 FULL JOIN t2 ON t1.key = t2.key; + * In that case for matched rows in t2.key we will use values from t1.key. + * However, in some cases we might need to adjust the type of column, e.g. t1.key :: LowCardinality(String) and t2.key :: String + * Also, the nullability of the column might be different. + * Returns the right column after with necessary adjustments. + */ + static ColumnWithTypeAndName copyLeftKeyColumnToRight( + const DataTypePtr & right_key_type, + const String & renamed_right_column, + const ColumnWithTypeAndName & left_column, + const IColumn::Filter * null_map_filter = nullptr) + { + ColumnWithTypeAndName right_column = left_column; + right_column.name = renamed_right_column; + + if (null_map_filter) + right_column.column = JoinCommon::filterWithBlanks(right_column.column, *null_map_filter); + + bool should_be_nullable = isNullableOrLowCardinalityNullable(right_key_type); + if (null_map_filter) + correctNullabilityInplace(right_column, should_be_nullable, *null_map_filter); + else + correctNullabilityInplace(right_column, should_be_nullable); + + if (!right_column.type->equals(*right_key_type)) + { + right_column.column = castColumnAccurate(right_column, right_key_type); + right_column.type = right_key_type; + } + + right_column.column = right_column.column->convertToFullColumnIfConst(); + return right_column; + } + + static void correctNullabilityInplace(ColumnWithTypeAndName & column, bool nullable) + { + if (nullable) + { + JoinCommon::convertColumnToNullable(column); + } + else + { + /// We have to replace values masked by NULLs with defaults. + if (column.column) + if (const auto * nullable_column = checkAndGetColumn(&*column.column)) + column.column = JoinCommon::filterWithBlanks(column.column, nullable_column->getNullMapColumn().getData(), true); + + JoinCommon::removeColumnNullability(column); + } + } + + static void correctNullabilityInplace(ColumnWithTypeAndName & column, bool nullable, const IColumn::Filter & negative_null_map) + { + if (nullable) + { + JoinCommon::convertColumnToNullable(column); + if (column.type->isNullable() && !negative_null_map.empty()) + { + MutableColumnPtr mutable_column = IColumn::mutate(std::move(column.column)); + assert_cast(*mutable_column).applyNegatedNullMap(negative_null_map); + column.column = std::move(mutable_column); + } + } + else + JoinCommon::removeColumnNullability(column); + } +}; + +/// Instantiate template class ahead in different .cpp files to avoid `too large translation unit`. +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; + +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; + +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; + +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +extern template class HashJoinMethods; +} + diff --git a/src/Interpreters/HashJoin/InnerHashJoin.cpp b/src/Interpreters/HashJoin/InnerHashJoin.cpp new file mode 100644 index 00000000000..85aedf3a8e5 --- /dev/null +++ b/src/Interpreters/HashJoin/InnerHashJoin.cpp @@ -0,0 +1,12 @@ + +#include + +namespace DB +{ +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +} diff --git a/src/Interpreters/HashJoin/JoinFeatures.h b/src/Interpreters/HashJoin/JoinFeatures.h new file mode 100644 index 00000000000..2f2bd1e29a2 --- /dev/null +++ b/src/Interpreters/HashJoin/JoinFeatures.h @@ -0,0 +1,28 @@ +#pragma once +#include +#include +namespace DB +{ +template +struct JoinFeatures +{ + static constexpr bool is_any_join = STRICTNESS == JoinStrictness::Any; + static constexpr bool is_any_or_semi_join = STRICTNESS == JoinStrictness::Any || STRICTNESS == JoinStrictness::RightAny || (STRICTNESS == JoinStrictness::Semi && KIND == JoinKind::Left); + static constexpr bool is_all_join = STRICTNESS == JoinStrictness::All; + static constexpr bool is_asof_join = STRICTNESS == JoinStrictness::Asof; + static constexpr bool is_semi_join = STRICTNESS == JoinStrictness::Semi; + static constexpr bool is_anti_join = STRICTNESS == JoinStrictness::Anti; + + static constexpr bool left = KIND == JoinKind::Left; + static constexpr bool right = KIND == JoinKind::Right; + static constexpr bool inner = KIND == JoinKind::Inner; + static constexpr bool full = KIND == JoinKind::Full; + + static constexpr bool need_replication = is_all_join || (is_any_join && right) || (is_semi_join && right); + static constexpr bool need_filter = !need_replication && (inner || right || (is_semi_join && left) || (is_anti_join && left)); + static constexpr bool add_missing = (left || full) && !is_semi_join; + + static constexpr bool need_flags = MapGetter::flagged; +}; + +} diff --git a/src/Interpreters/HashJoin/JoinUsedFlags.h b/src/Interpreters/HashJoin/JoinUsedFlags.h new file mode 100644 index 00000000000..bd41ba2073f --- /dev/null +++ b/src/Interpreters/HashJoin/JoinUsedFlags.h @@ -0,0 +1,154 @@ +#pragma once +#include +#include +#include +#include +#include +#include + +namespace DB +{ +namespace JoinStuff +{ +/// Flags needed to implement RIGHT and FULL JOINs. +class JoinUsedFlags +{ + using RawBlockPtr = const Block *; + using UsedFlagsForBlock = std::vector; + + /// For multiple dijuncts each empty in hashmap stores flags for particular block + /// For single dicunct we store all flags in `nullptr` entry, index is the offset in FindResult + std::unordered_map flags; + + bool need_flags; + +public: + /// Update size for vector with flags. + /// Calling this method invalidates existing flags. + /// It can be called several times, but all of them should happen before using this structure. + template + void reinit(size_t size) + { + if constexpr (MapGetter::flagged) + { + assert(flags[nullptr].size() <= size); + need_flags = true; + // For one disjunct clause case, we don't need to reinit each time we call addBlockToJoin. + // and there is no value inserted in this JoinUsedFlags before addBlockToJoin finish. + // So we reinit only when the hash table is rehashed to a larger size. + if (flags.empty() || flags[nullptr].size() < size) [[unlikely]] + { + flags[nullptr] = std::vector(size); + } + } + } + + template + void reinit(const Block * block_ptr) + { + if constexpr (MapGetter::flagged) + { + assert(flags[block_ptr].size() <= block_ptr->rows()); + need_flags = true; + flags[block_ptr] = std::vector(block_ptr->rows()); + } + } + + bool getUsedSafe(size_t i) const + { + return getUsedSafe(nullptr, i); + } + bool getUsedSafe(const Block * block_ptr, size_t row_idx) const + { + if (auto it = flags.find(block_ptr); it != flags.end()) + return it->second[row_idx].load(); + return !need_flags; + } + + template + void setUsed(const FindResult & f) + { + if constexpr (!use_flags) + return; + + /// Could be set simultaneously from different threads. + if constexpr (flag_per_row) + { + auto & mapped = f.getMapped(); + flags[mapped.block][mapped.row_num].store(true, std::memory_order_relaxed); + } + else + { + flags[nullptr][f.getOffset()].store(true, std::memory_order_relaxed); + } + } + + template + void setUsed(const Block * block, size_t row_num, size_t offset) + { + if constexpr (!use_flags) + return; + + /// Could be set simultaneously from different threads. + if constexpr (flag_per_row) + { + flags[block][row_num].store(true, std::memory_order_relaxed); + } + else + { + flags[nullptr][offset].store(true, std::memory_order_relaxed); + } + } + + template + bool getUsed(const FindResult & f) + { + if constexpr (!use_flags) + return true; + + if constexpr (flag_per_row) + { + auto & mapped = f.getMapped(); + return flags[mapped.block][mapped.row_num].load(); + } + else + { + return flags[nullptr][f.getOffset()].load(); + } + + } + + template + bool setUsedOnce(const FindResult & f) + { + if constexpr (!use_flags) + return true; + + if constexpr (flag_per_row) + { + auto & mapped = f.getMapped(); + + /// fast check to prevent heavy CAS with seq_cst order + if (flags[mapped.block][mapped.row_num].load(std::memory_order_relaxed)) + return false; + + bool expected = false; + return flags[mapped.block][mapped.row_num].compare_exchange_strong(expected, true); + } + else + { + auto off = f.getOffset(); + + /// fast check to prevent heavy CAS with seq_cst order + if (flags[nullptr][off].load(std::memory_order_relaxed)) + return false; + + bool expected = false; + return flags[nullptr][off].compare_exchange_strong(expected, true); + } + + } +}; + +} +} diff --git a/src/Interpreters/HashJoin/KeyGetter.h b/src/Interpreters/HashJoin/KeyGetter.h new file mode 100644 index 00000000000..35ff2bb6eb5 --- /dev/null +++ b/src/Interpreters/HashJoin/KeyGetter.h @@ -0,0 +1,73 @@ +#pragma once +#include + + +namespace DB +{ +template +class KeyGetterEmpty +{ +public: + struct MappedType + { + using mapped_type = Mapped; + }; + + using FindResult = ColumnsHashing::columns_hashing_impl::FindResultImpl; + + KeyGetterEmpty() = default; + + FindResult findKey(MappedType, size_t, const Arena &) { return FindResult(); } +}; + +template +struct KeyGetterForTypeImpl; + +constexpr bool use_offset = true; + +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodOneNumber; +}; +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodOneNumber; +}; +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodOneNumber; +}; +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodOneNumber; +}; +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodString; +}; +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodFixedString; +}; +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodKeysFixed; +}; +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodKeysFixed; +}; +template struct KeyGetterForTypeImpl +{ + using Type = ColumnsHashing::HashMethodHashed; +}; + +template +struct KeyGetterForType +{ + using Value = typename Data::value_type; + using Mapped_t = typename Data::mapped_type; + using Mapped = std::conditional_t, const Mapped_t, Mapped_t>; + using Type = typename KeyGetterForTypeImpl::Type; +}; +} diff --git a/src/Interpreters/HashJoin/KnowRowsHolder.h b/src/Interpreters/HashJoin/KnowRowsHolder.h new file mode 100644 index 00000000000..d51c96893c5 --- /dev/null +++ b/src/Interpreters/HashJoin/KnowRowsHolder.h @@ -0,0 +1,148 @@ +#pragma once +#include +#include +#include +#include +namespace DB +{ + +template +class KnownRowsHolder; + +/// Keep already joined rows to prevent duplication if many disjuncts +/// if for a particular pair of rows condition looks like TRUE or TRUE or TRUE +/// we want to have it once in resultset +template<> +class KnownRowsHolder +{ +public: + using Type = std::pair; + +private: + static const size_t MAX_LINEAR = 16; // threshold to switch from Array to Set + using ArrayHolder = std::array; + using SetHolder = std::set; + using SetHolderPtr = std::unique_ptr; + + ArrayHolder array_holder; + SetHolderPtr set_holder_ptr; + + size_t items; + +public: + KnownRowsHolder() + : items(0) + { + } + + + template + void add(InputIt from, InputIt to) + { + const size_t new_items = std::distance(from, to); + if (items + new_items <= MAX_LINEAR) + { + std::copy(from, to, &array_holder[items]); + } + else + { + if (items <= MAX_LINEAR) + { + set_holder_ptr = std::make_unique(); + set_holder_ptr->insert(std::cbegin(array_holder), std::cbegin(array_holder) + items); + } + set_holder_ptr->insert(from, to); + } + items += new_items; + } + + template + bool isKnown(const Needle & needle) + { + return items <= MAX_LINEAR + ? std::find(std::cbegin(array_holder), std::cbegin(array_holder) + items, needle) != std::cbegin(array_holder) + items + : set_holder_ptr->find(needle) != set_holder_ptr->end(); + } +}; + +template<> +class KnownRowsHolder +{ +public: + template + void add(InputIt, InputIt) + { + } + + template + static bool isKnown(const Needle &) + { + return false; + } +}; + +template +using FindResultImpl = ColumnsHashing::columns_hashing_impl::FindResultImpl; + + +template +void addFoundRowAll( + const typename Map::mapped_type & mapped, + AddedColumns & added, + IColumn::Offset & current_offset, + KnownRowsHolder & known_rows [[maybe_unused]], + JoinStuff::JoinUsedFlags * used_flags [[maybe_unused]]) +{ + if constexpr (add_missing) + added.applyLazyDefaults(); + + if constexpr (flag_per_row) + { + std::unique_ptr::Type>> new_known_rows_ptr; + + for (auto it = mapped.begin(); it.ok(); ++it) + { + if (!known_rows.isKnown(std::make_pair(it->block, it->row_num))) + { + added.appendFromBlock(*it->block, it->row_num, false); + ++current_offset; + if (!new_known_rows_ptr) + { + new_known_rows_ptr = std::make_unique::Type>>(); + } + new_known_rows_ptr->push_back(std::make_pair(it->block, it->row_num)); + if (used_flags) + { + used_flags->JoinStuff::JoinUsedFlags::setUsedOnce( + FindResultImpl(*it, true, 0)); + } + } + } + + if (new_known_rows_ptr) + { + known_rows.add(std::cbegin(*new_known_rows_ptr), std::cend(*new_known_rows_ptr)); + } + } + else + { + for (auto it = mapped.begin(); it.ok(); ++it) + { + added.appendFromBlock(*it->block, it->row_num, false); + ++current_offset; + } + } +} + +template +void addNotFoundRow(AddedColumns & added [[maybe_unused]], IColumn::Offset & current_offset [[maybe_unused]]) +{ + if constexpr (add_missing) + { + added.appendDefaultRow(); + if constexpr (need_offset) + ++current_offset; + } +} + +} diff --git a/src/Interpreters/HashJoin/LeftHashJoin.cpp b/src/Interpreters/HashJoin/LeftHashJoin.cpp new file mode 100644 index 00000000000..69e17ff70bd --- /dev/null +++ b/src/Interpreters/HashJoin/LeftHashJoin.cpp @@ -0,0 +1,11 @@ +#include + +namespace DB +{ +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +} diff --git a/src/Interpreters/HashJoin/RightHashJoin.cpp b/src/Interpreters/HashJoin/RightHashJoin.cpp new file mode 100644 index 00000000000..8e304754f5c --- /dev/null +++ b/src/Interpreters/HashJoin/RightHashJoin.cpp @@ -0,0 +1,11 @@ +#include + +namespace DB +{ +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +template class HashJoinMethods; +} diff --git a/src/Interpreters/HashTablesStatistics.cpp b/src/Interpreters/HashTablesStatistics.cpp new file mode 100644 index 00000000000..d66f1bbd1d3 --- /dev/null +++ b/src/Interpreters/HashTablesStatistics.cpp @@ -0,0 +1,117 @@ +#include + +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ +extern const int LOGICAL_ERROR; +} + +std::optional HashTablesStatistics::getSizeHint(const Params & params) +{ + if (!params.isCollectionAndUseEnabled()) + throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Collection and use of the statistics should be enabled."); + + std::lock_guard lock(mutex); + const auto cache = getHashTableStatsCache(params, lock); + if (const auto hint = cache->get(params.key)) + { + LOG_TRACE( + getLogger("HashTablesStatistics"), + "An entry for key={} found in cache: sum_of_sizes={}, median_size={}", + params.key, + hint->sum_of_sizes, + hint->median_size); + return *hint; + } + return std::nullopt; +} + +/// Collection and use of the statistics should be enabled. +void HashTablesStatistics::update(size_t sum_of_sizes, size_t median_size, const Params & params) +{ + if (!params.isCollectionAndUseEnabled()) + throw DB::Exception(DB::ErrorCodes::LOGICAL_ERROR, "Collection and use of the statistics should be enabled."); + + std::lock_guard lock(mutex); + const auto cache = getHashTableStatsCache(params, lock); + const auto hint = cache->get(params.key); + // We'll maintain the maximum among all the observed values until another prediction is much lower (that should indicate some change) + if (!hint || sum_of_sizes < hint->sum_of_sizes / 2 || hint->sum_of_sizes < sum_of_sizes || median_size < hint->median_size / 2 + || hint->median_size < median_size) + { + LOG_TRACE( + getLogger("HashTablesStatistics"), + "Statistics updated for key={}: new sum_of_sizes={}, median_size={}", + params.key, + sum_of_sizes, + median_size); + cache->set(params.key, std::make_shared(Entry{.sum_of_sizes = sum_of_sizes, .median_size = median_size})); + } +} + +std::optional HashTablesStatistics::getCacheStats() const +{ + std::lock_guard lock(mutex); + if (hash_table_stats) + { + size_t hits = 0, misses = 0; + hash_table_stats->getStats(hits, misses); + return DB::HashTablesCacheStatistics{.entries = hash_table_stats->count(), .hits = hits, .misses = misses}; + } + return std::nullopt; +} + +HashTablesStatistics::CachePtr HashTablesStatistics::getHashTableStatsCache(const Params & params, const std::lock_guard &) +{ + if (!hash_table_stats) + hash_table_stats = std::make_shared(params.max_entries_for_hash_table_stats * sizeof(Entry)); + return hash_table_stats; +} + +HashTablesStatistics & getHashTablesStatistics() +{ + static HashTablesStatistics hash_tables_stats; + return hash_tables_stats; +} + +std::optional getHashTablesCacheStatistics() +{ + return getHashTablesStatistics().getCacheStats(); +} + +std::optional getSizeHint(const DB::StatsCollectingParams & stats_collecting_params, size_t tables_cnt) +{ + if (stats_collecting_params.isCollectionAndUseEnabled()) + { + if (auto hint = DB::getHashTablesStatistics().getSizeHint(stats_collecting_params)) + { + const auto lower_limit = hint->sum_of_sizes / tables_cnt; + const auto upper_limit = stats_collecting_params.max_size_to_preallocate / tables_cnt; + if (hint->median_size > upper_limit) + { + /// Since we cannot afford to preallocate as much as needed, we would likely have to do at least one resize anyway. + /// Though it still sounds better than N resizes, but in actuality we saw that one big resize (remember, HT-s grow exponentially) + /// plus worse cache locality since we're dealing with big HT-s from the beginning yields worse performance. + /// So let's just do nothing. + LOG_TRACE( + getLogger("HashTablesStatistics"), + "No space were preallocated in hash tables because 'max_size_to_preallocate' has too small value: {}, " + "should be at least {}", + stats_collecting_params.max_size_to_preallocate, + hint->median_size * tables_cnt); + } + /// https://github.com/ClickHouse/ClickHouse/issues/44402#issuecomment-1359920703 + else if ((tables_cnt > 1 && hint->sum_of_sizes > 100'000) || hint->sum_of_sizes > 500'000) + { + return HashTablesStatistics::Entry{hint->sum_of_sizes, std::max(lower_limit, hint->median_size)}; + } + } + } + return std::nullopt; +} +} diff --git a/src/Interpreters/HashTablesStatistics.h b/src/Interpreters/HashTablesStatistics.h new file mode 100644 index 00000000000..7b4c4fcbfeb --- /dev/null +++ b/src/Interpreters/HashTablesStatistics.h @@ -0,0 +1,69 @@ +#pragma once + +#include + +namespace DB +{ + +struct HashTablesCacheStatistics +{ + size_t entries = 0; + size_t hits = 0; + size_t misses = 0; +}; + +struct StatsCollectingParams +{ + StatsCollectingParams() = default; + + StatsCollectingParams(UInt64 key_, bool enable_, size_t max_entries_for_hash_table_stats_, size_t max_size_to_preallocate_) + : key(enable_ ? key_ : 0) + , max_entries_for_hash_table_stats(max_entries_for_hash_table_stats_) + , max_size_to_preallocate(max_size_to_preallocate_) + { + } + + bool isCollectionAndUseEnabled() const { return key != 0; } + void disable() { key = 0; } + + UInt64 key = 0; + const size_t max_entries_for_hash_table_stats = 0; + const size_t max_size_to_preallocate = 0; +}; + +/** Collects observed HashTable-s sizes to avoid redundant intermediate resizes. + */ +class HashTablesStatistics +{ +public: + struct Entry + { + size_t sum_of_sizes; // used to determine if it's better to convert aggregation to two-level from the beginning + size_t median_size; // roughly the size we're going to preallocate on each thread + }; + + using Cache = DB::CacheBase; + using CachePtr = std::shared_ptr; + using Params = StatsCollectingParams; + + /// Collection and use of the statistics should be enabled. + std::optional getSizeHint(const Params & params); + + /// Collection and use of the statistics should be enabled. + void update(size_t sum_of_sizes, size_t median_size, const Params & params); + + std::optional getCacheStats() const; + +private: + CachePtr getHashTableStatsCache(const Params & params, const std::lock_guard &); + + mutable std::mutex mutex; + CachePtr hash_table_stats; +}; + +HashTablesStatistics & getHashTablesStatistics(); + +std::optional getHashTablesCacheStatistics(); + +std::optional getSizeHint(const DB::StatsCollectingParams & stats_collecting_params, size_t tables_cnt); +} diff --git a/src/Interpreters/IInterpreter.cpp b/src/Interpreters/IInterpreter.cpp index 148e73e43ce..a291199f24b 100644 --- a/src/Interpreters/IInterpreter.cpp +++ b/src/Interpreters/IInterpreter.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/Interpreters/IInterpreterUnionOrSelectQuery.cpp b/src/Interpreters/IInterpreterUnionOrSelectQuery.cpp index fed29b410db..f64b9540dbb 100644 --- a/src/Interpreters/IInterpreterUnionOrSelectQuery.cpp +++ b/src/Interpreters/IInterpreterUnionOrSelectQuery.cpp @@ -1,4 +1,6 @@ #include + +#include #include #include #include @@ -15,6 +17,29 @@ namespace DB { +IInterpreterUnionOrSelectQuery::IInterpreterUnionOrSelectQuery(const ASTPtr & query_ptr_, + const ContextMutablePtr & context_, const SelectQueryOptions & options_) + : query_ptr(query_ptr_) + , context(context_) + , options(options_) + , max_streams(context->getSettingsRef().max_threads) +{ + /// FIXME All code here will work with the old analyzer, however for views over Distributed tables + /// it's possible that new analyzer will be enabled in ::getQueryProcessingStage method + /// of the underlying storage when all other parts of infrastructure are not ready for it + /// (built with old analyzer). + context->setSetting("allow_experimental_analyzer", false); + + if (options.shard_num) + context->addSpecialScalar( + "_shard_num", + Block{{DataTypeUInt32().createColumnConst(1, *options.shard_num), std::make_shared(), "_shard_num"}}); + if (options.shard_count) + context->addSpecialScalar( + "_shard_count", + Block{{DataTypeUInt32().createColumnConst(1, *options.shard_count), std::make_shared(), "_shard_count"}}); +} + QueryPipelineBuilder IInterpreterUnionOrSelectQuery::buildQueryPipeline() { QueryPlan query_plan; diff --git a/src/Interpreters/IInterpreterUnionOrSelectQuery.h b/src/Interpreters/IInterpreterUnionOrSelectQuery.h index e4425a73505..ac8e90b5f52 100644 --- a/src/Interpreters/IInterpreterUnionOrSelectQuery.h +++ b/src/Interpreters/IInterpreterUnionOrSelectQuery.h @@ -17,21 +17,7 @@ public: { } - IInterpreterUnionOrSelectQuery(const ASTPtr & query_ptr_, const ContextMutablePtr & context_, const SelectQueryOptions & options_) - : query_ptr(query_ptr_) - , context(context_) - , options(options_) - , max_streams(context->getSettingsRef().max_threads) - { - if (options.shard_num) - context->addSpecialScalar( - "_shard_num", - Block{{DataTypeUInt32().createColumnConst(1, *options.shard_num), std::make_shared(), "_shard_num"}}); - if (options.shard_count) - context->addSpecialScalar( - "_shard_count", - Block{{DataTypeUInt32().createColumnConst(1, *options.shard_count), std::make_shared(), "_shard_count"}}); - } + IInterpreterUnionOrSelectQuery(const ASTPtr & query_ptr_, const ContextMutablePtr & context_, const SelectQueryOptions & options_); virtual void buildQueryPlan(QueryPlan & query_plan) = 0; QueryPipelineBuilder buildQueryPipeline(); diff --git a/src/Interpreters/IdentifierSemantic.cpp b/src/Interpreters/IdentifierSemantic.cpp index 36972aeb03d..24e594661e9 100644 --- a/src/Interpreters/IdentifierSemantic.cpp +++ b/src/Interpreters/IdentifierSemantic.cpp @@ -2,6 +2,8 @@ #include +#include + #include #include diff --git a/src/Interpreters/InJoinSubqueriesPreprocessor.cpp b/src/Interpreters/InJoinSubqueriesPreprocessor.cpp index 3b3ef928b42..22ecb6a5db2 100644 --- a/src/Interpreters/InJoinSubqueriesPreprocessor.cpp +++ b/src/Interpreters/InJoinSubqueriesPreprocessor.cpp @@ -11,6 +11,7 @@ #include #include #include +#include namespace DB diff --git a/src/Interpreters/InterpreterAlterNamedCollectionQuery.cpp b/src/Interpreters/InterpreterAlterNamedCollectionQuery.cpp index 79a17fd1844..0e83e2039f6 100644 --- a/src/Interpreters/InterpreterAlterNamedCollectionQuery.cpp +++ b/src/Interpreters/InterpreterAlterNamedCollectionQuery.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -13,14 +14,16 @@ namespace DB BlockIO InterpreterAlterNamedCollectionQuery::execute() { auto current_context = getContext(); - const auto & query = query_ptr->as(); + + const auto updated_query = removeOnClusterClauseIfNeeded(query_ptr, getContext()); + const auto & query = updated_query->as(); current_context->checkAccess(AccessType::ALTER_NAMED_COLLECTION, query.collection_name); if (!query.cluster.empty()) { DDLQueryOnClusterParams params; - return executeDDLQueryOnCluster(query_ptr, current_context, params); + return executeDDLQueryOnCluster(updated_query, current_context, params); } NamedCollectionFactory::instance().updateFromSQL(query); diff --git a/src/Interpreters/InterpreterAlterQuery.cpp b/src/Interpreters/InterpreterAlterQuery.cpp index c70a3397f4e..8f9c6bc533e 100644 --- a/src/Interpreters/InterpreterAlterQuery.cpp +++ b/src/Interpreters/InterpreterAlterQuery.cpp @@ -3,6 +3,8 @@ #include #include +#include +#include #include #include #include @@ -24,7 +26,6 @@ #include #include #include -#include #include #include @@ -175,10 +176,9 @@ BlockIO InterpreterAlterQuery::executeToTable(const ASTAlterQuery & alter) else throw Exception(ErrorCodes::LOGICAL_ERROR, "Wrong parameter type in ALTER query"); - if (!getContext()->getSettings().allow_experimental_statistics && ( - command_ast->type == ASTAlterCommand::ADD_STATISTICS || - command_ast->type == ASTAlterCommand::DROP_STATISTICS || - command_ast->type == ASTAlterCommand::MATERIALIZE_STATISTICS)) + if (!getContext()->getSettingsRef().allow_experimental_statistics + && (command_ast->type == ASTAlterCommand::ADD_STATISTICS || command_ast->type == ASTAlterCommand::DROP_STATISTICS + || command_ast->type == ASTAlterCommand::MATERIALIZE_STATISTICS)) throw Exception(ErrorCodes::INCORRECT_QUERY, "Alter table with statistics is now disabled. Turn on allow_experimental_statistics"); } diff --git a/src/Interpreters/InterpreterCheckQuery.cpp b/src/Interpreters/InterpreterCheckQuery.cpp index 4a84a7bf570..2b022a8cee0 100644 --- a/src/Interpreters/InterpreterCheckQuery.cpp +++ b/src/Interpreters/InterpreterCheckQuery.cpp @@ -2,6 +2,7 @@ #include #include +#include #include @@ -12,6 +13,8 @@ #include #include +#include + #include #include @@ -22,6 +25,7 @@ #include #include +#include #include #include #include @@ -91,7 +95,7 @@ Chunk getChunkFromCheckResult(const String & database, const String & table, con return Chunk(std::move(columns), 1); } -class TableCheckTask : public ChunkInfo +class TableCheckTask : public ChunkInfoCloneable { public: TableCheckTask(StorageID table_id, const std::variant & partition_or_part, ContextPtr context) @@ -110,6 +114,12 @@ public: context->checkAccess(AccessType::SHOW_TABLES, table_->getStorageID()); } + TableCheckTask(const TableCheckTask & other) + : table(other.table) + , check_data_tasks(other.check_data_tasks) + , is_finished(other.is_finished.load()) + {} + std::optional checkNext() const { if (isFinished()) @@ -121,8 +131,8 @@ public: std::this_thread::sleep_for(sleep_time); }); - IStorage::DataValidationTasksPtr check_data_tasks_ = check_data_tasks; - auto result = table->checkDataNext(check_data_tasks_); + IStorage::DataValidationTasksPtr tmp = check_data_tasks; + auto result = table->checkDataNext(tmp); is_finished = !result.has_value(); return result; } @@ -180,7 +190,7 @@ protected: /// source should return at least one row to start pipeline result.addColumn(ColumnUInt8::create(1, 1)); /// actual data stored in chunk info - result.setChunkInfo(std::move(current_check_task)); + result.getChunkInfos().add(std::move(current_check_task)); return result; } @@ -280,7 +290,7 @@ public: protected: void transform(Chunk & chunk) override { - auto table_check_task = std::dynamic_pointer_cast(chunk.getChunkInfo()); + auto table_check_task = chunk.getChunkInfos().get(); auto check_result = table_check_task->checkNext(); if (!check_result) { diff --git a/src/Interpreters/InterpreterCreateIndexQuery.cpp b/src/Interpreters/InterpreterCreateIndexQuery.cpp index a439cb672c8..1c6abe842d4 100644 --- a/src/Interpreters/InterpreterCreateIndexQuery.cpp +++ b/src/Interpreters/InterpreterCreateIndexQuery.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include diff --git a/src/Interpreters/InterpreterCreateNamedCollectionQuery.cpp b/src/Interpreters/InterpreterCreateNamedCollectionQuery.cpp index c71441daa8c..b4920b1729f 100644 --- a/src/Interpreters/InterpreterCreateNamedCollectionQuery.cpp +++ b/src/Interpreters/InterpreterCreateNamedCollectionQuery.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -13,14 +14,16 @@ namespace DB BlockIO InterpreterCreateNamedCollectionQuery::execute() { auto current_context = getContext(); - const auto & query = query_ptr->as(); + + const auto updated_query = removeOnClusterClauseIfNeeded(query_ptr, getContext()); + const auto & query = updated_query->as(); current_context->checkAccess(AccessType::CREATE_NAMED_COLLECTION, query.collection_name); if (!query.cluster.empty()) { DDLQueryOnClusterParams params; - return executeDDLQueryOnCluster(query_ptr, current_context, params); + return executeDDLQueryOnCluster(updated_query, current_context, params); } NamedCollectionFactory::instance().createFromSQL(query); diff --git a/src/Interpreters/InterpreterCreateQuery.cpp b/src/Interpreters/InterpreterCreateQuery.cpp index 7272e10b801..454cedae15c 100644 --- a/src/Interpreters/InterpreterCreateQuery.cpp +++ b/src/Interpreters/InterpreterCreateQuery.cpp @@ -5,7 +5,7 @@ #include #include -#include "Common/Exception.h" +#include #include #include #include @@ -14,9 +14,8 @@ #include #include #include +#include #include -#include -#include #include #include @@ -34,10 +33,11 @@ #include #include +#include #include #include -#include #include +#include #include #include @@ -81,13 +81,13 @@ #include #include -#include #include #include #include #include + namespace CurrentMetrics { extern const Metric AttachedTable; @@ -146,27 +146,27 @@ BlockIO InterpreterCreateQuery::createDatabase(ASTCreateQuery & create) } auto db_num_limit = getContext()->getGlobalContext()->getServerSettings().max_database_num_to_throw; - if (db_num_limit > 0) + if (db_num_limit > 0 && !internal) { size_t db_count = DatabaseCatalog::instance().getDatabases().size(); - std::vector system_databases = { + std::initializer_list system_databases = + { DatabaseCatalog::TEMPORARY_DATABASE, DatabaseCatalog::SYSTEM_DATABASE, DatabaseCatalog::INFORMATION_SCHEMA, DatabaseCatalog::INFORMATION_SCHEMA_UPPERCASE, - DatabaseCatalog::DEFAULT_DATABASE }; for (const auto & system_database : system_databases) { - if (db_count > 0 && DatabaseCatalog::instance().isDatabaseExist(system_database)) - db_count--; + if (db_count > 0 && DatabaseCatalog::instance().isDatabaseExist(std::string(system_database))) + --db_count; } if (db_count >= db_num_limit) throw Exception(ErrorCodes::TOO_MANY_DATABASES, - "Too many databases in the Clickhouse. " - "The limit (setting 'max_database_num_to_throw') is set to {}, current number of databases is {}", + "Too many databases. " + "The limit (server configuration parameter `max_database_num_to_throw`) is set to {}, the current number of databases is {}", db_num_limit, db_count); } @@ -898,6 +898,8 @@ InterpreterCreateQuery::TableProperties InterpreterCreateQuery::getTableProperti assert(as_database_saved.empty() && as_table_saved.empty()); std::swap(create.as_database, as_database_saved); std::swap(create.as_table, as_table_saved); + if (!as_table_saved.empty()) + create.is_create_empty = false; return properties; } @@ -947,7 +949,7 @@ namespace throw Exception(ErrorCodes::INCORRECT_QUERY, "Temporary tables cannot be created with Replicated, Shared or KeeperMap table engines"); } - void setDefaultTableEngine(ASTStorage &storage, DefaultTableEngine engine) + void setDefaultTableEngine(ASTStorage & storage, DefaultTableEngine engine) { if (engine == DefaultTableEngine::None) throw Exception(ErrorCodes::ENGINE_REQUIRED, "Table engine is not specified in CREATE query"); @@ -967,9 +969,6 @@ void InterpreterCreateQuery::setEngine(ASTCreateQuery & create) const if (create.is_dictionary || create.is_ordinary_view || create.is_live_view || create.is_window_view) return; - if (create.is_materialized_view && create.to_table_id) - return; - if (create.temporary) { /// Some part of storage definition is specified, but ENGINE is not: just set the one from default_temporary_table_engine setting. @@ -984,22 +983,44 @@ void InterpreterCreateQuery::setEngine(ASTCreateQuery & create) const } if (!create.storage->engine) - { setDefaultTableEngine(*create.storage, getContext()->getSettingsRef().default_temporary_table_engine.value); - } checkTemporaryTableEngineName(create.storage->engine->name); return; } + if (create.is_materialized_view) + { + /// A materialized view with an external target doesn't need a table engine. + if (create.is_materialized_view_with_external_target()) + return; + + if (auto to_engine = create.getTargetInnerEngine(ViewTarget::To)) + { + /// This materialized view already has a storage definition. + if (!to_engine->engine) + { + /// Some part of storage definition (such as PARTITION BY) is specified, but ENGINE is not: just set default one. + setDefaultTableEngine(*to_engine, getContext()->getSettingsRef().default_table_engine.value); + } + return; + } + } + if (create.storage) { - /// Some part of storage definition (such as PARTITION BY) is specified, but ENGINE is not: just set default one. + /// This table already has a storage definition. if (!create.storage->engine) + { + /// Some part of storage definition (such as PARTITION BY) is specified, but ENGINE is not: just set default one. setDefaultTableEngine(*create.storage, getContext()->getSettingsRef().default_table_engine.value); + } return; } + /// We'll try to extract a storage definition from clause `AS`: + /// CREATE TABLE table_name AS other_table_name + std::shared_ptr storage_def; if (!create.as_table.empty()) { /// NOTE Getting the structure from the table specified in the AS is done not atomically with the creation of the table. @@ -1015,12 +1036,14 @@ void InterpreterCreateQuery::setEngine(ASTCreateQuery & create) const if (as_create.is_ordinary_view) throw Exception(ErrorCodes::INCORRECT_QUERY, "Cannot CREATE a table AS {}, it is a View", qualified_name); - if (as_create.is_materialized_view && as_create.to_table_id) + if (as_create.is_materialized_view_with_external_target()) + { throw Exception( ErrorCodes::INCORRECT_QUERY, - "Cannot CREATE a table AS {}, it is a Materialized View without storage. Use \"AS `{}`\" instead", + "Cannot CREATE a table AS {}, it is a Materialized View without storage. Use \"AS {}\" instead", qualified_name, - as_create.to_table_id.getQualifiedName()); + as_create.getTargetTableID(ViewTarget::To).getFullTableName()); + } if (as_create.is_live_view) throw Exception(ErrorCodes::INCORRECT_QUERY, "Cannot CREATE a table AS {}, it is a Live View", qualified_name); @@ -1031,18 +1054,37 @@ void InterpreterCreateQuery::setEngine(ASTCreateQuery & create) const if (as_create.is_dictionary) throw Exception(ErrorCodes::INCORRECT_QUERY, "Cannot CREATE a table AS {}, it is a Dictionary", qualified_name); - if (as_create.storage) - create.set(create.storage, as_create.storage->ptr()); + if (as_create.is_materialized_view) + { + storage_def = as_create.getTargetInnerEngine(ViewTarget::To); + } else if (as_create.as_table_function) + { create.set(create.as_table_function, as_create.as_table_function->ptr()); + return; + } + else if (as_create.storage) + { + storage_def = typeid_cast>(as_create.storage->ptr()); + } else + { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot set engine, it's a bug."); - - return; + } } - create.set(create.storage, std::make_shared()); - setDefaultTableEngine(*create.storage, getContext()->getSettingsRef().default_table_engine.value); + if (!storage_def) + { + /// Set ENGINE by default. + storage_def = std::make_shared(); + setDefaultTableEngine(*storage_def, getContext()->getSettingsRef().default_table_engine.value); + } + + /// Use the found table engine to modify the create query. + if (create.is_materialized_view) + create.setTargetInnerEngine(ViewTarget::To, storage_def); + else + create.set(create.storage, storage_def); } void InterpreterCreateQuery::assertOrSetUUID(ASTCreateQuery & create, const DatabasePtr & database) const @@ -1084,11 +1126,11 @@ void InterpreterCreateQuery::assertOrSetUUID(ASTCreateQuery & create, const Data kind_upper, create.table); } - create.generateRandomUUID(); + create.generateRandomUUIDs(); } else { - bool has_uuid = create.uuid != UUIDHelpers::Nil || create.to_inner_uuid != UUIDHelpers::Nil; + bool has_uuid = (create.uuid != UUIDHelpers::Nil) || create.hasInnerUUIDs(); if (has_uuid && !is_on_cluster && !internal) { /// We don't show the following error message either @@ -1103,12 +1145,40 @@ void InterpreterCreateQuery::assertOrSetUUID(ASTCreateQuery & create, const Data /// The database doesn't support UUID so we'll ignore it. The UUID could be set here because of either /// a) the initiator of `ON CLUSTER` query generated it to ensure the same UUIDs are used on different hosts; or /// b) `RESTORE from backup` query generated it to ensure the same UUIDs are used on different hosts. - create.uuid = UUIDHelpers::Nil; - create.to_inner_uuid = UUIDHelpers::Nil; + create.resetUUIDs(); } } +namespace +{ + +void addTableDependencies(const ASTCreateQuery & create, const ASTPtr & query_ptr, const ContextPtr & context) +{ + QualifiedTableName qualified_name{create.getDatabase(), create.getTable()}; + auto ref_dependencies = getDependenciesFromCreateQuery(context->getGlobalContext(), qualified_name, query_ptr, context->getCurrentDatabase()); + auto loading_dependencies = getLoadingDependenciesFromCreateQuery(context->getGlobalContext(), qualified_name, query_ptr); + DatabaseCatalog::instance().addDependencies(qualified_name, ref_dependencies, loading_dependencies); +} + +void checkTableCanBeAddedWithNoCyclicDependencies(const ASTCreateQuery & create, const ASTPtr & query_ptr, const ContextPtr & context) +{ + QualifiedTableName qualified_name{create.getDatabase(), create.getTable()}; + auto ref_dependencies = getDependenciesFromCreateQuery(context->getGlobalContext(), qualified_name, query_ptr, context->getCurrentDatabase()); + auto loading_dependencies = getLoadingDependenciesFromCreateQuery(context->getGlobalContext(), qualified_name, query_ptr); + DatabaseCatalog::instance().checkTableCanBeAddedWithNoCyclicDependencies(qualified_name, ref_dependencies, loading_dependencies); +} + +bool isReplicated(const ASTStorage & storage) +{ + if (!storage.engine) + return false; + const auto & storage_name = storage.engine->name; + return storage_name.starts_with("Replicated") || storage_name.starts_with("Shared"); +} + +} + BlockIO InterpreterCreateQuery::createTable(ASTCreateQuery & create) { /// Temporary tables are created out of databases. @@ -1223,8 +1293,9 @@ BlockIO InterpreterCreateQuery::createTable(ASTCreateQuery & create) if (!create.temporary && !create.database) create.setDatabase(current_database); - if (create.to_table_id && create.to_table_id.database_name.empty()) - create.to_table_id.database_name = current_database; + + if (create.targets) + create.targets->setCurrentDatabase(current_database); if (create.select && create.isView()) { @@ -1258,12 +1329,9 @@ BlockIO InterpreterCreateQuery::createTable(ASTCreateQuery & create) TableProperties properties = getTablePropertiesAndNormalizeCreateQuery(create, mode); /// Check type compatible for materialized dest table and select columns - if (create.select && create.is_materialized_view && create.to_table_id && mode <= LoadingStrictnessLevel::CREATE) + if (create.is_materialized_view_with_external_target() && create.select && mode <= LoadingStrictnessLevel::CREATE) { - if (StoragePtr to_table = DatabaseCatalog::instance().tryGetTable( - {create.to_table_id.database_name, create.to_table_id.table_name, create.to_table_id.uuid}, - getContext() - )) + if (StoragePtr to_table = DatabaseCatalog::instance().tryGetTable(create.getTargetTableID(ViewTarget::To), getContext())) { Block input_block; @@ -1305,21 +1373,29 @@ BlockIO InterpreterCreateQuery::createTable(ASTCreateQuery & create) if (need_add_to_database) database = DatabaseCatalog::instance().tryGetDatabase(database_name); - if (database && database->getEngineName() == "Replicated" && create.select) + bool allow_heavy_create = getContext()->getSettingsRef().database_replicated_allow_heavy_create; + if (!allow_heavy_create && database && database->getEngineName() == "Replicated" && (create.select || create.is_populate)) { bool is_storage_replicated = false; - if (create.storage && create.storage->engine) + + if (create.storage && isReplicated(*create.storage)) + is_storage_replicated = true; + + if (create.targets) { - const auto & storage_name = create.storage->engine->name; - if (storage_name.starts_with("Replicated") || storage_name.starts_with("Shared")) - is_storage_replicated = true; + for (const auto & inner_table_engine : create.targets->getInnerEngines()) + { + if (isReplicated(*inner_table_engine)) + is_storage_replicated = true; + } } - const bool allow_create_select_for_replicated = create.isView() || create.is_create_empty || !is_storage_replicated; + const bool allow_create_select_for_replicated = (create.isView() && !create.is_populate) || create.is_create_empty || !is_storage_replicated; if (!allow_create_select_for_replicated) throw Exception( ErrorCodes::SUPPORT_IS_DISABLED, - "CREATE AS SELECT is not supported with Replicated databases. Use separate CREATE and INSERT queries"); + "CREATE AS SELECT and POPULATE is not supported with Replicated databases. Consider using separate CREATE and INSERT queries. " + "Alternatively, you can enable 'database_replicated_allow_heavy_create' setting to allow this operation, use with caution"); } if (database && database->shouldReplicateQuery(getContext(), query_ptr)) @@ -1354,11 +1430,7 @@ BlockIO InterpreterCreateQuery::createTable(ASTCreateQuery & create) return {}; /// If table has dependencies - add them to the graph - QualifiedTableName qualified_name{database_name, create.getTable()}; - auto ref_dependencies = getDependenciesFromCreateQuery(getContext()->getGlobalContext(), qualified_name, query_ptr); - auto loading_dependencies = getLoadingDependenciesFromCreateQuery(getContext()->getGlobalContext(), qualified_name, query_ptr); - DatabaseCatalog::instance().addDependencies(qualified_name, ref_dependencies, loading_dependencies); - + addTableDependencies(create, query_ptr, getContext()); return fillTableIfNeeded(create); } @@ -1510,6 +1582,9 @@ bool InterpreterCreateQuery::doCreateTable(ASTCreateQuery & create, throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot find UUID mapping for {}, it's a bug", create.uuid); } + /// Before actually creating the table, check if it will lead to cyclic dependencies. + checkTableCanBeAddedWithNoCyclicDependencies(create, query_ptr, getContext()); + StoragePtr res; /// NOTE: CREATE query may be rewritten by Storage creator or table function if (create.as_table_function) @@ -1576,13 +1651,13 @@ bool InterpreterCreateQuery::doCreateTable(ASTCreateQuery & create, } UInt64 table_num_limit = getContext()->getGlobalContext()->getServerSettings().max_table_num_to_throw; - if (table_num_limit > 0 && create.getDatabase() != DatabaseCatalog::SYSTEM_DATABASE) + if (table_num_limit > 0 && !internal) { UInt64 table_count = CurrentMetrics::get(CurrentMetrics::AttachedTable); if (table_count >= table_num_limit) throw Exception(ErrorCodes::TOO_MANY_TABLES, - "Too many tables in the Clickhouse. " - "The limit (setting 'max_table_num_to_throw') is set to {}, current number of tables is {}", + "Too many tables. " + "The limit (server configuration parameter `max_table_num_to_throw`) is set to {}, the current number of tables is {}", table_num_limit, table_count); } @@ -1621,6 +1696,9 @@ BlockIO InterpreterCreateQuery::doCreateOrReplaceTable(ASTCreateQuery & create, ContextMutablePtr create_context = Context::createCopy(current_context); create_context->setQueryContext(std::const_pointer_cast(current_context)); + /// Before actually creating/replacing the table, check if it will lead to cyclic dependencies. + checkTableCanBeAddedWithNoCyclicDependencies(create, query_ptr, create_context); + auto make_drop_context = [&]() -> ContextMutablePtr { ContextMutablePtr drop_context = Context::createCopy(current_context); @@ -1667,6 +1745,9 @@ BlockIO InterpreterCreateQuery::doCreateOrReplaceTable(ASTCreateQuery & create, assert(done); created = true; + /// If table has dependencies - add them to the graph + addTableDependencies(create, query_ptr, getContext()); + /// Try fill temporary table BlockIO fill_io = fillTableIfNeeded(create); executeTrivialBlockIO(fill_io, getContext()); @@ -1746,8 +1827,13 @@ BlockIO InterpreterCreateQuery::fillTableIfNeeded(const ASTCreateQuery & create) else insert->select = create.select->clone(); - return InterpreterInsertQuery(insert, getContext(), - getContext()->getSettingsRef().insert_allow_materialized_columns).execute(); + return InterpreterInsertQuery( + insert, + getContext(), + getContext()->getSettingsRef().insert_allow_materialized_columns, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false).execute(); } return {}; @@ -1760,7 +1846,7 @@ void InterpreterCreateQuery::prepareOnClusterQuery(ASTCreateQuery & create, Cont /// For CREATE query generate UUID on initiator, so it will be the same on all hosts. /// It will be ignored if database does not support UUIDs. - create.generateRandomUUID(); + create.generateRandomUUIDs(); /// For cross-replication cluster we cannot use UUID in replica path. String cluster_name_expanded = local_context->getMacros()->expand(cluster_name); @@ -1882,8 +1968,15 @@ AccessRightsElements InterpreterCreateQuery::getRequiredAccess() const } } - if (create.to_table_id) - required_access.emplace_back(AccessType::SELECT | AccessType::INSERT, create.to_table_id.database_name, create.to_table_id.table_name); + if (create.targets) + { + for (const auto & target : create.targets->targets) + { + const auto & target_id = target.table_id; + if (target_id) + required_access.emplace_back(AccessType::SELECT | AccessType::INSERT, target_id.database_name, target_id.table_name); + } + } if (create.storage && create.storage->engine) required_access.emplace_back(AccessType::TABLE_ENGINE, create.storage->engine->name); diff --git a/src/Interpreters/InterpreterCreateQuery.h b/src/Interpreters/InterpreterCreateQuery.h index 70ef29e6b07..3982ea2cabc 100644 --- a/src/Interpreters/InterpreterCreateQuery.h +++ b/src/Interpreters/InterpreterCreateQuery.h @@ -1,7 +1,6 @@ #pragma once #include -#include #include #include #include diff --git a/src/Interpreters/InterpreterDeleteQuery.cpp b/src/Interpreters/InterpreterDeleteQuery.cpp index 9cfb8e486cb..84fe0a9f4ae 100644 --- a/src/Interpreters/InterpreterDeleteQuery.cpp +++ b/src/Interpreters/InterpreterDeleteQuery.cpp @@ -2,6 +2,7 @@ #include #include +#include #include #include #include @@ -61,24 +62,7 @@ BlockIO InterpreterDeleteQuery::execute() auto table_lock = table->lockForShare(getContext()->getCurrentQueryId(), getContext()->getSettingsRef().lock_acquire_timeout); auto metadata_snapshot = table->getInMemoryMetadataPtr(); - if (table->supportsDelete()) - { - /// Convert to MutationCommand - MutationCommands mutation_commands; - MutationCommand mut_command; - - mut_command.type = MutationCommand::Type::DELETE; - mut_command.predicate = delete_query.predicate; - - mutation_commands.emplace_back(mut_command); - - table->checkMutationIsPossible(mutation_commands, getContext()->getSettingsRef()); - MutationsInterpreter::Settings settings(false); - MutationsInterpreter(table, metadata_snapshot, mutation_commands, getContext(), settings).validate(); - table->mutate(mutation_commands, getContext()); - return {}; - } - else if (table->supportsLightweightDelete()) + auto lightweightDelete = [&]() { if (!getContext()->getSettingsRef().enable_lightweight_delete) throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, @@ -105,17 +89,77 @@ BlockIO InterpreterDeleteQuery::execute() context->setSetting("mutations_sync", Field(context->getSettingsRef().lightweight_deletes_sync)); InterpreterAlterQuery alter_interpreter(alter_ast, context); return alter_interpreter.execute(); + }; + + if (table->supportsDelete()) + { + /// Convert to MutationCommand + MutationCommands mutation_commands; + MutationCommand mut_command; + + mut_command.type = MutationCommand::Type::DELETE; + mut_command.predicate = delete_query.predicate; + + mutation_commands.emplace_back(mut_command); + + table->checkMutationIsPossible(mutation_commands, getContext()->getSettingsRef()); + MutationsInterpreter::Settings settings(false); + MutationsInterpreter(table, metadata_snapshot, mutation_commands, getContext(), settings).validate(); + table->mutate(mutation_commands, getContext()); + return {}; + } + else if (table->supportsLightweightDelete()) + { + return lightweightDelete(); } else { - /// Currently just better exception for the case of a table with projection, - /// can act differently according to the setting. if (table->hasProjection()) { - throw Exception(ErrorCodes::NOT_IMPLEMENTED, - "DELETE query is not supported for table {} as it has projections. " - "User should drop all the projections manually before running the query", - table->getStorageID().getFullTableName()); + auto context = Context::createCopy(getContext()); + auto mode = context->getSettingsRef().lightweight_mutation_projection_mode; + if (mode == LightweightMutationProjectionMode::THROW) + { + throw Exception(ErrorCodes::NOT_IMPLEMENTED, + "DELETE query is not supported for table {} as it has projections. " + "User should drop all the projections manually before running the query", + table->getStorageID().getFullTableName()); + } + else if (mode == LightweightMutationProjectionMode::DROP) + { + std::vector all_projections = metadata_snapshot->projections.getAllRegisteredNames(); + + context->setSetting("mutations_sync", Field(context->getSettingsRef().lightweight_deletes_sync)); + + /// Drop projections first so that lightweight delete can be performed. + for (const auto & projection : all_projections) + { + String alter_query = + "ALTER TABLE " + table->getStorageID().getFullTableName() + + (delete_query.cluster.empty() ? "" : " ON CLUSTER " + backQuoteIfNeed(delete_query.cluster)) + + " DROP PROJECTION IF EXISTS " + projection; + + ParserAlterQuery parser; + ASTPtr alter_ast = parseQuery( + parser, + alter_query.data(), + alter_query.data() + alter_query.size(), + "ALTER query", + 0, + DBMS_DEFAULT_MAX_PARSER_DEPTH, + DBMS_DEFAULT_MAX_PARSER_BACKTRACKS); + + InterpreterAlterQuery alter_interpreter(alter_ast, context); + alter_interpreter.execute(); + } + } + else + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, + "Unrecognized lightweight_mutation_projection_mode, only throw and drop are allowed."); + } + + return lightweightDelete(); } throw Exception(ErrorCodes::BAD_ARGUMENTS, diff --git a/src/Interpreters/InterpreterDescribeQuery.cpp b/src/Interpreters/InterpreterDescribeQuery.cpp index 87b0eb89302..39fc85a5e23 100644 --- a/src/Interpreters/InterpreterDescribeQuery.cpp +++ b/src/Interpreters/InterpreterDescribeQuery.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/InterpreterDropIndexQuery.cpp b/src/Interpreters/InterpreterDropIndexQuery.cpp index f052aa201f1..8777532e4d0 100644 --- a/src/Interpreters/InterpreterDropIndexQuery.cpp +++ b/src/Interpreters/InterpreterDropIndexQuery.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include diff --git a/src/Interpreters/InterpreterDropNamedCollectionQuery.cpp b/src/Interpreters/InterpreterDropNamedCollectionQuery.cpp index 2edaef1b2f2..6233d21b439 100644 --- a/src/Interpreters/InterpreterDropNamedCollectionQuery.cpp +++ b/src/Interpreters/InterpreterDropNamedCollectionQuery.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -13,14 +14,16 @@ namespace DB BlockIO InterpreterDropNamedCollectionQuery::execute() { auto current_context = getContext(); - const auto & query = query_ptr->as(); + + const auto updated_query = removeOnClusterClauseIfNeeded(query_ptr, getContext()); + const auto & query = updated_query->as(); current_context->checkAccess(AccessType::DROP_NAMED_COLLECTION, query.collection_name); if (!query.cluster.empty()) { DDLQueryOnClusterParams params; - return executeDDLQueryOnCluster(query_ptr, current_context, params); + return executeDDLQueryOnCluster(updated_query, current_context, params); } NamedCollectionFactory::instance().removeFromSQL(query); diff --git a/src/Interpreters/InterpreterDropQuery.cpp b/src/Interpreters/InterpreterDropQuery.cpp index ba788a2e1f3..bad3e5277db 100644 --- a/src/Interpreters/InterpreterDropQuery.cpp +++ b/src/Interpreters/InterpreterDropQuery.cpp @@ -14,6 +14,7 @@ #include #include #include +#include #include #include "config.h" @@ -398,10 +399,9 @@ BlockIO InterpreterDropQuery::executeToDatabaseImpl(const ASTDropQuery & query, if (query.if_empty) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "DROP IF EMPTY is not implemented for databases"); - if (database->hasReplicationThread()) + if (!truncate && database->hasReplicationThread()) database->stopReplication(); - if (database->shouldBeEmptyOnDetach()) { /// Cancel restarting replicas in that database, wait for remaining RESTART queries to finish. diff --git a/src/Interpreters/InterpreterExplainQuery.cpp b/src/Interpreters/InterpreterExplainQuery.cpp index 3a06e1b2301..bedd9cb4a80 100644 --- a/src/Interpreters/InterpreterExplainQuery.cpp +++ b/src/Interpreters/InterpreterExplainQuery.cpp @@ -29,6 +29,7 @@ #include #include +#include #include #include @@ -43,6 +44,7 @@ namespace ErrorCodes extern const int UNKNOWN_SETTING; extern const int LOGICAL_ERROR; extern const int NOT_IMPLEMENTED; + extern const int BAD_ARGUMENTS; } namespace @@ -68,7 +70,7 @@ namespace static void visit(ASTSelectQuery & select, ASTPtr & node, Data & data) { /// we need to read statistic when `allow_statistics_optimize` is enabled. - bool only_analyze = !data.getContext()->getSettings().allow_statistics_optimize; + bool only_analyze = !data.getContext()->getSettingsRef().allow_statistics_optimize; InterpreterSelectQuery interpreter( node, data.getContext(), SelectQueryOptions(QueryProcessingStage::FetchColumns).analyze(only_analyze).modify()); @@ -170,6 +172,7 @@ struct QueryASTSettings struct QueryTreeSettings { bool run_passes = true; + bool dump_tree = true; bool dump_passes = false; bool dump_ast = false; Int64 passes = -1; @@ -179,6 +182,7 @@ struct QueryTreeSettings std::unordered_map> boolean_settings = { {"run_passes", run_passes}, + {"dump_tree", dump_tree}, {"dump_passes", dump_passes}, {"dump_ast", dump_ast} }; @@ -398,7 +402,11 @@ QueryPipeline InterpreterExplainQuery::executeImpl() throw Exception(ErrorCodes::INCORRECT_QUERY, "Only SELECT is supported for EXPLAIN QUERY TREE query"); auto settings = checkAndGetSettings(ast.getSettings()); + if (!settings.dump_tree && !settings.dump_ast) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Either 'dump_tree' or 'dump_ast' must be set for EXPLAIN QUERY TREE query"); + auto query_tree = buildQueryTree(ast.getExplainedQuery(), getContext()); + bool need_newline = false; if (settings.run_passes) { @@ -410,23 +418,26 @@ QueryPipeline InterpreterExplainQuery::executeImpl() if (settings.dump_passes) { query_tree_pass_manager.dump(buf, pass_index); - if (pass_index > 0) - buf << '\n'; + need_newline = true; } query_tree_pass_manager.run(query_tree, pass_index); + } + + if (settings.dump_tree) + { + if (need_newline) + buf << "\n\n"; query_tree->dumpTree(buf); - } - else - { - query_tree->dumpTree(buf); + need_newline = true; } if (settings.dump_ast) { - buf << '\n'; - buf << '\n'; + if (need_newline) + buf << "\n\n"; + query_tree->toAST()->format(IAST::FormatSettings(buf, false)); } @@ -524,7 +535,13 @@ QueryPipeline InterpreterExplainQuery::executeImpl() } else if (dynamic_cast(ast.getExplainedQuery().get())) { - InterpreterInsertQuery insert(ast.getExplainedQuery(), getContext()); + InterpreterInsertQuery insert( + ast.getExplainedQuery(), + getContext(), + /* allow_materialized */ false, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); auto io = insert.execute(); printPipeline(io.pipeline.getProcessors(), buf); } diff --git a/src/Interpreters/InterpreterFactory.cpp b/src/Interpreters/InterpreterFactory.cpp index 387d056ffe7..12b3b510098 100644 --- a/src/Interpreters/InterpreterFactory.cpp +++ b/src/Interpreters/InterpreterFactory.cpp @@ -60,6 +60,7 @@ #include #include #include +#include namespace ProfileEvents diff --git a/src/Interpreters/InterpreterInsertQuery.cpp b/src/Interpreters/InterpreterInsertQuery.cpp index f396db70d21..3906cef1995 100644 --- a/src/Interpreters/InterpreterInsertQuery.cpp +++ b/src/Interpreters/InterpreterInsertQuery.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -16,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -26,6 +28,7 @@ #include #include #include +#include #include #include #include @@ -35,9 +38,11 @@ #include #include #include +#include #include #include #include +#include "base/defines.h" namespace ProfileEvents @@ -394,28 +399,349 @@ Chain InterpreterInsertQuery::buildPreSinkChain( return out; } +std::pair, std::vector> InterpreterInsertQuery::buildPreAndSinkChains(size_t presink_streams, size_t sink_streams, StoragePtr table, const StorageMetadataPtr & metadata_snapshot, const Block & query_sample_block) +{ + chassert(presink_streams > 0); + chassert(sink_streams > 0); + + ThreadGroupPtr running_group; + if (current_thread) + running_group = current_thread->getThreadGroup(); + if (!running_group) + running_group = std::make_shared(getContext()); + + std::vector sink_chains; + std::vector presink_chains; + + for (size_t i = 0; i < sink_streams; ++i) + { + auto out = buildSink(table, metadata_snapshot, /* thread_status_holder= */ nullptr, + running_group, /* elapsed_counter_ms= */ nullptr); + + sink_chains.emplace_back(std::move(out)); + } + + for (size_t i = 0; i < presink_streams; ++i) + { + auto out = buildPreSinkChain(sink_chains[0].getInputHeader(), table, metadata_snapshot, query_sample_block); + presink_chains.emplace_back(std::move(out)); + } + + return {std::move(presink_chains), std::move(sink_chains)}; +} + + +QueryPipeline InterpreterInsertQuery::buildInsertSelectPipeline(ASTInsertQuery & query, StoragePtr table) +{ + const Settings & settings = getContext()->getSettingsRef(); + + auto metadata_snapshot = table->getInMemoryMetadataPtr(); + auto query_sample_block = getSampleBlock(query, table, metadata_snapshot, getContext(), no_destination, allow_materialized); + + bool is_trivial_insert_select = false; + + if (settings.optimize_trivial_insert_select) + { + const auto & select_query = query.select->as(); + const auto & selects = select_query.list_of_selects->children; + const auto & union_modes = select_query.list_of_modes; + + /// ASTSelectWithUnionQuery is not normalized now, so it may pass some queries which can be Trivial select queries + const auto mode_is_all = [](const auto & mode) { return mode == SelectUnionMode::UNION_ALL; }; + + is_trivial_insert_select = + std::all_of(union_modes.begin(), union_modes.end(), std::move(mode_is_all)) + && std::all_of(selects.begin(), selects.end(), isTrivialSelect); + } + + ContextPtr select_context = getContext(); + + if (is_trivial_insert_select) + { + /** When doing trivial INSERT INTO ... SELECT ... FROM table, + * don't need to process SELECT with more than max_insert_threads + * and it's reasonable to set block size for SELECT to the desired block size for INSERT + * to avoid unnecessary squashing. + */ + + Settings new_settings = select_context->getSettings(); + + new_settings.max_threads = std::max(1, settings.max_insert_threads); + + if (table->prefersLargeBlocks()) + { + if (settings.min_insert_block_size_rows) + new_settings.max_block_size = settings.min_insert_block_size_rows; + if (settings.min_insert_block_size_bytes) + new_settings.preferred_block_size_bytes = settings.min_insert_block_size_bytes; + } + + auto context_for_trivial_select = Context::createCopy(context); + context_for_trivial_select->setSettings(new_settings); + context_for_trivial_select->setInsertionTable(getContext()->getInsertionTable(), getContext()->getInsertionTableColumnNames()); + + select_context = context_for_trivial_select; + } + + QueryPipelineBuilder pipeline; + + { + auto select_query_options = SelectQueryOptions(QueryProcessingStage::Complete, 1); + + if (settings.allow_experimental_analyzer) + { + InterpreterSelectQueryAnalyzer interpreter_select_analyzer(query.select, select_context, select_query_options); + pipeline = interpreter_select_analyzer.buildQueryPipeline(); + } + else + { + InterpreterSelectWithUnionQuery interpreter_select(query.select, select_context, select_query_options); + pipeline = interpreter_select.buildQueryPipeline(); + } + } + + pipeline.dropTotalsAndExtremes(); + + /// Allow to insert Nullable into non-Nullable columns, NULL values will be added as defaults values. + if (getContext()->getSettingsRef().insert_null_as_default) + { + const auto & input_columns = pipeline.getHeader().getColumnsWithTypeAndName(); + const auto & query_columns = query_sample_block.getColumnsWithTypeAndName(); + const auto & output_columns = metadata_snapshot->getColumns(); + + if (input_columns.size() == query_columns.size()) + { + for (size_t col_idx = 0; col_idx < query_columns.size(); ++col_idx) + { + /// Change query sample block columns to Nullable to allow inserting nullable columns, where NULL values will be substituted with + /// default column values (in AddingDefaultsTransform), so all values will be cast correctly. + if (isNullableOrLowCardinalityNullable(input_columns[col_idx].type) + && !isNullableOrLowCardinalityNullable(query_columns[col_idx].type) + && !isVariant(query_columns[col_idx].type) + && !isDynamic(query_columns[col_idx].type) + && output_columns.has(query_columns[col_idx].name)) + { + query_sample_block.setColumn( + col_idx, + ColumnWithTypeAndName( + makeNullableOrLowCardinalityNullable(query_columns[col_idx].column), + makeNullableOrLowCardinalityNullable(query_columns[col_idx].type), + query_columns[col_idx].name)); + } + } + } + } + + auto actions_dag = ActionsDAG::makeConvertingActions( + pipeline.getHeader().getColumnsWithTypeAndName(), + query_sample_block.getColumnsWithTypeAndName(), + ActionsDAG::MatchColumnsMode::Position); + auto actions = std::make_shared(actions_dag, ExpressionActionsSettings::fromContext(getContext(), CompileExpressions::yes)); + + pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + { + return std::make_shared(in_header, actions); + }); + + /// We need to convert Sparse columns to full, because it's destination storage + /// may not support it or may have different settings for applying Sparse serialization. + pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + { + return std::make_shared(in_header); + }); + + pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + { + auto context_ptr = getContext(); + auto counting = std::make_shared(in_header, nullptr, context_ptr->getQuota()); + counting->setProcessListElement(context_ptr->getProcessListElement()); + counting->setProgressCallback(context_ptr->getProgressCallback()); + + return counting; + }); + + size_t num_select_threads = pipeline.getNumThreads(); + + pipeline.resize(1); + + if (shouldAddSquashingFroStorage(table)) + { + pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + { + return std::make_shared( + in_header, + table->prefersLargeBlocks() ? settings.min_insert_block_size_rows : settings.max_block_size, + table->prefersLargeBlocks() ? settings.min_insert_block_size_bytes : 0ULL); + }); + } + + pipeline.addSimpleTransform([&](const Block &in_header) -> ProcessorPtr + { + return std::make_shared(in_header); + }); + + if (!settings.insert_deduplication_token.value.empty()) + { + pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + { + return std::make_shared(settings.insert_deduplication_token.value, in_header); + }); + + pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + { + return std::make_shared(in_header); + }); + } + + /// Number of streams works like this: + /// * For the SELECT, use `max_threads`, or `max_insert_threads`, or whatever + /// InterpreterSelectQuery ends up with. + /// * Use `max_insert_threads` streams for various insert-preparation steps, e.g. + /// materializing and squashing (too slow to do in one thread). That's `presink_chains`. + /// * If the table supports parallel inserts, use max_insert_threads for writing to IStorage. + /// Otherwise ResizeProcessor them down to 1 stream. + + size_t presink_streams_size = std::max(settings.max_insert_threads, pipeline.getNumStreams()); + if (settings.max_insert_threads.changed) + presink_streams_size = std::max(1, settings.max_insert_threads); + + size_t sink_streams_size = table->supportsParallelInsert() ? std::max(1, settings.max_insert_threads) : 1; + + size_t views_involved = table->isView() || !DatabaseCatalog::instance().getDependentViews(table->getStorageID()).empty(); + if (!settings.parallel_view_processing && views_involved) + { + sink_streams_size = 1; + } + + auto [presink_chains, sink_chains] = buildPreAndSinkChains( + presink_streams_size, sink_streams_size, + table, metadata_snapshot, query_sample_block); + + pipeline.resize(presink_chains.size()); + + if (shouldAddSquashingFroStorage(table)) + { + pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + { + return std::make_shared( + in_header, + table->prefersLargeBlocks() ? settings.min_insert_block_size_rows : settings.max_block_size, + table->prefersLargeBlocks() ? settings.min_insert_block_size_bytes : 0ULL); + }); + } + + for (auto & chain : presink_chains) + pipeline.addResources(chain.detachResources()); + pipeline.addChains(std::move(presink_chains)); + + pipeline.resize(sink_streams_size); + + for (auto & chain : sink_chains) + pipeline.addResources(chain.detachResources()); + pipeline.addChains(std::move(sink_chains)); + + if (!settings.parallel_view_processing && views_involved) + { + /// Don't use more threads for INSERT than for SELECT to reduce memory consumption. + if (pipeline.getNumThreads() > num_select_threads) + pipeline.setMaxThreads(num_select_threads); + } + + pipeline.setSinks([&](const Block & cur_header, QueryPipelineBuilder::StreamType) -> ProcessorPtr + { + return std::make_shared(cur_header); + }); + + return QueryPipelineBuilder::getPipeline(std::move(pipeline)); +} + + +QueryPipeline InterpreterInsertQuery::buildInsertPipeline(ASTInsertQuery & query, StoragePtr table) +{ + const Settings & settings = getContext()->getSettingsRef(); + + auto metadata_snapshot = table->getInMemoryMetadataPtr(); + auto query_sample_block = getSampleBlock(query, table, metadata_snapshot, getContext(), no_destination, allow_materialized); + + Chain chain; + + { + auto [presink_chains, sink_chains] = buildPreAndSinkChains( + /* presink_streams */1, /* sink_streams */1, + table, metadata_snapshot, query_sample_block); + + chain = std::move(presink_chains.front()); + chain.appendChain(std::move(sink_chains.front())); + } + + if (!settings.insert_deduplication_token.value.empty()) + { + chain.addSource(std::make_shared(chain.getInputHeader())); + chain.addSource(std::make_shared(settings.insert_deduplication_token.value, chain.getInputHeader())); + } + + chain.addSource(std::make_shared(chain.getInputHeader())); + + if (shouldAddSquashingFroStorage(table)) + { + bool table_prefers_large_blocks = table->prefersLargeBlocks(); + + auto squashing = std::make_shared( + chain.getInputHeader(), + table_prefers_large_blocks ? settings.min_insert_block_size_rows : settings.max_block_size, + table_prefers_large_blocks ? settings.min_insert_block_size_bytes : 0ULL); + + chain.addSource(std::move(squashing)); + + auto balancing = std::make_shared( + chain.getInputHeader(), + table_prefers_large_blocks ? settings.min_insert_block_size_rows : settings.max_block_size, + table_prefers_large_blocks ? settings.min_insert_block_size_bytes : 0ULL); + + chain.addSource(std::move(balancing)); + } + + auto context_ptr = getContext(); + auto counting = std::make_shared(chain.getInputHeader(), nullptr, context_ptr->getQuota()); + counting->setProcessListElement(context_ptr->getProcessListElement()); + counting->setProgressCallback(context_ptr->getProgressCallback()); + chain.addSource(std::move(counting)); + + QueryPipeline pipeline = QueryPipeline(std::move(chain)); + + pipeline.setNumThreads(std::min(pipeline.getNumThreads(), settings.max_threads)); + pipeline.setConcurrencyControl(settings.use_concurrency_control); + + if (query.hasInlinedData() && !async_insert) + { + /// can execute without additional data + auto format = getInputFormatFromASTInsertQuery(query_ptr, true, query_sample_block, getContext(), nullptr); + for (auto && buffer : owned_buffers) + format->addBuffer(std::move(buffer)); + + auto pipe = getSourceFromInputFormat(query_ptr, std::move(format), getContext(), nullptr); + pipeline.complete(std::move(pipe)); + } + + return pipeline; +} + + BlockIO InterpreterInsertQuery::execute() { const Settings & settings = getContext()->getSettingsRef(); auto & query = query_ptr->as(); - QueryPipelineBuilder pipeline; - std::optional distributed_pipeline; - QueryPlanResourceHolder resources; StoragePtr table = getTable(query); checkStorageSupportsTransactionsIfNeeded(table, getContext()); - StoragePtr inner_table; - if (const auto * mv = dynamic_cast(table.get())) - inner_table = mv->getTargetTable(); - if (query.partition_by && !table->supportsPartitionBy()) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "PARTITION BY clause is not supported by storage"); auto table_lock = table->lockForShare(getContext()->getInitialQueryId(), settings.lock_acquire_timeout); - auto metadata_snapshot = table->getInMemoryMetadataPtr(); + auto metadata_snapshot = table->getInMemoryMetadataPtr(); auto query_sample_block = getSampleBlock(query, table, metadata_snapshot, getContext(), no_destination, allow_materialized); /// For table functions we check access while executing @@ -423,320 +749,45 @@ BlockIO InterpreterInsertQuery::execute() if (!query.table_function) getContext()->checkAccess(AccessType::INSERT, query.table_id, query_sample_block.getNames()); - if (query.select && settings.parallel_distributed_insert_select) - // Distributed INSERT SELECT - distributed_pipeline = table->distributedWrite(query, getContext()); - - std::vector presink_chains; - std::vector sink_chains; - if (!distributed_pipeline) + if (!allow_materialized) { - /// Number of streams works like this: - /// * For the SELECT, use `max_threads`, or `max_insert_threads`, or whatever - /// InterpreterSelectQuery ends up with. - /// * Use `max_insert_threads` streams for various insert-preparation steps, e.g. - /// materializing and squashing (too slow to do in one thread). That's `presink_chains`. - /// * If the table supports parallel inserts, use the same streams for writing to IStorage. - /// Otherwise ResizeProcessor them down to 1 stream. - /// * If it's not an INSERT SELECT, forget all that and use one stream. - size_t pre_streams_size = 1; - size_t sink_streams_size = 1; - - if (query.select) - { - bool is_trivial_insert_select = false; - - if (settings.optimize_trivial_insert_select) - { - const auto & select_query = query.select->as(); - const auto & selects = select_query.list_of_selects->children; - const auto & union_modes = select_query.list_of_modes; - - /// ASTSelectWithUnionQuery is not normalized now, so it may pass some queries which can be Trivial select queries - const auto mode_is_all = [](const auto & mode) { return mode == SelectUnionMode::UNION_ALL; }; - - is_trivial_insert_select = - std::all_of(union_modes.begin(), union_modes.end(), std::move(mode_is_all)) - && std::all_of(selects.begin(), selects.end(), isTrivialSelect); - } - - if (is_trivial_insert_select) - { - /** When doing trivial INSERT INTO ... SELECT ... FROM table, - * don't need to process SELECT with more than max_insert_threads - * and it's reasonable to set block size for SELECT to the desired block size for INSERT - * to avoid unnecessary squashing. - */ - - Settings new_settings = getContext()->getSettings(); - - new_settings.max_threads = std::max(1, settings.max_insert_threads); - - if (table->prefersLargeBlocks()) - { - if (settings.min_insert_block_size_rows) - new_settings.max_block_size = settings.min_insert_block_size_rows; - if (settings.min_insert_block_size_bytes) - new_settings.preferred_block_size_bytes = settings.min_insert_block_size_bytes; - } - - auto new_context = Context::createCopy(context); - new_context->setSettings(new_settings); - new_context->setInsertionTable(getContext()->getInsertionTable(), getContext()->getInsertionTableColumnNames()); - - auto select_query_options = SelectQueryOptions(QueryProcessingStage::Complete, 1); - - if (settings.allow_experimental_analyzer) - { - InterpreterSelectQueryAnalyzer interpreter_select_analyzer(query.select, new_context, select_query_options); - pipeline = interpreter_select_analyzer.buildQueryPipeline(); - } - else - { - InterpreterSelectWithUnionQuery interpreter_select(query.select, new_context, select_query_options); - pipeline = interpreter_select.buildQueryPipeline(); - } - } - else - { - /// Passing 1 as subquery_depth will disable limiting size of intermediate result. - auto select_query_options = SelectQueryOptions(QueryProcessingStage::Complete, 1); - - if (settings.allow_experimental_analyzer) - { - InterpreterSelectQueryAnalyzer interpreter_select_analyzer(query.select, getContext(), select_query_options); - pipeline = interpreter_select_analyzer.buildQueryPipeline(); - } - else - { - InterpreterSelectWithUnionQuery interpreter_select(query.select, getContext(), select_query_options); - pipeline = interpreter_select.buildQueryPipeline(); - } - } - - pipeline.dropTotalsAndExtremes(); - - if (settings.max_insert_threads > 1) - { - auto table_id = table->getStorageID(); - auto views = DatabaseCatalog::instance().getDependentViews(table_id); - - /// It breaks some views-related tests and we have dedicated `parallel_view_processing` for views, so let's just skip them. - /// Also it doesn't make sense to reshuffle data if storage doesn't support parallel inserts. - const bool resize_to_max_insert_threads = !table->isView() && views.empty() && table->supportsParallelInsert(); - pre_streams_size = resize_to_max_insert_threads ? settings.max_insert_threads - : std::min(settings.max_insert_threads, pipeline.getNumStreams()); - - /// Deduplication when passing insert_deduplication_token breaks if using more than one thread - if (!settings.insert_deduplication_token.toString().empty()) - { - LOG_DEBUG( - getLogger("InsertQuery"), - "Insert-select query using insert_deduplication_token, setting streams to 1 to avoid deduplication issues"); - pre_streams_size = 1; - } - - if (table->supportsParallelInsert()) - sink_streams_size = pre_streams_size; - } - - pipeline.resize(pre_streams_size); - - /// Allow to insert Nullable into non-Nullable columns, NULL values will be added as defaults values. - if (getContext()->getSettingsRef().insert_null_as_default) - { - const auto & input_columns = pipeline.getHeader().getColumnsWithTypeAndName(); - const auto & query_columns = query_sample_block.getColumnsWithTypeAndName(); - const auto & output_columns = metadata_snapshot->getColumns(); - - if (input_columns.size() == query_columns.size()) - { - for (size_t col_idx = 0; col_idx < query_columns.size(); ++col_idx) - { - /// Change query sample block columns to Nullable to allow inserting nullable columns, where NULL values will be substituted with - /// default column values (in AddingDefaultsTransform), so all values will be cast correctly. - if (isNullableOrLowCardinalityNullable(input_columns[col_idx].type) - && !isNullableOrLowCardinalityNullable(query_columns[col_idx].type) - && !isVariant(query_columns[col_idx].type) - && !isDynamic(query_columns[col_idx].type) - && output_columns.has(query_columns[col_idx].name)) - query_sample_block.setColumn(col_idx, ColumnWithTypeAndName(makeNullableOrLowCardinalityNullable(query_columns[col_idx].column), makeNullableOrLowCardinalityNullable(query_columns[col_idx].type), query_columns[col_idx].name)); - } - } - } - } - - ThreadGroupPtr running_group; - if (current_thread) - running_group = current_thread->getThreadGroup(); - if (!running_group) - running_group = std::make_shared(getContext()); - for (size_t i = 0; i < sink_streams_size; ++i) - { - auto out = buildSink(table, metadata_snapshot, /* thread_status_holder= */ nullptr, - running_group, /* elapsed_counter_ms= */ nullptr); - sink_chains.emplace_back(std::move(out)); - } - for (size_t i = 0; i < pre_streams_size; ++i) - { - auto out = buildPreSinkChain(sink_chains[0].getInputHeader(), table, metadata_snapshot, query_sample_block); - presink_chains.emplace_back(std::move(out)); - } + for (const auto & column : metadata_snapshot->getColumns()) + if (column.default_desc.kind == ColumnDefaultKind::Materialized && query_sample_block.has(column.name)) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Cannot insert column {}, because it is MATERIALIZED column.", column.name); } BlockIO res; - /// What type of query: INSERT or INSERT SELECT or INSERT WATCH? - if (distributed_pipeline) + if (query.select) { - res.pipeline = std::move(*distributed_pipeline); - } - else if (query.select) - { - const auto & header = presink_chains.at(0).getInputHeader(); - auto actions_dag = ActionsDAG::makeConvertingActions( - pipeline.getHeader().getColumnsWithTypeAndName(), - header.getColumnsWithTypeAndName(), - ActionsDAG::MatchColumnsMode::Position); - auto actions = std::make_shared(actions_dag, ExpressionActionsSettings::fromContext(getContext(), CompileExpressions::yes)); - - pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + if (settings.parallel_distributed_insert_select) { - return std::make_shared(in_header, actions); - }); - - /// We need to convert Sparse columns to full, because it's destination storage - /// may not support it or may have different settings for applying Sparse serialization. - pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr - { - return std::make_shared(in_header); - }); - - pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr - { - auto context_ptr = getContext(); - auto counting = std::make_shared(in_header, nullptr, context_ptr->getQuota()); - counting->setProcessListElement(context_ptr->getProcessListElement()); - counting->setProgressCallback(context_ptr->getProgressCallback()); - - return counting; - }); - - if (shouldAddSquashingFroStorage(table)) - { - bool table_prefers_large_blocks = table->prefersLargeBlocks(); - - size_t threads = presink_chains.size(); - - pipeline.resize(1); - - pipeline.addTransform(std::make_shared( - header, - table_prefers_large_blocks ? settings.min_insert_block_size_rows : settings.max_block_size, - table_prefers_large_blocks ? settings.min_insert_block_size_bytes : 0ULL)); - - pipeline.resize(threads); - - pipeline.addSimpleTransform([&](const Block & in_header) -> ProcessorPtr + auto distributed = table->distributedWrite(query, getContext()); + if (distributed) { - return std::make_shared( - in_header, - table_prefers_large_blocks ? settings.min_insert_block_size_rows : settings.max_block_size, - table_prefers_large_blocks ? settings.min_insert_block_size_bytes : 0ULL); - }); + res.pipeline = std::move(*distributed); + } + else + { + res.pipeline = buildInsertSelectPipeline(query, table); + } } - - size_t num_select_threads = pipeline.getNumThreads(); - - for (auto & chain : presink_chains) - resources = chain.detachResources(); - for (auto & chain : sink_chains) - resources = chain.detachResources(); - - pipeline.addChains(std::move(presink_chains)); - pipeline.resize(sink_chains.size()); - pipeline.addChains(std::move(sink_chains)); - - if (!settings.parallel_view_processing) + else { - /// Don't use more threads for INSERT than for SELECT to reduce memory consumption. - if (pipeline.getNumThreads() > num_select_threads) - pipeline.setMaxThreads(num_select_threads); + res.pipeline = buildInsertSelectPipeline(query, table); } - else if (pipeline.getNumThreads() < settings.max_threads) - { - /// It is possible for query to have max_threads=1, due to optimize_trivial_insert_select, - /// however in case of parallel_view_processing and multiple views, views can still be processed in parallel. - /// - /// Note, number of threads will be limited by buildPushingToViewsChain() to max_threads. - pipeline.setMaxThreads(settings.max_threads); - } - - pipeline.setSinks([&](const Block & cur_header, QueryPipelineBuilder::StreamType) -> ProcessorPtr - { - return std::make_shared(cur_header); - }); - - if (!allow_materialized) - { - for (const auto & column : metadata_snapshot->getColumns()) - if (column.default_desc.kind == ColumnDefaultKind::Materialized && header.has(column.name)) - throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Cannot insert column {}, because it is MATERIALIZED column.", column.name); - } - - res.pipeline = QueryPipelineBuilder::getPipeline(std::move(pipeline)); } else { - auto & chain = presink_chains.at(0); - chain.appendChain(std::move(sink_chains.at(0))); - - if (shouldAddSquashingFroStorage(table)) - { - bool table_prefers_large_blocks = table->prefersLargeBlocks(); - - auto squashing = std::make_shared( - chain.getInputHeader(), - table_prefers_large_blocks ? settings.min_insert_block_size_rows : settings.max_block_size, - table_prefers_large_blocks ? settings.min_insert_block_size_bytes : 0ULL); - - chain.addSource(std::move(squashing)); - - auto balancing = std::make_shared( - chain.getInputHeader(), - table_prefers_large_blocks ? settings.min_insert_block_size_rows : settings.max_block_size, - table_prefers_large_blocks ? settings.min_insert_block_size_bytes : 0ULL); - - chain.addSource(std::move(balancing)); - } - - auto context_ptr = getContext(); - auto counting = std::make_shared(chain.getInputHeader(), nullptr, context_ptr->getQuota()); - counting->setProcessListElement(context_ptr->getProcessListElement()); - counting->setProgressCallback(context_ptr->getProgressCallback()); - chain.addSource(std::move(counting)); - - res.pipeline = QueryPipeline(std::move(presink_chains[0])); - res.pipeline.setNumThreads(std::min(res.pipeline.getNumThreads(), settings.max_threads)); - res.pipeline.setConcurrencyControl(settings.use_concurrency_control); - - if (query.hasInlinedData() && !async_insert) - { - /// can execute without additional data - auto format = getInputFormatFromASTInsertQuery(query_ptr, true, query_sample_block, getContext(), nullptr); - for (auto && buffer : owned_buffers) - format->addBuffer(std::move(buffer)); - - auto pipe = getSourceFromInputFormat(query_ptr, std::move(format), getContext(), nullptr); - res.pipeline.complete(std::move(pipe)); - } + res.pipeline = buildInsertPipeline(query, table); } - res.pipeline.addResources(std::move(resources)); - res.pipeline.addStorageHolder(table); - if (inner_table) - res.pipeline.addStorageHolder(inner_table); + + if (const auto * mv = dynamic_cast(table.get())) + res.pipeline.addStorageHolder(mv->getTargetTable()); + + LOG_TEST(getLogger("InterpreterInsertQuery"), "Pipeline could use up to {} thread", res.pipeline.getNumThreads()); return res; } @@ -757,17 +808,27 @@ void InterpreterInsertQuery::extendQueryLogElemImpl(QueryLogElement & elem, Cont } } + void InterpreterInsertQuery::extendQueryLogElemImpl(QueryLogElement & elem, const ASTPtr &, ContextPtr context_) const { extendQueryLogElemImpl(elem, context_); } + void registerInterpreterInsertQuery(InterpreterFactory & factory) { auto create_fn = [] (const InterpreterFactory::Arguments & args) { - return std::make_unique(args.query, args.context, args.allow_materialized); + return std::make_unique( + args.query, + args.context, + args.allow_materialized, + /* no_squash */false, + /* no_destination */false, + /* async_insert */false); }; factory.registerInterpreter("InterpreterInsertQuery", create_fn); } + + } diff --git a/src/Interpreters/InterpreterInsertQuery.h b/src/Interpreters/InterpreterInsertQuery.h index bf73fb2a319..894c7c42144 100644 --- a/src/Interpreters/InterpreterInsertQuery.h +++ b/src/Interpreters/InterpreterInsertQuery.h @@ -23,10 +23,10 @@ public: InterpreterInsertQuery( const ASTPtr & query_ptr_, ContextPtr context_, - bool allow_materialized_ = false, - bool no_squash_ = false, - bool no_destination_ = false, - bool async_insert_ = false); + bool allow_materialized_, + bool no_squash_, + bool no_destination, + bool async_insert_); /** Prepare a request for execution. Return block streams * - the stream into which you can write data to execute the query, if INSERT; @@ -73,12 +73,17 @@ private: ASTPtr query_ptr; const bool allow_materialized; - const bool no_squash; - const bool no_destination; + bool no_squash = false; + bool no_destination = false; const bool async_insert; std::vector> owned_buffers; + std::pair, std::vector> buildPreAndSinkChains(size_t presink_streams, size_t sink_streams, StoragePtr table, const StorageMetadataPtr & metadata_snapshot, const Block & query_sample_block); + + QueryPipeline buildInsertSelectPipeline(ASTInsertQuery & query, StoragePtr table); + QueryPipeline buildInsertPipeline(ASTInsertQuery & query, StoragePtr table); + Chain buildSink( const StoragePtr & table, const StorageMetadataPtr & metadata_snapshot, diff --git a/src/Interpreters/InterpreterKillQueryQuery.cpp b/src/Interpreters/InterpreterKillQueryQuery.cpp index 6d6b1085ffb..7eb487ba7b3 100644 --- a/src/Interpreters/InterpreterKillQueryQuery.cpp +++ b/src/Interpreters/InterpreterKillQueryQuery.cpp @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/InterpreterRenameQuery.cpp b/src/Interpreters/InterpreterRenameQuery.cpp index eeb762b4d7e..c6e52590f89 100644 --- a/src/Interpreters/InterpreterRenameQuery.cpp +++ b/src/Interpreters/InterpreterRenameQuery.cpp @@ -9,6 +9,7 @@ #include #include #include +#include #include @@ -127,14 +128,23 @@ BlockIO InterpreterRenameQuery::executeToTables(const ASTRenameQuery & rename, c { StorageID from_table_id{elem.from_database_name, elem.from_table_name}; StorageID to_table_id{elem.to_database_name, elem.to_table_name}; - std::vector ref_dependencies; - std::vector loading_dependencies; + std::vector from_ref_dependencies; + std::vector from_loading_dependencies; + std::vector to_ref_dependencies; + std::vector to_loading_dependencies; - if (!exchange_tables) + if (exchange_tables) { + DatabaseCatalog::instance().checkTablesCanBeExchangedWithNoCyclicDependencies(from_table_id, to_table_id); + std::tie(from_ref_dependencies, from_loading_dependencies) = database_catalog.removeDependencies(from_table_id, false, false); + std::tie(to_ref_dependencies, to_loading_dependencies) = database_catalog.removeDependencies(to_table_id, false, false); + } + else + { + DatabaseCatalog::instance().checkTableCanBeRenamedWithNoCyclicDependencies(from_table_id, to_table_id); bool check_ref_deps = getContext()->getSettingsRef().check_referential_table_dependencies; bool check_loading_deps = !check_ref_deps && getContext()->getSettingsRef().check_table_dependencies; - std::tie(ref_dependencies, loading_dependencies) = database_catalog.removeDependencies(from_table_id, check_ref_deps, check_loading_deps); + std::tie(from_ref_dependencies, from_loading_dependencies) = database_catalog.removeDependencies(from_table_id, check_ref_deps, check_loading_deps); } try @@ -147,12 +157,17 @@ BlockIO InterpreterRenameQuery::executeToTables(const ASTRenameQuery & rename, c exchange_tables, rename.dictionary); - DatabaseCatalog::instance().addDependencies(to_table_id, ref_dependencies, loading_dependencies); + DatabaseCatalog::instance().addDependencies(to_table_id, from_ref_dependencies, from_loading_dependencies); + if (!to_ref_dependencies.empty() || !to_loading_dependencies.empty()) + DatabaseCatalog::instance().addDependencies(from_table_id, to_ref_dependencies, to_loading_dependencies); + } catch (...) { /// Restore dependencies if RENAME fails - DatabaseCatalog::instance().addDependencies(from_table_id, ref_dependencies, loading_dependencies); + DatabaseCatalog::instance().addDependencies(from_table_id, from_ref_dependencies, from_loading_dependencies); + if (!to_ref_dependencies.empty() || !to_loading_dependencies.empty()) + DatabaseCatalog::instance().addDependencies(to_table_id, to_ref_dependencies, to_loading_dependencies); throw; } } diff --git a/src/Interpreters/InterpreterSelectIntersectExceptQuery.cpp b/src/Interpreters/InterpreterSelectIntersectExceptQuery.cpp index 6eac2db20c9..b0da4caa3ac 100644 --- a/src/Interpreters/InterpreterSelectIntersectExceptQuery.cpp +++ b/src/Interpreters/InterpreterSelectIntersectExceptQuery.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp index c39c57d2169..0f24888cb79 100644 --- a/src/Interpreters/InterpreterSelectQuery.cpp +++ b/src/Interpreters/InterpreterSelectQuery.cpp @@ -75,7 +75,6 @@ #include #include -#include #include #include #include @@ -84,17 +83,20 @@ #include #include #include +#include +#include #include +#include #include #include #include #include #include +#include +#include #include #include #include -#include -#include namespace ProfileEvents @@ -211,11 +213,11 @@ InterpreterSelectQuery::InterpreterSelectQuery( {} InterpreterSelectQuery::InterpreterSelectQuery( - const ASTPtr & query_ptr_, - const ContextPtr & context_, - Pipe input_pipe_, - const SelectQueryOptions & options_) - : InterpreterSelectQuery(query_ptr_, context_, std::move(input_pipe_), nullptr, options_.copy().noSubquery()) + const ASTPtr & query_ptr_, + const ContextPtr & context_, + Pipe input_pipe_, + const SelectQueryOptions & options_) + : InterpreterSelectQuery(query_ptr_, context_, std::move(input_pipe_), nullptr, options_.copy().noSubquery()) {} InterpreterSelectQuery::InterpreterSelectQuery( @@ -224,18 +226,15 @@ InterpreterSelectQuery::InterpreterSelectQuery( const StoragePtr & storage_, const StorageMetadataPtr & metadata_snapshot_, const SelectQueryOptions & options_) - : InterpreterSelectQuery( - query_ptr_, context_, std::nullopt, storage_, options_.copy().noSubquery(), {}, metadata_snapshot_) -{ -} + : InterpreterSelectQuery(query_ptr_, context_, std::nullopt, storage_, options_.copy().noSubquery(), {}, metadata_snapshot_) +{} InterpreterSelectQuery::InterpreterSelectQuery( const ASTPtr & query_ptr_, const ContextPtr & context_, const SelectQueryOptions & options_, PreparedSetsPtr prepared_sets_) - : InterpreterSelectQuery( - query_ptr_, context_, std::nullopt, nullptr, options_, {}, {}, prepared_sets_) + : InterpreterSelectQuery(query_ptr_, context_, std::nullopt, nullptr, options_, {}, {}, prepared_sets_) {} InterpreterSelectQuery::~InterpreterSelectQuery() = default; @@ -565,7 +564,7 @@ InterpreterSelectQuery::InterpreterSelectQuery( settings.additional_table_filters, joined_tables.tablesWithColumns().front().table, *context); ASTPtr parallel_replicas_custom_filter_ast = nullptr; - if (storage && context->getParallelReplicasMode() == Context::ParallelReplicasMode::CUSTOM_KEY && !joined_tables.tablesWithColumns().empty()) + if (storage && context->canUseParallelReplicasCustomKey() && !joined_tables.tablesWithColumns().empty()) { if (settings.parallel_replicas_count > 1) { @@ -586,16 +585,28 @@ InterpreterSelectQuery::InterpreterSelectQuery( else if (settings.parallel_replica_offset > 0) { throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "Parallel replicas processing with custom_key has been requested " - "(setting 'max_parallel_replicas') but the table does not have custom_key defined for it " - "or it's invalid (settings `parallel_replicas_custom_key`)"); + ErrorCodes::BAD_ARGUMENTS, + "Parallel replicas processing with custom_key has been requested " + "(setting 'max_parallel_replicas') but the table does not have custom_key defined for it " + "or it's invalid (settings `parallel_replicas_custom_key`)"); } } + /// We disable prefer_localhost_replica because if one of the replicas is local it will create a single local plan + /// instead of executing the query with multiple replicas + /// We can enable this setting again for custom key parallel replicas when we can generate a plan that will use both a + /// local plan and remote replicas else if (auto * distributed = dynamic_cast(storage.get()); - distributed && context->canUseParallelReplicasCustomKey(*distributed->getCluster())) + distributed && context->canUseParallelReplicasCustomKeyForCluster(*distributed->getCluster())) { context->setSetting("distributed_group_by_no_merge", 2); + context->setSetting("prefer_localhost_replica", Field(0)); + } + else if ( + storage->isMergeTree() && (storage->supportsReplication() || settings.parallel_replicas_for_non_replicated_merge_tree) + && context->getClientInfo().distributed_depth == 0 + && context->canUseParallelReplicasCustomKeyForCluster(*context->getClusterForParallelReplicas())) + { + context->setSetting("prefer_localhost_replica", Field(0)); } } @@ -910,7 +921,7 @@ bool InterpreterSelectQuery::adjustParallelReplicasAfterAnalysis() UInt64 max_rows = maxBlockSizeByLimit(); if (settings.max_rows_to_read) max_rows = max_rows ? std::min(max_rows, settings.max_rows_to_read.value) : settings.max_rows_to_read; - query_info_copy.limit = max_rows; + query_info_copy.trivial_limit = max_rows; /// Apply filters to prewhere and add them to the query_info so we can filter out parts efficiently during row estimation applyFiltersToPrewhereInAnalysis(analysis_copy); @@ -1711,7 +1722,10 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, std::optional

getCurrentDataStream().header, join_clause.key_names_right); - if (settings.max_rows_in_set_to_optimize_join > 0 && kind_allows_filtering && has_non_const_keys) + if (settings.max_rows_in_set_to_optimize_join > 0 && join_type_allows_filtering && has_non_const_keys) { auto * left_set = add_create_set(query_plan, join_clause.key_names_left, JoinTableSide::Left); auto * right_set = add_create_set(*joined_plan, join_clause.key_names_right, JoinTableSide::Right); @@ -2445,13 +2459,13 @@ void InterpreterSelectQuery::executeFetchColumns(QueryProcessingStage::Enum proc if (local_limits.local_limits.size_limits.max_rows != 0) { if (max_block_limited < local_limits.local_limits.size_limits.max_rows) - query_info.limit = max_block_limited; + query_info.trivial_limit = max_block_limited; else if (local_limits.local_limits.size_limits.max_rows < std::numeric_limits::max()) /// Ask to read just enough rows to make the max_rows limit effective (so it has a chance to be triggered). - query_info.limit = 1 + local_limits.local_limits.size_limits.max_rows; + query_info.trivial_limit = 1 + local_limits.local_limits.size_limits.max_rows; } else { - query_info.limit = max_block_limited; + query_info.trivial_limit = max_block_limited; } } @@ -2603,10 +2617,10 @@ static Aggregator::Params getAggregatorParams( size_t group_by_two_level_threshold, size_t group_by_two_level_threshold_bytes) { - const auto stats_collecting_params = Aggregator::Params::StatsCollectingParams( - query_ptr, + const auto stats_collecting_params = StatsCollectingParams( + calculateCacheKey(query_ptr), settings.collect_hash_table_stats_during_aggregation, - settings.max_entries_for_hash_table_stats, + context.getServerSettings().max_entries_for_hash_table_stats, settings.max_size_to_preallocate_for_aggregation); return Aggregator::Params diff --git a/src/Interpreters/InterpreterSelectQuery.h b/src/Interpreters/InterpreterSelectQuery.h index d4ed19d45ea..ac1230a6eba 100644 --- a/src/Interpreters/InterpreterSelectQuery.h +++ b/src/Interpreters/InterpreterSelectQuery.h @@ -26,7 +26,6 @@ class Logger; namespace DB { -class SubqueryForSet; class InterpreterSelectWithUnionQuery; class Context; class QueryPlan; diff --git a/src/Interpreters/InterpreterSelectWithUnionQuery.cpp b/src/Interpreters/InterpreterSelectWithUnionQuery.cpp index cc1d7dd6531..e953a07b41d 100644 --- a/src/Interpreters/InterpreterSelectWithUnionQuery.cpp +++ b/src/Interpreters/InterpreterSelectWithUnionQuery.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include diff --git a/src/Interpreters/InterpreterShowColumnsQuery.cpp b/src/Interpreters/InterpreterShowColumnsQuery.cpp index f32ebceaa63..d8fff4e6026 100644 --- a/src/Interpreters/InterpreterShowColumnsQuery.cpp +++ b/src/Interpreters/InterpreterShowColumnsQuery.cpp @@ -3,6 +3,7 @@ #include #include +#include #include #include #include diff --git a/src/Interpreters/InterpreterShowCreateQuery.cpp b/src/Interpreters/InterpreterShowCreateQuery.cpp index 0fca7b64d5a..16add79d226 100644 --- a/src/Interpreters/InterpreterShowCreateQuery.cpp +++ b/src/Interpreters/InterpreterShowCreateQuery.cpp @@ -94,7 +94,8 @@ QueryPipeline InterpreterShowCreateQuery::executeImpl() { auto & create = create_query->as(); create.uuid = UUIDHelpers::Nil; - create.to_inner_uuid = UUIDHelpers::Nil; + if (create.targets) + create.targets->resetInnerUUIDs(); } MutableColumnPtr column = ColumnString::create(); diff --git a/src/Interpreters/InterpreterShowIndexesQuery.cpp b/src/Interpreters/InterpreterShowIndexesQuery.cpp index e8005ead91e..31f0404e123 100644 --- a/src/Interpreters/InterpreterShowIndexesQuery.cpp +++ b/src/Interpreters/InterpreterShowIndexesQuery.cpp @@ -33,12 +33,33 @@ String InterpreterShowIndexesQuery::getRewrittenQuery() String rewritten_query = fmt::format(R"( SELECT * FROM ( - (SELECT + (WITH + t1 AS ( + SELECT + name, + arrayJoin(splitByString(', ', primary_key)) AS pk_col + FROM + system.tables + WHERE + database = '{0}' + AND name = '{1}' + ), + t2 AS ( + SELECT + name, + pk_col, + row_number() OVER (ORDER BY 1) AS row_num + FROM + t1 + ) + SELECT name AS table, 1 AS non_unique, 'PRIMARY' AS key_name, - row_number() over (order by column_name) AS seq_in_index, - arrayJoin(splitByString(', ', primary_key)) AS column_name, + -- row_number() over (order by database) AS seq_in_index, + row_num AS seq_in_index, + -- arrayJoin(splitByString(', ', primary_key)) AS column_name, + pk_col, 'A' AS collation, 0 AS cardinality, NULL AS sub_part, @@ -49,10 +70,9 @@ FROM ( '' AS index_comment, 'YES' AS visible, '' AS expression - FROM system.tables - WHERE - database = '{0}' - AND name = '{1}') + FROM + t2 + ) UNION ALL ( SELECT table AS table, @@ -70,12 +90,13 @@ FROM ( '' AS index_comment, 'YES' AS visible, expr AS expression - FROM system.data_skipping_indices + FROM + system.data_skipping_indices WHERE database = '{0}' AND table = '{1}')) {2} -ORDER BY index_type, expression, column_name, seq_in_index;)", database, table, where_expression); +ORDER BY index_type, expression, seq_in_index;)", database, table, where_expression); /// Sorting is strictly speaking not necessary but 1. it is convenient for users, 2. SQL currently does not allow to /// sort the output of SHOW INDEXES otherwise (SELECT * FROM (SHOW INDEXES ...) ORDER BY ...) is rejected) and 3. some diff --git a/src/Interpreters/InterpreterShowTablesQuery.cpp b/src/Interpreters/InterpreterShowTablesQuery.cpp index 51038aaca46..78580f2ec02 100644 --- a/src/Interpreters/InterpreterShowTablesQuery.cpp +++ b/src/Interpreters/InterpreterShowTablesQuery.cpp @@ -121,9 +121,18 @@ String InterpreterShowTablesQuery::getRewrittenQuery() if (query.merges) { WriteBufferFromOwnString rewritten_query; - rewritten_query << "SELECT table, database, round((elapsed * (1 / merges.progress)) - merges.elapsed, 2) AS estimate_complete, round(elapsed,2) elapsed, " - "round(progress*100, 2) AS progress, is_mutation, formatReadableSize(total_size_bytes_compressed) AS size_compressed, " - "formatReadableSize(memory_usage) AS memory_usage FROM system.merges"; + rewritten_query << R"( + SELECT + table, + database, + merges.progress > 0 ? round(merges.elapsed * (1 - merges.progress) / merges.progress, 2) : NULL AS estimate_complete, + round(elapsed, 2) AS elapsed, + round(progress * 100, 2) AS progress, + is_mutation, + formatReadableSize(total_size_bytes_compressed) AS size_compressed, + formatReadableSize(memory_usage) AS memory_usage + FROM system.merges + )"; if (!query.like.empty()) { diff --git a/src/Interpreters/InterpreterSystemQuery.cpp b/src/Interpreters/InterpreterSystemQuery.cpp index d3526941b33..c284acfa308 100644 --- a/src/Interpreters/InterpreterSystemQuery.cpp +++ b/src/Interpreters/InterpreterSystemQuery.cpp @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/InterpreterTransactionControlQuery.cpp b/src/Interpreters/InterpreterTransactionControlQuery.cpp index 13872fbe3f5..e92d6f9c5e7 100644 --- a/src/Interpreters/InterpreterTransactionControlQuery.cpp +++ b/src/Interpreters/InterpreterTransactionControlQuery.cpp @@ -3,6 +3,7 @@ #include #include #include +#include namespace DB { diff --git a/src/Interpreters/InterpreterUndropQuery.cpp b/src/Interpreters/InterpreterUndropQuery.cpp index 920df3d6aed..8f935e951ef 100644 --- a/src/Interpreters/InterpreterUndropQuery.cpp +++ b/src/Interpreters/InterpreterUndropQuery.cpp @@ -64,7 +64,7 @@ BlockIO InterpreterUndropQuery::executeToTable(ASTUndropQuery & query) database->checkMetadataFilenameAvailability(table_id.table_name); - DatabaseCatalog::instance().dequeueDroppedTableCleanup(table_id); + DatabaseCatalog::instance().undropTable(table_id); return {}; } diff --git a/src/Interpreters/JoinSwitcher.cpp b/src/Interpreters/JoinSwitcher.cpp index 5ea347549c1..7f75b0d74b3 100644 --- a/src/Interpreters/JoinSwitcher.cpp +++ b/src/Interpreters/JoinSwitcher.cpp @@ -1,6 +1,6 @@ #include #include -#include +#include #include #include diff --git a/src/Interpreters/JoinUtils.cpp b/src/Interpreters/JoinUtils.cpp index 1788c9aca48..180a45d4295 100644 --- a/src/Interpreters/JoinUtils.cpp +++ b/src/Interpreters/JoinUtils.cpp @@ -554,7 +554,7 @@ static Blocks scatterBlockByHashImpl(const Strings & key_columns_names, const Bl for (const auto & key_name : key_columns_names) { ColumnPtr key_col = materializeColumn(block, key_name); - key_col->updateWeakHash32(hash); + hash.update(key_col->getWeakHash32()); } auto selector = hashToSelector(hash, sharder); diff --git a/src/Interpreters/JoinedTables.cpp b/src/Interpreters/JoinedTables.cpp index 457ed3ef4a6..0de2bf9cb1f 100644 --- a/src/Interpreters/JoinedTables.cpp +++ b/src/Interpreters/JoinedTables.cpp @@ -1,5 +1,6 @@ #include +#include #include #include diff --git a/src/Interpreters/MutationsInterpreter.cpp b/src/Interpreters/MutationsInterpreter.cpp index 6d3a4f30b34..480c6736bc5 100644 --- a/src/Interpreters/MutationsInterpreter.cpp +++ b/src/Interpreters/MutationsInterpreter.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -17,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -41,6 +43,7 @@ #include #include #include +#include namespace DB { diff --git a/src/Interpreters/MutationsNonDeterministicHelpers.cpp b/src/Interpreters/MutationsNonDeterministicHelpers.cpp index 7a4cb91acc0..bcff3e18b25 100644 --- a/src/Interpreters/MutationsNonDeterministicHelpers.cpp +++ b/src/Interpreters/MutationsNonDeterministicHelpers.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/NormalizeSelectWithUnionQueryVisitor.h b/src/Interpreters/NormalizeSelectWithUnionQueryVisitor.h index e8194f0dfe1..b2f55003da5 100644 --- a/src/Interpreters/NormalizeSelectWithUnionQueryVisitor.h +++ b/src/Interpreters/NormalizeSelectWithUnionQueryVisitor.h @@ -1,11 +1,10 @@ #pragma once -#include - +#include #include #include -#include +#include namespace DB { diff --git a/src/Interpreters/S3QueueLog.cpp b/src/Interpreters/ObjectStorageQueueLog.cpp similarity index 80% rename from src/Interpreters/S3QueueLog.cpp rename to src/Interpreters/ObjectStorageQueueLog.cpp index ba990a8ac25..c841984fd08 100644 --- a/src/Interpreters/S3QueueLog.cpp +++ b/src/Interpreters/ObjectStorageQueueLog.cpp @@ -8,19 +8,19 @@ #include #include #include -#include +#include namespace DB { -ColumnsDescription S3QueueLogElement::getColumnsDescription() +ColumnsDescription ObjectStorageQueueLogElement::getColumnsDescription() { auto status_datatype = std::make_shared( DataTypeEnum8::Values { - {"Processed", static_cast(S3QueueLogElement::S3QueueStatus::Processed)}, - {"Failed", static_cast(S3QueueLogElement::S3QueueStatus::Failed)}, + {"Processed", static_cast(ObjectStorageQueueLogElement::ObjectStorageQueueStatus::Processed)}, + {"Failed", static_cast(ObjectStorageQueueLogElement::ObjectStorageQueueStatus::Failed)}, }); return ColumnsDescription @@ -36,12 +36,11 @@ ColumnsDescription S3QueueLogElement::getColumnsDescription() {"status", status_datatype, "Status of the processing file"}, {"processing_start_time", std::make_shared(std::make_shared()), "Time of the start of processing the file"}, {"processing_end_time", std::make_shared(std::make_shared()), "Time of the end of processing the file"}, - {"ProfileEvents", std::make_shared(std::make_shared(), std::make_shared()), "Profile events collected while loading this file"}, {"exception", std::make_shared(), "Exception message if happened"}, }; } -void S3QueueLogElement::appendToBlock(MutableColumns & columns) const +void ObjectStorageQueueLogElement::appendToBlock(MutableColumns & columns) const { size_t i = 0; columns[i++]->insert(getFQDNOrHostName()); @@ -64,8 +63,6 @@ void S3QueueLogElement::appendToBlock(MutableColumns & columns) const else columns[i++]->insertDefault(); - ProfileEvents::dumpToMapColumn(counters_snapshot, columns[i++].get(), true); - columns[i++]->insert(exception); } diff --git a/src/Interpreters/S3QueueLog.h b/src/Interpreters/ObjectStorageQueueLog.h similarity index 68% rename from src/Interpreters/S3QueueLog.h rename to src/Interpreters/ObjectStorageQueueLog.h index 19e69c39247..669238d8dbb 100644 --- a/src/Interpreters/S3QueueLog.h +++ b/src/Interpreters/ObjectStorageQueueLog.h @@ -9,7 +9,7 @@ namespace DB { -struct S3QueueLogElement +struct ObjectStorageQueueLogElement { time_t event_time{}; @@ -20,18 +20,17 @@ struct S3QueueLogElement std::string file_name; size_t rows_processed = 0; - enum class S3QueueStatus : uint8_t + enum class ObjectStorageQueueStatus : uint8_t { Processed, Failed, }; - S3QueueStatus status; - ProfileEvents::Counters::Snapshot counters_snapshot; + ObjectStorageQueueStatus status; time_t processing_start_time; time_t processing_end_time; std::string exception; - static std::string name() { return "S3QueueLog"; } + static std::string name() { return "ObjectStorageQueueLog"; } static ColumnsDescription getColumnsDescription(); static NamesAndAliases getNamesAndAliases() { return {}; } @@ -39,9 +38,9 @@ struct S3QueueLogElement void appendToBlock(MutableColumns & columns) const; }; -class S3QueueLog : public SystemLog +class ObjectStorageQueueLog : public SystemLog { - using SystemLog::SystemLog; + using SystemLog::SystemLog; }; } diff --git a/src/Interpreters/OptimizeIfWithConstantConditionVisitor.cpp b/src/Interpreters/OptimizeIfWithConstantConditionVisitor.cpp index 20451fb20ad..48c9988b6fc 100644 --- a/src/Interpreters/OptimizeIfWithConstantConditionVisitor.cpp +++ b/src/Interpreters/OptimizeIfWithConstantConditionVisitor.cpp @@ -73,66 +73,55 @@ static bool tryExtractConstValueFromCondition(const ASTPtr & condition, bool & v return false; } -void OptimizeIfWithConstantConditionVisitor::visit(ASTPtr & current_ast) +void OptimizeIfWithConstantConditionVisitorData::visit(ASTFunction & function_node, ASTPtr & ast) { - if (!current_ast) - return; - checkStackSize(); - for (ASTPtr & child : current_ast->children) + if (function_node.name != "if") + return; + + if (!function_node.arguments) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Wrong number of arguments for function 'if' (0 instead of 3)"); + + if (function_node.arguments->children.size() != 3) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + "Wrong number of arguments for function 'if' ({} instead of 3)", + function_node.arguments->children.size()); + + const auto * args = function_node.arguments->as(); + + ASTPtr condition_expr = args->children[0]; + ASTPtr then_expr = args->children[1]; + ASTPtr else_expr = args->children[2]; + + bool condition; + if (tryExtractConstValueFromCondition(condition_expr, condition)) { - auto * function_node = child->as(); - if (!function_node || function_node->name != "if") + ASTPtr replace_ast = condition ? then_expr : else_expr; + ASTPtr child_copy = ast; + String replace_alias = replace_ast->tryGetAlias(); + String if_alias = ast->tryGetAlias(); + + if (replace_alias.empty()) { - visit(child); - continue; + replace_ast->setAlias(if_alias); + ast = replace_ast; + } + else + { + /// Only copy of one node is required here. + /// But IAST has only method for deep copy of subtree. + /// This can be a reason of performance degradation in case of deep queries. + ASTPtr replace_ast_deep_copy = replace_ast->clone(); + replace_ast_deep_copy->setAlias(if_alias); + ast = replace_ast_deep_copy; } - if (!function_node->arguments) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Wrong number of arguments for function 'if' (0 instead of 3)"); - - if (function_node->arguments->children.size() != 3) - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, - "Wrong number of arguments for function 'if' ({} instead of 3)", - function_node->arguments->children.size()); - - visit(function_node->arguments); - const auto * args = function_node->arguments->as(); - - ASTPtr condition_expr = args->children[0]; - ASTPtr then_expr = args->children[1]; - ASTPtr else_expr = args->children[2]; - - bool condition; - if (tryExtractConstValueFromCondition(condition_expr, condition)) + if (!if_alias.empty()) { - ASTPtr replace_ast = condition ? then_expr : else_expr; - ASTPtr child_copy = child; - String replace_alias = replace_ast->tryGetAlias(); - String if_alias = child->tryGetAlias(); - - if (replace_alias.empty()) - { - replace_ast->setAlias(if_alias); - child = replace_ast; - } - else - { - /// Only copy of one node is required here. - /// But IAST has only method for deep copy of subtree. - /// This can be a reason of performance degradation in case of deep queries. - ASTPtr replace_ast_deep_copy = replace_ast->clone(); - replace_ast_deep_copy->setAlias(if_alias); - child = replace_ast_deep_copy; - } - - if (!if_alias.empty()) - { - auto alias_it = aliases.find(if_alias); - if (alias_it != aliases.end() && alias_it->second.get() == child_copy.get()) - alias_it->second = child; - } + auto alias_it = aliases.find(if_alias); + if (alias_it != aliases.end() && alias_it->second.get() == child_copy.get()) + alias_it->second = ast; } } } diff --git a/src/Interpreters/OptimizeIfWithConstantConditionVisitor.h b/src/Interpreters/OptimizeIfWithConstantConditionVisitor.h index ad98f92bafd..3b46f90f07c 100644 --- a/src/Interpreters/OptimizeIfWithConstantConditionVisitor.h +++ b/src/Interpreters/OptimizeIfWithConstantConditionVisitor.h @@ -1,23 +1,24 @@ #pragma once #include +#include namespace DB { - -/// It removes Function_if node from AST if condition is constant. -/// TODO: rewrite with InDepthNodeVisitor -class OptimizeIfWithConstantConditionVisitor +struct OptimizeIfWithConstantConditionVisitorData { -public: - explicit OptimizeIfWithConstantConditionVisitor(Aliases & aliases_) + using TypeToVisit = ASTFunction; + + explicit OptimizeIfWithConstantConditionVisitorData(Aliases & aliases_) : aliases(aliases_) {} - void visit(ASTPtr & ast); - + void visit(ASTFunction & function_node, ASTPtr & ast); private: Aliases & aliases; }; +/// It removes Function_if node from AST if condition is constant. +using OptimizeIfWithConstantConditionVisitor = InDepthNodeVisitor, false>; + } diff --git a/src/Interpreters/PredicateExpressionsOptimizer.cpp b/src/Interpreters/PredicateExpressionsOptimizer.cpp index 8dc8c1c92cc..af67e62be1a 100644 --- a/src/Interpreters/PredicateExpressionsOptimizer.cpp +++ b/src/Interpreters/PredicateExpressionsOptimizer.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include diff --git a/src/Interpreters/PreparedSets.cpp b/src/Interpreters/PreparedSets.cpp index b04c8b1b725..43a04f8aff0 100644 --- a/src/Interpreters/PreparedSets.cpp +++ b/src/Interpreters/PreparedSets.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/ProcessorsProfileLog.h b/src/Interpreters/ProcessorsProfileLog.h index 8319d373f39..abece2604f2 100644 --- a/src/Interpreters/ProcessorsProfileLog.h +++ b/src/Interpreters/ProcessorsProfileLog.h @@ -25,11 +25,11 @@ struct ProcessorProfileLogElement String processor_name; /// Milliseconds spend in IProcessor::work() - UInt32 elapsed_us{}; + UInt64 elapsed_us{}; /// IProcessor::NeedData - UInt32 input_wait_elapsed_us{}; + UInt64 input_wait_elapsed_us{}; /// IProcessor::PortFull - UInt32 output_wait_elapsed_us{}; + UInt64 output_wait_elapsed_us{}; size_t input_rows{}; size_t input_bytes{}; diff --git a/src/Interpreters/QueryLog.cpp b/src/Interpreters/QueryLog.cpp index e63a2ae31aa..527159dc981 100644 --- a/src/Interpreters/QueryLog.cpp +++ b/src/Interpreters/QueryLog.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Interpreters/QueryLog.h b/src/Interpreters/QueryLog.h index bbaa7179757..d9cce4b9dd7 100644 --- a/src/Interpreters/QueryLog.h +++ b/src/Interpreters/QueryLog.h @@ -3,7 +3,7 @@ #include #include #include -#include +#include #include #include #include @@ -22,6 +22,8 @@ namespace ProfileEvents namespace DB { +struct Settings; + /** Allows to log information about queries execution: * - info about start of query execution; diff --git a/src/Interpreters/QueryViewsLog.h b/src/Interpreters/QueryViewsLog.h index 2de06fe3ddc..8f382ba6183 100644 --- a/src/Interpreters/QueryViewsLog.h +++ b/src/Interpreters/QueryViewsLog.h @@ -6,7 +6,7 @@ #include #include -#include +#include #include #include #include diff --git a/src/Interpreters/RewriteCountVariantsVisitor.cpp b/src/Interpreters/RewriteCountVariantsVisitor.cpp index f207bc51527..4a541c3765a 100644 --- a/src/Interpreters/RewriteCountVariantsVisitor.cpp +++ b/src/Interpreters/RewriteCountVariantsVisitor.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include diff --git a/src/Interpreters/RewriteFunctionToSubcolumnVisitor.cpp b/src/Interpreters/RewriteFunctionToSubcolumnVisitor.cpp deleted file mode 100644 index f0202199752..00000000000 --- a/src/Interpreters/RewriteFunctionToSubcolumnVisitor.cpp +++ /dev/null @@ -1,157 +0,0 @@ -#include -#include -#include -#include -#include -#include -#include - -namespace DB -{ - -namespace -{ - -ASTPtr transformToSubcolumn(const String & name_in_storage, const String & subcolumn_name) -{ - return std::make_shared(Nested::concatenateName(name_in_storage, subcolumn_name)); -} - -ASTPtr transformEmptyToSubcolumn(const String & name_in_storage, const String & subcolumn_name) -{ - auto ast = transformToSubcolumn(name_in_storage, subcolumn_name); - return makeASTFunction("equals", ast, std::make_shared(0u)); -} - -ASTPtr transformNotEmptyToSubcolumn(const String & name_in_storage, const String & subcolumn_name) -{ - auto ast = transformToSubcolumn(name_in_storage, subcolumn_name); - return makeASTFunction("notEquals", ast, std::make_shared(0u)); -} - -ASTPtr transformIsNotNullToSubcolumn(const String & name_in_storage, const String & subcolumn_name) -{ - auto ast = transformToSubcolumn(name_in_storage, subcolumn_name); - return makeASTFunction("not", ast); -} - -ASTPtr transformCountNullableToSubcolumn(const String & name_in_storage, const String & subcolumn_name) -{ - auto ast = transformToSubcolumn(name_in_storage, subcolumn_name); - return makeASTFunction("sum", makeASTFunction("not", ast)); -} - -ASTPtr transformMapContainsToSubcolumn(const String & name_in_storage, const String & subcolumn_name, const ASTPtr & arg) -{ - auto ast = transformToSubcolumn(name_in_storage, subcolumn_name); - return makeASTFunction("has", ast, arg); -} - -const std::unordered_map> unary_function_to_subcolumn = -{ - {"length", {TypeIndex::Array, "size0", transformToSubcolumn}}, - {"empty", {TypeIndex::Array, "size0", transformEmptyToSubcolumn}}, - {"notEmpty", {TypeIndex::Array, "size0", transformNotEmptyToSubcolumn}}, - {"isNull", {TypeIndex::Nullable, "null", transformToSubcolumn}}, - {"isNotNull", {TypeIndex::Nullable, "null", transformIsNotNullToSubcolumn}}, - {"count", {TypeIndex::Nullable, "null", transformCountNullableToSubcolumn}}, - {"mapKeys", {TypeIndex::Map, "keys", transformToSubcolumn}}, - {"mapValues", {TypeIndex::Map, "values", transformToSubcolumn}}, -}; - -const std::unordered_map> binary_function_to_subcolumn -{ - {"mapContains", {TypeIndex::Map, "keys", transformMapContainsToSubcolumn}}, -}; - -} - -void RewriteFunctionToSubcolumnData::visit(ASTFunction & function, ASTPtr & ast) const -{ - const auto & arguments = function.arguments->children; - if (arguments.empty() || arguments.size() > 2) - return; - - const auto * identifier = arguments[0]->as(); - if (!identifier) - return; - - const auto & columns = metadata_snapshot->getColumns(); - const auto & name_in_storage = identifier->name(); - - if (!columns.has(name_in_storage)) - return; - - const auto & column_type = columns.get(name_in_storage).type; - TypeIndex column_type_id = column_type->getTypeId(); - const auto & alias = function.tryGetAlias(); - - if (arguments.size() == 1) - { - auto it = unary_function_to_subcolumn.find(function.name); - if (it != unary_function_to_subcolumn.end()) - { - const auto & [type_id, subcolumn_name, transformer] = it->second; - if (column_type_id == type_id) - { - ast = transformer(name_in_storage, subcolumn_name); - ast->setAlias(alias); - } - } - } - else - { - if (function.name == "tupleElement" && column_type_id == TypeIndex::Tuple) - { - const auto * literal = arguments[1]->as(); - if (!literal) - return; - - String subcolumn_name; - auto value_type = literal->value.getType(); - if (value_type == Field::Types::UInt64) - { - const auto & type_tuple = assert_cast(*column_type); - auto index = literal->value.get(); - subcolumn_name = type_tuple.getNameByPosition(index); - } - else if (value_type == Field::Types::String) - subcolumn_name = literal->value.get(); - else - return; - - ast = transformToSubcolumn(name_in_storage, subcolumn_name); - ast->setAlias(alias); - } - else if (function.name == "variantElement" && column_type_id == TypeIndex::Variant) - { - const auto * literal = arguments[1]->as(); - if (!literal) - return; - - String subcolumn_name; - auto value_type = literal->value.getType(); - if (value_type != Field::Types::String) - return; - - subcolumn_name = literal->value.get(); - ast = transformToSubcolumn(name_in_storage, subcolumn_name); - ast->setAlias(alias); - } - else - { - auto it = binary_function_to_subcolumn.find(function.name); - if (it != binary_function_to_subcolumn.end()) - { - const auto & [type_id, subcolumn_name, transformer] = it->second; - if (column_type_id == type_id) - { - ast = transformer(name_in_storage, subcolumn_name, arguments[1]); - ast->setAlias(alias); - } - } - } - } -} - -} diff --git a/src/Interpreters/RewriteFunctionToSubcolumnVisitor.h b/src/Interpreters/RewriteFunctionToSubcolumnVisitor.h deleted file mode 100644 index 4d064bdee10..00000000000 --- a/src/Interpreters/RewriteFunctionToSubcolumnVisitor.h +++ /dev/null @@ -1,25 +0,0 @@ -#pragma once - -#include -#include - -namespace DB -{ - -class ASTFunction; - -/// Rewrites functions to subcolumns, if possible, to reduce amount of read data. -/// E.g. 'length(arr)' -> 'arr.size0', 'col IS NULL' -> 'col.null' -class RewriteFunctionToSubcolumnData -{ -public: - using TypeToVisit = ASTFunction; - void visit(ASTFunction & function, ASTPtr & ast) const; - - StorageMetadataPtr metadata_snapshot; -}; - -using RewriteFunctionToSubcolumnMatcher = OneTypeMatcher; -using RewriteFunctionToSubcolumnVisitor = InDepthNodeVisitor; - -} diff --git a/src/Interpreters/Session.cpp b/src/Interpreters/Session.cpp index bb8c415602f..5b378087707 100644 --- a/src/Interpreters/Session.cpp +++ b/src/Interpreters/Session.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include @@ -9,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -130,7 +132,7 @@ public: LOG_TRACE(log, "Reuse session from storage with session_id: {}, user_id: {}", key.second, key.first); - if (!session.unique()) + if (!isSharedPtrUnique(session)) throw Exception(ErrorCodes::SESSION_IS_LOCKED, "Session {} is locked by a concurrent client", session_id); return {session, false}; } @@ -156,7 +158,7 @@ public: return; } - if (!it->second.unique()) + if (!isSharedPtrUnique(it->second)) throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot close session {} with refcount {}", session_id, it->second.use_count()); sessions.erase(it); diff --git a/src/Interpreters/Squashing.cpp b/src/Interpreters/Squashing.cpp index f8b6a6542cc..488177c3b4f 100644 --- a/src/Interpreters/Squashing.cpp +++ b/src/Interpreters/Squashing.cpp @@ -1,34 +1,47 @@ #include #include +#include "Common/Logger.h" +#include "Common/logger_useful.h" #include +#include namespace DB { + namespace ErrorCodes { extern const int LOGICAL_ERROR; } Squashing::Squashing(Block header_, size_t min_block_size_rows_, size_t min_block_size_bytes_) - : header(header_) - , min_block_size_rows(min_block_size_rows_) + : min_block_size_rows(min_block_size_rows_) , min_block_size_bytes(min_block_size_bytes_) + , header(header_) { } Chunk Squashing::flush() { - return convertToChunk(std::move(chunks_to_merge_vec)); + if (!accumulated) + return {}; + + auto result = convertToChunk(extract()); + chassert(result); + return result; } Chunk Squashing::squash(Chunk && input_chunk) { - if (!input_chunk.hasChunkInfo()) + if (!input_chunk) return Chunk(); - const auto *info = getInfoFromChunk(input_chunk); - return squash(info->chunks); + auto squash_info = input_chunk.getChunkInfos().extract(); + + if (!squash_info) + throw Exception(ErrorCodes::LOGICAL_ERROR, "There is no ChunksToSquash in ChunkInfoPtr"); + + return squash(std::move(squash_info->chunks), std::move(input_chunk.getChunkInfos())); } Chunk Squashing::add(Chunk && input_chunk) @@ -37,67 +50,62 @@ Chunk Squashing::add(Chunk && input_chunk) return {}; /// Just read block is already enough. - if (isEnoughSize(input_chunk.getNumRows(), input_chunk.bytes())) + if (isEnoughSize(input_chunk)) { /// If no accumulated data, return just read block. - if (chunks_to_merge_vec.empty()) + if (!accumulated) { - chunks_to_merge_vec.push_back(std::move(input_chunk)); - Chunk res_chunk = convertToChunk(std::move(chunks_to_merge_vec)); - chunks_to_merge_vec.clear(); - return res_chunk; + accumulated.add(std::move(input_chunk)); + return convertToChunk(extract()); } /// Return accumulated data (maybe it has small size) and place new block to accumulated data. - Chunk res_chunk = convertToChunk(std::move(chunks_to_merge_vec)); - chunks_to_merge_vec.clear(); - changeCurrentSize(input_chunk.getNumRows(), input_chunk.bytes()); - chunks_to_merge_vec.push_back(std::move(input_chunk)); + Chunk res_chunk = convertToChunk(extract()); + accumulated.add(std::move(input_chunk)); return res_chunk; } /// Accumulated block is already enough. - if (isEnoughSize(accumulated_size.rows, accumulated_size.bytes)) + if (isEnoughSize()) { /// Return accumulated data and place new block to accumulated data. - Chunk res_chunk = convertToChunk(std::move(chunks_to_merge_vec)); - chunks_to_merge_vec.clear(); - changeCurrentSize(input_chunk.getNumRows(), input_chunk.bytes()); - chunks_to_merge_vec.push_back(std::move(input_chunk)); + Chunk res_chunk = convertToChunk(extract()); + accumulated.add(std::move(input_chunk)); return res_chunk; } /// Pushing data into accumulating vector - expandCurrentSize(input_chunk.getNumRows(), input_chunk.bytes()); - chunks_to_merge_vec.push_back(std::move(input_chunk)); + accumulated.add(std::move(input_chunk)); /// If accumulated data is big enough, we send it - if (isEnoughSize(accumulated_size.rows, accumulated_size.bytes)) - { - Chunk res_chunk = convertToChunk(std::move(chunks_to_merge_vec)); - changeCurrentSize(0, 0); - chunks_to_merge_vec.clear(); - return res_chunk; - } + if (isEnoughSize()) + return convertToChunk(extract()); + return {}; } -Chunk Squashing::convertToChunk(std::vector && chunks) const +Chunk Squashing::convertToChunk(CurrentData && data) const { - if (chunks.empty()) + if (data.chunks.empty()) return {}; auto info = std::make_shared(); - info->chunks = std::move(chunks); + info->chunks = std::move(data.chunks); - chunks.clear(); + // It is imortant that chunk is not empty, it has to have columns even if they are empty + // Sometimes there are could be no columns in header but not empty rows in chunks + // That happens when we intend to add defaults for the missing columns after + auto aggr_chunk = Chunk(header.getColumns(), 0); + if (header.columns() == 0) + aggr_chunk = Chunk(header.getColumns(), data.getRows()); - return Chunk(header.cloneEmptyColumns(), 0, info); + aggr_chunk.getChunkInfos().add(std::move(info)); + chassert(aggr_chunk); + return aggr_chunk; } -Chunk Squashing::squash(std::vector & input_chunks) +Chunk Squashing::squash(std::vector && input_chunks, Chunk::ChunkInfoCollection && infos) { - Chunk accumulated_chunk; std::vector mutable_columns = {}; size_t rows = 0; for (const Chunk & chunk : input_chunks) @@ -119,35 +127,17 @@ Chunk Squashing::squash(std::vector & input_chunks) for (size_t j = 0, size = mutable_columns.size(); j < size; ++j) { const auto source_column = columns[j]; - mutable_columns[j]->insertRangeFrom(*source_column, 0, source_column->size()); } } - accumulated_chunk.setColumns(std::move(mutable_columns), rows); - return accumulated_chunk; -} -const ChunksToSquash* Squashing::getInfoFromChunk(const Chunk & chunk) -{ - const auto& info = chunk.getChunkInfo(); - const auto * agg_info = typeid_cast(info.get()); + Chunk result; + result.setColumns(std::move(mutable_columns), rows); + result.setChunkInfos(infos); + result.getChunkInfos().appendIfUniq(std::move(input_chunks.back().getChunkInfos())); - if (!agg_info) - throw Exception(ErrorCodes::LOGICAL_ERROR, "There is no ChunksToSquash in ChunkInfoPtr"); - - return agg_info; -} - -void Squashing::expandCurrentSize(size_t rows, size_t bytes) -{ - accumulated_size.rows += rows; - accumulated_size.bytes += bytes; -} - -void Squashing::changeCurrentSize(size_t rows, size_t bytes) -{ - accumulated_size.rows = rows; - accumulated_size.bytes = bytes; + chassert(result); + return result; } bool Squashing::isEnoughSize(size_t rows, size_t bytes) const @@ -156,4 +146,29 @@ bool Squashing::isEnoughSize(size_t rows, size_t bytes) const || (min_block_size_rows && rows >= min_block_size_rows) || (min_block_size_bytes && bytes >= min_block_size_bytes); } + +bool Squashing::isEnoughSize() const +{ + return isEnoughSize(accumulated.getRows(), accumulated.getBytes()); +}; + +bool Squashing::isEnoughSize(const Chunk & chunk) const +{ + return isEnoughSize(chunk.getNumRows(), chunk.bytes()); +} + +void Squashing::CurrentData::add(Chunk && chunk) +{ + rows += chunk.getNumRows(); + bytes += chunk.bytes(); + chunks.push_back(std::move(chunk)); +} + +Squashing::CurrentData Squashing::extract() +{ + auto result = std::move(accumulated); + accumulated = {}; + return result; +} + } diff --git a/src/Interpreters/Squashing.h b/src/Interpreters/Squashing.h index d76cca60e41..71ed4c4185a 100644 --- a/src/Interpreters/Squashing.h +++ b/src/Interpreters/Squashing.h @@ -8,9 +8,18 @@ namespace DB { -struct ChunksToSquash : public ChunkInfo +class ChunksToSquash : public ChunkInfoCloneable { - mutable std::vector chunks = {}; +public: + ChunksToSquash() = default; + ChunksToSquash(const ChunksToSquash & other) + { + chunks.reserve(other.chunks.size()); + for (const auto & chunk: other.chunks) + chunks.push_back(chunk.clone()); + } + + std::vector chunks = {}; }; /** Merging consecutive passed blocks to specified minimum size. @@ -36,34 +45,37 @@ public: static Chunk squash(Chunk && input_chunk); Chunk flush(); - bool isDataLeft() - { - return !chunks_to_merge_vec.empty(); - } + void setHeader(Block header_) { header = std::move(header_); } + const Block & getHeader() const { return header; } - Block header; private: - struct CurrentSize + struct CurrentData { + std::vector chunks = {}; size_t rows = 0; size_t bytes = 0; + + explicit operator bool () const { return !chunks.empty(); } + size_t getRows() const { return rows; } + size_t getBytes() const { return bytes; } + void add(Chunk && chunk); }; - std::vector chunks_to_merge_vec = {}; - size_t min_block_size_rows; - size_t min_block_size_bytes; + const size_t min_block_size_rows; + const size_t min_block_size_bytes; + Block header; - CurrentSize accumulated_size; + CurrentData accumulated; - static const ChunksToSquash * getInfoFromChunk(const Chunk & chunk); + static Chunk squash(std::vector && input_chunks, Chunk::ChunkInfoCollection && infos); - static Chunk squash(std::vector & input_chunks); - - void expandCurrentSize(size_t rows, size_t bytes); - void changeCurrentSize(size_t rows, size_t bytes); + bool isEnoughSize() const; bool isEnoughSize(size_t rows, size_t bytes) const; + bool isEnoughSize(const Chunk & chunk) const; - Chunk convertToChunk(std::vector && chunks) const; + CurrentData extract(); + + Chunk convertToChunk(CurrentData && data) const; }; } diff --git a/src/Interpreters/SystemLog.cpp b/src/Interpreters/SystemLog.cpp index 7508d2a7e34..572481e6b12 100644 --- a/src/Interpreters/SystemLog.cpp +++ b/src/Interpreters/SystemLog.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -25,7 +26,7 @@ #include #include #include -#include +#include #include #include #include @@ -306,7 +307,8 @@ SystemLogs::SystemLogs(ContextPtr global_context, const Poco::Util::AbstractConf processors_profile_log = createSystemLog(global_context, "system", "processors_profile_log", config, "processors_profile_log", "Contains profiling information on processors level (building blocks for a pipeline for query execution."); asynchronous_insert_log = createSystemLog(global_context, "system", "asynchronous_insert_log", config, "asynchronous_insert_log", "Contains a history for all asynchronous inserts executed on current server."); backup_log = createSystemLog(global_context, "system", "backup_log", config, "backup_log", "Contains logging entries with the information about BACKUP and RESTORE operations."); - s3_queue_log = createSystemLog(global_context, "system", "s3queue_log", config, "s3queue_log", "Contains logging entries with the information files processes by S3Queue engine."); + s3_queue_log = createSystemLog(global_context, "system", "s3queue_log", config, "s3queue_log", "Contains logging entries with the information files processes by S3Queue engine."); + azure_queue_log = createSystemLog(global_context, "system", "azure_queue_log", config, "azure_queue_log", "Contains logging entries with the information files processes by S3Queue engine."); blob_storage_log = createSystemLog(global_context, "system", "blob_storage_log", config, "blob_storage_log", "Contains logging entries with information about various blob storage operations such as uploads and deletes."); if (query_log) @@ -355,10 +357,15 @@ SystemLogs::SystemLogs(ContextPtr global_context, const Poco::Util::AbstractConf if (blob_storage_log) logs.emplace_back(blob_storage_log.get()); + bool should_prepare = global_context->getServerSettings().prepare_system_log_tables_on_startup; try { for (auto & log : logs) + { log->startup(); + if (should_prepare) + log->prepareTable(); + } } catch (...) { @@ -537,7 +544,13 @@ void SystemLog::flushImpl(const std::vector & to_flush, insert_context->makeQueryContext(); addSettingsForQuery(insert_context, IAST::QueryKind::Insert); - InterpreterInsertQuery interpreter(query_ptr, insert_context); + InterpreterInsertQuery interpreter( + query_ptr, + insert_context, + /* allow_materialized */ false, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); BlockIO io = interpreter.execute(); PushingPipelineExecutor executor(io.pipeline); diff --git a/src/Interpreters/SystemLog.h b/src/Interpreters/SystemLog.h index c6f4a8ea9a0..0ac468b15ec 100644 --- a/src/Interpreters/SystemLog.h +++ b/src/Interpreters/SystemLog.h @@ -53,7 +53,7 @@ class FilesystemCacheLog; class FilesystemReadPrefetchesLog; class AsynchronousInsertLog; class BackupLog; -class S3QueueLog; +class ObjectStorageQueueLog; class BlobStorageLog; /// System logs should be destroyed in destructor of the last Context and before tables, @@ -76,7 +76,8 @@ struct SystemLogs std::shared_ptr error_log; /// Used to log errors. std::shared_ptr filesystem_cache_log; std::shared_ptr filesystem_read_prefetches_log; - std::shared_ptr s3_queue_log; + std::shared_ptr s3_queue_log; + std::shared_ptr azure_queue_log; /// Metrics from system.asynchronous_metrics. std::shared_ptr asynchronous_metric_log; /// OpenTelemetry trace spans. @@ -134,6 +135,12 @@ public: void stopFlushThread() override; + /** Creates new table if it does not exist. + * Renames old table if its structure is not suitable. + * This cannot be done in constructor to avoid deadlock while renaming a table under locked Context when SystemLog object is created. + */ + void prepareTable() override; + protected: LoggerPtr log; @@ -144,12 +151,6 @@ protected: StoragePtr getStorage() const; - /** Creates new table if it does not exist. - * Renames old table if its structure is not suitable. - * This cannot be done in constructor to avoid deadlock while renaming a table under locked Context when SystemLog object is created. - */ - void prepareTable() override; - /// Some tables can override settings for internal queries virtual void addSettingsForQuery(ContextMutablePtr & mutable_context, IAST::QueryKind query_kind) const; diff --git a/src/Interpreters/TableJoin.h b/src/Interpreters/TableJoin.h index 8e83233e54c..ed14fe4273c 100644 --- a/src/Interpreters/TableJoin.h +++ b/src/Interpreters/TableJoin.h @@ -2,7 +2,6 @@ #include #include -#include #include #include #include diff --git a/src/Interpreters/TemporaryDataOnDisk.cpp b/src/Interpreters/TemporaryDataOnDisk.cpp index a74b5bba2b9..7f0fb8cd6ca 100644 --- a/src/Interpreters/TemporaryDataOnDisk.cpp +++ b/src/Interpreters/TemporaryDataOnDisk.cpp @@ -3,6 +3,8 @@ #include #include +#include +#include #include #include #include @@ -224,25 +226,37 @@ struct TemporaryFileStream::OutputWriter bool finalized = false; }; -TemporaryFileStream::Reader::Reader(const String & path, const Block & header_, size_t size) - : in_file_buf(path, size ? std::min(DBMS_DEFAULT_BUFFER_SIZE, size) : DBMS_DEFAULT_BUFFER_SIZE) - , in_compressed_buf(in_file_buf) - , in_reader(in_compressed_buf, header_, DBMS_TCP_PROTOCOL_VERSION) +TemporaryFileStream::Reader::Reader(const String & path_, const Block & header_, size_t size_) + : path(path_) + , size(size_ ? std::min(size_, DBMS_DEFAULT_BUFFER_SIZE) : DBMS_DEFAULT_BUFFER_SIZE) + , header(header_) { LOG_TEST(getLogger("TemporaryFileStream"), "Reading {} from {}", header_.dumpStructure(), path); } -TemporaryFileStream::Reader::Reader(const String & path, size_t size) - : in_file_buf(path, size ? std::min(DBMS_DEFAULT_BUFFER_SIZE, size) : DBMS_DEFAULT_BUFFER_SIZE) - , in_compressed_buf(in_file_buf) - , in_reader(in_compressed_buf, DBMS_TCP_PROTOCOL_VERSION) +TemporaryFileStream::Reader::Reader(const String & path_, size_t size_) + : path(path_) + , size(size_ ? std::min(size_, DBMS_DEFAULT_BUFFER_SIZE) : DBMS_DEFAULT_BUFFER_SIZE) { LOG_TEST(getLogger("TemporaryFileStream"), "Reading from {}", path); } Block TemporaryFileStream::Reader::read() { - return in_reader.read(); + if (!in_reader) + { + if (fs::exists(path)) + in_file_buf = std::make_unique(path, size); + else + in_file_buf = std::make_unique(); + + in_compressed_buf = std::make_unique(*in_file_buf); + if (header.has_value()) + in_reader = std::make_unique(*in_compressed_buf, header.value(), DBMS_TCP_PROTOCOL_VERSION); + else + in_reader = std::make_unique(*in_compressed_buf, DBMS_TCP_PROTOCOL_VERSION); + } + return in_reader->read(); } TemporaryFileStream::TemporaryFileStream(TemporaryFileOnDiskHolder file_, const Block & header_, TemporaryDataOnDisk * parent_) diff --git a/src/Interpreters/TemporaryDataOnDisk.h b/src/Interpreters/TemporaryDataOnDisk.h index 488eed70da9..d541c93e031 100644 --- a/src/Interpreters/TemporaryDataOnDisk.h +++ b/src/Interpreters/TemporaryDataOnDisk.h @@ -151,9 +151,13 @@ public: Block read(); - ReadBufferFromFile in_file_buf; - CompressedReadBuffer in_compressed_buf; - NativeReader in_reader; + const std::string path; + const size_t size; + const std::optional header; + + std::unique_ptr in_file_buf; + std::unique_ptr in_compressed_buf; + std::unique_ptr in_reader; }; struct Stat diff --git a/src/Interpreters/ThreadStatusExt.cpp b/src/Interpreters/ThreadStatusExt.cpp index 9ca521a4ab3..1f7c6b1fe68 100644 --- a/src/Interpreters/ThreadStatusExt.cpp +++ b/src/Interpreters/ThreadStatusExt.cpp @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -233,7 +234,8 @@ void ThreadStatus::attachToGroupImpl(const ThreadGroupPtr & thread_group_) { /// Attach or init current thread to thread group and copy useful information from it thread_group = thread_group_; - thread_group->linkThread(thread_id); + if (!internal_thread) + thread_group->linkThread(thread_id); performance_counters.setParent(&thread_group->performance_counters); memory_tracker.setParent(&thread_group->memory_tracker); @@ -269,7 +271,8 @@ void ThreadStatus::detachFromGroup() /// Extract MemoryTracker out from query and user context memory_tracker.setParent(&total_memory_tracker); - thread_group->unlinkThread(); + if (!internal_thread) + thread_group->unlinkThread(); thread_group.reset(); diff --git a/src/Interpreters/TreeOptimizer.cpp b/src/Interpreters/TreeOptimizer.cpp index c331c8640d6..b872eb94fde 100644 --- a/src/Interpreters/TreeOptimizer.cpp +++ b/src/Interpreters/TreeOptimizer.cpp @@ -17,7 +17,6 @@ #include #include #include -#include #include #include #include @@ -564,12 +563,6 @@ void transformIfStringsIntoEnum(ASTPtr & query) ConvertStringsToEnumVisitor(convert_data).visit(query); } -void optimizeFunctionsToSubcolumns(ASTPtr & query, const StorageMetadataPtr & metadata_snapshot) -{ - RewriteFunctionToSubcolumnVisitor::Data data{metadata_snapshot}; - RewriteFunctionToSubcolumnVisitor(data).visit(query); -} - void optimizeOrLikeChain(ASTPtr & query) { ConvertFunctionOrLikeVisitor::Data data = {}; @@ -584,7 +577,8 @@ void TreeOptimizer::optimizeIf(ASTPtr & query, Aliases & aliases, bool if_chain_ optimizeMultiIfToIf(query); /// Optimize if with constant condition after constants was substituted instead of scalar subqueries. - OptimizeIfWithConstantConditionVisitor(aliases).visit(query); + OptimizeIfWithConstantConditionVisitorData visitor_data(aliases); + OptimizeIfWithConstantConditionVisitor(visitor_data).visit(query); if (if_chain_to_multiif) OptimizeIfChainsVisitor().visit(query); @@ -634,9 +628,6 @@ void TreeOptimizer::apply(ASTPtr & query, TreeRewriterResult & result, if (!select_query) throw Exception(ErrorCodes::LOGICAL_ERROR, "Select analyze for not select asts."); - if (settings.optimize_functions_to_subcolumns && result.storage_snapshot && result.storage->supportsSubcolumns()) - optimizeFunctionsToSubcolumns(query, result.storage_snapshot->metadata); - /// Move arithmetic operations out of aggregation functions if (settings.optimize_arithmetic_operations_in_aggregate_functions) optimizeAggregationFunctions(query); diff --git a/src/Interpreters/TreeRewriter.cpp b/src/Interpreters/TreeRewriter.cpp index a3c5a7ed3ed..6ce6f5e454e 100644 --- a/src/Interpreters/TreeRewriter.cpp +++ b/src/Interpreters/TreeRewriter.cpp @@ -1188,7 +1188,7 @@ bool TreeRewriterResult::collectUsedColumns(const ASTPtr & query, bool is_select } } - /// Check for dynamic subcolums in unknown required columns. + /// Check for dynamic subcolumns in unknown required columns. if (!unknown_required_source_columns.empty()) { for (const NameAndTypePair & pair : source_columns_ordinary) diff --git a/src/Interpreters/convertFieldToType.cpp b/src/Interpreters/convertFieldToType.cpp index 184c263dbdb..1a7c166c6a5 100644 --- a/src/Interpreters/convertFieldToType.cpp +++ b/src/Interpreters/convertFieldToType.cpp @@ -356,7 +356,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID for (size_t i = 0; i < src_arr_size; ++i) { res[i] = convertFieldToType(src_arr[i], element_type); - if (res[i].isNull() && !element_type.isNullable()) + if (res[i].isNull() && !canContainNull(element_type)) { // See the comment for Tuples below. have_unconvertible_element = true; @@ -384,25 +384,25 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID { const auto & element_type = *(type_tuple->getElements()[i]); res[i] = convertFieldToType(src_tuple[i], element_type); - if (!res[i].isNull() || element_type.isNullable()) - continue; - - /* - * Either the source element was Null, or the conversion did not - * succeed, because the source and the requested types of the - * element are compatible, but the value is not convertible - * (e.g. trying to convert -1 from Int8 to UInt8). In these - * cases, consider the whole tuple also compatible but not - * convertible. According to the specification of this function, - * we must return Null in this case. - * - * The following elements might be not even compatible, so it - * makes sense to check them to detect user errors. Remember - * that there is an unconvertible element, and try to process - * the remaining ones. The convertFieldToType for each element - * will throw if it detects incompatibility. - */ - have_unconvertible_element = true; + if (res[i].isNull() && !canContainNull(element_type)) + { + /* + * Either the source element was Null, or the conversion did not + * succeed, because the source and the requested types of the + * element are compatible, but the value is not convertible + * (e.g. trying to convert -1 from Int8 to UInt8). In these + * cases, consider the whole tuple also compatible but not + * convertible. According to the specification of this function, + * we must return Null in this case. + * + * The following elements might be not even compatible, so it + * makes sense to check them to detect user errors. Remember + * that there is an unconvertible element, and try to process + * the remaining ones. The convertFieldToType for each element + * will throw if it detects incompatibility. + */ + have_unconvertible_element = true; + } } return have_unconvertible_element ? Field(Null()) : Field(res); @@ -433,11 +433,11 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID updated_entry[0] = convertFieldToType(key, key_type); - if (updated_entry[0].isNull() && !key_type.isNullable()) + if (updated_entry[0].isNull() && !canContainNull(key_type)) have_unconvertible_element = true; updated_entry[1] = convertFieldToType(value, value_type); - if (updated_entry[1].isNull() && !value_type.isNullable()) + if (updated_entry[1].isNull() && !canContainNull(value_type)) have_unconvertible_element = true; res[i] = updated_entry; @@ -545,7 +545,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID catch (Exception & e) { if (e.code() == ErrorCodes::UNEXPECTED_DATA_AFTER_PARSED_VALUE) - throw Exception(ErrorCodes::TYPE_MISMATCH, "Cannot convert string {} to type {}", src.get(), type.getName()); + throw Exception(ErrorCodes::TYPE_MISMATCH, "Cannot convert string '{}' to type {}", src.get(), type.getName()); e.addMessage(fmt::format("while converting '{}' to {}", src.get(), type.getName())); throw; @@ -592,7 +592,7 @@ Field convertFieldToType(const Field & from_value, const IDataType & to_type, co Field convertFieldToTypeOrThrow(const Field & from_value, const IDataType & to_type, const IDataType * from_type_hint) { bool is_null = from_value.isNull(); - if (is_null && !to_type.isNullable() && !to_type.isLowCardinalityNullable()) + if (is_null && !canContainNull(to_type)) throw Exception(ErrorCodes::TYPE_MISMATCH, "Cannot convert NULL to {}", to_type.getName()); Field converted = convertFieldToType(from_value, to_type, from_type_hint); diff --git a/src/Interpreters/evaluateConstantExpression.cpp b/src/Interpreters/evaluateConstantExpression.cpp index 4e1a2bcf5ee..ac80ad2fb68 100644 --- a/src/Interpreters/evaluateConstantExpression.cpp +++ b/src/Interpreters/evaluateConstantExpression.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include #include @@ -24,6 +25,7 @@ #include #include #include + #include diff --git a/src/Interpreters/examples/CMakeLists.txt b/src/Interpreters/examples/CMakeLists.txt index 8bb7f9eeb98..4b1ec970b26 100644 --- a/src/Interpreters/examples/CMakeLists.txt +++ b/src/Interpreters/examples/CMakeLists.txt @@ -2,34 +2,34 @@ clickhouse_add_executable (hash_map hash_map.cpp) target_link_libraries (hash_map PRIVATE dbms clickhouse_functions ch_contrib::sparsehash) clickhouse_add_executable (hash_map_lookup hash_map_lookup.cpp) -target_link_libraries (hash_map_lookup PRIVATE clickhouse_common_io clickhouse_compression) +target_link_libraries (hash_map_lookup PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression) clickhouse_add_executable (hash_map3 hash_map3.cpp) -target_link_libraries (hash_map3 PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::farmhash ch_contrib::metrohash) +target_link_libraries (hash_map3 PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression ch_contrib::farmhash ch_contrib::metrohash) clickhouse_add_executable (hash_map_string hash_map_string.cpp) -target_link_libraries (hash_map_string PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::sparsehash) +target_link_libraries (hash_map_string PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression ch_contrib::sparsehash) clickhouse_add_executable (hash_map_string_2 hash_map_string_2.cpp) -target_link_libraries (hash_map_string_2 PRIVATE clickhouse_common_io clickhouse_compression) +target_link_libraries (hash_map_string_2 PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression) clickhouse_add_executable (hash_map_string_3 hash_map_string_3.cpp) -target_link_libraries (hash_map_string_3 PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::farmhash ch_contrib::metrohash) +target_link_libraries (hash_map_string_3 PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression ch_contrib::farmhash ch_contrib::metrohash) clickhouse_add_executable (hash_map_string_small hash_map_string_small.cpp) -target_link_libraries (hash_map_string_small PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::sparsehash) +target_link_libraries (hash_map_string_small PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression ch_contrib::sparsehash) clickhouse_add_executable (string_hash_map string_hash_map.cpp) -target_link_libraries (string_hash_map PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::sparsehash) +target_link_libraries (string_hash_map PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression ch_contrib::sparsehash) clickhouse_add_executable (string_hash_map_aggregation string_hash_map.cpp) -target_link_libraries (string_hash_map_aggregation PRIVATE clickhouse_common_io clickhouse_compression) +target_link_libraries (string_hash_map_aggregation PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression) clickhouse_add_executable (string_hash_set string_hash_set.cpp) -target_link_libraries (string_hash_set PRIVATE clickhouse_common_io clickhouse_compression) +target_link_libraries (string_hash_set PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression) clickhouse_add_executable (two_level_hash_map two_level_hash_map.cpp) -target_link_libraries (two_level_hash_map PRIVATE clickhouse_common_io clickhouse_compression ch_contrib::sparsehash) +target_link_libraries (two_level_hash_map PRIVATE clickhouse_common_io clickhouse_common_config clickhouse_compression ch_contrib::sparsehash) clickhouse_add_executable (jit_example jit_example.cpp) target_link_libraries (jit_example PRIVATE dbms) diff --git a/src/Interpreters/executeDDLQueryOnCluster.cpp b/src/Interpreters/executeDDLQueryOnCluster.cpp index e372f036073..1b57ad2b622 100644 --- a/src/Interpreters/executeDDLQueryOnCluster.cpp +++ b/src/Interpreters/executeDDLQueryOnCluster.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -237,6 +238,7 @@ private: Int64 timeout_seconds = 120; bool is_replicated_database = false; bool throw_on_timeout = true; + bool throw_on_timeout_only_active = false; bool only_running_hosts = false; bool timeout_exceeded = false; @@ -316,8 +318,8 @@ DDLQueryStatusSource::DDLQueryStatusSource( , log(getLogger("DDLQueryStatusSource")) { auto output_mode = context->getSettingsRef().distributed_ddl_output_mode; - throw_on_timeout = output_mode == DistributedDDLOutputMode::THROW || output_mode == DistributedDDLOutputMode::THROW_ONLY_ACTIVE - || output_mode == DistributedDDLOutputMode::NONE || output_mode == DistributedDDLOutputMode::NONE_ONLY_ACTIVE; + throw_on_timeout = output_mode == DistributedDDLOutputMode::THROW || output_mode == DistributedDDLOutputMode::NONE; + throw_on_timeout_only_active = output_mode == DistributedDDLOutputMode::THROW_ONLY_ACTIVE || output_mode == DistributedDDLOutputMode::NONE_ONLY_ACTIVE; if (hosts_to_wait) { @@ -451,7 +453,7 @@ Chunk DDLQueryStatusSource::generate() "({} of them are currently executing the task, {} are inactive). " "They are going to execute the query in background. Was waiting for {} seconds{}"; - if (throw_on_timeout) + if (throw_on_timeout || (throw_on_timeout_only_active && !stop_waiting_offline_hosts)) { if (!first_exception) first_exception = std::make_unique(Exception(ErrorCodes::TIMEOUT_EXCEEDED, @@ -536,7 +538,7 @@ Chunk DDLQueryStatusSource::generate() ExecutionStatus status(-1, "Cannot obtain error message"); /// Replicated database retries in case of error, it should not write error status. -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD bool need_check_status = true; #else bool need_check_status = !is_replicated_database; diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp index 9f33cbf1c27..ba6fc0f14a0 100644 --- a/src/Interpreters/executeQuery.cpp +++ b/src/Interpreters/executeQuery.cpp @@ -475,10 +475,9 @@ void logQueryFinish( processor_elem.processor_name = processor->getName(); - /// NOTE: convert this to UInt64 - processor_elem.elapsed_us = static_cast(processor->getElapsedUs()); - processor_elem.input_wait_elapsed_us = static_cast(processor->getInputWaitElapsedUs()); - processor_elem.output_wait_elapsed_us = static_cast(processor->getOutputWaitElapsedUs()); + processor_elem.elapsed_us = static_cast(processor->getElapsedNs() / 1000U); + processor_elem.input_wait_elapsed_us = static_cast(processor->getInputWaitElapsedNs() / 1000U); + processor_elem.output_wait_elapsed_us = static_cast(processor->getOutputWaitElapsedNs() / 1000U); auto stats = processor->getProcessorDataStats(); processor_elem.input_rows = stats.input_rows; @@ -1049,14 +1048,14 @@ static std::tuple executeQueryImpl( /// In case when the client had to retry some mini-INSERTs then they will be properly deduplicated /// by the source tables. This functionality is controlled by a setting `async_insert_deduplicate`. /// But then they will be glued together into a block and pushed through a chain of Materialized Views if any. - /// The process of forming such blocks is not deteministic so each time we retry mini-INSERTs the resulting + /// The process of forming such blocks is not deterministic so each time we retry mini-INSERTs the resulting /// block may be concatenated differently. /// That's why deduplication in dependent Materialized Views doesn't make sense in presence of async INSERTs. if (settings.throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert && settings.deduplicate_blocks_in_dependent_materialized_views) throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, - "Deduplication is dependent materialized view cannot work together with async inserts. "\ - "Please disable eiher `deduplicate_blocks_in_dependent_materialized_views` or `async_insert` setting."); + "Deduplication in dependent materialized view cannot work together with async inserts. "\ + "Please disable either `deduplicate_blocks_in_dependent_materialized_views` or `async_insert` setting."); quota = context->getQuota(); if (quota) diff --git a/src/Interpreters/formatWithPossiblyHidingSecrets.h b/src/Interpreters/formatWithPossiblyHidingSecrets.h index 25e1e7a5616..039bcbc2bca 100644 --- a/src/Interpreters/formatWithPossiblyHidingSecrets.h +++ b/src/Interpreters/formatWithPossiblyHidingSecrets.h @@ -2,6 +2,8 @@ #include "Access/ContextAccess.h" #include "Interpreters/Context.h" +#include + namespace DB { struct SecretHidingFormatSettings diff --git a/src/Interpreters/getColumnFromBlock.cpp b/src/Interpreters/getColumnFromBlock.cpp index 972e109afb3..89166bb2b3e 100644 --- a/src/Interpreters/getColumnFromBlock.cpp +++ b/src/Interpreters/getColumnFromBlock.cpp @@ -31,6 +31,36 @@ ColumnPtr tryGetColumnFromBlock(const Block & block, const NameAndTypePair & req return castColumn({elem_column, elem_type, ""}, requested_column.type); } +ColumnPtr tryGetSubcolumnFromBlock(const Block & block, const DataTypePtr & requested_column_type, const NameAndTypePair & requested_subcolumn) +{ + const auto * elem = block.findByName(requested_subcolumn.getNameInStorage()); + if (!elem) + return nullptr; + + auto subcolumn_name = requested_subcolumn.getSubcolumnName(); + /// If requested subcolumn is dynamic, we should first perform cast and then + /// extract the subcolumn, because the data of dynamic subcolumn can change after cast. + if ((elem->type->hasDynamicSubcolumns() || requested_column_type->hasDynamicSubcolumns()) && !elem->type->equals(*requested_column_type)) + { + auto casted_column = castColumn({elem->column, elem->type, ""}, requested_column_type); + auto elem_column = requested_column_type->tryGetSubcolumn(subcolumn_name, casted_column); + auto elem_type = requested_column_type->tryGetSubcolumnType(subcolumn_name); + + if (!elem_type || !elem_column) + return nullptr; + + return elem_column; + } + + auto elem_column = elem->type->tryGetSubcolumn(subcolumn_name, elem->column); + auto elem_type = elem->type->tryGetSubcolumnType(subcolumn_name); + + if (!elem_type || !elem_column) + return nullptr; + + return castColumn({elem_column, elem_type, ""}, requested_subcolumn.type); +} + ColumnPtr getColumnFromBlock(const Block & block, const NameAndTypePair & requested_column) { auto result_column = tryGetColumnFromBlock(block, requested_column); diff --git a/src/Interpreters/getColumnFromBlock.h b/src/Interpreters/getColumnFromBlock.h index 26500cfdd17..737ce9db555 100644 --- a/src/Interpreters/getColumnFromBlock.h +++ b/src/Interpreters/getColumnFromBlock.h @@ -9,5 +9,6 @@ namespace DB ColumnPtr getColumnFromBlock(const Block & block, const NameAndTypePair & requested_column); ColumnPtr tryGetColumnFromBlock(const Block & block, const NameAndTypePair & requested_column); +ColumnPtr tryGetSubcolumnFromBlock(const Block & block, const DataTypePtr & requested_column_type, const NameAndTypePair & requested_subcolumn); } diff --git a/src/Interpreters/getCustomKeyFilterForParallelReplicas.cpp b/src/Interpreters/getCustomKeyFilterForParallelReplicas.cpp index 43be7c5f043..6f902c7cd7e 100644 --- a/src/Interpreters/getCustomKeyFilterForParallelReplicas.cpp +++ b/src/Interpreters/getCustomKeyFilterForParallelReplicas.cpp @@ -1,5 +1,7 @@ #include +#include + #include #include #include diff --git a/src/Interpreters/getHeaderForProcessingStage.cpp b/src/Interpreters/getHeaderForProcessingStage.cpp index 06c5d424d2f..cf18cbbb54a 100644 --- a/src/Interpreters/getHeaderForProcessingStage.cpp +++ b/src/Interpreters/getHeaderForProcessingStage.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include diff --git a/src/Interpreters/interpretSubquery.cpp b/src/Interpreters/interpretSubquery.cpp index 6b42345f1d6..340f6d1d805 100644 --- a/src/Interpreters/interpretSubquery.cpp +++ b/src/Interpreters/interpretSubquery.cpp @@ -1,4 +1,5 @@ #include +#include #include #include diff --git a/src/Interpreters/joinDispatch.h b/src/Interpreters/joinDispatch.h index dccbe68fdb6..54c5c7dc83a 100644 --- a/src/Interpreters/joinDispatch.h +++ b/src/Interpreters/joinDispatch.h @@ -3,7 +3,7 @@ #include #include -#include +#include /** Used in implementation of Join to process different data structures. diff --git a/src/Interpreters/loadMetadata.cpp b/src/Interpreters/loadMetadata.cpp index 9c3922b8bda..4d17891f9f8 100644 --- a/src/Interpreters/loadMetadata.cpp +++ b/src/Interpreters/loadMetadata.cpp @@ -24,6 +24,7 @@ #include #include #include +#include #include @@ -384,7 +385,7 @@ static void maybeConvertOrdinaryDatabaseToAtomic(ContextMutablePtr context, cons if (database->getEngineName() != "Ordinary") return; - Strings permanently_detached_tables = database->getNamesOfPermanentlyDetachedTables(); + const Strings permanently_detached_tables = database->getNamesOfPermanentlyDetachedTables(); if (!permanently_detached_tables.empty()) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Cannot automatically convert database {} from Ordinary to Atomic, " diff --git a/src/Interpreters/parseColumnsListForTableFunction.cpp b/src/Interpreters/parseColumnsListForTableFunction.cpp index 3529863a623..b9fdaabede1 100644 --- a/src/Interpreters/parseColumnsListForTableFunction.cpp +++ b/src/Interpreters/parseColumnsListForTableFunction.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -21,6 +22,18 @@ extern const int ILLEGAL_COLUMN; } +DataTypeValidationSettings::DataTypeValidationSettings(const DB::Settings& settings) + : allow_suspicious_low_cardinality_types(settings.allow_suspicious_low_cardinality_types) + , allow_experimental_object_type(settings.allow_experimental_object_type) + , allow_suspicious_fixed_string_types(settings.allow_suspicious_fixed_string_types) + , allow_experimental_variant_type(settings.allow_experimental_variant_type) + , allow_suspicious_variant_types(settings.allow_suspicious_variant_types) + , validate_nested_types(settings.validate_experimental_and_suspicious_types_inside_nested_types) + , allow_experimental_dynamic_type(settings.allow_experimental_dynamic_type) +{ +} + + void validateDataType(const DataTypePtr & type_to_check, const DataTypeValidationSettings & settings) { auto validate_callback = [&](const IDataType & data_type) diff --git a/src/Interpreters/parseColumnsListForTableFunction.h b/src/Interpreters/parseColumnsListForTableFunction.h index e2d2bc97ff7..6f15c585e4f 100644 --- a/src/Interpreters/parseColumnsListForTableFunction.h +++ b/src/Interpreters/parseColumnsListForTableFunction.h @@ -2,28 +2,19 @@ #include #include -#include namespace DB { class Context; +struct Settings; struct DataTypeValidationSettings { DataTypeValidationSettings() = default; - explicit DataTypeValidationSettings(const Settings & settings) - : allow_suspicious_low_cardinality_types(settings.allow_suspicious_low_cardinality_types) - , allow_experimental_object_type(settings.allow_experimental_object_type) - , allow_suspicious_fixed_string_types(settings.allow_suspicious_fixed_string_types) - , allow_experimental_variant_type(settings.allow_experimental_variant_type) - , allow_suspicious_variant_types(settings.allow_suspicious_variant_types) - , validate_nested_types(settings.validate_experimental_and_suspicious_types_inside_nested_types) - , allow_experimental_dynamic_type(settings.allow_experimental_dynamic_type) - { - } + explicit DataTypeValidationSettings(const Settings & settings); bool allow_suspicious_low_cardinality_types = true; bool allow_experimental_object_type = true; diff --git a/src/Interpreters/removeOnClusterClauseIfNeeded.cpp b/src/Interpreters/removeOnClusterClauseIfNeeded.cpp index 44167fe7242..a6595d330fc 100644 --- a/src/Interpreters/removeOnClusterClauseIfNeeded.cpp +++ b/src/Interpreters/removeOnClusterClauseIfNeeded.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -15,6 +16,10 @@ #include #include #include +#include +#include +#include +#include namespace DB @@ -38,6 +43,13 @@ static bool isAccessControlQuery(const ASTPtr & query) || query->as(); } +static bool isNamedCollectionQuery(const ASTPtr & query) +{ + return query->as() + || query->as() + || query->as(); +} + ASTPtr removeOnClusterClauseIfNeeded(const ASTPtr & query, ContextPtr context, const WithoutOnClusterASTRewriteParams & params) { auto * query_on_cluster = dynamic_cast(query.get()); @@ -46,11 +58,14 @@ ASTPtr removeOnClusterClauseIfNeeded(const ASTPtr & query, ContextPtr context, c return query; if ((isUserDefinedFunctionQuery(query) - && context->getSettings().ignore_on_cluster_for_replicated_udf_queries + && context->getSettingsRef().ignore_on_cluster_for_replicated_udf_queries && context->getUserDefinedSQLObjectsStorage().isReplicated()) || (isAccessControlQuery(query) - && context->getSettings().ignore_on_cluster_for_replicated_access_entities_queries - && context->getAccessControl().containsStorage(ReplicatedAccessStorage::STORAGE_TYPE))) + && context->getSettingsRef().ignore_on_cluster_for_replicated_access_entities_queries + && context->getAccessControl().containsStorage(ReplicatedAccessStorage::STORAGE_TYPE)) + || (isNamedCollectionQuery(query) + && context->getSettingsRef().ignore_on_cluster_for_replicated_named_collections_queries + && NamedCollectionFactory::instance().usesReplicatedStorage())) { LOG_DEBUG(getLogger("removeOnClusterClauseIfNeeded"), "ON CLUSTER clause was ignored for query {}", query->getID()); return query_on_cluster->getRewrittenASTWithoutOnCluster(params); diff --git a/src/Interpreters/tests/gtest_convertFieldToType.cpp b/src/Interpreters/tests/gtest_convertFieldToType.cpp index c8a9d5aa2c0..0cac9a3b59d 100644 --- a/src/Interpreters/tests/gtest_convertFieldToType.cpp +++ b/src/Interpreters/tests/gtest_convertFieldToType.cpp @@ -147,7 +147,7 @@ INSTANTIATE_TEST_SUITE_P( DecimalField(DateTime64(123 * Day * 1'000'000), 6) } }) - ); +); INSTANTIATE_TEST_SUITE_P( DateTimeToDateTime64, @@ -179,3 +179,84 @@ INSTANTIATE_TEST_SUITE_P( }, }) ); + +INSTANTIATE_TEST_SUITE_P( + StringToNumber, + ConvertFieldToTypeTest, + ::testing::ValuesIn(std::initializer_list{ + { + "String", + Field("1"), + "Int8", + Field(1) + }, + { + "String", + Field("256"), + "Int8", + Field() + }, + { + "String", + Field("not a number"), + "Int8", + {} + }, + { + "String", + Field("1.1"), + "Int8", + {} /// we can not convert '1.1' to Int8 + }, + { + "String", + Field("1.1"), + "Float64", + Field(1.1) + }, + }) +); + +INSTANTIATE_TEST_SUITE_P( + NumberToString, + ConvertFieldToTypeTest, + ::testing::ValuesIn(std::initializer_list{ + { + "Int8", + Field(1), + "String", + Field("1") + }, + { + "Int8", + Field(-1), + "String", + Field("-1") + }, + { + "Float64", + Field(1.1), + "String", + Field("1.1") + }, + }) +); + +INSTANTIATE_TEST_SUITE_P( + StringToDate, + ConvertFieldToTypeTest, + ::testing::ValuesIn(std::initializer_list{ + { + "String", + Field("2024-07-12"), + "Date", + Field(static_cast(19916)) + }, + { + "String", + Field("not a date"), + "Date", + {} + }, + }) +); diff --git a/src/Loggers/OwnSplitChannel.cpp b/src/Loggers/OwnSplitChannel.cpp index dc51a13e01f..c1594361b2c 100644 --- a/src/Loggers/OwnSplitChannel.cpp +++ b/src/Loggers/OwnSplitChannel.cpp @@ -18,6 +18,9 @@ namespace DB void OwnSplitChannel::log(const Poco::Message & msg) { + if (!isLoggingEnabled()) + return; + #ifndef WITHOUT_TEXT_LOG auto logs_queue = CurrentThread::getInternalTextLogsQueue(); diff --git a/src/Loggers/OwnSplitChannel.h b/src/Loggers/OwnSplitChannel.h index 7ca27cf6584..88bb6b9ce76 100644 --- a/src/Loggers/OwnSplitChannel.h +++ b/src/Loggers/OwnSplitChannel.h @@ -1,6 +1,7 @@ #pragma once #include +#include #include #include #include diff --git a/src/Parsers/ASTCreateQuery.cpp b/src/Parsers/ASTCreateQuery.cpp index 3e5c6a9d86e..f0f782c0a63 100644 --- a/src/Parsers/ASTCreateQuery.cpp +++ b/src/Parsers/ASTCreateQuery.cpp @@ -2,6 +2,8 @@ #include #include #include +#include +#include #include #include #include @@ -240,12 +242,12 @@ ASTPtr ASTCreateQuery::clone() const res->set(res->columns_list, columns_list->clone()); if (storage) res->set(res->storage, storage->clone()); - if (inner_storage) - res->set(res->inner_storage, inner_storage->clone()); if (select) res->set(res->select, select->clone()); if (table_overrides) res->set(res->table_overrides, table_overrides->clone()); + if (targets) + res->set(res->targets, targets->clone()); if (dictionary) { @@ -265,6 +267,16 @@ ASTPtr ASTCreateQuery::clone() const return res; } +String ASTCreateQuery::getID(char delim) const +{ + String res = attach ? "AttachQuery" : "CreateQuery"; + String database = getDatabase(); + if (!database.empty()) + res += (delim + getDatabase()); + res += (delim + getTable()); + return res; +} + void ASTCreateQuery::formatQueryImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const { frame.need_parens = false; @@ -388,24 +400,32 @@ void ASTCreateQuery::formatQueryImpl(const FormatSettings & settings, FormatStat refresh_strategy->formatImpl(settings, state, frame); } - if (to_table_id) + if (auto to_table_id = getTargetTableID(ViewTarget::To)) { - assert((is_materialized_view || is_window_view) && to_inner_uuid == UUIDHelpers::Nil); - settings.ostr - << (settings.hilite ? hilite_keyword : "") << " TO " << (settings.hilite ? hilite_none : "") - << (!to_table_id.database_name.empty() ? backQuoteIfNeed(to_table_id.database_name) + "." : "") - << backQuoteIfNeed(to_table_id.table_name); + settings.ostr << " " << (settings.hilite ? hilite_keyword : "") << toStringView(Keyword::TO) + << (settings.hilite ? hilite_none : "") << " " + << (!to_table_id.database_name.empty() ? backQuoteIfNeed(to_table_id.database_name) + "." : "") + << backQuoteIfNeed(to_table_id.table_name); } - if (to_inner_uuid != UUIDHelpers::Nil) + if (auto to_inner_uuid = getTargetInnerUUID(ViewTarget::To); to_inner_uuid != UUIDHelpers::Nil) { - assert(is_materialized_view && !to_table_id); - settings.ostr << (settings.hilite ? hilite_keyword : "") << " TO INNER UUID " << (settings.hilite ? hilite_none : "") - << quoteString(toString(to_inner_uuid)); + settings.ostr << " " << (settings.hilite ? hilite_keyword : "") << toStringView(Keyword::TO_INNER_UUID) + << (settings.hilite ? hilite_none : "") << " " << quoteString(toString(to_inner_uuid)); } + bool should_add_empty = is_create_empty; + auto add_empty_if_needed = [&] + { + if (!should_add_empty) + return; + should_add_empty = false; + settings.ostr << (settings.hilite ? hilite_keyword : "") << " EMPTY" << (settings.hilite ? hilite_none : ""); + }; + if (!as_table.empty()) { + add_empty_if_needed(); settings.ostr << (settings.hilite ? hilite_keyword : "") << " AS " << (settings.hilite ? hilite_none : "") << (!as_database.empty() ? backQuoteIfNeed(as_database) + "." : "") << backQuoteIfNeed(as_table); @@ -423,6 +443,7 @@ void ASTCreateQuery::formatQueryImpl(const FormatSettings & settings, FormatStat frame.expression_list_always_start_on_new_line = false; } + add_empty_if_needed(); settings.ostr << (settings.hilite ? hilite_keyword : "") << " AS " << (settings.hilite ? hilite_none : ""); as_table_function->formatImpl(settings, state, frame); } @@ -450,14 +471,17 @@ void ASTCreateQuery::formatQueryImpl(const FormatSettings & settings, FormatStat frame.expression_list_always_start_on_new_line = false; - if (inner_storage) + if (storage) + storage->formatImpl(settings, state, frame); + + if (auto inner_storage = getTargetInnerEngine(ViewTarget::Inner)) { - settings.ostr << (settings.hilite ? hilite_keyword : "") << " INNER" << (settings.hilite ? hilite_none : ""); + settings.ostr << " " << (settings.hilite ? hilite_keyword : "") << toStringView(Keyword::INNER) << (settings.hilite ? hilite_none : ""); inner_storage->formatImpl(settings, state, frame); } - if (storage) - storage->formatImpl(settings, state, frame); + if (auto to_storage = getTargetInnerEngine(ViewTarget::To)) + to_storage->formatImpl(settings, state, frame); if (dictionary) dictionary->formatImpl(settings, state, frame); @@ -484,8 +508,8 @@ void ASTCreateQuery::formatQueryImpl(const FormatSettings & settings, FormatStat if (is_populate) settings.ostr << (settings.hilite ? hilite_keyword : "") << " POPULATE" << (settings.hilite ? hilite_none : ""); - else if (is_create_empty) - settings.ostr << (settings.hilite ? hilite_keyword : "") << " EMPTY" << (settings.hilite ? hilite_none : ""); + + add_empty_if_needed(); if (sql_security && supportSQLSecurity() && sql_security->as().type.has_value()) { @@ -517,48 +541,57 @@ bool ASTCreateQuery::isParameterizedView() const } -ASTCreateQuery::UUIDs::UUIDs(const ASTCreateQuery & query) - : uuid(query.uuid) - , to_inner_uuid(query.to_inner_uuid) +void ASTCreateQuery::generateRandomUUIDs() { + CreateQueryUUIDs{*this, /* generate_random= */ true}.copyToQuery(*this); } -String ASTCreateQuery::UUIDs::toString() const +void ASTCreateQuery::resetUUIDs() { - WriteBufferFromOwnString out; - out << "{" << uuid << "," << to_inner_uuid << "}"; - return out.str(); + CreateQueryUUIDs{}.copyToQuery(*this); } -ASTCreateQuery::UUIDs ASTCreateQuery::UUIDs::fromString(const String & str) + +StorageID ASTCreateQuery::getTargetTableID(ViewTarget::Kind target_kind) const { - ReadBufferFromString in{str}; - ASTCreateQuery::UUIDs res; - in >> "{" >> res.uuid >> "," >> res.to_inner_uuid >> "}"; - return res; + if (targets) + return targets->getTableID(target_kind); + return StorageID::createEmpty(); } -ASTCreateQuery::UUIDs ASTCreateQuery::generateRandomUUID(bool always_generate_new_uuid) +bool ASTCreateQuery::hasTargetTableID(ViewTarget::Kind target_kind) const { - if (always_generate_new_uuid) - setUUID({}); - - if (uuid == UUIDHelpers::Nil) - uuid = UUIDHelpers::generateV4(); - - /// If destination table (to_table_id) is not specified for materialized view, - /// then MV will create inner table. We should generate UUID of inner table here. - bool need_uuid_for_inner_table = !attach && is_materialized_view && !to_table_id; - if (need_uuid_for_inner_table && (to_inner_uuid == UUIDHelpers::Nil)) - to_inner_uuid = UUIDHelpers::generateV4(); - - return UUIDs{*this}; + if (targets) + return targets->hasTableID(target_kind); + return false; } -void ASTCreateQuery::setUUID(const UUIDs & uuids) +UUID ASTCreateQuery::getTargetInnerUUID(ViewTarget::Kind target_kind) const { - uuid = uuids.uuid; - to_inner_uuid = uuids.to_inner_uuid; + if (targets) + return targets->getInnerUUID(target_kind); + return UUIDHelpers::Nil; +} + +bool ASTCreateQuery::hasInnerUUIDs() const +{ + if (targets) + return targets->hasInnerUUIDs(); + return false; +} + +std::shared_ptr ASTCreateQuery::getTargetInnerEngine(ViewTarget::Kind target_kind) const +{ + if (targets) + return targets->getInnerEngine(target_kind); + return nullptr; +} + +void ASTCreateQuery::setTargetInnerEngine(ViewTarget::Kind target_kind, ASTPtr storage_def) +{ + if (!targets) + set(targets, std::make_shared()); + targets->setInnerEngine(target_kind, storage_def); } } diff --git a/src/Parsers/ASTCreateQuery.h b/src/Parsers/ASTCreateQuery.h index 6fbf045915b..a95010aea31 100644 --- a/src/Parsers/ASTCreateQuery.h +++ b/src/Parsers/ASTCreateQuery.h @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -15,6 +16,7 @@ namespace DB class ASTFunction; class ASTSetQuery; class ASTSelectWithUnionQuery; +struct CreateQueryUUIDs; class ASTStorage : public IAST @@ -101,17 +103,15 @@ public: bool has_uuid{false}; // CREATE TABLE x UUID '...' ASTColumns * columns_list = nullptr; - - StorageID to_table_id = StorageID::createEmpty(); /// For CREATE MATERIALIZED VIEW mv TO table. - UUID to_inner_uuid = UUIDHelpers::Nil; /// For materialized view with inner table - ASTStorage * inner_storage = nullptr; /// For window view with inner table ASTStorage * storage = nullptr; + ASTPtr watermark_function; ASTPtr lateness_function; String as_database; String as_table; IAST * as_table_function = nullptr; ASTSelectWithUnionQuery * select = nullptr; + ASTViewTargets * targets = nullptr; IAST * comment = nullptr; ASTPtr sql_security = nullptr; @@ -136,7 +136,7 @@ public: bool create_or_replace{false}; /** Get the text that identifies this element. */ - String getID(char delim) const override { return (attach ? "AttachQuery" : "CreateQuery") + (delim + getDatabase()) + delim + getTable(); } + String getID(char delim) const override; ASTPtr clone() const override; @@ -153,17 +153,26 @@ public: QueryKind getQueryKind() const override { return QueryKind::Create; } - struct UUIDs - { - UUID uuid = UUIDHelpers::Nil; - UUID to_inner_uuid = UUIDHelpers::Nil; - UUIDs() = default; - explicit UUIDs(const ASTCreateQuery & query); - String toString() const; - static UUIDs fromString(const String & str); - }; - UUIDs generateRandomUUID(bool always_generate_new_uuid = false); - void setUUID(const UUIDs & uuids); + /// Generates a random UUID for this create query if it's not specified already. + /// The function also generates random UUIDs for inner target tables if this create query implies that + /// (for example, if it's a `CREATE MATERIALIZED VIEW` query with an inner storage). + void generateRandomUUIDs(); + + /// Removes UUID from this create query. + /// The function also removes UUIDs for inner target tables from this create query (see also generateRandomUUID()). + void resetUUIDs(); + + /// Returns information about a target table. + /// If that information isn't specified in this create query (or even not allowed) then the function returns an empty value. + StorageID getTargetTableID(ViewTarget::Kind target_kind) const; + bool hasTargetTableID(ViewTarget::Kind target_kind) const; + UUID getTargetInnerUUID(ViewTarget::Kind target_kind) const; + bool hasInnerUUIDs() const; + std::shared_ptr getTargetInnerEngine(ViewTarget::Kind target_kind) const; + void setTargetInnerEngine(ViewTarget::Kind target_kind, ASTPtr storage_def); + + bool is_materialized_view_with_external_target() const { return is_materialized_view && hasTargetTableID(ViewTarget::To); } + bool is_materialized_view_with_inner_table() const { return is_materialized_view && !hasTargetTableID(ViewTarget::To); } protected: void formatQueryImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const override; @@ -171,8 +180,8 @@ protected: void forEachPointerToChild(std::function f) override { f(reinterpret_cast(&columns_list)); - f(reinterpret_cast(&inner_storage)); f(reinterpret_cast(&storage)); + f(reinterpret_cast(&targets)); f(reinterpret_cast(&as_table_function)); f(reinterpret_cast(&select)); f(reinterpret_cast(&comment)); diff --git a/src/Parsers/ASTFunction.cpp b/src/Parsers/ASTFunction.cpp index 21c845a6e13..efa69078f7b 100644 --- a/src/Parsers/ASTFunction.cpp +++ b/src/Parsers/ASTFunction.cpp @@ -7,7 +7,6 @@ #include #include #include -#include #include #include #include @@ -20,9 +19,6 @@ #include #include #include -#include - -#include using namespace std::literals; @@ -334,19 +330,23 @@ void ASTFunction::formatImplWithoutAlias(const FormatSettings & settings, Format const auto * literal = arguments->children[0]->as(); const auto * function = arguments->children[0]->as(); + const auto * subquery = arguments->children[0]->as(); bool is_tuple = literal && literal->value.getType() == Field::Types::Tuple; - // do not add parentheses for tuple literal, otherwise extra parens will be added `-((3, 7, 3), 1)` -> `-(((3, 7, 3), 1))` + /// Do not add parentheses for tuple literal, otherwise extra parens will be added `-((3, 7, 3), 1)` -> `-(((3, 7, 3), 1))` bool literal_need_parens = literal && !is_tuple; - // negate always requires parentheses, otherwise -(-1) will be printed as --1 - bool inside_parens = name == "negate" && (literal_need_parens || (function && function->name == "negate")); + /// Negate always requires parentheses, otherwise -(-1) will be printed as --1 + /// Also extra parentheses are needed for subqueries, because NOT can be parsed as a function: + /// not(SELECT 1) cannot be parsed, while not((SELECT 1)) can. + bool inside_parens = (name == "negate" && (literal_need_parens || (function && function->name == "negate"))) + || (subquery && name == "not"); /// We DO need parentheses around a single literal /// For example, SELECT (NOT 0) + (NOT 0) cannot be transformed into SELECT NOT 0 + NOT 0, since /// this is equal to SELECT NOT (0 + NOT 0) bool outside_parens = frame.need_parens && !inside_parens; - // do not add extra parentheses for functions inside negate, i.e. -(-toUInt64(-(1))) + /// Do not add extra parentheses for functions inside negate, i.e. -(-toUInt64(-(1))) if (inside_parens) nested_need_parens.need_parens = false; @@ -409,25 +409,26 @@ void ASTFunction::formatImplWithoutAlias(const FormatSettings & settings, Format { const char * operators[] = { - "multiply", " * ", - "divide", " / ", - "modulo", " % ", - "plus", " + ", - "minus", " - ", - "notEquals", " != ", - "lessOrEquals", " <= ", - "greaterOrEquals", " >= ", - "less", " < ", - "greater", " > ", - "equals", " = ", - "like", " LIKE ", - "ilike", " ILIKE ", - "notLike", " NOT LIKE ", - "notILike", " NOT ILIKE ", - "in", " IN ", - "notIn", " NOT IN ", - "globalIn", " GLOBAL IN ", - "globalNotIn", " GLOBAL NOT IN ", + "multiply", " * ", + "divide", " / ", + "modulo", " % ", + "plus", " + ", + "minus", " - ", + "notEquals", " != ", + "lessOrEquals", " <= ", + "greaterOrEquals", " >= ", + "less", " < ", + "greater", " > ", + "equals", " = ", + "isNotDistinctFrom", " <=> ", + "like", " LIKE ", + "ilike", " ILIKE ", + "notLike", " NOT LIKE ", + "notILike", " NOT ILIKE ", + "in", " IN ", + "notIn", " NOT IN ", + "globalIn", " GLOBAL IN ", + "globalNotIn", " GLOBAL NOT IN ", nullptr }; @@ -632,6 +633,7 @@ void ASTFunction::formatImplWithoutAlias(const FormatSettings & settings, Format settings.ostr << ", "; if (arguments->children[i]->as()) settings.ostr << "SETTINGS "; + nested_dont_need_parens.list_element_index = i; arguments->children[i]->formatImpl(settings, state, nested_dont_need_parens); } settings.ostr << (settings.hilite ? hilite_operator : "") << ']' << (settings.hilite ? hilite_none : ""); @@ -642,12 +644,14 @@ void ASTFunction::formatImplWithoutAlias(const FormatSettings & settings, Format { settings.ostr << (settings.hilite ? hilite_operator : "") << ((frame.need_parens && !alias.empty()) ? "tuple" : "") << '(' << (settings.hilite ? hilite_none : ""); + for (size_t i = 0; i < arguments->children.size(); ++i) { if (i != 0) settings.ostr << ", "; if (arguments->children[i]->as()) settings.ostr << "SETTINGS "; + nested_dont_need_parens.list_element_index = i; arguments->children[i]->formatImpl(settings, state, nested_dont_need_parens); } settings.ostr << (settings.hilite ? hilite_operator : "") << ')' << (settings.hilite ? hilite_none : ""); @@ -663,6 +667,7 @@ void ASTFunction::formatImplWithoutAlias(const FormatSettings & settings, Format settings.ostr << ", "; if (arguments->children[i]->as()) settings.ostr << "SETTINGS "; + nested_dont_need_parens.list_element_index = i; arguments->children[i]->formatImpl(settings, state, nested_dont_need_parens); } settings.ostr << (settings.hilite ? hilite_operator : "") << ')' << (settings.hilite ? hilite_none : ""); diff --git a/src/Parsers/ASTTablesInSelectQuery.cpp b/src/Parsers/ASTTablesInSelectQuery.cpp index e782bad797e..b6d42513aa7 100644 --- a/src/Parsers/ASTTablesInSelectQuery.cpp +++ b/src/Parsers/ASTTablesInSelectQuery.cpp @@ -235,7 +235,13 @@ void ASTTableJoin::formatImplAfterTable(const FormatSettings & settings, FormatS else if (on_expression) { settings.ostr << (settings.hilite ? hilite_keyword : "") << " ON " << (settings.hilite ? hilite_none : ""); + /// If there is an alias for the whole expression parens should be added, otherwise it will be invalid syntax + bool on_has_alias = !on_expression->tryGetAlias().empty(); + if (on_has_alias) + settings.ostr << "("; on_expression->formatImpl(settings, state, frame); + if (on_has_alias) + settings.ostr << ")"; } } @@ -243,7 +249,7 @@ void ASTTableJoin::formatImplAfterTable(const FormatSettings & settings, FormatS void ASTTableJoin::formatImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const { formatImplBeforeTable(settings, state, frame); - settings.ostr << " ... "; + settings.ostr << " ..."; formatImplAfterTable(settings, state, frame); } diff --git a/src/Parsers/ASTViewTargets.cpp b/src/Parsers/ASTViewTargets.cpp new file mode 100644 index 00000000000..8ee98e704df --- /dev/null +++ b/src/Parsers/ASTViewTargets.cpp @@ -0,0 +1,300 @@ +#include + +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; + extern const int LOGICAL_ERROR; +} + + +std::string_view toString(ViewTarget::Kind kind) +{ + switch (kind) + { + case ViewTarget::To: return "to"; + case ViewTarget::Inner: return "inner"; + } + throw Exception(ErrorCodes::LOGICAL_ERROR, "{} doesn't support kind {}", __FUNCTION__, kind); +} + +void parseFromString(ViewTarget::Kind & out, std::string_view str) +{ + for (auto kind : magic_enum::enum_values()) + { + if (toString(kind) == str) + { + out = kind; + return; + } + } + throw Exception(ErrorCodes::BAD_ARGUMENTS, "{}: Unexpected string {}", __FUNCTION__, str); +} + + +std::vector ASTViewTargets::getKinds() const +{ + std::vector kinds; + kinds.reserve(targets.size()); + for (const auto & target : targets) + kinds.push_back(target.kind); + return kinds; +} + + +void ASTViewTargets::setTableID(ViewTarget::Kind kind, const StorageID & table_id_) +{ + for (auto & target : targets) + { + if (target.kind == kind) + { + target.table_id = table_id_; + return; + } + } + if (table_id_) + targets.emplace_back(kind).table_id = table_id_; +} + +StorageID ASTViewTargets::getTableID(ViewTarget::Kind kind) const +{ + if (const auto * target = tryGetTarget(kind)) + return target->table_id; + return StorageID::createEmpty(); +} + +bool ASTViewTargets::hasTableID(ViewTarget::Kind kind) const +{ + if (const auto * target = tryGetTarget(kind)) + return !target->table_id.empty(); + return false; +} + +void ASTViewTargets::setCurrentDatabase(const String & current_database) +{ + for (auto & target : targets) + { + auto & table_id = target.table_id; + if (!table_id.table_name.empty() && table_id.database_name.empty()) + table_id.database_name = current_database; + } +} + +void ASTViewTargets::setInnerUUID(ViewTarget::Kind kind, const UUID & inner_uuid_) +{ + for (auto & target : targets) + { + if (target.kind == kind) + { + target.inner_uuid = inner_uuid_; + return; + } + } + if (inner_uuid_ != UUIDHelpers::Nil) + targets.emplace_back(kind).inner_uuid = inner_uuid_; +} + +UUID ASTViewTargets::getInnerUUID(ViewTarget::Kind kind) const +{ + if (const auto * target = tryGetTarget(kind)) + return target->inner_uuid; + return UUIDHelpers::Nil; +} + +bool ASTViewTargets::hasInnerUUID(ViewTarget::Kind kind) const +{ + return getInnerUUID(kind) != UUIDHelpers::Nil; +} + +void ASTViewTargets::resetInnerUUIDs() +{ + for (auto & target : targets) + target.inner_uuid = UUIDHelpers::Nil; +} + +bool ASTViewTargets::hasInnerUUIDs() const +{ + for (const auto & target : targets) + { + if (target.inner_uuid != UUIDHelpers::Nil) + return true; + } + return false; +} + +void ASTViewTargets::setInnerEngine(ViewTarget::Kind kind, ASTPtr storage_def) +{ + auto new_inner_engine = typeid_cast>(storage_def); + if (!new_inner_engine && storage_def) + throw Exception(DB::ErrorCodes::LOGICAL_ERROR, "Bad cast from type {} to ASTStorage", storage_def->getID()); + + for (auto & target : targets) + { + if (target.kind == kind) + { + if (target.inner_engine == new_inner_engine) + return; + if (new_inner_engine) + children.push_back(new_inner_engine); + if (target.inner_engine) + std::erase(children, target.inner_engine); + target.inner_engine = new_inner_engine; + return; + } + } + + if (new_inner_engine) + { + targets.emplace_back(kind).inner_engine = new_inner_engine; + children.push_back(new_inner_engine); + } +} + +std::shared_ptr ASTViewTargets::getInnerEngine(ViewTarget::Kind kind) const +{ + if (const auto * target = tryGetTarget(kind)) + return target->inner_engine; + return nullptr; +} + +std::vector> ASTViewTargets::getInnerEngines() const +{ + std::vector> res; + res.reserve(targets.size()); + for (const auto & target : targets) + { + if (target.inner_engine) + res.push_back(target.inner_engine); + } + return res; +} + +const ViewTarget * ASTViewTargets::tryGetTarget(ViewTarget::Kind kind) const +{ + for (const auto & target : targets) + { + if (target.kind == kind) + return ⌖ + } + return nullptr; +} + +ASTPtr ASTViewTargets::clone() const +{ + auto res = std::make_shared(*this); + res->children.clear(); + for (auto & target : res->targets) + { + if (target.inner_engine) + { + target.inner_engine = typeid_cast>(target.inner_engine->clone()); + res->children.push_back(target.inner_engine); + } + } + return res; +} + +void ASTViewTargets::formatImpl(const FormatSettings & s, FormatState & state, FormatStateStacked frame) const +{ + for (const auto & target : targets) + formatTarget(target, s, state, frame); +} + +void ASTViewTargets::formatTarget(ViewTarget::Kind kind, const FormatSettings & s, FormatState & state, FormatStateStacked frame) const +{ + for (const auto & target : targets) + { + if (target.kind == kind) + formatTarget(target, s, state, frame); + } +} + +void ASTViewTargets::formatTarget(const ViewTarget & target, const FormatSettings & s, FormatState & state, FormatStateStacked frame) +{ + if (target.table_id) + { + auto keyword = getKeywordForTableID(target.kind); + if (!keyword) + throw Exception(ErrorCodes::LOGICAL_ERROR, "No keyword for table name of kind {}", toString(target.kind)); + s.ostr << " " << (s.hilite ? hilite_keyword : "") << toStringView(*keyword) + << (s.hilite ? hilite_none : "") << " " + << (!target.table_id.database_name.empty() ? backQuoteIfNeed(target.table_id.database_name) + "." : "") + << backQuoteIfNeed(target.table_id.table_name); + } + + if (target.inner_uuid != UUIDHelpers::Nil) + { + auto keyword = getKeywordForInnerUUID(target.kind); + if (!keyword) + throw Exception(ErrorCodes::LOGICAL_ERROR, "No prefix keyword for inner UUID of kind {}", toString(target.kind)); + s.ostr << " " << (s.hilite ? hilite_keyword : "") << toStringView(*keyword) + << (s.hilite ? hilite_none : "") << " " << quoteString(toString(target.inner_uuid)); + } + + if (target.inner_engine) + { + auto keyword = getKeywordForInnerStorage(target.kind); + if (!keyword) + throw Exception(ErrorCodes::LOGICAL_ERROR, "No prefix keyword for table engine of kind {}", toString(target.kind)); + s.ostr << " " << (s.hilite ? hilite_keyword : "") << toStringView(*keyword) << (s.hilite ? hilite_none : ""); + target.inner_engine->formatImpl(s, state, frame); + } +} + +std::optional ASTViewTargets::getKeywordForTableID(ViewTarget::Kind kind) +{ + switch (kind) + { + case ViewTarget::To: return Keyword::TO; /// TO mydb.mydata + case ViewTarget::Inner: return std::nullopt; + } + UNREACHABLE(); +} + +std::optional ASTViewTargets::getKeywordForInnerStorage(ViewTarget::Kind kind) +{ + switch (kind) + { + case ViewTarget::To: return std::nullopt; /// ENGINE = MergeTree() + case ViewTarget::Inner: return Keyword::INNER; /// INNER ENGINE = MergeTree() + } + UNREACHABLE(); +} + +std::optional ASTViewTargets::getKeywordForInnerUUID(ViewTarget::Kind kind) +{ + switch (kind) + { + case ViewTarget::To: return Keyword::TO_INNER_UUID; /// TO INNER UUID 'XXX' + case ViewTarget::Inner: return std::nullopt; + } + UNREACHABLE(); +} + +void ASTViewTargets::forEachPointerToChild(std::function f) +{ + for (auto & target : targets) + { + if (target.inner_engine) + { + ASTStorage * new_inner_engine = target.inner_engine.get(); + f(reinterpret_cast(&new_inner_engine)); + if (new_inner_engine != target.inner_engine.get()) + { + if (new_inner_engine) + target.inner_engine = typeid_cast>(new_inner_engine->ptr()); + else + target.inner_engine.reset(); + } + } + } +} + +} diff --git a/src/Parsers/ASTViewTargets.h b/src/Parsers/ASTViewTargets.h new file mode 100644 index 00000000000..12182919f0e --- /dev/null +++ b/src/Parsers/ASTViewTargets.h @@ -0,0 +1,115 @@ +#pragma once + +#include +#include + + +namespace DB +{ +class ASTStorage; +enum class Keyword : size_t; + +/// Information about target tables (external or inner) of a materialized view or a window view. +/// See ASTViewTargets for more details. +struct ViewTarget +{ + enum Kind + { + /// If `kind == ViewTarget::To` then `ViewTarget` contains information about the "TO" table of a materialized view or a window view: + /// CREATE MATERIALIZED VIEW db.mv_name {TO [db.]to_target | ENGINE to_engine} AS SELECT ... + /// or + /// CREATE WINDOW VIEW db.wv_name {TO [db.]to_target | ENGINE to_engine} AS SELECT ... + To, + + /// If `kind == ViewTarget::Inner` then `ViewTarget` contains information about the "INNER" table of a window view: + /// CREATE WINDOW VIEW db.wv_name {INNER ENGINE inner_engine} AS SELECT ... + Inner, + }; + + Kind kind = To; + + /// StorageID of the target table, if it's not inner. + /// That storage ID can be seen for example after "TO" in a statement like CREATE MATERIALIZED VIEW ... TO ... + StorageID table_id = StorageID::createEmpty(); + + /// UUID of the target table, if it's inner. + /// The UUID is calculated automatically and can be seen for example after "TO INNER UUID" in a statement like + /// CREATE MATERIALIZED VIEW ... TO INNER UUID ... + UUID inner_uuid = UUIDHelpers::Nil; + + /// Table engine of the target table, if it's inner. + /// That engine can be seen for example after "ENGINE" in a statement like CREATE MATERIALIZED VIEW ... ENGINE ... + std::shared_ptr inner_engine; +}; + +/// Converts ViewTarget::Kind to a string. +std::string_view toString(ViewTarget::Kind kind); +void parseFromString(ViewTarget::Kind & out, std::string_view str); + + +/// Information about all target tables (external or inner) of a view. +/// +/// For example, for a materialized view: +/// CREATE MATERIALIZED VIEW db.mv_name [TO [db.]to_target | ENGINE to_engine] AS SELECT ... +/// this class contains information about the "TO" table: its name and database (if it's external), its UUID and engine (if it's inner). +/// +/// For a window view: +/// CREATE WINDOW VIEW db.wv_name [TO [db.]to_target | ENGINE to_engine] [INNER ENGINE inner_engine] AS SELECT ... +/// this class contains information about both the "TO" table and the "INNER" table. +class ASTViewTargets : public IAST +{ +public: + std::vector targets; + + /// Sets the StorageID of the target table, if it's not inner. + /// That storage ID can be seen for example after "TO" in a statement like CREATE MATERIALIZED VIEW ... TO ... + void setTableID(ViewTarget::Kind kind, const StorageID & table_id_); + StorageID getTableID(ViewTarget::Kind kind) const; + bool hasTableID(ViewTarget::Kind kind) const; + + /// Replaces an empty database in the StorageID of the target table with a specified database. + void setCurrentDatabase(const String & current_database); + + /// Sets the UUID of the target table, if it's inner. + /// The UUID is calculated automatically and can be seen for example after "TO INNER UUID" in a statement like + /// CREATE MATERIALIZED VIEW ... TO INNER UUID ... + void setInnerUUID(ViewTarget::Kind kind, const UUID & inner_uuid_); + UUID getInnerUUID(ViewTarget::Kind kind) const; + bool hasInnerUUID(ViewTarget::Kind kind) const; + + void resetInnerUUIDs(); + bool hasInnerUUIDs() const; + + /// Sets the table engine of the target table, if it's inner. + /// That engine can be seen for example after "ENGINE" in a statement like CREATE MATERIALIZED VIEW ... ENGINE ... + void setInnerEngine(ViewTarget::Kind kind, ASTPtr storage_def); + std::shared_ptr getInnerEngine(ViewTarget::Kind kind) const; + std::vector> getInnerEngines() const; + + /// Returns a list of all kinds of views in this ASTViewTargets. + std::vector getKinds() const; + + /// Returns information about a target table. + /// The function returns null if such target doesn't exist. + const ViewTarget * tryGetTarget(ViewTarget::Kind kind) const; + + String getID(char) const override { return "ViewTargets"; } + + ASTPtr clone() const override; + + void formatImpl(const FormatSettings & s, FormatState & state, FormatStateStacked frame) const override; + + /// Formats information only about a specific target table. + void formatTarget(ViewTarget::Kind kind, const FormatSettings & s, FormatState & state, FormatStateStacked frame) const; + static void formatTarget(const ViewTarget & target, const FormatSettings & s, FormatState & state, FormatStateStacked frame); + + /// Helper functions for class ParserViewTargets. Returns a prefix keyword matching a specified target kind. + static std::optional getKeywordForTableID(ViewTarget::Kind kind); + static std::optional getKeywordForInnerUUID(ViewTarget::Kind kind); + static std::optional getKeywordForInnerStorage(ViewTarget::Kind kind); + +protected: + void forEachPointerToChild(std::function f) override; +}; + +} diff --git a/src/Parsers/Access/ASTAuthenticationData.cpp b/src/Parsers/Access/ASTAuthenticationData.cpp index dab93ba3de5..386ed900960 100644 --- a/src/Parsers/Access/ASTAuthenticationData.cpp +++ b/src/Parsers/Access/ASTAuthenticationData.cpp @@ -112,7 +112,7 @@ void ASTAuthenticationData::formatImpl(const FormatSettings & settings, FormatSt } case AuthenticationType::SSL_CERTIFICATE: { - prefix = "CN"; + prefix = ssl_cert_subject_type.value(); parameters = true; break; } diff --git a/src/Parsers/Access/ASTAuthenticationData.h b/src/Parsers/Access/ASTAuthenticationData.h index de166bdf234..7f0644b3437 100644 --- a/src/Parsers/Access/ASTAuthenticationData.h +++ b/src/Parsers/Access/ASTAuthenticationData.h @@ -33,6 +33,7 @@ public: std::optional getPassword() const; std::optional getSalt() const; + std::optional ssl_cert_subject_type; /// CN or SubjectAltName /// If type is empty we use the default password type. /// AuthenticationType::NO_PASSWORD is specified explicitly. diff --git a/src/Parsers/Access/ASTCreateUserQuery.cpp b/src/Parsers/Access/ASTCreateUserQuery.cpp index 02735568a04..6f0ccc76797 100644 --- a/src/Parsers/Access/ASTCreateUserQuery.cpp +++ b/src/Parsers/Access/ASTCreateUserQuery.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include diff --git a/src/Parsers/Access/ParserCreateUserQuery.cpp b/src/Parsers/Access/ParserCreateUserQuery.cpp index d4729ab796a..d4a8813e9e4 100644 --- a/src/Parsers/Access/ParserCreateUserQuery.cpp +++ b/src/Parsers/Access/ParserCreateUserQuery.cpp @@ -20,7 +20,6 @@ #include #include #include - #include "config.h" namespace DB @@ -65,7 +64,7 @@ namespace bool expect_hash = false; bool expect_ldap_server_name = false; bool expect_kerberos_realm = false; - bool expect_common_names = false; + bool expect_ssl_cert_subjects = false; bool expect_public_ssh_key = false; bool expect_http_auth_server = false; @@ -82,7 +81,7 @@ namespace else if (check_type == AuthenticationType::KERBEROS) expect_kerberos_realm = true; else if (check_type == AuthenticationType::SSL_CERTIFICATE) - expect_common_names = true; + expect_ssl_cert_subjects = true; else if (check_type == AuthenticationType::SSH_KEY) expect_public_ssh_key = true; else if (check_type == AuthenticationType::HTTP) @@ -122,9 +121,10 @@ namespace ASTPtr value; ASTPtr parsed_salt; - ASTPtr common_names; ASTPtr public_ssh_keys; ASTPtr http_auth_scheme; + ASTPtr ssl_cert_subjects; + std::optional ssl_cert_subject_type; if (expect_password || expect_hash) { @@ -153,12 +153,19 @@ namespace return false; } } - else if (expect_common_names) + else if (expect_ssl_cert_subjects) { - if (!ParserKeyword{Keyword::CN}.ignore(pos, expected)) + for (const Keyword &keyword : {Keyword::CN, Keyword::SAN}) + if (ParserKeyword{keyword}.ignore(pos, expected)) + { + ssl_cert_subject_type = toStringView(keyword); + break; + } + + if (!ssl_cert_subject_type) return false; - if (!ParserList{std::make_unique(), std::make_unique(TokenType::Comma), false}.parse(pos, common_names, expected)) + if (!ParserList{std::make_unique(), std::make_unique(TokenType::Comma), false}.parse(pos, ssl_cert_subjects, expected)) return false; } else if (expect_public_ssh_key) @@ -166,7 +173,7 @@ namespace if (!ParserKeyword{Keyword::BY}.ignore(pos, expected)) return false; - if (!ParserList{std::make_unique(), std::make_unique(TokenType::Comma), false}.parse(pos, common_names, expected)) + if (!ParserList{std::make_unique(), std::make_unique(TokenType::Comma), false}.parse(pos, public_ssh_keys, expected)) return false; } else if (expect_http_auth_server) @@ -195,8 +202,11 @@ namespace if (parsed_salt) auth_data->children.push_back(std::move(parsed_salt)); - if (common_names) - auth_data->children = std::move(common_names->children); + if (ssl_cert_subjects) + { + auth_data->ssl_cert_subject_type = ssl_cert_subject_type.value(); + auth_data->children = std::move(ssl_cert_subjects->children); + } if (public_ssh_keys) auth_data->children = std::move(public_ssh_keys->children); diff --git a/src/Parsers/CommonParsers.h b/src/Parsers/CommonParsers.h index 93642d94880..0ae9ee4833c 100644 --- a/src/Parsers/CommonParsers.h +++ b/src/Parsers/CommonParsers.h @@ -418,6 +418,7 @@ namespace DB MR_MACROS(SALT, "SALT") \ MR_MACROS(SAMPLE_BY, "SAMPLE BY") \ MR_MACROS(SAMPLE, "SAMPLE") \ + MR_MACROS(SAN, "SAN") \ MR_MACROS(SCHEME, "SCHEME") \ MR_MACROS(SECOND, "SECOND") \ MR_MACROS(SECONDS, "SECONDS") \ diff --git a/src/Parsers/CreateQueryUUIDs.cpp b/src/Parsers/CreateQueryUUIDs.cpp new file mode 100644 index 00000000000..4dfee67b537 --- /dev/null +++ b/src/Parsers/CreateQueryUUIDs.cpp @@ -0,0 +1,168 @@ +#include + +#include +#include +#include +#include + + +namespace DB +{ + +CreateQueryUUIDs::CreateQueryUUIDs(const ASTCreateQuery & query, bool generate_random, bool force_random) +{ + if (!generate_random || !force_random) + { + uuid = query.uuid; + if (query.targets) + { + for (const auto & target : query.targets->targets) + setTargetInnerUUID(target.kind, target.inner_uuid); + } + } + + if (generate_random) + { + if (uuid == UUIDHelpers::Nil) + uuid = UUIDHelpers::generateV4(); + + /// For an ATTACH query we should never generate UUIDs for its inner target tables + /// because for an ATTACH query those inner target tables probably already exist and can be accessible by names. + /// If we generate random UUIDs for already existing tables then those UUIDs will not be correct making those inner target table inaccessible. + /// Thus it's not safe for example to replace + /// "ATTACH MATERIALIZED VIEW mv AS SELECT a FROM b" with + /// "ATTACH MATERIALIZED VIEW mv TO INNER UUID "XXXX" AS SELECT a FROM b" + /// This replacement is safe only for CREATE queries when inner target tables don't exist yet. + if (!query.attach) + { + auto generate_target_uuid = [&](ViewTarget::Kind target_kind) + { + if ((query.getTargetInnerUUID(target_kind) == UUIDHelpers::Nil) && query.getTargetTableID(target_kind).empty()) + setTargetInnerUUID(target_kind, UUIDHelpers::generateV4()); + }; + + /// If destination table (to_table_id) is not specified for materialized view, + /// then MV will create inner table. We should generate UUID of inner table here. + if (query.is_materialized_view) + generate_target_uuid(ViewTarget::To); + } + } +} + +bool CreateQueryUUIDs::empty() const +{ + if (uuid != UUIDHelpers::Nil) + return false; + for (const auto & [_, inner_uuid] : targets_inner_uuids) + { + if (inner_uuid != UUIDHelpers::Nil) + return false; + } + return true; +} + +String CreateQueryUUIDs::toString() const +{ + WriteBufferFromOwnString out; + out << "{"; + bool need_comma = false; + auto add_name_and_uuid_to_string = [&](std::string_view name_, const UUID & uuid_) + { + if (std::exchange(need_comma, true)) + out << ", "; + out << "\"" << name_ << "\": \"" << uuid_ << "\""; + }; + if (uuid != UUIDHelpers::Nil) + add_name_and_uuid_to_string("uuid", uuid); + for (const auto & [kind, inner_uuid] : targets_inner_uuids) + { + if (inner_uuid != UUIDHelpers::Nil) + add_name_and_uuid_to_string(::DB::toString(kind), inner_uuid); + } + out << "}"; + return out.str(); +} + +CreateQueryUUIDs CreateQueryUUIDs::fromString(const String & str) +{ + ReadBufferFromString in{str}; + CreateQueryUUIDs res; + skipWhitespaceIfAny(in); + in >> "{"; + skipWhitespaceIfAny(in); + char c; + while (in.peek(c) && c != '}') + { + String name; + String value; + readDoubleQuotedString(name, in); + skipWhitespaceIfAny(in); + in >> ":"; + skipWhitespaceIfAny(in); + readDoubleQuotedString(value, in); + skipWhitespaceIfAny(in); + if (name == "uuid") + { + res.uuid = parse(value); + } + else + { + ViewTarget::Kind kind; + parseFromString(kind, name); + res.setTargetInnerUUID(kind, parse(value)); + } + if (in.peek(c) && c == ',') + { + in.ignore(1); + skipWhitespaceIfAny(in); + } + } + in >> "}"; + return res; +} + +void CreateQueryUUIDs::setTargetInnerUUID(ViewTarget::Kind kind, const UUID & new_inner_uuid) +{ + for (auto & pair : targets_inner_uuids) + { + if (pair.first == kind) + { + pair.second = new_inner_uuid; + return; + } + } + if (new_inner_uuid != UUIDHelpers::Nil) + targets_inner_uuids.emplace_back(kind, new_inner_uuid); +} + +UUID CreateQueryUUIDs::getTargetInnerUUID(ViewTarget::Kind kind) const +{ + for (const auto & pair : targets_inner_uuids) + { + if (pair.first == kind) + return pair.second; + } + return UUIDHelpers::Nil; +} + +void CreateQueryUUIDs::copyToQuery(ASTCreateQuery & query) const +{ + query.uuid = uuid; + + if (query.targets) + query.targets->resetInnerUUIDs(); + + if (!targets_inner_uuids.empty()) + { + if (!query.targets) + query.set(query.targets, std::make_shared()); + + for (const auto & [kind, inner_uuid] : targets_inner_uuids) + { + if (inner_uuid != UUIDHelpers::Nil) + query.targets->setInnerUUID(kind, inner_uuid); + } + } +} + +} diff --git a/src/Parsers/CreateQueryUUIDs.h b/src/Parsers/CreateQueryUUIDs.h new file mode 100644 index 00000000000..419dad24b35 --- /dev/null +++ b/src/Parsers/CreateQueryUUIDs.h @@ -0,0 +1,40 @@ +#pragma once + +#include + + +namespace DB +{ +class ASTCreateQuery; + +/// The UUID of a table or a database defined with a CREATE QUERY along with the UUIDs of its inner targets. +struct CreateQueryUUIDs +{ + CreateQueryUUIDs() = default; + + /// Collect UUIDs from ASTCreateQuery. + /// Parameters: + /// `generate_random` - if it's true then unspecified in the query UUIDs will be generated randomly; + /// `force_random` - if it's true then all UUIDs (even specified in the query) will be (re)generated randomly. + explicit CreateQueryUUIDs(const ASTCreateQuery & query, bool generate_random = false, bool force_random = false); + + bool empty() const; + explicit operator bool() const { return !empty(); } + + String toString() const; + static CreateQueryUUIDs fromString(const String & str); + + void setTargetInnerUUID(ViewTarget::Kind kind, const UUID & new_inner_uuid); + UUID getTargetInnerUUID(ViewTarget::Kind kind) const; + + /// Copies UUIDs to ASTCreateQuery. + void copyToQuery(ASTCreateQuery & query) const; + + /// UUID of the table. + UUID uuid = UUIDHelpers::Nil; + + /// UUIDs of its target table (or tables). + std::vector> targets_inner_uuids; +}; + +} diff --git a/src/Parsers/ExpressionElementParsers.cpp b/src/Parsers/ExpressionElementParsers.cpp index 5997452bcf3..d4fc9a4bc4d 100644 --- a/src/Parsers/ExpressionElementParsers.cpp +++ b/src/Parsers/ExpressionElementParsers.cpp @@ -1129,11 +1129,11 @@ inline static bool makeHexOrBinStringLiteral(IParser::Pos & pos, ASTPtr & node, if (hex) { - hexStringDecode(str_begin, str_end, res_pos); + hexStringDecode(str_begin, str_end, res_pos, word_size); } else { - binStringDecode(str_begin, str_end, res_pos); + binStringDecode(str_begin, str_end, res_pos, word_size); } return makeStringLiteral(pos, node, String(reinterpret_cast(res.data()), (res_pos - res_begin - 1))); diff --git a/src/Parsers/ExpressionElementParsers.h b/src/Parsers/ExpressionElementParsers.h index 4e3f29bfe0c..0209e785bff 100644 --- a/src/Parsers/ExpressionElementParsers.h +++ b/src/Parsers/ExpressionElementParsers.h @@ -9,7 +9,7 @@ namespace DB { -/** The SELECT subquery is in parenthesis. +/** The SELECT subquery, in parentheses. */ class ParserSubquery : public IParserBase { diff --git a/src/Parsers/ExpressionListParsers.cpp b/src/Parsers/ExpressionListParsers.cpp index fff8383e7b3..f97c042e91e 100644 --- a/src/Parsers/ExpressionListParsers.cpp +++ b/src/Parsers/ExpressionListParsers.cpp @@ -2743,7 +2743,7 @@ Action ParserExpressionImpl::tryParseOperator(Layers & layers, IParser::Pos & po /// 'AND' can be both boolean function and part of the '... BETWEEN ... AND ...' operator if (op.function_name == "and" && layers.back()->between_counter) { - layers.back()->between_counter--; + --layers.back()->between_counter; op = finish_between_operator; } diff --git a/src/Parsers/ParserCreateQuery.cpp b/src/Parsers/ParserCreateQuery.cpp index 014dc7bd3bf..1cc565a41d1 100644 --- a/src/Parsers/ParserCreateQuery.cpp +++ b/src/Parsers/ParserCreateQuery.cpp @@ -22,6 +22,7 @@ #include #include #include +#include #include #include @@ -693,7 +694,8 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe ASTPtr table; ASTPtr columns_list; - ASTPtr storage; + std::shared_ptr storage; + ASTPtr targets; ASTPtr as_database; ASTPtr as_table; ASTPtr as_table_function; @@ -773,6 +775,17 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe return true; } + auto parse_storage = [&] + { + chassert(!storage); + ASTPtr ast; + if (!storage_p.parse(pos, ast, expected)) + return false; + + storage = typeid_cast>(ast); + return true; + }; + auto need_parse_as_select = [&is_create_empty, &pos, &expected]() { if (ParserKeyword{Keyword::EMPTY_AS}.ignore(pos, expected)) @@ -798,7 +811,7 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe if (!s_rparen.ignore(pos, expected)) return false; - auto storage_parse_result = storage_p.parse(pos, storage, expected); + auto storage_parse_result = parse_storage(); if ((storage_parse_result || is_temporary) && need_parse_as_select()) { @@ -820,7 +833,7 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe */ else { - storage_p.parse(pos, storage, expected); + parse_storage(); /// CREATE|ATTACH TABLE ... AS ... if (need_parse_as_select()) @@ -843,7 +856,7 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe /// Optional - ENGINE can be specified. if (!storage) - storage_p.parse(pos, storage, expected); + parse_storage(); } } } @@ -904,6 +917,7 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe tryGetIdentifierNameInto(as_database, query->as_database); tryGetIdentifierNameInto(as_table, query->as_table); query->set(query->select, select); + query->set(query->targets, targets); query->is_create_empty = is_create_empty; if (from_path) @@ -977,6 +991,13 @@ bool ParserCreateLiveViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e return false; } + std::shared_ptr targets; + if (to_table) + { + targets = std::make_shared(); + targets->setTableID(ViewTarget::To, to_table->as()->getTableId()); + } + /// Optional - a list of columns can be specified. It must fully comply with SELECT. if (s_lparen.ignore(pos, expected)) { @@ -1017,14 +1038,12 @@ bool ParserCreateLiveViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e if (query->table) query->children.push_back(query->table); - if (to_table) - query->to_table_id = to_table->as()->getTableId(); - query->set(query->columns_list, columns_list); tryGetIdentifierNameInto(as_database, query->as_database); tryGetIdentifierNameInto(as_table, query->as_table); query->set(query->select, select); + query->set(query->targets, targets); if (comment) query->set(query->comment, comment); @@ -1139,6 +1158,18 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & storage_p.parse(pos, storage, expected); } + std::shared_ptr targets; + if (to_table || storage || inner_storage) + { + targets = std::make_shared(); + if (to_table) + targets->setTableID(ViewTarget::To, to_table->as()->getTableId()); + if (storage) + targets->setInnerEngine(ViewTarget::To, storage); + if (inner_storage) + targets->setInnerEngine(ViewTarget::Inner, inner_storage); + } + // WATERMARK if (s_watermark.ignore(pos, expected)) { @@ -1176,6 +1207,7 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & if (!select_p.parse(pos, select, expected)) return false; + auto comment = parseComment(pos, expected); auto query = std::make_shared(); node = query; @@ -1194,13 +1226,11 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & query->children.push_back(query->database); if (query->table) query->children.push_back(query->table); - - if (to_table) - query->to_table_id = to_table->as()->getTableId(); + if (comment) + query->set(query->comment, comment); query->set(query->columns_list, columns_list); - query->set(query->storage, storage); - query->set(query->inner_storage, inner_storage); + query->is_watermark_strictly_ascending = is_watermark_strictly_ascending; query->is_watermark_ascending = is_watermark_ascending; query->is_watermark_bounded = is_watermark_bounded; @@ -1213,6 +1243,7 @@ bool ParserCreateWindowViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & tryGetIdentifierNameInto(as_database, query->as_database); tryGetIdentifierNameInto(as_table, query->as_table); query->set(query->select, select); + query->set(query->targets, targets); return true; } @@ -1436,6 +1467,7 @@ bool ParserCreateDatabaseQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e return true; } + bool ParserCreateViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected) { ParserKeyword s_create(Keyword::CREATE); @@ -1622,13 +1654,8 @@ bool ParserCreateViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec if (query->table) query->children.push_back(query->table); - if (to_table) - query->to_table_id = to_table->as()->getTableId(); - if (to_inner_uuid) - query->to_inner_uuid = parseFromString(to_inner_uuid->as()->value.get()); - query->set(query->columns_list, columns_list); - query->set(query->storage, storage); + if (refresh_strategy) query->set(query->refresh_strategy, refresh_strategy); if (comment) @@ -1639,29 +1666,41 @@ bool ParserCreateViewQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec if (query->columns_list && query->columns_list->primary_key) { /// If engine is not set will use default one - if (!query->storage) - query->set(query->storage, std::make_shared()); - else if (query->storage->primary_key) + if (!storage) + storage = std::make_shared(); + auto & storage_ref = typeid_cast(*storage); + if (storage_ref.primary_key) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Multiple primary keys are not allowed."); - - query->storage->primary_key = query->columns_list->primary_key; - + storage_ref.primary_key = query->columns_list->primary_key; } if (query->columns_list && (query->columns_list->primary_key_from_columns)) { /// If engine is not set will use default one - if (!query->storage) - query->set(query->storage, std::make_shared()); - else if (query->storage->primary_key) + if (!storage) + storage = std::make_shared(); + auto & storage_ref = typeid_cast(*storage); + if (storage_ref.primary_key) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Multiple primary keys are not allowed."); + storage_ref.primary_key = query->columns_list->primary_key_from_columns; + } - query->storage->primary_key = query->columns_list->primary_key_from_columns; + std::shared_ptr targets; + if (to_table || to_inner_uuid || storage) + { + targets = std::make_shared(); + if (to_table) + targets->setTableID(ViewTarget::To, to_table->as()->getTableId()); + if (to_inner_uuid) + targets->setInnerUUID(ViewTarget::To, parseFromString(to_inner_uuid->as()->value.safeGet())); + if (storage) + targets->setInnerEngine(ViewTarget::To, storage); } tryGetIdentifierNameInto(as_database, query->as_database); tryGetIdentifierNameInto(as_table, query->as_table); query->set(query->select, select); + query->set(query->targets, targets); return true; } diff --git a/src/Parsers/ParserDescribeTableQuery.cpp b/src/Parsers/ParserDescribeTableQuery.cpp index 92c0cfacd9b..22bbfdb03e1 100644 --- a/src/Parsers/ParserDescribeTableQuery.cpp +++ b/src/Parsers/ParserDescribeTableQuery.cpp @@ -11,15 +11,12 @@ namespace DB { - bool ParserDescribeTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected) { ParserKeyword s_describe(Keyword::DESCRIBE); ParserKeyword s_desc(Keyword::DESC); ParserKeyword s_table(Keyword::TABLE); ParserKeyword s_settings(Keyword::SETTINGS); - ParserToken s_dot(TokenType::Dot); - ParserIdentifier name_p; ParserSetQuery parser_settings(true); ASTPtr database; @@ -53,5 +50,4 @@ bool ParserDescribeTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & ex return true; } - } diff --git a/src/Parsers/ParserViewTargets.cpp b/src/Parsers/ParserViewTargets.cpp new file mode 100644 index 00000000000..8f010882cdd --- /dev/null +++ b/src/Parsers/ParserViewTargets.cpp @@ -0,0 +1,88 @@ +#include + +#include +#include +#include +#include +#include + + +namespace DB +{ + +ParserViewTargets::ParserViewTargets() +{ + for (auto kind : magic_enum::enum_values()) + accept_kinds.push_back(kind); +} + +bool ParserViewTargets::parseImpl(Pos & pos, ASTPtr & node, Expected & expected) +{ + ParserStringLiteral literal_p; + ParserStorage storage_p{ParserStorage::TABLE_ENGINE}; + ParserCompoundIdentifier table_name_p(/*table_name_with_optional_uuid*/ true, /*allow_query_parameter*/ true); + + std::shared_ptr res; + + auto result = [&] -> ASTViewTargets & + { + if (!res) + res = std::make_shared(); + return *res; + }; + + for (;;) + { + auto start = pos; + for (auto kind : accept_kinds) + { + auto current = pos; + + auto keyword = ASTViewTargets::getKeywordForInnerUUID(kind); + if (keyword && ParserKeyword{*keyword}.ignore(pos, expected)) + { + ASTPtr ast; + if (literal_p.parse(pos, ast, expected)) + { + result().setInnerUUID(kind, parseFromString(ast->as()->value.safeGet())); + break; + } + } + pos = current; + + keyword = ASTViewTargets::getKeywordForInnerStorage(kind); + if (keyword && ParserKeyword{*keyword}.ignore(pos, expected)) + { + ASTPtr ast; + if (storage_p.parse(pos, ast, expected)) + { + result().setInnerEngine(kind, ast); + break; + } + } + pos = current; + + keyword = ASTViewTargets::getKeywordForTableID(kind); + if (keyword && ParserKeyword{*keyword}.ignore(pos, expected)) + { + ASTPtr ast; + if (table_name_p.parse(pos, ast, expected)) + { + result().setTableID(kind, ast->as()->getTableId()); + break; + } + } + pos = current; + } + if (pos == start) + break; + } + + if (!res || res->targets.empty()) + return false; + + node = res; + return true; +} + +} diff --git a/src/Parsers/ParserViewTargets.h b/src/Parsers/ParserViewTargets.h new file mode 100644 index 00000000000..3af3c0b8df3 --- /dev/null +++ b/src/Parsers/ParserViewTargets.h @@ -0,0 +1,29 @@ +#pragma once + +#include +#include + + +namespace DB +{ + +/// Parses information about target tables (external or inner) of a materialized view or a window view. +/// The function parses one or multiple parts of a CREATE query looking like this: +/// TO db.table_name +/// TO INNER UUID 'XXX' +/// {ENGINE / INNER ENGINE} TableEngine(arguments) [ORDER BY ...] [SETTINGS ...] +/// Returns ASTViewTargets if succeeded. +class ParserViewTargets : public IParserBase +{ +public: + ParserViewTargets(); + explicit ParserViewTargets(const std::vector & accept_kinds_) : accept_kinds(accept_kinds_) { } + +protected: + const char * getName() const override { return "ViewTargets"; } + bool parseImpl(Pos & pos, ASTPtr & node, Expected & expected) override; + + std::vector accept_kinds; +}; + +} diff --git a/src/Parsers/examples/CMakeLists.txt b/src/Parsers/examples/CMakeLists.txt index 261f234081c..07f60601acd 100644 --- a/src/Parsers/examples/CMakeLists.txt +++ b/src/Parsers/examples/CMakeLists.txt @@ -1,7 +1,7 @@ set(SRCS) clickhouse_add_executable(lexer lexer.cpp ${SRCS}) -target_link_libraries(lexer PRIVATE clickhouse_parsers) +target_link_libraries(lexer PRIVATE clickhouse_parsers clickhouse_common_config) clickhouse_add_executable(select_parser select_parser.cpp ${SRCS} "../../Server/ServerType.cpp") target_link_libraries(select_parser PRIVATE dbms) diff --git a/src/Parsers/isUnquotedIdentifier.cpp b/src/Parsers/isUnquotedIdentifier.cpp new file mode 100644 index 00000000000..26cb3992a50 --- /dev/null +++ b/src/Parsers/isUnquotedIdentifier.cpp @@ -0,0 +1,33 @@ +#include + +#include +#include + +namespace DB +{ + +bool isUnquotedIdentifier(const String & name) +{ + auto is_keyword = [&name](Keyword keyword) + { + auto s = toStringView(keyword); + if (name.size() != s.size()) + return false; + return strncasecmp(s.data(), name.data(), s.size()) == 0; + }; + + /// Special keywords are parsed as literals instead of identifiers. + if (is_keyword(Keyword::NULL_KEYWORD) || is_keyword(Keyword::TRUE_KEYWORD) || is_keyword(Keyword::FALSE_KEYWORD)) + return false; + + Lexer lexer(name.data(), name.data() + name.size()); + + auto maybe_ident = lexer.nextToken(); + + if (maybe_ident.type != TokenType::BareWord) + return false; + + return lexer.nextToken().isEnd(); +} + +} diff --git a/src/Parsers/isUnquotedIdentifier.h b/src/Parsers/isUnquotedIdentifier.h new file mode 100644 index 00000000000..9c9f9239eb3 --- /dev/null +++ b/src/Parsers/isUnquotedIdentifier.h @@ -0,0 +1,18 @@ +#pragma once + +#include + +namespace DB +{ + +/// Checks if the input string @name is a valid unquoted identifier. +/// +/// Example Usage: +/// abc -> true (valid unquoted identifier) +/// 123 -> false (identifiers cannot start with digits) +/// `123` -> false (quoted identifiers are not considered) +/// `abc` -> false (quoted identifiers are not considered) +/// null -> false (reserved literal keyword) +bool isUnquotedIdentifier(const String & name); + +} diff --git a/src/Planner/CollectSets.cpp b/src/Planner/CollectSets.cpp index 52a0d748d63..9cac4e61380 100644 --- a/src/Planner/CollectSets.cpp +++ b/src/Planner/CollectSets.cpp @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Planner/Planner.cpp b/src/Planner/Planner.cpp index 2d42ed73223..5640d84731f 100644 --- a/src/Planner/Planner.cpp +++ b/src/Planner/Planner.cpp @@ -3,6 +3,8 @@ #include #include #include +#include +#include #include #include @@ -39,6 +41,7 @@ #include #include +#include #include #include @@ -370,10 +373,10 @@ Aggregator::Params getAggregatorParams(const PlannerContextPtr & planner_context const auto & query_context = planner_context->getQueryContext(); const Settings & settings = query_context->getSettingsRef(); - const auto stats_collecting_params = Aggregator::Params::StatsCollectingParams( - select_query_info.query, + const auto stats_collecting_params = StatsCollectingParams( + calculateCacheKey(select_query_info.query), settings.collect_hash_table_stats_during_aggregation, - settings.max_entries_for_hash_table_stats, + query_context->getServerSettings().max_entries_for_hash_table_stats, settings.max_size_to_preallocate_for_aggregation); auto aggregate_descriptions = aggregation_analysis_result.aggregate_descriptions; diff --git a/src/Planner/PlannerActionsVisitor.cpp b/src/Planner/PlannerActionsVisitor.cpp index 7a12d5d690d..a0c7f9b2516 100644 --- a/src/Planner/PlannerActionsVisitor.cpp +++ b/src/Planner/PlannerActionsVisitor.cpp @@ -30,6 +30,8 @@ #include #include +#include + namespace DB { diff --git a/src/Planner/PlannerExpressionAnalysis.cpp b/src/Planner/PlannerExpressionAnalysis.cpp index ceb506d1bbb..2b67c96d843 100644 --- a/src/Planner/PlannerExpressionAnalysis.cpp +++ b/src/Planner/PlannerExpressionAnalysis.cpp @@ -23,6 +23,8 @@ #include #include +#include + namespace DB { @@ -85,14 +87,14 @@ bool canRemoveConstantFromGroupByKey(const ConstantNode & root) else if (function_node) { /// Do not allow removing constants like `hostName()` - if (!function_node->getFunctionOrThrow()->isDeterministic()) + if (function_node->getFunctionOrThrow()->isServerConstant()) return false; for (const auto & child : function_node->getArguments()) nodes.push(child.get()); } - else - return false; + // else + // return false; } return true; diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 43b223172e6..4fedacbb48a 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -1,5 +1,7 @@ #include +#include + #include #include @@ -53,7 +55,7 @@ #include #include #include -#include +#include #include #include #include @@ -75,7 +77,6 @@ namespace ErrorCodes extern const int INVALID_JOIN_ON_EXPRESSION; extern const int LOGICAL_ERROR; extern const int NOT_IMPLEMENTED; - extern const int SYNTAX_ERROR; extern const int ACCESS_DENIED; extern const int PARAMETER_OUT_OF_BOUND; extern const int TOO_MANY_COLUMNS; @@ -647,7 +648,8 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres auto table_expression_query_info = select_query_info; table_expression_query_info.table_expression = table_expression; table_expression_query_info.filter_actions_dag = table_expression_data.getFilterActions(); - table_expression_query_info.analyzer_can_use_parallel_replicas_on_follower = table_node == planner_context->getGlobalPlannerContext()->parallel_replicas_table; + table_expression_query_info.current_table_chosen_for_reading_with_parallel_replicas + = table_node == planner_context->getGlobalPlannerContext()->parallel_replicas_table; size_t max_streams = settings.max_threads; size_t max_threads_execute_query = settings.max_threads; @@ -693,14 +695,14 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres if (select_query_info.local_storage_limits.local_limits.size_limits.max_rows != 0) { if (max_block_size_limited < select_query_info.local_storage_limits.local_limits.size_limits.max_rows) - table_expression_query_info.limit = max_block_size_limited; + table_expression_query_info.trivial_limit = max_block_size_limited; /// Ask to read just enough rows to make the max_rows limit effective (so it has a chance to be triggered). else if (select_query_info.local_storage_limits.local_limits.size_limits.max_rows < std::numeric_limits::max()) - table_expression_query_info.limit = 1 + select_query_info.local_storage_limits.local_limits.size_limits.max_rows; + table_expression_query_info.trivial_limit = 1 + select_query_info.local_storage_limits.local_limits.size_limits.max_rows; } else { - table_expression_query_info.limit = max_block_size_limited; + table_expression_query_info.trivial_limit = max_block_size_limited; } } @@ -834,7 +836,7 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres if (row_policy_filter_info.actions) table_expression_data.setRowLevelFilterActions(row_policy_filter_info.actions); - if (query_context->getParallelReplicasMode() == Context::ParallelReplicasMode::CUSTOM_KEY) + if (query_context->canUseParallelReplicasCustomKey()) { if (settings.parallel_replicas_count > 1) { @@ -843,9 +845,14 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres add_filter(parallel_replicas_custom_key_filter_info, "Parallel replicas custom key filter"); } else if (auto * distributed = typeid_cast(storage.get()); - distributed && query_context->canUseParallelReplicasCustomKey(*distributed->getCluster())) + distributed && query_context->canUseParallelReplicasCustomKeyForCluster(*distributed->getCluster())) { planner_context->getMutableQueryContext()->setSetting("distributed_group_by_no_merge", 2); + /// We disable prefer_localhost_replica because if one of the replicas is local it will create a single local plan + /// instead of executing the query with multiple replicas + /// We can enable this setting again for custom key parallel replicas when we can generate a plan that will use both a + /// local plan and remote replicas + planner_context->getMutableQueryContext()->setSetting("prefer_localhost_replica", Field{0}); } } @@ -857,6 +864,15 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres from_stage = storage->getQueryProcessingStage( query_context, select_query_options.to_stage, storage_snapshot, table_expression_query_info); + /// It is just a safety check needed until we have a proper sending plan to replicas. + /// If we have a non-trivial storage like View it might create its own Planner inside read(), run findTableForParallelReplicas() + /// and find some other table that might be used for reading with parallel replicas. It will lead to errors. + const bool other_table_already_chosen_for_reading_with_parallel_replicas + = planner_context->getGlobalPlannerContext()->parallel_replicas_table + && !table_expression_query_info.current_table_chosen_for_reading_with_parallel_replicas; + if (other_table_already_chosen_for_reading_with_parallel_replicas) + planner_context->getMutableQueryContext()->setSetting("allow_experimental_parallel_reading_from_replicas", Field(0)); + storage->read( query_plan, columns_names, @@ -879,7 +895,7 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres }; /// query_plan can be empty if there is nothing to read - if (query_plan.isInitialized() && parallel_replicas_enabled_for_storage(storage, settings) && query_context->canUseParallelReplicasOnInitiator()) + if (query_plan.isInitialized() && parallel_replicas_enabled_for_storage(storage, settings)) { // (1) find read step QueryPlan::Node * node = query_plan.getRootNode(); @@ -906,54 +922,78 @@ JoinTreeQueryPlan buildQueryPlanForTableExpression(QueryTreeNodePtr table_expres } chassert(reading); - - // (2) if it's ReadFromMergeTree - run index analysis and check number of rows to read - if (settings.parallel_replicas_min_number_of_rows_per_replica > 0) + if (query_context->canUseParallelReplicasCustomKey() && query_context->getClientInfo().distributed_depth == 0) { - auto result_ptr = reading->selectRangesToRead(); - - UInt64 rows_to_read = result_ptr->selected_rows; - if (table_expression_query_info.limit > 0 && table_expression_query_info.limit < rows_to_read) - rows_to_read = table_expression_query_info.limit; - - if (max_block_size_limited && (max_block_size_limited < rows_to_read)) - rows_to_read = max_block_size_limited; - - const size_t number_of_replicas_to_use = rows_to_read / settings.parallel_replicas_min_number_of_rows_per_replica; - LOG_TRACE( - getLogger("Planner"), - "Estimated {} rows to read. It is enough work for {} parallel replicas", - rows_to_read, - number_of_replicas_to_use); - - if (number_of_replicas_to_use <= 1) + if (auto cluster = query_context->getClusterForParallelReplicas(); + query_context->canUseParallelReplicasCustomKeyForCluster(*cluster)) { - planner_context->getMutableQueryContext()->setSetting( - "allow_experimental_parallel_reading_from_replicas", Field(0)); - planner_context->getMutableQueryContext()->setSetting("max_parallel_replicas", UInt64{1}); - LOG_DEBUG(getLogger("Planner"), "Disabling parallel replicas because there aren't enough rows to read"); - } - else if (number_of_replicas_to_use < settings.max_parallel_replicas) - { - planner_context->getMutableQueryContext()->setSetting("max_parallel_replicas", number_of_replicas_to_use); - LOG_DEBUG(getLogger("Planner"), "Reducing the number of replicas to use to {}", number_of_replicas_to_use); + planner_context->getMutableQueryContext()->setSetting("prefer_localhost_replica", Field{0}); + auto modified_query_info = select_query_info; + modified_query_info.cluster = std::move(cluster); + from_stage = QueryProcessingStage::WithMergeableStateAfterAggregationAndLimit; + QueryPlan query_plan_parallel_replicas; + ClusterProxy::executeQueryWithParallelReplicasCustomKey( + query_plan_parallel_replicas, + storage->getStorageID(), + modified_query_info, + storage->getInMemoryMetadataPtr()->getColumns(), + storage_snapshot, + from_stage, + table_expression_query_info.query_tree, + query_context); + query_plan = std::move(query_plan_parallel_replicas); } } - - // (3) if parallel replicas still enabled - replace reading step - if (planner_context->getQueryContext()->canUseParallelReplicasOnInitiator()) + else if (query_context->canUseParallelReplicasOnInitiator()) { - from_stage = QueryProcessingStage::WithMergeableState; - QueryPlan query_plan_parallel_replicas; - ClusterProxy::executeQueryWithParallelReplicas( - query_plan_parallel_replicas, - storage->getStorageID(), - from_stage, - table_expression_query_info.query_tree, - table_expression_query_info.planner_context, - query_context, - table_expression_query_info.storage_limits); - query_plan = std::move(query_plan_parallel_replicas); + // (2) if it's ReadFromMergeTree - run index analysis and check number of rows to read + if (settings.parallel_replicas_min_number_of_rows_per_replica > 0) + { + auto result_ptr = reading->selectRangesToRead(); + + UInt64 rows_to_read = result_ptr->selected_rows; + if (table_expression_query_info.trivial_limit > 0 && table_expression_query_info.trivial_limit < rows_to_read) + rows_to_read = table_expression_query_info.trivial_limit; + + if (max_block_size_limited && (max_block_size_limited < rows_to_read)) + rows_to_read = max_block_size_limited; + + const size_t number_of_replicas_to_use = rows_to_read / settings.parallel_replicas_min_number_of_rows_per_replica; + LOG_TRACE( + getLogger("Planner"), + "Estimated {} rows to read. It is enough work for {} parallel replicas", + rows_to_read, + number_of_replicas_to_use); + + if (number_of_replicas_to_use <= 1) + { + planner_context->getMutableQueryContext()->setSetting( + "allow_experimental_parallel_reading_from_replicas", Field(0)); + planner_context->getMutableQueryContext()->setSetting("max_parallel_replicas", UInt64{1}); + LOG_DEBUG(getLogger("Planner"), "Disabling parallel replicas because there aren't enough rows to read"); + } + else if (number_of_replicas_to_use < settings.max_parallel_replicas) + { + planner_context->getMutableQueryContext()->setSetting("max_parallel_replicas", number_of_replicas_to_use); + LOG_DEBUG(getLogger("Planner"), "Reducing the number of replicas to use to {}", number_of_replicas_to_use); + } + } + + // (3) if parallel replicas still enabled - replace reading step + if (planner_context->getQueryContext()->canUseParallelReplicasOnInitiator()) + { + from_stage = QueryProcessingStage::WithMergeableState; + QueryPlan query_plan_parallel_replicas; + ClusterProxy::executeQueryWithParallelReplicas( + query_plan_parallel_replicas, + storage->getStorageID(), + from_stage, + table_expression_query_info.query_tree, + table_expression_query_info.planner_context, + query_context, + table_expression_query_info.storage_limits); + query_plan = std::move(query_plan_parallel_replicas); + } } } @@ -1356,12 +1396,7 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ { if (!join_clause.hasASOF()) throw Exception(ErrorCodes::INVALID_JOIN_ON_EXPRESSION, - "JOIN {} no inequality in ASOF JOIN ON section.", - join_node.formatASTForErrorMessage()); - - if (table_join_clause.key_names_left.size() <= 1) - throw Exception(ErrorCodes::SYNTAX_ERROR, - "JOIN {} ASOF join needs at least one equi-join column", + "JOIN {} no inequality in ASOF JOIN ON section", join_node.formatASTForErrorMessage()); } @@ -1483,7 +1518,9 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ { const auto & join_clause = table_join->getOnlyClause(); - bool kind_allows_filtering = isInner(join_kind) || isLeft(join_kind) || isRight(join_kind); + bool join_type_allows_filtering = (join_strictness == JoinStrictness::All || join_strictness == JoinStrictness::Any) + && (isInner(join_kind) || isLeft(join_kind) || isRight(join_kind)); + auto has_non_const = [](const Block & block, const auto & keys) { @@ -1503,7 +1540,7 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ bool has_non_const_keys = has_non_const(left_plan.getCurrentDataStream().header, join_clause.key_names_left) && has_non_const(right_plan.getCurrentDataStream().header, join_clause.key_names_right); - if (settings.max_rows_in_set_to_optimize_join > 0 && kind_allows_filtering && has_non_const_keys) + if (settings.max_rows_in_set_to_optimize_join > 0 && join_type_allows_filtering && has_non_const_keys) { auto * left_set = add_create_set(left_plan, join_clause.key_names_left, JoinTableSide::Left); auto * right_set = add_create_set(right_plan, join_clause.key_names_right, JoinTableSide::Right); diff --git a/src/Planner/PlannerJoins.cpp b/src/Planner/PlannerJoins.cpp index 84efdd21336..d9360a58240 100644 --- a/src/Planner/PlannerJoins.cpp +++ b/src/Planner/PlannerJoins.cpp @@ -29,7 +29,7 @@ #include #include -#include +#include #include #include #include @@ -43,6 +43,9 @@ #include #include +#include +#include + namespace DB { @@ -528,7 +531,7 @@ JoinClausesAndActions buildJoinClausesAndActions( size_t join_clause_key_nodes_size = join_clause.getLeftKeyNodes().size(); if (join_clause_key_nodes_size == 0) - throw Exception(ErrorCodes::INVALID_JOIN_ON_EXPRESSION, "JOIN {} cannot get JOIN keys", + throw Exception(ErrorCodes::INVALID_JOIN_ON_EXPRESSION, "Cannot determine join keys in {}", join_node.formatASTForErrorMessage()); for (size_t i = 0; i < join_clause_key_nodes_size; ++i) @@ -768,10 +771,8 @@ std::shared_ptr tryDirectJoin(const std::shared_ptr(table_join, right_table_expression_header, storage, right_table_expression_header_with_storage_column_names); } - } - static std::shared_ptr tryCreateJoin(JoinAlgorithm algorithm, std::shared_ptr & table_join, const QueryTreeNodePtr & right_table_expression, @@ -802,13 +803,20 @@ static std::shared_ptr tryCreateJoin(JoinAlgorithm algorithm, algorithm == JoinAlgorithm::PARALLEL_HASH || algorithm == JoinAlgorithm::DEFAULT) { + auto query_context = planner_context->getQueryContext(); if (table_join->allowParallelHashJoin()) { - auto query_context = planner_context->getQueryContext(); - return std::make_shared(query_context, table_join, query_context->getSettings().max_threads, right_table_expression_header); + const auto & settings = query_context->getSettingsRef(); + StatsCollectingParams params{ + calculateCacheKey(table_join, right_table_expression), + settings.collect_hash_table_stats_during_joins, + query_context->getServerSettings().max_entries_for_hash_table_stats, + settings.max_size_to_preallocate_for_joins}; + return std::make_shared( + query_context, table_join, query_context->getSettingsRef().max_threads, right_table_expression_header, params); } - return std::make_shared(table_join, right_table_expression_header); + return std::make_shared(table_join, right_table_expression_header, query_context->getSettingsRef().join_any_take_last_row); } if (algorithm == JoinAlgorithm::FULL_SORTING_MERGE) diff --git a/src/Planner/PlannerSorting.cpp b/src/Planner/PlannerSorting.cpp index 611a26f78fa..1c6e820fed1 100644 --- a/src/Planner/PlannerSorting.cpp +++ b/src/Planner/PlannerSorting.cpp @@ -1,5 +1,7 @@ #include +#include + #include #include diff --git a/src/Planner/PlannerWindowFunctions.cpp b/src/Planner/PlannerWindowFunctions.cpp index ce74d82c08d..225852de5a7 100644 --- a/src/Planner/PlannerWindowFunctions.cpp +++ b/src/Planner/PlannerWindowFunctions.cpp @@ -4,6 +4,8 @@ #include #include +#include + #include #include diff --git a/src/Planner/Utils.cpp b/src/Planner/Utils.cpp index 18a6d297838..3c54c57a28c 100644 --- a/src/Planner/Utils.cpp +++ b/src/Planner/Utils.cpp @@ -33,6 +33,8 @@ #include #include +#include + #include #include #include diff --git a/src/Planner/findParallelReplicasQuery.cpp b/src/Planner/findParallelReplicasQuery.cpp index f2bc1f060d8..c89a70be541 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Processors/Chunk.cpp b/src/Processors/Chunk.cpp index 2631f665f9c..4466be5b3a7 100644 --- a/src/Processors/Chunk.cpp +++ b/src/Processors/Chunk.cpp @@ -19,14 +19,6 @@ Chunk::Chunk(DB::Columns columns_, UInt64 num_rows_) : columns(std::move(columns checkNumRowsIsConsistent(); } -Chunk::Chunk(Columns columns_, UInt64 num_rows_, ChunkInfoPtr chunk_info_) - : columns(std::move(columns_)) - , num_rows(num_rows_) - , chunk_info(std::move(chunk_info_)) -{ - checkNumRowsIsConsistent(); -} - static Columns unmuteColumns(MutableColumns && mutable_columns) { Columns columns; @@ -43,17 +35,11 @@ Chunk::Chunk(MutableColumns columns_, UInt64 num_rows_) checkNumRowsIsConsistent(); } -Chunk::Chunk(MutableColumns columns_, UInt64 num_rows_, ChunkInfoPtr chunk_info_) - : columns(unmuteColumns(std::move(columns_))) - , num_rows(num_rows_) - , chunk_info(std::move(chunk_info_)) -{ - checkNumRowsIsConsistent(); -} - Chunk Chunk::clone() const { - return Chunk(getColumns(), getNumRows(), chunk_info); + auto tmp = Chunk(getColumns(), getNumRows()); + tmp.setChunkInfos(chunk_infos.clone()); + return tmp; } void Chunk::setColumns(Columns columns_, UInt64 num_rows_) @@ -125,7 +111,7 @@ void Chunk::addColumn(size_t position, ColumnPtr column) if (position >= columns.size()) throw Exception(ErrorCodes::POSITION_OUT_OF_BOUND, "Position {} out of bound in Chunk::addColumn(), max position = {}", - position, columns.size() - 1); + position, !columns.empty() ? columns.size() - 1 : 0); if (empty()) num_rows = column->size(); else if (column->size() != num_rows) @@ -143,7 +129,7 @@ void Chunk::erase(size_t position) if (position >= columns.size()) throw Exception(ErrorCodes::POSITION_OUT_OF_BOUND, "Position {} out of bound in Chunk::erase(), max position = {}", - toString(position), toString(columns.size() - 1)); + toString(position), toString(!columns.empty() ? columns.size() - 1 : 0)); columns.erase(columns.begin() + position); } diff --git a/src/Processors/Chunk.h b/src/Processors/Chunk.h index 4f753798eaa..1348966c0d3 100644 --- a/src/Processors/Chunk.h +++ b/src/Processors/Chunk.h @@ -1,7 +1,9 @@ #pragma once +#include #include -#include + +#include namespace DB { @@ -9,11 +11,29 @@ namespace DB class ChunkInfo { public: - virtual ~ChunkInfo() = default; + using Ptr = std::shared_ptr; + ChunkInfo() = default; + ChunkInfo(const ChunkInfo&) = default; + ChunkInfo(ChunkInfo&&) = default; + + virtual Ptr clone() const = 0; + virtual ~ChunkInfo() = default; }; -using ChunkInfoPtr = std::shared_ptr; + +template +class ChunkInfoCloneable : public ChunkInfo +{ +public: + ChunkInfoCloneable() = default; + ChunkInfoCloneable(const ChunkInfoCloneable & other) = default; + + Ptr clone() const override + { + return std::static_pointer_cast(std::make_shared(*static_cast(this))); + } +}; /** * Chunk is a list of columns with the same length. @@ -32,26 +52,26 @@ using ChunkInfoPtr = std::shared_ptr; class Chunk { public: + using ChunkInfoCollection = CollectionOfDerivedItems; + Chunk() = default; Chunk(const Chunk & other) = delete; Chunk(Chunk && other) noexcept : columns(std::move(other.columns)) , num_rows(other.num_rows) - , chunk_info(std::move(other.chunk_info)) + , chunk_infos(std::move(other.chunk_infos)) { other.num_rows = 0; } Chunk(Columns columns_, UInt64 num_rows_); - Chunk(Columns columns_, UInt64 num_rows_, ChunkInfoPtr chunk_info_); Chunk(MutableColumns columns_, UInt64 num_rows_); - Chunk(MutableColumns columns_, UInt64 num_rows_, ChunkInfoPtr chunk_info_); Chunk & operator=(const Chunk & other) = delete; Chunk & operator=(Chunk && other) noexcept { columns = std::move(other.columns); - chunk_info = std::move(other.chunk_info); + chunk_infos = std::move(other.chunk_infos); num_rows = other.num_rows; other.num_rows = 0; return *this; @@ -62,15 +82,15 @@ public: void swap(Chunk & other) noexcept { columns.swap(other.columns); - chunk_info.swap(other.chunk_info); std::swap(num_rows, other.num_rows); + chunk_infos.swap(other.chunk_infos); } void clear() { num_rows = 0; columns.clear(); - chunk_info.reset(); + chunk_infos.clear(); } const Columns & getColumns() const { return columns; } @@ -81,9 +101,9 @@ public: /** Get empty columns with the same types as in block. */ MutableColumns cloneEmptyColumns() const; - const ChunkInfoPtr & getChunkInfo() const { return chunk_info; } - bool hasChunkInfo() const { return chunk_info != nullptr; } - void setChunkInfo(ChunkInfoPtr chunk_info_) { chunk_info = std::move(chunk_info_); } + ChunkInfoCollection & getChunkInfos() { return chunk_infos; } + const ChunkInfoCollection & getChunkInfos() const { return chunk_infos; } + void setChunkInfos(ChunkInfoCollection chunk_infos_) { chunk_infos = std::move(chunk_infos_); } UInt64 getNumRows() const { return num_rows; } UInt64 getNumColumns() const { return columns.size(); } @@ -107,7 +127,7 @@ public: private: Columns columns; UInt64 num_rows = 0; - ChunkInfoPtr chunk_info; + ChunkInfoCollection chunk_infos; void checkNumRowsIsConsistent(); }; @@ -117,11 +137,15 @@ using Chunks = std::vector; /// AsyncInsert needs two kinds of information: /// - offsets of different sub-chunks /// - tokens of different sub-chunks, which are assigned by setting `insert_deduplication_token`. -class AsyncInsertInfo : public ChunkInfo +class AsyncInsertInfo : public ChunkInfoCloneable { public: AsyncInsertInfo() = default; - explicit AsyncInsertInfo(const std::vector & offsets_, const std::vector & tokens_) : offsets(offsets_), tokens(tokens_) {} + AsyncInsertInfo(const AsyncInsertInfo & other) = default; + AsyncInsertInfo(const std::vector & offsets_, const std::vector & tokens_) + : offsets(offsets_) + , tokens(tokens_) + {} std::vector offsets; std::vector tokens; @@ -130,9 +154,11 @@ public: using AsyncInsertInfoPtr = std::shared_ptr; /// Extension to support delayed defaults. AddingDefaultsProcessor uses it to replace missing values with column defaults. -class ChunkMissingValues : public ChunkInfo +class ChunkMissingValues : public ChunkInfoCloneable { public: + ChunkMissingValues(const ChunkMissingValues & other) = default; + using RowsBitMask = std::vector; /// a bit per row for a column const RowsBitMask & getDefaultsBitmask(size_t column_idx) const; diff --git a/src/Processors/Executors/ExecutingGraph.cpp b/src/Processors/Executors/ExecutingGraph.cpp index 27f6a454b24..6d5b60d8159 100644 --- a/src/Processors/Executors/ExecutingGraph.cpp +++ b/src/Processors/Executors/ExecutingGraph.cpp @@ -292,7 +292,7 @@ bool ExecutingGraph::updateNode(uint64_t pid, Queue & queue, Queue & async_queue } else if (last_status == IProcessor::Status::NeedData && status != IProcessor::Status::NeedData) { - processor.input_wait_elapsed_us += processor.input_wait_watch.elapsedMicroseconds(); + processor.input_wait_elapsed_ns += processor.input_wait_watch.elapsedNanoseconds(); } /// PortFull @@ -302,7 +302,7 @@ bool ExecutingGraph::updateNode(uint64_t pid, Queue & queue, Queue & async_queue } else if (last_status == IProcessor::Status::PortFull && status != IProcessor::Status::PortFull) { - processor.output_wait_elapsed_us += processor.output_wait_watch.elapsedMicroseconds(); + processor.output_wait_elapsed_ns += processor.output_wait_watch.elapsedNanoseconds(); } } } diff --git a/src/Processors/Executors/ExecutionThreadContext.cpp b/src/Processors/Executors/ExecutionThreadContext.cpp index 05669725f9a..17b6773ad83 100644 --- a/src/Processors/Executors/ExecutionThreadContext.cpp +++ b/src/Processors/Executors/ExecutionThreadContext.cpp @@ -103,10 +103,10 @@ bool ExecutionThreadContext::executeTask() if (profile_processors) { - UInt64 elapsed_microseconds = execution_time_watch->elapsedMicroseconds(); - node->processor->elapsed_us += elapsed_microseconds; + UInt64 elapsed_ns = execution_time_watch->elapsedNanoseconds(); + node->processor->elapsed_ns += elapsed_ns; if (trace_processors) - span->addAttribute("execution_time_ms", elapsed_microseconds); + span->addAttribute("execution_time_ms", elapsed_ns / 1000U); } #ifndef NDEBUG execution_time_ns += execution_time_watch->elapsed(); diff --git a/src/Processors/Executors/PipelineExecutor.cpp b/src/Processors/Executors/PipelineExecutor.cpp index 49ec9999521..82cad471a29 100644 --- a/src/Processors/Executors/PipelineExecutor.cpp +++ b/src/Processors/Executors/PipelineExecutor.cpp @@ -12,6 +12,7 @@ #include #include #include +#include #ifndef NDEBUG #include diff --git a/src/Processors/Executors/PullingAsyncPipelineExecutor.cpp b/src/Processors/Executors/PullingAsyncPipelineExecutor.cpp index d27002197d2..d9fab88fe1f 100644 --- a/src/Processors/Executors/PullingAsyncPipelineExecutor.cpp +++ b/src/Processors/Executors/PullingAsyncPipelineExecutor.cpp @@ -147,13 +147,10 @@ bool PullingAsyncPipelineExecutor::pull(Block & block, uint64_t milliseconds) block = lazy_format->getPort(IOutputFormat::PortKind::Main).getHeader().cloneWithColumns(chunk.detachColumns()); - if (auto chunk_info = chunk.getChunkInfo()) + if (auto agg_info = chunk.getChunkInfos().get()) { - if (const auto * agg_info = typeid_cast(chunk_info.get())) - { - block.info.bucket_num = agg_info->bucket_num; - block.info.is_overflows = agg_info->is_overflows; - } + block.info.bucket_num = agg_info->bucket_num; + block.info.is_overflows = agg_info->is_overflows; } return true; diff --git a/src/Processors/Executors/PullingPipelineExecutor.cpp b/src/Processors/Executors/PullingPipelineExecutor.cpp index cbf73c5cb07..25c15d40c9a 100644 --- a/src/Processors/Executors/PullingPipelineExecutor.cpp +++ b/src/Processors/Executors/PullingPipelineExecutor.cpp @@ -73,13 +73,10 @@ bool PullingPipelineExecutor::pull(Block & block) } block = pulling_format->getPort(IOutputFormat::PortKind::Main).getHeader().cloneWithColumns(chunk.detachColumns()); - if (auto chunk_info = chunk.getChunkInfo()) + if (auto agg_info = chunk.getChunkInfos().get()) { - if (const auto * agg_info = typeid_cast(chunk_info.get())) - { - block.info.bucket_num = agg_info->bucket_num; - block.info.is_overflows = agg_info->is_overflows; - } + block.info.bucket_num = agg_info->bucket_num; + block.info.is_overflows = agg_info->is_overflows; } return true; diff --git a/src/Processors/Formats/Impl/BinaryRowInputFormat.cpp b/src/Processors/Formats/Impl/BinaryRowInputFormat.cpp index ac5da172210..c5336f3bcc7 100644 --- a/src/Processors/Formats/Impl/BinaryRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/BinaryRowInputFormat.cpp @@ -4,6 +4,7 @@ #include #include #include +#include namespace DB { @@ -55,10 +56,25 @@ std::vector BinaryFormatReader::readNames() template std::vector BinaryFormatReader::readTypes() { - auto types = readHeaderRow(); - for (const auto & type_name : types) - read_data_types.push_back(DataTypeFactory::instance().get(type_name)); - return types; + read_data_types.reserve(read_columns); + Names type_names; + if (format_settings.binary.decode_types_in_binary_format) + { + type_names.reserve(read_columns); + for (size_t i = 0; i < read_columns; ++i) + { + read_data_types.push_back(decodeDataType(*in)); + type_names.push_back(read_data_types.back()->getName()); + } + } + else + { + type_names = readHeaderRow(); + for (const auto & type_name : type_names) + read_data_types.push_back(DataTypeFactory::instance().get(type_name)); + } + + return type_names; } template diff --git a/src/Processors/Formats/Impl/BinaryRowOutputFormat.cpp b/src/Processors/Formats/Impl/BinaryRowOutputFormat.cpp index ff904f61d22..d4c2348d080 100644 --- a/src/Processors/Formats/Impl/BinaryRowOutputFormat.cpp +++ b/src/Processors/Formats/Impl/BinaryRowOutputFormat.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include @@ -35,9 +36,15 @@ void BinaryRowOutputFormat::writePrefix() if (with_types) { - for (size_t i = 0; i < columns; ++i) + if (format_settings.binary.encode_types_in_binary_format) { - writeStringBinary(header.safeGetByPosition(i).type->getName(), out); + for (size_t i = 0; i < columns; ++i) + encodeDataType(header.safeGetByPosition(i).type, out); + } + else + { + for (size_t i = 0; i < columns; ++i) + writeStringBinary(header.safeGetByPosition(i).type->getName(), out); } } } diff --git a/src/Processors/Formats/Impl/CSVRowInputFormat.cpp b/src/Processors/Formats/Impl/CSVRowInputFormat.cpp index dd7d6c6b024..b7f84748f61 100644 --- a/src/Processors/Formats/Impl/CSVRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/CSVRowInputFormat.cpp @@ -616,7 +616,7 @@ void registerCSVSchemaReader(FormatFactory & factory) { String result = getAdditionalFormatInfoByEscapingRule(settings, FormatSettings::EscapingRule::CSV); if (!with_names) - result += fmt::format(", column_names_for_schema_inference={}, try_detect_header={}", settings.column_names_for_schema_inference, settings.csv.try_detect_header); + result += fmt::format(", column_names_for_schema_inference={}, try_detect_header={}, skip_first_lines={}", settings.column_names_for_schema_inference, settings.csv.try_detect_header, settings.csv.skip_first_lines); return result; }); } diff --git a/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp b/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp index 9d056b42101..06e8668cd7c 100644 --- a/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp +++ b/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include @@ -227,7 +228,12 @@ private: return true; } - String column_name = "_dummy_" + std::to_string(replaced_literals.size()); + /// When generating placeholder names, ensure that we use names + /// requiring quotes to be valid identifiers. This prevents the + /// tuple() function from generating named tuples. Otherwise, + /// inserting named tuples with different names into another named + /// tuple will result in only default values being inserted. + String column_name = "-dummy-" + std::to_string(replaced_literals.size()); replaced_literals.emplace_back(literal, column_name, force_nullable); setDataType(replaced_literals.back()); ast = std::make_shared(column_name); diff --git a/src/Processors/Formats/Impl/FormRowInputFormat.cpp b/src/Processors/Formats/Impl/FormRowInputFormat.cpp index d3c6f3798cc..698efb896ea 100644 --- a/src/Processors/Formats/Impl/FormRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/FormRowInputFormat.cpp @@ -1,7 +1,9 @@ -#include +#include #include "Formats/EscapingRuleUtils.h" #include +#include + namespace DB { diff --git a/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp b/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp index 8855a1bc28d..0e3972e8c45 100644 --- a/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp @@ -46,6 +46,15 @@ JSONEachRowRowInputFormat::JSONEachRowRowInputFormat( { const auto & header = getPort().getHeader(); name_map = header.getNamesToIndexesMap(); + if (format_settings_.json.case_insensitive_column_matching) + { + for (auto & it : name_map) + { + StringRef key = it.first; + String lower_case_key = transformFieldNameToLowerCase(key); + lower_case_name_map[lower_case_key] = key; + } + } if (format_settings_.import_nested_json) { for (size_t i = 0; i != header.columns(); ++i) @@ -171,7 +180,15 @@ void JSONEachRowRowInputFormat::readJSONObject(MutableColumns & columns) skipUnknownField(name_ref); continue; } - const size_t column_index = columnIndex(name_ref, key_index); + size_t column_index = 0; + if (format_settings.json.case_insensitive_column_matching) + { + String lower_case_name = transformFieldNameToLowerCase(name_ref); + StringRef field_name_ref = lower_case_name_map[lower_case_name]; + column_index = columnIndex(field_name_ref, key_index); + } + else + column_index = columnIndex(name_ref, key_index); if (unlikely(ssize_t(column_index) < 0)) { diff --git a/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.h b/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.h index d97aa2dad8d..8a1cef8fa9f 100644 --- a/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.h +++ b/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.h @@ -55,7 +55,13 @@ private: virtual void readRowStart(MutableColumns &) {} virtual void skipRowStart() {} - + String transformFieldNameToLowerCase(const StringRef & field_name) + { + String field_name_str = field_name.toString(); + std::transform(field_name_str.begin(), field_name_str.end(), field_name_str.begin(), + [](unsigned char c) { return std::tolower(c); }); + return field_name_str; + } /// Buffer for the read from the stream field name. Used when you have to copy it. /// Also, if processing of Nested data is in progress, it holds the common prefix /// of the nested column names (so that appending the field name to it produces @@ -74,7 +80,8 @@ private: /// Hash table match `field name -> position in the block`. NOTE You can use perfect hash map. Block::NameMap name_map; - + /// Hash table match `lower_case field name -> field name in the block`. + std::unordered_map lower_case_name_map; /// Cached search results for previous row (keyed as index in JSON object) - used as a hint. std::vector prev_positions; diff --git a/src/Processors/Formats/Impl/NativeFormat.cpp b/src/Processors/Formats/Impl/NativeFormat.cpp index a7a49ab6a8c..38fac60eef6 100644 --- a/src/Processors/Formats/Impl/NativeFormat.cpp +++ b/src/Processors/Formats/Impl/NativeFormat.cpp @@ -21,9 +21,7 @@ public: buf, header_, 0, - settings.skip_unknown_fields, - settings.null_as_default, - settings.native.allow_types_conversion, + settings, settings.defaults_for_omitted_fields ? &block_missing_values : nullptr)) , header(header_) {} @@ -72,9 +70,9 @@ private: class NativeOutputFormat final : public IOutputFormat { public: - NativeOutputFormat(WriteBuffer & buf, const Block & header, UInt64 client_protocol_version = 0) + NativeOutputFormat(WriteBuffer & buf, const Block & header, const FormatSettings & settings, UInt64 client_protocol_version = 0) : IOutputFormat(header, buf) - , writer(buf, client_protocol_version, header) + , writer(buf, client_protocol_version, header, settings) { } @@ -103,14 +101,17 @@ private: class NativeSchemaReader : public ISchemaReader { public: - explicit NativeSchemaReader(ReadBuffer & in_) : ISchemaReader(in_) {} + explicit NativeSchemaReader(ReadBuffer & in_, const FormatSettings & settings_) : ISchemaReader(in_), settings(settings_) {} NamesAndTypesList readSchema() override { - auto reader = NativeReader(in, 0); + auto reader = NativeReader(in, 0, settings); auto block = reader.read(); return block.getNamesAndTypesList(); } + +private: + const FormatSettings settings; }; @@ -134,16 +135,16 @@ void registerOutputFormatNative(FormatFactory & factory) const Block & sample, const FormatSettings & settings) { - return std::make_shared(buf, sample, settings.client_protocol_version); + return std::make_shared(buf, sample, settings, settings.client_protocol_version); }); } void registerNativeSchemaReader(FormatFactory & factory) { - factory.registerSchemaReader("Native", [](ReadBuffer & buf, const FormatSettings &) + factory.registerSchemaReader("Native", [](ReadBuffer & buf, const FormatSettings & settings) { - return std::make_shared(buf); + return std::make_shared(buf, settings); }); } diff --git a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp index dcd5a531b05..c10969b02b7 100644 --- a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp @@ -900,6 +900,11 @@ bool NativeORCBlockInputFormat::prepareStripeReader() orc::RowReaderOptions row_reader_options; row_reader_options.includeTypes(include_indices); + if (format_settings.orc.read_use_writer_time_zone) + { + String writer_time_zone = current_stripe_info->getWriterTimezone(); + row_reader_options.setTimezoneName(writer_time_zone); + } row_reader_options.range(current_stripe_info->getOffset(), current_stripe_info->getLength()); if (format_settings.orc.filter_push_down && sarg) { diff --git a/src/Processors/Formats/Impl/NpyRowInputFormat.cpp b/src/Processors/Formats/Impl/NpyRowInputFormat.cpp index 65e0f9dd192..773cbc9268e 100644 --- a/src/Processors/Formats/Impl/NpyRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/NpyRowInputFormat.cpp @@ -445,6 +445,9 @@ bool NpyRowInputFormat::readRow(MutableColumns & columns, RowReadExtension & /* elements_in_current_column *= header.shape[i]; } + if (typeid_cast(current_column)) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected nesting level of column '{}', expected {}", column->getName(), header.shape.size() - 1); + for (size_t i = 0; i != elements_in_current_column; ++i) readValue(current_column); diff --git a/src/Processors/Formats/Impl/Parquet/ParquetRecordReader.cpp b/src/Processors/Formats/Impl/Parquet/ParquetRecordReader.cpp index 9a15789f267..18a0db7484e 100644 --- a/src/Processors/Formats/Impl/Parquet/ParquetRecordReader.cpp +++ b/src/Processors/Formats/Impl/Parquet/ParquetRecordReader.cpp @@ -113,7 +113,7 @@ private: { throw Exception( ErrorCodes::PARQUET_EXCEPTION, - "Unsupported logical type: {} and physical type: {} for field =={}=={}", + "Unsupported logical type: {} and physical type: {} for field `{}`{}", col_descriptor.logical_type()->ToString(), col_descriptor.physical_type(), col_descriptor.name(), msg); } }; diff --git a/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp b/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp index a5d334f4f1d..9e499e2c400 100644 --- a/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp @@ -179,7 +179,9 @@ void ParquetBlockOutputFormat::consume(Chunk chunk) columns[i]->insertRangeFrom(*concatenated.getColumns()[i], offset, count); Chunks piece; - piece.emplace_back(std::move(columns), count, concatenated.getChunkInfo()); + piece.emplace_back(std::move(columns), count); + piece.back().setChunkInfos(concatenated.getChunkInfos()); + writeRowGroup(std::move(piece)); } } diff --git a/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h b/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h index 4c52300fbd1..698efecd4b2 100644 --- a/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h +++ b/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h @@ -74,7 +74,8 @@ void registerPrettyFormatWithNoEscapesAndMonoBlock(FormatFactory & factory, cons const Block & sample, const FormatSettings & format_settings) { - bool color = !no_escapes && format_settings.pretty.color.valueOr(format_settings.is_writing_to_terminal); + bool color = !no_escapes + && (format_settings.pretty.color == 1 || (format_settings.pretty.color == 2 && format_settings.is_writing_to_terminal)); return std::make_shared(buf, sample, format_settings, mono_block, color); }); if (!mono_block) diff --git a/src/Processors/Formats/Impl/TabSeparatedRowInputFormat.cpp b/src/Processors/Formats/Impl/TabSeparatedRowInputFormat.cpp index 6d4dcba9e60..d2e17e92924 100644 --- a/src/Processors/Formats/Impl/TabSeparatedRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/TabSeparatedRowInputFormat.cpp @@ -440,9 +440,10 @@ void registerTSVSchemaReader(FormatFactory & factory) settings, is_raw ? FormatSettings::EscapingRule::Raw : FormatSettings::EscapingRule::Escaped); if (!with_names) result += fmt::format( - ", column_names_for_schema_inference={}, try_detect_header={}", + ", column_names_for_schema_inference={}, try_detect_header={}, skip_first_lines={}", settings.column_names_for_schema_inference, - settings.tsv.try_detect_header); + settings.tsv.try_detect_header, + settings.tsv.skip_first_lines); return result; }); } diff --git a/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp b/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp index 1493779ec2d..c23d717d52f 100644 --- a/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Processors/Formats/InputFormatErrorsLogger.cpp b/src/Processors/Formats/InputFormatErrorsLogger.cpp index 814c4679cf9..b5aa936df07 100644 --- a/src/Processors/Formats/InputFormatErrorsLogger.cpp +++ b/src/Processors/Formats/InputFormatErrorsLogger.cpp @@ -5,6 +5,7 @@ #include #include #include +#include namespace DB diff --git a/src/Processors/IAccumulatingTransform.cpp b/src/Processors/IAccumulatingTransform.cpp index 4136fc5a5f2..46be6e74693 100644 --- a/src/Processors/IAccumulatingTransform.cpp +++ b/src/Processors/IAccumulatingTransform.cpp @@ -8,8 +8,9 @@ namespace ErrorCodes } IAccumulatingTransform::IAccumulatingTransform(Block input_header, Block output_header) - : IProcessor({std::move(input_header)}, {std::move(output_header)}), - input(inputs.front()), output(outputs.front()) + : IProcessor({std::move(input_header)}, {std::move(output_header)}) + , input(inputs.front()) + , output(outputs.front()) { } diff --git a/src/Processors/IProcessor.h b/src/Processors/IProcessor.h index 6f779e7a8d4..02f7b6b3d12 100644 --- a/src/Processors/IProcessor.h +++ b/src/Processors/IProcessor.h @@ -303,9 +303,9 @@ public: IQueryPlanStep * getQueryPlanStep() const { return query_plan_step; } size_t getQueryPlanStepGroup() const { return query_plan_step_group; } - uint64_t getElapsedUs() const { return elapsed_us; } - uint64_t getInputWaitElapsedUs() const { return input_wait_elapsed_us; } - uint64_t getOutputWaitElapsedUs() const { return output_wait_elapsed_us; } + uint64_t getElapsedNs() const { return elapsed_ns; } + uint64_t getInputWaitElapsedNs() const { return input_wait_elapsed_ns; } + uint64_t getOutputWaitElapsedNs() const { return output_wait_elapsed_ns; } struct ProcessorDataStats { @@ -369,21 +369,21 @@ protected: private: /// For: - /// - elapsed_us + /// - elapsed_ns friend class ExecutionThreadContext; /// For - /// - input_wait_elapsed_us - /// - output_wait_elapsed_us + /// - input_wait_elapsed_ns + /// - output_wait_elapsed_ns friend class ExecutingGraph; std::string processor_description; /// For processors_profile_log - uint64_t elapsed_us = 0; + uint64_t elapsed_ns = 0; Stopwatch input_wait_watch; - uint64_t input_wait_elapsed_us = 0; + uint64_t input_wait_elapsed_ns = 0; Stopwatch output_wait_watch; - uint64_t output_wait_elapsed_us = 0; + uint64_t output_wait_elapsed_ns = 0; size_t stream_number = NO_STREAM; diff --git a/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.cpp b/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.cpp index 466adf93538..86675bcb237 100644 --- a/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.cpp @@ -53,13 +53,11 @@ void FinishAggregatingInOrderAlgorithm::consume(Input & input, size_t source_num if (!input.chunk.hasRows()) return; - const auto & info = input.chunk.getChunkInfo(); - if (!info) + if (input.chunk.getChunkInfos().empty()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk info was not set for chunk in FinishAggregatingInOrderAlgorithm"); Int64 allocated_bytes = 0; - /// Will be set by AggregatingInOrderTransform during local aggregation; will be nullptr during merging on initiator. - if (const auto * arenas_info = typeid_cast(info.get())) + if (auto arenas_info = input.chunk.getChunkInfos().get()) allocated_bytes = arenas_info->allocated_bytes; states[source_num] = State{input.chunk, description, allocated_bytes}; @@ -136,7 +134,7 @@ Chunk FinishAggregatingInOrderAlgorithm::prepareToMerge() info->chunk_num = chunk_num++; Chunk chunk; - chunk.setChunkInfo(std::move(info)); + chunk.getChunkInfos().add(std::move(info)); return chunk; } @@ -163,7 +161,7 @@ void FinishAggregatingInOrderAlgorithm::addToAggregation() chunks.emplace_back(std::move(new_columns), current_rows); } - chunks.back().setChunkInfo(std::make_shared()); + chunks.back().getChunkInfos().add(std::make_shared()); states[i].current_row = states[i].to_row; /// We assume that sizes in bytes of rows are almost the same. diff --git a/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h b/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h index bcf4e759024..e4f22deec8d 100644 --- a/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h +++ b/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h @@ -6,18 +6,22 @@ namespace DB { /// To carry part level if chunk is produced by a merge tree source -class MergeTreePartLevelInfo : public ChunkInfo +class MergeTreePartLevelInfo : public ChunkInfoCloneable { public: MergeTreePartLevelInfo() = delete; - explicit MergeTreePartLevelInfo(ssize_t part_level) : origin_merge_tree_part_level(part_level) { } + explicit MergeTreePartLevelInfo(ssize_t part_level) + : origin_merge_tree_part_level(part_level) + { } + MergeTreePartLevelInfo(const MergeTreePartLevelInfo & other) = default; + size_t origin_merge_tree_part_level = 0; }; inline size_t getPartLevelFromChunk(const Chunk & chunk) { - const auto & info = chunk.getChunkInfo(); - if (const auto * part_level_info = typeid_cast(info.get())) + const auto part_level_info = chunk.getChunkInfos().get(); + if (part_level_info) return part_level_info->origin_merge_tree_part_level; return 0; } diff --git a/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.cpp index 7b2c7d82a01..cd347d371d9 100644 --- a/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.cpp @@ -17,7 +17,7 @@ namespace ErrorCodes static IMergingAlgorithm::Status emitChunk(detail::SharedChunkPtr & chunk, bool finished = false) { - chunk->setChunkInfo(std::make_shared(std::move(chunk->replace_final_selection))); + chunk->getChunkInfos().add(std::make_shared(std::move(chunk->replace_final_selection))); return IMergingAlgorithm::Status(std::move(*chunk), finished); } diff --git a/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h index a3ccccf0845..2f23f2a5c4d 100644 --- a/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h +++ b/src/Processors/Merges/Algorithms/ReplacingSortedAlgorithm.h @@ -3,6 +3,7 @@ #include #include #include +#include namespace Poco { @@ -14,11 +15,13 @@ namespace DB /** Use in skipping final to keep list of indices of selected row after merging final */ -struct ChunkSelectFinalIndices : public ChunkInfo +struct ChunkSelectFinalIndices : public ChunkInfoCloneable { + explicit ChunkSelectFinalIndices(MutableColumnPtr select_final_indices_); + ChunkSelectFinalIndices(const ChunkSelectFinalIndices & other) = default; + const ColumnPtr column_holder; const ColumnUInt64 * select_final_indices = nullptr; - explicit ChunkSelectFinalIndices(MutableColumnPtr select_final_indices_); }; /** Merges several sorted inputs into one. diff --git a/src/Processors/Merges/IMergingTransform.cpp b/src/Processors/Merges/IMergingTransform.cpp index fbb47969b2f..b1b0182a113 100644 --- a/src/Processors/Merges/IMergingTransform.cpp +++ b/src/Processors/Merges/IMergingTransform.cpp @@ -157,7 +157,7 @@ IProcessor::Status IMergingTransformBase::prepare() bool is_port_full = !output.canPush(); /// Push if has data. - if ((state.output_chunk || state.output_chunk.hasChunkInfo()) && !is_port_full) + if ((state.output_chunk || !state.output_chunk.getChunkInfos().empty()) && !is_port_full) output.push(std::move(state.output_chunk)); if (!is_initialized) diff --git a/src/Processors/Merges/IMergingTransform.h b/src/Processors/Merges/IMergingTransform.h index c218f622870..be629271736 100644 --- a/src/Processors/Merges/IMergingTransform.h +++ b/src/Processors/Merges/IMergingTransform.h @@ -129,7 +129,7 @@ public: IMergingAlgorithm::Status status = algorithm.merge(); - if ((status.chunk && status.chunk.hasRows()) || status.chunk.hasChunkInfo()) + if ((status.chunk && status.chunk.hasRows()) || !status.chunk.getChunkInfos().empty()) { // std::cerr << "Got chunk with " << status.chunk.getNumRows() << " rows" << std::endl; state.output_chunk = std::move(status.chunk); diff --git a/src/Processors/QueryPlan/BufferChunksTransform.cpp b/src/Processors/QueryPlan/BufferChunksTransform.cpp new file mode 100644 index 00000000000..3601a68d36e --- /dev/null +++ b/src/Processors/QueryPlan/BufferChunksTransform.cpp @@ -0,0 +1,85 @@ +#include + +namespace DB +{ + +BufferChunksTransform::BufferChunksTransform( + const Block & header_, + size_t max_rows_to_buffer_, + size_t max_bytes_to_buffer_, + size_t limit_) + : IProcessor({header_}, {header_}) + , input(inputs.front()) + , output(outputs.front()) + , max_rows_to_buffer(max_rows_to_buffer_) + , max_bytes_to_buffer(max_bytes_to_buffer_) + , limit(limit_) +{ +} + +IProcessor::Status BufferChunksTransform::prepare() +{ + if (output.isFinished()) + { + chunks = {}; + input.close(); + return Status::Finished; + } + + if (input.isFinished() && chunks.empty()) + { + output.finish(); + return Status::Finished; + } + + if (output.canPush()) + { + input.setNeeded(); + + if (!chunks.empty()) + { + auto chunk = std::move(chunks.front()); + chunks.pop(); + + num_buffered_rows -= chunk.getNumRows(); + num_buffered_bytes -= chunk.bytes(); + + output.push(std::move(chunk)); + } + else if (input.hasData()) + { + auto chunk = pullChunk(); + output.push(std::move(chunk)); + } + } + + if (input.hasData() && (num_buffered_rows < max_rows_to_buffer || num_buffered_bytes < max_bytes_to_buffer)) + { + auto chunk = pullChunk(); + num_buffered_rows += chunk.getNumRows(); + num_buffered_bytes += chunk.bytes(); + chunks.push(std::move(chunk)); + } + + if (num_buffered_rows >= max_rows_to_buffer && num_buffered_bytes >= max_bytes_to_buffer) + { + input.setNotNeeded(); + return Status::PortFull; + } + + input.setNeeded(); + return Status::NeedData; +} + +Chunk BufferChunksTransform::pullChunk() +{ + auto chunk = input.pull(); + num_processed_rows += chunk.getNumRows(); + + if (limit && num_processed_rows >= limit) + input.close(); + + return chunk; +} + +} diff --git a/src/Processors/QueryPlan/BufferChunksTransform.h b/src/Processors/QueryPlan/BufferChunksTransform.h new file mode 100644 index 00000000000..752f9910734 --- /dev/null +++ b/src/Processors/QueryPlan/BufferChunksTransform.h @@ -0,0 +1,42 @@ +#pragma once +#include +#include + +namespace DB +{ + +/// Transform that buffers chunks from the input +/// up to the certain limit and pushes chunks to +/// the output whenever it is ready. It can be used +/// to increase parallelism of execution, for example +/// when it is adeded before MergingSortedTransform. +class BufferChunksTransform : public IProcessor +{ +public: + /// OR condition is used for the limits on rows and bytes. + BufferChunksTransform( + const Block & header_, + size_t max_rows_to_buffer_, + size_t max_bytes_to_buffer_, + size_t limit_); + + Status prepare() override; + String getName() const override { return "BufferChunks"; } + +private: + Chunk pullChunk(); + + InputPort & input; + OutputPort & output; + + size_t max_rows_to_buffer; + size_t max_bytes_to_buffer; + size_t limit; + + std::queue chunks; + size_t num_buffered_rows = 0; + size_t num_buffered_bytes = 0; + size_t num_processed_rows = 0; +}; + +} diff --git a/src/Processors/QueryPlan/CreateSetAndFilterOnTheFlyStep.h b/src/Processors/QueryPlan/CreateSetAndFilterOnTheFlyStep.h index 70ae4c6156e..27171511703 100644 --- a/src/Processors/QueryPlan/CreateSetAndFilterOnTheFlyStep.h +++ b/src/Processors/QueryPlan/CreateSetAndFilterOnTheFlyStep.h @@ -1,4 +1,6 @@ #pragma once + +#include #include #include diff --git a/src/Processors/QueryPlan/DistributedCreateLocalPlan.cpp b/src/Processors/QueryPlan/DistributedCreateLocalPlan.cpp index 1f4f271fa6e..14f58585d3d 100644 --- a/src/Processors/QueryPlan/DistributedCreateLocalPlan.cpp +++ b/src/Processors/QueryPlan/DistributedCreateLocalPlan.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include diff --git a/src/Processors/QueryPlan/Optimizations/Optimizations.h b/src/Processors/QueryPlan/Optimizations/Optimizations.h index b1ab5561958..c48bdf1552a 100644 --- a/src/Processors/QueryPlan/Optimizations/Optimizations.h +++ b/src/Processors/QueryPlan/Optimizations/Optimizations.h @@ -46,6 +46,10 @@ size_t trySplitFilter(QueryPlan::Node * node, QueryPlan::Nodes & nodes); /// Replace chain `FilterStep -> ExpressionStep` to single FilterStep size_t tryMergeExpressions(QueryPlan::Node * parent_node, QueryPlan::Nodes &); +/// Replace chain `FilterStep -> FilterStep` to single FilterStep +/// Note: this breaks short-circuit logic, so it is disabled for now. +size_t tryMergeFilters(QueryPlan::Node * parent_node, QueryPlan::Nodes &); + /// Move FilterStep down if possible. /// May split FilterStep and push down only part of it. size_t tryPushDownFilter(QueryPlan::Node * parent_node, QueryPlan::Nodes & nodes); @@ -81,11 +85,12 @@ size_t tryAggregatePartitionsIndependently(QueryPlan::Node * node, QueryPlan::No inline const auto & getOptimizations() { - static const std::array optimizations = {{ + static const std::array optimizations = {{ {tryLiftUpArrayJoin, "liftUpArrayJoin", &QueryPlanOptimizationSettings::lift_up_array_join}, {tryPushDownLimit, "pushDownLimit", &QueryPlanOptimizationSettings::push_down_limit}, {trySplitFilter, "splitFilter", &QueryPlanOptimizationSettings::split_filter}, {tryMergeExpressions, "mergeExpressions", &QueryPlanOptimizationSettings::merge_expressions}, + {tryMergeFilters, "mergeFilters", &QueryPlanOptimizationSettings::merge_filters}, {tryPushDownFilter, "pushDownFilter", &QueryPlanOptimizationSettings::filter_push_down}, {tryConvertOuterJoinToInnerJoin, "convertOuterJoinToInnerJoin", &QueryPlanOptimizationSettings::convert_outer_join_to_inner_join}, {tryExecuteFunctionsAfterSorting, "liftUpFunctions", &QueryPlanOptimizationSettings::execute_functions_after_sorting}, diff --git a/src/Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.cpp b/src/Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.cpp index 2738de1ff5f..4d984133efd 100644 --- a/src/Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.cpp +++ b/src/Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.cpp @@ -20,6 +20,8 @@ QueryPlanOptimizationSettings QueryPlanOptimizationSettings::fromSettings(const settings.merge_expressions = from.query_plan_enable_optimizations && from.query_plan_merge_expressions; + settings.merge_filters = from.query_plan_enable_optimizations && from.query_plan_merge_filters; + settings.filter_push_down = from.query_plan_enable_optimizations && from.query_plan_filter_push_down; settings.convert_outer_join_to_inner_join = from.query_plan_enable_optimizations && from.query_plan_convert_outer_join_to_inner_join; diff --git a/src/Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.h b/src/Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.h index 85042cea4ed..539ff2eafbb 100644 --- a/src/Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.h +++ b/src/Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.h @@ -31,6 +31,9 @@ struct QueryPlanOptimizationSettings /// If merge-expressions optimization is enabled. bool merge_expressions = true; + /// If merge-filters optimization is enabled. + bool merge_filters = false; + /// If filter push down optimization is enabled. bool filter_push_down = true; diff --git a/src/Processors/QueryPlan/Optimizations/convertOuterJoinToInnerJoin.cpp b/src/Processors/QueryPlan/Optimizations/convertOuterJoinToInnerJoin.cpp index d90f0e152e7..d9296f10a98 100644 --- a/src/Processors/QueryPlan/Optimizations/convertOuterJoinToInnerJoin.cpp +++ b/src/Processors/QueryPlan/Optimizations/convertOuterJoinToInnerJoin.cpp @@ -23,7 +23,10 @@ size_t tryConvertOuterJoinToInnerJoin(QueryPlan::Node * parent_node, QueryPlan:: return 0; const auto & table_join = join->getJoin()->getTableJoin(); - if (table_join.strictness() == JoinStrictness::Asof) + + /// Any JOIN issue https://github.com/ClickHouse/ClickHouse/issues/66447 + /// Anti JOIN issue https://github.com/ClickHouse/ClickHouse/issues/67156 + if (table_join.strictness() != JoinStrictness::All) return 0; /// TODO: Support join_use_nulls diff --git a/src/Processors/QueryPlan/Optimizations/mergeExpressions.cpp b/src/Processors/QueryPlan/Optimizations/mergeExpressions.cpp index 6ace1b3b5ce..118abdd701f 100644 --- a/src/Processors/QueryPlan/Optimizations/mergeExpressions.cpp +++ b/src/Processors/QueryPlan/Optimizations/mergeExpressions.cpp @@ -34,7 +34,6 @@ size_t tryMergeExpressions(QueryPlan::Node * parent_node, QueryPlan::Nodes &) auto * parent_expr = typeid_cast(parent.get()); auto * parent_filter = typeid_cast(parent.get()); auto * child_expr = typeid_cast(child.get()); - auto * child_filter = typeid_cast(child.get()); if (parent_expr && child_expr) { @@ -76,7 +75,23 @@ size_t tryMergeExpressions(QueryPlan::Node * parent_node, QueryPlan::Nodes &) parent_node->children.swap(child_node->children); return 1; } - else if (parent_filter && child_filter) + + return 0; +} +size_t tryMergeFilters(QueryPlan::Node * parent_node, QueryPlan::Nodes &) +{ + if (parent_node->children.size() != 1) + return false; + + QueryPlan::Node * child_node = parent_node->children.front(); + + auto & parent = parent_node->step; + auto & child = child_node->step; + + auto * parent_filter = typeid_cast(parent.get()); + auto * child_filter = typeid_cast(child.get()); + + if (parent_filter && child_filter) { const auto & child_actions = child_filter->getExpression(); const auto & parent_actions = parent_filter->getExpression(); diff --git a/src/Processors/QueryPlan/Optimizations/optimizePrewhere.cpp b/src/Processors/QueryPlan/Optimizations/optimizePrewhere.cpp index 1badd315200..3cb7f8a5d83 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizePrewhere.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizePrewhere.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -8,6 +9,7 @@ #include #include #include + namespace DB { diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index 537555afa2a..28eb4da2e17 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -22,9 +22,10 @@ #include #include #include -#include "Storages/KeyDescription.h" +#include #include #include +#include #include @@ -919,15 +920,23 @@ void optimizeReadInOrder(QueryPlan::Node & node, QueryPlan::Nodes & nodes) { auto & union_node = node.children.front(); - std::vector infos; + bool use_buffering = false; const SortDescription * max_sort_descr = nullptr; + + std::vector infos; infos.reserve(node.children.size()); + for (auto * child : union_node->children) { infos.push_back(buildInputOrderInfo(*sorting, *child, steps_to_update)); - if (infos.back() && (!max_sort_descr || max_sort_descr->size() < infos.back()->sort_description_for_merging.size())) - max_sort_descr = &infos.back()->sort_description_for_merging; + if (infos.back()) + { + if (!max_sort_descr || max_sort_descr->size() < infos.back()->sort_description_for_merging.size()) + max_sort_descr = &infos.back()->sort_description_for_merging; + + use_buffering |= infos.back()->limit == 0; + } } if (!max_sort_descr || max_sort_descr->empty()) @@ -972,12 +981,13 @@ void optimizeReadInOrder(QueryPlan::Node & node, QueryPlan::Nodes & nodes) } } - sorting->convertToFinishSorting(*max_sort_descr); + sorting->convertToFinishSorting(*max_sort_descr, use_buffering); } else if (auto order_info = buildInputOrderInfo(*sorting, *node.children.front(), steps_to_update)) { - sorting->convertToFinishSorting(order_info->sort_description_for_merging); - /// update data stream's sorting properties + /// Use buffering only if have filter or don't have limit. + bool use_buffering = order_info->limit == 0; + sorting->convertToFinishSorting(order_info->sort_description_for_merging, use_buffering); updateStepsDataStreams(steps_to_update); } } @@ -1091,7 +1101,7 @@ size_t tryReuseStorageOrderingForWindowFunctions(QueryPlan::Node * parent_node, bool can_read = read_from_merge_tree->requestReadingInOrder(order_info->used_prefix_of_sorting_key_size, order_info->direction, order_info->limit); if (!can_read) return 0; - sorting->convertToFinishSorting(order_info->sort_description_for_merging); + sorting->convertToFinishSorting(order_info->sort_description_for_merging, false); } return 0; diff --git a/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp b/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp index 70327bc95b4..9e2c2dcc96d 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp @@ -17,6 +17,7 @@ #include #include +#include #include #include #include @@ -515,10 +516,11 @@ AggregateProjectionCandidates getAggregateProjectionCandidates( if (!candidates.minmax_projection) { - auto it = std::find_if(agg_projections.begin(), agg_projections.end(), [&](const auto * projection) - { - return projection->name == context->getSettings().preferred_optimize_projection_name.value; - }); + auto it = std::find_if( + agg_projections.begin(), + agg_projections.end(), + [&](const auto * projection) + { return projection->name == context->getSettingsRef().preferred_optimize_projection_name.value; }); if (it != agg_projections.end()) { diff --git a/src/Processors/QueryPlan/Optimizations/optimizeUseNormalProjection.cpp b/src/Processors/QueryPlan/Optimizations/optimizeUseNormalProjection.cpp index 0af3869ccf1..291800c2c86 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeUseNormalProjection.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeUseNormalProjection.cpp @@ -7,9 +7,11 @@ #include #include #include +#include #include #include #include + #include namespace DB::QueryPlanOptimizations @@ -110,10 +112,10 @@ std::optional optimizeUseNormalProjections(Stack & stack, QueryPlan::Nod return {}; ContextPtr context = reading->getContext(); - auto it = std::find_if(normal_projections.begin(), normal_projections.end(), [&](const auto * projection) - { - return projection->name == context->getSettings().preferred_optimize_projection_name.value; - }); + auto it = std::find_if( + normal_projections.begin(), + normal_projections.end(), + [&](const auto * projection) { return projection->name == context->getSettingsRef().preferred_optimize_projection_name.value; }); if (it != normal_projections.end()) { diff --git a/src/Processors/QueryPlan/Optimizations/projectionsCommon.cpp b/src/Processors/QueryPlan/Optimizations/projectionsCommon.cpp index af1578d6af8..4d60a1135a6 100644 --- a/src/Processors/QueryPlan/Optimizations/projectionsCommon.cpp +++ b/src/Processors/QueryPlan/Optimizations/projectionsCommon.cpp @@ -5,6 +5,7 @@ #include #include +#include #include #include #include diff --git a/src/Processors/QueryPlan/ReadFromMemoryStorageStep.cpp b/src/Processors/QueryPlan/ReadFromMemoryStorageStep.cpp index 2e7693b1b36..6dc0c021a14 100644 --- a/src/Processors/QueryPlan/ReadFromMemoryStorageStep.cpp +++ b/src/Processors/QueryPlan/ReadFromMemoryStorageStep.cpp @@ -30,12 +30,15 @@ public: std::shared_ptr> parallel_execution_index_, InitializerFunc initializer_func_ = {}) : ISource(storage_snapshot->getSampleBlockForColumns(column_names_)) - , column_names_and_types(storage_snapshot->getColumnsByNames( + , requested_column_names_and_types(storage_snapshot->getColumnsByNames( GetColumnsOptions(GetColumnsOptions::All).withSubcolumns().withExtendedObjects(), column_names_)) , data(data_) , parallel_execution_index(parallel_execution_index_) , initializer_func(std::move(initializer_func_)) { + auto all_column_names_and_types = storage_snapshot->getColumns(GetColumnsOptions(GetColumnsOptions::All).withSubcolumns().withExtendedObjects()); + for (const auto & [name, type] : all_column_names_and_types) + all_names_to_types[name] = type; } String getName() const override { return "Memory"; } @@ -59,17 +62,20 @@ protected: const Block & src = (*data)[current_index]; Columns columns; - size_t num_columns = column_names_and_types.size(); + size_t num_columns = requested_column_names_and_types.size(); columns.reserve(num_columns); - auto name_and_type = column_names_and_types.begin(); + auto name_and_type = requested_column_names_and_types.begin(); for (size_t i = 0; i < num_columns; ++i) { - columns.emplace_back(tryGetColumnFromBlock(src, *name_and_type)); + if (name_and_type->isSubcolumn()) + columns.emplace_back(tryGetSubcolumnFromBlock(src, all_names_to_types[name_and_type->getNameInStorage()], *name_and_type)); + else + columns.emplace_back(tryGetColumnFromBlock(src, *name_and_type)); ++name_and_type; } - fillMissingColumns(columns, src.rows(), column_names_and_types, column_names_and_types, {}, nullptr); + fillMissingColumns(columns, src.rows(), requested_column_names_and_types, requested_column_names_and_types, {}, nullptr); assert(std::all_of(columns.begin(), columns.end(), [](const auto & column) { return column != nullptr; })); return Chunk(std::move(columns), src.rows()); @@ -88,7 +94,9 @@ private: } } - const NamesAndTypesList column_names_and_types; + const NamesAndTypesList requested_column_names_and_types; + /// Map (name -> type) for all columns from the storage header. + std::unordered_map all_names_to_types; size_t execution_index = 0; std::shared_ptr data; std::shared_ptr> parallel_execution_index; diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index aba3f6ff2da..a7ce92fe8e0 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include @@ -40,6 +41,7 @@ #include #include #include +#include #include #include #include @@ -250,9 +252,9 @@ void ReadFromMergeTree::AnalysisResult::checkLimits(const Settings & settings, c { /// Fail fast if estimated number of rows to read exceeds the limit size_t total_rows_estimate = selected_rows; - if (query_info_.limit > 0 && total_rows_estimate > query_info_.limit) + if (query_info_.trivial_limit > 0 && total_rows_estimate > query_info_.trivial_limit) { - total_rows_estimate = query_info_.limit; + total_rows_estimate = query_info_.trivial_limit; } limits.check(total_rows_estimate, 0, "rows (controlled by 'max_rows_to_read' setting)", ErrorCodes::TOO_MANY_ROWS); leaf_limits.check( @@ -383,7 +385,7 @@ Pipe ReadFromMergeTree::readFromPoolParallelReplicas( pool, std::move(algorithm), prewhere_info, actions_settings, block_size_copy, reader_settings); - auto source = std::make_shared(std::move(processor)); + auto source = std::make_shared(std::move(processor), data.getLogName()); pipes.emplace_back(std::move(source)); } @@ -398,8 +400,8 @@ Pipe ReadFromMergeTree::readFromPool( { size_t total_rows = parts_with_range.getRowsCountAllParts(); - if (query_info.limit > 0 && query_info.limit < total_rows) - total_rows = query_info.limit; + if (query_info.trivial_limit > 0 && query_info.trivial_limit < total_rows) + total_rows = query_info.trivial_limit; const auto & settings = context->getSettingsRef(); @@ -436,7 +438,7 @@ Pipe ReadFromMergeTree::readFromPool( * Because time spend during filling per thread tasks can be greater than whole query * execution for big tables with small limit. */ - bool use_prefetched_read_pool = query_info.limit == 0 && (allow_prefetched_remote || allow_prefetched_local); + bool use_prefetched_read_pool = query_info.trivial_limit == 0 && (allow_prefetched_remote || allow_prefetched_local); if (use_prefetched_read_pool) { @@ -482,7 +484,7 @@ Pipe ReadFromMergeTree::readFromPool( pool, std::move(algorithm), prewhere_info, actions_settings, block_size_copy, reader_settings); - auto source = std::make_shared(std::move(processor)); + auto source = std::make_shared(std::move(processor), data.getLogName()); if (i == 0) source->addTotalRowsApprox(total_rows); @@ -563,9 +565,8 @@ Pipe ReadFromMergeTree::readInOrder( /// Actually it means that parallel reading from replicas enabled /// and we have to collaborate with initiator. /// In this case we won't set approximate rows, because it will be accounted multiple times. - /// Also do not count amount of read rows if we read in order of sorting key, - /// because we don't know actual amount of read rows in case when limit is set. - bool set_rows_approx = !is_parallel_reading_from_replicas && !reader_settings.read_in_order; + const auto in_order_limit = query_info.input_order_info ? query_info.input_order_info->limit : 0; + const bool set_total_rows_approx = !is_parallel_reading_from_replicas; Pipes pipes; for (size_t i = 0; i < parts_with_ranges.size(); ++i) @@ -573,8 +574,10 @@ Pipe ReadFromMergeTree::readInOrder( const auto & part_with_ranges = parts_with_ranges[i]; UInt64 total_rows = part_with_ranges.getRowsCount(); - if (query_info.limit > 0 && query_info.limit < total_rows) - total_rows = query_info.limit; + if (query_info.trivial_limit > 0 && query_info.trivial_limit < total_rows) + total_rows = query_info.trivial_limit; + else if (in_order_limit > 0 && in_order_limit < total_rows) + total_rows = in_order_limit; LOG_TRACE(log, "Reading {} ranges in{}order from part {}, approx. {} rows starting from {}", part_with_ranges.ranges.size(), @@ -594,8 +597,8 @@ Pipe ReadFromMergeTree::readInOrder( processor->addPartLevelToChunk(isQueryWithFinal()); - auto source = std::make_shared(std::move(processor)); - if (set_rows_approx) + auto source = std::make_shared(std::move(processor), data.getLogName()); + if (set_total_rows_approx) source->addTotalRowsApprox(total_rows); pipes.emplace_back(std::move(source)); @@ -940,6 +943,8 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( PoolSettings pool_settings { + .threads = num_streams, + .sum_marks = parts_with_ranges.getMarksCountAllParts(), .min_marks_for_concurrent_read = info.min_marks_for_concurrent_read, .preferred_block_size_bytes = settings.preferred_block_size_bytes, .use_uncompressed_cache = info.use_uncompressed_cache, diff --git a/src/Processors/QueryPlan/ReadFromRemote.cpp b/src/Processors/QueryPlan/ReadFromRemote.cpp index 84c2515e8ca..cf11052cd59 100644 --- a/src/Processors/QueryPlan/ReadFromRemote.cpp +++ b/src/Processors/QueryPlan/ReadFromRemote.cpp @@ -16,11 +16,12 @@ #include #include #include +#include #include #include #include #include - +#include #include namespace DB @@ -361,7 +362,6 @@ ReadFromParallelRemoteReplicasStep::ReadFromParallelRemoteReplicasStep( ASTPtr query_ast_, ClusterPtr cluster_, const StorageID & storage_id_, - ParallelReplicasReadingCoordinatorPtr coordinator_, Block header_, QueryProcessingStage::Enum stage_, ContextMutablePtr context_, @@ -374,7 +374,6 @@ ReadFromParallelRemoteReplicasStep::ReadFromParallelRemoteReplicasStep( , cluster(cluster_) , query_ast(query_ast_) , storage_id(storage_id_) - , coordinator(std::move(coordinator_)) , stage(std::move(stage_)) , context(context_) , throttler(throttler_) @@ -411,8 +410,8 @@ void ReadFromParallelRemoteReplicasStep::initializePipeline(QueryPipelineBuilder auto timeouts = ConnectionTimeouts::getTCPTimeoutsWithFailover(current_settings); const auto & shard = cluster->getShardsInfo().at(0); - size_t all_replicas_count = current_settings.max_parallel_replicas; - if (all_replicas_count > shard.getAllNodeCount()) + size_t max_replicas_to_use = current_settings.max_parallel_replicas; + if (max_replicas_to_use > shard.getAllNodeCount()) { LOG_INFO( getLogger("ReadFromParallelRemoteReplicasStep"), @@ -420,14 +419,14 @@ void ReadFromParallelRemoteReplicasStep::initializePipeline(QueryPipelineBuilder "Will use the latter number to execute the query.", current_settings.max_parallel_replicas, shard.getAllNodeCount()); - all_replicas_count = shard.getAllNodeCount(); + max_replicas_to_use = shard.getAllNodeCount(); } std::vector shuffled_pool; - if (all_replicas_count < shard.getAllNodeCount()) + if (max_replicas_to_use < shard.getAllNodeCount()) { shuffled_pool = shard.pool->getShuffledPools(current_settings); - shuffled_pool.resize(all_replicas_count); + shuffled_pool.resize(max_replicas_to_use); } else { @@ -437,11 +436,13 @@ void ReadFromParallelRemoteReplicasStep::initializePipeline(QueryPipelineBuilder shuffled_pool = shard.pool->getShuffledPools(current_settings, priority_func); } - for (size_t i=0; i < all_replicas_count; ++i) + coordinator + = std::make_shared(max_replicas_to_use, current_settings.parallel_replicas_mark_segment_size); + + for (size_t i=0; i < max_replicas_to_use; ++i) { IConnections::ReplicaInfo replica_info { - .all_replicas_count = all_replicas_count, /// we should use this number specifically because efficiency of data distribution by consistent hash depends on it. .number_of_current_replica = i, }; diff --git a/src/Processors/QueryPlan/ReadFromRemote.h b/src/Processors/QueryPlan/ReadFromRemote.h index eb15269155a..1adb26b2915 100644 --- a/src/Processors/QueryPlan/ReadFromRemote.h +++ b/src/Processors/QueryPlan/ReadFromRemote.h @@ -70,7 +70,6 @@ public: ASTPtr query_ast_, ClusterPtr cluster_, const StorageID & storage_id_, - ParallelReplicasReadingCoordinatorPtr coordinator_, Block header_, QueryProcessingStage::Enum stage_, ContextMutablePtr context_, diff --git a/src/Processors/QueryPlan/ReadFromStreamLikeEngine.cpp b/src/Processors/QueryPlan/ReadFromStreamLikeEngine.cpp index 4a257bba922..51117ad3c9b 100644 --- a/src/Processors/QueryPlan/ReadFromStreamLikeEngine.cpp +++ b/src/Processors/QueryPlan/ReadFromStreamLikeEngine.cpp @@ -1,5 +1,6 @@ #include +#include #include #include diff --git a/src/Processors/QueryPlan/ReadFromSystemNumbersStep.cpp b/src/Processors/QueryPlan/ReadFromSystemNumbersStep.cpp index 5dbf6fa3318..4136e2d58b4 100644 --- a/src/Processors/QueryPlan/ReadFromSystemNumbersStep.cpp +++ b/src/Processors/QueryPlan/ReadFromSystemNumbersStep.cpp @@ -12,6 +12,7 @@ #include #include #include +#include #include @@ -35,18 +36,33 @@ inline void iotaWithStepOptimized(T * begin, size_t count, T first_value, T step iotaWithStep(begin, count, first_value, step); } +/// The range is defined as [start, end) +UInt64 itemCountInRange(UInt64 start, UInt64 end, UInt64 step) +{ + const auto range_count = end - start; + if (step == 1) + return range_count; + + return (range_count - 1) / step + 1; +} + class NumbersSource : public ISource { public: - NumbersSource(UInt64 block_size_, UInt64 offset_, std::optional limit_, UInt64 chunk_step_, const std::string & column_name, UInt64 step_) + NumbersSource( + UInt64 block_size_, + UInt64 offset_, + std::optional end_, + const std::string & column_name, + UInt64 step_in_chunk_, + UInt64 step_between_chunks_) : ISource(createHeader(column_name)) , block_size(block_size_) , next(offset_) - , chunk_step(chunk_step_) - , step(step_) + , end(end_) + , step_in_chunk(step_in_chunk_) + , step_between_chunks(step_between_chunks_) { - if (limit_.has_value()) - end = limit_.value() + offset_; } String getName() const override { return "Numbers"; } @@ -63,7 +79,10 @@ protected: { if (end.value() <= next) return {}; - real_block_size = std::min(block_size, end.value() - next); + + auto max_items_to_generate = itemCountInRange(next, *end, step_in_chunk); + + real_block_size = std::min(block_size, max_items_to_generate); } auto column = ColumnUInt64::create(real_block_size); ColumnUInt64::Container & vec = column->getData(); @@ -73,21 +92,20 @@ protected: UInt64 * current_end = &vec[real_block_size]; - iotaWithStepOptimized(pos, static_cast(current_end - pos), curr, step); + iotaWithStepOptimized(pos, static_cast(current_end - pos), curr, step_in_chunk); - next += chunk_step; + next += step_between_chunks; progress(column->size(), column->byteSize()); - return {Columns{std::move(column)}, real_block_size}; } private: UInt64 block_size; UInt64 next; - UInt64 chunk_step; std::optional end; /// not included - UInt64 step; + UInt64 step_in_chunk; + UInt64 step_between_chunks; }; struct RangeWithStep @@ -393,7 +411,7 @@ ReadFromSystemNumbersStep::ReadFromSystemNumbersStep( , num_streams{num_streams_} , limit_length_and_offset(InterpreterSelectQuery::getLimitLengthAndOffset(query_info.query->as(), context)) , should_pushdown_limit(shouldPushdownLimit(query_info, limit_length_and_offset.first)) - , query_info_limit(query_info.limit) + , query_info_limit(query_info.trivial_limit) , storage_limits(query_info.storage_limits) { storage_snapshot->check(column_names); @@ -548,41 +566,47 @@ Pipe ReadFromSystemNumbersStep::makePipe() return pipe; } + const auto end = std::invoke( + [&]() -> std::optional + { + if (numbers_storage.limit.has_value()) + return *(numbers_storage.limit) + numbers_storage.offset; + return {}; + }); + /// Fall back to NumbersSource + /// Range in a single block + const auto block_range = max_block_size * numbers_storage.step; + /// Step between chunks in a single source. + /// It is bigger than block_range in case of multiple threads, because we have to account for other sources as well. + const auto step_between_chunks = num_streams * block_range; for (size_t i = 0; i < num_streams; ++i) { + const auto source_offset = i * block_range; + if (numbers_storage.limit.has_value() && *numbers_storage.limit < source_offset) + break; + + const auto source_start = numbers_storage.offset + source_offset; + auto source = std::make_shared( max_block_size, - numbers_storage.offset + i * max_block_size * numbers_storage.step, - numbers_storage.limit, - num_streams * max_block_size * numbers_storage.step, + source_start, + end, numbers_storage.column_name, - numbers_storage.step); + numbers_storage.step, + step_between_chunks); - if (numbers_storage.limit && i == 0) + if (end && i == 0) { - auto rows_appr = (*numbers_storage.limit - 1) / numbers_storage.step + 1; - if (limit > 0 && limit < rows_appr) - rows_appr = query_info_limit; - source->addTotalRowsApprox(rows_appr); + UInt64 rows_approx = itemCountInRange(numbers_storage.offset, *end, numbers_storage.step); + if (limit > 0 && limit < rows_approx) + rows_approx = query_info_limit; + source->addTotalRowsApprox(rows_approx); } pipe.addSource(std::move(source)); } - if (numbers_storage.limit) - { - size_t i = 0; - auto storage_limit = (*numbers_storage.limit - 1) / numbers_storage.step + 1; - /// This formula is how to split 'limit' elements to 'num_streams' chunks almost uniformly. - pipe.addSimpleTransform( - [&](const Block & header) - { - ++i; - return std::make_shared(header, storage_limit * i / num_streams - storage_limit * (i - 1) / num_streams, 0); - }); - } - return pipe; } diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp index 8f40e523b42..48fad9f5fdb 100644 --- a/src/Processors/QueryPlan/SortingStep.cpp +++ b/src/Processors/QueryPlan/SortingStep.cpp @@ -1,5 +1,3 @@ -#include -#include #include #include #include @@ -8,12 +6,16 @@ #include #include #include +#include #include #include +#include #include #include +#include + namespace CurrentMetrics { extern const Metric TemporaryFilesForSort; @@ -38,6 +40,7 @@ SortingStep::Settings::Settings(const Context & context) tmp_data = context.getTempDataOnDisk(); min_free_disk_space = settings.min_free_disk_space_for_temporary_data; max_block_bytes = settings.prefer_external_sort_block_bytes; + read_in_order_use_buffering = settings.read_in_order_use_buffering; } SortingStep::Settings::Settings(size_t max_block_size_) @@ -153,10 +156,11 @@ void SortingStep::updateLimit(size_t limit_) } } -void SortingStep::convertToFinishSorting(SortDescription prefix_description_) +void SortingStep::convertToFinishSorting(SortDescription prefix_description_, bool use_buffering_) { type = Type::FinishSorting; prefix_description = std::move(prefix_description_); + use_buffering = use_buffering_; } void SortingStep::scatterByPartitionIfNeeded(QueryPipelineBuilder& pipeline) @@ -244,6 +248,14 @@ void SortingStep::mergingSorted(QueryPipelineBuilder & pipeline, const SortDescr /// If there are several streams, then we merge them into one if (pipeline.getNumStreams() > 1) { + if (use_buffering && sort_settings.read_in_order_use_buffering) + { + pipeline.addSimpleTransform([&](const Block & header) + { + return std::make_shared(header, sort_settings.max_block_size, sort_settings.max_block_bytes, limit_); + }); + } + auto transform = std::make_shared( pipeline.getHeader(), pipeline.getNumStreams(), @@ -373,9 +385,8 @@ void SortingStep::transformPipeline(QueryPipelineBuilder & pipeline, const Build mergingSorted(pipeline, prefix_description, (need_finish_sorting ? 0 : limit)); if (need_finish_sorting) - { finishSorting(pipeline, prefix_description, result_description, limit); - } + return; } diff --git a/src/Processors/QueryPlan/SortingStep.h b/src/Processors/QueryPlan/SortingStep.h index 49dcf9f3121..b4a49394a13 100644 --- a/src/Processors/QueryPlan/SortingStep.h +++ b/src/Processors/QueryPlan/SortingStep.h @@ -28,6 +28,7 @@ public: TemporaryDataOnDiskScopePtr tmp_data = nullptr; size_t min_free_disk_space = 0; size_t max_block_bytes = 0; + size_t read_in_order_use_buffering = 0; explicit Settings(const Context & context); explicit Settings(size_t max_block_size_); @@ -80,7 +81,7 @@ public: const SortDescription & getSortDescription() const { return result_description; } - void convertToFinishSorting(SortDescription prefix_description); + void convertToFinishSorting(SortDescription prefix_description, bool use_buffering_); Type getType() const { return type; } const Settings & getSettings() const { return sort_settings; } @@ -126,6 +127,7 @@ private: UInt64 limit; bool always_read_till_end = false; + bool use_buffering = false; Settings sort_settings; diff --git a/src/Processors/QueryPlan/TotalsHavingStep.cpp b/src/Processors/QueryPlan/TotalsHavingStep.cpp index ac5e144bf4a..10384f6296a 100644 --- a/src/Processors/QueryPlan/TotalsHavingStep.cpp +++ b/src/Processors/QueryPlan/TotalsHavingStep.cpp @@ -5,6 +5,7 @@ #include #include #include +#include namespace DB { diff --git a/src/Processors/Sinks/RemoteSink.h b/src/Processors/Sinks/RemoteSink.h index 30cf958c072..c05cc1defcb 100644 --- a/src/Processors/Sinks/RemoteSink.h +++ b/src/Processors/Sinks/RemoteSink.h @@ -20,7 +20,7 @@ public: } String getName() const override { return "RemoteSink"; } - void consume (Chunk chunk) override { write(RemoteInserter::getHeader().cloneWithColumns(chunk.detachColumns())); } + void consume (Chunk & chunk) override { write(RemoteInserter::getHeader().cloneWithColumns(chunk.getColumns())); } void onFinish() override { RemoteInserter::onFinish(); } }; diff --git a/src/Processors/Sinks/SinkToStorage.cpp b/src/Processors/Sinks/SinkToStorage.cpp index 5f9f9f9b1a1..36bb70f493f 100644 --- a/src/Processors/Sinks/SinkToStorage.cpp +++ b/src/Processors/Sinks/SinkToStorage.cpp @@ -15,9 +15,8 @@ void SinkToStorage::onConsume(Chunk chunk) */ Nested::validateArraySizes(getHeader().cloneWithColumns(chunk.getColumns())); - consume(chunk.clone()); - if (!lastBlockIsDuplicate()) - cur_chunk = std::move(chunk); + consume(chunk); + cur_chunk = std::move(chunk); } SinkToStorage::GenerateResult SinkToStorage::onGenerate() diff --git a/src/Processors/Sinks/SinkToStorage.h b/src/Processors/Sinks/SinkToStorage.h index 023bbd8b094..c728fa87b1e 100644 --- a/src/Processors/Sinks/SinkToStorage.h +++ b/src/Processors/Sinks/SinkToStorage.h @@ -18,8 +18,7 @@ public: void addTableLock(const TableLockHolder & lock) { table_locks.push_back(lock); } protected: - virtual void consume(Chunk chunk) = 0; - virtual bool lastBlockIsDuplicate() const { return false; } + virtual void consume(Chunk & chunk) = 0; private: std::vector table_locks; @@ -38,7 +37,7 @@ class NullSinkToStorage : public SinkToStorage public: using SinkToStorage::SinkToStorage; std::string getName() const override { return "NullSinkToStorage"; } - void consume(Chunk) override {} + void consume(Chunk &) override {} }; using SinkPtr = std::shared_ptr; diff --git a/src/Processors/Sources/BlocksSource.h b/src/Processors/Sources/BlocksSource.h index ec0dc9609f1..7ac460c14e2 100644 --- a/src/Processors/Sources/BlocksSource.h +++ b/src/Processors/Sources/BlocksSource.h @@ -43,7 +43,10 @@ protected: info->bucket_num = res.info.bucket_num; info->is_overflows = res.info.is_overflows; - return Chunk(res.getColumns(), res.rows(), std::move(info)); + auto chunk = Chunk(res.getColumns(), res.rows()); + chunk.getChunkInfos().add(std::move(info)); + + return chunk; } private: diff --git a/src/Processors/Sources/MySQLSource.cpp b/src/Processors/Sources/MySQLSource.cpp index 985a82a7b17..5d533a7747e 100644 --- a/src/Processors/Sources/MySQLSource.cpp +++ b/src/Processors/Sources/MySQLSource.cpp @@ -2,7 +2,9 @@ #if USE_MYSQL #include + #include +#include #include #include #include diff --git a/src/Processors/Sources/MySQLSource.h b/src/Processors/Sources/MySQLSource.h index 4ae5af22dab..b99e48a7759 100644 --- a/src/Processors/Sources/MySQLSource.h +++ b/src/Processors/Sources/MySQLSource.h @@ -6,11 +6,12 @@ #include #include #include -#include namespace DB { +struct Settings; + struct StreamSettings { /// Check if setting is enabled, otherwise use common `max_block_size` setting. diff --git a/src/Processors/Sources/PostgreSQLSource.cpp b/src/Processors/Sources/PostgreSQLSource.cpp index f18c63ed385..a3d6fd691d8 100644 --- a/src/Processors/Sources/PostgreSQLSource.cpp +++ b/src/Processors/Sources/PostgreSQLSource.cpp @@ -191,6 +191,20 @@ PostgreSQLSource::~PostgreSQLSource() { try { + if (stream) + { + /** Internally libpqxx::stream_from runs PostgreSQL copy query `COPY query TO STDOUT`. + * During transaction abort we try to execute PostgreSQL `ROLLBACK` command and if + * copy query is not cancelled, we wait until it finishes. + */ + tx->conn().cancel_query(); + + /** If stream is not closed, libpqxx::stream_from closes stream in destructor, but that way + * exception is added into transaction pending error and we can potentially ignore exception message. + */ + stream->close(); + } + stream.reset(); tx.reset(); } diff --git a/src/Processors/Sources/RecursiveCTESource.cpp b/src/Processors/Sources/RecursiveCTESource.cpp index 221198c622a..40e72e9cbb1 100644 --- a/src/Processors/Sources/RecursiveCTESource.cpp +++ b/src/Processors/Sources/RecursiveCTESource.cpp @@ -17,6 +17,8 @@ #include #include +#include + namespace DB { diff --git a/src/Processors/Sources/RemoteSource.cpp b/src/Processors/Sources/RemoteSource.cpp index 3d7dd3f76b8..1578bd389c9 100644 --- a/src/Processors/Sources/RemoteSource.cpp +++ b/src/Processors/Sources/RemoteSource.cpp @@ -176,7 +176,7 @@ std::optional RemoteSource::tryGenerate() auto info = std::make_shared(); info->bucket_num = block.info.bucket_num; info->is_overflows = block.info.is_overflows; - chunk.setChunkInfo(std::move(info)); + chunk.getChunkInfos().add(std::move(info)); } return chunk; diff --git a/src/Processors/Sources/SourceFromSingleChunk.cpp b/src/Processors/Sources/SourceFromSingleChunk.cpp index 00f40a34361..9abe0504d10 100644 --- a/src/Processors/Sources/SourceFromSingleChunk.cpp +++ b/src/Processors/Sources/SourceFromSingleChunk.cpp @@ -5,7 +5,9 @@ namespace DB { -SourceFromSingleChunk::SourceFromSingleChunk(Block header, Chunk chunk_) : ISource(std::move(header)), chunk(std::move(chunk_)) {} +SourceFromSingleChunk::SourceFromSingleChunk(Block header, Chunk chunk_) : ISource(std::move(header)), chunk(std::move(chunk_)) +{ +} SourceFromSingleChunk::SourceFromSingleChunk(Block data) : ISource(data.cloneEmpty()), chunk(data.getColumns(), data.rows()) { @@ -20,7 +22,7 @@ SourceFromSingleChunk::SourceFromSingleChunk(Block data) : ISource(data.cloneEmp auto info = std::make_shared(); info->bucket_num = data.info.bucket_num; info->is_overflows = data.info.is_overflows; - chunk.setChunkInfo(std::move(info)); + chunk.getChunkInfos().add(std::move(info)); } } diff --git a/src/Processors/TTL/TTLAggregationAlgorithm.cpp b/src/Processors/TTL/TTLAggregationAlgorithm.cpp index 45e8a96412e..2d7a37d0abe 100644 --- a/src/Processors/TTL/TTLAggregationAlgorithm.cpp +++ b/src/Processors/TTL/TTLAggregationAlgorithm.cpp @@ -1,5 +1,6 @@ -#include +#include #include +#include namespace DB { diff --git a/src/Processors/Transforms/AggregatingInOrderTransform.cpp b/src/Processors/Transforms/AggregatingInOrderTransform.cpp index 9ffe15d0f85..45b0960ec8f 100644 --- a/src/Processors/Transforms/AggregatingInOrderTransform.cpp +++ b/src/Processors/Transforms/AggregatingInOrderTransform.cpp @@ -332,7 +332,7 @@ void AggregatingInOrderTransform::generate() variants.aggregates_pool = variants.aggregates_pools.at(0).get(); /// Pass info about used memory by aggregate functions further. - to_push_chunk.setChunkInfo(std::make_shared(cur_block_bytes)); + to_push_chunk.getChunkInfos().add(std::make_shared(cur_block_bytes)); cur_block_bytes = 0; cur_block_size = 0; @@ -351,11 +351,12 @@ FinalizeAggregatedTransform::FinalizeAggregatedTransform(Block header, Aggregati void FinalizeAggregatedTransform::transform(Chunk & chunk) { if (params->final) - finalizeChunk(chunk, aggregates_mask); - else if (!chunk.getChunkInfo()) { - auto info = std::make_shared(); - chunk.setChunkInfo(std::move(info)); + finalizeChunk(chunk, aggregates_mask); + } + else if (!chunk.getChunkInfos().get()) + { + chunk.getChunkInfos().add(std::make_shared()); } } diff --git a/src/Processors/Transforms/AggregatingInOrderTransform.h b/src/Processors/Transforms/AggregatingInOrderTransform.h index 5d50e97f552..41a0d7fc7f1 100644 --- a/src/Processors/Transforms/AggregatingInOrderTransform.h +++ b/src/Processors/Transforms/AggregatingInOrderTransform.h @@ -5,6 +5,7 @@ #include #include #include +#include namespace DB { @@ -12,10 +13,12 @@ namespace DB struct InputOrderInfo; using InputOrderInfoPtr = std::shared_ptr; -struct ChunkInfoWithAllocatedBytes : public ChunkInfo +struct ChunkInfoWithAllocatedBytes : public ChunkInfoCloneable { + ChunkInfoWithAllocatedBytes(const ChunkInfoWithAllocatedBytes & other) = default; explicit ChunkInfoWithAllocatedBytes(Int64 allocated_bytes_) : allocated_bytes(allocated_bytes_) {} + Int64 allocated_bytes; }; diff --git a/src/Processors/Transforms/AggregatingTransform.cpp b/src/Processors/Transforms/AggregatingTransform.cpp index 65f0612d738..fda03174c53 100644 --- a/src/Processors/Transforms/AggregatingTransform.cpp +++ b/src/Processors/Transforms/AggregatingTransform.cpp @@ -35,7 +35,7 @@ Chunk convertToChunk(const Block & block) UInt64 num_rows = block.rows(); Chunk chunk(block.getColumns(), num_rows); - chunk.setChunkInfo(std::move(info)); + chunk.getChunkInfos().add(std::move(info)); return chunk; } @@ -44,15 +44,11 @@ namespace { const AggregatedChunkInfo * getInfoFromChunk(const Chunk & chunk) { - const auto & info = chunk.getChunkInfo(); - if (!info) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk info was not set for chunk."); - - const auto * agg_info = typeid_cast(info.get()); + auto agg_info = chunk.getChunkInfos().get(); if (!agg_info) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk should have AggregatedChunkInfo."); - return agg_info; + return agg_info.get(); } /// Reads chunks from file in native format. Provide chunks with aggregation info. @@ -210,11 +206,7 @@ private: void process(Chunk && chunk) { - if (!chunk.hasChunkInfo()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected chunk with chunk info in {}", getName()); - - const auto & info = chunk.getChunkInfo(); - const auto * chunks_to_merge = typeid_cast(info.get()); + auto chunks_to_merge = chunk.getChunkInfos().get(); if (!chunks_to_merge) throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected chunk with ChunksToMerge info in {}", getName()); @@ -783,7 +775,7 @@ void AggregatingTransform::initGenerate() { /// Just a reasonable constant, matches default value for the setting `preferred_block_size_bytes` static constexpr size_t oneMB = 1024 * 1024; - return std::make_shared(header, params->params.max_block_size, oneMB); + return std::make_shared(header, params->params.max_block_size, oneMB); }); } /// AggregatingTransform::expandPipeline expects single output port. diff --git a/src/Processors/Transforms/AggregatingTransform.h b/src/Processors/Transforms/AggregatingTransform.h index e167acde067..95983c39d1e 100644 --- a/src/Processors/Transforms/AggregatingTransform.h +++ b/src/Processors/Transforms/AggregatingTransform.h @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include @@ -19,7 +20,7 @@ namespace CurrentMetrics namespace DB { -class AggregatedChunkInfo : public ChunkInfo +class AggregatedChunkInfo : public ChunkInfoCloneable { public: bool is_overflows = false; diff --git a/src/Processors/Transforms/ApplySquashingTransform.h b/src/Processors/Transforms/ApplySquashingTransform.h index 965a084bb13..49a6581e685 100644 --- a/src/Processors/Transforms/ApplySquashingTransform.h +++ b/src/Processors/Transforms/ApplySquashingTransform.h @@ -27,18 +27,12 @@ public: } ExceptionKeepingTransform::work(); - if (finish_chunk) - { - data.chunk = std::move(finish_chunk); - ready_output = true; - } } protected: void onConsume(Chunk chunk) override { - if (auto res_chunk = DB::Squashing::squash(std::move(chunk))) - cur_chunk.setColumns(res_chunk.getColumns(), res_chunk.getNumRows()); + cur_chunk = Squashing::squash(std::move(chunk)); } GenerateResult onGenerate() override @@ -48,16 +42,10 @@ protected: res.is_done = true; return res; } - void onFinish() override - { - auto chunk = DB::Squashing::squash({}); - finish_chunk.setColumns(chunk.getColumns(), chunk.getNumRows()); - } private: Squashing squashing; Chunk cur_chunk; - Chunk finish_chunk; }; } diff --git a/src/Processors/Transforms/CheckConstraintsTransform.cpp b/src/Processors/Transforms/CheckConstraintsTransform.cpp index e43aa6028da..cdae8c23a3e 100644 --- a/src/Processors/Transforms/CheckConstraintsTransform.cpp +++ b/src/Processors/Transforms/CheckConstraintsTransform.cpp @@ -10,6 +10,7 @@ #include #include #include +#include namespace DB @@ -31,6 +32,7 @@ CheckConstraintsTransform::CheckConstraintsTransform( , table_id(table_id_) , constraints_to_check(constraints_.filterConstraints(ConstraintsDescription::ConstraintType::CHECK)) , expressions(constraints_.getExpressions(context_, header.getNamesAndTypesList())) + , context(std::move(context_)) { } @@ -39,6 +41,10 @@ void CheckConstraintsTransform::onConsume(Chunk chunk) { if (chunk.getNumRows() > 0) { + if (rows_written == 0) + for (const auto & expression : expressions) + VirtualColumnUtils::buildSetsForDAG(expression->getActionsDAG(), context); + Block block_to_calculate = getInputPort().getHeader().cloneWithColumns(chunk.getColumns()); for (size_t i = 0; i < expressions.size(); ++i) { diff --git a/src/Processors/Transforms/CheckConstraintsTransform.h b/src/Processors/Transforms/CheckConstraintsTransform.h index 09833ff396b..f92d0ab855e 100644 --- a/src/Processors/Transforms/CheckConstraintsTransform.h +++ b/src/Processors/Transforms/CheckConstraintsTransform.h @@ -35,6 +35,7 @@ private: StorageID table_id; const ASTs constraints_to_check; const ConstraintsExpressions expressions; + ContextPtr context; size_t rows_written = 0; Chunk cur_chunk; }; diff --git a/src/Processors/Transforms/CountingTransform.cpp b/src/Processors/Transforms/CountingTransform.cpp index 3dfb9fe178f..2c6b3bd8638 100644 --- a/src/Processors/Transforms/CountingTransform.cpp +++ b/src/Processors/Transforms/CountingTransform.cpp @@ -1,6 +1,7 @@ - -#include #include + +#include +#include #include #include diff --git a/src/Processors/Transforms/DeduplicationTokenTransforms.cpp b/src/Processors/Transforms/DeduplicationTokenTransforms.cpp new file mode 100644 index 00000000000..841090f029e --- /dev/null +++ b/src/Processors/Transforms/DeduplicationTokenTransforms.cpp @@ -0,0 +1,238 @@ +#include + +#include + +#include +#include +#include + + +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +void RestoreChunkInfosTransform::transform(Chunk & chunk) +{ + chunk.getChunkInfos().appendIfUniq(chunk_infos.clone()); +} + +namespace DeduplicationToken +{ + +String TokenInfo::getToken() const +{ + if (!isDefined()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "token is not defined, stage {}, token {}", stage, debugToken()); + + return getTokenImpl(); +} + +String TokenInfo::getTokenImpl() const +{ + String result; + result.reserve(getTotalSize()); + + for (const auto & part : parts) + { + if (!result.empty()) + result.append(":"); + result.append(part); + } + + return result; +} + +String TokenInfo::debugToken() const +{ + return getTokenImpl(); +} + +void TokenInfo::addChunkHash(String part) +{ + if (stage == UNDEFINED && empty()) + stage = DEFINE_SOURCE_WITH_HASHES; + + if (stage != DEFINE_SOURCE_WITH_HASHES) + throw Exception(ErrorCodes::LOGICAL_ERROR, "token is in wrong stage {}, token {}", stage, debugToken()); + + addTokenPart(std::move(part)); +} + +void TokenInfo::finishChunkHashes() +{ + if (stage == UNDEFINED && empty()) + stage = DEFINE_SOURCE_WITH_HASHES; + + if (stage != DEFINE_SOURCE_WITH_HASHES) + throw Exception(ErrorCodes::LOGICAL_ERROR, "token is in wrong stage {}, token {}", stage, debugToken()); + + stage = DEFINED; +} + +void TokenInfo::setUserToken(const String & token) +{ + if (stage == UNDEFINED && empty()) + stage = DEFINE_SOURCE_USER_TOKEN; + + if (stage != DEFINE_SOURCE_USER_TOKEN) + throw Exception(ErrorCodes::LOGICAL_ERROR, "token is in wrong stage {}, token {}", stage, debugToken()); + + addTokenPart(fmt::format("user-token-{}", token)); +} + +void TokenInfo::setSourceWithUserToken(size_t block_number) +{ + if (stage != DEFINE_SOURCE_USER_TOKEN) + throw Exception(ErrorCodes::LOGICAL_ERROR, "token is in wrong stage {}, token {}", stage, debugToken()); + + addTokenPart(fmt::format("source-number-{}", block_number)); + + stage = DEFINED; +} + +void TokenInfo::setViewID(const String & id) +{ + if (stage == DEFINED) + stage = DEFINE_VIEW; + + if (stage != DEFINE_VIEW) + throw Exception(ErrorCodes::LOGICAL_ERROR, "token is in wrong stage {}, token {}", stage, debugToken()); + + addTokenPart(fmt::format("view-id-{}", id)); +} + +void TokenInfo::setViewBlockNumber(size_t block_number) +{ + if (stage != DEFINE_VIEW) + throw Exception(ErrorCodes::LOGICAL_ERROR, "token is in wrong stage {}, token {}", stage, debugToken()); + + addTokenPart(fmt::format("view-block-{}", block_number)); + + stage = DEFINED; +} + +void TokenInfo::reset() +{ + stage = UNDEFINED; + parts.clear(); +} + +void TokenInfo::addTokenPart(String part) +{ + parts.push_back(std::move(part)); +} + +size_t TokenInfo::getTotalSize() const +{ + if (parts.empty()) + return 0; + + size_t size = 0; + for (const auto & part : parts) + size += part.size(); + + // we reserve more size here to be able to add delimenter between parts. + return size + parts.size() - 1; +} + +#ifdef ABORT_ON_LOGICAL_ERROR +void CheckTokenTransform::transform(Chunk & chunk) +{ + auto token_info = chunk.getChunkInfos().get(); + + if (!token_info) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk has to have DedupTokenInfo as ChunkInfo, {}", debug); + } + + LOG_TEST(log, "debug: {}, token: {}, columns {} rows {}", debug, token_info->debugToken(), chunk.getNumColumns(), chunk.getNumRows()); +} +#endif + +String DefineSourceWithChunkHashTransform::getChunkHash(const Chunk & chunk) +{ + SipHash hash; + for (const auto & colunm : chunk.getColumns()) + colunm->updateHashFast(hash); + + const auto hash_value = hash.get128(); + return toString(hash_value.items[0]) + "_" + toString(hash_value.items[1]); +} + + +void DefineSourceWithChunkHashTransform::transform(Chunk & chunk) +{ + auto token_info = chunk.getChunkInfos().get(); + + if (!token_info) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "TokenInfo is expected for consumed chunk in DefineSourceWithChunkHashesTransform"); + + if (token_info->isDefined()) + return; + + token_info->addChunkHash(getChunkHash(chunk)); + token_info->finishChunkHashes(); +} + +void SetUserTokenTransform::transform(Chunk & chunk) +{ + auto token_info = chunk.getChunkInfos().get(); + if (!token_info) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "TokenInfo is expected for consumed chunk in SetUserTokenTransform"); + token_info->setUserToken(user_token); +} + +void SetSourceBlockNumberTransform::transform(Chunk & chunk) +{ + auto token_info = chunk.getChunkInfos().get(); + if (!token_info) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "TokenInfo is expected for consumed chunk in SetSourceBlockNumberTransform"); + token_info->setSourceWithUserToken(block_number++); +} + +void SetViewIDTransform::transform(Chunk & chunk) +{ + auto token_info = chunk.getChunkInfos().get(); + if (!token_info) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "TokenInfo is expected for consumed chunk in SetViewIDTransform"); + token_info->setViewID(view_id); +} + +void SetViewBlockNumberTransform::transform(Chunk & chunk) +{ + auto token_info = chunk.getChunkInfos().get(); + if (!token_info) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "TokenInfo is expected for consumed chunk in SetViewBlockNumberTransform"); + token_info->setViewBlockNumber(block_number++); +} + +void ResetTokenTransform::transform(Chunk & chunk) +{ + auto token_info = chunk.getChunkInfos().get(); + if (!token_info) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "TokenInfo is expected for consumed chunk in ResetTokenTransform"); + + token_info->reset(); +} + +} +} diff --git a/src/Processors/Transforms/DeduplicationTokenTransforms.h b/src/Processors/Transforms/DeduplicationTokenTransforms.h new file mode 100644 index 00000000000..d6aff9e1370 --- /dev/null +++ b/src/Processors/Transforms/DeduplicationTokenTransforms.h @@ -0,0 +1,237 @@ +#pragma once + +#include +#include + +#include +#include "Common/Logger.h" + + +namespace DB +{ + class RestoreChunkInfosTransform : public ISimpleTransform + { + public: + RestoreChunkInfosTransform(Chunk::ChunkInfoCollection chunk_infos_, const Block & header_) + : ISimpleTransform(header_, header_, true) + , chunk_infos(std::move(chunk_infos_)) + {} + + String getName() const override { return "RestoreChunkInfosTransform"; } + + void transform(Chunk & chunk) override; + + private: + Chunk::ChunkInfoCollection chunk_infos; + }; + + +namespace DeduplicationToken +{ + class TokenInfo : public ChunkInfoCloneable + { + public: + TokenInfo() = default; + TokenInfo(const TokenInfo & other) = default; + + String getToken() const; + String debugToken() const; + + bool empty() const { return parts.empty(); } + + bool isDefined() const { return stage == DEFINED; } + + void addChunkHash(String part); + void finishChunkHashes(); + + void setUserToken(const String & token); + void setSourceWithUserToken(size_t block_number); + + void setViewID(const String & id); + void setViewBlockNumber(size_t block_number); + + void reset(); + + private: + String getTokenImpl() const; + + void addTokenPart(String part); + size_t getTotalSize() const; + + /* Token has to be prepared in a particular order. + * BuildingStage ensures that token is expanded according the following order. + * Firstly token is expanded with information about the source. + * It could be done with two ways: add several hash sums from the source chunks or provide user defined deduplication token and its sequentional block number. + * + * transition // method + * UNDEFINED -> DEFINE_SOURCE_WITH_HASHES // addChunkHash + * DEFINE_SOURCE_WITH_HASHES -> DEFINE_SOURCE_WITH_HASHES // addChunkHash + * DEFINE_SOURCE_WITH_HASHES -> DEFINED // defineSourceWithChankHashes + * + * transition // method + * UNDEFINED -> DEFINE_SOURCE_USER_TOKEN // setUserToken + * DEFINE_SOURCE_USER_TOKEN -> DEFINED // defineSourceWithUserToken + * + * After token is defined, it could be extended with view id and view block number. Actually it has to be expanded with view details if there is one or several views. + * + * transition // method + * DEFINED -> DEFINE_VIEW // setViewID + * DEFINE_VIEW -> DEFINED // defineViewID + */ + + enum BuildingStage + { + UNDEFINED, + DEFINE_SOURCE_WITH_HASHES, + DEFINE_SOURCE_USER_TOKEN, + DEFINE_VIEW, + DEFINED, + }; + + BuildingStage stage = UNDEFINED; + std::vector parts; + }; + + +#ifdef ABORT_ON_LOGICAL_ERROR + /// use that class only with debug builds in CI for introspection + class CheckTokenTransform : public ISimpleTransform + { + public: + CheckTokenTransform(String debug_, const Block & header_) + : ISimpleTransform(header_, header_, true) + , debug(std::move(debug_)) + { + } + + String getName() const override { return "DeduplicationToken::CheckTokenTransform"; } + + void transform(Chunk & chunk) override; + + private: + String debug; + LoggerPtr log = getLogger("CheckInsertDeduplicationTokenTransform"); + }; +#endif + + + class AddTokenInfoTransform : public ISimpleTransform + { + public: + explicit AddTokenInfoTransform(const Block & header_) + : ISimpleTransform(header_, header_, true) + { + } + + String getName() const override { return "DeduplicationToken::AddTokenInfoTransform"; } + + void transform(Chunk & chunk) override + { + chunk.getChunkInfos().add(std::make_shared()); + } + }; + + + class DefineSourceWithChunkHashTransform : public ISimpleTransform + { + public: + explicit DefineSourceWithChunkHashTransform(const Block & header_) + : ISimpleTransform(header_, header_, true) + { + } + + String getName() const override { return "DeduplicationToken::DefineSourceWithChunkHashesTransform"; } + + // Usually MergeTreeSink/ReplicatedMergeTreeSink calls addChunkHash for the deduplication token with hashes from the parts. + // But if there is some table with different engine, we still need to define the source of the data in deduplication token + // We use that transform to define the source as a hash of entire block in deduplication token + void transform(Chunk & chunk) override; + + static String getChunkHash(const Chunk & chunk); + }; + + class ResetTokenTransform : public ISimpleTransform + { + public: + explicit ResetTokenTransform(const Block & header_) + : ISimpleTransform(header_, header_, true) + { + } + + String getName() const override { return "DeduplicationToken::ResetTokenTransform"; } + + void transform(Chunk & chunk) override; + }; + + + class SetUserTokenTransform : public ISimpleTransform + { + public: + SetUserTokenTransform(String user_token_, const Block & header_) + : ISimpleTransform(header_, header_, true) + , user_token(std::move(user_token_)) + { + } + + String getName() const override { return "DeduplicationToken::SetUserTokenTransform"; } + + void transform(Chunk & chunk) override; + + private: + String user_token; + }; + + + class SetSourceBlockNumberTransform : public ISimpleTransform + { + public: + explicit SetSourceBlockNumberTransform(const Block & header_) + : ISimpleTransform(header_, header_, true) + { + } + + String getName() const override { return "DeduplicationToken::SetSourceBlockNumberTransform"; } + + void transform(Chunk & chunk) override; + + private: + size_t block_number = 0; + }; + + + class SetViewIDTransform : public ISimpleTransform + { + public: + SetViewIDTransform(String view_id_, const Block & header_) + : ISimpleTransform(header_, header_, true) + , view_id(std::move(view_id_)) + { + } + + String getName() const override { return "DeduplicationToken::SetViewIDTransform"; } + + void transform(Chunk & chunk) override; + + private: + String view_id; + }; + + + class SetViewBlockNumberTransform : public ISimpleTransform + { + public: + explicit SetViewBlockNumberTransform(const Block & header_) + : ISimpleTransform(header_, header_, true) + { + } + + String getName() const override { return "DeduplicationToken::SetViewBlockNumberTransform"; } + + void transform(Chunk & chunk) override; + + private: + size_t block_number = 0; + }; + +} +} diff --git a/src/Processors/Transforms/DistinctSortedChunkTransform.h b/src/Processors/Transforms/DistinctSortedChunkTransform.h index 188e3d5c4c7..cdb20410a5b 100644 --- a/src/Processors/Transforms/DistinctSortedChunkTransform.h +++ b/src/Processors/Transforms/DistinctSortedChunkTransform.h @@ -5,6 +5,7 @@ #include #include #include +#include namespace DB { diff --git a/src/Processors/Transforms/DistinctSortedTransform.h b/src/Processors/Transforms/DistinctSortedTransform.h index 86c5c968e4b..52819f1afd7 100644 --- a/src/Processors/Transforms/DistinctSortedTransform.h +++ b/src/Processors/Transforms/DistinctSortedTransform.h @@ -1,6 +1,7 @@ #pragma once #include +#include #include #include #include diff --git a/src/Processors/Transforms/ExpressionTransform.cpp b/src/Processors/Transforms/ExpressionTransform.cpp index 2fbd2c21b8d..04fabc9a3c6 100644 --- a/src/Processors/Transforms/ExpressionTransform.cpp +++ b/src/Processors/Transforms/ExpressionTransform.cpp @@ -1,5 +1,7 @@ #include #include + + namespace DB { diff --git a/src/Processors/Transforms/JoiningTransform.cpp b/src/Processors/Transforms/JoiningTransform.cpp index 3e2a9462e54..ca204bcb482 100644 --- a/src/Processors/Transforms/JoiningTransform.cpp +++ b/src/Processors/Transforms/JoiningTransform.cpp @@ -365,10 +365,9 @@ IProcessor::Status DelayedJoinedBlocksWorkerTransform::prepare() return Status::Finished; } - if (!data.chunk.hasChunkInfo()) + task = data.chunk.getChunkInfos().get(); + if (!task) throw Exception(ErrorCodes::LOGICAL_ERROR, "DelayedJoinedBlocksWorkerTransform must have chunk info"); - - task = std::dynamic_pointer_cast(data.chunk.getChunkInfo()); } else { @@ -479,7 +478,7 @@ IProcessor::Status DelayedJoinedBlocksTransform::prepare() if (output.isFinished()) continue; Chunk chunk; - chunk.setChunkInfo(std::make_shared()); + chunk.getChunkInfos().add(std::make_shared()); output.push(std::move(chunk)); output.finish(); } @@ -496,7 +495,7 @@ IProcessor::Status DelayedJoinedBlocksTransform::prepare() { Chunk chunk; auto task = std::make_shared(delayed_blocks, left_delayed_stream_finished_counter); - chunk.setChunkInfo(task); + chunk.getChunkInfos().add(std::move(task)); output.push(std::move(chunk)); } delayed_blocks = nullptr; diff --git a/src/Processors/Transforms/JoiningTransform.h b/src/Processors/Transforms/JoiningTransform.h index a308af03662..5f6d9d6fff2 100644 --- a/src/Processors/Transforms/JoiningTransform.h +++ b/src/Processors/Transforms/JoiningTransform.h @@ -1,6 +1,7 @@ #pragma once #include - +#include +#include namespace DB { @@ -111,11 +112,12 @@ private: }; -class DelayedBlocksTask : public ChunkInfo +class DelayedBlocksTask : public ChunkInfoCloneable { public: DelayedBlocksTask() = default; + DelayedBlocksTask(const DelayedBlocksTask & other) = default; explicit DelayedBlocksTask(IBlocksStreamPtr delayed_blocks_, JoiningTransform::FinishCounterPtr left_delayed_stream_finish_counter_) : delayed_blocks(std::move(delayed_blocks_)) , left_delayed_stream_finish_counter(left_delayed_stream_finish_counter_) diff --git a/src/Processors/Transforms/MaterializingTransform.cpp b/src/Processors/Transforms/MaterializingTransform.cpp index 1eaa5458d37..9ae80e21a68 100644 --- a/src/Processors/Transforms/MaterializingTransform.cpp +++ b/src/Processors/Transforms/MaterializingTransform.cpp @@ -1,6 +1,7 @@ #include #include + namespace DB { diff --git a/src/Processors/Transforms/MemoryBoundMerging.h b/src/Processors/Transforms/MemoryBoundMerging.h index 607087fb39c..d7bc320173b 100644 --- a/src/Processors/Transforms/MemoryBoundMerging.h +++ b/src/Processors/Transforms/MemoryBoundMerging.h @@ -150,11 +150,7 @@ private: if (!chunk.hasRows()) return; - const auto & info = chunk.getChunkInfo(); - if (!info) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk info was not set for chunk in SortingAggregatedForMemoryBoundMergingTransform."); - - const auto * agg_info = typeid_cast(info.get()); + const auto & agg_info = chunk.getChunkInfos().get(); if (!agg_info) throw Exception( ErrorCodes::LOGICAL_ERROR, "Chunk should have AggregatedChunkInfo in SortingAggregatedForMemoryBoundMergingTransform."); diff --git a/src/Processors/Transforms/MergeJoinTransform.cpp b/src/Processors/Transforms/MergeJoinTransform.cpp index 159a3244fe9..e96a75d277b 100644 --- a/src/Processors/Transforms/MergeJoinTransform.cpp +++ b/src/Processors/Transforms/MergeJoinTransform.cpp @@ -9,7 +9,6 @@ #include #include -#include #include #include #include @@ -19,6 +18,7 @@ #include #include #include +#include #include @@ -34,13 +34,20 @@ namespace ErrorCodes namespace { -FullMergeJoinCursorPtr createCursor(const Block & block, const Names & columns) +FullMergeJoinCursorPtr createCursor(const Block & block, const Names & columns, JoinStrictness strictness) { SortDescription desc; desc.reserve(columns.size()); for (const auto & name : columns) desc.emplace_back(name); - return std::make_unique(materializeBlock(block), desc); + return std::make_unique(block, desc, strictness == JoinStrictness::Asof); +} + +bool ALWAYS_INLINE isNullAt(const IColumn & column, size_t row) +{ + if (const auto * nullable_column = checkAndGetColumn(&column)) + return nullable_column->isNullAt(row); + return false; } template @@ -54,7 +61,7 @@ int nullableCompareAt(const IColumn & left_column, const IColumn & right_column, if (left_nullable && right_nullable) { int res = left_nullable->compareAt(lhs_pos, rhs_pos, right_column, null_direction_hint); - if (res) + if (res != 0) return res; /// NULL != NULL case @@ -90,9 +97,10 @@ int nullableCompareAt(const IColumn & left_column, const IColumn & right_column, int ALWAYS_INLINE compareCursors(const SortCursorImpl & lhs, size_t lpos, const SortCursorImpl & rhs, size_t rpos, + size_t key_length, int null_direction_hint) { - for (size_t i = 0; i < lhs.sort_columns_size; ++i) + for (size_t i = 0; i < key_length; ++i) { /// TODO(@vdimir): use nullableCompareAt only if there's nullable columns int cmp = nullableCompareAt(*lhs.sort_columns[i], *rhs.sort_columns[i], lpos, rpos, null_direction_hint); @@ -104,13 +112,18 @@ int ALWAYS_INLINE compareCursors(const SortCursorImpl & lhs, size_t lpos, int ALWAYS_INLINE compareCursors(const SortCursorImpl & lhs, const SortCursorImpl & rhs, int null_direction_hint) { - return compareCursors(lhs, lhs.getRow(), rhs, rhs.getRow(), null_direction_hint); + return compareCursors(lhs, lhs.getRow(), rhs, rhs.getRow(), lhs.sort_columns_size, null_direction_hint); +} + +int compareAsofCursors(const FullMergeJoinCursor & lhs, const FullMergeJoinCursor & rhs, int null_direction_hint) +{ + return nullableCompareAt(*lhs.getAsofColumn(), *rhs.getAsofColumn(), lhs->getRow(), rhs->getRow(), null_direction_hint); } bool ALWAYS_INLINE totallyLess(SortCursorImpl & lhs, SortCursorImpl & rhs, int null_direction_hint) { /// The last row of left cursor is less than the current row of the right cursor. - int cmp = compareCursors(lhs, lhs.rows - 1, rhs, rhs.getRow(), null_direction_hint); + int cmp = compareCursors(lhs, lhs.rows - 1, rhs, rhs.getRow(), lhs.sort_columns_size, null_direction_hint); return cmp < 0; } @@ -222,19 +235,135 @@ Chunk getRowFromChunk(const Chunk & chunk, size_t pos) return result; } -void inline addRange(PaddedPODArray & left_map, size_t start, size_t end) +void inline addRange(PaddedPODArray & values, UInt64 start, UInt64 end) { assert(end > start); - for (size_t i = start; i < end; ++i) - left_map.push_back(i); + for (UInt64 i = start; i < end; ++i) + values.push_back(i); } -void inline addMany(PaddedPODArray & left_or_right_map, size_t idx, size_t num) +void inline addMany(PaddedPODArray & values, UInt64 value, size_t num) { - for (size_t i = 0; i < num; ++i) - left_or_right_map.push_back(idx); + values.resize_fill(values.size() + num, value); +} } +JoinKeyRow::JoinKeyRow(const FullMergeJoinCursor & cursor, size_t pos) +{ + row.reserve(cursor->sort_columns.size()); + for (const auto & col : cursor->sort_columns) + { + auto new_col = col->cloneEmpty(); + new_col->insertFrom(*col, pos); + row.push_back(std::move(new_col)); + } + if (const IColumn * asof_column = cursor.getAsofColumn()) + { + if (const auto * nullable_asof_column = checkAndGetColumn(asof_column)) + { + /// We save matched column, and since NULL do not match anything, we can't use it as a key + chassert(!nullable_asof_column->isNullAt(pos)); + asof_column = nullable_asof_column->getNestedColumnPtr().get(); + } + auto new_col = asof_column->cloneEmpty(); + new_col->insertFrom(*asof_column, pos); + row.push_back(std::move(new_col)); + } +} + +void JoinKeyRow::reset() +{ + row.clear(); +} + +bool JoinKeyRow::equals(const FullMergeJoinCursor & cursor) const +{ + if (row.empty()) + return false; + + for (size_t i = 0; i < cursor->sort_columns_size; ++i) + { + // int cmp = this->row[i]->compareAt(0, cursor->getRow(), *(cursor->sort_columns[i]), cursor->desc[i].nulls_direction); + int cmp = nullableCompareAt(*this->row[i], *cursor->sort_columns[i], 0, cursor->getRow(), cursor->desc[i].nulls_direction); + if (cmp != 0) + return false; + } + return true; +} + +bool JoinKeyRow::asofMatch(const FullMergeJoinCursor & cursor, ASOFJoinInequality asof_inequality) const +{ + chassert(this->row.size() == cursor->sort_columns_size + 1); + if (!equals(cursor)) + return false; + + const auto & asof_row = row.back(); + if (isNullAt(*asof_row, 0) || isNullAt(*cursor.getAsofColumn(), cursor->getRow())) + return false; + + int cmp = 0; + if (const auto * nullable_column = checkAndGetColumn(cursor.getAsofColumn())) + cmp = nullable_column->getNestedColumn().compareAt(cursor->getRow(), 0, *asof_row, 1); + else + cmp = cursor.getAsofColumn()->compareAt(cursor->getRow(), 0, *asof_row, 1); + + return (asof_inequality == ASOFJoinInequality::Less && cmp < 0) + || (asof_inequality == ASOFJoinInequality::LessOrEquals && cmp <= 0) + || (asof_inequality == ASOFJoinInequality::Greater && cmp > 0) + || (asof_inequality == ASOFJoinInequality::GreaterOrEquals && cmp >= 0); +} + +void AnyJoinState::set(size_t source_num, const FullMergeJoinCursor & cursor) +{ + assert(cursor->rows); + keys[source_num] = JoinKeyRow(cursor, cursor->rows - 1); +} + +void AnyJoinState::reset(size_t source_num) +{ + keys[source_num].reset(); + value.clear(); +} + +void AnyJoinState::setValue(Chunk value_) +{ + value = std::move(value_); +} + +bool AnyJoinState::empty() const { return keys[0].row.empty() && keys[1].row.empty(); } + + +void AsofJoinState::set(const FullMergeJoinCursor & rcursor, size_t rpos) +{ + key = JoinKeyRow(rcursor, rpos); + value = rcursor.getCurrent().clone(); + value_row = rpos; +} + +void AsofJoinState::reset() +{ + key.reset(); + value.clear(); +} + +FullMergeJoinCursor::FullMergeJoinCursor(const Block & sample_block_, const SortDescription & description_, bool is_asof) + : sample_block(materializeBlock(sample_block_).cloneEmpty()) + , desc(description_) +{ + if (desc.empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Got empty sort description for FullMergeJoinCursor"); + + if (is_asof) + { + /// For ASOF join prefix of sort description is used for equality comparison + /// and the last column is used for inequality comparison and is handled separately + + auto asof_column_description = desc.back(); + desc.pop_back(); + + chassert(asof_column_description.direction == 1 && asof_column_description.nulls_direction == 1); + asof_column_position = sample_block.getPositionByName(asof_column_description.column_name); + } } const Chunk & FullMergeJoinCursor::getCurrent() const @@ -260,6 +389,10 @@ void FullMergeJoinCursor::setChunk(Chunk && chunk) return; } + // should match the structure of sample_block (after materialization) + convertToFullIfConst(chunk); + convertToFullIfSparse(chunk); + current_chunk = std::move(chunk); cursor = SortCursorImpl(sample_block, current_chunk.getColumns(), desc); } @@ -269,48 +402,103 @@ bool FullMergeJoinCursor::fullyCompleted() const return !cursor.isValid() && recieved_all_blocks; } +String FullMergeJoinCursor::dump() const +{ + Strings row_dump; + if (cursor.isValid()) + { + Field val; + for (size_t i = 0; i < cursor.sort_columns_size; ++i) + { + cursor.sort_columns[i]->get(cursor.getRow(), val); + row_dump.push_back(val.dump()); + } + + if (const auto * asof_column = getAsofColumn()) + { + asof_column->get(cursor.getRow(), val); + row_dump.push_back(val.dump()); + } + } + + return fmt::format("<{}/{}{}>[{}]", + cursor.getRow(), cursor.rows, + recieved_all_blocks ? "(finished)" : "", + fmt::join(row_dump, ", ")); +} + MergeJoinAlgorithm::MergeJoinAlgorithm( - JoinPtr table_join_, + JoinKind kind_, + JoinStrictness strictness_, + const TableJoin::JoinOnClause & on_clause_, const Blocks & input_headers, size_t max_block_size_) - : table_join(table_join_) + : kind(kind_) + , strictness(strictness_) , max_block_size(max_block_size_) , log(getLogger("MergeJoinAlgorithm")) { if (input_headers.size() != 2) throw Exception(ErrorCodes::LOGICAL_ERROR, "MergeJoinAlgorithm requires exactly two inputs"); - auto strictness = table_join->getTableJoin().strictness(); - if (strictness != JoinStrictness::Any && strictness != JoinStrictness::All) + if (strictness != JoinStrictness::Any && strictness != JoinStrictness::All && strictness != JoinStrictness::Asof) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "MergeJoinAlgorithm is not implemented for strictness {}", strictness); - auto kind = table_join->getTableJoin().kind(); + if (strictness == JoinStrictness::Asof) + { + if (kind != JoinKind::Left && kind != JoinKind::Inner) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "MergeJoinAlgorithm does not implement ASOF {} join", kind); + } + if (!isInner(kind) && !isLeft(kind) && !isRight(kind) && !isFull(kind)) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "MergeJoinAlgorithm is not implemented for kind {}", kind); - const auto & join_on = table_join->getTableJoin().getOnlyClause(); - - if (join_on.on_filter_condition_left || join_on.on_filter_condition_right) + if (on_clause_.on_filter_condition_left || on_clause_.on_filter_condition_right) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "MergeJoinAlgorithm does not support ON filter conditions"); cursors = { - createCursor(input_headers[0], join_on.key_names_left), - createCursor(input_headers[1], join_on.key_names_right) + createCursor(input_headers[0], on_clause_.key_names_left, strictness), + createCursor(input_headers[1], on_clause_.key_names_right, strictness), }; +} - for (const auto & [left_key, right_key] : table_join->getTableJoin().leftToRightKeyRemap()) +MergeJoinAlgorithm::MergeJoinAlgorithm( + JoinPtr join_ptr, + const Blocks & input_headers, + size_t max_block_size_) + : MergeJoinAlgorithm( + join_ptr->getTableJoin().kind(), + join_ptr->getTableJoin().strictness(), + join_ptr->getTableJoin().getOnlyClause(), + input_headers, + max_block_size_) +{ + for (const auto & [left_key, right_key] : join_ptr->getTableJoin().leftToRightKeyRemap()) { size_t left_idx = input_headers[0].getPositionByName(left_key); size_t right_idx = input_headers[1].getPositionByName(right_key); left_to_right_key_remap[left_idx] = right_idx; } - const auto *smjPtr = typeid_cast(table_join.get()); + const auto *smjPtr = typeid_cast(join_ptr.get()); if (smjPtr) { null_direction_hint = smjPtr->getNullDirection(); } + if (strictness == JoinStrictness::Asof) + setAsofInequality(join_ptr->getTableJoin().getAsofInequality()); +} + +void MergeJoinAlgorithm::setAsofInequality(ASOFJoinInequality asof_inequality_) +{ + if (strictness != JoinStrictness::Asof) + throw Exception(ErrorCodes::LOGICAL_ERROR, "setAsofInequality is only supported for ASOF joins"); + + if (asof_inequality_ == ASOFJoinInequality::None) + throw Exception(ErrorCodes::LOGICAL_ERROR, "ASOF inequality cannot be None"); + + asof_inequality = asof_inequality_; } void MergeJoinAlgorithm::logElapsed(double seconds) @@ -398,7 +586,7 @@ struct AllJoinImpl size_t lnum = nextDistinct(left_cursor.cursor); size_t rnum = nextDistinct(right_cursor.cursor); - bool all_fit_in_block = std::max(left_map.size(), right_map.size()) + lnum * rnum <= max_block_size; + bool all_fit_in_block = !max_block_size || std::max(left_map.size(), right_map.size()) + lnum * rnum <= max_block_size; bool have_all_ranges = left_cursor.cursor.isValid() && right_cursor.cursor.isValid(); if (all_fit_in_block && have_all_ranges) { @@ -412,7 +600,7 @@ struct AllJoinImpl else { assert(state == nullptr); - state = std::make_unique(left_cursor.cursor, lpos, right_cursor.cursor, rpos); + state = std::make_unique(left_cursor, lpos, right_cursor, rpos); state->addRange(0, left_cursor.getCurrent().clone(), lpos, lnum); state->addRange(1, right_cursor.getCurrent().clone(), rpos, rnum); return; @@ -457,6 +645,17 @@ void dispatchKind(JoinKind kind, Args && ... args) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Unsupported join kind: \"{}\"", kind); } +MutableColumns MergeJoinAlgorithm::getEmptyResultColumns() const +{ + MutableColumns result_cols; + for (size_t i = 0; i < 2; ++i) + { + for (const auto & col : cursors[i]->sampleColumns()) + result_cols.push_back(col->cloneEmpty()); + } + return result_cols; +} + std::optional MergeJoinAlgorithm::handleAllJoinState() { if (all_join_state && all_join_state->finished()) @@ -470,7 +669,7 @@ std::optional MergeJoinAlgorithm::handleAllJoinState /// Accumulate blocks with same key in all_join_state for (size_t i = 0; i < 2; ++i) { - if (cursors[i]->cursor.isValid() && all_join_state->keys[i].equals(cursors[i]->cursor)) + if (cursors[i]->cursor.isValid() && all_join_state->keys[i].equals(*cursors[i])) { size_t pos = cursors[i]->cursor.getRow(); size_t num = nextDistinct(cursors[i]->cursor); @@ -490,15 +689,10 @@ std::optional MergeJoinAlgorithm::handleAllJoinState stat.max_blocks_loaded = std::max(stat.max_blocks_loaded, all_join_state->blocksStored()); /// join all rows with current key - MutableColumns result_cols; - for (size_t i = 0; i < 2; ++i) - { - for (const auto & col : cursors[i]->sampleColumns()) - result_cols.push_back(col->cloneEmpty()); - } + MutableColumns result_cols = getEmptyResultColumns(); size_t total_rows = 0; - while (total_rows < max_block_size) + while (!max_block_size || total_rows < max_block_size) { const auto & left_range = all_join_state->getLeft(); const auto & right_range = all_join_state->getRight(); @@ -523,7 +717,52 @@ std::optional MergeJoinAlgorithm::handleAllJoinState return {}; } -MergeJoinAlgorithm::Status MergeJoinAlgorithm::allJoin(JoinKind kind) +std::optional MergeJoinAlgorithm::handleAsofJoinState() +{ + if (strictness != JoinStrictness::Asof) + return {}; + + if (!cursors[1]->fullyCompleted()) + return {}; + + auto & left_cursor = *cursors[0]; + const auto & left_columns = left_cursor.getCurrent().getColumns(); + + MutableColumns result_cols = getEmptyResultColumns(); + + while (left_cursor->isValid() && asof_join_state.hasMatch(left_cursor, asof_inequality)) + { + size_t i = 0; + for (const auto & col : left_columns) + result_cols[i++]->insertFrom(*col, left_cursor->getRow()); + for (const auto & col : asof_join_state.value.getColumns()) + result_cols[i++]->insertFrom(*col, asof_join_state.value_row); + chassert(i == result_cols.size()); + left_cursor->next(); + } + + while (isLeft(kind) && left_cursor->isValid()) + { + /// return row with default values at right side + size_t i = 0; + for (const auto & col : left_columns) + result_cols[i++]->insertFrom(*col, left_cursor->getRow()); + for (; i < result_cols.size(); ++i) + result_cols[i]->insertDefault(); + chassert(i == result_cols.size()); + + left_cursor->next(); + } + + size_t result_rows = result_cols.empty() ? 0 : result_cols.front()->size(); + if (result_rows) + return Status(Chunk(std::move(result_cols), result_rows)); + + return {}; +} + + +MergeJoinAlgorithm::Status MergeJoinAlgorithm::allJoin() { PaddedPODArray idx_map[2]; @@ -586,7 +825,7 @@ struct AnyJoinImpl FullMergeJoinCursor & right_cursor, PaddedPODArray & left_map, PaddedPODArray & right_map, - AnyJoinState & state, + AnyJoinState & any_join_state, int null_direction_hint) { assert(enabled); @@ -647,21 +886,21 @@ struct AnyJoinImpl } } - /// Remember index of last joined row to propagate it to next block + /// Remember last joined row to propagate it to next block - state.setValue({}); + any_join_state.setValue({}); if (!left_cursor->isValid()) { - state.set(0, left_cursor.cursor); + any_join_state.set(0, left_cursor); if (cmp == 0 && isLeft(kind)) - state.setValue(getRowFromChunk(right_cursor.getCurrent(), rpos)); + any_join_state.setValue(getRowFromChunk(right_cursor.getCurrent(), rpos)); } if (!right_cursor->isValid()) { - state.set(1, right_cursor.cursor); + any_join_state.set(1, right_cursor); if (cmp == 0 && isRight(kind)) - state.setValue(getRowFromChunk(left_cursor.getCurrent(), lpos)); + any_join_state.setValue(getRowFromChunk(left_cursor.getCurrent(), lpos)); } } }; @@ -671,40 +910,34 @@ std::optional MergeJoinAlgorithm::handleAnyJoinState if (any_join_state.empty()) return {}; - auto kind = table_join->getTableJoin().kind(); - Chunk result; for (size_t source_num = 0; source_num < 2; ++source_num) { auto & current = *cursors[source_num]; - auto & state = any_join_state; - if (any_join_state.keys[source_num].equals(current.cursor)) + if (any_join_state.keys[source_num].equals(current)) { size_t start_pos = current->getRow(); size_t length = nextDistinct(current.cursor); if (length && isLeft(kind) && source_num == 0) { - if (state.value) - result = copyChunkResized(current.getCurrent(), state.value, start_pos, length); + if (any_join_state.value) + result = copyChunkResized(current.getCurrent(), any_join_state.value, start_pos, length); else result = createBlockWithDefaults(source_num, start_pos, length); } if (length && isRight(kind) && source_num == 1) { - if (state.value) - result = copyChunkResized(state.value, current.getCurrent(), start_pos, length); + if (any_join_state.value) + result = copyChunkResized(any_join_state.value, current.getCurrent(), start_pos, length); else result = createBlockWithDefaults(source_num, start_pos, length); } - /// We've found row with other key, no need to skip more rows with current key if (current->isValid()) - { - state.keys[source_num].reset(); - } + any_join_state.keys[source_num].reset(); } else { @@ -717,7 +950,7 @@ std::optional MergeJoinAlgorithm::handleAnyJoinState return {}; } -MergeJoinAlgorithm::Status MergeJoinAlgorithm::anyJoin(JoinKind kind) +MergeJoinAlgorithm::Status MergeJoinAlgorithm::anyJoin() { if (auto result = handleAnyJoinState()) return std::move(*result); @@ -762,10 +995,151 @@ MergeJoinAlgorithm::Status MergeJoinAlgorithm::anyJoin(JoinKind kind) return Status(std::move(result)); } + +MergeJoinAlgorithm::Status MergeJoinAlgorithm::asofJoin() +{ + auto & left_cursor = *cursors[0]; + if (!left_cursor->isValid()) + return Status(0); + + auto & right_cursor = *cursors[1]; + if (!right_cursor->isValid()) + return Status(1); + + const auto & left_columns = left_cursor.getCurrent().getColumns(); + const auto & right_columns = right_cursor.getCurrent().getColumns(); + + MutableColumns result_cols = getEmptyResultColumns(); + + while (left_cursor->isValid() && right_cursor->isValid()) + { + auto lpos = left_cursor->getRow(); + auto rpos = right_cursor->getRow(); + auto cmp = compareCursors(*left_cursor, *right_cursor, null_direction_hint); + if (cmp == 0) + { + if (isNullAt(*left_cursor.getAsofColumn(), lpos)) + cmp = -1; + if (isNullAt(*right_cursor.getAsofColumn(), rpos)) + cmp = 1; + } + + if (cmp == 0) + { + auto asof_cmp = compareAsofCursors(left_cursor, right_cursor, null_direction_hint); + + if ((asof_inequality == ASOFJoinInequality::Less && asof_cmp <= -1) + || (asof_inequality == ASOFJoinInequality::LessOrEquals && asof_cmp <= 0)) + { + /// First row in right table that is greater (or equal) than current row in left table + /// matches asof join condition the best + size_t i = 0; + for (const auto & col : left_columns) + result_cols[i++]->insertFrom(*col, lpos); + for (const auto & col : right_columns) + result_cols[i++]->insertFrom(*col, rpos); + chassert(i == result_cols.size()); + + left_cursor->next(); + continue; + } + + if (asof_inequality == ASOFJoinInequality::Less || asof_inequality == ASOFJoinInequality::LessOrEquals) + { + /// Asof condition is not (yet) satisfied, skip row in right table + right_cursor->next(); + continue; + } + + if ((asof_inequality == ASOFJoinInequality::Greater && asof_cmp >= 1) + || (asof_inequality == ASOFJoinInequality::GreaterOrEquals && asof_cmp >= 0)) + { + /// condition is satisfied, remember this row and move next to try to find better match + asof_join_state.set(right_cursor, rpos); + right_cursor->next(); + continue; + } + + if (asof_inequality == ASOFJoinInequality::Greater || asof_inequality == ASOFJoinInequality::GreaterOrEquals) + { + /// Asof condition is not satisfied anymore, use last matched row from right table + if (asof_join_state.hasMatch(left_cursor, asof_inequality)) + { + size_t i = 0; + for (const auto & col : left_columns) + result_cols[i++]->insertFrom(*col, lpos); + for (const auto & col : asof_join_state.value.getColumns()) + result_cols[i++]->insertFrom(*col, asof_join_state.value_row); + chassert(i == result_cols.size()); + } + else + { + asof_join_state.reset(); + if (isLeft(kind)) + { + /// return row with default values at right side + size_t i = 0; + for (const auto & col : left_columns) + result_cols[i++]->insertFrom(*col, lpos); + for (; i < result_cols.size(); ++i) + result_cols[i]->insertDefault(); + chassert(i == result_cols.size()); + } + } + left_cursor->next(); + continue; + } + + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "TODO: implement ASOF equality join"); + } + else if (cmp < 0) + { + if (asof_join_state.hasMatch(left_cursor, asof_inequality)) + { + size_t i = 0; + for (const auto & col : left_columns) + result_cols[i++]->insertFrom(*col, lpos); + for (const auto & col : asof_join_state.value.getColumns()) + result_cols[i++]->insertFrom(*col, asof_join_state.value_row); + chassert(i == result_cols.size()); + left_cursor->next(); + continue; + } + else + { + asof_join_state.reset(); + } + + /// no matches for rows in left table, just pass them through + size_t num = nextDistinct(*left_cursor); + + if (isLeft(kind) && num) + { + /// return them with default values at right side + size_t i = 0; + for (const auto & col : left_columns) + result_cols[i++]->insertRangeFrom(*col, lpos, num); + for (; i < result_cols.size(); ++i) + result_cols[i]->insertManyDefaults(num); + chassert(i == result_cols.size()); + } + } + else + { + /// skip rows in right table until we find match for current row in left table + nextDistinct(*right_cursor); + } + } + size_t num_rows = result_cols.empty() ? 0 : result_cols.front()->size(); + return Status(Chunk(std::move(result_cols), num_rows)); +} + + /// if `source_num == 0` get data from left cursor and fill defaults at right /// otherwise - vice versa Chunk MergeJoinAlgorithm::createBlockWithDefaults(size_t source_num, size_t start, size_t num_rows) const { + ColumnRawPtrs cols; { const auto & columns_left = source_num == 0 ? cursors[0]->getCurrent().getColumns() : cursors[0]->sampleColumns(); @@ -788,7 +1162,6 @@ Chunk MergeJoinAlgorithm::createBlockWithDefaults(size_t source_num, size_t star cols.push_back(col.get()); } } - Chunk result_chunk; copyColumnsResized(cols, start, num_rows, result_chunk); return result_chunk; @@ -804,7 +1177,6 @@ Chunk MergeJoinAlgorithm::createBlockWithDefaults(size_t source_num) IMergingAlgorithm::Status MergeJoinAlgorithm::merge() { - auto kind = table_join->getTableJoin().kind(); if (!cursors[0]->cursor.isValid() && !cursors[0]->fullyCompleted()) return Status(0); @@ -812,11 +1184,11 @@ IMergingAlgorithm::Status MergeJoinAlgorithm::merge() if (!cursors[1]->cursor.isValid() && !cursors[1]->fullyCompleted()) return Status(1); - if (auto result = handleAllJoinState()) - { return std::move(*result); - } + + if (auto result = handleAsofJoinState()) + return std::move(*result); if (cursors[0]->fullyCompleted() || cursors[1]->fullyCompleted()) { @@ -830,7 +1202,7 @@ IMergingAlgorithm::Status MergeJoinAlgorithm::merge() } /// check if blocks are not intersecting at all - if (int cmp = totallyCompare(cursors[0]->cursor, cursors[1]->cursor, null_direction_hint); cmp != 0) + if (int cmp = totallyCompare(cursors[0]->cursor, cursors[1]->cursor, null_direction_hint); cmp != 0 && strictness != JoinStrictness::Asof) { if (cmp < 0) { @@ -849,13 +1221,14 @@ IMergingAlgorithm::Status MergeJoinAlgorithm::merge() } } - auto strictness = table_join->getTableJoin().strictness(); - if (strictness == JoinStrictness::Any) - return anyJoin(kind); + return anyJoin(); if (strictness == JoinStrictness::All) - return allJoin(kind); + return allJoin(); + + if (strictness == JoinStrictness::Asof) + return asofJoin(); throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Unsupported strictness '{}'", strictness); } @@ -874,9 +1247,26 @@ MergeJoinTransform::MergeJoinTransform( /* always_read_till_end_= */ false, /* empty_chunk_on_finish_= */ true, table_join, input_headers, max_block_size) - , log(getLogger("MergeJoinTransform")) { - LOG_TRACE(log, "Use MergeJoinTransform"); +} + +MergeJoinTransform::MergeJoinTransform( + JoinKind kind_, + JoinStrictness strictness_, + const TableJoin::JoinOnClause & on_clause_, + const Blocks & input_headers, + const Block & output_header, + size_t max_block_size, + UInt64 limit_hint_) + : IMergingTransform( + input_headers, + output_header, + /* have_all_inputs_= */ true, + limit_hint_, + /* always_read_till_end_= */ false, + /* empty_chunk_on_finish_= */ true, + kind_, strictness_, on_clause_, input_headers, max_block_size) +{ } void MergeJoinTransform::onFinish() diff --git a/src/Processors/Transforms/MergeJoinTransform.h b/src/Processors/Transforms/MergeJoinTransform.h index cf9331abd59..d37a0b9f3ae 100644 --- a/src/Processors/Transforms/MergeJoinTransform.h +++ b/src/Processors/Transforms/MergeJoinTransform.h @@ -8,6 +8,7 @@ #include #include #include +#include #include @@ -19,6 +20,7 @@ #include #include #include +#include namespace Poco { class Logger; } @@ -35,57 +37,28 @@ using FullMergeJoinCursorPtr = std::unique_ptr; /// Used instead of storing previous block struct JoinKeyRow { - std::vector row; - JoinKeyRow() = default; - explicit JoinKeyRow(const SortCursorImpl & impl_, size_t pos) - { - row.reserve(impl_.sort_columns.size()); - for (const auto & col : impl_.sort_columns) - { - auto new_col = col->cloneEmpty(); - new_col->insertFrom(*col, pos); - row.push_back(std::move(new_col)); - } - } + JoinKeyRow(const FullMergeJoinCursor & cursor, size_t pos); - void reset() - { - row.clear(); - } + bool equals(const FullMergeJoinCursor & cursor) const; + bool asofMatch(const FullMergeJoinCursor & cursor, ASOFJoinInequality asof_inequality) const; - bool equals(const SortCursorImpl & impl) const - { - if (row.empty()) - return false; + void reset(); - assert(this->row.size() == impl.sort_columns_size); - for (size_t i = 0; i < impl.sort_columns_size; ++i) - { - int cmp = this->row[i]->compareAt(0, impl.getRow(), *impl.sort_columns[i], impl.desc[i].nulls_direction); - if (cmp != 0) - return false; - } - return true; - } + std::vector row; }; /// Remembers previous key if it was joined in previous block class AnyJoinState : boost::noncopyable { public: - AnyJoinState() = default; + void set(size_t source_num, const FullMergeJoinCursor & cursor); + void setValue(Chunk value_); - void set(size_t source_num, const SortCursorImpl & cursor) - { - assert(cursor.rows); - keys[source_num] = JoinKeyRow(cursor, cursor.rows - 1); - } + void reset(size_t source_num); - void setValue(Chunk value_) { value = std::move(value_); } - - bool empty() const { return keys[0].row.empty() && keys[1].row.empty(); } + bool empty() const; /// current keys JoinKeyRow keys[2]; @@ -118,8 +91,8 @@ public: Chunk chunk; }; - AllJoinState(const SortCursorImpl & lcursor, size_t lpos, - const SortCursorImpl & rcursor, size_t rpos) + AllJoinState(const FullMergeJoinCursor & lcursor, size_t lpos, + const FullMergeJoinCursor & rcursor, size_t rpos) : keys{JoinKeyRow(lcursor, lpos), JoinKeyRow(rcursor, rpos)} { } @@ -187,17 +160,32 @@ private: size_t ridx = 0; }; + +class AsofJoinState : boost::noncopyable +{ +public: + void set(const FullMergeJoinCursor & rcursor, size_t rpos); + void reset(); + + bool hasMatch(const FullMergeJoinCursor & cursor, ASOFJoinInequality asof_inequality) const + { + if (value.empty()) + return false; + return key.asofMatch(cursor, asof_inequality); + } + + JoinKeyRow key; + Chunk value; + size_t value_row = 0; +}; + /* * Wrapper for SortCursorImpl */ class FullMergeJoinCursor : boost::noncopyable { public: - explicit FullMergeJoinCursor(const Block & sample_block_, const SortDescription & description_) - : sample_block(sample_block_.cloneEmpty()) - , desc(description_) - { - } + explicit FullMergeJoinCursor(const Block & sample_block_, const SortDescription & description_, bool is_asof = false); bool fullyCompleted() const; void setChunk(Chunk && chunk); @@ -207,17 +195,31 @@ public: SortCursorImpl * operator-> () { return &cursor; } const SortCursorImpl * operator-> () const { return &cursor; } + SortCursorImpl & operator* () { return cursor; } + const SortCursorImpl & operator* () const { return cursor; } + SortCursorImpl cursor; const Block & sampleBlock() const { return sample_block; } Columns sampleColumns() const { return sample_block.getColumns(); } + const IColumn * getAsofColumn() const + { + if (!asof_column_position) + return nullptr; + return cursor.all_columns[*asof_column_position]; + } + + String dump() const; + private: Block sample_block; SortDescription desc; Chunk current_chunk; bool recieved_all_blocks = false; + + std::optional asof_column_position; }; /* @@ -227,22 +229,33 @@ private: class MergeJoinAlgorithm final : public IMergingAlgorithm { public: - explicit MergeJoinAlgorithm(JoinPtr table_join, const Blocks & input_headers, size_t max_block_size_); + MergeJoinAlgorithm(JoinKind kind_, + JoinStrictness strictness_, + const TableJoin::JoinOnClause & on_clause_, + const Blocks & input_headers, + size_t max_block_size_); + + MergeJoinAlgorithm(JoinPtr join_ptr, const Blocks & input_headers, size_t max_block_size_); const char * getName() const override { return "MergeJoinAlgorithm"; } void initialize(Inputs inputs) override; void consume(Input & input, size_t source_num) override; Status merge() override; - void logElapsed(double seconds); + void setAsofInequality(ASOFJoinInequality asof_inequality_); + void logElapsed(double seconds); private: std::optional handleAnyJoinState(); - Status anyJoin(JoinKind kind); + Status anyJoin(); std::optional handleAllJoinState(); - Status allJoin(JoinKind kind); + Status allJoin(); + std::optional handleAsofJoinState(); + Status asofJoin(); + + MutableColumns getEmptyResultColumns() const; Chunk createBlockWithDefaults(size_t source_num); Chunk createBlockWithDefaults(size_t source_num, size_t start, size_t num_rows) const; @@ -250,12 +263,15 @@ private: std::unordered_map left_to_right_key_remap; std::array cursors; + ASOFJoinInequality asof_inequality = ASOFJoinInequality::None; - /// Keep some state to make connection between data in different blocks + /// Keep some state to make handle data from different blocks AnyJoinState any_join_state; std::unique_ptr all_join_state; + AsofJoinState asof_join_state; - JoinPtr table_join; + JoinKind kind; + JoinStrictness strictness; size_t max_block_size; int null_direction_hint = 1; @@ -285,12 +301,21 @@ public: size_t max_block_size, UInt64 limit_hint = 0); + MergeJoinTransform( + JoinKind kind_, + JoinStrictness strictness_, + const TableJoin::JoinOnClause & on_clause_, + const Blocks & input_headers, + const Block & output_header, + size_t max_block_size, + UInt64 limit_hint_ = 0); + String getName() const override { return "MergeJoinTransform"; } + void setAsofInequality(ASOFJoinInequality asof_inequality_) { algorithm.setAsofInequality(asof_inequality_); } + protected: void onFinish() override; - - LoggerPtr log; }; } diff --git a/src/Processors/Transforms/MergingAggregatedMemoryEfficientTransform.cpp b/src/Processors/Transforms/MergingAggregatedMemoryEfficientTransform.cpp index fc40c6894bb..ea9ebb0f96e 100644 --- a/src/Processors/Transforms/MergingAggregatedMemoryEfficientTransform.cpp +++ b/src/Processors/Transforms/MergingAggregatedMemoryEfficientTransform.cpp @@ -30,10 +30,10 @@ void GroupingAggregatedTransform::pushData(Chunks chunks, Int32 bucket, bool is_ auto info = std::make_shared(); info->bucket_num = bucket; info->is_overflows = is_overflows; - info->chunks = std::make_unique(std::move(chunks)); + info->chunks = std::make_shared(std::move(chunks)); Chunk chunk; - chunk.setChunkInfo(std::move(info)); + chunk.getChunkInfos().add(std::move(info)); output.push(std::move(chunk)); } @@ -255,11 +255,10 @@ void GroupingAggregatedTransform::addChunk(Chunk chunk, size_t input) if (!chunk.hasRows()) return; - const auto & info = chunk.getChunkInfo(); - if (!info) + if (chunk.getChunkInfos().empty()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk info was not set for chunk in GroupingAggregatedTransform."); - if (const auto * agg_info = typeid_cast(info.get())) + if (auto agg_info = chunk.getChunkInfos().get()) { Int32 bucket = agg_info->bucket_num; bool is_overflows = agg_info->is_overflows; @@ -275,7 +274,7 @@ void GroupingAggregatedTransform::addChunk(Chunk chunk, size_t input) last_bucket_number[input] = bucket; } } - else if (typeid_cast(info.get())) + else if (chunk.getChunkInfos().get()) { single_level_chunks.emplace_back(std::move(chunk)); } @@ -304,7 +303,11 @@ void GroupingAggregatedTransform::work() Int32 bucket = cur_block.info.bucket_num; auto chunk_info = std::make_shared(); chunk_info->bucket_num = bucket; - chunks_map[bucket].emplace_back(Chunk(cur_block.getColumns(), cur_block.rows(), std::move(chunk_info))); + + auto chunk = Chunk(cur_block.getColumns(), cur_block.rows()); + chunk.getChunkInfos().add(std::move(chunk_info)); + + chunks_map[bucket].emplace_back(std::move(chunk)); } } } @@ -319,9 +322,7 @@ MergingAggregatedBucketTransform::MergingAggregatedBucketTransform( void MergingAggregatedBucketTransform::transform(Chunk & chunk) { - const auto & info = chunk.getChunkInfo(); - const auto * chunks_to_merge = typeid_cast(info.get()); - + auto chunks_to_merge = chunk.getChunkInfos().get(); if (!chunks_to_merge) throw Exception(ErrorCodes::LOGICAL_ERROR, "MergingAggregatedSimpleTransform chunk must have ChunkInfo with type ChunksToMerge."); @@ -330,11 +331,10 @@ void MergingAggregatedBucketTransform::transform(Chunk & chunk) BlocksList blocks_list; for (auto & cur_chunk : *chunks_to_merge->chunks) { - const auto & cur_info = cur_chunk.getChunkInfo(); - if (!cur_info) + if (cur_chunk.getChunkInfos().empty()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk info was not set for chunk in MergingAggregatedBucketTransform."); - if (const auto * agg_info = typeid_cast(cur_info.get())) + if (auto agg_info = cur_chunk.getChunkInfos().get()) { Block block = header.cloneWithColumns(cur_chunk.detachColumns()); block.info.is_overflows = agg_info->is_overflows; @@ -342,7 +342,7 @@ void MergingAggregatedBucketTransform::transform(Chunk & chunk) blocks_list.emplace_back(std::move(block)); } - else if (typeid_cast(cur_info.get())) + else if (cur_chunk.getChunkInfos().get()) { Block block = header.cloneWithColumns(cur_chunk.detachColumns()); block.info.is_overflows = false; @@ -361,7 +361,7 @@ void MergingAggregatedBucketTransform::transform(Chunk & chunk) res_info->is_overflows = chunks_to_merge->is_overflows; res_info->bucket_num = chunks_to_merge->bucket_num; res_info->chunk_num = chunks_to_merge->chunk_num; - chunk.setChunkInfo(std::move(res_info)); + chunk.getChunkInfos().add(std::move(res_info)); auto block = params->aggregator.mergeBlocks(blocks_list, params->final, is_cancelled); @@ -405,11 +405,7 @@ bool SortingAggregatedTransform::tryPushChunk() void SortingAggregatedTransform::addChunk(Chunk chunk, size_t from_input) { - const auto & info = chunk.getChunkInfo(); - if (!info) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk info was not set for chunk in SortingAggregatedTransform."); - - const auto * agg_info = typeid_cast(info.get()); + auto agg_info = chunk.getChunkInfos().get(); if (!agg_info) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk should have AggregatedChunkInfo in SortingAggregatedTransform."); diff --git a/src/Processors/Transforms/MergingAggregatedMemoryEfficientTransform.h b/src/Processors/Transforms/MergingAggregatedMemoryEfficientTransform.h index 77ee3034ffc..3a3c1bd9c1e 100644 --- a/src/Processors/Transforms/MergingAggregatedMemoryEfficientTransform.h +++ b/src/Processors/Transforms/MergingAggregatedMemoryEfficientTransform.h @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -142,9 +143,9 @@ private: void addChunk(Chunk chunk, size_t from_input); }; -struct ChunksToMerge : public ChunkInfo +struct ChunksToMerge : public ChunkInfoCloneable { - std::unique_ptr chunks; + std::shared_ptr chunks; Int32 bucket_num = -1; bool is_overflows = false; UInt64 chunk_num = 0; // chunk number in order of generation, used during memory bound merging to restore chunks order diff --git a/src/Processors/Transforms/MergingAggregatedTransform.cpp b/src/Processors/Transforms/MergingAggregatedTransform.cpp index ad723da7527..446e60a0b81 100644 --- a/src/Processors/Transforms/MergingAggregatedTransform.cpp +++ b/src/Processors/Transforms/MergingAggregatedTransform.cpp @@ -32,11 +32,10 @@ void MergingAggregatedTransform::consume(Chunk chunk) total_input_rows += input_rows; ++total_input_blocks; - const auto & info = chunk.getChunkInfo(); - if (!info) + if (chunk.getChunkInfos().empty()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk info was not set for chunk in MergingAggregatedTransform."); - if (const auto * agg_info = typeid_cast(info.get())) + if (auto agg_info = chunk.getChunkInfos().get()) { /** If the remote servers used a two-level aggregation method, * then blocks will contain information about the number of the bucket. @@ -49,7 +48,7 @@ void MergingAggregatedTransform::consume(Chunk chunk) bucket_to_blocks[agg_info->bucket_num].emplace_back(std::move(block)); } - else if (typeid_cast(info.get())) + else if (chunk.getChunkInfos().get()) { auto block = getInputPort().getHeader().cloneWithColumns(chunk.getColumns()); block.info.is_overflows = false; @@ -89,7 +88,8 @@ Chunk MergingAggregatedTransform::generate() UInt64 num_rows = block.rows(); Chunk chunk(block.getColumns(), num_rows); - chunk.setChunkInfo(std::move(info)); + + chunk.getChunkInfos().add(std::move(info)); return chunk; } diff --git a/src/Processors/Transforms/PlanSquashingTransform.cpp b/src/Processors/Transforms/PlanSquashingTransform.cpp index 0f433165f14..ee4dfa6a64e 100644 --- a/src/Processors/Transforms/PlanSquashingTransform.cpp +++ b/src/Processors/Transforms/PlanSquashingTransform.cpp @@ -10,20 +10,20 @@ namespace ErrorCodes } PlanSquashingTransform::PlanSquashingTransform( - const Block & header, size_t min_block_size_rows, size_t min_block_size_bytes) - : IInflatingTransform(header, header), squashing(header, min_block_size_rows, min_block_size_bytes) + Block header_, size_t min_block_size_rows, size_t min_block_size_bytes) + : IInflatingTransform(header_, header_) + , squashing(header_, min_block_size_rows, min_block_size_bytes) { } void PlanSquashingTransform::consume(Chunk chunk) { - if (Chunk current_chunk = squashing.add(std::move(chunk)); current_chunk.hasChunkInfo()) - squashed_chunk.swap(current_chunk); + squashed_chunk = squashing.add(std::move(chunk)); } Chunk PlanSquashingTransform::generate() { - if (!squashed_chunk.hasChunkInfo()) + if (!squashed_chunk) throw Exception(ErrorCodes::LOGICAL_ERROR, "Can't generate chunk in SimpleSquashingChunksTransform"); Chunk result_chunk; @@ -33,12 +33,11 @@ Chunk PlanSquashingTransform::generate() bool PlanSquashingTransform::canGenerate() { - return squashed_chunk.hasChunkInfo(); + return bool(squashed_chunk); } Chunk PlanSquashingTransform::getRemaining() { - Chunk current_chunk = squashing.flush(); - return current_chunk; + return squashing.flush(); } } diff --git a/src/Processors/Transforms/PlanSquashingTransform.h b/src/Processors/Transforms/PlanSquashingTransform.h index 4ad2ec2d089..e6db245499e 100644 --- a/src/Processors/Transforms/PlanSquashingTransform.h +++ b/src/Processors/Transforms/PlanSquashingTransform.h @@ -10,7 +10,7 @@ class PlanSquashingTransform : public IInflatingTransform { public: PlanSquashingTransform( - const Block & header, size_t min_block_size_rows, size_t min_block_size_bytes); + Block header_, size_t min_block_size_rows, size_t min_block_size_bytes); String getName() const override { return "PlanSquashingTransform"; } @@ -23,7 +23,6 @@ protected: private: Squashing squashing; Chunk squashed_chunk; - Chunk finish_chunk; }; } diff --git a/src/Processors/Transforms/ScatterByPartitionTransform.cpp b/src/Processors/Transforms/ScatterByPartitionTransform.cpp index 6e3cdc0fda1..16d265c9bcb 100644 --- a/src/Processors/Transforms/ScatterByPartitionTransform.cpp +++ b/src/Processors/Transforms/ScatterByPartitionTransform.cpp @@ -109,7 +109,7 @@ void ScatterByPartitionTransform::generateOutputChunks() hash.reset(num_rows); for (const auto & column_number : key_columns) - columns[column_number]->updateWeakHash32(hash); + hash.update(columns[column_number]->getWeakHash32()); const auto & hash_data = hash.getData(); IColumn::Selector selector(num_rows); diff --git a/src/Processors/Transforms/SelectByIndicesTransform.h b/src/Processors/Transforms/SelectByIndicesTransform.h index 480ab1a0f61..b44f5a3203e 100644 --- a/src/Processors/Transforms/SelectByIndicesTransform.h +++ b/src/Processors/Transforms/SelectByIndicesTransform.h @@ -26,7 +26,7 @@ public: void transform(Chunk & chunk) override { size_t num_rows = chunk.getNumRows(); - const auto * select_final_indices_info = typeid_cast(chunk.getChunkInfo().get()); + auto select_final_indices_info = chunk.getChunkInfos().extract(); if (!select_final_indices_info || !select_final_indices_info->select_final_indices) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk passed to SelectByIndicesTransform without indices column"); @@ -41,7 +41,6 @@ public: chunk.setColumns(std::move(columns), index_column->size()); } - chunk.setChunkInfo(nullptr); } }; diff --git a/src/Processors/Transforms/SquashingTransform.cpp b/src/Processors/Transforms/SquashingTransform.cpp index 34b733cde5e..490a57d4e23 100644 --- a/src/Processors/Transforms/SquashingTransform.cpp +++ b/src/Processors/Transforms/SquashingTransform.cpp @@ -1,12 +1,14 @@ +#include #include #include +#include "Processors/Chunk.h" namespace DB { namespace ErrorCodes { -extern const int LOGICAL_ERROR; + extern const int LOGICAL_ERROR; } SquashingTransform::SquashingTransform( @@ -18,9 +20,7 @@ SquashingTransform::SquashingTransform( void SquashingTransform::onConsume(Chunk chunk) { - Chunk planned_chunk = squashing.add(std::move(chunk)); - if (planned_chunk.hasChunkInfo()) - cur_chunk = DB::Squashing::squash(std::move(planned_chunk)); + cur_chunk = Squashing::squash(squashing.add(std::move(chunk))); } SquashingTransform::GenerateResult SquashingTransform::onGenerate() @@ -33,10 +33,7 @@ SquashingTransform::GenerateResult SquashingTransform::onGenerate() void SquashingTransform::onFinish() { - Chunk chunk = squashing.flush(); - if (chunk.hasChunkInfo()) - chunk = DB::Squashing::squash(std::move(chunk)); - finish_chunk.setColumns(chunk.getColumns(), chunk.getNumRows()); + finish_chunk = Squashing::squash(squashing.flush()); } void SquashingTransform::work() @@ -49,6 +46,7 @@ void SquashingTransform::work() } ExceptionKeepingTransform::work(); + if (finish_chunk) { data.chunk = std::move(finish_chunk); @@ -56,53 +54,36 @@ void SquashingTransform::work() } } -SimpleSquashingTransform::SimpleSquashingTransform( +SimpleSquashingChunksTransform::SimpleSquashingChunksTransform( const Block & header, size_t min_block_size_rows, size_t min_block_size_bytes) - : ISimpleTransform(header, header, false) + : IInflatingTransform(header, header) , squashing(header, min_block_size_rows, min_block_size_bytes) { } -void SimpleSquashingTransform::transform(Chunk & chunk) +void SimpleSquashingChunksTransform::consume(Chunk chunk) { - if (!finished) - { - Chunk planned_chunk = squashing.add(std::move(chunk)); - if (planned_chunk.hasChunkInfo()) - chunk = DB::Squashing::squash(std::move(planned_chunk)); - } - else - { - if (chunk.hasRows()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk expected to be empty, otherwise it will be lost"); - - chunk = squashing.flush(); - if (chunk.hasChunkInfo()) - chunk = DB::Squashing::squash(std::move(chunk)); - } + squashed_chunk = Squashing::squash(squashing.add(std::move(chunk))); } -IProcessor::Status SimpleSquashingTransform::prepare() +Chunk SimpleSquashingChunksTransform::generate() { - if (!finished && input.isFinished()) - { - if (output.isFinished()) - return Status::Finished; + if (squashed_chunk.empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Can't generate chunk in SimpleSquashingChunksTransform"); - if (!output.canPush()) - return Status::PortFull; - - if (has_output) - { - output.pushData(std::move(output_data)); - has_output = false; - return Status::PortFull; - } - - finished = true; - /// On the next call to transform() we will return all data buffered in `squashing` (if any) - return Status::Ready; - } - return ISimpleTransform::prepare(); + Chunk result; + result.swap(squashed_chunk); + return result; } + +bool SimpleSquashingChunksTransform::canGenerate() +{ + return !squashed_chunk.empty(); +} + +Chunk SimpleSquashingChunksTransform::getRemaining() +{ + return Squashing::squash(squashing.flush()); +} + } diff --git a/src/Processors/Transforms/SquashingTransform.h b/src/Processors/Transforms/SquashingTransform.h index c5b727ac6ec..092f58f2fe0 100644 --- a/src/Processors/Transforms/SquashingTransform.h +++ b/src/Processors/Transforms/SquashingTransform.h @@ -2,6 +2,7 @@ #include #include +#include #include #include @@ -29,22 +30,22 @@ private: Chunk finish_chunk; }; -/// Doesn't care about propagating exceptions and thus doesn't throw LOGICAL_ERROR if the following transform closes its input port. -class SimpleSquashingTransform : public ISimpleTransform +class SimpleSquashingChunksTransform : public IInflatingTransform { public: - explicit SimpleSquashingTransform(const Block & header, size_t min_block_size_rows, size_t min_block_size_bytes); + explicit SimpleSquashingChunksTransform(const Block & header, size_t min_block_size_rows, size_t min_block_size_bytes); String getName() const override { return "SimpleSquashingTransform"; } protected: - void transform(Chunk &) override; - - IProcessor::Status prepare() override; + void consume(Chunk chunk) override; + bool canGenerate() override; + Chunk generate() override; + Chunk getRemaining() override; private: Squashing squashing; - - bool finished = false; + Chunk squashed_chunk; }; + } diff --git a/src/Processors/Transforms/TotalsHavingTransform.cpp b/src/Processors/Transforms/TotalsHavingTransform.cpp index aa86879e62c..567997c24ab 100644 --- a/src/Processors/Transforms/TotalsHavingTransform.cpp +++ b/src/Processors/Transforms/TotalsHavingTransform.cpp @@ -6,6 +6,7 @@ #include #include +#include #include #include @@ -150,11 +151,7 @@ void TotalsHavingTransform::transform(Chunk & chunk) /// Block with values not included in `max_rows_to_group_by`. We'll postpone it. if (overflow_row) { - const auto & info = chunk.getChunkInfo(); - if (!info) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk info was not set for chunk in TotalsHavingTransform."); - - const auto * agg_info = typeid_cast(info.get()); + const auto & agg_info = chunk.getChunkInfos().get(); if (!agg_info) throw Exception(ErrorCodes::LOGICAL_ERROR, "Chunk should have AggregatedChunkInfo in TotalsHavingTransform."); diff --git a/src/Processors/Transforms/WindowTransform.cpp b/src/Processors/Transforms/WindowTransform.cpp index b9f61d30182..2b255c5120e 100644 --- a/src/Processors/Transforms/WindowTransform.cpp +++ b/src/Processors/Transforms/WindowTransform.cpp @@ -17,6 +17,9 @@ #include #include +#include +#include + #include @@ -53,7 +56,10 @@ namespace ErrorCodes { extern const int BAD_ARGUMENTS; extern const int NOT_IMPLEMENTED; + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; extern const int ILLEGAL_TYPE_OF_ARGUMENT; + extern const int TOO_FEW_ARGUMENTS_FOR_FUNCTION; + extern const int TOO_MANY_ARGUMENTS_FOR_FUNCTION; } // Interface for true window functions. It's not much of an interface, they just @@ -71,6 +77,9 @@ public: size_t function_index) const = 0; virtual std::optional getDefaultFrame() const { return {}; } + + /// Is the frame type supported by this function. + virtual bool checkWindowFrameType(const WindowTransform * /*transform*/) const { return true; } }; // Compares ORDER BY column values at given rows to find the boundaries of frame: @@ -402,6 +411,19 @@ WindowTransform::WindowTransform(const Block & input_header_, } } } + + for (const auto & workspace : workspaces) + { + if (workspace.window_function_impl) + { + if (!workspace.window_function_impl->checkWindowFrameType(this)) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unsupported window frame type for function '{}'", + workspace.aggregate_function->getName()); + } + } + + } } WindowTransform::~WindowTransform() @@ -1609,6 +1631,34 @@ struct WindowFunctionHelpers { recurrent_detail::setValueToOutputColumn(transform, function_index, value); } + + ALWAYS_INLINE static bool checkPartitionEnterFirstRow(const WindowTransform * transform) { return transform->current_row_number == 1; } + + ALWAYS_INLINE static bool checkPartitionEnterLastRow(const WindowTransform * transform) + { + /// This is for fast check. + if (!transform->partition_ended) + return false; + + auto current_row = transform->current_row; + /// checkPartitionEnterLastRow is called on each row, also move on current_row.row here. + current_row.row++; + const auto & partition_end_row = transform->partition_end; + + /// The partition end is reached, when following is true + /// - current row is the partition end row, + /// - or current row is the last row of all input. + if (current_row != partition_end_row) + { + /// when current row is not the partition end row, we need to check whether it's the last + /// input row. + if (current_row.row < transform->blockRowsNumber(current_row)) + return false; + if (partition_end_row.block != current_row.block + 1 || partition_end_row.row) + return false; + } + return true; + } }; template @@ -1663,7 +1713,7 @@ struct WindowFunctionExponentialTimeDecayedSum final : public StatefulWindowFunc { if (parameters_.size() != 1) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly one parameter", name_); } return applyVisitor(FieldVisitorConvertToNumber(), parameters_[0]); @@ -1676,7 +1726,7 @@ struct WindowFunctionExponentialTimeDecayedSum final : public StatefulWindowFunc { if (argument_types.size() != 2) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly two arguments", name_); } @@ -1760,7 +1810,7 @@ struct WindowFunctionExponentialTimeDecayedMax final : public WindowFunction { if (parameters_.size() != 1) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly one parameter", name_); } return applyVisitor(FieldVisitorConvertToNumber(), parameters_[0]); @@ -1773,7 +1823,7 @@ struct WindowFunctionExponentialTimeDecayedMax final : public WindowFunction { if (argument_types.size() != 2) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly two arguments", name_); } @@ -1835,7 +1885,7 @@ struct WindowFunctionExponentialTimeDecayedCount final : public StatefulWindowFu { if (parameters_.size() != 1) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly one parameter", name_); } return applyVisitor(FieldVisitorConvertToNumber(), parameters_[0]); @@ -1848,7 +1898,7 @@ struct WindowFunctionExponentialTimeDecayedCount final : public StatefulWindowFu { if (argument_types.size() != 1) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly one argument", name_); } @@ -1921,7 +1971,7 @@ struct WindowFunctionExponentialTimeDecayedAvg final : public StatefulWindowFunc { if (parameters_.size() != 1) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly one parameter", name_); } return applyVisitor(FieldVisitorConvertToNumber(), parameters_[0]); @@ -1934,7 +1984,7 @@ struct WindowFunctionExponentialTimeDecayedAvg final : public StatefulWindowFunc { if (argument_types.size() != 2) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly two arguments", name_); } @@ -2058,8 +2108,6 @@ namespace const WindowTransform * transform, size_t function_index, const DataTypes & argument_types); - - static void checkWindowFrameType(const WindowTransform * transform); }; } @@ -2071,7 +2119,7 @@ struct WindowFunctionNtile final : public StatefulWindowFunction : StatefulWindowFunction(name_, argument_types_, parameters_, std::make_shared()) { if (argument_types.size() != 1) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Function {} takes exactly one argument", name_); + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly one argument", name_); auto type_id = argument_types[0]->getTypeId(); if (type_id != TypeIndex::UInt8 && type_id != TypeIndex::UInt16 && type_id != TypeIndex::UInt32 && type_id != TypeIndex::UInt64) @@ -2080,6 +2128,29 @@ struct WindowFunctionNtile final : public StatefulWindowFunction bool allocatesMemoryInArena() const override { return false; } + bool checkWindowFrameType(const WindowTransform * transform) const override + { + if (transform->order_by_indices.empty()) + { + LOG_ERROR(getLogger("WindowFunctionNtile"), "Window frame for 'ntile' function must have ORDER BY clause"); + return false; + } + + // We must wait all for the partition end and get the total rows number in this + // partition. So before the end of this partition, there is no any block could be + // dropped out. + bool is_frame_supported = transform->window_description.frame.begin_type == WindowFrame::BoundaryType::Unbounded + && transform->window_description.frame.end_type == WindowFrame::BoundaryType::Unbounded; + if (!is_frame_supported) + { + LOG_ERROR( + getLogger("WindowFunctionNtile"), + "Window frame for function 'ntile' should be 'ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING'"); + return false; + } + return true; + } + std::optional getDefaultFrame() const override { WindowFrame frame; @@ -2106,7 +2177,6 @@ namespace { if (!buckets) [[unlikely]] { - checkWindowFrameType(transform); const auto & current_block = transform->blockAt(transform->current_row); const auto & workspace = transform->workspaces[function_index]; const auto & arg_col = *current_block.original_input_columns[workspace.argument_column_indices[0]]; @@ -2124,11 +2194,11 @@ namespace if (!buckets) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Argument of 'ntile' funtcion must be greater than zero"); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Argument of 'ntile' function must be greater than zero"); } } // new partition - if (transform->current_row_number == 1) [[unlikely]] + if (WindowFunctionHelpers::checkPartitionEnterFirstRow(transform)) [[unlikely]] { current_partition_rows = 0; current_partition_inserted_row = 0; @@ -2137,25 +2207,9 @@ namespace current_partition_rows++; // Only do the action when we meet the last row in this partition. - if (!transform->partition_ended) + if (!WindowFunctionHelpers::checkPartitionEnterLastRow(transform)) return; - else - { - auto current_row = transform->current_row; - current_row.row++; - const auto & end_row = transform->partition_end; - if (current_row != end_row) - { - if (current_row.row < transform->blockRowsNumber(current_row)) - return; - if (end_row.block != current_row.block + 1 || end_row.row) - { - return; - } - // else, current_row is the last input row. - } - } auto bucket_capacity = current_partition_rows / buckets; auto capacity_diff = current_partition_rows - bucket_capacity * buckets; @@ -2193,23 +2247,115 @@ namespace bucket_num += 1; } } +} - void NtileState::checkWindowFrameType(const WindowTransform * transform) +namespace +{ +struct PercentRankState +{ + RowNumber start_row; + UInt64 current_partition_rows = 0; +}; +} + +struct WindowFunctionPercentRank final : public StatefulWindowFunction +{ +public: + WindowFunctionPercentRank(const std::string & name_, + const DataTypes & argument_types_, const Array & parameters_) + : StatefulWindowFunction(name_, argument_types_, parameters_, std::make_shared()) + {} + + bool allocatesMemoryInArena() const override { return false; } + + bool checkWindowFrameType(const WindowTransform * transform) const override { - if (transform->order_by_indices.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Window frame for 'ntile' function must have ORDER BY clause"); + if (transform->window_description.frame.type != WindowFrame::FrameType::RANGE + || transform->window_description.frame.begin_type != WindowFrame::BoundaryType::Unbounded + || transform->window_description.frame.end_type != WindowFrame::BoundaryType::Current) + { + LOG_ERROR( + getLogger("WindowFunctionPercentRank"), + "Window frame for function 'percent_rank' should be 'RANGE BETWEEN UNBOUNDED PRECEDING AND CURRENT'"); + return false; + } + return true; + } - // We must wait all for the partition end and get the total rows number in this - // partition. So before the end of this partition, there is no any block could be - // dropped out. - bool is_frame_supported = transform->window_description.frame.begin_type == WindowFrame::BoundaryType::Unbounded - && transform->window_description.frame.end_type == WindowFrame::BoundaryType::Unbounded; - if (!is_frame_supported) + std::optional getDefaultFrame() const override + { + WindowFrame frame; + frame.type = WindowFrame::FrameType::RANGE; + frame.begin_type = WindowFrame::BoundaryType::Unbounded; + frame.end_type = WindowFrame::BoundaryType::Current; + return frame; + } + + void windowInsertResultInto(const WindowTransform * transform, size_t function_index) const override + { + auto & state = getWorkspaceState(transform, function_index); + if (WindowFunctionHelpers::checkPartitionEnterFirstRow(transform)) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Window frame for function 'ntile' should be 'ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING'"); + state.current_partition_rows = 0; + state.start_row = transform->current_row; + } + + insertRankIntoColumn(transform, function_index); + state.current_partition_rows++; + + if (!WindowFunctionHelpers::checkPartitionEnterLastRow(transform)) + { + return; + } + + UInt64 remaining_rows = state.current_partition_rows; + Float64 percent_rank_denominator = remaining_rows == 1 ? 1 : remaining_rows - 1; + + while (remaining_rows > 0) + { + auto block_rows_number = transform->blockRowsNumber(state.start_row); + auto available_block_rows = block_rows_number - state.start_row.row; + if (available_block_rows <= remaining_rows) + { + /// This partition involves multiple blocks. Finish current block and move on to the + /// next block. + auto & to_column = *transform->blockAt(state.start_row).output_columns[function_index]; + auto & data = assert_cast(to_column).getData(); + for (size_t i = state.start_row.row; i < block_rows_number; ++i) + data[i] = (data[i] - 1) / percent_rank_denominator; + + state.start_row.block++; + state.start_row.row = 0; + remaining_rows -= available_block_rows; + } + else + { + /// The partition ends in current block.s + auto & to_column = *transform->blockAt(state.start_row).output_columns[function_index]; + auto & data = assert_cast(to_column).getData(); + for (size_t i = state.start_row.row, n = state.start_row.row + remaining_rows; i < n; ++i) + { + data[i] = (data[i] - 1) / percent_rank_denominator; + } + state.start_row.row += remaining_rows; + remaining_rows = 0; + } } } -} + + + inline PercentRankState & getWorkspaceState(const WindowTransform * transform, size_t function_index) const + { + const auto & workspace = transform->workspaces[function_index]; + return getState(workspace); + } + + inline void insertRankIntoColumn(const WindowTransform * transform, size_t function_index) const + { + auto & to_column = *transform->blockAt(transform->current_row).output_columns[function_index]; + assert_cast(to_column).getData().push_back(static_cast(transform->peer_group_start_row_number)); + } +}; // ClickHouse-specific variant of lag/lead that respects the window frame. template @@ -2261,7 +2407,7 @@ struct WindowFunctionLagLeadInFrame final : public WindowFunction if (argument_types.size() > 3) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::TOO_MANY_ARGUMENTS_FOR_FUNCTION, "Function '{}' accepts at most 3 arguments, {} given", name, argument_types.size()); } @@ -2271,7 +2417,7 @@ struct WindowFunctionLagLeadInFrame final : public WindowFunction { if (argument_types_.empty()) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::TOO_FEW_ARGUMENTS_FOR_FUNCTION, "Function {} takes at least one argument", name_); } @@ -2361,7 +2507,7 @@ struct WindowFunctionNthValue final : public WindowFunction { if (argument_types_.size() != 2) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes exactly two arguments", name_); } @@ -2435,7 +2581,7 @@ struct NonNegativeDerivativeParams if (argument_types.size() != 2 && argument_types.size() != 3) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Function {} takes 2 or 3 arguments", name_); } @@ -2573,35 +2719,42 @@ void registerWindowFunctions(AggregateFunctionFactory & factory) { return std::make_shared(name, argument_types, parameters); - }, properties}, AggregateFunctionFactory::CaseInsensitive); + }, properties}, AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("dense_rank", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) { return std::make_shared(name, argument_types, parameters); - }, properties}, AggregateFunctionFactory::CaseInsensitive); + }, properties}, AggregateFunctionFactory::Case::Insensitive); + + factory.registerFunction("percent_rank", {[](const std::string & name, + const DataTypes & argument_types, const Array & parameters, const Settings *) + { + return std::make_shared(name, argument_types, + parameters); + }, properties}, AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("row_number", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) { return std::make_shared(name, argument_types, parameters); - }, properties}, AggregateFunctionFactory::CaseInsensitive); + }, properties}, AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("ntile", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) { return std::make_shared(name, argument_types, parameters); - }, properties}, AggregateFunctionFactory::CaseInsensitive); + }, properties}, AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("nth_value", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) { return std::make_shared( name, argument_types, parameters); - }, properties}, AggregateFunctionFactory::CaseInsensitive); + }, properties}, AggregateFunctionFactory::Case::Insensitive); factory.registerFunction("lagInFrame", {[](const std::string & name, const DataTypes & argument_types, const Array & parameters, const Settings *) diff --git a/src/Processors/Transforms/buildPushingToViewsChain.cpp b/src/Processors/Transforms/buildPushingToViewsChain.cpp index 25fbf13b0e7..98d66ed77c3 100644 --- a/src/Processors/Transforms/buildPushingToViewsChain.cpp +++ b/src/Processors/Transforms/buildPushingToViewsChain.cpp @@ -5,7 +5,9 @@ #include #include #include +#include #include +#include #include #include #include @@ -16,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -24,9 +27,13 @@ #include #include #include +#include +#include +#include #include #include +#include namespace ProfileEvents @@ -105,7 +112,7 @@ private: class ExecutingInnerQueryFromViewTransform final : public ExceptionKeepingTransform { public: - ExecutingInnerQueryFromViewTransform(const Block & header, ViewRuntimeData & view_, ViewsDataPtr views_data_); + ExecutingInnerQueryFromViewTransform(const Block & header, ViewRuntimeData & view_, ViewsDataPtr views_data_, bool disable_deduplication_for_children_); String getName() const override { return "ExecutingInnerQueryFromView"; } @@ -116,6 +123,7 @@ protected: private: ViewsDataPtr views_data; ViewRuntimeData & view; + bool disable_deduplication_for_children; struct State { @@ -138,7 +146,7 @@ class PushingToLiveViewSink final : public SinkToStorage public: PushingToLiveViewSink(const Block & header, StorageLiveView & live_view_, StoragePtr storage_holder_, ContextPtr context_); String getName() const override { return "PushingToLiveViewSink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; private: StorageLiveView & live_view; @@ -152,7 +160,7 @@ class PushingToWindowViewSink final : public SinkToStorage public: PushingToWindowViewSink(const Block & header, StorageWindowView & window_view_, StoragePtr storage_holder_, ContextPtr context_); String getName() const override { return "PushingToWindowViewSink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; private: StorageWindowView & window_view; @@ -216,45 +224,10 @@ std::optional generateViewChain( const auto & insert_settings = insert_context->getSettingsRef(); - // Do not deduplicate insertions into MV if the main insertion is Ok if (disable_deduplication_for_children) { insert_context->setSetting("insert_deduplicate", Field{false}); } - else if (insert_settings.update_insert_deduplication_token_in_dependent_materialized_views && - !insert_settings.insert_deduplication_token.value.empty()) - { - /** Update deduplication token passed to dependent MV with current view id. So it is possible to properly handle - * deduplication in complex INSERT flows. - * - * Example: - * - * landing -┬--> mv_1_1 ---> ds_1_1 ---> mv_2_1 --┬-> ds_2_1 ---> mv_3_1 ---> ds_3_1 - * | | - * └--> mv_1_2 ---> ds_1_2 ---> mv_2_2 --┘ - * - * Here we want to avoid deduplication for two different blocks generated from `mv_2_1` and `mv_2_2` that will - * be inserted into `ds_2_1`. - * - * We are forced to use view id instead of table id because there are some possible INSERT flows where no tables - * are involved. - * - * Example: - * - * landing -┬--> mv_1_1 --┬-> ds_1_1 - * | | - * └--> mv_1_2 --┘ - * - */ - auto insert_deduplication_token = insert_settings.insert_deduplication_token.value; - - if (view_id.hasUUID()) - insert_deduplication_token += "_" + toString(view_id.uuid); - else - insert_deduplication_token += "_" + view_id.getFullNameNotQuoted(); - - insert_context->setSetting("insert_deduplication_token", insert_deduplication_token); - } // Processing of blocks for MVs is done block by block, and there will // be no parallel reading after (plus it is not a costless operation) @@ -361,7 +334,13 @@ std::optional generateViewChain( insert_columns.emplace_back(column.name); } - InterpreterInsertQuery interpreter(nullptr, insert_context, false, false, false); + InterpreterInsertQuery interpreter( + nullptr, + insert_context, + /* allow_materialized */ false, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); /// TODO: remove sql_security_type check after we turn `ignore_empty_sql_security_in_create_view_query=false` bool check_access = !materialized_view->hasInnerTable() && materialized_view->getInMemoryMetadataPtr()->sql_security_type; @@ -378,6 +357,10 @@ std::optional generateViewChain( table_prefers_large_blocks ? settings.min_insert_block_size_bytes : 0ULL)); } +#ifdef ABORT_ON_LOGICAL_ERROR + out.addSource(std::make_shared("Before squashing", out.getInputHeader())); +#endif + auto counting = std::make_shared(out.getInputHeader(), current_thread, insert_context->getQuota()); counting->setProcessListElement(insert_context->getProcessListElement()); counting->setProgressCallback(insert_context->getProgressCallback()); @@ -420,11 +403,19 @@ std::optional generateViewChain( if (type == QueryViewsLogElement::ViewType::MATERIALIZED) { +#ifdef ABORT_ON_LOGICAL_ERROR + out.addSource(std::make_shared("Right after Inner query", out.getInputHeader())); +#endif + auto executing_inner_query = std::make_shared( - storage_header, views_data->views.back(), views_data); + storage_header, views_data->views.back(), views_data, disable_deduplication_for_children); executing_inner_query->setRuntimeData(view_thread_status, view_counter_ms); out.addSource(std::move(executing_inner_query)); + +#ifdef ABORT_ON_LOGICAL_ERROR + out.addSource(std::make_shared("Right before Inner query", out.getInputHeader())); +#endif } return out; @@ -465,11 +456,7 @@ Chain buildPushingToViewsChain( */ result_chain.addTableLock(storage->lockForShare(context->getInitialQueryId(), context->getSettingsRef().lock_acquire_timeout)); - /// If the "root" table deduplicates blocks, there are no need to make deduplication for children - /// Moreover, deduplication for AggregatingMergeTree children could produce false positives due to low size of inserting blocks - bool disable_deduplication_for_children = false; - if (!context->getSettingsRef().deduplicate_blocks_in_dependent_materialized_views) - disable_deduplication_for_children = !no_destination && storage->supportsDeduplication(); + bool disable_deduplication_for_children = !context->getSettingsRef().deduplicate_blocks_in_dependent_materialized_views; auto table_id = storage->getStorageID(); auto views = DatabaseCatalog::instance().getDependentViews(table_id); @@ -560,12 +547,25 @@ Chain buildPushingToViewsChain( auto sink = std::make_shared(live_view_header, *live_view, storage, context); sink->setRuntimeData(thread_status, elapsed_counter_ms); result_chain.addSource(std::move(sink)); + + result_chain.addSource(std::make_shared(result_chain.getInputHeader())); } else if (auto * window_view = dynamic_cast(storage.get())) { auto sink = std::make_shared(window_view->getInputHeader(), *window_view, storage, context); sink->setRuntimeData(thread_status, elapsed_counter_ms); result_chain.addSource(std::move(sink)); + + result_chain.addSource(std::make_shared(result_chain.getInputHeader())); + } + else if (dynamic_cast(storage.get())) + { + auto sink = storage->write(query_ptr, metadata_snapshot, context, async_insert); + metadata_snapshot->check(sink->getHeader().getColumnsWithTypeAndName()); + sink->setRuntimeData(thread_status, elapsed_counter_ms); + result_chain.addSource(std::move(sink)); + + result_chain.addSource(std::make_shared(result_chain.getInputHeader())); } /// Do not push to destination table if the flag is set else if (!no_destination) @@ -573,8 +573,15 @@ Chain buildPushingToViewsChain( auto sink = storage->write(query_ptr, metadata_snapshot, context, async_insert); metadata_snapshot->check(sink->getHeader().getColumnsWithTypeAndName()); sink->setRuntimeData(thread_status, elapsed_counter_ms); + + result_chain.addSource(std::make_shared(sink->getHeader())); + result_chain.addSource(std::move(sink)); } + else + { + result_chain.addSource(std::make_shared(storage_header)); + } if (result_chain.empty()) result_chain.addSink(std::make_shared(storage_header)); @@ -590,7 +597,7 @@ Chain buildPushingToViewsChain( return result_chain; } -static QueryPipeline process(Block block, ViewRuntimeData & view, const ViewsData & views_data) +static QueryPipeline process(Block block, ViewRuntimeData & view, const ViewsData & views_data, Chunk::ChunkInfoCollection && chunk_infos, bool disable_deduplication_for_children) { const auto & context = view.context; @@ -637,6 +644,19 @@ static QueryPipeline process(Block block, ViewRuntimeData & view, const ViewsDat pipeline.getHeader(), std::make_shared(std::move(converting)))); + pipeline.addTransform(std::make_shared(std::move(chunk_infos), pipeline.getHeader())); + + if (!disable_deduplication_for_children) + { + String materialize_view_id = view.table_id.hasUUID() ? toString(view.table_id.uuid) : view.table_id.getFullNameNotQuoted(); + pipeline.addTransform(std::make_shared(std::move(materialize_view_id), pipeline.getHeader())); + pipeline.addTransform(std::make_shared(pipeline.getHeader())); + } + else + { + pipeline.addTransform(std::make_shared(pipeline.getHeader())); + } + return QueryPipelineBuilder::getPipeline(std::move(pipeline)); } @@ -728,17 +748,19 @@ IProcessor::Status CopyingDataToViewsTransform::prepare() ExecutingInnerQueryFromViewTransform::ExecutingInnerQueryFromViewTransform( const Block & header, ViewRuntimeData & view_, - std::shared_ptr views_data_) + std::shared_ptr views_data_, + bool disable_deduplication_for_children_) : ExceptionKeepingTransform(header, view_.sample_block) , views_data(std::move(views_data_)) , view(view_) + , disable_deduplication_for_children(disable_deduplication_for_children_) { } void ExecutingInnerQueryFromViewTransform::onConsume(Chunk chunk) { - auto block = getInputPort().getHeader().cloneWithColumns(chunk.getColumns()); - state.emplace(process(block, view, *views_data)); + auto block = getInputPort().getHeader().cloneWithColumns(chunk.detachColumns()); + state.emplace(process(std::move(block), view, *views_data, std::move(chunk.getChunkInfos()), disable_deduplication_for_children)); } @@ -770,10 +792,10 @@ PushingToLiveViewSink::PushingToLiveViewSink(const Block & header, StorageLiveVi { } -void PushingToLiveViewSink::consume(Chunk chunk) +void PushingToLiveViewSink::consume(Chunk & chunk) { Progress local_progress(chunk.getNumRows(), chunk.bytes(), 0); - live_view.writeBlock(getHeader().cloneWithColumns(chunk.detachColumns()), context); + live_view.writeBlock(live_view, getHeader().cloneWithColumns(chunk.getColumns()), std::move(chunk.getChunkInfos()), context); if (auto process = context->getProcessListElement()) process->updateProgressIn(local_progress); @@ -793,11 +815,11 @@ PushingToWindowViewSink::PushingToWindowViewSink( { } -void PushingToWindowViewSink::consume(Chunk chunk) +void PushingToWindowViewSink::consume(Chunk & chunk) { Progress local_progress(chunk.getNumRows(), chunk.bytes(), 0); StorageWindowView::writeIntoWindowView( - window_view, getHeader().cloneWithColumns(chunk.detachColumns()), context); + window_view, getHeader().cloneWithColumns(chunk.getColumns()), std::move(chunk.getChunkInfos()), context); if (auto process = context->getProcessListElement()) process->updateProgressIn(local_progress); diff --git a/src/Processors/Transforms/getSourceFromASTInsertQuery.cpp b/src/Processors/Transforms/getSourceFromASTInsertQuery.cpp index a2a42f27c3f..c33e87815b1 100644 --- a/src/Processors/Transforms/getSourceFromASTInsertQuery.cpp +++ b/src/Processors/Transforms/getSourceFromASTInsertQuery.cpp @@ -14,7 +14,8 @@ #include #include #include "IO/CompressionMethod.h" -#include "Parsers/ASTLiteral.h" +#include +#include namespace DB diff --git a/src/Processors/tests/gtest_exception_on_incorrect_pipeline.cpp b/src/Processors/tests/gtest_exception_on_incorrect_pipeline.cpp index ce5992c2548..364d7c69071 100644 --- a/src/Processors/tests/gtest_exception_on_incorrect_pipeline.cpp +++ b/src/Processors/tests/gtest_exception_on_incorrect_pipeline.cpp @@ -50,7 +50,7 @@ TEST(Processors, PortsNotConnected) processors->emplace_back(std::move(source)); processors->emplace_back(std::move(sink)); -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD try { QueryStatusPtr element; diff --git a/src/Processors/tests/gtest_full_sorting_join.cpp b/src/Processors/tests/gtest_full_sorting_join.cpp new file mode 100644 index 00000000000..f678d7984e8 --- /dev/null +++ b/src/Processors/tests/gtest_full_sorting_join.cpp @@ -0,0 +1,768 @@ +#include + +#include +#include +#include +#include +#include + +#include +#include +#include + +#include + +#include + +#include +#include +#include +#include +#include +#include + +#include +#include + + +#include + +using namespace DB; + +namespace +{ + +QueryPipeline buildJoinPipeline( + std::shared_ptr left_source, + std::shared_ptr right_source, + size_t key_length = 1, + JoinKind kind = JoinKind::Inner, + JoinStrictness strictness = JoinStrictness::All, + ASOFJoinInequality asof_inequality = ASOFJoinInequality::None) +{ + Blocks inputs; + inputs.emplace_back(left_source->getPort().getHeader()); + inputs.emplace_back(right_source->getPort().getHeader()); + + Block out_header; + for (const auto & input : inputs) + { + for (ColumnWithTypeAndName column : input) + { + if (&input == &inputs.front()) + column.name = "t1." + column.name; + else + column.name = "t2." + column.name; + out_header.insert(column); + } + } + + TableJoin::JoinOnClause on_clause; + for (size_t i = 0; i < key_length; ++i) + { + on_clause.key_names_left.emplace_back(inputs[0].getByPosition(i).name); + on_clause.key_names_right.emplace_back(inputs[1].getByPosition(i).name); + } + + auto joining = std::make_shared( + kind, + strictness, + on_clause, + inputs, out_header, /* max_block_size = */ 0); + + if (asof_inequality != ASOFJoinInequality::None) + joining->setAsofInequality(asof_inequality); + + chassert(joining->getInputs().size() == 2); + + connect(left_source->getPort(), joining->getInputs().front()); + connect(right_source->getPort(), joining->getInputs().back()); + + auto * output_port = &joining->getOutputPort(); + + auto processors = std::make_shared(); + processors->emplace_back(std::move(left_source)); + processors->emplace_back(std::move(right_source)); + processors->emplace_back(std::move(joining)); + + QueryPipeline pipeline(QueryPlanResourceHolder{}, processors, output_port); + return pipeline; +} + + +std::shared_ptr oneColumnSource(const std::vector> & values) +{ + Block header = { + ColumnWithTypeAndName(std::make_shared(), "key"), + ColumnWithTypeAndName(std::make_shared(), "idx"), + }; + + UInt64 idx = 0; + Chunks chunks; + for (const auto & chunk_values : values) + { + auto key_column = ColumnUInt64::create(); + auto idx_column = ColumnUInt64::create(); + + for (auto n : chunk_values) + { + key_column->insertValue(n); + idx_column->insertValue(idx); + ++idx; + } + chunks.emplace_back(Chunk(Columns{std::move(key_column), std::move(idx_column)}, chunk_values.size())); + } + return std::make_shared(header, std::move(chunks)); +} + +class SourceChunksBuilder +{ +public: + + explicit SourceChunksBuilder(const Block & header_) + : header(header_) + { + current_chunk = header.cloneEmptyColumns(); + chassert(!current_chunk.empty()); + } + + void setBreakProbability(pcg64 & rng_) + { + /// random probability with possibility to have exact 0.0 and 1.0 values + break_prob = std::uniform_int_distribution(0, 5)(rng_) / static_cast(5); + rng = &rng_; + } + + void addRow(const std::vector & row) + { + chassert(row.size() == current_chunk.size()); + for (size_t i = 0; i < current_chunk.size(); ++i) + current_chunk[i]->insert(row[i]); + + if (rng && std::uniform_real_distribution<>(0.0, 1.0)(*rng) < break_prob) + addChunk(); + } + + void addChunk() + { + if (current_chunk.front()->empty()) + return; + + size_t rows = current_chunk.front()->size(); + chunks.emplace_back(std::move(current_chunk), rows); + current_chunk = header.cloneEmptyColumns(); + } + + std::shared_ptr getSource() + { + addChunk(); + + /// copy chunk to allow reusing same builder + Chunks chunks_copy; + chunks_copy.reserve(chunks.size()); + for (const auto & chunk : chunks) + chunks_copy.emplace_back(chunk.clone()); + return std::make_shared(header, std::move(chunks_copy)); + } + +private: + Block header; + Chunks chunks; + MutableColumns current_chunk; + + pcg64 * rng = nullptr; + double break_prob = 0.0; +}; + + +std::vector> getValuesFromBlock(const Block & block, const Names & names) +{ + std::vector> result; + for (size_t i = 0; i < block.rows(); ++i) + { + auto & row = result.emplace_back(); + for (const auto & name : names) + block.getByName(name).column->get(i, row.emplace_back()); + } + return result; +} + + +Block executePipeline(QueryPipeline && pipeline) +{ + PullingPipelineExecutor executor(pipeline); + + Blocks result_blocks; + while (true) + { + Block block; + bool is_ok = executor.pull(block); + if (!is_ok) + break; + result_blocks.emplace_back(std::move(block)); + } + + return concatenateBlocks(result_blocks); +} + +template +void assertColumnVectorEq(const typename ColumnVector::Container & expected, const Block & block, const std::string & name) +{ + const auto * actual = typeid_cast *>(block.getByName(name).column.get()); + ASSERT_TRUE(actual) << "unexpected column type: " << block.getByName(name).column->dumpStructure() << "expected: " << typeid(ColumnVector).name(); + + auto get_first_diff = [&]() -> String + { + const auto & actual_data = actual->getData(); + size_t num_rows = std::min(expected.size(), actual_data.size()); + for (size_t i = 0; i < num_rows; ++i) + { + if (expected[i] != actual_data[i]) + return fmt::format(", expected: {}, actual: {} at row {}", expected[i], actual_data[i], i); + } + return ""; + }; + + EXPECT_EQ(actual->getData().size(), expected.size()); + ASSERT_EQ(actual->getData(), expected) << "column name: " << name << get_first_diff(); +} + +template +void assertColumnEq(const IColumn & expected, const Block & block, const std::string & name) +{ + const ColumnPtr & actual = block.getByName(name).column; + ASSERT_TRUE(checkColumn(*actual)); + ASSERT_TRUE(checkColumn(expected)); + EXPECT_EQ(actual->size(), expected.size()); + + auto dump_val = [](const IColumn & col, size_t i) -> String + { + Field value; + col.get(i, value); + return value.dump(); + }; + + size_t num_rows = std::min(actual->size(), expected.size()); + for (size_t i = 0; i < num_rows; ++i) + ASSERT_EQ(actual->compareAt(i, i, expected, 1), 0) << dump_val(*actual, i) << " != " << dump_val(expected, i) << " at row " << i; +} + +template +T getRandomFrom(pcg64 & rng, const std::initializer_list & opts) +{ + std::vector options(opts.begin(), opts.end()); + size_t idx = std::uniform_int_distribution(0, options.size() - 1)(rng); + return options[idx]; +} + +void generateNextKey(pcg64 & rng, UInt64 & k1, String & k2) +{ + size_t str_len = std::uniform_int_distribution<>(1, 10)(rng); + String new_k2 = getRandomASCIIString(str_len, rng); + if (new_k2.compare(k2) <= 0) + ++k1; + k2 = new_k2; +} + +bool isStrict(ASOFJoinInequality inequality) +{ + return inequality == ASOFJoinInequality::Less || inequality == ASOFJoinInequality::Greater; +} + +} + +class FullSortingJoinTest : public ::testing::Test +{ +public: + FullSortingJoinTest() = default; + + void SetUp() override + { + Poco::AutoPtr channel(new Poco::ConsoleChannel(std::cerr)); + Poco::Logger::root().setChannel(channel); + if (const char * test_log_level = std::getenv("TEST_LOG_LEVEL")) // NOLINT(concurrency-mt-unsafe) + Poco::Logger::root().setLevel(test_log_level); + else + Poco::Logger::root().setLevel("none"); + + + UInt64 seed = randomSeed(); + if (const char * random_seed = std::getenv("TEST_RANDOM_SEED")) // NOLINT(concurrency-mt-unsafe) + seed = std::stoull(random_seed); + std::cout << "TEST_RANDOM_SEED=" << seed << std::endl; + rng = pcg64(seed); + } + + void TearDown() override + { + } + + pcg64 rng; +}; + +TEST_F(FullSortingJoinTest, AllAnyOneKey) +try +{ + { + SCOPED_TRACE("Inner All"); + Block result = executePipeline(buildJoinPipeline( + oneColumnSource({ {1, 2, 3, 4, 5} }), + oneColumnSource({ {1}, {2}, {3}, {4}, {5} }), + 1, JoinKind::Inner, JoinStrictness::All)); + + assertColumnVectorEq(ColumnUInt64::Container({0, 1, 2, 3, 4}), result, "t1.idx"); + assertColumnVectorEq(ColumnUInt64::Container({0, 1, 2, 3, 4}), result, "t2.idx"); + } + { + SCOPED_TRACE("Inner Any"); + Block result = executePipeline(buildJoinPipeline( + oneColumnSource({ {1, 2, 3, 4, 5} }), + oneColumnSource({ {1}, {2}, {3}, {4}, {5} }), + 1, JoinKind::Inner, JoinStrictness::Any)); + assertColumnVectorEq(ColumnUInt64::Container({0, 1, 2, 3, 4}), result, "t1.idx"); + assertColumnVectorEq(ColumnUInt64::Container({0, 1, 2, 3, 4}), result, "t2.idx"); + } + { + SCOPED_TRACE("Inner All"); + Block result = executePipeline(buildJoinPipeline( + oneColumnSource({ {2, 2, 2}, {2, 3}, {3, 5} }), + oneColumnSource({ {1, 1, 1}, {2, 2}, {3, 4} }), + 1, JoinKind::Inner, JoinStrictness::All)); + assertColumnVectorEq(ColumnUInt64::Container({0, 1, 2, 0, 1, 2, 3, 3, 4, 5}), result, "t1.idx"); + assertColumnVectorEq(ColumnUInt64::Container({3, 3, 3, 4, 4, 4, 3, 4, 5, 5}), result, "t2.idx"); + } + { + SCOPED_TRACE("Inner Any"); + Block result = executePipeline(buildJoinPipeline( + oneColumnSource({ {2, 2, 2}, {2, 3}, {3, 5} }), + oneColumnSource({ {1, 1, 1}, {2, 2}, {3, 4} }), + 1, JoinKind::Inner, JoinStrictness::Any)); + assertColumnVectorEq(ColumnUInt64::Container({0, 4}), result, "t1.idx"); + assertColumnVectorEq(ColumnUInt64::Container({3, 5}), result, "t2.idx"); + } + { + SCOPED_TRACE("Inner Any"); + Block result = executePipeline(buildJoinPipeline( + oneColumnSource({ {2, 2, 2, 2}, {3}, {3, 5} }), + oneColumnSource({ {1, 1, 1, 2}, {2}, {3, 4} }), + 1, JoinKind::Inner, JoinStrictness::Any)); + assertColumnVectorEq(ColumnUInt64::Container({0, 4}), result, "t1.idx"); + assertColumnVectorEq(ColumnUInt64::Container({3, 5}), result, "t2.idx"); + } + { + + SCOPED_TRACE("Left Any"); + Block result = executePipeline(buildJoinPipeline( + oneColumnSource({ {2, 2, 2}, {2, 3}, {3, 5} }), + oneColumnSource({ {1, 1, 1}, {2, 2}, {3, 4} }), + 1, JoinKind::Left, JoinStrictness::Any)); + assertColumnVectorEq(ColumnUInt64::Container({0, 1, 2, 3, 4, 5, 6}), result, "t1.idx"); + assertColumnVectorEq(ColumnUInt64::Container({3, 3, 3, 3, 5, 5, 0}), result, "t2.idx"); + } + { + SCOPED_TRACE("Left Any"); + Block result = executePipeline(buildJoinPipeline( + oneColumnSource({ {2, 2, 2, 2}, {3}, {3, 5} }), + oneColumnSource({ {1, 1, 1, 2}, {2}, {3, 4} }), + 1, JoinKind::Left, JoinStrictness::Any)); + assertColumnVectorEq(ColumnUInt64::Container({0, 1, 2, 3, 4, 5, 6}), result, "t1.idx"); + assertColumnVectorEq(ColumnUInt64::Container({3, 3, 3, 3, 5, 5, 0}), result, "t2.idx"); + } +} +catch (Exception & e) +{ + std::cout << e.getStackTraceString() << std::endl; + throw; +} + + +TEST_F(FullSortingJoinTest, AnySimple) +try +{ + JoinKind kind = getRandomFrom(rng, {JoinKind::Inner, JoinKind::Left, JoinKind::Right}); + + SourceChunksBuilder left_source({ + {std::make_shared(), "k1"}, + {std::make_shared(), "k2"}, + {std::make_shared(), "attr"}, + }); + + SourceChunksBuilder right_source({ + {std::make_shared(), "k1"}, + {std::make_shared(), "k2"}, + {std::make_shared(), "attr"}, + }); + + left_source.setBreakProbability(rng); + right_source.setBreakProbability(rng); + + size_t num_keys = std::uniform_int_distribution<>(100, 1000)(rng); + + auto expected_left = ColumnString::create(); + auto expected_right = ColumnString::create(); + + UInt64 k1 = 1; + String k2; + + auto get_attr = [&](const String & side, size_t idx) -> String + { + return toString(k1) + "_" + k2 + "_" + side + "_" + toString(idx); + }; + + for (size_t i = 0; i < num_keys; ++i) + { + generateNextKey(rng, k1, k2); + + /// Key is present in left, right or both tables. Both tables is more probable. + size_t key_presence = std::uniform_int_distribution<>(0, 10)(rng); + + size_t num_rows_left = key_presence == 0 ? 0 : std::uniform_int_distribution<>(1, 10)(rng); + for (size_t j = 0; j < num_rows_left; ++j) + left_source.addRow({k1, k2, get_attr("left", j)}); + + size_t num_rows_right = key_presence == 1 ? 0 : std::uniform_int_distribution<>(1, 10)(rng); + for (size_t j = 0; j < num_rows_right; ++j) + right_source.addRow({k1, k2, get_attr("right", j)}); + + String left_attr = num_rows_left ? get_attr("left", 0) : ""; + String right_attr = num_rows_right ? get_attr("right", 0) : ""; + + if (kind == JoinKind::Inner && num_rows_left && num_rows_right) + { + expected_left->insert(left_attr); + expected_right->insert(right_attr); + } + else if (kind == JoinKind::Left) + { + for (size_t j = 0; j < num_rows_left; ++j) + { + expected_left->insert(get_attr("left", j)); + expected_right->insert(right_attr); + } + } + else if (kind == JoinKind::Right) + { + for (size_t j = 0; j < num_rows_right; ++j) + { + expected_left->insert(left_attr); + expected_right->insert(get_attr("right", j)); + } + } + } + + Block result_block = executePipeline(buildJoinPipeline( + left_source.getSource(), right_source.getSource(), /* key_length = */ 2, + kind, JoinStrictness::Any)); + assertColumnEq(*expected_left, result_block, "t1.attr"); + assertColumnEq(*expected_right, result_block, "t2.attr"); +} +catch (Exception & e) +{ + std::cout << e.getStackTraceString() << std::endl; + throw; +} + +TEST_F(FullSortingJoinTest, AsofSimple) +try +{ + SourceChunksBuilder left_source({ + {std::make_shared(), "key"}, + {std::make_shared(), "t"}, + }); + left_source.addRow({"AMZN", 3}); + left_source.addRow({"AMZN", 4}); + left_source.addRow({"AMZN", 6}); + left_source.addRow({"SBUX", 10}); + + SourceChunksBuilder right_source({ + {std::make_shared(), "key"}, + {std::make_shared(), "t"}, + {std::make_shared(), "value"}, + }); + right_source.addRow({"AAPL", 1, 97}); + right_source.addChunk(); + right_source.addRow({"AAPL", 2, 98}); + right_source.addRow({"AAPL", 3, 99}); + right_source.addRow({"AMZN", 1, 100}); + right_source.addRow({"AMZN", 2, 110}); + right_source.addChunk(); + right_source.addRow({"AMZN", 2, 110}); + right_source.addChunk(); + right_source.addRow({"AMZN", 4, 130}); + right_source.addRow({"AMZN", 5, 140}); + right_source.addRow({"SBUX", 8, 180}); + right_source.addChunk(); + right_source.addRow({"SBUX", 9, 190}); + + { + Block result_block = executePipeline(buildJoinPipeline( + left_source.getSource(), right_source.getSource(), /* key_length = */ 2, + JoinKind::Inner, JoinStrictness::Asof, ASOFJoinInequality::LessOrEquals)); + auto values = getValuesFromBlock(result_block, {"t1.key", "t1.t", "t2.t", "t2.value"}); + + ASSERT_EQ(values, (std::vector>{ + {"AMZN", 3u, 4u, 130u}, + {"AMZN", 4u, 4u, 130u}, + })); + } + + { + Block result_block = executePipeline(buildJoinPipeline( + left_source.getSource(), right_source.getSource(), /* key_length = */ 2, + JoinKind::Inner, JoinStrictness::Asof, ASOFJoinInequality::GreaterOrEquals)); + auto values = getValuesFromBlock(result_block, {"t1.key", "t1.t", "t2.t", "t2.value"}); + + ASSERT_EQ(values, (std::vector>{ + {"AMZN", 3u, 2u, 110u}, + {"AMZN", 4u, 4u, 130u}, + {"AMZN", 6u, 5u, 140u}, + {"SBUX", 10u, 9u, 190u}, + })); + } +} +catch (Exception & e) +{ + std::cout << e.getStackTraceString() << std::endl; + throw; +} + + +TEST_F(FullSortingJoinTest, AsofOnlyColumn) +try +{ + auto left_source = oneColumnSource({ {3}, {3, 3, 3}, {3, 5, 5, 6}, {9, 9}, {10, 20} }); + + SourceChunksBuilder right_source_builder({ + {std::make_shared(), "t"}, + {std::make_shared(), "value"}, + }); + + right_source_builder.setBreakProbability(rng); + + for (const auto & row : std::vector>{ {1, 101}, {2, 102}, {4, 104}, {5, 105}, {11, 111}, {15, 115} }) + right_source_builder.addRow(row); + + auto right_source = right_source_builder.getSource(); + + auto pipeline = buildJoinPipeline( + left_source, right_source, /* key_length = */ 1, + JoinKind::Inner, JoinStrictness::Asof, ASOFJoinInequality::LessOrEquals); + + Block result_block = executePipeline(std::move(pipeline)); + + ASSERT_EQ( + assert_cast(result_block.getByName("t1.key").column.get())->getData(), + (ColumnUInt64::Container{3, 3, 3, 3, 3, 5, 5, 6, 9, 9, 10}) + ); + + ASSERT_EQ( + assert_cast(result_block.getByName("t2.t").column.get())->getData(), + (ColumnUInt64::Container{4, 4, 4, 4, 4, 5, 5, 11, 11, 11, 11}) + ); + + ASSERT_EQ( + assert_cast(result_block.getByName("t2.value").column.get())->getData(), + (ColumnUInt64::Container{104, 104, 104, 104, 104, 105, 105, 111, 111, 111, 111}) + ); +} +catch (Exception & e) +{ + std::cout << e.getStackTraceString() << std::endl; + throw; +} + +TEST_F(FullSortingJoinTest, AsofLessGeneratedTestData) +try +{ + /// Generate data random and build expected result at the same time. + + /// Test specific combinations of join kind and inequality per each run + auto join_kind = getRandomFrom(rng, { JoinKind::Inner, JoinKind::Left }); + auto asof_inequality = getRandomFrom(rng, { ASOFJoinInequality::Less, ASOFJoinInequality::LessOrEquals }); + + SCOPED_TRACE(fmt::format("{} {}", join_kind, asof_inequality)); + + /// Key is complex, `k1, k2` for equality and `t` for asof + SourceChunksBuilder left_source_builder({ + {std::make_shared(), "k1"}, + {std::make_shared(), "k2"}, + {std::make_shared(), "t"}, + {std::make_shared(), "attr"}, + }); + + SourceChunksBuilder right_source_builder({ + {std::make_shared(), "k1"}, + {std::make_shared(), "k2"}, + {std::make_shared(), "t"}, + {std::make_shared(), "attr"}, + }); + + /// How small generated block should be + left_source_builder.setBreakProbability(rng); + right_source_builder.setBreakProbability(rng); + + /// We are going to generate sorted data and remember expected result + ColumnInt64::Container expected; + + UInt64 k1 = 1; + String k2; + auto key_num_total = std::uniform_int_distribution<>(1, 1000)(rng); + for (size_t key_num = 0; key_num < key_num_total; ++key_num) + { + /// Generate new key greater than previous + generateNextKey(rng, k1, k2); + + Int64 left_t = 0; + /// Generate several rows for the key + size_t num_left_rows = std::uniform_int_distribution<>(1, 100)(rng); + for (size_t i = 0; i < num_left_rows; ++i) + { + /// t is strictly greater than previous + left_t += std::uniform_int_distribution<>(1, 10)(rng); + + auto left_arrtibute_value = 10 * left_t; + left_source_builder.addRow({k1, k2, left_t, left_arrtibute_value}); + expected.push_back(left_arrtibute_value); + + auto num_matches = 1 + std::poisson_distribution<>(4)(rng); + /// Generate several matches in the right table + auto right_t = left_t; + for (size_t j = 0; j < num_matches; ++j) + { + int min_step = isStrict(asof_inequality) ? 1 : 0; + right_t += std::uniform_int_distribution<>(min_step, 3)(rng); + + /// First row should match + bool is_match = j == 0; + right_source_builder.addRow({k1, k2, right_t, is_match ? 10 * left_arrtibute_value : -1}); + } + /// Next left_t should be greater than right_t not to match with previous rows + left_t = right_t; + } + + /// generate some rows with greater left_t to check that they are not matched + num_left_rows = std::bernoulli_distribution(0.5)(rng) ? std::uniform_int_distribution<>(1, 100)(rng) : 0; + for (size_t i = 0; i < num_left_rows; ++i) + { + left_t += std::uniform_int_distribution<>(1, 10)(rng); + left_source_builder.addRow({k1, k2, left_t, -10 * left_t}); + + if (join_kind == JoinKind::Left) + expected.push_back(-10 * left_t); + } + } + + Block result_block = executePipeline(buildJoinPipeline( + left_source_builder.getSource(), right_source_builder.getSource(), + /* key_length = */ 3, + join_kind, JoinStrictness::Asof, asof_inequality)); + + assertColumnVectorEq(expected, result_block, "t1.attr"); + + for (auto & e : expected) + /// Non matched rows from left table have negative attr + /// Value if attribute in right table is 10 times greater than in left table + e = e < 0 ? 0 : 10 * e; + + assertColumnVectorEq(expected, result_block, "t2.attr"); +} +catch (Exception & e) +{ + std::cout << e.getStackTraceString() << std::endl; + throw; +} + +TEST_F(FullSortingJoinTest, AsofGreaterGeneratedTestData) +try +{ + /// Generate data random and build expected result at the same time. + + /// Test specific combinations of join kind and inequality per each run + auto join_kind = getRandomFrom(rng, { JoinKind::Inner, JoinKind::Left }); + auto asof_inequality = getRandomFrom(rng, { ASOFJoinInequality::Greater, ASOFJoinInequality::GreaterOrEquals }); + + SCOPED_TRACE(fmt::format("{} {}", join_kind, asof_inequality)); + + SourceChunksBuilder left_source_builder({ + {std::make_shared(), "k1"}, + {std::make_shared(), "k2"}, + {std::make_shared(), "t"}, + {std::make_shared(), "attr"}, + }); + + SourceChunksBuilder right_source_builder({ + {std::make_shared(), "k1"}, + {std::make_shared(), "k2"}, + {std::make_shared(), "t"}, + {std::make_shared(), "attr"}, + }); + + left_source_builder.setBreakProbability(rng); + right_source_builder.setBreakProbability(rng); + + ColumnInt64::Container expected; + + UInt64 k1 = 1; + String k2; + UInt64 left_t = 0; + + auto key_num_total = std::uniform_int_distribution<>(1, 1000)(rng); + for (size_t key_num = 0; key_num < key_num_total; ++key_num) + { + /// Generate new key greater than previous + generateNextKey(rng, k1, k2); + + /// Generate some rows with smaller left_t to check that they are not matched + size_t num_left_rows = std::bernoulli_distribution(0.5)(rng) ? std::uniform_int_distribution<>(1, 100)(rng) : 0; + for (size_t i = 0; i < num_left_rows; ++i) + { + left_t += std::uniform_int_distribution<>(1, 10)(rng); + left_source_builder.addRow({k1, k2, left_t, -10 * left_t}); + + if (join_kind == JoinKind::Left) + expected.push_back(-10 * left_t); + } + + if (std::bernoulli_distribution(0.1)(rng)) + continue; + + size_t num_right_matches = std::uniform_int_distribution<>(1, 10)(rng); + auto right_t = left_t + std::uniform_int_distribution<>(isStrict(asof_inequality) ? 0 : 1, 10)(rng); + auto attribute_value = 10 * right_t; + for (size_t j = 0; j < num_right_matches; ++j) + { + right_t += std::uniform_int_distribution<>(0, 3)(rng); + bool is_match = j == num_right_matches - 1; + right_source_builder.addRow({k1, k2, right_t, is_match ? 10 * attribute_value : -1}); + } + + /// Next left_t should be greater than (or equals) right_t to match with previous rows + left_t = right_t + std::uniform_int_distribution<>(isStrict(asof_inequality) ? 1 : 0, 100)(rng); + size_t num_left_matches = std::uniform_int_distribution<>(1, 100)(rng); + for (size_t j = 0; j < num_left_matches; ++j) + { + left_t += std::uniform_int_distribution<>(0, 3)(rng); + left_source_builder.addRow({k1, k2, left_t, attribute_value}); + expected.push_back(attribute_value); + } + } + + Block result_block = executePipeline(buildJoinPipeline( + left_source_builder.getSource(), right_source_builder.getSource(), + /* key_length = */ 3, + join_kind, JoinStrictness::Asof, asof_inequality)); + + assertColumnVectorEq(expected, result_block, "t1.attr"); + + for (auto & e : expected) + /// Non matched rows from left table have negative attr + /// Value if attribute in right table is 10 times greater than in left table + e = e < 0 ? 0 : 10 * e; + + assertColumnVectorEq(expected, result_block, "t2.attr"); +} +catch (Exception & e) +{ + std::cout << e.getStackTraceString() << std::endl; + throw; +} diff --git a/src/QueryPipeline/QueryPipelineBuilder.h b/src/QueryPipeline/QueryPipelineBuilder.h index f0b2ead687e..a9e5b1535c0 100644 --- a/src/QueryPipeline/QueryPipelineBuilder.h +++ b/src/QueryPipeline/QueryPipelineBuilder.h @@ -193,7 +193,7 @@ public: return concurrency_control; } - void addResources(QueryPlanResourceHolder resources_) { resources = std::move(resources_); } + void addResources(QueryPlanResourceHolder resources_) { resources.append(std::move(resources_)); } void setQueryIdHolder(std::shared_ptr query_id_holder) { resources.query_id_holders.emplace_back(std::move(query_id_holder)); } void addContext(ContextPtr context) { resources.interpreter_context.emplace_back(std::move(context)); } diff --git a/src/QueryPipeline/QueryPlanResourceHolder.cpp b/src/QueryPipeline/QueryPlanResourceHolder.cpp index 2cd4dc42a83..bb2be2c8ffb 100644 --- a/src/QueryPipeline/QueryPlanResourceHolder.cpp +++ b/src/QueryPipeline/QueryPlanResourceHolder.cpp @@ -5,7 +5,7 @@ namespace DB { -QueryPlanResourceHolder & QueryPlanResourceHolder::operator=(QueryPlanResourceHolder && rhs) noexcept +QueryPlanResourceHolder & QueryPlanResourceHolder::append(QueryPlanResourceHolder && rhs) noexcept { table_locks.insert(table_locks.end(), rhs.table_locks.begin(), rhs.table_locks.end()); storage_holders.insert(storage_holders.end(), rhs.storage_holders.begin(), rhs.storage_holders.end()); @@ -16,6 +16,12 @@ QueryPlanResourceHolder & QueryPlanResourceHolder::operator=(QueryPlanResourceHo return *this; } +QueryPlanResourceHolder & QueryPlanResourceHolder::operator=(QueryPlanResourceHolder && rhs) noexcept +{ + append(std::move(rhs)); + return *this; +} + QueryPlanResourceHolder::QueryPlanResourceHolder() = default; QueryPlanResourceHolder::QueryPlanResourceHolder(QueryPlanResourceHolder &&) noexcept = default; QueryPlanResourceHolder::~QueryPlanResourceHolder() = default; diff --git a/src/QueryPipeline/QueryPlanResourceHolder.h b/src/QueryPipeline/QueryPlanResourceHolder.h index ed9eb68b7ba..10f7f39ab09 100644 --- a/src/QueryPipeline/QueryPlanResourceHolder.h +++ b/src/QueryPipeline/QueryPlanResourceHolder.h @@ -20,8 +20,11 @@ struct QueryPlanResourceHolder QueryPlanResourceHolder(QueryPlanResourceHolder &&) noexcept; ~QueryPlanResourceHolder(); + QueryPlanResourceHolder & operator=(QueryPlanResourceHolder &) = delete; + /// Custom move assignment does not destroy data from lhs. It appends data from rhs to lhs. QueryPlanResourceHolder & operator=(QueryPlanResourceHolder &&) noexcept; + QueryPlanResourceHolder & append(QueryPlanResourceHolder &&) noexcept; /// Some processors may implicitly use Context or temporary Storage created by Interpreter. /// But lifetime of Streams is not nested in lifetime of Interpreters, so we have to store it here, diff --git a/src/QueryPipeline/RemoteQueryExecutor.cpp b/src/QueryPipeline/RemoteQueryExecutor.cpp index 1686a101bde..b08f2002f64 100644 --- a/src/QueryPipeline/RemoteQueryExecutor.cpp +++ b/src/QueryPipeline/RemoteQueryExecutor.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -104,8 +105,16 @@ RemoteQueryExecutor::RemoteQueryExecutor( connection_entries.emplace_back(std::move(result.entry)); } + else + { + chassert(!fail_message.empty()); + if (result.entry.isNull()) + LOG_DEBUG(log, "Failed to connect to replica {}. {}", pool->getAddress(), fail_message); + else + LOG_DEBUG(log, "Replica is not usable for remote query execution: {}. {}", pool->getAddress(), fail_message); + } - auto res = std::make_unique(std::move(connection_entries), current_settings, throttler); + auto res = std::make_unique(std::move(connection_entries), context, throttler); if (extension_ && extension_->replica_info) res->setReplicaInfo(*extension_->replica_info); @@ -127,7 +136,7 @@ RemoteQueryExecutor::RemoteQueryExecutor( { create_connections = [this, &connection, throttler, extension_](AsyncCallback) { - auto res = std::make_unique(connection, context->getSettingsRef(), throttler); + auto res = std::make_unique(connection, context, throttler); if (extension_ && extension_->replica_info) res->setReplicaInfo(*extension_->replica_info); return res; @@ -148,7 +157,7 @@ RemoteQueryExecutor::RemoteQueryExecutor( { create_connections = [this, connection_ptr, throttler, extension_](AsyncCallback) { - auto res = std::make_unique(connection_ptr, context->getSettingsRef(), throttler); + auto res = std::make_unique(connection_ptr, context, throttler); if (extension_ && extension_->replica_info) res->setReplicaInfo(*extension_->replica_info); return res; @@ -169,7 +178,7 @@ RemoteQueryExecutor::RemoteQueryExecutor( { create_connections = [this, connections_, throttler, extension_](AsyncCallback) mutable { - auto res = std::make_unique(std::move(connections_), context->getSettingsRef(), throttler); + auto res = std::make_unique(std::move(connections_), context, throttler); if (extension_ && extension_->replica_info) res->setReplicaInfo(*extension_->replica_info); return res; @@ -234,7 +243,7 @@ RemoteQueryExecutor::RemoteQueryExecutor( timeouts, current_settings, pool_mode, std::move(async_callback), skip_unavailable_endpoints, priority_func); } - auto res = std::make_unique(std::move(connection_entries), current_settings, throttler); + auto res = std::make_unique(std::move(connection_entries), context, throttler); if (extension && extension->replica_info) res->setReplicaInfo(*extension->replica_info); return res; @@ -899,4 +908,10 @@ void RemoteQueryExecutor::setProfileInfoCallback(ProfileInfoCallback callback) std::lock_guard guard(was_cancelled_mutex); profile_info_callback = std::move(callback); } + +bool RemoteQueryExecutor::needToSkipUnavailableShard() const +{ + return context->getSettingsRef().skip_unavailable_shards && (0 == connections->size()); +} + } diff --git a/src/QueryPipeline/RemoteQueryExecutor.h b/src/QueryPipeline/RemoteQueryExecutor.h index 6b1539bd08e..04a59cc3b7e 100644 --- a/src/QueryPipeline/RemoteQueryExecutor.h +++ b/src/QueryPipeline/RemoteQueryExecutor.h @@ -218,7 +218,7 @@ public: IConnections & getConnections() { return *connections; } - bool needToSkipUnavailableShard() const { return context->getSettingsRef().skip_unavailable_shards && (0 == connections->size()); } + bool needToSkipUnavailableShard() const; bool isReplicaUnavailable() const { return extension && extension->parallel_reading_coordinator && connections->size() == 0; } diff --git a/src/QueryPipeline/SizeLimits.cpp b/src/QueryPipeline/SizeLimits.cpp index 76832b1f951..4161f3f365f 100644 --- a/src/QueryPipeline/SizeLimits.cpp +++ b/src/QueryPipeline/SizeLimits.cpp @@ -2,7 +2,6 @@ #include #include #include -#include namespace ProfileEvents diff --git a/src/QueryPipeline/tests/gtest_check_sorted_stream.cpp b/src/QueryPipeline/tests/gtest_check_sorted_stream.cpp index c8ab2e3a973..34bc2eb2b5e 100644 --- a/src/QueryPipeline/tests/gtest_check_sorted_stream.cpp +++ b/src/QueryPipeline/tests/gtest_check_sorted_stream.cpp @@ -133,7 +133,7 @@ TEST(CheckSortedTransform, CheckBadLastRow) EXPECT_NO_THROW(executor.pull(chunk)); EXPECT_NO_THROW(executor.pull(chunk)); -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD EXPECT_THROW(executor.pull(chunk), DB::Exception); #endif } @@ -158,7 +158,7 @@ TEST(CheckSortedTransform, CheckUnsortedBlock1) Chunk chunk; -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD EXPECT_THROW(executor.pull(chunk), DB::Exception); #endif } @@ -181,7 +181,7 @@ TEST(CheckSortedTransform, CheckUnsortedBlock2) PullingPipelineExecutor executor(pipeline); Chunk chunk; -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD EXPECT_THROW(executor.pull(chunk), DB::Exception); #endif } @@ -204,7 +204,7 @@ TEST(CheckSortedTransform, CheckUnsortedBlock3) PullingPipelineExecutor executor(pipeline); Chunk chunk; -#ifndef ABORT_ON_LOGICAL_ERROR +#ifndef DEBUG_OR_SANITIZER_BUILD EXPECT_THROW(executor.pull(chunk), DB::Exception); #endif } diff --git a/src/Server/CloudPlacementInfo.cpp b/src/Server/CloudPlacementInfo.cpp index 0790f825a45..d8810bb30de 100644 --- a/src/Server/CloudPlacementInfo.cpp +++ b/src/Server/CloudPlacementInfo.cpp @@ -11,6 +11,11 @@ namespace DB { +namespace ErrorCodes +{ +extern const int LOGICAL_ERROR; +} + namespace PlacementInfo { @@ -46,7 +51,15 @@ PlacementInfo & PlacementInfo::instance() } void PlacementInfo::initialize(const Poco::Util::AbstractConfiguration & config) +try { + if (!config.has(DB::PlacementInfo::PLACEMENT_CONFIG_PREFIX)) + { + availability_zone = ""; + initialized = true; + return; + } + use_imds = config.getBool(getConfigPath("use_imds"), false); if (use_imds) @@ -67,14 +80,17 @@ void PlacementInfo::initialize(const Poco::Util::AbstractConfiguration & config) LOG_DEBUG(log, "Loaded info: availability_zone: {}", availability_zone); initialized = true; } +catch (...) +{ + tryLogCurrentException("Failed to get availability zone"); + availability_zone = ""; + initialized = true; +} std::string PlacementInfo::getAvailabilityZone() const { if (!initialized) - { - LOG_WARNING(log, "Placement info has not been loaded"); - return ""; - } + throw Exception(ErrorCodes::LOGICAL_ERROR, "Placement info has not been loaded"); return availability_zone; } diff --git a/src/Server/GRPCServer.cpp b/src/Server/GRPCServer.cpp index 10b59751b22..d8a4d7f0e1f 100644 --- a/src/Server/GRPCServer.cpp +++ b/src/Server/GRPCServer.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -1082,8 +1083,8 @@ namespace read_buffer = wrapReadBufferWithCompressionMethod(std::move(read_buffer), input_compression_method); assert(!pipeline); - auto source = query_context->getInputFormat( - input_format, *read_buffer, header, query_context->getSettings().max_insert_block_size); + auto source + = query_context->getInputFormat(input_format, *read_buffer, header, query_context->getSettingsRef().max_insert_block_size); pipeline = std::make_unique(std::move(source)); pipeline_executor = std::make_unique(*pipeline); @@ -1151,8 +1152,7 @@ namespace external_table_context->applySettingsChanges(settings_changes); } auto in = external_table_context->getInputFormat( - format, *buf, metadata_snapshot->getSampleBlock(), - external_table_context->getSettings().max_insert_block_size); + format, *buf, metadata_snapshot->getSampleBlock(), external_table_context->getSettingsRef().max_insert_block_size); QueryPipelineBuilder cur_pipeline; cur_pipeline.init(Pipe(std::move(in))); @@ -1735,10 +1735,19 @@ namespace class GRPCServer::Runner { public: - explicit Runner(GRPCServer & owner_) : owner(owner_) {} + explicit Runner(GRPCServer & owner_) : owner(owner_), log(owner.log) {} ~Runner() { + try + { + stop(); + } + catch (...) + { + tryLogCurrentException(log, "~Runner"); + } + if (queue_thread.joinable()) queue_thread.join(); } @@ -1756,13 +1765,27 @@ public: } catch (...) { - tryLogCurrentException("GRPCServer"); + tryLogCurrentException(log, "run"); } }; queue_thread = ThreadFromGlobalPool{runner_function}; } - void stop() { stopReceivingNewCalls(); } + void stop() + { + std::lock_guard lock{mutex}; + should_stop = true; + + if (current_calls.empty()) + { + /// If there are no current calls then we call shutdownQueue() to signal the queue to stop waiting for next events. + /// The following line will make CompletionQueue::Next() stop waiting if the queue is empty and return false instead. + shutdownQueue(); + + /// If there are some current calls then we can't call shutdownQueue() right now because we want to let the current calls finish. + /// In this case function shutdownQueue() will be called later in run(). + } + } size_t getNumCurrentCalls() const { @@ -1789,12 +1812,6 @@ private: [this, call_type](bool ok) { onNewCall(call_type, ok); }); } - void stopReceivingNewCalls() - { - std::lock_guard lock{mutex}; - should_stop = true; - } - void onNewCall(CallType call_type, bool responder_started_ok) { std::lock_guard lock{mutex}; @@ -1827,38 +1844,47 @@ private: void run() { setThreadName("GRPCServerQueue"); - while (true) + + bool ok = false; + void * tag = nullptr; + + while (owner.queue->Next(&tag, &ok)) { - { - std::lock_guard lock{mutex}; - finished_calls.clear(); /// Destroy finished calls. - - /// If (should_stop == true) we continue processing until there is no active calls. - if (should_stop && current_calls.empty()) - { - bool all_responders_gone = std::all_of( - responders_for_new_calls.begin(), responders_for_new_calls.end(), - [](std::unique_ptr & responder) { return !responder; }); - if (all_responders_gone) - break; - } - } - - bool ok = false; - void * tag = nullptr; - if (!owner.queue->Next(&tag, &ok)) - { - /// Queue shutted down. - break; - } - auto & callback = *static_cast(tag); callback(ok); + + std::lock_guard lock{mutex}; + finished_calls.clear(); /// Destroy finished calls. + + /// If (should_stop == true) we continue processing while there are current calls. + if (should_stop && current_calls.empty()) + shutdownQueue(); } + + /// CompletionQueue::Next() returns false if the queue is fully drained and shut down. + } + + /// Shutdown the queue if that isn't done yet. + void shutdownQueue() + { + chassert(should_stop); + if (queue_is_shut_down) + return; + + queue_is_shut_down = true; + + /// Server should be shut down before CompletionQueue. + if (owner.grpc_server) + owner.grpc_server->Shutdown(); + + if (owner.queue) + owner.queue->Shutdown(); } GRPCServer & owner; + LoggerRawPtr log; ThreadFromGlobalPool queue_thread; + bool queue_is_shut_down = false; std::vector> responders_for_new_calls; std::map> current_calls; std::vector> finished_calls; @@ -1876,16 +1902,6 @@ GRPCServer::GRPCServer(IServer & iserver_, const Poco::Net::SocketAddress & addr GRPCServer::~GRPCServer() { - /// Server should be shutdown before CompletionQueue. - if (grpc_server) - grpc_server->Shutdown(); - - /// Completion Queue should be shutdown before destroying the runner, - /// because the runner is now probably executing CompletionQueue::Next() on queue_thread - /// which is blocked until an event is available or the queue is shutting down. - if (queue) - queue->Shutdown(); - runner.reset(); } diff --git a/src/Server/HTTP/authenticateUserByHTTP.cpp b/src/Server/HTTP/authenticateUserByHTTP.cpp new file mode 100644 index 00000000000..ac43bfd64c0 --- /dev/null +++ b/src/Server/HTTP/authenticateUserByHTTP.cpp @@ -0,0 +1,265 @@ +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#if USE_SSL +#include +#endif + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int AUTHENTICATION_FAILED; + extern const int BAD_ARGUMENTS; + extern const int SUPPORT_IS_DISABLED; +} + + +namespace +{ + /// Throws an exception that multiple authorization schemes are used simultaneously. + [[noreturn]] void throwMultipleAuthenticationMethods(std::string_view method1, std::string_view method2) + { + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, + "Invalid authentication: it is not allowed to use {} and {} simultaneously", method1, method2); + } + + /// Checks that a specified user name is not empty, and throws an exception if it's empty. + void checkUserNameNotEmpty(const String & user_name, std::string_view method) + { + if (user_name.empty()) + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Got an empty user name from {}", method); + } +} + + +bool authenticateUserByHTTP( + const HTTPServerRequest & request, + const HTMLForm & params, + HTTPServerResponse & response, + Session & session, + std::unique_ptr & request_credentials, + ContextPtr global_context, + LoggerPtr log) +{ + /// Get the credentials created by the previous call of authenticateUserByHTTP() while handling the previous HTTP request. + auto current_credentials = std::move(request_credentials); + + /// The user and password can be passed by headers (similar to X-Auth-*), + /// which is used by load balancers to pass authentication information. + std::string user = request.get("X-ClickHouse-User", ""); + std::string password = request.get("X-ClickHouse-Key", ""); + std::string quota_key = request.get("X-ClickHouse-Quota", ""); + bool has_auth_headers = !user.empty() || !password.empty(); + + /// The header 'X-ClickHouse-SSL-Certificate-Auth: on' enables checking the common name + /// extracted from the SSL certificate used for this connection instead of checking password. + bool has_ssl_certificate_auth = (request.get("X-ClickHouse-SSL-Certificate-Auth", "") == "on"); + + /// User name and password can be passed using HTTP Basic auth or query parameters + /// (both methods are insecure). + bool has_http_credentials = request.hasCredentials(); + bool has_credentials_in_query_params = params.has("user") || params.has("password"); + + std::string spnego_challenge; + SSLCertificateSubjects certificate_subjects; + + if (has_ssl_certificate_auth) + { +#if USE_SSL + /// For SSL certificate authentication we extract the user name from the "X-ClickHouse-User" HTTP header. + checkUserNameNotEmpty(user, "X-ClickHouse HTTP headers"); + + /// It is prohibited to mix different authorization schemes. + if (!password.empty()) + throwMultipleAuthenticationMethods("SSL certificate authentication", "authentication via password"); + if (has_http_credentials) + throwMultipleAuthenticationMethods("SSL certificate authentication", "Authorization HTTP header"); + if (has_credentials_in_query_params) + throwMultipleAuthenticationMethods("SSL certificate authentication", "authentication via parameters"); + + if (request.havePeerCertificate()) + certificate_subjects = extractSSLCertificateSubjects(request.peerCertificate()); + + if (certificate_subjects.empty()) + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, + "Invalid authentication: SSL certificate authentication requires nonempty certificate's Common Name or Subject Alternative Name"); +#else + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, + "SSL certificate authentication disabled because ClickHouse was built without SSL library"); +#endif + } + else if (has_auth_headers) + { + checkUserNameNotEmpty(user, "X-ClickHouse HTTP headers"); + + /// It is prohibited to mix different authorization schemes. + if (has_http_credentials) + throwMultipleAuthenticationMethods("X-ClickHouse HTTP headers", "Authorization HTTP header"); + if (has_credentials_in_query_params) + throwMultipleAuthenticationMethods("X-ClickHouse HTTP headers", "authentication via parameters"); + } + else if (has_http_credentials) + { + /// It is prohibited to mix different authorization schemes. + if (has_credentials_in_query_params) + throwMultipleAuthenticationMethods("Authorization HTTP header", "authentication via parameters"); + + std::string scheme; + std::string auth_info; + request.getCredentials(scheme, auth_info); + + if (Poco::icompare(scheme, "Basic") == 0) + { + Poco::Net::HTTPBasicCredentials credentials(auth_info); + user = credentials.getUsername(); + password = credentials.getPassword(); + checkUserNameNotEmpty(user, "Authorization HTTP header"); + } + else if (Poco::icompare(scheme, "Negotiate") == 0) + { + spnego_challenge = auth_info; + + if (spnego_challenge.empty()) + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: SPNEGO challenge is empty"); + } + else + { + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: '{}' HTTP Authorization scheme is not supported", scheme); + } + } + else + { + /// If the user name is not set we assume it's the 'default' user. + user = params.get("user", "default"); + password = params.get("password", ""); + checkUserNameNotEmpty(user, "authentication via parameters"); + } + + if (!certificate_subjects.empty()) + { + chassert(!user.empty()); + if (!current_credentials) + current_credentials = std::make_unique(user, std::move(certificate_subjects)); + + auto * certificate_credentials = dynamic_cast(current_credentials.get()); + if (!certificate_credentials) + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: expected SSL certificate authorization scheme"); + } + else if (!spnego_challenge.empty()) + { + if (!current_credentials) + current_credentials = global_context->makeGSSAcceptorContext(); + + auto * gss_acceptor_context = dynamic_cast(current_credentials.get()); + if (!gss_acceptor_context) + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: unexpected 'Negotiate' HTTP Authorization scheme expected"); + +#pragma clang diagnostic push +#pragma clang diagnostic ignored "-Wunreachable-code" + const auto spnego_response = base64Encode(gss_acceptor_context->processToken(base64Decode(spnego_challenge), log)); +#pragma clang diagnostic pop + + if (!spnego_response.empty()) + response.set("WWW-Authenticate", "Negotiate " + spnego_response); + + if (!gss_acceptor_context->isFailed() && !gss_acceptor_context->isReady()) + { + if (spnego_response.empty()) + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: 'Negotiate' HTTP Authorization failure"); + + response.setStatusAndReason(HTTPResponse::HTTP_UNAUTHORIZED); + response.send(); + /// Keep the credentials for next HTTP request. A client can handle HTTP_UNAUTHORIZED and send us more credentials with the next HTTP request. + request_credentials = std::move(current_credentials); + return false; + } + } + else // I.e., now using user name and password strings ("Basic"). + { + if (!current_credentials) + current_credentials = std::make_unique(); + + auto * basic_credentials = dynamic_cast(current_credentials.get()); + if (!basic_credentials) + throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: expected 'Basic' HTTP Authorization scheme"); + + chassert(!user.empty()); + basic_credentials->setUserName(user); + basic_credentials->setPassword(password); + } + + if (params.has("quota_key")) + { + if (!quota_key.empty()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, + "Invalid authentication: it is not allowed " + "to use quota key as HTTP header and as parameter simultaneously"); + + quota_key = params.get("quota_key"); + } + + /// Set client info. It will be used for quota accounting parameters in 'setUser' method. + + session.setHTTPClientInfo(request); + session.setQuotaClientKey(quota_key); + + /// Extract the last entry from comma separated list of forwarded_for addresses. + /// Only the last proxy can be trusted (if any). + String forwarded_address = session.getClientInfo().getLastForwardedFor(); + try + { + if (!forwarded_address.empty() && global_context->getConfigRef().getBool("auth_use_forwarded_address", false)) + session.authenticate(*current_credentials, Poco::Net::SocketAddress(forwarded_address, request.clientAddress().port())); + else + session.authenticate(*current_credentials, request.clientAddress()); + } + catch (const Authentication::Require & required_credentials) + { + current_credentials = std::make_unique(); + + if (required_credentials.getRealm().empty()) + response.set("WWW-Authenticate", "Basic"); + else + response.set("WWW-Authenticate", "Basic realm=\"" + required_credentials.getRealm() + "\""); + + response.setStatusAndReason(HTTPResponse::HTTP_UNAUTHORIZED); + response.send(); + /// Keep the credentials for next HTTP request. A client can handle HTTP_UNAUTHORIZED and send us more credentials with the next HTTP request. + request_credentials = std::move(current_credentials); + return false; + } + catch (const Authentication::Require & required_credentials) + { + current_credentials = global_context->makeGSSAcceptorContext(); + + if (required_credentials.getRealm().empty()) + response.set("WWW-Authenticate", "Negotiate"); + else + response.set("WWW-Authenticate", "Negotiate realm=\"" + required_credentials.getRealm() + "\""); + + response.setStatusAndReason(HTTPResponse::HTTP_UNAUTHORIZED); + response.send(); + /// Keep the credentials for next HTTP request. A client can handle HTTP_UNAUTHORIZED and send us more credentials with the next HTTP request. + request_credentials = std::move(current_credentials); + return false; + } + + return true; +} + +} diff --git a/src/Server/HTTP/authenticateUserByHTTP.h b/src/Server/HTTP/authenticateUserByHTTP.h new file mode 100644 index 00000000000..3b5a04cae68 --- /dev/null +++ b/src/Server/HTTP/authenticateUserByHTTP.h @@ -0,0 +1,31 @@ +#pragma once + +#include +#include + + +namespace DB +{ +class HTTPServerRequest; +class HTMLForm; +class HTTPServerResponse; +class Session; +class Credentials; + +/// Authenticates a user via HTTP protocol and initializes a session. +/// Usually retrieves the name and the password for that user from either the request's headers or from the query parameters. +/// Returns true when the user successfully authenticated, +/// the session instance will be configured accordingly, and the request_credentials instance will be dropped. +/// Returns false when the user is not authenticated yet, and the HTTP_UNAUTHORIZED response is sent with the "WWW-Authenticate" header, +/// in this case the `request_credentials` instance must be preserved until the next request or until any exception. +/// Throws an exception if authentication failed. +bool authenticateUserByHTTP( + const HTTPServerRequest & request, + const HTMLForm & params, + HTTPServerResponse & response, + Session & session, + std::unique_ptr & request_credentials, + ContextPtr global_context, + LoggerPtr log); + +} diff --git a/src/Server/HTTP/exceptionCodeToHTTPStatus.cpp b/src/Server/HTTP/exceptionCodeToHTTPStatus.cpp new file mode 100644 index 00000000000..6de57217aac --- /dev/null +++ b/src/Server/HTTP/exceptionCodeToHTTPStatus.cpp @@ -0,0 +1,158 @@ +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; + extern const int CANNOT_PARSE_TEXT; + extern const int CANNOT_PARSE_ESCAPE_SEQUENCE; + extern const int CANNOT_PARSE_QUOTED_STRING; + extern const int CANNOT_PARSE_DATE; + extern const int CANNOT_PARSE_DATETIME; + extern const int CANNOT_PARSE_NUMBER; + extern const int CANNOT_PARSE_DOMAIN_VALUE_FROM_STRING; + extern const int CANNOT_PARSE_IPV4; + extern const int CANNOT_PARSE_IPV6; + extern const int CANNOT_PARSE_UUID; + extern const int CANNOT_PARSE_INPUT_ASSERTION_FAILED; + extern const int CANNOT_SCHEDULE_TASK; + extern const int CANNOT_OPEN_FILE; + extern const int CANNOT_COMPILE_REGEXP; + extern const int DUPLICATE_COLUMN; + extern const int ILLEGAL_COLUMN; + extern const int THERE_IS_NO_COLUMN; + extern const int UNKNOWN_ELEMENT_IN_AST; + extern const int UNKNOWN_TYPE_OF_AST_NODE; + extern const int TOO_DEEP_AST; + extern const int TOO_BIG_AST; + extern const int UNEXPECTED_AST_STRUCTURE; + extern const int VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE; + + extern const int SYNTAX_ERROR; + + extern const int INCORRECT_DATA; + extern const int TYPE_MISMATCH; + + extern const int UNKNOWN_TABLE; + extern const int UNKNOWN_FUNCTION; + extern const int UNKNOWN_IDENTIFIER; + extern const int UNKNOWN_TYPE; + extern const int UNKNOWN_STORAGE; + extern const int UNKNOWN_DATABASE; + extern const int UNKNOWN_SETTING; + extern const int UNKNOWN_DIRECTION_OF_SORTING; + extern const int UNKNOWN_AGGREGATE_FUNCTION; + extern const int UNKNOWN_FORMAT; + extern const int UNKNOWN_DATABASE_ENGINE; + extern const int UNKNOWN_TYPE_OF_QUERY; + extern const int UNKNOWN_ROLE; + + extern const int QUERY_IS_TOO_LARGE; + + extern const int NOT_IMPLEMENTED; + extern const int SOCKET_TIMEOUT; + + extern const int UNKNOWN_USER; + extern const int WRONG_PASSWORD; + extern const int REQUIRED_PASSWORD; + extern const int AUTHENTICATION_FAILED; + extern const int SET_NON_GRANTED_ROLE; + + extern const int HTTP_LENGTH_REQUIRED; + + extern const int TIMEOUT_EXCEEDED; +} + + +Poco::Net::HTTPResponse::HTTPStatus exceptionCodeToHTTPStatus(int exception_code) +{ + using namespace Poco::Net; + + if (exception_code == ErrorCodes::REQUIRED_PASSWORD) + { + return HTTPResponse::HTTP_UNAUTHORIZED; + } + else if (exception_code == ErrorCodes::UNKNOWN_USER || + exception_code == ErrorCodes::WRONG_PASSWORD || + exception_code == ErrorCodes::AUTHENTICATION_FAILED || + exception_code == ErrorCodes::SET_NON_GRANTED_ROLE) + { + return HTTPResponse::HTTP_FORBIDDEN; + } + else if (exception_code == ErrorCodes::BAD_ARGUMENTS || + exception_code == ErrorCodes::CANNOT_COMPILE_REGEXP || + exception_code == ErrorCodes::CANNOT_PARSE_TEXT || + exception_code == ErrorCodes::CANNOT_PARSE_ESCAPE_SEQUENCE || + exception_code == ErrorCodes::CANNOT_PARSE_QUOTED_STRING || + exception_code == ErrorCodes::CANNOT_PARSE_DATE || + exception_code == ErrorCodes::CANNOT_PARSE_DATETIME || + exception_code == ErrorCodes::CANNOT_PARSE_NUMBER || + exception_code == ErrorCodes::CANNOT_PARSE_DOMAIN_VALUE_FROM_STRING || + exception_code == ErrorCodes::CANNOT_PARSE_IPV4 || + exception_code == ErrorCodes::CANNOT_PARSE_IPV6 || + exception_code == ErrorCodes::CANNOT_PARSE_INPUT_ASSERTION_FAILED || + exception_code == ErrorCodes::CANNOT_PARSE_UUID || + exception_code == ErrorCodes::DUPLICATE_COLUMN || + exception_code == ErrorCodes::ILLEGAL_COLUMN || + exception_code == ErrorCodes::UNKNOWN_ELEMENT_IN_AST || + exception_code == ErrorCodes::UNKNOWN_TYPE_OF_AST_NODE || + exception_code == ErrorCodes::THERE_IS_NO_COLUMN || + exception_code == ErrorCodes::TOO_DEEP_AST || + exception_code == ErrorCodes::TOO_BIG_AST || + exception_code == ErrorCodes::UNEXPECTED_AST_STRUCTURE || + exception_code == ErrorCodes::SYNTAX_ERROR || + exception_code == ErrorCodes::INCORRECT_DATA || + exception_code == ErrorCodes::TYPE_MISMATCH || + exception_code == ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE) + { + return HTTPResponse::HTTP_BAD_REQUEST; + } + else if (exception_code == ErrorCodes::UNKNOWN_TABLE || + exception_code == ErrorCodes::UNKNOWN_FUNCTION || + exception_code == ErrorCodes::UNKNOWN_IDENTIFIER || + exception_code == ErrorCodes::UNKNOWN_TYPE || + exception_code == ErrorCodes::UNKNOWN_STORAGE || + exception_code == ErrorCodes::UNKNOWN_DATABASE || + exception_code == ErrorCodes::UNKNOWN_SETTING || + exception_code == ErrorCodes::UNKNOWN_DIRECTION_OF_SORTING || + exception_code == ErrorCodes::UNKNOWN_AGGREGATE_FUNCTION || + exception_code == ErrorCodes::UNKNOWN_FORMAT || + exception_code == ErrorCodes::UNKNOWN_DATABASE_ENGINE || + exception_code == ErrorCodes::UNKNOWN_TYPE_OF_QUERY || + exception_code == ErrorCodes::UNKNOWN_ROLE) + { + return HTTPResponse::HTTP_NOT_FOUND; + } + else if (exception_code == ErrorCodes::QUERY_IS_TOO_LARGE) + { + return HTTPResponse::HTTP_REQUESTENTITYTOOLARGE; + } + else if (exception_code == ErrorCodes::NOT_IMPLEMENTED) + { + return HTTPResponse::HTTP_NOT_IMPLEMENTED; + } + else if (exception_code == ErrorCodes::SOCKET_TIMEOUT || + exception_code == ErrorCodes::CANNOT_OPEN_FILE) + { + return HTTPResponse::HTTP_SERVICE_UNAVAILABLE; + } + else if (exception_code == ErrorCodes::HTTP_LENGTH_REQUIRED) + { + return HTTPResponse::HTTP_LENGTH_REQUIRED; + } + else if (exception_code == ErrorCodes::TIMEOUT_EXCEEDED) + { + return HTTPResponse::HTTP_REQUEST_TIMEOUT; + } + else if (exception_code == ErrorCodes::CANNOT_SCHEDULE_TASK) + { + return HTTPResponse::HTTP_SERVICE_UNAVAILABLE; + } + + return HTTPResponse::HTTP_INTERNAL_SERVER_ERROR; +} + +} diff --git a/src/Server/HTTP/exceptionCodeToHTTPStatus.h b/src/Server/HTTP/exceptionCodeToHTTPStatus.h new file mode 100644 index 00000000000..aadec7aac5a --- /dev/null +++ b/src/Server/HTTP/exceptionCodeToHTTPStatus.h @@ -0,0 +1,11 @@ +#pragma once +#include + + +namespace DB +{ + +/// Converts Exception code to HTTP status code. +Poco::Net::HTTPResponse::HTTPStatus exceptionCodeToHTTPStatus(int exception_code); + +} diff --git a/src/Server/HTTP/sendExceptionToHTTPClient.cpp b/src/Server/HTTP/sendExceptionToHTTPClient.cpp new file mode 100644 index 00000000000..022a763a9a2 --- /dev/null +++ b/src/Server/HTTP/sendExceptionToHTTPClient.cpp @@ -0,0 +1,81 @@ +#include + +#include +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int HTTP_LENGTH_REQUIRED; + extern const int REQUIRED_PASSWORD; +} + + +void sendExceptionToHTTPClient( + const String & exception_message, + int exception_code, + HTTPServerRequest & request, + HTTPServerResponse & response, + WriteBufferFromHTTPServerResponse * out, + LoggerPtr log) +{ + setHTTPResponseStatusAndHeadersForException(exception_code, request, response, out, log); + + if (!out) + { + /// If nothing was sent yet. + WriteBufferFromHTTPServerResponse out_for_message{response, request.getMethod() == HTTPRequest::HTTP_HEAD, DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT}; + + out_for_message.writeln(exception_message); + out_for_message.finalize(); + } + else + { + /// If buffer has data, and that data wasn't sent yet, then no need to send that data + bool data_sent = (out->count() != out->offset()); + + if (!data_sent) + out->position() = out->buffer().begin(); + + out->writeln(exception_message); + out->finalize(); + } +} + + +void setHTTPResponseStatusAndHeadersForException( + int exception_code, HTTPServerRequest & request, HTTPServerResponse & response, WriteBufferFromHTTPServerResponse * out, LoggerPtr log) +{ + if (out) + out->setExceptionCode(exception_code); + else + response.set("X-ClickHouse-Exception-Code", toString(exception_code)); + + /// If HTTP method is POST and Keep-Alive is turned on, we should try to read the whole request body + /// to avoid reading part of the current request body in the next request. + if (request.getMethod() == Poco::Net::HTTPRequest::HTTP_POST && response.getKeepAlive() + && exception_code != ErrorCodes::HTTP_LENGTH_REQUIRED) + { + try + { + if (!request.getStream().eof()) + request.getStream().ignoreAll(); + } + catch (...) + { + tryLogCurrentException(log, "Cannot read remaining request body during exception handling"); + response.setKeepAlive(false); + } + } + + if (exception_code == ErrorCodes::REQUIRED_PASSWORD) + response.requireAuthentication("ClickHouse server HTTP API"); + else + response.setStatusAndReason(exceptionCodeToHTTPStatus(exception_code)); +} +} diff --git a/src/Server/HTTP/sendExceptionToHTTPClient.h b/src/Server/HTTP/sendExceptionToHTTPClient.h new file mode 100644 index 00000000000..31fda88d900 --- /dev/null +++ b/src/Server/HTTP/sendExceptionToHTTPClient.h @@ -0,0 +1,27 @@ +#pragma once + +#include +#include + + +namespace DB +{ +class HTTPServerRequest; +class HTTPServerResponse; +class WriteBufferFromHTTPServerResponse; + +/// Sends an exception to HTTP client. This function doesn't handle its own exceptions so it needs to be wrapped in try-catch. +/// Argument `out` may be either created from `response` or be nullptr (if it wasn't created before the exception). +void sendExceptionToHTTPClient( + const String & exception_message, + int exception_code, + HTTPServerRequest & request, + HTTPServerResponse & response, + WriteBufferFromHTTPServerResponse * out, + LoggerPtr log); + +/// Sets "X-ClickHouse-Exception-Code" header and the correspondent HTTP status in the response for an exception. +/// This is a part of what sendExceptionToHTTPClient() does. +void setHTTPResponseStatusAndHeadersForException( + int exception_code, HTTPServerRequest & request, HTTPServerResponse & response, WriteBufferFromHTTPServerResponse * out, LoggerPtr log); +} diff --git a/src/Server/HTTP/setReadOnlyIfHTTPMethodIdempotent.cpp b/src/Server/HTTP/setReadOnlyIfHTTPMethodIdempotent.cpp new file mode 100644 index 00000000000..48a3b39a3e7 --- /dev/null +++ b/src/Server/HTTP/setReadOnlyIfHTTPMethodIdempotent.cpp @@ -0,0 +1,25 @@ +#include + +#include +#include +#include + + +namespace DB +{ + +void setReadOnlyIfHTTPMethodIdempotent(ContextMutablePtr context, const String & http_method) +{ + /// Anything else beside HTTP POST should be readonly queries. + if (http_method != HTTPServerRequest::HTTP_POST) + { + /// 'readonly' setting values mean: + /// readonly = 0 - any query is allowed, client can change any setting. + /// readonly = 1 - only readonly queries are allowed, client can't change settings. + /// readonly = 2 - only readonly queries are allowed, client can change any setting except 'readonly'. + if (context->getSettingsRef().readonly == 0) + context->setSetting("readonly", 2); + } +} + +} diff --git a/src/Server/HTTP/setReadOnlyIfHTTPMethodIdempotent.h b/src/Server/HTTP/setReadOnlyIfHTTPMethodIdempotent.h new file mode 100644 index 00000000000..c46f2032d82 --- /dev/null +++ b/src/Server/HTTP/setReadOnlyIfHTTPMethodIdempotent.h @@ -0,0 +1,12 @@ +#pragma once + +#include + + +namespace DB +{ + +/// Sets readonly = 2 if the current HTTP method is not HTTP POST and if readonly is not set already. +void setReadOnlyIfHTTPMethodIdempotent(ContextMutablePtr context, const String & http_method); + +} diff --git a/src/Server/HTTPHandler.cpp b/src/Server/HTTPHandler.cpp index 3241e22fa35..c8a58527f2c 100644 --- a/src/Server/HTTPHandler.cpp +++ b/src/Server/HTTPHandler.cpp @@ -1,14 +1,11 @@ #include -#include #include -#include -#include -#include -#include #include #include #include +#include +#include #include #include #include @@ -37,19 +34,15 @@ #include #include #include +#include +#include +#include +#include + +#include #include "config.h" -#include -#include -#include -#include -#include -#include -#include -#include -#include - #include #include #include @@ -58,78 +51,19 @@ #include #include -#if USE_SSL -#include -#endif - namespace DB { namespace ErrorCodes { - extern const int BAD_ARGUMENTS; extern const int LOGICAL_ERROR; extern const int CANNOT_COMPILE_REGEXP; - extern const int CANNOT_OPEN_FILE; - extern const int CANNOT_PARSE_TEXT; - extern const int CANNOT_PARSE_ESCAPE_SEQUENCE; - extern const int CANNOT_PARSE_QUOTED_STRING; - extern const int CANNOT_PARSE_DATE; - extern const int CANNOT_PARSE_DATETIME; - extern const int CANNOT_PARSE_NUMBER; - extern const int CANNOT_PARSE_DOMAIN_VALUE_FROM_STRING; - extern const int CANNOT_PARSE_IPV4; - extern const int CANNOT_PARSE_IPV6; - extern const int CANNOT_PARSE_UUID; - extern const int CANNOT_PARSE_INPUT_ASSERTION_FAILED; - extern const int CANNOT_SCHEDULE_TASK; - extern const int DUPLICATE_COLUMN; - extern const int ILLEGAL_COLUMN; - extern const int THERE_IS_NO_COLUMN; - extern const int UNKNOWN_ELEMENT_IN_AST; - extern const int UNKNOWN_TYPE_OF_AST_NODE; - extern const int TOO_DEEP_AST; - extern const int TOO_BIG_AST; - extern const int UNEXPECTED_AST_STRUCTURE; - extern const int VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE; - extern const int SYNTAX_ERROR; - - extern const int INCORRECT_DATA; - extern const int TYPE_MISMATCH; - - extern const int UNKNOWN_TABLE; - extern const int UNKNOWN_FUNCTION; - extern const int UNKNOWN_IDENTIFIER; - extern const int UNKNOWN_TYPE; - extern const int UNKNOWN_STORAGE; - extern const int UNKNOWN_DATABASE; - extern const int UNKNOWN_SETTING; - extern const int UNKNOWN_DIRECTION_OF_SORTING; - extern const int UNKNOWN_AGGREGATE_FUNCTION; - extern const int UNKNOWN_FORMAT; - extern const int UNKNOWN_DATABASE_ENGINE; - extern const int UNKNOWN_TYPE_OF_QUERY; - extern const int UNKNOWN_ROLE; extern const int NO_ELEMENTS_IN_CONFIG; - extern const int QUERY_IS_TOO_LARGE; - - extern const int NOT_IMPLEMENTED; - extern const int SOCKET_TIMEOUT; - - extern const int UNKNOWN_USER; - extern const int WRONG_PASSWORD; - extern const int REQUIRED_PASSWORD; - extern const int AUTHENTICATION_FAILED; - extern const int SET_NON_GRANTED_ROLE; - extern const int INVALID_SESSION_TIMEOUT; extern const int HTTP_LENGTH_REQUIRED; - extern const int SUPPORT_IS_DISABLED; - - extern const int TIMEOUT_EXCEEDED; } namespace @@ -171,115 +105,6 @@ void processOptionsRequest(HTTPServerResponse & response, const Poco::Util::Laye } } -static String base64Decode(const String & encoded) -{ - String decoded; - Poco::MemoryInputStream istr(encoded.data(), encoded.size()); - Poco::Base64Decoder decoder(istr); - Poco::StreamCopier::copyToString(decoder, decoded); - return decoded; -} - -static String base64Encode(const String & decoded) -{ - std::ostringstream ostr; // STYLE_CHECK_ALLOW_STD_STRING_STREAM - ostr.exceptions(std::ios::failbit); - Poco::Base64Encoder encoder(ostr); - encoder.rdbuf()->setLineLength(0); - encoder << decoded; - encoder.close(); - return ostr.str(); -} - -static Poco::Net::HTTPResponse::HTTPStatus exceptionCodeToHTTPStatus(int exception_code) -{ - using namespace Poco::Net; - - if (exception_code == ErrorCodes::REQUIRED_PASSWORD) - { - return HTTPResponse::HTTP_UNAUTHORIZED; - } - else if (exception_code == ErrorCodes::UNKNOWN_USER || - exception_code == ErrorCodes::WRONG_PASSWORD || - exception_code == ErrorCodes::AUTHENTICATION_FAILED || - exception_code == ErrorCodes::SET_NON_GRANTED_ROLE) - { - return HTTPResponse::HTTP_FORBIDDEN; - } - else if (exception_code == ErrorCodes::BAD_ARGUMENTS || - exception_code == ErrorCodes::CANNOT_COMPILE_REGEXP || - exception_code == ErrorCodes::CANNOT_PARSE_TEXT || - exception_code == ErrorCodes::CANNOT_PARSE_ESCAPE_SEQUENCE || - exception_code == ErrorCodes::CANNOT_PARSE_QUOTED_STRING || - exception_code == ErrorCodes::CANNOT_PARSE_DATE || - exception_code == ErrorCodes::CANNOT_PARSE_DATETIME || - exception_code == ErrorCodes::CANNOT_PARSE_NUMBER || - exception_code == ErrorCodes::CANNOT_PARSE_DOMAIN_VALUE_FROM_STRING || - exception_code == ErrorCodes::CANNOT_PARSE_IPV4 || - exception_code == ErrorCodes::CANNOT_PARSE_IPV6 || - exception_code == ErrorCodes::CANNOT_PARSE_INPUT_ASSERTION_FAILED || - exception_code == ErrorCodes::CANNOT_PARSE_UUID || - exception_code == ErrorCodes::DUPLICATE_COLUMN || - exception_code == ErrorCodes::ILLEGAL_COLUMN || - exception_code == ErrorCodes::UNKNOWN_ELEMENT_IN_AST || - exception_code == ErrorCodes::UNKNOWN_TYPE_OF_AST_NODE || - exception_code == ErrorCodes::THERE_IS_NO_COLUMN || - exception_code == ErrorCodes::TOO_DEEP_AST || - exception_code == ErrorCodes::TOO_BIG_AST || - exception_code == ErrorCodes::UNEXPECTED_AST_STRUCTURE || - exception_code == ErrorCodes::SYNTAX_ERROR || - exception_code == ErrorCodes::INCORRECT_DATA || - exception_code == ErrorCodes::TYPE_MISMATCH || - exception_code == ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE) - { - return HTTPResponse::HTTP_BAD_REQUEST; - } - else if (exception_code == ErrorCodes::UNKNOWN_TABLE || - exception_code == ErrorCodes::UNKNOWN_FUNCTION || - exception_code == ErrorCodes::UNKNOWN_IDENTIFIER || - exception_code == ErrorCodes::UNKNOWN_TYPE || - exception_code == ErrorCodes::UNKNOWN_STORAGE || - exception_code == ErrorCodes::UNKNOWN_DATABASE || - exception_code == ErrorCodes::UNKNOWN_SETTING || - exception_code == ErrorCodes::UNKNOWN_DIRECTION_OF_SORTING || - exception_code == ErrorCodes::UNKNOWN_AGGREGATE_FUNCTION || - exception_code == ErrorCodes::UNKNOWN_FORMAT || - exception_code == ErrorCodes::UNKNOWN_DATABASE_ENGINE || - exception_code == ErrorCodes::UNKNOWN_TYPE_OF_QUERY || - exception_code == ErrorCodes::UNKNOWN_ROLE) - { - return HTTPResponse::HTTP_NOT_FOUND; - } - else if (exception_code == ErrorCodes::QUERY_IS_TOO_LARGE) - { - return HTTPResponse::HTTP_REQUESTENTITYTOOLARGE; - } - else if (exception_code == ErrorCodes::NOT_IMPLEMENTED) - { - return HTTPResponse::HTTP_NOT_IMPLEMENTED; - } - else if (exception_code == ErrorCodes::SOCKET_TIMEOUT || - exception_code == ErrorCodes::CANNOT_OPEN_FILE) - { - return HTTPResponse::HTTP_SERVICE_UNAVAILABLE; - } - else if (exception_code == ErrorCodes::HTTP_LENGTH_REQUIRED) - { - return HTTPResponse::HTTP_LENGTH_REQUIRED; - } - else if (exception_code == ErrorCodes::TIMEOUT_EXCEEDED) - { - return HTTPResponse::HTTP_REQUEST_TIMEOUT; - } - else if (exception_code == ErrorCodes::CANNOT_SCHEDULE_TASK) - { - return HTTPResponse::HTTP_SERVICE_UNAVAILABLE; - } - - return HTTPResponse::HTTP_INTERNAL_SERVER_ERROR; -} - - static std::chrono::steady_clock::duration parseSessionTimeout( const Poco::Util::AbstractConfiguration & config, const HTMLForm & params) @@ -356,204 +181,9 @@ HTTPHandler::HTTPHandler(IServer & server_, const std::string & name, const HTTP HTTPHandler::~HTTPHandler() = default; -bool HTTPHandler::authenticateUser( - HTTPServerRequest & request, - HTMLForm & params, - HTTPServerResponse & response) +bool HTTPHandler::authenticateUser(HTTPServerRequest & request, HTMLForm & params, HTTPServerResponse & response) { - using namespace Poco::Net; - - /// The user and password can be passed by headers (similar to X-Auth-*), - /// which is used by load balancers to pass authentication information. - std::string user = request.get("X-ClickHouse-User", ""); - std::string password = request.get("X-ClickHouse-Key", ""); - std::string quota_key = request.get("X-ClickHouse-Quota", ""); - - /// The header 'X-ClickHouse-SSL-Certificate-Auth: on' enables checking the common name - /// extracted from the SSL certificate used for this connection instead of checking password. - bool has_ssl_certificate_auth = (request.get("X-ClickHouse-SSL-Certificate-Auth", "") == "on"); - bool has_auth_headers = !user.empty() || !password.empty() || has_ssl_certificate_auth; - - /// User name and password can be passed using HTTP Basic auth or query parameters - /// (both methods are insecure). - bool has_http_credentials = request.hasCredentials(); - bool has_credentials_in_query_params = params.has("user") || params.has("password"); - - std::string spnego_challenge; - std::string certificate_common_name; - - if (has_auth_headers) - { - /// It is prohibited to mix different authorization schemes. - if (has_http_credentials) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, - "Invalid authentication: it is not allowed " - "to use SSL certificate authentication and Authorization HTTP header simultaneously"); - if (has_credentials_in_query_params) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, - "Invalid authentication: it is not allowed " - "to use SSL certificate authentication and authentication via parameters simultaneously simultaneously"); - - if (has_ssl_certificate_auth) - { -#if USE_SSL - if (!password.empty()) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, - "Invalid authentication: it is not allowed " - "to use SSL certificate authentication and authentication via password simultaneously"); - - if (request.havePeerCertificate()) - certificate_common_name = request.peerCertificate().commonName(); - - if (certificate_common_name.empty()) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, - "Invalid authentication: SSL certificate authentication requires nonempty certificate's Common Name"); -#else - throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, - "SSL certificate authentication disabled because ClickHouse was built without SSL library"); -#endif - } - } - else if (has_http_credentials) - { - /// It is prohibited to mix different authorization schemes. - if (has_credentials_in_query_params) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, - "Invalid authentication: it is not allowed " - "to use Authorization HTTP header and authentication via parameters simultaneously"); - - std::string scheme; - std::string auth_info; - request.getCredentials(scheme, auth_info); - - if (Poco::icompare(scheme, "Basic") == 0) - { - HTTPBasicCredentials credentials(auth_info); - user = credentials.getUsername(); - password = credentials.getPassword(); - } - else if (Poco::icompare(scheme, "Negotiate") == 0) - { - spnego_challenge = auth_info; - - if (spnego_challenge.empty()) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: SPNEGO challenge is empty"); - } - else - { - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: '{}' HTTP Authorization scheme is not supported", scheme); - } - } - else - { - /// If the user name is not set we assume it's the 'default' user. - user = params.get("user", "default"); - password = params.get("password", ""); - } - - if (!certificate_common_name.empty()) - { - if (!request_credentials) - request_credentials = std::make_unique(user, certificate_common_name); - - auto * certificate_credentials = dynamic_cast(request_credentials.get()); - if (!certificate_credentials) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: expected SSL certificate authorization scheme"); - } - else if (!spnego_challenge.empty()) - { - if (!request_credentials) - request_credentials = server.context()->makeGSSAcceptorContext(); - - auto * gss_acceptor_context = dynamic_cast(request_credentials.get()); - if (!gss_acceptor_context) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: unexpected 'Negotiate' HTTP Authorization scheme expected"); - -#pragma clang diagnostic push -#pragma clang diagnostic ignored "-Wunreachable-code" - const auto spnego_response = base64Encode(gss_acceptor_context->processToken(base64Decode(spnego_challenge), log)); -#pragma clang diagnostic pop - - if (!spnego_response.empty()) - response.set("WWW-Authenticate", "Negotiate " + spnego_response); - - if (!gss_acceptor_context->isFailed() && !gss_acceptor_context->isReady()) - { - if (spnego_response.empty()) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: 'Negotiate' HTTP Authorization failure"); - - response.setStatusAndReason(HTTPResponse::HTTP_UNAUTHORIZED); - response.send(); - return false; - } - } - else // I.e., now using user name and password strings ("Basic"). - { - if (!request_credentials) - request_credentials = std::make_unique(); - - auto * basic_credentials = dynamic_cast(request_credentials.get()); - if (!basic_credentials) - throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Invalid authentication: expected 'Basic' HTTP Authorization scheme"); - - basic_credentials->setUserName(user); - basic_credentials->setPassword(password); - } - - if (params.has("quota_key")) - { - if (!quota_key.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, - "Invalid authentication: it is not allowed " - "to use quota key as HTTP header and as parameter simultaneously"); - - quota_key = params.get("quota_key"); - } - - /// Set client info. It will be used for quota accounting parameters in 'setUser' method. - - session->setHTTPClientInfo(request); - session->setQuotaClientKey(quota_key); - - /// Extract the last entry from comma separated list of forwarded_for addresses. - /// Only the last proxy can be trusted (if any). - String forwarded_address = session->getClientInfo().getLastForwardedFor(); - try - { - if (!forwarded_address.empty() && server.config().getBool("auth_use_forwarded_address", false)) - session->authenticate(*request_credentials, Poco::Net::SocketAddress(forwarded_address, request.clientAddress().port())); - else - session->authenticate(*request_credentials, request.clientAddress()); - } - catch (const Authentication::Require & required_credentials) - { - request_credentials = std::make_unique(); - - if (required_credentials.getRealm().empty()) - response.set("WWW-Authenticate", "Basic"); - else - response.set("WWW-Authenticate", "Basic realm=\"" + required_credentials.getRealm() + "\""); - - response.setStatusAndReason(HTTPResponse::HTTP_UNAUTHORIZED); - response.send(); - return false; - } - catch (const Authentication::Require & required_credentials) - { - request_credentials = server.context()->makeGSSAcceptorContext(); - - if (required_credentials.getRealm().empty()) - response.set("WWW-Authenticate", "Negotiate"); - else - response.set("WWW-Authenticate", "Negotiate realm=\"" + required_credentials.getRealm() + "\""); - - response.setStatusAndReason(HTTPResponse::HTTP_UNAUTHORIZED); - response.send(); - return false; - } - - request_credentials.reset(); - return true; + return authenticateUserByHTTP(request, params, response, *session, request_credentials, server.context(), log); } @@ -725,10 +355,22 @@ void HTTPHandler::processQuery( std::unique_ptr in; - static const NameSet reserved_param_names{"compress", "decompress", "user", "password", "quota_key", "query_id", "stacktrace", "role", - "buffer_size", "wait_end_of_query", "session_id", "session_timeout", "session_check", "client_protocol_version", "close_session"}; + auto roles = params.getAll("role"); + if (!roles.empty()) + context->setCurrentRoles(roles); - Names reserved_param_suffixes; + std::string database = request.get("X-ClickHouse-Database", params.get("database", "")); + if (!database.empty()) + context->setCurrentDatabase(database); + + std::string default_format = request.get("X-ClickHouse-Format", params.get("default_format", "")); + if (!default_format.empty()) + context->setDefaultFormat(default_format); + + /// Anything else beside HTTP POST should be readonly queries. + setReadOnlyIfHTTPMethodIdempotent(context, request.getMethod()); + + bool has_external_data = startsWith(request.getContentType(), "multipart/form-data"); auto param_could_be_skipped = [&] (const String & name) { @@ -736,87 +378,36 @@ void HTTPHandler::processQuery( if (name.empty()) return true; + /// Some parameters (database, default_format, everything used in the code above) do not + /// belong to the Settings class. + static const NameSet reserved_param_names{"compress", "decompress", "user", "password", "quota_key", "query_id", "stacktrace", "role", + "buffer_size", "wait_end_of_query", "session_id", "session_timeout", "session_check", "client_protocol_version", "close_session", + "database", "default_format"}; + if (reserved_param_names.contains(name)) return true; - for (const String & suffix : reserved_param_suffixes) + /// For external data we also want settings. + if (has_external_data) { - if (endsWith(name, suffix)) - return true; + /// Skip unneeded parameters to avoid confusing them later with context settings or query parameters. + /// It is a bug and ambiguity with `date_time_input_format` and `low_cardinality_allow_in_native_format` formats/settings. + static const Names reserved_param_suffixes = {"_format", "_types", "_structure"}; + for (const String & suffix : reserved_param_suffixes) + { + if (endsWith(name, suffix)) + return true; + } } return false; }; - auto roles = params.getAll("role"); - if (!roles.empty()) - { - const auto & access_control = context->getAccessControl(); - const auto & user = context->getUser(); - std::vector roles_ids(roles.size()); - for (size_t i = 0; i < roles.size(); i++) - { - auto role_id = access_control.getID(roles[i]); - if (user->granted_roles.isGranted(role_id)) - roles_ids[i] = role_id; - else - throw Exception(ErrorCodes::SET_NON_GRANTED_ROLE, "Role {} should be granted to set as a current", roles[i].get()); - } - context->setCurrentRoles(roles_ids); - } - /// Settings can be overridden in the query. - /// Some parameters (database, default_format, everything used in the code above) do not - /// belong to the Settings class. - - /// 'readonly' setting values mean: - /// readonly = 0 - any query is allowed, client can change any setting. - /// readonly = 1 - only readonly queries are allowed, client can't change settings. - /// readonly = 2 - only readonly queries are allowed, client can change any setting except 'readonly'. - - /// In theory if initially readonly = 0, the client can change any setting and then set readonly - /// to some other value. - const auto & settings = context->getSettingsRef(); - - /// Anything else beside HTTP POST should be readonly queries. - if (request.getMethod() != HTTPServerRequest::HTTP_POST) - { - if (settings.readonly == 0) - context->setSetting("readonly", 2); - } - - bool has_external_data = startsWith(request.getContentType(), "multipart/form-data"); - - if (has_external_data) - { - /// Skip unneeded parameters to avoid confusing them later with context settings or query parameters. - reserved_param_suffixes.reserve(3); - /// It is a bug and ambiguity with `date_time_input_format` and `low_cardinality_allow_in_native_format` formats/settings. - reserved_param_suffixes.emplace_back("_format"); - reserved_param_suffixes.emplace_back("_types"); - reserved_param_suffixes.emplace_back("_structure"); - } - - std::string database = request.get("X-ClickHouse-Database", ""); - std::string default_format = request.get("X-ClickHouse-Format", ""); - SettingsChanges settings_changes; for (const auto & [key, value] : params) { - if (key == "database") - { - if (database.empty()) - database = value; - } - else if (key == "default_format") - { - if (default_format.empty()) - default_format = value; - } - else if (param_could_be_skipped(key)) - { - } - else + if (!param_could_be_skipped(key)) { /// Other than query parameters are treated as settings. if (!customizeQueryParam(context, key, value)) @@ -824,15 +415,9 @@ void HTTPHandler::processQuery( } } - if (!database.empty()) - context->setCurrentDatabase(database); - - if (!default_format.empty()) - context->setDefaultFormat(default_format); - - /// For external data we also want settings context->checkSettingsConstraints(settings_changes, SettingSource::QUERY); context->applySettingsChanges(settings_changes); + const auto & settings = context->getSettingsRef(); /// Set the query id supplied by the user, if any, and also update the OpenTelemetry fields. context->setCurrentQueryId(params.get("query_id", request.get("X-ClickHouse-Query-Id", ""))); @@ -934,7 +519,7 @@ void HTTPHandler::processQuery( { bool with_stacktrace = (params.getParsed("stacktrace", false) && server.config().getBool("enable_http_stacktrace", true)); ExecutionStatus status = ExecutionStatus::fromCurrentException("", with_stacktrace); - formatExceptionForClient(status.code, request, response, used_output); + setHTTPResponseStatusAndHeadersForException(status.code, request, response, used_output.out_holder.get(), log); current_output_format.setException(status.message); current_output_format.finalize(); used_output.exception_is_written = true; @@ -968,7 +553,7 @@ void HTTPHandler::trySendExceptionToClient( const std::string & s, int exception_code, HTTPServerRequest & request, HTTPServerResponse & response, Output & used_output) try { - formatExceptionForClient(exception_code, request, response, used_output); + setHTTPResponseStatusAndHeadersForException(exception_code, request, response, used_output.out_holder.get(), log); if (!used_output.out_holder && !used_output.exception_is_written) { @@ -1030,38 +615,6 @@ catch (...) used_output.cancel(); } -void HTTPHandler::formatExceptionForClient(int exception_code, HTTPServerRequest & request, HTTPServerResponse & response, Output & used_output) -{ - if (used_output.out_holder) - used_output.out_holder->setExceptionCode(exception_code); - else - response.set("X-ClickHouse-Exception-Code", toString(exception_code)); - - /// FIXME: make sure that no one else is reading from the same stream at the moment. - - /// If HTTP method is POST and Keep-Alive is turned on, we should try to read the whole request body - /// to avoid reading part of the current request body in the next request. - if (request.getMethod() == Poco::Net::HTTPRequest::HTTP_POST && response.getKeepAlive() - && exception_code != ErrorCodes::HTTP_LENGTH_REQUIRED) - { - try - { - if (!request.getStream().eof()) - request.getStream().ignoreAll(); - } - catch (...) - { - tryLogCurrentException(log, "Cannot read remaining request body during exception handling"); - response.setKeepAlive(false); - } - } - - if (exception_code == ErrorCodes::REQUIRED_PASSWORD) - response.requireAuthentication("ClickHouse server HTTP API"); - else - response.setStatusAndReason(exceptionCodeToHTTPStatus(exception_code)); -} - void HTTPHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event & write_event) { setThreadName("HTTPHandler"); diff --git a/src/Server/HTTPHandler.h b/src/Server/HTTPHandler.h index c78c45826f0..6580b317f6e 100644 --- a/src/Server/HTTPHandler.h +++ b/src/Server/HTTPHandler.h @@ -173,12 +173,6 @@ private: HTTPServerResponse & response, Output & used_output); - void formatExceptionForClient( - int exception_code, - HTTPServerRequest & request, - HTTPServerResponse & response, - Output & used_output); - static void pushDelayedResults(Output & used_output); }; diff --git a/src/Server/InterserverIOHTTPHandler.cpp b/src/Server/InterserverIOHTTPHandler.cpp index 45c28babe3a..e46021c8e68 100644 --- a/src/Server/InterserverIOHTTPHandler.cpp +++ b/src/Server/InterserverIOHTTPHandler.cpp @@ -3,6 +3,7 @@ #include #include +#include #include #include #include diff --git a/src/Server/KeeperReadinessHandler.cpp b/src/Server/KeeperReadinessHandler.cpp index c973be040c8..1972c3263b2 100644 --- a/src/Server/KeeperReadinessHandler.cpp +++ b/src/Server/KeeperReadinessHandler.cpp @@ -5,6 +5,7 @@ #include #include +#include #include #include #include diff --git a/src/Server/KeeperTCPHandler.cpp b/src/Server/KeeperTCPHandler.cpp index 4612e2e9fa8..b61df45133a 100644 --- a/src/Server/KeeperTCPHandler.cpp +++ b/src/Server/KeeperTCPHandler.cpp @@ -2,33 +2,37 @@ #if USE_NURAFT -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include -#ifdef POCO_HAVE_FD_EPOLL - #include -#else - #include -#endif +# ifdef POCO_HAVE_FD_EPOLL +# include +# else +# include +# endif + +namespace ProfileEvents +{ + extern const Event KeeperTotalElapsedMicroseconds; +} namespace DB @@ -397,13 +401,11 @@ void KeeperTCPHandler::runImpl() } auto response_fd = poll_wrapper->getResponseFD(); - auto response_callback = [responses_ = this->responses, response_fd](const Coordination::ZooKeeperResponsePtr & response) + auto response_callback = [my_responses = this->responses, + response_fd](const Coordination::ZooKeeperResponsePtr & response, Coordination::ZooKeeperRequestPtr request) { - if (!responses_->push(response)) - throw Exception(ErrorCodes::SYSTEM_ERROR, - "Could not push response with xid {} and zxid {}", - response->xid, - response->zxid); + if (!my_responses->push(RequestWithResponse{response, std::move(request)})) + throw Exception(ErrorCodes::SYSTEM_ERROR, "Could not push response with xid {} and zxid {}", response->xid, response->zxid); UInt8 single_byte = 1; [[maybe_unused]] ssize_t result = write(response_fd, &single_byte, sizeof(single_byte)); @@ -411,12 +413,12 @@ void KeeperTCPHandler::runImpl() keeper_dispatcher->registerSession(session_id, response_callback); Stopwatch logging_stopwatch; + auto operation_max_ms = keeper_dispatcher->getKeeperContext()->getCoordinationSettings()->log_slow_connection_operation_threshold_ms; auto log_long_operation = [&](const String & operation) { - constexpr UInt64 operation_max_ms = 500; auto elapsed_ms = logging_stopwatch.elapsedMilliseconds(); if (operation_max_ms < elapsed_ms) - LOG_TEST(log, "{} for session {} took {} ms", operation, session_id, elapsed_ms); + LOG_INFO(log, "{} for session {} took {} ms", operation, session_id, elapsed_ms); logging_stopwatch.restart(); }; @@ -467,19 +469,20 @@ void KeeperTCPHandler::runImpl() /// became inconsistent and race condition is possible. while (result.responses_count != 0) { - Coordination::ZooKeeperResponsePtr response; + RequestWithResponse request_with_response; - if (!responses->tryPop(response)) + if (!responses->tryPop(request_with_response)) throw Exception(ErrorCodes::LOGICAL_ERROR, "We must have ready response, but queue is empty. It's a bug."); log_long_operation("Waiting for response to be ready"); + auto & response = request_with_response.response; if (response->xid == close_xid) { LOG_DEBUG(log, "Session #{} successfully closed", session_id); return; } - updateStats(response); + updateStats(response, request_with_response.request); packageSent(); response->write(getWriteBuffer()); @@ -606,16 +609,29 @@ void KeeperTCPHandler::packageReceived() keeper_dispatcher->incrementPacketsReceived(); } -void KeeperTCPHandler::updateStats(Coordination::ZooKeeperResponsePtr & response) +void KeeperTCPHandler::updateStats(Coordination::ZooKeeperResponsePtr & response, const Coordination::ZooKeeperRequestPtr & request) { /// update statistics ignoring watch response and heartbeat. if (response->xid != Coordination::WATCH_XID && response->getOpNum() != Coordination::OpNum::Heartbeat) { - Int64 elapsed = (Poco::Timestamp() - operations[response->xid]) / 1000; - conn_stats.updateLatency(elapsed); + Int64 elapsed = (Poco::Timestamp() - operations[response->xid]); + ProfileEvents::increment(ProfileEvents::KeeperTotalElapsedMicroseconds, elapsed); + Int64 elapsed_ms = elapsed / 1000; + + if (request && elapsed_ms > static_cast(keeper_dispatcher->getKeeperContext()->getCoordinationSettings()->log_slow_total_threshold_ms)) + { + LOG_INFO( + log, + "Total time to process a request in session {} took too long ({}ms).\nRequest info: {}", + session_id, + elapsed_ms, + request->toString(/*short_format=*/true)); + } + + conn_stats.updateLatency(elapsed_ms); operations.erase(response->xid); - keeper_dispatcher->updateKeeperStatLatency(elapsed); + keeper_dispatcher->updateKeeperStatLatency(elapsed_ms); last_op.set(std::make_unique(LastOp{ .name = Coordination::toString(response->getOpNum()), diff --git a/src/Server/KeeperTCPHandler.h b/src/Server/KeeperTCPHandler.h index c1c522eee89..7c2b8acf624 100644 --- a/src/Server/KeeperTCPHandler.h +++ b/src/Server/KeeperTCPHandler.h @@ -26,7 +26,13 @@ namespace DB struct SocketInterruptablePollWrapper; using SocketInterruptablePollWrapperPtr = std::unique_ptr; -using ThreadSafeResponseQueue = ConcurrentBoundedQueue; +struct RequestWithResponse +{ + Coordination::ZooKeeperResponsePtr response; + Coordination::ZooKeeperRequestPtr request; /// it can be nullptr for some responses +}; + +using ThreadSafeResponseQueue = ConcurrentBoundedQueue; using ThreadSafeResponseQueuePtr = std::shared_ptr; struct LastOp; @@ -104,7 +110,7 @@ private: void packageSent(); void packageReceived(); - void updateStats(Coordination::ZooKeeperResponsePtr & response); + void updateStats(Coordination::ZooKeeperResponsePtr & response, const Coordination::ZooKeeperRequestPtr & request); Poco::Timestamp established; diff --git a/src/Server/MySQLHandler.cpp b/src/Server/MySQLHandler.cpp index 9471509ad4b..b6d795b1e69 100644 --- a/src/Server/MySQLHandler.cpp +++ b/src/Server/MySQLHandler.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -204,6 +205,10 @@ void MySQLHandler::run() session->setClientConnectionId(connection_id); + const Settings & settings = server.context()->getSettingsRef(); + socket().setReceiveTimeout(settings.receive_timeout); + socket().setSendTimeout(settings.send_timeout); + in = std::make_shared(socket(), read_event); out = std::make_shared(socket(), write_event); packet_endpoint = std::make_shared(*in, *out, sequence_id); @@ -451,6 +456,7 @@ void MySQLHandler::comQuery(ReadBuffer & payload, bool binary_protocol) // Settings replacements if (!should_replace) + { for (auto const & [mysql_setting, clickhouse_setting] : settings_replacements) { const auto replacement_query_opt = setSettingReplacementQuery(query, mysql_setting, clickhouse_setting); @@ -461,6 +467,7 @@ void MySQLHandler::comQuery(ReadBuffer & payload, bool binary_protocol) break; } } + } auto query_context = session->makeQueryContext(); query_context->setCurrentQueryId(fmt::format("mysql:{}:{}", connection_id, toString(UUIDHelpers::generateV4()))); @@ -470,6 +477,10 @@ void MySQLHandler::comQuery(ReadBuffer & payload, bool binary_protocol) settings.prefer_column_name_to_alias = true; query_context->setSettings(settings); + /// Update timeouts + socket().setReceiveTimeout(settings.receive_timeout); + socket().setSendTimeout(settings.send_timeout); + CurrentThread::QueryScope query_scope{query_context}; std::atomic affected_rows {0}; @@ -643,7 +654,11 @@ void MySQLHandlerSSL::finishHandshakeSSL( client_capabilities = ssl_request.capability_flags; max_packet_size = ssl_request.max_packet_size ? ssl_request.max_packet_size : MAX_PACKET_LENGTH; secure_connection = true; + ss = std::make_shared(SecureStreamSocket::attach(socket(), SSLManager::instance().defaultServerContext())); + ss->setReceiveTimeout(socket().getReceiveTimeout()); + ss->setSendTimeout(socket().getSendTimeout()); + in = std::make_shared(*ss); out = std::make_shared(*ss); sequence_id = 2; diff --git a/src/Server/MySQLHandler.h b/src/Server/MySQLHandler.h index 2f891ebf810..f632ef51c00 100644 --- a/src/Server/MySQLHandler.h +++ b/src/Server/MySQLHandler.h @@ -8,6 +8,7 @@ #include #include #include +#include #include "IO/ReadBufferFromString.h" #include "IServer.h" diff --git a/src/Server/NotFoundHandler.cpp b/src/Server/NotFoundHandler.cpp index 38f56921c89..304021ee8dc 100644 --- a/src/Server/NotFoundHandler.cpp +++ b/src/Server/NotFoundHandler.cpp @@ -1,6 +1,7 @@ #include #include +#include #include namespace DB diff --git a/src/Server/PostgreSQLHandler.cpp b/src/Server/PostgreSQLHandler.cpp index 8ba8421e6f0..cde31b4c58a 100644 --- a/src/Server/PostgreSQLHandler.cpp +++ b/src/Server/PostgreSQLHandler.cpp @@ -13,6 +13,7 @@ #include #include #include +#include #if USE_SSL # include diff --git a/src/Server/PrometheusRequestHandler.cpp b/src/Server/PrometheusRequestHandler.cpp index dff960f7031..87c106c3fc0 100644 --- a/src/Server/PrometheusRequestHandler.cpp +++ b/src/Server/PrometheusRequestHandler.cpp @@ -18,9 +18,6 @@ void PrometheusRequestHandler::handleRequest(HTTPServerRequest & request, HTTPSe { try { - /// Raw config reference is used here to avoid dependency on Context and ServerSettings. - /// This is painful, because this class is also used in a build with CLICKHOUSE_KEEPER_STANDALONE_BUILD=1 - /// And there ordinary Context is replaced with a tiny clone. const auto & config = server.config(); unsigned keep_alive_timeout = config.getUInt("keep_alive_timeout", DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT); @@ -33,15 +30,8 @@ void PrometheusRequestHandler::handleRequest(HTTPServerRequest & request, HTTPSe response.setContentType("text/plain; version=0.0.4; charset=UTF-8"); WriteBufferFromHTTPServerResponse wb(response, request.getMethod() == Poco::Net::HTTPRequest::HTTP_HEAD, keep_alive_timeout, write_event); - try - { - metrics_writer->write(wb); - wb.finalize(); - } - catch (...) - { - wb.finalize(); - } + metrics_writer->write(wb); + wb.finalize(); } catch (...) { diff --git a/src/Server/ProtocolServerAdapter.cpp b/src/Server/ProtocolServerAdapter.cpp index 8d14a849894..b41ad2376f1 100644 --- a/src/Server/ProtocolServerAdapter.cpp +++ b/src/Server/ProtocolServerAdapter.cpp @@ -1,7 +1,7 @@ #include #include -#if USE_GRPC && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_GRPC #include #endif @@ -37,7 +37,7 @@ ProtocolServerAdapter::ProtocolServerAdapter( { } -#if USE_GRPC && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_GRPC class ProtocolServerAdapter::GRPCServerAdapterImpl : public Impl { public: diff --git a/src/Server/ProtocolServerAdapter.h b/src/Server/ProtocolServerAdapter.h index dd11c1dfc58..76a6776ed9c 100644 --- a/src/Server/ProtocolServerAdapter.h +++ b/src/Server/ProtocolServerAdapter.h @@ -23,7 +23,7 @@ public: ProtocolServerAdapter & operator =(ProtocolServerAdapter && src) = default; ProtocolServerAdapter(const std::string & listen_host_, const char * port_name_, const std::string & description_, std::unique_ptr tcp_server_); -#if USE_GRPC && !defined(CLICKHOUSE_KEEPER_STANDALONE_BUILD) +#if USE_GRPC ProtocolServerAdapter(const std::string & listen_host_, const char * port_name_, const std::string & description_, std::unique_ptr grpc_server_); #endif diff --git a/src/Server/ProxyV1Handler.cpp b/src/Server/ProxyV1Handler.cpp index d5e6ab23360..5b33a119cef 100644 --- a/src/Server/ProxyV1Handler.cpp +++ b/src/Server/ProxyV1Handler.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/Server/ReplicasStatusHandler.cpp b/src/Server/ReplicasStatusHandler.cpp index 67823117758..f43357db0a8 100644 --- a/src/Server/ReplicasStatusHandler.cpp +++ b/src/Server/ReplicasStatusHandler.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include @@ -7,6 +8,7 @@ #include #include #include +#include #include #include diff --git a/src/Server/StaticRequestHandler.cpp b/src/Server/StaticRequestHandler.cpp index 331b7a84857..f3981dea9fb 100644 --- a/src/Server/StaticRequestHandler.cpp +++ b/src/Server/StaticRequestHandler.cpp @@ -4,13 +4,14 @@ #include "HTTPHandlerFactory.h" #include "HTTPResponseHeaderWriter.h" +#include #include #include #include -#include #include -#include +#include #include +#include #include diff --git a/src/Server/TCPHandler.cpp b/src/Server/TCPHandler.cpp index 14a2bceebf1..c5dfe3e6e5f 100644 --- a/src/Server/TCPHandler.cpp +++ b/src/Server/TCPHandler.cpp @@ -1,48 +1,48 @@ -#include -#include -#include #include #include #include #include -#include #include -#include -#include -#include -#include -#include -#include -#include -#include -#include +#include +#include +#include #include #include -#include -#include -#include -#include -#include +#include +#include +#include #include #include -#include -#include +#include +#include +#include +#include +#include +#include +#include #include #include #include +#include +#include +#include +#include #include -#include #include #include -#include -#include -#include -#include -#include -#include +#include +#include +#include +#include #include +#include +#include +#include +#include +#include +#include +#include #include -#include #include #include @@ -61,6 +61,8 @@ #include +#include + using namespace std::literals; using namespace DB; @@ -387,7 +389,7 @@ void TCPHandler::runImpl() query_scope.emplace(query_context, /* fatal_error_callback */ [this] { - std::lock_guard lock(fatal_error_mutex); + std::lock_guard lock(out_mutex); sendLogs(); }); @@ -475,7 +477,7 @@ void TCPHandler::runImpl() Stopwatch watch; CurrentMetrics::Increment callback_metric_increment(CurrentMetrics::ReadTaskRequestsSent); - std::lock_guard lock(task_callback_mutex); + std::scoped_lock lock(out_mutex, task_callback_mutex); if (state.cancellation_status == CancellationStatus::FULLY_CANCELLED) return {}; @@ -491,7 +493,7 @@ void TCPHandler::runImpl() { Stopwatch watch; CurrentMetrics::Increment callback_metric_increment(CurrentMetrics::MergeTreeAllRangesAnnouncementsSent); - std::lock_guard lock(task_callback_mutex); + std::scoped_lock lock(out_mutex, task_callback_mutex); if (state.cancellation_status == CancellationStatus::FULLY_CANCELLED) return; @@ -505,7 +507,7 @@ void TCPHandler::runImpl() { Stopwatch watch; CurrentMetrics::Increment callback_metric_increment(CurrentMetrics::MergeTreeReadTaskRequestsSent); - std::lock_guard lock(task_callback_mutex); + std::scoped_lock lock(out_mutex, task_callback_mutex); if (state.cancellation_status == CancellationStatus::FULLY_CANCELLED) return std::nullopt; @@ -553,7 +555,7 @@ void TCPHandler::runImpl() { auto callback = [this]() { - std::scoped_lock lock(task_callback_mutex, fatal_error_mutex); + std::scoped_lock lock(out_mutex, task_callback_mutex); if (getQueryCancellationStatus() == CancellationStatus::FULLY_CANCELLED) return true; @@ -572,7 +574,7 @@ void TCPHandler::runImpl() finish_or_cancel(); - std::lock_guard lock(task_callback_mutex); + std::lock_guard lock(out_mutex); /// Send final progress after calling onFinish(), since it will update the progress. /// @@ -595,7 +597,7 @@ void TCPHandler::runImpl() break; { - std::lock_guard lock(task_callback_mutex); + std::lock_guard lock(out_mutex); sendLogs(); sendEndOfStream(); } @@ -666,7 +668,7 @@ void TCPHandler::runImpl() // Server should die on std logic errors in debug, like with assert() // or ErrorCodes::LOGICAL_ERROR. This helps catch these errors in // tests. -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD catch (const std::logic_error & e) { state.io.onException(); @@ -888,12 +890,11 @@ AsynchronousInsertQueue::PushResult TCPHandler::processAsyncInsertQuery(Asynchro while (readDataNext()) { - squashing.header = state.block_for_insert; - auto planned_chunk = squashing.add({state.block_for_insert.getColumns(), state.block_for_insert.rows()}); - if (planned_chunk.hasChunkInfo()) + squashing.setHeader(state.block_for_insert.cloneEmpty()); + auto result_chunk = Squashing::squash(squashing.add({state.block_for_insert.getColumns(), state.block_for_insert.rows()})); + if (result_chunk) { - Chunk result_chunk = DB::Squashing::squash(std::move(planned_chunk)); - auto result = state.block_for_insert.cloneWithColumns(result_chunk.getColumns()); + auto result = squashing.getHeader().cloneWithColumns(result_chunk.detachColumns()); return PushResult { .status = PushResult::TOO_MUCH_DATA, @@ -902,12 +903,13 @@ AsynchronousInsertQueue::PushResult TCPHandler::processAsyncInsertQuery(Asynchro } } - auto planned_chunk = squashing.flush(); - Chunk result_chunk; - if (planned_chunk.hasChunkInfo()) - result_chunk = DB::Squashing::squash(std::move(planned_chunk)); + Chunk result_chunk = Squashing::squash(squashing.flush()); + if (!result_chunk) + { + return insert_queue.pushQueryWithBlock(state.parsed_query, squashing.getHeader(), query_context); + } - auto result = squashing.header.cloneWithColumns(result_chunk.getColumns()); + auto result = squashing.getHeader().cloneWithColumns(result_chunk.detachColumns()); return insert_queue.pushQueryWithBlock(state.parsed_query, std::move(result), query_context); } @@ -961,8 +963,8 @@ void TCPHandler::processInsertQuery() if (settings.throw_if_deduplication_in_dependent_materialized_views_enabled_with_async_insert && settings.deduplicate_blocks_in_dependent_materialized_views) throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, - "Deduplication is dependent materialized view cannot work together with async inserts. "\ - "Please disable eiher `deduplicate_blocks_in_dependent_materialized_views` or `async_insert` setting."); + "Deduplication in dependent materialized view cannot work together with async inserts. "\ + "Please disable either `deduplicate_blocks_in_dependent_materialized_views` or `async_insert` setting."); auto result = processAsyncInsertQuery(*insert_queue); if (result.status == AsynchronousInsertQueue::PushResult::OK) @@ -1014,7 +1016,7 @@ void TCPHandler::processOrdinaryQuery() if (query_context->getSettingsRef().allow_experimental_query_deduplication) { - std::lock_guard lock(task_callback_mutex); + std::lock_guard lock(out_mutex); sendPartUUIDs(); } @@ -1024,18 +1026,29 @@ void TCPHandler::processOrdinaryQuery() if (header) { - std::lock_guard lock(task_callback_mutex); + std::lock_guard lock(out_mutex); sendData(header); } } /// Defer locking to cover a part of the scope below and everything after it - std::unique_lock progress_lock(task_callback_mutex, std::defer_lock); + std::unique_lock out_lock(out_mutex, std::defer_lock); { PullingAsyncPipelineExecutor executor(pipeline); CurrentMetrics::Increment query_thread_metric_increment{CurrentMetrics::QueryThread}; + /// The following may happen: + /// * current thread is holding the lock + /// * because of the exception we unwind the stack and call the destructor of `executor` + /// * the destructor calls cancel() and waits for all query threads to finish + /// * at the same time one of the query threads is trying to acquire the lock, e.g. inside `merge_tree_read_task_callback` + /// * deadlock + SCOPE_EXIT({ + if (out_lock.owns_lock()) + out_lock.unlock(); + }); + Block block; while (executor.pull(block, interactive_delay / 1000)) { @@ -1056,6 +1069,9 @@ void TCPHandler::processOrdinaryQuery() executor.cancelReading(); } + lock.unlock(); + out_lock.lock(); + if (after_send_progress.elapsed() / 1000 >= interactive_delay) { /// Some time passed and there is a progress. @@ -1071,12 +1087,13 @@ void TCPHandler::processOrdinaryQuery() if (!state.io.null_format) sendData(block); } + + out_lock.unlock(); } /// This lock wasn't acquired before and we make .lock() call here - /// so everything under this line is covered even together - /// with sendProgress() out of the scope - progress_lock.lock(); + /// so everything under this line is covered. + out_lock.lock(); /** If data has run out, we will send the profiling data and total values to * the last zero block to be able to use @@ -1102,6 +1119,7 @@ void TCPHandler::processOrdinaryQuery() last_sent_snapshots.clear(); } + out_lock.lock(); sendProgress(); } @@ -1485,7 +1503,7 @@ void TCPHandler::receiveHello() try { session->authenticate( - SSLCertificateCredentials{user, secure_socket.peerCertificate().commonName()}, + SSLCertificateCredentials{user, extractSSLCertificateSubjects(secure_socket.peerCertificate())}, getClientAddress(client_info)); return; } @@ -1870,7 +1888,7 @@ void TCPHandler::receiveQuery() #endif } - query_context = session->makeQueryContext(std::move(client_info)); + query_context = session->makeQueryContext(client_info); /// Sets the default database if it wasn't set earlier for the session context. if (is_interserver_mode && !default_database.empty()) @@ -1885,6 +1903,16 @@ void TCPHandler::receiveQuery() /// /// Settings /// + + /// FIXME: Remove when allow_experimental_analyzer will become obsolete. + /// Analyzer became Beta in 24.3 and started to be enabled by default. + /// We have to disable it for ourselves to make sure we don't have different settings on + /// different servers. + if (query_kind == ClientInfo::QueryKind::SECONDARY_QUERY + && client_info.getVersionNumber() < VersionNumber(23, 3, 0) + && !passed_settings.allow_experimental_analyzer.changed) + passed_settings.set("allow_experimental_analyzer", false); + auto settings_changes = passed_settings.changes(); query_kind = query_context->getClientInfo().query_kind; if (query_kind == ClientInfo::QueryKind::INITIAL_QUERY) @@ -2092,6 +2120,7 @@ void TCPHandler::initBlockOutput(const Block & block) *state.maybe_compressed_out, client_tcp_protocol_version, block.cloneEmpty(), + std::nullopt, !query_settings.low_cardinality_allow_in_native_format); } } @@ -2106,6 +2135,7 @@ void TCPHandler::initLogsBlockOutput(const Block & block) *out, client_tcp_protocol_version, block.cloneEmpty(), + std::nullopt, !query_settings.low_cardinality_allow_in_native_format); } } @@ -2120,6 +2150,7 @@ void TCPHandler::initProfileEventsBlockOutput(const Block & block) *out, client_tcp_protocol_version, block.cloneEmpty(), + std::nullopt, !query_settings.low_cardinality_allow_in_native_format); } } diff --git a/src/Server/TCPHandler.h b/src/Server/TCPHandler.h index 77899d247e6..74afb5a14a5 100644 --- a/src/Server/TCPHandler.h +++ b/src/Server/TCPHandler.h @@ -226,8 +226,13 @@ private: std::optional nonce; String cluster; + /// `out_mutex` protects `out` (WriteBuffer). + /// So it is used for method sendData(), sendProgress(), sendLogs(), etc. + std::mutex out_mutex; + /// `task_callback_mutex` protects tasks callbacks. + /// Inside these callbacks we might also change cancellation status, + /// so it also protects cancellation status checks. std::mutex task_callback_mutex; - std::mutex fatal_error_mutex; /// At the moment, only one ongoing query in the connection is supported at a time. QueryState state; diff --git a/src/Server/WebUIRequestHandler.cpp b/src/Server/WebUIRequestHandler.cpp index 68d3ff0b325..a3d098014e7 100644 --- a/src/Server/WebUIRequestHandler.cpp +++ b/src/Server/WebUIRequestHandler.cpp @@ -5,8 +5,9 @@ #include #include -#include +#include #include +#include #include #include diff --git a/src/Storages/AlterCommands.cpp b/src/Storages/AlterCommands.cpp index 35a5e95e643..7891042bb96 100644 --- a/src/Storages/AlterCommands.cpp +++ b/src/Storages/AlterCommands.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include @@ -33,9 +34,9 @@ #include #include #include +#include #include #include -#include #include diff --git a/src/Storages/ColumnsDescription.cpp b/src/Storages/ColumnsDescription.cpp index c07583cd39d..da749812167 100644 --- a/src/Storages/ColumnsDescription.cpp +++ b/src/Storages/ColumnsDescription.cpp @@ -701,15 +701,15 @@ std::optional ColumnsDescription::tryGetColumn(const GetColumns auto jt = subcolumns.get<0>().find(column_name); if (jt != subcolumns.get<0>().end()) return *jt; - } - /// Check for dynamic subcolumns. - auto [ordinary_column_name, dynamic_subcolumn_name] = Nested::splitName(column_name); - it = columns.get<1>().find(ordinary_column_name); - if (it != columns.get<1>().end() && it->type->hasDynamicSubcolumns()) - { - if (auto dynamic_subcolumn_type = it->type->tryGetSubcolumnType(dynamic_subcolumn_name)) - return NameAndTypePair(ordinary_column_name, dynamic_subcolumn_name, it->type, dynamic_subcolumn_type); + /// Check for dynamic subcolumns. + auto [ordinary_column_name, dynamic_subcolumn_name] = Nested::splitName(column_name); + it = columns.get<1>().find(ordinary_column_name); + if (it != columns.get<1>().end() && it->type->hasDynamicSubcolumns()) + { + if (auto dynamic_subcolumn_type = it->type->tryGetSubcolumnType(dynamic_subcolumn_name)) + return NameAndTypePair(ordinary_column_name, dynamic_subcolumn_name, it->type, dynamic_subcolumn_type); + } } return {}; diff --git a/src/Storages/Distributed/DistributedAsyncInsertBatch.cpp b/src/Storages/Distributed/DistributedAsyncInsertBatch.cpp index 06d4c185840..e1facec5b40 100644 --- a/src/Storages/Distributed/DistributedAsyncInsertBatch.cpp +++ b/src/Storages/Distributed/DistributedAsyncInsertBatch.cpp @@ -196,6 +196,16 @@ void DistributedAsyncInsertBatch::readText(ReadBuffer & in) UInt64 idx; in >> idx >> "\n"; files.push_back(std::filesystem::absolute(fmt::format("{}/{}.bin", parent.path, idx)).string()); + + ReadBufferFromFile header_buffer(files.back()); + const DistributedAsyncInsertHeader & header = DistributedAsyncInsertHeader::read(header_buffer, parent.log); + total_bytes += total_bytes; + + if (header.rows) + { + total_rows += header.rows; + total_bytes += header.bytes; + } } recovered = true; diff --git a/src/Storages/Distributed/DistributedSink.cpp b/src/Storages/Distributed/DistributedSink.cpp index e556bda2561..b2ce62caf0a 100644 --- a/src/Storages/Distributed/DistributedSink.cpp +++ b/src/Storages/Distributed/DistributedSink.cpp @@ -32,6 +32,7 @@ #include #include #include +#include #include @@ -134,7 +135,7 @@ DistributedSink::DistributedSink( } -void DistributedSink::consume(Chunk chunk) +void DistributedSink::consume(Chunk & chunk) { if (is_first_chunk) { @@ -142,7 +143,7 @@ void DistributedSink::consume(Chunk chunk) is_first_chunk = false; } - auto ordinary_block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto ordinary_block = getHeader().cloneWithColumns(chunk.getColumns()); if (insert_sync) writeSync(ordinary_block); @@ -420,7 +421,13 @@ DistributedSink::runWritingJob(JobReplica & job, const Block & current_block, si /// to resolve tables (in InterpreterInsertQuery::getTable()) auto copy_query_ast = query_ast->clone(); - InterpreterInsertQuery interp(copy_query_ast, job.local_context, allow_materialized); + InterpreterInsertQuery interp( + copy_query_ast, + job.local_context, + allow_materialized, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); auto block_io = interp.execute(); job.pipeline = std::move(block_io.pipeline); @@ -715,7 +722,13 @@ void DistributedSink::writeToLocal(const Cluster::ShardInfo & shard_info, const try { - InterpreterInsertQuery interp(query_ast, context, allow_materialized); + InterpreterInsertQuery interp( + query_ast, + context, + allow_materialized, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); auto block_io = interp.execute(); PushingPipelineExecutor executor(block_io.pipeline); diff --git a/src/Storages/Distributed/DistributedSink.h b/src/Storages/Distributed/DistributedSink.h index a4c95633595..5b7396f2c6f 100644 --- a/src/Storages/Distributed/DistributedSink.h +++ b/src/Storages/Distributed/DistributedSink.h @@ -49,7 +49,7 @@ public: const Names & columns_to_send_); String getName() const override { return "DistributedSink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onFinish() override; private: diff --git a/src/Storages/FileLog/FileLogSettings.cpp b/src/Storages/FileLog/FileLogSettings.cpp index 8e245285b9a..cf2d6b64d7c 100644 --- a/src/Storages/FileLog/FileLogSettings.cpp +++ b/src/Storages/FileLog/FileLogSettings.cpp @@ -3,6 +3,7 @@ #include #include #include +#include namespace DB diff --git a/src/Storages/FileLog/StorageFileLog.cpp b/src/Storages/FileLog/StorageFileLog.cpp index abd4b4ce23b..0f9bd8b6ff9 100644 --- a/src/Storages/FileLog/StorageFileLog.cpp +++ b/src/Storages/FileLog/StorageFileLog.cpp @@ -740,7 +740,14 @@ bool StorageFileLog::streamToViews() auto new_context = Context::createCopy(getContext()); - InterpreterInsertQuery interpreter(insert, new_context, false, true, true); + InterpreterInsertQuery interpreter( + insert, + new_context, + /* allow_materialized */ false, + /* no_squash */ true, + /* no_destination */ true, + /* async_isnert */ false); + auto block_io = interpreter.execute(); /// Each stream responsible for closing it's files and store meta diff --git a/src/Storages/Hive/StorageHive.cpp b/src/Storages/Hive/StorageHive.cpp index 28d8128e052..9063437c065 100644 --- a/src/Storages/Hive/StorageHive.cpp +++ b/src/Storages/Hive/StorageHive.cpp @@ -321,7 +321,7 @@ public: Chunk generateChunkFromMetadata() { - size_t num_rows = std::min(current_file_remained_rows, UInt64(getContext()->getSettings().max_block_size)); + size_t num_rows = std::min(current_file_remained_rows, UInt64(getContext()->getSettingsRef().max_block_size)); current_file_remained_rows -= num_rows; Block source_block; @@ -964,7 +964,7 @@ HiveFiles StorageHive::collectHiveFiles( /// Hive files to collect HiveFiles hive_files; Int64 hit_parttions_num = 0; - Int64 hive_max_query_partitions = context_->getSettings().max_partitions_to_read; + Int64 hive_max_query_partitions = context_->getSettingsRef().max_partitions_to_read; /// Mutext to protect hive_files, which maybe appended in multiple threads std::mutex hive_files_mutex; ThreadPool pool{CurrentMetrics::StorageHiveThreads, CurrentMetrics::StorageHiveThreadsActive, CurrentMetrics::StorageHiveThreadsScheduled, max_threads}; diff --git a/src/Storages/IStorage.cpp b/src/Storages/IStorage.cpp index 1f7ac23eb82..3b9c6ff28fa 100644 --- a/src/Storages/IStorage.cpp +++ b/src/Storages/IStorage.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include diff --git a/src/Storages/IStorage.h b/src/Storages/IStorage.h index 98afd844046..6217470780d 100644 --- a/src/Storages/IStorage.h +++ b/src/Storages/IStorage.h @@ -165,6 +165,8 @@ public: /// Returns true if the storage supports reading of subcolumns of complex types. virtual bool supportsSubcolumns() const { return false; } + /// Returns true if storage supports optimizations of functions by reading subcolumns. + virtual bool supportsOptimizationToSubcolumns() const { return supportsSubcolumns(); } /// Returns true if the storage supports transactions for SELECT, INSERT and ALTER queries. /// Storage may throw an exception later if some query kind is not fully supported. diff --git a/src/Storages/IStorageCluster.cpp b/src/Storages/IStorageCluster.cpp index 9c5b29ae265..63467603d16 100644 --- a/src/Storages/IStorageCluster.cpp +++ b/src/Storages/IStorageCluster.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include diff --git a/src/Storages/IStorageCluster.h b/src/Storages/IStorageCluster.h index f3283247672..893cf222556 100644 --- a/src/Storages/IStorageCluster.h +++ b/src/Storages/IStorageCluster.h @@ -37,7 +37,10 @@ public: QueryProcessingStage::Enum getQueryProcessingStage(ContextPtr, QueryProcessingStage::Enum, const StorageSnapshotPtr &, SelectQueryInfo &) const override; - bool isRemote() const override { return true; } + bool isRemote() const final { return true; } + bool supportsSubcolumns() const override { return true; } + bool supportsOptimizationToSubcolumns() const override { return false; } + bool supportsTrivialCountOptimization(const StorageSnapshotPtr &, ContextPtr) const override { return true; } protected: virtual void updateBeforeRead(const ContextPtr &) {} diff --git a/src/Storages/IStorage_fwd.h b/src/Storages/IStorage_fwd.h index b9243b029b0..4cbc586a745 100644 --- a/src/Storages/IStorage_fwd.h +++ b/src/Storages/IStorage_fwd.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include @@ -9,9 +10,10 @@ namespace DB { class IStorage; +struct SnapshotDetachedTable; using ConstStoragePtr = std::shared_ptr; using StoragePtr = std::shared_ptr; using Tables = std::map; - +using SnapshotDetachedTables = std::map; } diff --git a/src/Storages/Kafka/StorageKafka.cpp b/src/Storages/Kafka/StorageKafka.cpp index f5c5d093ce1..3aad64a0cfb 100644 --- a/src/Storages/Kafka/StorageKafka.cpp +++ b/src/Storages/Kafka/StorageKafka.cpp @@ -418,8 +418,11 @@ namespace } StorageKafka::StorageKafka( - const StorageID & table_id_, ContextPtr context_, - const ColumnsDescription & columns_, std::unique_ptr kafka_settings_, + const StorageID & table_id_, + ContextPtr context_, + const ColumnsDescription & columns_, + const String & comment, + std::unique_ptr kafka_settings_, const String & collection_name_) : IStorage(table_id_) , WithContext(context_->getGlobalContext()) @@ -451,6 +454,7 @@ StorageKafka::StorageKafka( StorageInMemoryMetadata storage_metadata; storage_metadata.setColumns(columns_); + storage_metadata.setComment(comment); setInMemoryMetadata(storage_metadata); setVirtuals(createVirtuals(kafka_settings->kafka_handle_error_mode)); @@ -1099,7 +1103,13 @@ bool StorageKafka::streamToViews() // Create a stream for each consumer and join them in a union stream // Only insert into dependent views and expect that input blocks contain virtual columns - InterpreterInsertQuery interpreter(insert, kafka_context, false, true, true); + InterpreterInsertQuery interpreter( + insert, + kafka_context, + /* allow_materialized */ false, + /* no_squash */ true, + /* no_destination */ true, + /* async_isnert */ false); auto block_io = interpreter.execute(); // Create a stream for each consumer and join them in a union stream @@ -1311,7 +1321,7 @@ void registerStorageKafka(StorageFactory & factory) "See https://clickhouse.com/docs/en/engines/table-engines/integrations/kafka/#configuration"); } - return std::make_shared(args.table_id, args.getContext(), args.columns, std::move(kafka_settings), collection_name); + return std::make_shared(args.table_id, args.getContext(), args.columns, args.comment, std::move(kafka_settings), collection_name); }; factory.registerStorage("Kafka", creator_fn, StorageFactory::StorageFeatures{ .supports_settings = true, }); diff --git a/src/Storages/Kafka/StorageKafka.h b/src/Storages/Kafka/StorageKafka.h index fa4affbda36..31e1a6076b6 100644 --- a/src/Storages/Kafka/StorageKafka.h +++ b/src/Storages/Kafka/StorageKafka.h @@ -40,6 +40,7 @@ public: const StorageID & table_id_, ContextPtr context_, const ColumnsDescription & columns_, + const String & comment, std::unique_ptr kafka_settings_, const String & collection_name_); diff --git a/src/Storages/LiveView/LiveViewSink.h b/src/Storages/LiveView/LiveViewSink.h index 792133ced64..9803fa0a160 100644 --- a/src/Storages/LiveView/LiveViewSink.h +++ b/src/Storages/LiveView/LiveViewSink.h @@ -71,9 +71,9 @@ public: new_hash.reset(); } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { - auto block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto block = getHeader().cloneWithColumns(chunk.getColumns()); block.updateHash(*new_hash); new_blocks->push_back(std::move(block)); } diff --git a/src/Storages/LiveView/StorageLiveView.cpp b/src/Storages/LiveView/StorageLiveView.cpp index 57a1ea302f9..71b1a0a73c9 100644 --- a/src/Storages/LiveView/StorageLiveView.cpp +++ b/src/Storages/LiveView/StorageLiveView.cpp @@ -21,12 +21,14 @@ limitations under the License. */ #include #include #include +#include #include #include #include #include #include #include +#include #include #include @@ -330,7 +332,7 @@ Pipe StorageLiveView::watch( return reader; } -void StorageLiveView::writeBlock(const Block & block, ContextPtr local_context) +void StorageLiveView::writeBlock(StorageLiveView & live_view, Block && block, Chunk::ChunkInfoCollection && chunk_infos, ContextPtr local_context) { auto output = std::make_shared(*this); @@ -407,6 +409,21 @@ void StorageLiveView::writeBlock(const Block & block, ContextPtr local_context) builder = interpreter.buildQueryPipeline(); } + builder.addSimpleTransform([&](const Block & cur_header) + { + return std::make_shared(chunk_infos.clone(), cur_header); + }); + + String live_view_id = live_view.getStorageID().hasUUID() ? toString(live_view.getStorageID().uuid) : live_view.getStorageID().getFullNameNotQuoted(); + builder.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared(live_view_id, stream_header); + }); + builder.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared(stream_header); + }); + builder.addSimpleTransform([&](const Block & cur_header) { return std::make_shared(cur_header); diff --git a/src/Storages/LiveView/StorageLiveView.h b/src/Storages/LiveView/StorageLiveView.h index 91daac32c7b..12d8e898347 100644 --- a/src/Storages/LiveView/StorageLiveView.h +++ b/src/Storages/LiveView/StorageLiveView.h @@ -118,7 +118,7 @@ public: return 0; } - void writeBlock(const Block & block, ContextPtr context); + void writeBlock(StorageLiveView & live_view, Block && block, Chunk::ChunkInfoCollection && chunk_infos, ContextPtr context); void refresh(); diff --git a/src/Storages/MaterializedView/RefreshTask.cpp b/src/Storages/MaterializedView/RefreshTask.cpp index bc8cb0ce69a..aa8f51d5295 100644 --- a/src/Storages/MaterializedView/RefreshTask.cpp +++ b/src/Storages/MaterializedView/RefreshTask.cpp @@ -3,6 +3,7 @@ #include #include +#include #include #include #include @@ -303,7 +304,7 @@ void RefreshTask::refreshTask() { PreformattedMessage message = getCurrentExceptionMessageAndPattern(true); auto text = message.text; - message.text = fmt::format("Refresh failed: {}", message.text); + message.text = fmt::format("Refresh view {} failed: {}", view->getStorageID().getFullTableName(), message.text); LOG_ERROR(log, message); exception = text; } @@ -356,7 +357,7 @@ void RefreshTask::refreshTask() stop_requested = true; tryLogCurrentException(log, "Unexpected exception in refresh scheduling, please investigate. The view will be stopped."); -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD abortOnFailedAssertion("Unexpected exception in refresh scheduling"); #endif } @@ -377,7 +378,13 @@ void RefreshTask::executeRefreshUnlocked(std::shared_ptr +#include #include #include #include @@ -54,7 +55,7 @@ ApproximateNearestNeighborInformation::Metric stringToMetric(std::string_view me ApproximateNearestNeighborCondition::ApproximateNearestNeighborCondition(const SelectQueryInfo & query_info, ContextPtr context) : block_with_constants(KeyCondition::getBlockWithConstants(query_info.query, query_info.syntax_analyzer_result, context)) , index_granularity(context->getMergeTreeSettings().index_granularity) - , max_limit_for_ann_queries(context->getSettings().max_limit_for_ann_queries) + , max_limit_for_ann_queries(context->getSettingsRef().max_limit_for_ann_queries) , index_is_useful(checkQueryStructure(query_info)) {} diff --git a/src/Storages/MergeTree/AsyncBlockIDsCache.cpp b/src/Storages/MergeTree/AsyncBlockIDsCache.cpp index 9d64592ed64..6606b00e287 100644 --- a/src/Storages/MergeTree/AsyncBlockIDsCache.cpp +++ b/src/Storages/MergeTree/AsyncBlockIDsCache.cpp @@ -1,7 +1,8 @@ +#include +#include +#include #include #include -#include -#include #include diff --git a/src/Storages/MergeTree/DataPartStorageOnDiskBase.cpp b/src/Storages/MergeTree/DataPartStorageOnDiskBase.cpp index 378a1944396..df151e8478f 100644 --- a/src/Storages/MergeTree/DataPartStorageOnDiskBase.cpp +++ b/src/Storages/MergeTree/DataPartStorageOnDiskBase.cpp @@ -73,12 +73,7 @@ std::optional DataPartStorageOnDiskBase::getRelativePathForPrefix(Logger for (int try_no = 0; try_no < 10; ++try_no) { - if (prefix.empty()) - res = part_dir + (try_no ? "_try" + DB::toString(try_no) : ""); - else if (prefix.ends_with("_")) - res = prefix + part_dir + (try_no ? "_try" + DB::toString(try_no) : ""); - else - res = prefix + "_" + part_dir + (try_no ? "_try" + DB::toString(try_no) : ""); + res = getPartDirForPrefix(prefix, detached, try_no); if (!volume->getDisk()->exists(full_relative_path / res)) return res; @@ -101,6 +96,36 @@ std::optional DataPartStorageOnDiskBase::getRelativePathForPrefix(Logger return res; } +String DataPartStorageOnDiskBase::getPartDirForPrefix(const String & prefix, bool detached, int try_no) const +{ + /// This function joins `prefix` and the part name and an attempt number returning something like "__". + String res = prefix; + if (!prefix.empty() && !prefix.ends_with("_")) + res += "_"; + + /// During RESTORE temporary part directories are created with names like "tmp_restore_all_2_2_0-XXXXXXXX". + /// To detach such a directory we need to rename it replacing "tmp_restore_" with a specified prefix, + /// and a random suffix with an attempt number. + String part_name; + if (detached && part_dir.starts_with("tmp_restore_")) + { + part_name = part_dir.substr(strlen("tmp_restore_")); + size_t endpos = part_name.find('-'); + if (endpos != String::npos) + part_name.erase(endpos, String::npos); + } + + if (!part_name.empty()) + res += part_name; + else + res += part_dir; + + if (try_no) + res += "_try" + DB::toString(try_no); + + return res; +} + bool DataPartStorageOnDiskBase::looksLikeBrokenDetachedPartHasTheSameContent(const String & detached_part_path, std::optional & original_checksums_content, std::optional & original_files_list) const diff --git a/src/Storages/MergeTree/DataPartStorageOnDiskBase.h b/src/Storages/MergeTree/DataPartStorageOnDiskBase.h index 81353d4e20b..1707efc2d4d 100644 --- a/src/Storages/MergeTree/DataPartStorageOnDiskBase.h +++ b/src/Storages/MergeTree/DataPartStorageOnDiskBase.h @@ -148,6 +148,9 @@ private: /// Actual file name may be the same as expected /// or be the name of the file with packed data. virtual NameSet getActualFileNamesOnDisk(const NameSet & file_names) const = 0; + + /// Returns the destination path for the part directory while copying a detached part. + String getPartDirForPrefix(const String & prefix, bool detached, int try_no) const; }; } diff --git a/src/Storages/MergeTree/DataPartsExchange.cpp b/src/Storages/MergeTree/DataPartsExchange.cpp index b7fdcc63e1b..061ee356203 100644 --- a/src/Storages/MergeTree/DataPartsExchange.cpp +++ b/src/Storages/MergeTree/DataPartsExchange.cpp @@ -15,6 +15,8 @@ #include #include #include +#include +#include #include #include #include @@ -223,14 +225,18 @@ void Service::processQuery(const HTMLForm & params, ReadBuffer & /*body*/, Write } catch (const Exception & e) { - if (e.code() != ErrorCodes::ABORTED && e.code() != ErrorCodes::CANNOT_WRITE_TO_OSTREAM) + if (e.code() != ErrorCodes::CANNOT_WRITE_TO_OSTREAM + && !isRetryableException(std::current_exception())) + { report_broken_part(); + } throw; } catch (...) { - report_broken_part(); + if (!isRetryableException(std::current_exception())) + report_broken_part(); throw; } } diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index f3a54092a75..3a44359b537 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -375,6 +376,12 @@ void IMergeTreeDataPart::unloadIndex() index_loaded = false; } +bool IMergeTreeDataPart::isIndexLoaded() const +{ + std::scoped_lock lock(index_mutex); + return index_loaded; +} + void IMergeTreeDataPart::setName(const String & new_name) { mutable_name = new_name; @@ -645,15 +652,12 @@ size_t IMergeTreeDataPart::getFileSizeOrZero(const String & file_name) const return checksum->second.file_size; } -String IMergeTreeDataPart::getColumnNameWithMinimumCompressedSize(bool with_subcolumns) const +String IMergeTreeDataPart::getColumnNameWithMinimumCompressedSize(const NamesAndTypesList & available_columns) const { - auto options = GetColumnsOptions(GetColumnsOptions::AllPhysical).withSubcolumns(with_subcolumns); - auto columns_list = columns_description.get(options); - std::optional minimum_size_column; UInt64 minimum_size = std::numeric_limits::max(); - for (const auto & column : columns_list) + for (const auto & column : available_columns) { if (!hasColumnFiles(column)) continue; @@ -735,10 +739,25 @@ void IMergeTreeDataPart::loadColumnsChecksumsIndexes(bool require_columns_checks } catch (...) { - /// Don't scare people with broken part error + /// Don't scare people with broken part error if it's retryable. if (!isRetryableException(std::current_exception())) + { LOG_ERROR(storage.log, "Part {} is broken and needs manual correction", getDataPartStorage().getFullPath()); + if (Exception * e = exception_cast(std::current_exception())) + { + /// Probably there is something wrong with files of this part. + /// So it can be helpful to add to the error message some information about those files. + String files_in_part; + for (auto it = getDataPartStorage().iterate(); it->isValid(); it->next()) + files_in_part += fmt::format("{}{} ({} bytes)", (files_in_part.empty() ? "" : ", "), it->name(), getDataPartStorage().getFileSize(it->name())); + if (!files_in_part.empty()) + e->addMessage("Part contains files: {}", files_in_part); + if (isEmpty()) + e->addMessage("Part is empty"); + } + } + // There could be conditions that data part to be loaded is broken, but some of meta infos are already written // into metadata before exception, need to clean them all. metadata_manager->deleteAll(/*include_projection*/ true); @@ -1307,6 +1326,17 @@ void IMergeTreeDataPart::loadRowsCount() auto buf = metadata_manager->read("count.txt"); readIntText(rows_count, *buf); assertEOF(*buf); + + if (!index_granularity.empty() && rows_count < index_granularity.getTotalRows() && index_granularity_info.fixed_index_granularity) + { + /// Adjust last granule size to match the number of rows in the part in case of fixed index_granularity. + index_granularity.popMark(); + index_granularity.appendMark(rows_count % index_granularity_info.fixed_index_granularity); + if (rows_count != index_granularity.getTotalRows()) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Index granularity total rows in part {} does not match rows_count: {}, instead of {}", + name, index_granularity.getTotalRows(), rows_count); + } }; if (index_granularity.empty()) @@ -1577,7 +1607,7 @@ void IMergeTreeDataPart::loadColumns(bool require) if (getFileNameForColumn(column)) loaded_columns.push_back(column); - if (columns.empty()) + if (loaded_columns.empty()) throw Exception(ErrorCodes::NO_FILE_IN_DATA_PART, "No columns in part {}", name); if (!is_readonly_storage) @@ -2322,21 +2352,26 @@ String IMergeTreeDataPart::getUniqueId() const return getDataPartStorage().getUniqueId(); } +UInt128 IMergeTreeDataPart::getPartBlockIDHash() const +{ + SipHash hash; + checksums.computeTotalChecksumDataOnly(hash); + return hash.get128(); +} + String IMergeTreeDataPart::getZeroLevelPartBlockID(std::string_view token) const { if (info.level != 0) throw Exception(ErrorCodes::LOGICAL_ERROR, "Trying to get block id for non zero level part {}", name); - SipHash hash; if (token.empty()) { - checksums.computeTotalChecksumDataOnly(hash); - } - else - { - hash.update(token.data(), token.size()); + const auto hash_value = getPartBlockIDHash(); + return info.partition_id + "_" + toString(hash_value.items[0]) + "_" + toString(hash_value.items[1]); } + SipHash hash; + hash.update(token.data(), token.size()); const auto hash_value = hash.get128(); return info.partition_id + "_" + toString(hash_value.items[0]) + "_" + toString(hash_value.items[1]); } diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.h b/src/Storages/MergeTree/IMergeTreeDataPart.h index c9b3ec26df0..85ef0472ce7 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.h +++ b/src/Storages/MergeTree/IMergeTreeDataPart.h @@ -196,7 +196,9 @@ public: /// Returns the name of a column with minimum compressed size (as returned by getColumnSize()). /// If no checksums are present returns the name of the first physically existing column. - String getColumnNameWithMinimumCompressedSize(bool with_subcolumns) const; + /// We pass a list of available columns since the ones available in the current storage snapshot might be smaller + /// than the one the table has (e.g a DROP COLUMN happened) and we don't want to get a column not in the snapshot + String getColumnNameWithMinimumCompressedSize(const NamesAndTypesList & available_columns) const; bool contains(const IMergeTreeDataPart & other) const { return info.contains(other.info); } @@ -210,6 +212,7 @@ public: /// Compute part block id for zero level part. Otherwise throws an exception. /// If token is not empty, block id is calculated based on it instead of block data + UInt128 getPartBlockIDHash() const; String getZeroLevelPartBlockID(std::string_view token) const; void setName(const String & new_name); @@ -369,6 +372,7 @@ public: void setIndex(const Columns & cols_); void setIndex(Columns && cols_); void unloadIndex(); + bool isIndexLoaded() const; /// For data in RAM ('index') UInt64 getIndexSizeInBytes() const; diff --git a/src/Storages/MergeTree/IMergeTreeDataPartInfoForReader.h b/src/Storages/MergeTree/IMergeTreeDataPartInfoForReader.h index ccc88079daa..7d4fb1df1c2 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPartInfoForReader.h +++ b/src/Storages/MergeTree/IMergeTreeDataPartInfoForReader.h @@ -47,7 +47,7 @@ public: virtual std::optional getColumnPosition(const String & column_name) const = 0; - virtual String getColumnNameWithMinimumCompressedSize(bool with_subcolumns) const = 0; + virtual String getColumnNameWithMinimumCompressedSize(const NamesAndTypesList & available_columns) const = 0; virtual const MergeTreeDataPartChecksums & getChecksums() const = 0; diff --git a/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp b/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp index 6152da78395..c87f66b64f3 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp @@ -1,4 +1,5 @@ #include +#include namespace DB { @@ -71,9 +72,21 @@ IMergeTreeDataPartWriter::IMergeTreeDataPartWriter( Columns IMergeTreeDataPartWriter::releaseIndexColumns() { - return Columns( - std::make_move_iterator(index_columns.begin()), - std::make_move_iterator(index_columns.end())); + /// The memory for index was allocated without thread memory tracker. + /// We need to deallocate it in shrinkToFit without memory tracker as well. + MemoryTrackerBlockerInThread temporarily_disable_memory_tracker; + + Columns result; + result.reserve(index_columns.size()); + + for (auto & column : index_columns) + { + column->shrinkToFit(); + result.push_back(std::move(column)); + } + + index_columns.clear(); + return result; } SerializationPtr IMergeTreeDataPartWriter::getSerialization(const String & column_name) const diff --git a/src/Storages/MergeTree/IMergeTreeDataPartWriter.h b/src/Storages/MergeTree/IMergeTreeDataPartWriter.h index d9e9a433827..adaa4eddb98 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPartWriter.h +++ b/src/Storages/MergeTree/IMergeTreeDataPartWriter.h @@ -1,11 +1,10 @@ #pragma once +#include #include -#include #include #include #include -#include #include #include @@ -13,6 +12,9 @@ namespace DB { +struct MergeTreeSettings; +using MergeTreeSettingsPtr = std::shared_ptr; + Block getBlockAndPermute(const Block & block, const Names & names, const IColumn::Permutation * permutation); Block permuteBlockIfNeeded(const Block & block, const IColumn::Permutation * permutation); diff --git a/src/Storages/MergeTree/IMergedBlockOutputStream.cpp b/src/Storages/MergeTree/IMergedBlockOutputStream.cpp index 89c813ab233..70e838e666a 100644 --- a/src/Storages/MergeTree/IMergedBlockOutputStream.cpp +++ b/src/Storages/MergeTree/IMergedBlockOutputStream.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include diff --git a/src/Storages/MergeTree/IMergedBlockOutputStream.h b/src/Storages/MergeTree/IMergedBlockOutputStream.h index a9b058418ea..cfcfb177e05 100644 --- a/src/Storages/MergeTree/IMergedBlockOutputStream.h +++ b/src/Storages/MergeTree/IMergedBlockOutputStream.h @@ -1,16 +1,18 @@ #pragma once #include -#include -#include -#include #include #include +#include +#include #include namespace DB { +struct MergeTreeSettings; +using MergeTreeSettingsPtr = std::shared_ptr; + class IMergedBlockOutputStream { public: diff --git a/src/Storages/MergeTree/IPartMetadataManager.h b/src/Storages/MergeTree/IPartMetadataManager.h index cef1d10e4ad..e817421f7d0 100644 --- a/src/Storages/MergeTree/IPartMetadataManager.h +++ b/src/Storages/MergeTree/IPartMetadataManager.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include #include diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index 7e4b1db4c89..40eef8e7431 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -874,46 +875,6 @@ static Field applyFunctionForField( return (*col)[0]; } -/// The case when arguments may have types different than in the primary key. -static std::pair applyFunctionForFieldOfUnknownType( - const FunctionBasePtr & func, - const DataTypePtr & arg_type, - const Field & arg_value) -{ - ColumnsWithTypeAndName arguments{{ arg_type->createColumnConst(1, arg_value), arg_type, "x" }}; - DataTypePtr return_type = func->getResultType(); - - auto col = func->execute(arguments, return_type, 1); - - Field result = (*col)[0]; - - return {std::move(result), std::move(return_type)}; -} - - -/// Same as above but for binary operators -static std::pair applyBinaryFunctionForFieldOfUnknownType( - const FunctionOverloadResolverPtr & func, - const DataTypePtr & arg_type, - const Field & arg_value, - const DataTypePtr & arg_type2, - const Field & arg_value2) -{ - ColumnsWithTypeAndName arguments{ - {arg_type->createColumnConst(1, arg_value), arg_type, "x"}, {arg_type2->createColumnConst(1, arg_value2), arg_type2, "y"}}; - - FunctionBasePtr func_base = func->build(arguments); - - DataTypePtr return_type = func_base->getResultType(); - - auto col = func_base->execute(arguments, return_type, 1); - - Field result = (*col)[0]; - - return {std::move(result), std::move(return_type)}; -} - - static FieldRef applyFunction(const FunctionBasePtr & func, const DataTypePtr & current_type, const FieldRef & field) { /// Fallback for fields without block reference. @@ -940,164 +901,92 @@ static FieldRef applyFunction(const FunctionBasePtr & func, const DataTypePtr & return {field.columns, field.row_idx, result_idx}; } -/** When table's key has expression with these functions from a column, - * and when a column in a query is compared with a constant, such as: - * CREATE TABLE (x String) ORDER BY toDate(x) - * SELECT ... WHERE x LIKE 'Hello%' - * we want to apply the function to the constant for index analysis, - * but should modify it to pass on un-parsable values. - */ -static std::set date_time_parsing_functions = { - "toDate", - "toDate32", - "toDateTime", - "toDateTime64", - "parseDateTimeBestEffort", - "parseDateTimeBestEffortUS", - "parseDateTime32BestEffort", - "parseDateTime64BestEffort", - "parseDateTime", - "parseDateTimeInJodaSyntax", -}; - -/** The key functional expression constraint may be inferred from a plain column in the expression. - * For example, if the key contains `toStartOfHour(Timestamp)` and query contains `WHERE Timestamp >= now()`, - * it can be assumed that if `toStartOfHour()` is monotonic on [now(), inf), the `toStartOfHour(Timestamp) >= toStartOfHour(now())` - * condition also holds, so the index may be used to select only parts satisfying this condition. - * - * To check the assumption, we'd need to assert that the inverse function to this transformation is also monotonic, however the - * inversion isn't exported (or even viable for not strictly monotonic functions such as `toStartOfHour()`). - * Instead, we can qualify only functions that do not transform the range (for example rounding), - * which while not strictly monotonic, are monotonic everywhere on the input range. - */ -bool KeyCondition::transformConstantWithValidFunctions( - ContextPtr context, - const String & expr_name, - size_t & out_key_column_num, - DataTypePtr & out_key_column_type, - Field & out_value, - DataTypePtr & out_type, - std::function always_monotonic) const +/// Sequentially applies functions to the column, returns `true` +/// if all function arguments are compatible with functions +/// signatures, and none of the functions produce `NULL` output. +/// +/// After functions chain execution, fills result column and its type. +bool applyFunctionChainToColumn( + const ColumnPtr & in_column, + const DataTypePtr & in_data_type, + const std::vector & functions, + ColumnPtr & out_column, + DataTypePtr & out_data_type) { - const auto & sample_block = key_expr->getSampleBlock(); + // Remove LowCardinality from input column, and convert it to regular one + auto result_column = in_column->convertToFullIfNeeded(); + auto result_type = removeLowCardinality(in_data_type); - for (const auto & node : key_expr->getNodes()) + // In case function sequence is empty, return full non-LowCardinality column + if (functions.empty()) { - auto it = key_columns.find(node.result_name); - if (it != key_columns.end()) - { - std::stack chain; - - const auto * cur_node = &node; - bool is_valid_chain = true; - - while (is_valid_chain) - { - if (cur_node->result_name == expr_name) - break; - - chain.push(cur_node); - - if (cur_node->type == ActionsDAG::ActionType::FUNCTION && cur_node->children.size() <= 2) - { - is_valid_chain = always_monotonic(*cur_node->function_base, *cur_node->result_type); - - const ActionsDAG::Node * next_node = nullptr; - for (const auto * arg : cur_node->children) - { - if (arg->column && isColumnConst(*arg->column)) - continue; - - if (next_node) - is_valid_chain = false; - - next_node = arg; - } - - if (!next_node) - is_valid_chain = false; - - cur_node = next_node; - } - else if (cur_node->type == ActionsDAG::ActionType::ALIAS) - cur_node = cur_node->children.front(); - else - is_valid_chain = false; - } - - if (is_valid_chain) - { - out_type = removeLowCardinality(out_type); - auto const_type = removeLowCardinality(cur_node->result_type); - auto const_column = out_type->createColumnConst(1, out_value); - auto const_value = (*castColumnAccurateOrNull({const_column, out_type, ""}, const_type))[0]; - - if (const_value.isNull()) - return false; - - while (!chain.empty()) - { - const auto * func = chain.top(); - chain.pop(); - - if (func->type != ActionsDAG::ActionType::FUNCTION) - continue; - - const auto & func_name = func->function_base->getName(); - auto func_base = func->function_base; - const auto & arg_types = func_base->getArgumentTypes(); - if (date_time_parsing_functions.contains(func_name) && !arg_types.empty() && isStringOrFixedString(arg_types[0])) - { - auto func_or_null = FunctionFactory::instance().get(func_name + "OrNull", context); - ColumnsWithTypeAndName arguments; - int i = 0; - for (const auto & type : func->function_base->getArgumentTypes()) - arguments.push_back({nullptr, type, fmt::format("_{}", i++)}); - - func_base = func_or_null->build(arguments); - } - - if (func->children.size() == 1) - { - std::tie(const_value, const_type) - = applyFunctionForFieldOfUnknownType(func_base, const_type, const_value); - } - else if (func->children.size() == 2) - { - const auto * left = func->children[0]; - const auto * right = func->children[1]; - if (left->column && isColumnConst(*left->column)) - { - auto left_arg_type = left->result_type; - auto left_arg_value = (*left->column)[0]; - std::tie(const_value, const_type) = applyBinaryFunctionForFieldOfUnknownType( - FunctionFactory::instance().get(func_base->getName(), context), - left_arg_type, left_arg_value, const_type, const_value); - } - else - { - auto right_arg_type = right->result_type; - auto right_arg_value = (*right->column)[0]; - std::tie(const_value, const_type) = applyBinaryFunctionForFieldOfUnknownType( - FunctionFactory::instance().get(func_base->getName(), context), - const_type, const_value, right_arg_type, right_arg_value); - } - } - - if (const_value.isNull()) - return false; - } - - out_key_column_num = it->second; - out_key_column_type = sample_block.getByName(it->first).type; - out_value = const_value; - out_type = const_type; - return true; - } - } + out_column = result_column; + out_data_type = result_type; + return true; } - return false; + // If first function arguments are empty, cannot transform input column + if (functions[0]->getArgumentTypes().empty()) + { + return false; + } + + // And cast it to the argument type of the first function in the chain + auto in_argument_type = functions[0]->getArgumentTypes()[0]; + if (canBeSafelyCasted(result_type, in_argument_type)) + { + result_column = castColumnAccurate({result_column, result_type, ""}, in_argument_type); + result_type = in_argument_type; + } + // If column cannot be casted accurate, casting with OrNull, and in case all + // values has been casted (no nulls), unpacking nested column from nullable. + // In case any further functions require Nullable input, they'll be able + // to cast it. + else + { + result_column = castColumnAccurateOrNull({result_column, result_type, ""}, in_argument_type); + const auto & result_column_nullable = assert_cast(*result_column); + const auto & null_map_data = result_column_nullable.getNullMapData(); + for (char8_t i : null_map_data) + { + if (i != 0) + return false; + } + result_column = result_column_nullable.getNestedColumnPtr(); + result_type = removeNullable(in_argument_type); + } + + for (const auto & func : functions) + { + if (func->getArgumentTypes().empty()) + return false; + + auto argument_type = func->getArgumentTypes()[0]; + if (!canBeSafelyCasted(result_type, argument_type)) + return false; + + result_column = castColumnAccurate({result_column, result_type, ""}, argument_type); + result_column = func->execute({{result_column, argument_type, ""}}, func->getResultType(), result_column->size()); + result_type = func->getResultType(); + + // Transforming nullable columns to the nested ones, in case no nulls found + if (result_column->isNullable()) + { + const auto & result_column_nullable = assert_cast(*result_column); + const auto & null_map_data = result_column_nullable.getNullMapData(); + for (char8_t i : null_map_data) + { + if (i != 0) + return false; + } + result_column = result_column_nullable.getNestedColumnPtr(); + result_type = removeNullable(func->getResultType()); + } + } + out_column = result_column; + out_data_type = result_type; + + return true; } bool KeyCondition::canConstantBeWrappedByMonotonicFunctions( @@ -1118,13 +1007,13 @@ bool KeyCondition::canConstantBeWrappedByMonotonicFunctions( if (out_value.isNull()) return false; - return transformConstantWithValidFunctions( + MonotonicFunctionsChain transform_functions; + auto can_transform_constant = extractMonotonicFunctionsChainFromKey( node.getTreeContext().getQueryContext(), expr_name, out_key_column_num, out_key_column_type, - out_value, - out_type, + transform_functions, [](const IFunctionBase & func, const IDataType & type) { if (!func.hasInformationAboutMonotonicity()) @@ -1138,6 +1027,27 @@ bool KeyCondition::canConstantBeWrappedByMonotonicFunctions( } return true; }); + + if (!can_transform_constant) + return false; + + auto const_column = out_type->createColumnConst(1, out_value); + + ColumnPtr transformed_const_column; + DataTypePtr transformed_const_type; + bool constant_transformed = applyFunctionChainToColumn( + const_column, + out_type, + transform_functions, + transformed_const_column, + transformed_const_type); + + if (!constant_transformed) + return false; + + out_value = (*transformed_const_column)[0]; + out_type = transformed_const_type; + return true; } /// Looking for possible transformation of `column = constant` into `partition_expr = function(constant)` @@ -1173,28 +1083,48 @@ bool KeyCondition::canConstantBeWrappedByFunctions( if (out_value.isNull()) return false; - return transformConstantWithValidFunctions( + MonotonicFunctionsChain transform_functions; + auto can_transform_constant = extractMonotonicFunctionsChainFromKey( node.getTreeContext().getQueryContext(), expr_name, out_key_column_num, out_key_column_type, - out_value, + transform_functions, + [](const IFunctionBase & func, const IDataType &) { return func.isDeterministic(); }); + + if (!can_transform_constant) + return false; + + auto const_column = out_type->createColumnConst(1, out_value); + + ColumnPtr transformed_const_column; + DataTypePtr transformed_const_type; + bool constant_transformed = applyFunctionChainToColumn( + const_column, out_type, - [](const IFunctionBase & func, const IDataType &) - { - return func.isDeterministic(); - }); + transform_functions, + transformed_const_column, + transformed_const_type); + + if (!constant_transformed) + return false; + + out_value = (*transformed_const_column)[0]; + out_type = transformed_const_type; + return true; } bool KeyCondition::tryPrepareSetIndex( const RPNBuilderFunctionTreeNode & func, RPNElement & out, - size_t & out_key_column_num) + size_t & out_key_column_num, + bool & is_constant_transformed) { const auto & left_arg = func.getArgumentAt(0); out_key_column_num = 0; std::vector indexes_mapping; + std::vector set_transforming_chains; DataTypes data_types; auto get_key_tuple_position_mapping = [&](const RPNBuilderTreeNode & node, size_t tuple_index) @@ -1203,6 +1133,7 @@ bool KeyCondition::tryPrepareSetIndex( index_mapping.tuple_index = tuple_index; DataTypePtr data_type; std::optional key_space_filling_curve_argument_pos; + MonotonicFunctionsChain set_transforming_chain; if (isKeyPossiblyWrappedByMonotonicFunctions( node, index_mapping.key_index, key_space_filling_curve_argument_pos, data_type, index_mapping.functions) && !key_space_filling_curve_argument_pos) /// We don't support the analysis of space-filling curves and IN set. @@ -1210,6 +1141,15 @@ bool KeyCondition::tryPrepareSetIndex( indexes_mapping.push_back(index_mapping); data_types.push_back(data_type); out_key_column_num = std::max(out_key_column_num, index_mapping.key_index); + set_transforming_chains.push_back(set_transforming_chain); + } + // For partition index, checking if set can be transformed to prune any partitions + else if (single_point && canSetValuesBeWrappedByFunctions(node, index_mapping.key_index, data_type, set_transforming_chain)) + { + indexes_mapping.push_back(index_mapping); + data_types.push_back(data_type); + out_key_column_num = std::max(out_key_column_num, index_mapping.key_index); + set_transforming_chains.push_back(set_transforming_chain); } }; @@ -1275,6 +1215,23 @@ bool KeyCondition::tryPrepareSetIndex( auto set_element_type = set_types[set_element_index]; auto set_column = set_columns[set_element_index]; + if (!set_transforming_chains[indexes_mapping_index].empty()) + { + ColumnPtr transformed_set_column; + DataTypePtr transformed_set_type; + if (!applyFunctionChainToColumn( + set_column, + set_element_type, + set_transforming_chains[indexes_mapping_index], + transformed_set_column, + transformed_set_type)) + return false; + + set_column = transformed_set_column; + set_element_type = transformed_set_type; + is_constant_transformed = true; + } + if (canBeSafelyCasted(set_element_type, key_column_type)) { set_columns[set_element_index] = castColumn({set_column, set_element_type, {}}, key_column_type); @@ -1571,6 +1528,191 @@ bool KeyCondition::isKeyPossiblyWrappedByMonotonicFunctionsImpl( return false; } +/** When table's key has expression with these functions from a column, + * and when a column in a query is compared with a constant, such as: + * CREATE TABLE (x String) ORDER BY toDate(x) + * SELECT ... WHERE x LIKE 'Hello%' + * we want to apply the function to the constant for index analysis, + * but should modify it to pass on un-parsable values. + */ +static std::set date_time_parsing_functions = { + "toDate", + "toDate32", + "toDateTime", + "toDateTime64", + "parseDateTimeBestEffort", + "parseDateTimeBestEffortUS", + "parseDateTime32BestEffort", + "parseDateTime64BestEffort", + "parseDateTime", + "parseDateTimeInJodaSyntax", +}; + +/** The key functional expression constraint may be inferred from a plain column in the expression. + * For example, if the key contains `toStartOfHour(Timestamp)` and query contains `WHERE Timestamp >= now()`, + * it can be assumed that if `toStartOfHour()` is monotonic on [now(), inf), the `toStartOfHour(Timestamp) >= toStartOfHour(now())` + * condition also holds, so the index may be used to select only parts satisfying this condition. + * + * To check the assumption, we'd need to assert that the inverse function to this transformation is also monotonic, however the + * inversion isn't exported (or even viable for not strictly monotonic functions such as `toStartOfHour()`). + * Instead, we can qualify only functions that do not transform the range (for example rounding), + * which while not strictly monotonic, are monotonic everywhere on the input range. + */ +bool KeyCondition::extractMonotonicFunctionsChainFromKey( + ContextPtr context, + const String & expr_name, + size_t & out_key_column_num, + DataTypePtr & out_key_column_type, + MonotonicFunctionsChain & out_functions_chain, + std::function always_monotonic) const +{ + const auto & sample_block = key_expr->getSampleBlock(); + + for (const auto & node : key_expr->getNodes()) + { + auto it = key_columns.find(node.result_name); + if (it != key_columns.end()) + { + std::stack chain; + + const auto * cur_node = &node; + bool is_valid_chain = true; + + while (is_valid_chain) + { + if (cur_node->result_name == expr_name) + break; + + chain.push(cur_node); + + if (cur_node->type == ActionsDAG::ActionType::FUNCTION && cur_node->children.size() <= 2) + { + is_valid_chain = always_monotonic(*cur_node->function_base, *cur_node->result_type); + + const ActionsDAG::Node * next_node = nullptr; + for (const auto * arg : cur_node->children) + { + if (arg->column && isColumnConst(*arg->column)) + continue; + + if (next_node) + is_valid_chain = false; + + next_node = arg; + } + + if (!next_node) + is_valid_chain = false; + + cur_node = next_node; + } + else if (cur_node->type == ActionsDAG::ActionType::ALIAS) + cur_node = cur_node->children.front(); + else + is_valid_chain = false; + } + + if (is_valid_chain) + { + while (!chain.empty()) + { + const auto * func = chain.top(); + chain.pop(); + + if (func->type != ActionsDAG::ActionType::FUNCTION) + continue; + + auto func_name = func->function_base->getName(); + auto func_base = func->function_base; + + ColumnsWithTypeAndName arguments; + ColumnWithTypeAndName const_arg; + FunctionWithOptionalConstArg::Kind kind = FunctionWithOptionalConstArg::Kind::NO_CONST; + + if (date_time_parsing_functions.contains(func_name)) + { + const auto & arg_types = func_base->getArgumentTypes(); + if (!arg_types.empty() && isStringOrFixedString(arg_types[0])) + { + func_name = func_name + "OrNull"; + } + + } + + auto func_builder = FunctionFactory::instance().tryGet(func_name, context); + + if (func->children.size() == 1) + { + arguments.push_back({nullptr, removeLowCardinality(func->children[0]->result_type), ""}); + } + else if (func->children.size() == 2) + { + const auto * left = func->children[0]; + const auto * right = func->children[1]; + if (left->column && isColumnConst(*left->column)) + { + const_arg = {left->result_type->createColumnConst(0, (*left->column)[0]), left->result_type, ""}; + arguments.push_back(const_arg); + arguments.push_back({nullptr, removeLowCardinality(right->result_type), ""}); + kind = FunctionWithOptionalConstArg::Kind::LEFT_CONST; + } + else + { + const_arg = {right->result_type->createColumnConst(0, (*right->column)[0]), right->result_type, ""}; + arguments.push_back({nullptr, removeLowCardinality(left->result_type), ""}); + arguments.push_back(const_arg); + kind = FunctionWithOptionalConstArg::Kind::RIGHT_CONST; + } + } + + auto out_func = func_builder->build(arguments); + if (kind == FunctionWithOptionalConstArg::Kind::NO_CONST) + out_functions_chain.push_back(out_func); + else + out_functions_chain.push_back(std::make_shared(out_func, const_arg, kind)); + } + + out_key_column_num = it->second; + out_key_column_type = sample_block.getByName(it->first).type; + return true; + } + } + } + + return false; +} + +bool KeyCondition::canSetValuesBeWrappedByFunctions( + const RPNBuilderTreeNode & node, + size_t & out_key_column_num, + DataTypePtr & out_key_res_column_type, + MonotonicFunctionsChain & out_functions_chain) +{ + // Checking if column name matches any of key subexpressions + String expr_name = node.getColumnName(); + + if (array_joined_column_names.contains(expr_name)) + return false; + + if (!key_subexpr_names.contains(expr_name)) + { + expr_name = node.getColumnNameWithModuloLegacy(); + + if (!key_subexpr_names.contains(expr_name)) + return false; + } + + return extractMonotonicFunctionsChainFromKey( + node.getTreeContext().getQueryContext(), + expr_name, + out_key_column_num, + out_key_res_column_type, + out_functions_chain, + [](const IFunctionBase & func, const IDataType &) + { + return func.isDeterministic(); + }); +} static void castValueToType(const DataTypePtr & desired_type, Field & src_value, const DataTypePtr & src_type, const String & node_column_name) { @@ -1649,7 +1791,7 @@ bool KeyCondition::extractAtomFromTree(const RPNBuilderTreeNode & node, RPNEleme if (functionIsInOrGlobalInOperator(func_name)) { - if (tryPrepareSetIndex(func, out, key_column_num)) + if (tryPrepareSetIndex(func, out, key_column_num, is_constant_transformed)) { key_arg_pos = 0; is_set_const = true; diff --git a/src/Storages/MergeTree/KeyCondition.h b/src/Storages/MergeTree/KeyCondition.h index 6e5956706aa..9e2218d7a29 100644 --- a/src/Storages/MergeTree/KeyCondition.h +++ b/src/Storages/MergeTree/KeyCondition.h @@ -14,6 +14,7 @@ #include #include +#include "DataTypes/Serializations/ISerialization.h" namespace DB @@ -253,13 +254,12 @@ private: DataTypePtr & out_key_column_type, std::vector & out_functions_chain); - bool transformConstantWithValidFunctions( + bool extractMonotonicFunctionsChainFromKey( ContextPtr context, const String & expr_name, size_t & out_key_column_num, DataTypePtr & out_key_column_type, - Field & out_value, - DataTypePtr & out_type, + MonotonicFunctionsChain & out_functions_chain, std::function always_monotonic) const; bool canConstantBeWrappedByMonotonicFunctions( @@ -276,13 +276,25 @@ private: Field & out_value, DataTypePtr & out_type); + /// Checks if node is a subexpression of any of key columns expressions, + /// wrapped by deterministic functions, and if so, returns `true`, and + /// specifies key column position / type. Besides that it produces the + /// chain of functions which should be executed on set, to transform it + /// into key column values. + bool canSetValuesBeWrappedByFunctions( + const RPNBuilderTreeNode & node, + size_t & out_key_column_num, + DataTypePtr & out_key_res_column_type, + MonotonicFunctionsChain & out_functions_chain); + /// If it's possible to make an RPNElement /// that will filter values (possibly tuples) by the content of 'prepared_set', /// do it and return true. bool tryPrepareSetIndex( const RPNBuilderFunctionTreeNode & func, RPNElement & out, - size_t & out_key_column_num); + size_t & out_key_column_num, + bool & is_constant_transformed); /// Checks that the index can not be used. /// diff --git a/src/Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h b/src/Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h index f5111ccaacc..aff1cf0edb0 100644 --- a/src/Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h +++ b/src/Storages/MergeTree/LoadedMergeTreeDataPartInfoForReader.h @@ -36,7 +36,10 @@ public: AlterConversionsPtr getAlterConversions() const override { return alter_conversions; } - String getColumnNameWithMinimumCompressedSize(bool with_subcolumns) const override { return data_part->getColumnNameWithMinimumCompressedSize(with_subcolumns); } + String getColumnNameWithMinimumCompressedSize(const NamesAndTypesList & available_columns) const override + { + return data_part->getColumnNameWithMinimumCompressedSize(available_columns); + } const MergeTreeDataPartChecksums & getChecksums() const override { return data_part->checksums; } diff --git a/src/Storages/MergeTree/MergeFromLogEntryTask.cpp b/src/Storages/MergeTree/MergeFromLogEntryTask.cpp index 79efb0ca8b3..4c947487d21 100644 --- a/src/Storages/MergeTree/MergeFromLogEntryTask.cpp +++ b/src/Storages/MergeTree/MergeFromLogEntryTask.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/MergeTree/MergeTask.cpp b/src/Storages/MergeTree/MergeTask.cpp index 7ab8fa2430a..fc64fae9a58 100644 --- a/src/Storages/MergeTree/MergeTask.cpp +++ b/src/Storages/MergeTree/MergeTask.cpp @@ -7,6 +7,7 @@ #include #include +#include #include #include #include @@ -16,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -422,6 +424,16 @@ bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare() return false; } +bool MergeTask::enabledBlockNumberColumn(GlobalRuntimeContextPtr global_ctx) +{ + return global_ctx->data->getSettings()->enable_block_number_column && global_ctx->metadata_snapshot->getGroupByTTLs().empty(); +} + +bool MergeTask::enabledBlockOffsetColumn(GlobalRuntimeContextPtr global_ctx) +{ + return global_ctx->data->getSettings()->enable_block_offset_column && global_ctx->metadata_snapshot->getGroupByTTLs().empty(); +} + void MergeTask::addGatheringColumn(GlobalRuntimeContextPtr global_ctx, const String & name, const DataTypePtr & type) { if (global_ctx->storage_columns.contains(name)) @@ -555,18 +567,18 @@ bool MergeTask::VerticalMergeStage::prepareVerticalMergeForAllColumns() const if (!reread_buf) throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot read temporary file {}", ctx->rows_sources_uncompressed_write_buf->getFileName()); - auto * reread_buffer_raw = dynamic_cast(reread_buf.get()); + auto * reread_buffer_raw = dynamic_cast(reread_buf.get()); if (!reread_buffer_raw) { const auto & reread_buf_ref = *reread_buf; - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected ReadBufferFromFile, but got {}", demangle(typeid(reread_buf_ref).name())); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected ReadBufferFromFileBase, but got {}", demangle(typeid(reread_buf_ref).name())); } /// Move ownership from std::unique_ptr to std::unique_ptr for CompressedReadBufferFromFile. /// First, release ownership from unique_ptr to base type. reread_buf.release(); /// NOLINT(bugprone-unused-return-value,hicpp-ignored-remove-result): we already have the pointer value in `reread_buffer_raw` /// Then, move ownership to unique_ptr to concrete type. - std::unique_ptr reread_buffer_from_file(reread_buffer_raw); + std::unique_ptr reread_buffer_from_file(reread_buffer_raw); /// CompressedReadBufferFromFile expects std::unique_ptr as argument. ctx->rows_sources_read_buf = std::make_unique(std::move(reread_buffer_from_file)); diff --git a/src/Storages/MergeTree/MergeTask.h b/src/Storages/MergeTree/MergeTask.h index 56909d1b7a0..8b0f2130e8e 100644 --- a/src/Storages/MergeTree/MergeTask.h +++ b/src/Storages/MergeTree/MergeTask.h @@ -408,15 +408,8 @@ private: Stages::const_iterator stages_iterator = stages.begin(); - static bool enabledBlockNumberColumn(GlobalRuntimeContextPtr global_ctx) - { - return global_ctx->data->getSettings()->enable_block_number_column && global_ctx->metadata_snapshot->getGroupByTTLs().empty(); - } - - static bool enabledBlockOffsetColumn(GlobalRuntimeContextPtr global_ctx) - { - return global_ctx->data->getSettings()->enable_block_offset_column && global_ctx->metadata_snapshot->getGroupByTTLs().empty(); - } + static bool enabledBlockNumberColumn(GlobalRuntimeContextPtr global_ctx); + static bool enabledBlockOffsetColumn(GlobalRuntimeContextPtr global_ctx); static void addGatheringColumn(GlobalRuntimeContextPtr global_ctx, const String & name, const DataTypePtr & type); }; diff --git a/src/Storages/MergeTree/MergeTreeBackgroundExecutor.cpp b/src/Storages/MergeTree/MergeTreeBackgroundExecutor.cpp index 8cb0badc19b..4a4deb07eee 100644 --- a/src/Storages/MergeTree/MergeTreeBackgroundExecutor.cpp +++ b/src/Storages/MergeTree/MergeTreeBackgroundExecutor.cpp @@ -155,6 +155,10 @@ void printExceptionWithRespectToAbort(LoggerPtr log, const String & query_id) { std::rethrow_exception(ex); } + catch (const TestException &) // NOLINT + { + /// Exception from a unit test, ignore it. + } catch (const Exception & e) { NOEXCEPT_SCOPE({ diff --git a/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp b/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp index 570387a7046..aaa4ecd8eee 100644 --- a/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp +++ b/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp @@ -127,7 +127,8 @@ NameSet injectRequiredColumns( */ if (!have_at_least_one_physical_column) { - const auto minimum_size_column_name = data_part_info_for_reader.getColumnNameWithMinimumCompressedSize(with_subcolumns); + auto available_columns = storage_snapshot->metadata->getColumns().get(options); + const auto minimum_size_column_name = data_part_info_for_reader.getColumnNameWithMinimumCompressedSize(available_columns); columns.push_back(minimum_size_column_name); /// correctly report added column injected_columns.insert(columns.back()); diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp index f9cc65871fe..30a4a7caa0f 100644 --- a/src/Storages/MergeTree/MergeTreeData.cpp +++ b/src/Storages/MergeTree/MergeTreeData.cpp @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -73,6 +74,7 @@ #include #include #include +#include #include #include #include @@ -87,6 +89,7 @@ #include #include +#include #include #include @@ -283,13 +286,12 @@ void MergeTreeData::initializeDirectoriesAndFormatVersion(const std::string & re } } - - // When data path or file not exists, ignore the format_version check + /// When data path or file not exists, ignore the format_version check if (!attach || !read_format_version) { format_version = min_format_version; - // try to write to first non-readonly disk + /// Try to write to first non-readonly disk for (const auto & disk : getStoragePolicy()->getDisks()) { if (disk->isBroken()) @@ -474,7 +476,7 @@ StoragePolicyPtr MergeTreeData::getStoragePolicy() const ConditionSelectivityEstimator MergeTreeData::getConditionSelectivityEstimatorByPredicate( const StorageSnapshotPtr & storage_snapshot, const ActionsDAGPtr & filter_dag, ContextPtr local_context) const { - if (!local_context->getSettings().allow_statistics_optimize) + if (!local_context->getSettingsRef().allow_statistics_optimize) return {}; const auto & parts = assert_cast(*storage_snapshot->data).parts; @@ -497,8 +499,9 @@ ConditionSelectivityEstimator MergeTreeData::getConditionSelectivityEstimatorByP { auto stats = part->loadStatistics(); /// TODO: We only have one stats file for every part. + result.addRows(part->rows_count); for (const auto & stat : stats) - result.merge(part->info.getPartNameV1(), part->rows_count, stat); + result.merge(part->info.getPartNameV1(), stat); } catch (...) { @@ -513,8 +516,9 @@ ConditionSelectivityEstimator MergeTreeData::getConditionSelectivityEstimatorByP if (!partition_pruner.canBePruned(*part)) { auto stats = part->loadStatistics(); + result.addRows(part->rows_count); for (const auto & stat : stats) - result.merge(part->info.getPartNameV1(), part->rows_count, stat); + result.merge(part->info.getPartNameV1(), stat); } } catch (...) @@ -1142,7 +1146,7 @@ std::optional MergeTreeData::totalRowsByPartitionPredicateImpl( auto metadata_snapshot = getInMemoryMetadataPtr(); auto virtual_columns_block = getBlockWithVirtualsForFilter(metadata_snapshot, {parts[0]}); - auto filter_dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag->getOutputs().at(0), nullptr); + auto filter_dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag->getOutputs().at(0), nullptr, /*allow_non_deterministic_functions=*/ false); if (!filter_dag) return {}; @@ -1759,11 +1763,14 @@ void MergeTreeData::loadDataParts(bool skip_sanity_checks, std::optional runner(getActivePartsLoadingThreadPool().get(), "ActiveParts"); + bool all_disks_are_readonly = true; for (size_t i = 0; i < disks.size(); ++i) { const auto & disk_ptr = disks[i]; if (disk_ptr->isBroken()) continue; + if (!disk_ptr->isReadOnly()) + all_disks_are_readonly = false; auto & disk_parts = parts_to_load_by_disk[i]; auto & unexpected_disk_parts = unexpected_parts_to_load_by_disk[i]; @@ -1916,7 +1923,6 @@ void MergeTreeData::loadDataParts(bool skip_sanity_checks, std::optionalrenameToDetached("broken-on-start"); /// detached parts must not have '_' in prefixes @@ -1961,7 +1967,8 @@ void MergeTreeData::loadDataParts(bool skip_sanity_checks, std::optionalremoval_state.store(DataPartRemovalState::NON_UNIQUE_OWNERSHIP, std::memory_order_relaxed); skipped_parts.push_back(part->info); @@ -4358,13 +4365,13 @@ bool MergeTreeData::tryRemovePartImmediately(DataPartPtr && part) part.reset(); - if (!((*it)->getState() == DataPartState::Outdated && it->unique())) + if (!((*it)->getState() == DataPartState::Outdated && isSharedPtrUnique(*it))) { if ((*it)->getState() != DataPartState::Outdated) LOG_WARNING(log, "Cannot immediately remove part {} because it's not in Outdated state " "usage counter {}", part_name_with_state, it->use_count()); - if (!it->unique()) + if (!isSharedPtrUnique(*it)) LOG_WARNING(log, "Cannot immediately remove part {} because someone using it right now " "usage counter {}", part_name_with_state, it->use_count()); return false; @@ -4430,7 +4437,7 @@ size_t MergeTreeData::getNumberOfOutdatedPartsWithExpiredRemovalTime() const for (const auto & part : outdated_parts_range) { auto part_remove_time = part->remove_time.load(std::memory_order_relaxed); - if (part_remove_time <= time_now && time_now - part_remove_time >= getSettings()->old_parts_lifetime.totalSeconds() && part.unique()) + if (part_remove_time <= time_now && time_now - part_remove_time >= getSettings()->old_parts_lifetime.totalSeconds() && isSharedPtrUnique(part)) ++res; } @@ -5544,12 +5551,17 @@ public: attachIfAllPartsRestored(); } - String getTemporaryDirectory(const DiskPtr & disk) + String getTemporaryDirectory(const DiskPtr & disk, const String & part_name) { std::lock_guard lock{mutex}; - auto it = temp_dirs.find(disk); - if (it == temp_dirs.end()) - it = temp_dirs.emplace(disk, std::make_shared(disk, "tmp/")).first; + auto it = temp_part_dirs.find(part_name); + if (it == temp_part_dirs.end()) + { + auto temp_part_dir = std::make_shared(disk, fs::path{storage->getRelativeDataPath()} / ("tmp_restore_" + part_name + "-")); + /// Attaching parts will rename them so it's expected for a temporary part directory not to exist anymore in the end. + temp_part_dir->setShowWarningIfRemoved(false); + it = temp_part_dirs.emplace(part_name, temp_part_dir).first; + } return it->second->getRelativePath(); } @@ -5567,7 +5579,7 @@ private: storage->attachRestoredParts(std::move(parts)); parts.clear(); - temp_dirs.clear(); + temp_part_dirs.clear(); num_parts = 0; } @@ -5576,7 +5588,7 @@ private: size_t num_parts = 0; size_t num_broken_parts = 0; MutableDataPartsVector parts; - std::map> temp_dirs; + std::map> temp_part_dirs; mutable std::mutex mutex; }; @@ -5629,9 +5641,11 @@ void MergeTreeData::restorePartFromBackup(std::shared_ptr r String part_name = part_info.getPartNameAndCheckFormat(format_version); auto backup = restored_parts_holder->getBackup(); + /// Find all files of this part in the backup. + Strings filenames = backup->listFiles(part_path_in_backup, /* recursive= */ true); + /// Calculate the total size of the part. UInt64 total_size_of_part = 0; - Strings filenames = backup->listFiles(part_path_in_backup, /* recursive= */ true); fs::path part_path_in_backup_fs = part_path_in_backup; for (const String & filename : filenames) total_size_of_part += backup->getFileSize(part_path_in_backup_fs / filename); @@ -5641,11 +5655,9 @@ void MergeTreeData::restorePartFromBackup(std::shared_ptr r /// Calculate paths, for example: /// part_name = 0_1_1_0 /// part_path_in_backup = /data/test/table/0_1_1_0 - /// tmp_dir = tmp/1aaaaaa - /// tmp_part_dir = tmp/1aaaaaa/data/test/table/0_1_1_0 + /// temp_part_dir = /var/lib/clickhouse/data/test/table/tmp_restore_all_0_1_1_0-XXXXXXXX auto disk = reservation->getDisk(); - fs::path temp_dir = restored_parts_holder->getTemporaryDirectory(disk); - fs::path temp_part_dir = temp_dir / part_path_in_backup_fs.relative_path(); + fs::path temp_part_dir = restored_parts_holder->getTemporaryDirectory(disk, part_name); /// Subdirectories in the part's directory. It's used to restore projections. std::unordered_set subdirs; @@ -5672,22 +5684,25 @@ void MergeTreeData::restorePartFromBackup(std::shared_ptr r reservation->update(reservation->getSize() - file_size); } - if (auto part = loadPartRestoredFromBackup(disk, temp_part_dir.parent_path(), part_name, detach_if_broken)) + if (auto part = loadPartRestoredFromBackup(part_name, disk, temp_part_dir, detach_if_broken)) restored_parts_holder->addPart(part); else restored_parts_holder->increaseNumBrokenParts(); } -MergeTreeData::MutableDataPartPtr MergeTreeData::loadPartRestoredFromBackup(const DiskPtr & disk, const String & temp_dir, const String & part_name, bool detach_if_broken) const +MergeTreeData::MutableDataPartPtr MergeTreeData::loadPartRestoredFromBackup(const String & part_name, const DiskPtr & disk, const String & temp_part_dir, bool detach_if_broken) const { MutableDataPartPtr part; auto single_disk_volume = std::make_shared(disk->getName(), disk, 0); + fs::path full_part_dir{temp_part_dir}; + String parent_part_dir = full_part_dir.parent_path(); + String part_dir_name = full_part_dir.filename(); - /// Load this part from the directory `tmp_part_dir`. + /// Load this part from the directory `temp_part_dir`. auto load_part = [&] { - MergeTreeDataPartBuilder builder(*this, part_name, single_disk_volume, temp_dir, part_name); + MergeTreeDataPartBuilder builder(*this, part_name, single_disk_volume, parent_part_dir, part_dir_name); builder.withPartFormatFromDisk(); part = std::move(builder).build(); part->version.setCreationTID(Tx::PrehistoricTID, nullptr); @@ -5702,7 +5717,7 @@ MergeTreeData::MutableDataPartPtr MergeTreeData::loadPartRestoredFromBackup(cons if (!part) { /// Make a fake data part only to copy its files to /detached/. - part = MergeTreeDataPartBuilder{*this, part_name, single_disk_volume, temp_dir, part_name} + part = MergeTreeDataPartBuilder{*this, part_name, single_disk_volume, parent_part_dir, part_dir_name} .withPartStorageType(MergeTreeDataPartStorageType::Full) .withPartType(MergeTreeDataPartType::Wide) .build(); @@ -6170,6 +6185,11 @@ bool MergeTreeData::hasProjection() const return false; } +bool MergeTreeData::areAsynchronousInsertsEnabled() const +{ + return getSettings()->async_insert; +} + MergeTreeData::ProjectionPartsVector MergeTreeData::getAllProjectionPartsVector(MergeTreeData::DataPartStateVector * out_states) const { ProjectionPartsVector res; @@ -7075,6 +7095,20 @@ QueryProcessingStage::Enum MergeTreeData::getQueryProcessingStage( /// with new analyzer, Planner make decision regarding parallel replicas usage, and so about processing stage on reading if (!query_context->getSettingsRef().allow_experimental_analyzer) { + const auto & settings = query_context->getSettingsRef(); + if (query_context->canUseParallelReplicasCustomKey()) + { + if (query_context->getClientInfo().distributed_depth > 0) + return QueryProcessingStage::FetchColumns; + + if (!supportsReplication() && !settings.parallel_replicas_for_non_replicated_merge_tree) + return QueryProcessingStage::Enum::FetchColumns; + + if (to_stage >= QueryProcessingStage::WithMergeableState + && query_context->canUseParallelReplicasCustomKeyForCluster(*query_context->getClusterForParallelReplicas())) + return QueryProcessingStage::WithMergeableStateAfterAggregationAndLimit; + } + if (query_context->getClientInfo().collaborate_with_initiator) return QueryProcessingStage::Enum::FetchColumns; @@ -7086,7 +7120,7 @@ QueryProcessingStage::Enum MergeTreeData::getQueryProcessingStage( return QueryProcessingStage::Enum::WithMergeableState; /// For non-replicated MergeTree we allow them only if parallel_replicas_for_non_replicated_merge_tree is enabled - if (query_context->getSettingsRef().parallel_replicas_for_non_replicated_merge_tree) + if (settings.parallel_replicas_for_non_replicated_merge_tree) return QueryProcessingStage::Enum::WithMergeableState; } } @@ -7111,8 +7145,8 @@ UInt64 MergeTreeData::estimateNumberOfRowsToRead( query_context->getSettingsRef().max_threads); UInt64 total_rows = result_ptr->selected_rows; - if (query_info.limit > 0 && query_info.limit < total_rows) - total_rows = query_info.limit; + if (query_info.trivial_limit > 0 && query_info.trivial_limit < total_rows) + total_rows = query_info.trivial_limit; return total_rows; } @@ -7377,6 +7411,13 @@ std::pair MergeTreeData::cloneAn return std::make_pair(dst_data_part, std::move(temporary_directory_lock)); } +bool MergeTreeData::canUseAdaptiveGranularity() const +{ + const auto settings = getSettings(); + return settings->index_granularity_bytes != 0 + && (settings->enable_mixed_granularity_parts || !has_non_adaptive_index_granularity_parts); +} + String MergeTreeData::getFullPathOnDisk(const DiskPtr & disk) const { return disk->getPath() + relative_data_path; @@ -8606,6 +8647,38 @@ void MergeTreeData::unloadPrimaryKeys() } } +size_t MergeTreeData::unloadPrimaryKeysOfOutdatedParts() +{ + /// If the method is already called from another thread, then we don't need to do anything. + std::unique_lock lock(unload_primary_key_mutex, std::defer_lock); + if (!lock.try_lock()) + return 0; + + DataPartsVector parts_to_unload_index; + + { + auto parts_lock = lockParts(); + auto parts_range = getDataPartsStateRange(DataPartState::Outdated); + + for (const auto & part : parts_range) + { + /// Outdated part may be hold by SELECT query and still needs the index. + /// This check requires lock of index_mutex but if outdated part is unique then there is no + /// contention on it, so it's relatively cheap and it's ok to check under a global parts lock. + if (isSharedPtrUnique(part) && part->isIndexLoaded()) + parts_to_unload_index.push_back(part); + } + } + + for (const auto & part : parts_to_unload_index) + { + const_cast(*part).unloadIndex(); + LOG_TEST(log, "Unloaded primary key for outdated part {}", part->name); + } + + return parts_to_unload_index.size(); +} + void MergeTreeData::verifySortingKey(const KeyDescription & sorting_key) { /// Aggregate functions already forbidden, but SimpleAggregateFunction are not diff --git a/src/Storages/MergeTree/MergeTreeData.h b/src/Storages/MergeTree/MergeTreeData.h index c6f736a4afd..d185a5262fd 100644 --- a/src/Storages/MergeTree/MergeTreeData.h +++ b/src/Storages/MergeTree/MergeTreeData.h @@ -18,7 +18,6 @@ #include #include #include -#include #include #include #include @@ -65,6 +64,7 @@ using BackupEntries = std::vector; +struct MergeTreeSettings; struct WriteSettings; /// Auxiliary struct holding information about the future merged or mutated part. @@ -441,7 +441,7 @@ public: bool hasProjection() const override; - bool areAsynchronousInsertsEnabled() const override { return getSettings()->async_insert; } + bool areAsynchronousInsertsEnabled() const override; bool supportsTrivialCountOptimization(const StorageSnapshotPtr &, ContextPtr) const override; @@ -860,12 +860,7 @@ public: /// Returns true if table can create new parts with adaptive granularity /// Has additional constraint in replicated version - virtual bool canUseAdaptiveGranularity() const - { - const auto settings = getSettings(); - return settings->index_granularity_bytes != 0 && - (settings->enable_mixed_granularity_parts || !has_non_adaptive_index_granularity_parts); - } + virtual bool canUseAdaptiveGranularity() const; /// Get constant pointer to storage settings. /// Copy this pointer into your scope and you will @@ -1096,8 +1091,13 @@ public: static VirtualColumnsDescription createVirtuals(const StorageInMemoryMetadata & metadata); + /// Unloads primary keys of all parts. void unloadPrimaryKeys(); + /// Unloads primary keys of outdated parts that are not used by any query. + /// Returns the number of parts for which index was unloaded. + size_t unloadPrimaryKeysOfOutdatedParts(); + protected: friend class IMergeTreeDataPart; friend class MergeTreeDataMergerMutator; @@ -1143,7 +1143,7 @@ protected: struct TagByInfo{}; struct TagByStateAndInfo{}; - void initializeDirectoriesAndFormatVersion(const std::string & relative_data_path_, bool attach, const std::string & date_column_name, bool need_create_directories=true); + void initializeDirectoriesAndFormatVersion(const std::string & relative_data_path_, bool attach, const std::string & date_column_name, bool need_create_directories = true); static const MergeTreePartInfo & dataPartPtrToInfo(const DataPartPtr & part) { @@ -1260,6 +1260,8 @@ protected: std::mutex grab_old_parts_mutex; /// The same for clearOldTemporaryDirectories. std::mutex clear_old_temporary_directories_mutex; + /// The same for unloadPrimaryKeysOfOutdatedParts. + std::mutex unload_primary_key_mutex; void checkProperties( const StorageInMemoryMetadata & new_metadata, @@ -1471,7 +1473,7 @@ protected: /// Restores the parts of this table from backup. void restorePartsFromBackup(RestorerFromBackup & restorer, const String & data_path_in_backup, const std::optional & partitions); void restorePartFromBackup(std::shared_ptr restored_parts_holder, const MergeTreePartInfo & part_info, const String & part_path_in_backup, bool detach_if_broken) const; - MutableDataPartPtr loadPartRestoredFromBackup(const DiskPtr & disk, const String & temp_dir, const String & part_name, bool detach_if_broken) const; + MutableDataPartPtr loadPartRestoredFromBackup(const String & part_name, const DiskPtr & disk, const String & temp_part_dir, bool detach_if_broken) const; /// Attaches restored parts to the storage. virtual void attachRestoredParts(MutableDataPartsVector && parts) = 0; @@ -1728,14 +1730,6 @@ struct CurrentlySubmergingEmergingTagger ~CurrentlySubmergingEmergingTagger(); }; - -/// TODO: move it somewhere -[[ maybe_unused ]] static bool needSyncPart(size_t input_rows, size_t input_bytes, const MergeTreeSettings & settings) -{ - return ((settings.min_rows_to_fsync_after_merge && input_rows >= settings.min_rows_to_fsync_after_merge) - || (settings.min_compressed_bytes_to_fsync_after_merge && input_bytes >= settings.min_compressed_bytes_to_fsync_after_merge)); -} - /// Look at MutationCommands if it contains mutations for AlterConversions, update the counter. /// Return true if the counter had been updated bool updateAlterConversionsMutations(const MutationCommands & commands, std::atomic & alter_conversions_mutations, bool remove); diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp index 2d49e1df19b..140a226f2d1 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -117,11 +118,11 @@ UInt64 MergeTreeDataMergerMutator::getMaxSourcePartSizeForMutation() const occupied >= data_settings->max_number_of_mutations_for_replica) return 0; - /// DataPart can be store only at one disk. Get maximum reservable free space at all disks. + /// A DataPart can be stored only at a single disk. Get the maximum reservable free space at all disks. UInt64 disk_space = data.getStoragePolicy()->getMaxUnreservedFreeSpace(); auto max_tasks_count = data.getContext()->getMergeMutateExecutor()->getMaxTasksCount(); - /// Allow mutations only if there are enough threads, leave free threads for merges else + /// Allow mutations only if there are enough threads, otherwise, leave free threads for merges. if (occupied <= 1 || max_tasks_count - occupied >= data_settings->number_of_free_entries_in_pool_to_execute_mutation) return static_cast(disk_space / DISK_USAGE_COEFFICIENT_TO_RESERVE); diff --git a/src/Storages/MergeTree/MergeTreeDataPartWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWide.cpp index 379c465a409..5bab523a9f1 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWide.cpp @@ -266,10 +266,13 @@ void MergeTreeDataPartWide::doCheckConsistency(bool require_part_metadata) const bool MergeTreeDataPartWide::hasColumnFiles(const NameAndTypePair & column) const { + auto serialization = tryGetSerialization(column.name); + if (!serialization) + return false; auto marks_file_extension = index_granularity_info.mark_type.getFileExtension(); bool res = true; - getSerialization(column.name)->enumerateStreams([&](const auto & substream_path) + serialization->enumerateStreams([&](const auto & substream_path) { auto stream_name = getStreamNameForColumn(column, substream_path, checksums); if (!stream_name || !checksums.files.contains(*stream_name + marks_file_extension)) @@ -298,6 +301,11 @@ std::optional MergeTreeDataPartWide::getColumnModificationTime(const Str std::optional MergeTreeDataPartWide::getFileNameForColumn(const NameAndTypePair & column) const { std::optional filename; + + /// Fallback for the case when serializations was not loaded yet (called from loadColumns()) + if (getSerializations().empty()) + return getStreamNameForColumn(column, {}, DATA_FILE_EXTENSION, getDataPartStorage()); + getSerialization(column.name)->enumerateStreams([&](const ISerialization::SubstreamPath & substream_path) { if (!filename.has_value()) @@ -309,6 +317,7 @@ std::optional MergeTreeDataPartWide::getFileNameForColumn(const NameAndT filename = getStreamNameForColumn(column, substream_path, DATA_FILE_EXTENSION, getDataPartStorage()); } }); + return filename; } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp index 21d046c76f2..52d12c9db7d 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp @@ -154,7 +154,8 @@ void writeColumnSingleGranule( const SerializationPtr & serialization, ISerialization::OutputStreamGetter stream_getter, size_t from_row, - size_t number_of_rows) + size_t number_of_rows, + const MergeTreeWriterSettings & settings) { ISerialization::SerializeBinaryBulkStatePtr state; ISerialization::SerializeBinaryBulkSettings serialize_settings; @@ -162,6 +163,7 @@ void writeColumnSingleGranule( serialize_settings.getter = stream_getter; serialize_settings.position_independent_encoding = true; serialize_settings.low_cardinality_max_dictionary_size = 0; + serialize_settings.use_compact_variant_discriminators_serialization = settings.use_compact_variant_discriminators_serialization; serialize_settings.dynamic_write_statistics = ISerialization::SerializeBinaryBulkSettings::DynamicStatisticsMode::PREFIX; serialization->serializeBinaryBulkStatePrefix(*column.column, serialize_settings, state); @@ -259,7 +261,7 @@ void MergeTreeDataPartWriterCompact::writeDataBlock(const Block & block, const G writeColumnSingleGranule( block.getByName(name_and_type->name), getSerialization(name_and_type->name), - stream_getter, granule.start_row, granule.rows_to_write); + stream_getter, granule.start_row, granule.rows_to_write, settings); /// Each type always have at least one substream prev_stream->hashing_buf.next(); diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp index a576720294f..6dc7e649b06 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include @@ -254,6 +255,12 @@ void MergeTreeDataPartWriterOnDisk::initPrimaryIndex() index_compressor_stream = std::make_unique(*index_file_hashing_stream, primary_key_compression_codec, settings.primary_key_compress_block_size); index_source_hashing_stream = std::make_unique(*index_compressor_stream); } + + const auto & primary_key_types = metadata_snapshot->getPrimaryKey().data_types; + index_serializations.reserve(primary_key_types.size()); + + for (const auto & type : primary_key_types) + index_serializations.push_back(type->getDefaultSerialization()); } } @@ -299,22 +306,33 @@ void MergeTreeDataPartWriterOnDisk::initSkipIndices() store = std::make_shared(stream_name, data_part_storage, data_part_storage, storage_settings->max_digestion_size_per_segment); gin_index_stores[stream_name] = store; } + skip_indices_aggregators.push_back(skip_index->createIndexAggregatorForPart(store, settings)); skip_index_accumulated_marks.push_back(0); } } +void MergeTreeDataPartWriterOnDisk::calculateAndSerializePrimaryIndexRow(const Block & index_block, size_t row) +{ + chassert(index_block.columns() == index_serializations.size()); + auto & index_stream = compress_primary_key ? *index_source_hashing_stream : *index_file_hashing_stream; + + for (size_t i = 0; i < index_block.columns(); ++i) + { + const auto & column = index_block.getByPosition(i).column; + + index_columns[i]->insertFrom(*column, row); + index_serializations[i]->serializeBinary(*column, row, index_stream, {}); + } +} + void MergeTreeDataPartWriterOnDisk::calculateAndSerializePrimaryIndex(const Block & primary_index_block, const Granules & granules_to_write) { - size_t primary_columns_num = primary_index_block.columns(); + if (!metadata_snapshot->hasPrimaryKey()) + return; + if (index_columns.empty()) - { - index_types = primary_index_block.getDataTypes(); - index_columns.resize(primary_columns_num); - last_block_index_columns.resize(primary_columns_num); - for (size_t i = 0; i < primary_columns_num; ++i) - index_columns[i] = primary_index_block.getByPosition(i).column->cloneEmpty(); - } + index_columns = primary_index_block.cloneEmptyColumns(); { /** While filling index (index_columns), disable memory tracker. @@ -328,22 +346,14 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializePrimaryIndex(const Bloc /// Write index. The index contains Primary Key value for each `index_granularity` row. for (const auto & granule : granules_to_write) { - if (metadata_snapshot->hasPrimaryKey() && granule.mark_on_start) - { - for (size_t j = 0; j < primary_columns_num; ++j) - { - const auto & primary_column = primary_index_block.getByPosition(j); - index_columns[j]->insertFrom(*primary_column.column, granule.start_row); - primary_column.type->getDefaultSerialization()->serializeBinary( - *primary_column.column, granule.start_row, compress_primary_key ? *index_source_hashing_stream : *index_file_hashing_stream, {}); - } - } + if (granule.mark_on_start) + calculateAndSerializePrimaryIndexRow(primary_index_block, granule.start_row); } } - /// store last index row to write final mark at the end of column - for (size_t j = 0; j < primary_columns_num; ++j) - last_block_index_columns[j] = primary_index_block.getByPosition(j).column; + /// Store block with last index row to write final mark at the end of column + if (with_final_mark) + last_index_block = primary_index_block; } void MergeTreeDataPartWriterOnDisk::calculateAndSerializeStatistics(const Block & block) @@ -420,17 +430,11 @@ void MergeTreeDataPartWriterOnDisk::fillPrimaryIndexChecksums(MergeTreeData::Dat if (index_file_hashing_stream) { - if (write_final_mark) + if (write_final_mark && last_index_block) { - for (size_t j = 0; j < index_columns.size(); ++j) - { - const auto & column = *last_block_index_columns[j]; - size_t last_row_number = column.size() - 1; - index_columns[j]->insertFrom(column, last_row_number); - index_types[j]->getDefaultSerialization()->serializeBinary( - column, last_row_number, compress_primary_key ? *index_source_hashing_stream : *index_file_hashing_stream, {}); - } - last_block_index_columns.clear(); + MemoryTrackerBlockerInThread temporarily_disable_memory_tracker; + calculateAndSerializePrimaryIndexRow(last_index_block, last_index_block.rows() - 1); + last_index_block.clear(); } if (compress_primary_key) diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h index bdf0fdb7f32..8d84442981e 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h @@ -173,10 +173,10 @@ protected: std::unique_ptr index_source_hashing_stream; bool compress_primary_key; - DataTypes index_types; - /// Index columns from the last block - /// It's written to index file in the `writeSuffixAndFinalizePart` method - Columns last_block_index_columns; + /// Last block with index columns. + /// It's written to index file in the `writeSuffixAndFinalizePart` method. + Block last_index_block; + Serializations index_serializations; bool data_written = false; @@ -193,6 +193,7 @@ private: void initStatistics(); virtual void fillIndexGranularity(size_t index_granularity_for_block, size_t rows_in_block) = 0; + void calculateAndSerializePrimaryIndexRow(const Block & index_block, size_t row); struct ExecutionStatistics { diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp index 5ba326cef0c..3fbabe1dd52 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp @@ -7,6 +7,7 @@ #include #include #include +#include namespace DB { @@ -433,6 +434,7 @@ void MergeTreeDataPartWriterWide::writeColumn( if (inserted) { ISerialization::SerializeBinaryBulkSettings serialize_settings; + serialize_settings.use_compact_variant_discriminators_serialization = settings.use_compact_variant_discriminators_serialization; serialize_settings.getter = createStreamGetter(name_and_type, offset_columns); serialization->serializeBinaryBulkStatePrefix(column, serialize_settings, it->second); } @@ -441,6 +443,7 @@ void MergeTreeDataPartWriterWide::writeColumn( serialize_settings.getter = createStreamGetter(name_and_type, offset_columns); serialize_settings.low_cardinality_max_dictionary_size = settings.low_cardinality_max_dictionary_size; serialize_settings.low_cardinality_use_single_dictionary_for_part = settings.low_cardinality_use_single_dictionary_for_part; + serialize_settings.use_compact_variant_discriminators_serialization = settings.use_compact_variant_discriminators_serialization; for (const auto & granule : granules) { @@ -558,7 +561,10 @@ void MergeTreeDataPartWriterWide::validateColumnOfFixedSize(const NameAndTypePai if (index_granularity_rows != index_granularity.getMarkRows(mark_num)) { - throw Exception( + /// With fixed granularity we can have last mark with less rows than granularity + const bool is_last_mark = (mark_num + 1 == index_granularity.getMarksCount()); + if (!index_granularity_info.fixed_index_granularity || !is_last_mark) + throw Exception( ErrorCodes::LOGICAL_ERROR, "Incorrect mark rows for part {} for mark #{}" " (compressed offset {}, decompressed offset {}), in-memory {}, on disk {}, total marks {}", @@ -627,6 +633,7 @@ void MergeTreeDataPartWriterWide::fillDataChecksums(MergeTreeDataPartChecksums & ISerialization::SerializeBinaryBulkSettings serialize_settings; serialize_settings.low_cardinality_max_dictionary_size = settings.low_cardinality_max_dictionary_size; serialize_settings.low_cardinality_use_single_dictionary_for_part = settings.low_cardinality_use_single_dictionary_for_part; + serialize_settings.use_compact_variant_discriminators_serialization = settings.use_compact_variant_discriminators_serialization; WrittenOffsetColumns offset_columns; if (rows_written_in_last_mark > 0) { @@ -821,7 +828,14 @@ void MergeTreeDataPartWriterWide::adjustLastMarkIfNeedAndFlushToDisk(size_t new_ /// Without offset rows_written_in_last_mark = 0; } + + if (compute_granularity) + { + index_granularity.popMark(); + index_granularity.appendMark(new_rows_in_last_mark); + } } + } } diff --git a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp index 2e287ff3042..c5214f0f3d3 100644 --- a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp +++ b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include @@ -36,6 +37,7 @@ #include #include +#include #include #include #include diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp index 5c8aa32949d..ee3ac4207cc 100644 --- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp +++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp @@ -12,12 +12,14 @@ #include #include #include +#include #include #include #include #include #include #include +#include #include @@ -313,6 +315,12 @@ BlocksWithPartition MergeTreeDataWriter::splitBlockIntoParts( for (size_t i = 0; i < async_insert_info_with_partition.size(); ++i) { + if (async_insert_info_with_partition[i] == nullptr) + { + LOG_ERROR(getLogger("MergeTreeDataWriter"), "The {}th element in async_insert_info_with_partition is nullptr. There are totally {} partitions in the insert. Selector content is {}", + i, partitions_count, fmt::join(selector.begin(), selector.end(), ",")); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected error for async deduplicated insert, please check error logs"); + } result[i].offsets = std::move(async_insert_info_with_partition[i]->offsets); result[i].tokens = std::move(async_insert_info_with_partition[i]->tokens); } diff --git a/src/Storages/MergeTree/MergeTreeIOSettings.cpp b/src/Storages/MergeTree/MergeTreeIOSettings.cpp new file mode 100644 index 00000000000..58c3bd28d6a --- /dev/null +++ b/src/Storages/MergeTree/MergeTreeIOSettings.cpp @@ -0,0 +1,37 @@ +#include +#include +#include +#include +#include + +namespace DB +{ + +MergeTreeWriterSettings::MergeTreeWriterSettings( + const Settings & global_settings, + const WriteSettings & query_write_settings_, + const MergeTreeSettingsPtr & storage_settings, + bool can_use_adaptive_granularity_, + bool rewrite_primary_key_, + bool blocks_are_granules_size_) + : min_compress_block_size( + storage_settings->min_compress_block_size ? storage_settings->min_compress_block_size : global_settings.min_compress_block_size) + , max_compress_block_size( + storage_settings->max_compress_block_size ? storage_settings->max_compress_block_size : global_settings.max_compress_block_size) + , marks_compression_codec(storage_settings->marks_compression_codec) + , marks_compress_block_size(storage_settings->marks_compress_block_size) + , compress_primary_key(storage_settings->compress_primary_key) + , primary_key_compression_codec(storage_settings->primary_key_compression_codec) + , primary_key_compress_block_size(storage_settings->primary_key_compress_block_size) + , can_use_adaptive_granularity(can_use_adaptive_granularity_) + , rewrite_primary_key(rewrite_primary_key_) + , blocks_are_granules_size(blocks_are_granules_size_) + , query_write_settings(query_write_settings_) + , max_threads_for_annoy_index_creation(global_settings.max_threads_for_annoy_index_creation) + , low_cardinality_max_dictionary_size(global_settings.low_cardinality_max_dictionary_size) + , low_cardinality_use_single_dictionary_for_part(global_settings.low_cardinality_use_single_dictionary_for_part != 0) + , use_compact_variant_discriminators_serialization(storage_settings->use_compact_variant_discriminators_serialization) +{ +} + +} diff --git a/src/Storages/MergeTree/MergeTreeIOSettings.h b/src/Storages/MergeTree/MergeTreeIOSettings.h index a9125b4047e..c79ca1e66ee 100644 --- a/src/Storages/MergeTree/MergeTreeIOSettings.h +++ b/src/Storages/MergeTree/MergeTreeIOSettings.h @@ -1,15 +1,17 @@ #pragma once #include -#include -#include -#include #include #include +#include namespace DB { +struct MergeTreeSettings; +using MergeTreeSettingsPtr = std::shared_ptr; +struct Settings; + class MMappedFileCache; using MMappedFileCachePtr = std::shared_ptr; @@ -58,26 +60,7 @@ struct MergeTreeWriterSettings const MergeTreeSettingsPtr & storage_settings, bool can_use_adaptive_granularity_, bool rewrite_primary_key_, - bool blocks_are_granules_size_ = false) - : min_compress_block_size( - storage_settings->min_compress_block_size ? storage_settings->min_compress_block_size : global_settings.min_compress_block_size) - , max_compress_block_size( - storage_settings->max_compress_block_size ? storage_settings->max_compress_block_size - : global_settings.max_compress_block_size) - , marks_compression_codec(storage_settings->marks_compression_codec) - , marks_compress_block_size(storage_settings->marks_compress_block_size) - , compress_primary_key(storage_settings->compress_primary_key) - , primary_key_compression_codec(storage_settings->primary_key_compression_codec) - , primary_key_compress_block_size(storage_settings->primary_key_compress_block_size) - , can_use_adaptive_granularity(can_use_adaptive_granularity_) - , rewrite_primary_key(rewrite_primary_key_) - , blocks_are_granules_size(blocks_are_granules_size_) - , query_write_settings(query_write_settings_) - , max_threads_for_annoy_index_creation(global_settings.max_threads_for_annoy_index_creation) - , low_cardinality_max_dictionary_size(global_settings.low_cardinality_max_dictionary_size) - , low_cardinality_use_single_dictionary_for_part(global_settings.low_cardinality_use_single_dictionary_for_part != 0) - { - } + bool blocks_are_granules_size_ = false); size_t min_compress_block_size; size_t max_compress_block_size; @@ -98,6 +81,7 @@ struct MergeTreeWriterSettings size_t low_cardinality_max_dictionary_size; bool low_cardinality_use_single_dictionary_for_part; + bool use_compact_variant_discriminators_serialization; }; } diff --git a/src/Storages/MergeTree/MergeTreeIndexAnnoy.cpp b/src/Storages/MergeTree/MergeTreeIndexAnnoy.cpp index e492ca0aec2..c66aa2373e7 100644 --- a/src/Storages/MergeTree/MergeTreeIndexAnnoy.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexAnnoy.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -230,7 +231,7 @@ MergeTreeIndexConditionAnnoy::MergeTreeIndexConditionAnnoy( ContextPtr context) : ann_condition(query, context) , distance_function(distance_function_) - , search_k(context->getSettings().annoy_index_search_k_nodes) + , search_k(context->getSettingsRef().annoy_index_search_k_nodes) {} bool MergeTreeIndexConditionAnnoy::mayBeTrueOnGranule(MergeTreeIndexGranulePtr /*idx_granule*/) const diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp b/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp index 62bb31dff68..067a692a3b5 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp @@ -1,5 +1,6 @@ -#include #include +#include +#include namespace fs = std::filesystem; diff --git a/src/Storages/MergeTree/MergeTreeIndexSet.cpp b/src/Storages/MergeTree/MergeTreeIndexSet.cpp index b11cbf1e034..284d47ef9e7 100644 --- a/src/Storages/MergeTree/MergeTreeIndexSet.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexSet.cpp @@ -44,10 +44,12 @@ MergeTreeIndexGranuleSet::MergeTreeIndexGranuleSet( const String & index_name_, const Block & index_sample_block_, size_t max_rows_, - MutableColumns && mutable_columns_) + MutableColumns && mutable_columns_, + std::vector && set_hyperrectangle_) : index_name(index_name_) , max_rows(max_rows_) , block(index_sample_block_.cloneWithColumns(std::move(mutable_columns_))) + , set_hyperrectangle(std::move(set_hyperrectangle_)) { } @@ -106,6 +108,10 @@ void MergeTreeIndexGranuleSet::deserializeBinary(ReadBuffer & istr, MergeTreeInd settings.getter = [&](ISerialization::SubstreamPath) -> ReadBuffer * { return &istr; }; settings.position_independent_encoding = false; + set_hyperrectangle.clear(); + Field min_val; + Field max_val; + for (size_t i = 0; i < num_columns; ++i) { auto & elem = block.getByPosition(i); @@ -116,6 +122,13 @@ void MergeTreeIndexGranuleSet::deserializeBinary(ReadBuffer & istr, MergeTreeInd serialization->deserializeBinaryBulkStatePrefix(settings, state, nullptr); serialization->deserializeBinaryBulkWithMultipleStreams(elem.column, rows_to_read, settings, state, nullptr); + + if (const auto * column_nullable = typeid_cast(elem.column.get())) + column_nullable->getExtremesNullLast(min_val, max_val); + else + elem.column->getExtremes(min_val, max_val); + + set_hyperrectangle.emplace_back(min_val, true, max_val, true); } } @@ -182,10 +195,29 @@ void MergeTreeIndexAggregatorSet::update(const Block & block, size_t * pos, size if (has_new_data) { + FieldRef field_min; + FieldRef field_max; for (size_t i = 0; i < columns.size(); ++i) { auto filtered_column = block.getByName(index_columns[i]).column->filter(filter, block.rows()); columns[i]->insertRangeFrom(*filtered_column, 0, filtered_column->size()); + + if (const auto * column_nullable = typeid_cast(filtered_column.get())) + column_nullable->getExtremesNullLast(field_min, field_max); + else + filtered_column->getExtremes(field_min, field_max); + + if (set_hyperrectangle.size() <= i) + { + set_hyperrectangle.emplace_back(field_min, true, field_max, true); + } + else + { + set_hyperrectangle[i].left + = applyVisitor(FieldVisitorAccurateLess(), set_hyperrectangle[i].left, field_min) ? set_hyperrectangle[i].left : field_min; + set_hyperrectangle[i].right + = applyVisitor(FieldVisitorAccurateLess(), set_hyperrectangle[i].right, field_max) ? field_max : set_hyperrectangle[i].right; + } } } @@ -221,7 +253,7 @@ bool MergeTreeIndexAggregatorSet::buildFilter( MergeTreeIndexGranulePtr MergeTreeIndexAggregatorSet::getGranuleAndReset() { - auto granule = std::make_shared(index_name, index_sample_block, max_rows, std::move(columns)); + auto granule = std::make_shared(index_name, index_sample_block, max_rows, std::move(columns), std::move(set_hyperrectangle)); switch (data.type) { @@ -240,17 +272,22 @@ MergeTreeIndexGranulePtr MergeTreeIndexAggregatorSet::getGranuleAndReset() return granule; } +KeyCondition buildCondition(const IndexDescription & index, const ActionsDAGPtr & filter_actions_dag, ContextPtr context) +{ + return KeyCondition{filter_actions_dag, context, index.column_names, index.expression}; +} MergeTreeIndexConditionSet::MergeTreeIndexConditionSet( - const String & index_name_, - const Block & index_sample_block, size_t max_rows_, const ActionsDAGPtr & filter_dag, - ContextPtr context) - : index_name(index_name_) + ContextPtr context, + const IndexDescription & index_description) + : index_name(index_description.name) , max_rows(max_rows_) + , index_data_types(index_description.data_types) + , condition(buildCondition(index_description, filter_dag, context)) { - for (const auto & name : index_sample_block.getNames()) + for (const auto & name : index_description.sample_block.getNames()) if (!key_columns.contains(name)) key_columns.insert(name); @@ -293,6 +330,9 @@ bool MergeTreeIndexConditionSet::mayBeTrueOnGranule(MergeTreeIndexGranulePtr idx if (size == 0 || (max_rows != 0 && size > max_rows)) return true; + if (!condition.checkInHyperrectangle(granule.set_hyperrectangle, index_data_types).can_be_true) + return false; + Block result = granule.block; actions->execute(result); @@ -546,7 +586,7 @@ MergeTreeIndexAggregatorPtr MergeTreeIndexSet::createIndexAggregator(const Merge MergeTreeIndexConditionPtr MergeTreeIndexSet::createIndexCondition( const ActionsDAGPtr & filter_actions_dag, ContextPtr context) const { - return std::make_shared(index.name, index.sample_block, max_rows, filter_actions_dag, context); + return std::make_shared(max_rows, filter_actions_dag, context, index); } MergeTreeIndexPtr setIndexCreator(const IndexDescription & index) diff --git a/src/Storages/MergeTree/MergeTreeIndexSet.h b/src/Storages/MergeTree/MergeTreeIndexSet.h index 6efc2effafd..168262360fc 100644 --- a/src/Storages/MergeTree/MergeTreeIndexSet.h +++ b/src/Storages/MergeTree/MergeTreeIndexSet.h @@ -22,7 +22,8 @@ struct MergeTreeIndexGranuleSet final : public IMergeTreeIndexGranule const String & index_name_, const Block & index_sample_block_, size_t max_rows_, - MutableColumns && columns_); + MutableColumns && columns_, + std::vector && set_hyperrectangle_); void serializeBinary(WriteBuffer & ostr) const override; void deserializeBinary(ReadBuffer & istr, MergeTreeIndexVersion version) override; @@ -36,6 +37,7 @@ struct MergeTreeIndexGranuleSet final : public IMergeTreeIndexGranule const size_t max_rows; Block block; + std::vector set_hyperrectangle; }; @@ -73,6 +75,7 @@ private: ClearableSetVariants data; Sizes key_sizes; MutableColumns columns; + std::vector set_hyperrectangle; }; @@ -80,11 +83,10 @@ class MergeTreeIndexConditionSet final : public IMergeTreeIndexCondition { public: MergeTreeIndexConditionSet( - const String & index_name_, - const Block & index_sample_block, size_t max_rows_, const ActionsDAGPtr & filter_dag, - ContextPtr context); + ContextPtr context, + const IndexDescription & index_description); bool alwaysUnknownOrTrue() const override; @@ -119,6 +121,9 @@ private: std::unordered_set key_columns; ExpressionActionsPtr actions; String actions_output_column_name; + + DataTypes index_data_types; + KeyCondition condition; }; diff --git a/src/Storages/MergeTree/MergeTreePartInfo.h b/src/Storages/MergeTree/MergeTreePartInfo.h index 9bb79e21144..f128722b03b 100644 --- a/src/Storages/MergeTree/MergeTreePartInfo.h +++ b/src/Storages/MergeTree/MergeTreePartInfo.h @@ -101,9 +101,8 @@ struct MergeTreePartInfo bool isFakeDropRangePart() const { - /// Another max level was previously used for REPLACE/MOVE PARTITION - auto another_max_level = std::numeric_limits::max(); - return level == MergeTreePartInfo::MAX_LEVEL || level == another_max_level; + /// LEGACY_MAX_LEVEL was previously used for REPLACE/MOVE PARTITION + return level == MergeTreePartInfo::MAX_LEVEL || level == MergeTreePartInfo::LEGACY_MAX_LEVEL; } String getPartNameAndCheckFormat(MergeTreeDataFormatVersion format_version) const; diff --git a/src/Storages/MergeTree/MergeTreePartsMover.cpp b/src/Storages/MergeTree/MergeTreePartsMover.cpp index 1db70162bff..9223d6fd5b1 100644 --- a/src/Storages/MergeTree/MergeTreePartsMover.cpp +++ b/src/Storages/MergeTree/MergeTreePartsMover.cpp @@ -1,5 +1,6 @@ -#include #include +#include +#include #include #include diff --git a/src/Storages/MergeTree/MergeTreePrefetchedReadPool.cpp b/src/Storages/MergeTree/MergeTreePrefetchedReadPool.cpp index 2c249f7b63b..26595fbb36d 100644 --- a/src/Storages/MergeTree/MergeTreePrefetchedReadPool.cpp +++ b/src/Storages/MergeTree/MergeTreePrefetchedReadPool.cpp @@ -14,6 +14,7 @@ #include #include #include +#include namespace ProfileEvents @@ -378,7 +379,6 @@ void MergeTreePrefetchedReadPool::fillPerThreadTasks(size_t threads, size_t sum_ for (const auto & part : per_part_statistics) total_size_approx += part.sum_marks * part.approx_size_of_mark; - size_t min_prefetch_step_marks = pool_settings.min_marks_for_concurrent_read; for (size_t i = 0; i < per_part_infos.size(); ++i) { auto & part_stat = per_part_statistics[i]; @@ -393,29 +393,7 @@ void MergeTreePrefetchedReadPool::fillPerThreadTasks(size_t threads, size_t sum_ 1, static_cast(std::round(static_cast(settings.filesystem_prefetch_step_bytes) / part_stat.approx_size_of_mark))); } - /// This limit is important to avoid spikes of slow aws getObject requests when parallelizing within one file. - /// (The default is taken from here https://docs.aws.amazon.com/whitepapers/latest/s3-optimizing-performance-best-practices/use-byte-range-fetches.html). - if (part_stat.approx_size_of_mark - && settings.filesystem_prefetch_min_bytes_for_single_read_task - && part_stat.approx_size_of_mark < settings.filesystem_prefetch_min_bytes_for_single_read_task) - { - const size_t min_prefetch_step_marks_by_total_cols = static_cast( - std::ceil(static_cast(settings.filesystem_prefetch_min_bytes_for_single_read_task) / part_stat.approx_size_of_mark)); - - /// At least one task to start working on it right now and another one to prefetch in the meantime. - const size_t new_min_prefetch_step_marks = std::min(min_prefetch_step_marks_by_total_cols, sum_marks / threads / 2); - if (min_prefetch_step_marks < new_min_prefetch_step_marks) - { - LOG_DEBUG(log, "Increasing min prefetch step from {} to {}", min_prefetch_step_marks, new_min_prefetch_step_marks); - min_prefetch_step_marks = new_min_prefetch_step_marks; - } - } - - if (part_stat.prefetch_step_marks < min_prefetch_step_marks) - { - LOG_DEBUG(log, "Increasing prefetch step from {} to {}", part_stat.prefetch_step_marks, min_prefetch_step_marks); - part_stat.prefetch_step_marks = min_prefetch_step_marks; - } + part_stat.prefetch_step_marks = std::max(part_stat.prefetch_step_marks, per_part_infos[i]->min_marks_per_task); LOG_DEBUG( log, @@ -432,11 +410,10 @@ void MergeTreePrefetchedReadPool::fillPerThreadTasks(size_t threads, size_t sum_ LOG_DEBUG( log, - "Sum marks: {}, threads: {}, min_marks_per_thread: {}, min prefetch step marks: {}, prefetches limit: {}, total_size_approx: {}", + "Sum marks: {}, threads: {}, min_marks_per_thread: {}, prefetches limit: {}, total_size_approx: {}", sum_marks, threads, min_marks_per_thread, - min_prefetch_step_marks, settings.filesystem_prefetches_limit, total_size_approx); diff --git a/src/Storages/MergeTree/MergeTreeReadPool.cpp b/src/Storages/MergeTree/MergeTreeReadPool.cpp index dc1ba030f45..cc321cd5a4d 100644 --- a/src/Storages/MergeTree/MergeTreeReadPool.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPool.cpp @@ -7,6 +7,7 @@ #include #include #include +#include #include @@ -25,14 +26,6 @@ extern const int CANNOT_SCHEDULE_TASK; extern const int LOGICAL_ERROR; } -size_t getApproxSizeOfPart(const IMergeTreeDataPart & part, const Names & columns_to_read) -{ - ColumnSize columns_size{}; - for (const auto & col_name : columns_to_read) - columns_size.add(part.getColumnSize(col_name)); - return columns_size.data_compressed; -} - MergeTreeReadPool::MergeTreeReadPool( RangesInDataParts && parts_, VirtualFields shared_virtual_fields_, @@ -53,37 +46,9 @@ MergeTreeReadPool::MergeTreeReadPool( column_names_, settings_, context_) - , min_marks_for_concurrent_read(pool_settings.min_marks_for_concurrent_read) , backoff_settings{context_->getSettingsRef()} , backoff_state{pool_settings.threads} { - if (std::ranges::count(is_part_on_remote_disk, true)) - { - const auto & settings = context_->getSettingsRef(); - - size_t total_compressed_bytes = 0; - size_t total_marks = 0; - for (const auto & part : parts_ranges) - { - const auto & columns = settings.merge_tree_determine_task_size_by_prewhere_columns && prewhere_info - ? prewhere_info->prewhere_actions->getRequiredColumnsNames() - : column_names_; - - total_compressed_bytes += getApproxSizeOfPart(*part.data_part, columns); - total_marks += part.getMarksCount(); - } - - if (total_marks) - { - const auto min_bytes_per_task = settings.merge_tree_min_bytes_per_task_for_remote_reading; - const auto avg_mark_bytes = std::max(total_compressed_bytes / total_marks, 1); - /// We're taking min here because number of tasks shouldn't be too low - it will make task stealing impossible. - const auto heuristic_min_marks = std::min(total_marks / pool_settings.threads, min_bytes_per_task / avg_mark_bytes); - - min_marks_for_concurrent_read = std::max(heuristic_min_marks, min_marks_for_concurrent_read); - } - } - fillPerThreadInfo(pool_settings.threads, pool_settings.sum_marks); } @@ -128,15 +93,16 @@ MergeTreeReadTaskPtr MergeTreeReadPool::getTask(size_t task_idx, MergeTreeReadTa const auto part_idx = thread_task.part_idx; auto & marks_in_part = thread_tasks.sum_marks_in_parts.back(); + const auto min_marks_per_task = per_part_infos[part_idx]->min_marks_per_task; size_t need_marks; if (is_part_on_remote_disk[part_idx] && !pool_settings.use_const_size_tasks_for_remote_reading) need_marks = marks_in_part; else /// Get whole part to read if it is small enough. - need_marks = std::min(marks_in_part, min_marks_for_concurrent_read); + need_marks = std::min(marks_in_part, min_marks_per_task); /// Do not leave too little rows in part for next time. - if (marks_in_part > need_marks && marks_in_part - need_marks < min_marks_for_concurrent_read / 2) + if (marks_in_part > need_marks && marks_in_part - need_marks < min_marks_per_task / 2) need_marks = marks_in_part; MarkRanges ranges_to_get_from_part; @@ -255,8 +221,6 @@ void MergeTreeReadPool::fillPerThreadInfo(size_t threads, size_t sum_marks) parts_queue.push(std::move(info.second)); } - LOG_DEBUG(log, "min_marks_for_concurrent_read={}", min_marks_for_concurrent_read); - const size_t min_marks_per_thread = (sum_marks - 1) / threads + 1; for (size_t i = 0; i < threads && !parts_queue.empty(); ++i) @@ -269,15 +233,14 @@ void MergeTreeReadPool::fillPerThreadInfo(size_t threads, size_t sum_marks) auto & part_with_ranges = current_parts.back().part; size_t & marks_in_part = current_parts.back().sum_marks; const auto part_idx = current_parts.back().part_idx; + const auto min_marks_per_task = per_part_infos[part_idx]->min_marks_per_task; /// Do not get too few rows from part. - if (marks_in_part >= min_marks_for_concurrent_read && - need_marks < min_marks_for_concurrent_read) - need_marks = min_marks_for_concurrent_read; + if (marks_in_part >= min_marks_per_task && need_marks < min_marks_per_task) + need_marks = min_marks_per_task; /// Do not leave too few rows in part for next time. - if (marks_in_part > need_marks && - marks_in_part - need_marks < min_marks_for_concurrent_read) + if (marks_in_part > need_marks && marks_in_part - need_marks < min_marks_per_task) need_marks = marks_in_part; MarkRanges ranges_to_get_from_part; @@ -335,4 +298,12 @@ void MergeTreeReadPool::fillPerThreadInfo(size_t threads, size_t sum_marks) } } +MergeTreeReadPool::BackoffSettings::BackoffSettings(const DB::Settings& settings) + : min_read_latency_ms(settings.read_backoff_min_latency_ms.totalMilliseconds()), + max_throughput(settings.read_backoff_max_throughput), + min_interval_between_events_ms(settings.read_backoff_min_interval_between_events_ms.totalMilliseconds()), + min_events(settings.read_backoff_min_events), + min_concurrency(settings.read_backoff_min_concurrency) +{} + } diff --git a/src/Storages/MergeTree/MergeTreeReadPool.h b/src/Storages/MergeTree/MergeTreeReadPool.h index cb0e8a9657f..7f0de21e1a4 100644 --- a/src/Storages/MergeTree/MergeTreeReadPool.h +++ b/src/Storages/MergeTree/MergeTreeReadPool.h @@ -64,12 +64,7 @@ public: size_t min_concurrency = 1; /// Constants above is just an example. - explicit BackoffSettings(const Settings & settings) - : min_read_latency_ms(settings.read_backoff_min_latency_ms.totalMilliseconds()), - max_throughput(settings.read_backoff_max_throughput), - min_interval_between_events_ms(settings.read_backoff_min_interval_between_events_ms.totalMilliseconds()), - min_events(settings.read_backoff_min_events), - min_concurrency(settings.read_backoff_min_concurrency) {} + explicit BackoffSettings(const Settings & settings); BackoffSettings() : min_read_latency_ms(0) {} }; @@ -78,7 +73,6 @@ private: void fillPerThreadInfo(size_t threads, size_t sum_marks); mutable std::mutex mutex; - size_t min_marks_for_concurrent_read = 0; /// State to track numbers of slow reads. struct BackoffState diff --git a/src/Storages/MergeTree/MergeTreeReadPoolBase.cpp b/src/Storages/MergeTree/MergeTreeReadPoolBase.cpp index 0ea19370d45..46482bc0959 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolBase.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPoolBase.cpp @@ -1,6 +1,10 @@ #include -#include + +#include #include +#include + +#include namespace DB @@ -34,10 +38,58 @@ MergeTreeReadPoolBase::MergeTreeReadPoolBase( , header(storage_snapshot->getSampleBlockForColumns(column_names)) , profile_callback([this](ReadBufferFromFileBase::ProfileInfo info_) { profileFeedback(info_); }) { - fillPerPartInfos(); + fillPerPartInfos(context_->getSettingsRef()); } -void MergeTreeReadPoolBase::fillPerPartInfos() +static size_t getApproxSizeOfPart(const IMergeTreeDataPart & part, const Names & columns_to_read) +{ + ColumnSize columns_size{}; + for (const auto & col_name : columns_to_read) + columns_size.add(part.getColumnSize(col_name)); + /// For compact parts we don't know individual column sizes, let's use whole part size as approximation + return columns_size.data_compressed ? columns_size.data_compressed : part.getBytesOnDisk(); +} + +static size_t calculateMinMarksPerTask( + const RangesInDataPart & part, + const Names & columns_to_read, + PrewhereInfoPtr prewhere_info, + const MergeTreeReadPoolBase::PoolSettings & pool_settings, + const Settings & settings) +{ + size_t min_marks_per_task = pool_settings.min_marks_for_concurrent_read; + const size_t part_marks_count = part.getMarksCount(); + if (part_marks_count && part.data_part->isStoredOnRemoteDisk()) + { + /// We assume that most of the time prewhere does it's job good meaning that lion's share of the rows is filtered out. + /// Which means in turn that for most of the rows we will read only the columns from prewhere clause. + /// So it makes sense to use only them for the estimation. + const auto & columns = settings.merge_tree_determine_task_size_by_prewhere_columns && prewhere_info + ? prewhere_info->prewhere_actions->getRequiredColumnsNames() + : columns_to_read; + const size_t part_compressed_bytes = getApproxSizeOfPart(*part.data_part, columns); + + const auto avg_mark_bytes = std::max(part_compressed_bytes / part_marks_count, 1); + const auto min_bytes_per_task = settings.merge_tree_min_bytes_per_task_for_remote_reading; + /// We're taking min here because number of tasks shouldn't be too low - it will make task stealing impossible. + /// We also create at least two tasks per thread to have something to steal from a slow thread. + const auto heuristic_min_marks + = std::min(pool_settings.sum_marks / pool_settings.threads / 2, min_bytes_per_task / avg_mark_bytes); + if (heuristic_min_marks > min_marks_per_task) + { + LOG_TEST( + &Poco::Logger::get("MergeTreeReadPoolBase"), + "Increasing min_marks_per_task from {} to {} based on columns size heuristic", + min_marks_per_task, + heuristic_min_marks); + min_marks_per_task = heuristic_min_marks; + } + } + LOG_TEST(&Poco::Logger::get("MergeTreeReadPoolBase"), "Will use min_marks_per_task={}", min_marks_per_task); + return min_marks_per_task; +} + +void MergeTreeReadPoolBase::fillPerPartInfos(const Settings & settings) { per_part_infos.reserve(parts_ranges.size()); is_part_on_remote_disk.reserve(parts_ranges.size()); @@ -101,6 +153,8 @@ void MergeTreeReadPoolBase::fillPerPartInfos() } is_part_on_remote_disk.push_back(part_with_ranges.data_part->isStoredOnRemoteDisk()); + read_task_info.min_marks_per_task + = calculateMinMarksPerTask(part_with_ranges, column_names, prewhere_info, pool_settings, settings); per_part_infos.push_back(std::make_shared(std::move(read_task_info))); } } diff --git a/src/Storages/MergeTree/MergeTreeReadPoolBase.h b/src/Storages/MergeTree/MergeTreeReadPoolBase.h index 1b5bfec5898..123f7538ba8 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolBase.h +++ b/src/Storages/MergeTree/MergeTreeReadPoolBase.h @@ -48,7 +48,7 @@ protected: const UncompressedCachePtr owned_uncompressed_cache; const Block header; - void fillPerPartInfos(); + void fillPerPartInfos(const Settings & settings); std::vector getPerPartSumMarks() const; MergeTreeReadTaskPtr createTask( diff --git a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.cpp b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.cpp index 38035d97f56..33eaf5a49bd 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.cpp @@ -33,7 +33,11 @@ MergeTreeReadPoolParallelReplicas::MergeTreeReadPoolParallelReplicas( context_) , extension(std::move(extension_)) , coordination_mode(CoordinationMode::Default) + , min_marks_per_task(pool_settings.min_marks_for_concurrent_read) { + for (const auto & info : per_part_infos) + min_marks_per_task = std::max(min_marks_per_task, info->min_marks_per_task); + extension.all_callback( InitialAllRangesAnnouncement(coordination_mode, parts_ranges.getDescriptions(), extension.number_of_current_replica)); } @@ -50,7 +54,7 @@ MergeTreeReadTaskPtr MergeTreeReadPoolParallelReplicas::getTask(size_t /*task_id auto result = extension.callback(ParallelReadRequest( coordination_mode, extension.number_of_current_replica, - pool_settings.min_marks_for_concurrent_read * pool_settings.threads, + min_marks_per_task * pool_settings.threads, /// For Default coordination mode we don't need to pass part names. RangesInDataPartsDescription{})); @@ -76,9 +80,9 @@ MergeTreeReadTaskPtr MergeTreeReadPoolParallelReplicas::getTask(size_t /*task_id MarkRanges ranges_to_read; size_t current_sum_marks = 0; - while (current_sum_marks < pool_settings.min_marks_for_concurrent_read && !current_task.ranges.empty()) + while (current_sum_marks < min_marks_per_task && !current_task.ranges.empty()) { - auto diff = pool_settings.min_marks_for_concurrent_read - current_sum_marks; + auto diff = min_marks_per_task - current_sum_marks; auto range = current_task.ranges.front(); if (range.getNumberOfMarks() > diff) { diff --git a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.h b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.h index ca159edb91c..6ba63cc2c9a 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.h +++ b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.h @@ -32,6 +32,7 @@ private: const ParallelReadingExtension extension; const CoordinationMode coordination_mode; + size_t min_marks_per_task{0}; RangesInDataPartsDescription buffered_ranges; bool no_more_tasks_available{false}; LoggerPtr log = getLogger("MergeTreeReadPoolParallelReplicas"); diff --git a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.cpp b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.cpp index 01c0a9f91be..6b5cf978423 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.cpp @@ -32,7 +32,11 @@ MergeTreeReadPoolParallelReplicasInOrder::MergeTreeReadPoolParallelReplicasInOrd context_) , extension(std::move(extension_)) , mode(mode_) + , min_marks_per_task(pool_settings.min_marks_for_concurrent_read) { + for (const auto & info : per_part_infos) + min_marks_per_task = std::max(min_marks_per_task, info->min_marks_per_task); + for (const auto & part : parts_ranges) request.push_back({part.data_part->info, MarkRanges{}}); @@ -76,12 +80,8 @@ MergeTreeReadTaskPtr MergeTreeReadPoolParallelReplicasInOrder::getTask(size_t ta if (no_more_tasks) return nullptr; - auto response = extension.callback(ParallelReadRequest( - mode, - extension.number_of_current_replica, - pool_settings.min_marks_for_concurrent_read * request.size(), - request - )); + auto response + = extension.callback(ParallelReadRequest(mode, extension.number_of_current_replica, min_marks_per_task * request.size(), request)); if (!response || response->description.empty() || response->finish) { diff --git a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.h b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.h index 4fe3f7a699c..22841bea212 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.h +++ b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.h @@ -30,6 +30,7 @@ private: const ParallelReadingExtension extension; const CoordinationMode mode; + size_t min_marks_per_task{0}; bool no_more_tasks{false}; RangesInDataPartsDescription request; RangesInDataPartsDescription buffered_tasks; diff --git a/src/Storages/MergeTree/MergeTreeReadTask.h b/src/Storages/MergeTree/MergeTreeReadTask.h index 794e70a0fbb..e90a07e0b55 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.h +++ b/src/Storages/MergeTree/MergeTreeReadTask.h @@ -68,6 +68,8 @@ struct MergeTreeReadTaskInfo MergeTreeBlockSizePredictorPtr shared_size_predictor; /// TODO: comment VirtualFields const_virtual_fields; + /// The amount of data to read per task based on size of the queried columns. + size_t min_marks_per_task = 0; }; using MergeTreeReadTaskInfoPtr = std::shared_ptr; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 78b67de1a7e..a6298aab3d9 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -145,8 +145,12 @@ ChunkAndProgress MergeTreeSelectProcessor::read() ordered_columns.push_back(res.block.getByName(name).column); } + auto chunk = Chunk(ordered_columns, res.row_count); + if (add_part_level) + chunk.getChunkInfos().add(std::make_shared(task->getInfo().data_part->info.level)); + return ChunkAndProgress{ - .chunk = Chunk(ordered_columns, res.row_count, add_part_level ? std::make_shared(task->getInfo().data_part->info.level) : nullptr), + .chunk = std::move(chunk), .num_read_rows = res.num_read_rows, .num_read_bytes = res.num_read_bytes, .is_finished = false}; diff --git a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp index 02f8d6f4f6a..311720728e7 100644 --- a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp +++ b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include @@ -14,16 +15,11 @@ #include #include #include +#include namespace DB { -namespace ErrorCodes -{ - extern const int MEMORY_LIMIT_EXCEEDED; -} - - /// Lightweight (in terms of logic) stream for reading single part from /// MergeTree, used for merges and mutations. /// @@ -264,7 +260,10 @@ try ++it; } - return Chunk(std::move(res_columns), rows_read, add_part_level ? std::make_shared(data_part->info.level) : nullptr); + auto result = Chunk(std::move(res_columns), rows_read); + if (add_part_level) + result.getChunkInfos().add(std::make_shared(data_part->info.level)); + return result; } } else @@ -277,7 +276,7 @@ try catch (...) { /// Suspicion of the broken part. A part is added to the queue for verification. - if (getCurrentExceptionCode() != ErrorCodes::MEMORY_LIMIT_EXCEEDED) + if (!isRetryableException(std::current_exception())) storage.reportBrokenPart(data_part); throw; } diff --git a/src/Storages/MergeTree/MergeTreeSettings.cpp b/src/Storages/MergeTree/MergeTreeSettings.cpp index 5d6f08d3c53..c968ad84936 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeSettings.cpp @@ -115,6 +115,16 @@ void MergeTreeSettings::loadFromQuery(ASTStorage & storage_def, ContextPtr conte #undef ADD_IF_ABSENT } +bool MergeTreeSettings::isReadonlySetting(const String & name) +{ + return name == "index_granularity" || name == "index_granularity_bytes" || name == "enable_mixed_granularity_parts"; +} + +bool MergeTreeSettings::isPartFormatSetting(const String & name) +{ + return name == "min_bytes_for_wide_part" || name == "min_rows_for_wide_part"; +} + void MergeTreeSettings::sanityCheck(size_t background_pool_tasks) const { if (number_of_free_entries_in_pool_to_execute_mutation > background_pool_tasks) diff --git a/src/Storages/MergeTree/MergeTreeSettings.h b/src/Storages/MergeTree/MergeTreeSettings.h index 1f8d6abebd2..f5ada81cf55 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.h +++ b/src/Storages/MergeTree/MergeTreeSettings.h @@ -43,6 +43,7 @@ struct Settings; M(UInt64, compact_parts_max_granules_to_buffer, 128, "Only available in ClickHouse Cloud", 0) \ M(UInt64, compact_parts_merge_max_bytes_to_prefetch_part, 16 * 1024 * 1024, "Only available in ClickHouse Cloud", 0) \ M(Bool, load_existing_rows_count_for_old_parts, false, "Whether to load existing_rows_count for existing parts. If false, existing_rows_count will be equal to rows_count for existing parts.", 0) \ + M(Bool, use_compact_variant_discriminators_serialization, true, "Use compact version of Variant discriminators serialization.", 0) \ \ /** Merge settings. */ \ M(UInt64, merge_max_block_size, 8192, "How many rows in blocks should be formed for merge operations. By default has the same value as `index_granularity`.", 0) \ @@ -267,17 +268,8 @@ struct MergeTreeSettings : public BaseSettings, public /// NOTE: will rewrite the AST to add immutable settings. void loadFromQuery(ASTStorage & storage_def, ContextPtr context, bool is_attach); - /// We check settings after storage creation - static bool isReadonlySetting(const String & name) - { - return name == "index_granularity" || name == "index_granularity_bytes" - || name == "enable_mixed_granularity_parts"; - } - - static bool isPartFormatSetting(const String & name) - { - return name == "min_bytes_for_wide_part" || name == "min_rows_for_wide_part"; - } + static bool isReadonlySetting(const String & name); + static bool isPartFormatSetting(const String & name); /// Check that the values are sane taking also query-level settings into account. void sanityCheck(size_t background_pool_tasks) const; @@ -294,4 +286,10 @@ namespace MergeTreeColumnSettings void validate(const SettingsChanges & changes); } +[[maybe_unused]] static bool needSyncPart(size_t input_rows, size_t input_bytes, const MergeTreeSettings & settings) +{ + return ( + (settings.min_rows_to_fsync_after_merge && input_rows >= settings.min_rows_to_fsync_after_merge) + || (settings.min_compressed_bytes_to_fsync_after_merge && input_bytes >= settings.min_compressed_bytes_to_fsync_after_merge)); +} } diff --git a/src/Storages/MergeTree/MergeTreeSink.cpp b/src/Storages/MergeTree/MergeTreeSink.cpp index 05751e0fa6f..0ca4276b858 100644 --- a/src/Storages/MergeTree/MergeTreeSink.cpp +++ b/src/Storages/MergeTree/MergeTreeSink.cpp @@ -1,14 +1,22 @@ #include #include #include +#include #include #include +#include + namespace ProfileEvents { extern const Event DuplicatedInsertedBlocks; } +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + namespace DB { @@ -58,12 +66,12 @@ void MergeTreeSink::onCancel() { } -void MergeTreeSink::consume(Chunk chunk) +void MergeTreeSink::consume(Chunk & chunk) { if (num_blocks_processed > 0) storage.delayInsertOrThrowIfNeeded(nullptr, context, false); - auto block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto block = getHeader().cloneWithColumns(chunk.getColumns()); if (!storage_snapshot->object_columns.empty()) convertDynamicColumnsToTuples(block, storage_snapshot); @@ -76,6 +84,18 @@ void MergeTreeSink::consume(Chunk chunk) size_t streams = 0; bool support_parallel_write = false; + auto token_info = chunk.getChunkInfos().get(); + if (!token_info) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "TokenInfo is expected for consumed chunk in MergeTreeSink for table: {}", + storage.getStorageID().getNameForLogs()); + + const bool need_to_define_dedup_token = !token_info->isDefined(); + + String block_dedup_token; + if (token_info->isDefined()) + block_dedup_token = token_info->getToken(); + for (auto & current_block : part_blocks) { ProfileEvents::Counters part_counters; @@ -100,22 +120,16 @@ void MergeTreeSink::consume(Chunk chunk) if (!temp_part.part) continue; + if (need_to_define_dedup_token) + { + chassert(temp_part.part); + const auto hash_value = temp_part.part->getPartBlockIDHash(); + token_info->addChunkHash(toString(hash_value.items[0]) + "_" + toString(hash_value.items[1])); + } + if (!support_parallel_write && temp_part.part->getDataPartStorage().supportParallelWrite()) support_parallel_write = true; - String block_dedup_token; - if (storage.getDeduplicationLog()) - { - const String & dedup_token = settings.insert_deduplication_token; - if (!dedup_token.empty()) - { - /// multiple blocks can be inserted within the same insert query - /// an ordinal number is added to dedup token to generate a distinctive block id for each block - block_dedup_token = fmt::format("{}_{}", dedup_token, chunk_dedup_seqnum); - ++chunk_dedup_seqnum; - } - } - size_t max_insert_delayed_streams_for_parallel_write; if (settings.max_insert_delayed_streams_for_parallel_write.changed) @@ -127,6 +141,7 @@ void MergeTreeSink::consume(Chunk chunk) /// In case of too much columns/parts in block, flush explicitly. streams += temp_part.streams.size(); + if (streams > max_insert_delayed_streams_for_parallel_write) { finishDelayedChunk(); @@ -143,11 +158,16 @@ void MergeTreeSink::consume(Chunk chunk) { .temp_part = std::move(temp_part), .elapsed_ns = elapsed_ns, - .block_dedup_token = std::move(block_dedup_token), + .block_dedup_token = block_dedup_token, .part_counters = std::move(part_counters), }); } + if (need_to_define_dedup_token) + { + token_info->finishChunkHashes(); + } + finishDelayedChunk(); delayed_chunk = std::make_unique(); delayed_chunk->partitions = std::move(partitions); @@ -160,6 +180,8 @@ void MergeTreeSink::finishDelayedChunk() if (!delayed_chunk) return; + const Settings & settings = context->getSettingsRef(); + for (auto & partition : delayed_chunk->partitions) { ProfileEventsScope scoped_attach(&partition.part_counters); @@ -178,7 +200,8 @@ void MergeTreeSink::finishDelayedChunk() storage.fillNewPartName(part, lock); auto * deduplication_log = storage.getDeduplicationLog(); - if (deduplication_log) + + if (settings.insert_deduplicate && deduplication_log) { const String block_id = part->getZeroLevelPartBlockID(partition.block_dedup_token); auto res = deduplication_log->addPart(block_id, part->info); diff --git a/src/Storages/MergeTree/MergeTreeSink.h b/src/Storages/MergeTree/MergeTreeSink.h index cf6715a3415..90976020d52 100644 --- a/src/Storages/MergeTree/MergeTreeSink.h +++ b/src/Storages/MergeTree/MergeTreeSink.h @@ -25,7 +25,7 @@ public: ~MergeTreeSink() override; String getName() const override { return "MergeTreeSink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onStart() override; void onFinish() override; void onCancel() override; @@ -36,7 +36,6 @@ private: size_t max_parts_per_block; ContextPtr context; StorageSnapshotPtr storage_snapshot; - UInt64 chunk_dedup_seqnum = 0; /// input chunk ordinal number in case of dedup token UInt64 num_blocks_processed = 0; /// We can delay processing for previous chunk and start writing a new one. diff --git a/src/Storages/MergeTree/MergeTreeSource.cpp b/src/Storages/MergeTree/MergeTreeSource.cpp index fcf2dd76e3f..e323b9f9ee7 100644 --- a/src/Storages/MergeTree/MergeTreeSource.cpp +++ b/src/Storages/MergeTree/MergeTreeSource.cpp @@ -133,9 +133,8 @@ private: }; #endif -MergeTreeSource::MergeTreeSource(MergeTreeSelectProcessorPtr processor_) - : ISource(processor_->getHeader()) - , processor(std::move(processor_)) +MergeTreeSource::MergeTreeSource(MergeTreeSelectProcessorPtr processor_, const std::string & log_name_) + : ISource(processor_->getHeader()), processor(std::move(processor_)), log_name(log_name_) { #if defined(OS_LINUX) if (processor->getSettings().use_asynchronous_read_from_pool) @@ -207,7 +206,7 @@ std::optional MergeTreeSource::tryGenerate() try { - OpenTelemetry::SpanHolder span{"MergeTreeSource::tryGenerate()"}; + OpenTelemetry::SpanHolder span{fmt::format("MergeTreeSource({})::tryGenerate", log_name)}; holder->setResult(processor->read()); } catch (...) @@ -222,7 +221,7 @@ std::optional MergeTreeSource::tryGenerate() } #endif - OpenTelemetry::SpanHolder span{"MergeTreeSource::tryGenerate()"}; + OpenTelemetry::SpanHolder span{fmt::format("MergeTreeSource({})::tryGenerate", log_name)}; return processReadResult(processor->read()); } diff --git a/src/Storages/MergeTree/MergeTreeSource.h b/src/Storages/MergeTree/MergeTreeSource.h index 655f0ee6ebe..fc39b4f9b09 100644 --- a/src/Storages/MergeTree/MergeTreeSource.h +++ b/src/Storages/MergeTree/MergeTreeSource.h @@ -12,7 +12,7 @@ struct ChunkAndProgress; class MergeTreeSource final : public ISource { public: - explicit MergeTreeSource(MergeTreeSelectProcessorPtr processor_); + explicit MergeTreeSource(MergeTreeSelectProcessorPtr processor_, const std::string & log_name_); ~MergeTreeSource() override; std::string getName() const override; @@ -30,6 +30,7 @@ protected: private: MergeTreeSelectProcessorPtr processor; + const std::string log_name; #if defined(OS_LINUX) struct AsyncReadingState; diff --git a/src/Storages/MergeTree/MergeTreeWhereOptimizer.cpp b/src/Storages/MergeTree/MergeTreeWhereOptimizer.cpp index a9a5fddace4..44ab12bde2a 100644 --- a/src/Storages/MergeTree/MergeTreeWhereOptimizer.cpp +++ b/src/Storages/MergeTree/MergeTreeWhereOptimizer.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -232,7 +233,7 @@ static bool isConditionGood(const RPNBuilderTreeNode & condition, const NameSet else if (type == Field::Types::Float64) { const auto value = output_value.get(); - return value < threshold || threshold < value; + return value < -threshold || threshold < value; } return false; diff --git a/src/Storages/MergeTree/MergedBlockOutputStream.cpp b/src/Storages/MergeTree/MergedBlockOutputStream.cpp index 164658c914e..38869aebaa5 100644 --- a/src/Storages/MergeTree/MergedBlockOutputStream.cpp +++ b/src/Storages/MergeTree/MergedBlockOutputStream.cpp @@ -4,6 +4,7 @@ #include #include #include +#include namespace DB @@ -93,6 +94,7 @@ struct MergedBlockOutputStream::Finalizer::Impl void MergedBlockOutputStream::Finalizer::finish() { std::unique_ptr to_finish = std::move(impl); + impl.reset(); if (to_finish) to_finish->finish(); } @@ -130,7 +132,19 @@ MergedBlockOutputStream::Finalizer::Finalizer(Finalizer &&) noexcept = default; MergedBlockOutputStream::Finalizer & MergedBlockOutputStream::Finalizer::operator=(Finalizer &&) noexcept = default; MergedBlockOutputStream::Finalizer::Finalizer(std::unique_ptr impl_) : impl(std::move(impl_)) {} -MergedBlockOutputStream::Finalizer::~Finalizer() = default; +MergedBlockOutputStream::Finalizer::~Finalizer() +{ + try + { + if (impl) + finish(); + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + } +} + void MergedBlockOutputStream::finalizePart( const MergeTreeMutableDataPartPtr & new_part, diff --git a/src/Storages/MergeTree/MergedColumnOnlyOutputStream.cpp b/src/Storages/MergeTree/MergedColumnOnlyOutputStream.cpp index 5ae6517a236..c167ac87317 100644 --- a/src/Storages/MergeTree/MergedColumnOnlyOutputStream.cpp +++ b/src/Storages/MergeTree/MergedColumnOnlyOutputStream.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include diff --git a/src/Storages/MergeTree/MutateFromLogEntryTask.cpp b/src/Storages/MergeTree/MutateFromLogEntryTask.cpp index 4c96cbf2c97..73084f487b9 100644 --- a/src/Storages/MergeTree/MutateFromLogEntryTask.cpp +++ b/src/Storages/MergeTree/MutateFromLogEntryTask.cpp @@ -2,6 +2,7 @@ #include #include +#include #include #include diff --git a/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp b/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp index 20f387137e7..9aec074deae 100644 --- a/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp +++ b/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp @@ -3,6 +3,7 @@ #include #include #include +#include namespace DB { diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp index a552ee89aee..9a775db73e2 100644 --- a/src/Storages/MergeTree/MutateTask.cpp +++ b/src/Storages/MergeTree/MutateTask.cpp @@ -1,8 +1,10 @@ #include +#include #include #include #include +#include #include #include #include @@ -19,11 +21,13 @@ #include #include #include +#include #include #include #include #include #include +#include #include #include #include @@ -1297,6 +1301,7 @@ void PartMergerWriter::prepare() bool PartMergerWriter::mutateOriginalPartAndPrepareProjections() { Block cur_block; + Block projection_header; if (MutationHelpers::checkOperationIsNotCanceled(*ctx->merges_blocker, ctx->mutate_entry) && ctx->mutating_executor->pull(cur_block)) { if (ctx->minmax_idx) @@ -1314,14 +1319,12 @@ bool PartMergerWriter::mutateOriginalPartAndPrepareProjections() ProfileEventTimeIncrement watch(ProfileEvents::MutateTaskProjectionsCalculationMicroseconds); Block block_to_squash = projection.calculate(cur_block, ctx->context); - projection_squashes[i].header = block_to_squash; - Chunk planned_chunk = projection_squashes[i].add({block_to_squash.getColumns(), block_to_squash.rows()}); + projection_squashes[i].setHeader(block_to_squash.cloneEmpty()); - if (planned_chunk.hasChunkInfo()) + Chunk squashed_chunk = Squashing::squash(projection_squashes[i].add({block_to_squash.getColumns(), block_to_squash.rows()})); + if (squashed_chunk) { - Chunk projection_chunk = DB::Squashing::squash(std::move(planned_chunk)); - - auto result = block_to_squash.cloneWithColumns(projection_chunk.getColumns()); + auto result = projection_squashes[i].getHeader().cloneWithColumns(squashed_chunk.detachColumns()); auto tmp_part = MergeTreeDataWriter::writeTempProjectionPart( *ctx->data, ctx->log, result, projection, ctx->new_data_part.get(), ++block_num); tmp_part.finalize(); @@ -1342,12 +1345,10 @@ bool PartMergerWriter::mutateOriginalPartAndPrepareProjections() { const auto & projection = *ctx->projections_to_build[i]; auto & projection_squash_plan = projection_squashes[i]; - auto planned_chunk = projection_squash_plan.flush(); - if (planned_chunk.hasChunkInfo()) + auto squashed_chunk = Squashing::squash(projection_squash_plan.flush()); + if (squashed_chunk) { - Chunk projection_chunk = DB::Squashing::squash(std::move(planned_chunk)); - - auto result = projection_squash_plan.header.cloneWithColumns(projection_chunk.getColumns()); + auto result = projection_squash_plan.getHeader().cloneWithColumns(squashed_chunk.detachColumns()); auto temp_part = MergeTreeDataWriter::writeTempProjectionPart( *ctx->data, ctx->log, result, projection, ctx->new_data_part.get(), ++block_num); temp_part.finalize(); diff --git a/src/Storages/MergeTree/ParallelReplicasReadingCoordinator.cpp b/src/Storages/MergeTree/ParallelReplicasReadingCoordinator.cpp index 5a84c6fd684..f46b4de10b7 100644 --- a/src/Storages/MergeTree/ParallelReplicasReadingCoordinator.cpp +++ b/src/Storages/MergeTree/ParallelReplicasReadingCoordinator.cpp @@ -444,6 +444,9 @@ void DefaultCoordinator::doHandleInitialAllRangesAnnouncement(InitialAllRangesAn ErrorCodes::LOGICAL_ERROR, "Replica number ({}) is bigger than total replicas count ({})", replica_num, stats.size()); ++stats[replica_num].number_of_requests; + + if (replica_status[replica_num].is_announcement_received) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Duplicate announcement received for replica number {}", replica_num); replica_status[replica_num].is_announcement_received = true; LOG_DEBUG(log, "Sent initial requests: {} Replicas count: {}", sent_initial_requests, replicas_count); diff --git a/src/Storages/MergeTree/PartMovesBetweenShardsOrchestrator.cpp b/src/Storages/MergeTree/PartMovesBetweenShardsOrchestrator.cpp index 4228d7b70b6..10c7bef72fc 100644 --- a/src/Storages/MergeTree/PartMovesBetweenShardsOrchestrator.cpp +++ b/src/Storages/MergeTree/PartMovesBetweenShardsOrchestrator.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include diff --git a/src/Storages/MergeTree/RPNBuilder.cpp b/src/Storages/MergeTree/RPNBuilder.cpp index 4a18d606bb7..93087b0d3fe 100644 --- a/src/Storages/MergeTree/RPNBuilder.cpp +++ b/src/Storages/MergeTree/RPNBuilder.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include diff --git a/src/Storages/MergeTree/ReplicatedMergeMutateTaskBase.cpp b/src/Storages/MergeTree/ReplicatedMergeMutateTaskBase.cpp index 2fc5238827d..24929365b72 100644 --- a/src/Storages/MergeTree/ReplicatedMergeMutateTaskBase.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeMutateTaskBase.cpp @@ -1,8 +1,9 @@ #include -#include #include +#include #include +#include #include diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.cpp index 336d19692d4..6e22a3515bc 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeCleanupThread.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeCleanupThread.cpp index e034918ef57..7aef249c366 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeCleanupThread.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeCleanupThread.cpp @@ -1,13 +1,14 @@ +#include +#include #include #include -#include -#include #include #include #include #include +#include namespace DB @@ -175,6 +176,8 @@ Float32 ReplicatedMergeTreeCleanupThread::iterate() cleaned_part_like += storage.clearEmptyParts(); } + cleaned_part_like += storage.unloadPrimaryKeysOfOutdatedParts(); + /// We need to measure the number of removed objects somehow (for better scheduling), /// but just summing the number of removed async blocks, logs, and empty parts does not make any sense. /// So we are trying to (approximately) measure the number of inserted blocks/parts, so we will be able to compare apples to apples. diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeMergeStrategyPicker.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeMergeStrategyPicker.cpp index 192f0d23f96..3e64a4c7c52 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeMergeStrategyPicker.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeMergeStrategyPicker.cpp @@ -1,6 +1,7 @@ +#include +#include #include #include -#include #include #include diff --git a/src/Storages/MergeTree/ReplicatedMergeTreePartCheckThread.cpp b/src/Storages/MergeTree/ReplicatedMergeTreePartCheckThread.cpp index d7601e6e638..dc242a7b084 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreePartCheckThread.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreePartCheckThread.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp index 30ba95c46f0..627bda3f8bf 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp index 35f355d1d9b..268cead2a0e 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index 4b4f4c33e7d..63e1c12a15f 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -3,20 +3,24 @@ #include #include #include +#include #include "Common/Exception.h" #include #include #include #include #include +#include #include #include +#include #include #include #include #include #include + namespace ProfileEvents { extern const Event DuplicatedInsertedBlocks; @@ -253,12 +257,12 @@ size_t ReplicatedMergeTreeSinkImpl::checkQuorumPrecondition(const } template -void ReplicatedMergeTreeSinkImpl::consume(Chunk chunk) +void ReplicatedMergeTreeSinkImpl::consume(Chunk & chunk) { if (num_blocks_processed > 0) storage.delayInsertOrThrowIfNeeded(&storage.partial_shutdown_event, context, false); - auto block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto block = getHeader().cloneWithColumns(chunk.getColumns()); const auto & settings = context->getSettingsRef(); @@ -284,13 +288,25 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk chunk) if constexpr (async_insert) { - const auto & chunk_info = chunk.getChunkInfo(); - if (const auto * async_insert_info_ptr = typeid_cast(chunk_info.get())) + const auto async_insert_info_ptr = chunk.getChunkInfos().get(); + if (async_insert_info_ptr) async_insert_info = std::make_shared(async_insert_info_ptr->offsets, async_insert_info_ptr->tokens); else throw Exception(ErrorCodes::LOGICAL_ERROR, "No chunk info for async inserts"); } + String block_dedup_token; + auto token_info = chunk.getChunkInfos().get(); + if (!token_info) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "TokenInfo is expected for consumed chunk in ReplicatedMergeTreeSink for table: {}", + storage.getStorageID().getNameForLogs()); + + const bool need_to_define_dedup_token = !token_info->isDefined(); + + if (token_info->isDefined()) + block_dedup_token = token_info->getToken(); + auto part_blocks = MergeTreeDataWriter::splitBlockIntoParts(std::move(block), max_parts_per_block, metadata_snapshot, context, async_insert_info); using DelayedPartition = typename ReplicatedMergeTreeSinkImpl::DelayedChunk::Partition; @@ -342,23 +358,10 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk chunk) } else { - if (deduplicate) { - String block_dedup_token; - /// We add the hash from the data and partition identifier to deduplication ID. /// That is, do not insert the same data to the same partition twice. - - const String & dedup_token = settings.insert_deduplication_token; - if (!dedup_token.empty()) - { - /// multiple blocks can be inserted within the same insert query - /// an ordinal number is added to dedup token to generate a distinctive block id for each block - block_dedup_token = fmt::format("{}_{}", dedup_token, chunk_dedup_seqnum); - ++chunk_dedup_seqnum; - } - block_id = temp_part.part->getZeroLevelPartBlockID(block_dedup_token); LOG_DEBUG(log, "Wrote block with ID '{}', {} rows{}", block_id, current_block.block.rows(), quorumLogMessage(replicas_num)); } @@ -366,6 +369,13 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk chunk) { LOG_DEBUG(log, "Wrote block with {} rows{}", current_block.block.rows(), quorumLogMessage(replicas_num)); } + + if (need_to_define_dedup_token) + { + chassert(temp_part.part); + const auto hash_value = temp_part.part->getPartBlockIDHash(); + token_info->addChunkHash(toString(hash_value.items[0]) + "_" + toString(hash_value.items[1])); + } } profile_events_scope.reset(); @@ -411,17 +421,15 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk chunk) )); } + if (need_to_define_dedup_token) + { + token_info->finishChunkHashes(); + } + finishDelayedChunk(zookeeper); delayed_chunk = std::make_unique(); delayed_chunk->partitions = std::move(partitions); - /// If deduplicated data should not be inserted into MV, we need to set proper - /// value for `last_block_is_duplicate`, which is possible only after the part is committed. - /// Othervide we can delay commit. - /// TODO: we can also delay commit if there is no MVs. - if (!settings.deduplicate_blocks_in_dependent_materialized_views) - finishDelayedChunk(zookeeper); - ++num_blocks_processed; } @@ -431,8 +439,6 @@ void ReplicatedMergeTreeSinkImpl::finishDelayedChunk(const ZooKeeperWithF if (!delayed_chunk) return; - last_block_is_duplicate = false; - for (auto & partition : delayed_chunk->partitions) { ProfileEventsScope scoped_attach(&partition.part_counters); @@ -445,8 +451,6 @@ void ReplicatedMergeTreeSinkImpl::finishDelayedChunk(const ZooKeeperWithF { bool deduplicated = commitPart(zookeeper, part, partition.block_id, delayed_chunk->replicas_num).second; - last_block_is_duplicate = last_block_is_duplicate || deduplicated; - /// Set a special error code if the block is duplicate int error = (deduplicate && deduplicated) ? ErrorCodes::INSERT_WAS_DEDUPLICATED : 0; auto counters_snapshot = std::make_shared(partition.part_counters.getPartiallyAtomicSnapshot()); @@ -535,7 +539,7 @@ bool ReplicatedMergeTreeSinkImpl::writeExistingPart(MergeTreeData::Mutabl ProfileEventsScope profile_events_scope; String original_part_dir = part->getDataPartStorage().getPartDirectory(); - auto try_rollback_part_rename = [this, &part, &original_part_dir]() + auto try_rollback_part_rename = [this, &part, &original_part_dir] () { if (original_part_dir == part->getDataPartStorage().getPartDirectory()) return; @@ -1151,8 +1155,16 @@ void ReplicatedMergeTreeSinkImpl::onStart() template void ReplicatedMergeTreeSinkImpl::onFinish() { - auto zookeeper = storage.getZooKeeper(); - finishDelayedChunk(std::make_shared(zookeeper)); + const auto & settings = context->getSettingsRef(); + + ZooKeeperWithFaultInjectionPtr zookeeper = ZooKeeperWithFaultInjection::createInstance( + settings.insert_keeper_fault_injection_probability, + settings.insert_keeper_fault_injection_seed, + storage.getZooKeeper(), + "ReplicatedMergeTreeSink::onFinish", + log); + + finishDelayedChunk(zookeeper); } template diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.h b/src/Storages/MergeTree/ReplicatedMergeTreeSink.h index 39623c20584..7d025361717 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.h +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.h @@ -51,7 +51,7 @@ public: ~ReplicatedMergeTreeSinkImpl() override; void onStart() override; - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onFinish() override; String getName() const override { return "ReplicatedMergeTreeSink"; } @@ -59,16 +59,6 @@ public: /// For ATTACHing existing data on filesystem. bool writeExistingPart(MergeTreeData::MutableDataPartPtr & part); - /// For proper deduplication in MaterializedViews - bool lastBlockIsDuplicate() const override - { - /// If MV is responsible for deduplication, block is not considered duplicating. - if (context->getSettingsRef().deduplicate_blocks_in_dependent_materialized_views) - return false; - - return last_block_is_duplicate; - } - struct DelayedChunk; private: std::vector detectConflictsInAsyncBlockIDs(const std::vector & ids); @@ -126,7 +116,6 @@ private: bool allow_attach_while_readonly = false; bool quorum_parallel = false; const bool deduplicate = true; - bool last_block_is_duplicate = false; UInt64 num_blocks_processed = 0; LoggerPtr log; diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeTableMetadata.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeTableMetadata.cpp index 287a4d20543..c4bae5352cb 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeTableMetadata.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeTableMetadata.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include diff --git a/src/Storages/MergeTree/StorageFromMergeTreeDataPart.h b/src/Storages/MergeTree/StorageFromMergeTreeDataPart.h index a94508ad41f..82ebbf09988 100644 --- a/src/Storages/MergeTree/StorageFromMergeTreeDataPart.h +++ b/src/Storages/MergeTree/StorageFromMergeTreeDataPart.h @@ -1,25 +1,17 @@ #pragma once +#include #include #include #include #include -#include -#include -#include -#include #include -#include -#include namespace DB { -namespace ErrorCodes -{ - extern const int LOGICAL_ERROR; -} +class QueryPlan; /// A Storage that allows reading from a single MergeTree data part. class StorageFromMergeTreeDataPart final : public IStorage @@ -46,20 +38,7 @@ public: String getName() const override { return "FromMergeTreeDataPart"; } - StorageSnapshotPtr getStorageSnapshot( - const StorageMetadataPtr & metadata_snapshot, ContextPtr /*query_context*/) const override - { - const auto & storage_columns = metadata_snapshot->getColumns(); - if (!hasDynamicSubcolumns(storage_columns)) - return std::make_shared(*this, metadata_snapshot); - - auto data_parts = storage.getDataPartsVectorForInternalUsage(); - - auto object_columns = getConcreteObjectColumns( - data_parts.begin(), data_parts.end(), storage_columns, [](const auto & part) -> const auto & { return part->getColumns(); }); - - return std::make_shared(*this, metadata_snapshot, std::move(object_columns)); - } + StorageSnapshotPtr getStorageSnapshot(const StorageMetadataPtr & metadata_snapshot, ContextPtr /*query_context*/) const override; void read( QueryPlan & query_plan, @@ -69,21 +48,7 @@ public: ContextPtr context, QueryProcessingStage::Enum /*processed_stage*/, size_t max_block_size, - size_t num_streams) override - { - query_plan.addStep(MergeTreeDataSelectExecutor(storage) - .readFromParts( - parts, - alter_conversions, - column_names, - storage_snapshot, - query_info, - context, - max_block_size, - num_streams, - nullptr, - analysis_result_ptr)); - } + size_t num_streams) override; bool supportsPrewhere() const override { return true; } @@ -102,12 +67,7 @@ public: return storage.getPartitionIDFromQuery(ast, context); } - bool materializeTTLRecalculateOnly() const - { - if (parts.empty()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "parts must not be empty for materializeTTLRecalculateOnly"); - return parts.front()->storage.getSettings()->materialize_ttl_recalculate_only; - } + bool materializeTTLRecalculateOnly() const; bool hasLightweightDeletedMask() const override { diff --git a/src/Storages/MergeTree/checkDataPart.cpp b/src/Storages/MergeTree/checkDataPart.cpp index 525960d5314..fb86d9e7603 100644 --- a/src/Storages/MergeTree/checkDataPart.cpp +++ b/src/Storages/MergeTree/checkDataPart.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include @@ -35,11 +36,13 @@ namespace ErrorCodes extern const int CANNOT_ALLOCATE_MEMORY; extern const int CANNOT_MUNMAP; extern const int CANNOT_MREMAP; + extern const int CANNOT_SCHEDULE_TASK; extern const int UNEXPECTED_FILE_IN_DATA_PART; extern const int NO_FILE_IN_DATA_PART; extern const int NETWORK_ERROR; extern const int SOCKET_TIMEOUT; extern const int BROKEN_PROJECTION; + extern const int ABORTED; } @@ -84,7 +87,9 @@ bool isRetryableException(std::exception_ptr exception_ptr) { return isNotEnoughMemoryErrorCode(e.code()) || e.code() == ErrorCodes::NETWORK_ERROR - || e.code() == ErrorCodes::SOCKET_TIMEOUT; + || e.code() == ErrorCodes::SOCKET_TIMEOUT + || e.code() == ErrorCodes::CANNOT_SCHEDULE_TASK + || e.code() == ErrorCodes::ABORTED; } catch (const Poco::Net::NetException &) { @@ -328,16 +333,21 @@ static IMergeTreeDataPart::Checksums checkDataPart( projections_on_disk.erase(projection_file); } - if (throw_on_broken_projection && !broken_projections_message.empty()) + if (throw_on_broken_projection) { - throw Exception(ErrorCodes::BROKEN_PROJECTION, "{}", broken_projections_message); - } + if (!broken_projections_message.empty()) + { + throw Exception(ErrorCodes::BROKEN_PROJECTION, "{}", broken_projections_message); + } - if (require_checksums && !projections_on_disk.empty()) - { - throw Exception(ErrorCodes::UNEXPECTED_FILE_IN_DATA_PART, - "Found unexpected projection directories: {}", - fmt::join(projections_on_disk, ",")); + /// This one is actually not broken, just redundant files on disk which + /// MergeTree will never use. + if (require_checksums && !projections_on_disk.empty()) + { + throw Exception(ErrorCodes::UNEXPECTED_FILE_IN_DATA_PART, + "Found unexpected projection directories: {}", + fmt::join(projections_on_disk, ",")); + } } if (is_cancelled()) diff --git a/src/Storages/MergeTree/registerStorageMergeTree.cpp b/src/Storages/MergeTree/registerStorageMergeTree.cpp index d234103e52b..3f0603f6900 100644 --- a/src/Storages/MergeTree/registerStorageMergeTree.cpp +++ b/src/Storages/MergeTree/registerStorageMergeTree.cpp @@ -1,11 +1,14 @@ #include #include #include +#include #include #include #include #include +#include +#include #include #include #include diff --git a/src/Storages/MergeTree/tests/gtest_executor.cpp b/src/Storages/MergeTree/tests/gtest_executor.cpp index 6f34eb4dfbd..c7057ce87c6 100644 --- a/src/Storages/MergeTree/tests/gtest_executor.cpp +++ b/src/Storages/MergeTree/tests/gtest_executor.cpp @@ -34,7 +34,7 @@ public: auto choice = distribution(generator); if (choice == 0) - throw std::runtime_error("Unlucky..."); + throw TestException(); return false; } @@ -48,7 +48,7 @@ public: { auto choice = distribution(generator); if (choice == 0) - throw std::runtime_error("Unlucky..."); + throw TestException(); } Priority getPriority() const override { return {}; } diff --git a/src/Storages/MessageQueueSink.cpp b/src/Storages/MessageQueueSink.cpp index 4fb81d69070..36899011e33 100644 --- a/src/Storages/MessageQueueSink.cpp +++ b/src/Storages/MessageQueueSink.cpp @@ -40,7 +40,7 @@ void MessageQueueSink::onFinish() producer->finish(); } -void MessageQueueSink::consume(Chunk chunk) +void MessageQueueSink::consume(Chunk & chunk) { const auto & columns = chunk.getColumns(); if (columns.empty()) diff --git a/src/Storages/MessageQueueSink.h b/src/Storages/MessageQueueSink.h index b3c1e61734f..4a9248c6c4d 100644 --- a/src/Storages/MessageQueueSink.h +++ b/src/Storages/MessageQueueSink.h @@ -35,7 +35,7 @@ public: String getName() const override { return storage_name + "Sink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onStart() override; void onFinish() override; diff --git a/src/Storages/MySQL/MySQLSettings.cpp b/src/Storages/MySQL/MySQLSettings.cpp index fd53174f4f6..1c88e4a94e9 100644 --- a/src/Storages/MySQL/MySQLSettings.cpp +++ b/src/Storages/MySQL/MySQLSettings.cpp @@ -6,6 +6,7 @@ #include #include #include +#include namespace DB diff --git a/src/Storages/NATS/StorageNATS.cpp b/src/Storages/NATS/StorageNATS.cpp index 0b88a9e8929..9d728c3395f 100644 --- a/src/Storages/NATS/StorageNATS.cpp +++ b/src/Storages/NATS/StorageNATS.cpp @@ -49,6 +49,7 @@ StorageNATS::StorageNATS( const StorageID & table_id_, ContextPtr context_, const ColumnsDescription & columns_, + const String & comment, std::unique_ptr nats_settings_, LoadingStrictnessLevel mode) : IStorage(table_id_) @@ -87,6 +88,7 @@ StorageNATS::StorageNATS( StorageInMemoryMetadata storage_metadata; storage_metadata.setColumns(columns_); + storage_metadata.setComment(comment); setInMemoryMetadata(storage_metadata); setVirtuals(createVirtuals(nats_settings->nats_handle_error_mode)); @@ -644,7 +646,13 @@ bool StorageNATS::streamToViews() insert->table_id = table_id; // Only insert into dependent views and expect that input blocks contain virtual columns - InterpreterInsertQuery interpreter(insert, nats_context, false, true, true); + InterpreterInsertQuery interpreter( + insert, + nats_context, + /* allow_materialized */ false, + /* no_squash */ true, + /* no_destination */ true, + /* async_isnert */ false); auto block_io = interpreter.execute(); auto storage_snapshot = getStorageSnapshot(getInMemoryMetadataPtr(), getContext()); @@ -754,7 +762,7 @@ void registerStorageNATS(StorageFactory & factory) if (!nats_settings->nats_subjects.changed) throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "You must specify `nats_subjects` setting"); - return std::make_shared(args.table_id, args.getContext(), args.columns, std::move(nats_settings), args.mode); + return std::make_shared(args.table_id, args.getContext(), args.columns, args.comment, std::move(nats_settings), args.mode); }; factory.registerStorage("NATS", creator_fn, StorageFactory::StorageFeatures{ .supports_settings = true, }); diff --git a/src/Storages/NATS/StorageNATS.h b/src/Storages/NATS/StorageNATS.h index 41d77acfde6..5fca8cb0163 100644 --- a/src/Storages/NATS/StorageNATS.h +++ b/src/Storages/NATS/StorageNATS.h @@ -23,6 +23,7 @@ public: const StorageID & table_id_, ContextPtr context_, const ColumnsDescription & columns_, + const String & comment, std::unique_ptr nats_settings_, LoadingStrictnessLevel mode); diff --git a/src/Storages/NamedCollectionsHelpers.cpp b/src/Storages/NamedCollectionsHelpers.cpp index ba90f21c907..b58aed573ae 100644 --- a/src/Storages/NamedCollectionsHelpers.cpp +++ b/src/Storages/NamedCollectionsHelpers.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include @@ -117,7 +118,7 @@ MutableNamedCollectionPtr tryGetNamedCollectionWithOverrides( if (asts.size() == 1) return collection_copy; - const auto allow_override_by_default = context->getSettings().allow_named_collection_override_by_default; + const auto allow_override_by_default = context->getSettingsRef().allow_named_collection_override_by_default; for (auto * it = std::next(asts.begin()); it != asts.end(); ++it) { @@ -162,7 +163,7 @@ MutableNamedCollectionPtr tryGetNamedCollectionWithOverrides( Poco::Util::AbstractConfiguration::Keys keys; config.keys(config_prefix, keys); - const auto allow_override_by_default = context->getSettings().allow_named_collection_override_by_default; + const auto allow_override_by_default = context->getSettingsRef().allow_named_collection_override_by_default; for (const auto & key : keys) { if (collection_copy->isOverridable(key, allow_override_by_default)) diff --git a/src/Storages/NamedCollectionsHelpers.h b/src/Storages/NamedCollectionsHelpers.h index b4aea096c59..f444a581eb6 100644 --- a/src/Storages/NamedCollectionsHelpers.h +++ b/src/Storages/NamedCollectionsHelpers.h @@ -9,14 +9,14 @@ #include +namespace DB +{ + namespace ErrorCodes { extern const int BAD_ARGUMENTS; } -namespace DB -{ - /// Helper function to get named collection for table engine. /// Table engines have collection name as first argument of ast and other arguments are key-value overrides. MutableNamedCollectionPtr tryGetNamedCollectionWithOverrides( diff --git a/src/Storages/ObjectStorage/Azure/Configuration.cpp b/src/Storages/ObjectStorage/Azure/Configuration.cpp index 163f08be420..f0a0a562b92 100644 --- a/src/Storages/ObjectStorage/Azure/Configuration.cpp +++ b/src/Storages/ObjectStorage/Azure/Configuration.cpp @@ -1,4 +1,5 @@ #include +#include #if USE_AZURE_BLOB_STORAGE @@ -12,6 +13,8 @@ #include #include #include +#include +#include #include #include @@ -39,72 +42,19 @@ const std::unordered_set optional_configuration_keys = { "storage_account_url", }; -using AzureClient = Azure::Storage::Blobs::BlobContainerClient; -using AzureClientPtr = std::unique_ptr; - -namespace -{ - bool isConnectionString(const std::string & candidate) - { - return !candidate.starts_with("http"); - } - - template - bool containerExists(T & blob_service_client, const std::string & container_name) - { - Azure::Storage::Blobs::ListBlobContainersOptions options; - options.Prefix = container_name; - options.PageSizeHint = 1; - - auto containers_list_response = blob_service_client.ListBlobContainers(options); - auto containers_list = containers_list_response.BlobContainers; - - auto it = std::find_if( - containers_list.begin(), containers_list.end(), - [&](const auto & c) { return c.Name == container_name; }); - return it != containers_list.end(); - } -} - -Poco::URI StorageAzureConfiguration::getConnectionURL() const -{ - if (!is_connection_string) - return Poco::URI(connection_url); - - auto parsed_connection_string = Azure::Storage::_internal::ParseConnectionString(connection_url); - return Poco::URI(parsed_connection_string.BlobServiceUrl.GetAbsoluteUrl()); -} - void StorageAzureConfiguration::check(ContextPtr context) const { - context->getGlobalContext()->getRemoteHostFilter().checkURL(getConnectionURL()); + auto url = Poco::URI(connection_params.getConnectionURL()); + context->getGlobalContext()->getRemoteHostFilter().checkURL(url); Configuration::check(context); } StorageAzureConfiguration::StorageAzureConfiguration(const StorageAzureConfiguration & other) : Configuration(other) { - connection_url = other.connection_url; - is_connection_string = other.is_connection_string; - account_name = other.account_name; - account_key = other.account_key; - container = other.container; blob_path = other.blob_path; blobs_paths = other.blobs_paths; -} - -AzureObjectStorage::SettingsPtr StorageAzureConfiguration::createSettings(ContextPtr context) -{ - const auto & context_settings = context->getSettingsRef(); - auto settings_ptr = std::make_unique(); - settings_ptr->max_single_part_upload_size = context_settings.azure_max_single_part_upload_size; - settings_ptr->max_single_read_retries = context_settings.azure_max_single_read_retries; - settings_ptr->list_object_keys_size = static_cast(context_settings.azure_list_object_keys_size); - settings_ptr->strict_upload_part_size = context_settings.azure_strict_upload_part_size; - settings_ptr->max_upload_part_size = context_settings.azure_max_upload_part_size; - settings_ptr->max_blocks_in_multipart_upload = context_settings.azure_max_blocks_in_multipart_upload; - settings_ptr->min_upload_part_size = context_settings.azure_min_upload_part_size; - return settings_ptr; + connection_params = other.connection_params; } StorageObjectStorage::QuerySettings StorageAzureConfiguration::getQuerySettings(const ContextPtr & context) const @@ -125,147 +75,59 @@ StorageObjectStorage::QuerySettings StorageAzureConfiguration::getQuerySettings( ObjectStoragePtr StorageAzureConfiguration::createObjectStorage(ContextPtr context, bool is_readonly) /// NOLINT { assertInitialized(); - auto client = createClient(is_readonly, /* attempt_to_create_container */true); - auto settings = createSettings(context); + + auto settings = AzureBlobStorage::getRequestSettings(context->getSettingsRef()); + auto client = AzureBlobStorage::getContainerClient(connection_params, is_readonly); + return std::make_unique( - "AzureBlobStorage", std::move(client), std::move(settings), container, getConnectionURL().toString()); + "AzureBlobStorage", + connection_params.createForContainer(), + std::move(settings), + connection_params.getContainer(), + connection_params.getConnectionURL()); } -AzureClientPtr StorageAzureConfiguration::createClient(bool is_read_only, bool attempt_to_create_container) +static AzureBlobStorage::ConnectionParams getConnectionParams( + const String & connection_url, + const String & container_name, + const std::optional & account_name, + const std::optional & account_key, + const ContextPtr & local_context) { - using namespace Azure::Storage::Blobs; + AzureBlobStorage::ConnectionParams connection_params; + auto request_settings = AzureBlobStorage::getRequestSettings(local_context->getSettingsRef()); - AzureClientPtr result; - - if (is_connection_string) + if (account_name && account_key) { - auto managed_identity_credential = std::make_shared(); - auto blob_service_client = std::make_unique(BlobServiceClient::CreateFromConnectionString(connection_url)); - result = std::make_unique(BlobContainerClient::CreateFromConnectionString(connection_url, container)); - - if (attempt_to_create_container) - { - bool container_exists = containerExists(*blob_service_client, container); - if (!container_exists) - { - if (is_read_only) - throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "AzureBlobStorage container does not exist '{}'", - container); - - try - { - result->CreateIfNotExists(); - } - catch (const Azure::Storage::StorageException & e) - { - if (!(e.StatusCode == Azure::Core::Http::HttpStatusCode::Conflict - && e.ReasonPhrase == "The specified container already exists.")) - { - throw; - } - } - } - } + connection_params.endpoint.storage_account_url = connection_url; + connection_params.endpoint.container_name = container_name; + connection_params.auth_method = std::make_shared(*account_name, *account_key); + connection_params.client_options = AzureBlobStorage::getClientOptions(*request_settings, /*for_disk=*/ false); } else { - std::shared_ptr storage_shared_key_credential; - if (account_name.has_value() && account_key.has_value()) - { - storage_shared_key_credential - = std::make_shared(*account_name, *account_key); - } - - std::unique_ptr blob_service_client; - std::shared_ptr managed_identity_credential; - if (storage_shared_key_credential) - { - blob_service_client = std::make_unique(connection_url, storage_shared_key_credential); - } - else - { - managed_identity_credential = std::make_shared(); - blob_service_client = std::make_unique(connection_url, managed_identity_credential); - } - - std::string final_url; - size_t pos = connection_url.find('?'); - if (pos != std::string::npos) - { - auto url_without_sas = connection_url.substr(0, pos); - final_url = url_without_sas + (url_without_sas.back() == '/' ? "" : "/") + container - + connection_url.substr(pos); - } - else - final_url - = connection_url + (connection_url.back() == '/' ? "" : "/") + container; - - if (!attempt_to_create_container) - { - if (storage_shared_key_credential) - return std::make_unique(final_url, storage_shared_key_credential); - else - return std::make_unique(final_url, managed_identity_credential); - } - - bool container_exists = containerExists(*blob_service_client, container); - if (container_exists) - { - if (storage_shared_key_credential) - result = std::make_unique(final_url, storage_shared_key_credential); - else - result = std::make_unique(final_url, managed_identity_credential); - } - else - { - if (is_read_only) - throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "AzureBlobStorage container does not exist '{}'", - container); - try - { - result = std::make_unique(blob_service_client->CreateBlobContainer(container).Value); - } catch (const Azure::Storage::StorageException & e) - { - if (e.StatusCode == Azure::Core::Http::HttpStatusCode::Conflict - && e.ReasonPhrase == "The specified container already exists.") - { - if (storage_shared_key_credential) - result = std::make_unique(final_url, storage_shared_key_credential); - else - result = std::make_unique(final_url, managed_identity_credential); - } - else - { - throw; - } - } - } + AzureBlobStorage::processURL(connection_url, container_name, connection_params.endpoint, connection_params.auth_method); + connection_params.client_options = AzureBlobStorage::getClientOptions(*request_settings, /*for_disk=*/ false); } - return result; + return connection_params; } - void StorageAzureConfiguration::fromNamedCollection(const NamedCollection & collection, ContextPtr) +void StorageAzureConfiguration::fromNamedCollection(const NamedCollection & collection, ContextPtr context) { validateNamedCollection(collection, required_configuration_keys, optional_configuration_keys); + String connection_url; + String container_name; + std::optional account_name; + std::optional account_key; + if (collection.has("connection_string")) - { connection_url = collection.get("connection_string"); - is_connection_string = true; - } - - if (collection.has("storage_account_url")) - { + else if (collection.has("storage_account_url")) connection_url = collection.get("storage_account_url"); - is_connection_string = false; - } - container = collection.get("container"); + container_name = collection.get("container"); blob_path = collection.get("blob_path"); if (collection.has("account_name")) @@ -279,6 +141,7 @@ AzureClientPtr StorageAzureConfiguration::createClient(bool is_read_only, bool a compression_method = collection.getOrDefault("compression_method", collection.getOrDefault("compression", "auto")); blobs_paths = {blob_path}; + connection_params = getConnectionParams(connection_url, container_name, account_name, account_key, context); } void StorageAzureConfiguration::fromAST(ASTs & engine_args, ContextPtr context, bool with_structure) @@ -296,12 +159,14 @@ void StorageAzureConfiguration::fromAST(ASTs & engine_args, ContextPtr context, std::unordered_map engine_args_to_idx; - connection_url = checkAndGetLiteralArgument(engine_args[0], "connection_string/storage_account_url"); - is_connection_string = isConnectionString(connection_url); - container = checkAndGetLiteralArgument(engine_args[1], "container"); + String connection_url = checkAndGetLiteralArgument(engine_args[0], "connection_string/storage_account_url"); + String container_name = checkAndGetLiteralArgument(engine_args[1], "container"); blob_path = checkAndGetLiteralArgument(engine_args[2], "blobpath"); + std::optional account_name; + std::optional account_key; + auto is_format_arg = [] (const std::string & s) -> bool { return s == "auto" || FormatFactory::instance().getAllFormats().contains(Poco::toLower(s)); @@ -358,7 +223,9 @@ void StorageAzureConfiguration::fromAST(ASTs & engine_args, ContextPtr context, account_key = checkAndGetLiteralArgument(engine_args[4], "account_key"); auto sixth_arg = checkAndGetLiteralArgument(engine_args[5], "format/account_name"); if (is_format_arg(sixth_arg)) + { format = sixth_arg; + } else { if (with_structure) @@ -400,6 +267,7 @@ void StorageAzureConfiguration::fromAST(ASTs & engine_args, ContextPtr context, } blobs_paths = {blob_path}; + connection_params = getConnectionParams(connection_url, container_name, account_name, account_key, context); } void StorageAzureConfiguration::addStructureAndFormatToArgs( diff --git a/src/Storages/ObjectStorage/Azure/Configuration.h b/src/Storages/ObjectStorage/Azure/Configuration.h index bbaa82c51ba..4e6bfbc0745 100644 --- a/src/Storages/ObjectStorage/Azure/Configuration.h +++ b/src/Storages/ObjectStorage/Azure/Configuration.h @@ -35,8 +35,8 @@ public: const Paths & getPaths() const override { return blobs_paths; } void setPaths(const Paths & paths) override { blobs_paths = paths; } - String getNamespace() const override { return container; } - String getDataSourceDescription() const override { return std::filesystem::path(connection_url) / container; } + String getNamespace() const override { return connection_params.getContainer(); } + String getDataSourceDescription() const override { return std::filesystem::path(connection_params.getConnectionURL()) / connection_params.getContainer(); } StorageObjectStorage::QuerySettings getQuerySettings(const ContextPtr &) const override; void check(ContextPtr context) const override; @@ -54,22 +54,9 @@ protected: void fromNamedCollection(const NamedCollection & collection, ContextPtr context) override; void fromAST(ASTs & args, ContextPtr context, bool with_structure) override; - using AzureClient = Azure::Storage::Blobs::BlobContainerClient; - using AzureClientPtr = std::unique_ptr; - - std::string connection_url; - bool is_connection_string; - - std::optional account_name; - std::optional account_key; - - std::string container; std::string blob_path; std::vector blobs_paths; - - AzureClientPtr createClient(bool is_read_only, bool attempt_to_create_container); - AzureObjectStorage::SettingsPtr createSettings(ContextPtr local_context); - Poco::URI getConnectionURL() const; + AzureBlobStorage::ConnectionParams connection_params; }; } diff --git a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp index 38bf3112ee2..c896a760597 100644 --- a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp @@ -4,19 +4,43 @@ #include #if USE_AWS_S3 && USE_PARQUET -#include + +#include +#include +#include +#include +#include + +#include +#include +#include + #include #include #include -#include -#include -#include -#include -#include -#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + #include +#include #include -#include +#include +#include +#include + +namespace fs = std::filesystem; namespace DB { @@ -25,10 +49,14 @@ namespace ErrorCodes { extern const int INCORRECT_DATA; extern const int BAD_ARGUMENTS; + extern const int LOGICAL_ERROR; + extern const int NOT_IMPLEMENTED; } -struct DeltaLakeMetadata::Impl +struct DeltaLakeMetadataImpl { + using ConfigurationPtr = DeltaLakeMetadata::ConfigurationPtr; + ObjectStoragePtr object_storage; ConfigurationPtr configuration; ContextPtr context; @@ -37,7 +65,7 @@ struct DeltaLakeMetadata::Impl * Useful links: * - https://github.com/delta-io/delta/blob/master/PROTOCOL.md#data-files */ - Impl(ObjectStoragePtr object_storage_, + DeltaLakeMetadataImpl(ObjectStoragePtr object_storage_, ConfigurationPtr configuration_, ContextPtr context_) : object_storage(object_storage_) @@ -74,10 +102,18 @@ struct DeltaLakeMetadata::Impl * An action changes one aspect of the table's state, for example, adding or removing a file. * Note: it is not a valid json, but a list of json's, so we read it in a while cycle. */ - std::set processMetadataFiles() + struct DeltaLakeMetadata + { + NamesAndTypesList schema; + Strings data_files; + DataLakePartitionColumns partition_columns; + }; + DeltaLakeMetadata processMetadataFiles() { std::set result_files; - const auto checkpoint_version = getCheckpointIfExists(result_files); + NamesAndTypesList current_schema; + DataLakePartitionColumns current_partition_columns; + const auto checkpoint_version = getCheckpointIfExists(result_files, current_schema, current_partition_columns); if (checkpoint_version) { @@ -90,7 +126,7 @@ struct DeltaLakeMetadata::Impl if (!object_storage->exists(StoredObject(file_path))) break; - processMetadataFile(file_path, result_files); + processMetadataFile(file_path, current_schema, current_partition_columns, result_files); } LOG_TRACE( @@ -101,10 +137,10 @@ struct DeltaLakeMetadata::Impl { const auto keys = listFiles(*object_storage, *configuration, deltalake_metadata_directory, metadata_file_suffix); for (const String & key : keys) - processMetadataFile(key, result_files); + processMetadataFile(key, current_schema, current_partition_columns, result_files); } - return result_files; + return DeltaLakeMetadata{current_schema, Strings(result_files.begin(), result_files.end()), current_partition_columns}; } /** @@ -136,10 +172,20 @@ struct DeltaLakeMetadata::Impl * \"nullCount\":{\"col-6c990940-59bb-4709-8f2e-17083a82c01a\":0,\"col-763cd7e2-7627-4d8e-9fb7-9e85d0c8845b\":0}}"}} * " */ - void processMetadataFile(const String & key, std::set & result) const + + /// Read metadata file and fill `file_schema`, `file_parition_columns`, `result`. + /// `result` is a list of data files. + /// `file_schema` is a common schema for all files. + /// Schema evolution is not supported, so we check that all files have the same schema. + /// `file_partiion_columns` is information about partition columns of data files. + void processMetadataFile( + const String & metadata_file_path, + NamesAndTypesList & file_schema, + DataLakePartitionColumns & file_partition_columns, + std::set & result) { auto read_settings = context->getReadSettings(); - auto buf = object_storage->readObject(StoredObject(key), read_settings); + auto buf = object_storage->readObject(StoredObject(metadata_file_path), read_settings); char c; while (!buf->eof()) @@ -157,20 +203,250 @@ struct DeltaLakeMetadata::Impl if (json_str.empty()) continue; - const JSON json(json_str); - if (json.has("add")) + Poco::JSON::Parser parser; + Poco::Dynamic::Var json = parser.parse(json_str); + Poco::JSON::Object::Ptr object = json.extract(); + + std::ostringstream oss; // STYLE_CHECK_ALLOW_STD_STRING_STREAM + object->stringify(oss); + LOG_TEST(log, "Metadata: {}", oss.str()); + + if (object->has("metaData")) { - const auto path = json["add"]["path"].getString(); - result.insert(std::filesystem::path(configuration->getPath()) / path); + const auto metadata_object = object->get("metaData").extract(); + const auto schema_object = metadata_object->getValue("schemaString"); + + Poco::JSON::Parser p; + Poco::Dynamic::Var fields_json = parser.parse(schema_object); + const Poco::JSON::Object::Ptr & fields_object = fields_json.extract(); + + auto current_schema = parseMetadata(fields_object); + if (file_schema.empty()) + { + file_schema = current_schema; + } + else if (file_schema != current_schema) + { + throw Exception(ErrorCodes::NOT_IMPLEMENTED, + "Reading from files with different schema is not possible " + "({} is different from {})", + file_schema.toString(), current_schema.toString()); + } } - else if (json.has("remove")) + + if (object->has("add")) { - const auto path = json["remove"]["path"].getString(); - result.erase(std::filesystem::path(configuration->getPath()) / path); + auto add_object = object->get("add").extract(); + auto path = add_object->getValue("path"); + result.insert(fs::path(configuration->getPath()) / path); + + auto filename = fs::path(path).filename().string(); + auto it = file_partition_columns.find(filename); + if (it == file_partition_columns.end()) + { + if (add_object->has("partitionValues")) + { + auto partition_values = add_object->get("partitionValues").extract(); + if (partition_values->size()) + { + auto & current_partition_columns = file_partition_columns[filename]; + for (const auto & partition_name : partition_values->getNames()) + { + const auto value = partition_values->getValue(partition_name); + auto name_and_type = file_schema.tryGetByName(partition_name); + if (!name_and_type) + { + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "No such column in schema: {} (schema: {})", + partition_name, file_schema.toNamesAndTypesDescription()); + } + + auto field = getFieldValue(value, name_and_type->type); + current_partition_columns.emplace_back(*name_and_type, field); + + LOG_TEST(log, "Partition {} value is {} (for {})", partition_name, value, filename); + } + } + } + } + } + else if (object->has("remove")) + { + auto path = object->get("remove").extract()->getValue("path"); + result.erase(fs::path(configuration->getPath()) / path); } } } + NamesAndTypesList parseMetadata(const Poco::JSON::Object::Ptr & metadata_json) + { + NamesAndTypesList schema; + const auto fields = metadata_json->get("fields").extract(); + for (size_t i = 0; i < fields->size(); ++i) + { + const auto field = fields->getObject(static_cast(i)); + auto column_name = field->getValue("name"); + auto type = field->getValue("type"); + auto is_nullable = field->getValue("nullable"); + + std::string physical_name; + auto schema_metadata_object = field->get("metadata").extract(); + if (schema_metadata_object->has("delta.columnMapping.physicalName")) + physical_name = schema_metadata_object->getValue("delta.columnMapping.physicalName"); + else + physical_name = column_name; + + LOG_TEST(log, "Found column: {}, type: {}, nullable: {}, physical name: {}", + column_name, type, is_nullable, physical_name); + + schema.push_back({physical_name, getFieldType(field, "type", is_nullable)}); + } + return schema; + } + + DataTypePtr getFieldType(const Poco::JSON::Object::Ptr & field, const String & type_key, bool is_nullable) + { + if (field->isObject(type_key)) + return getComplexTypeFromObject(field->getObject(type_key)); + + auto type = field->get(type_key); + if (type.isString()) + { + const String & type_name = type.extract(); + auto data_type = getSimpleTypeByName(type_name); + return is_nullable ? makeNullable(data_type) : data_type; + } + + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected 'type' field: {}", type.toString()); + } + + Field getFieldValue(const String & value, DataTypePtr data_type) + { + DataTypePtr check_type; + if (data_type->isNullable()) + check_type = static_cast(data_type.get())->getNestedType(); + else + check_type = data_type; + + WhichDataType which(check_type->getTypeId()); + if (which.isStringOrFixedString()) + return value; + else if (which.isInt8()) + return parse(value); + else if (which.isUInt8()) + return parse(value); + else if (which.isInt16()) + return parse(value); + else if (which.isUInt16()) + return parse(value); + else if (which.isInt32()) + return parse(value); + else if (which.isUInt32()) + return parse(value); + else if (which.isInt64()) + return parse(value); + else if (which.isUInt64()) + return parse(value); + else if (which.isFloat32()) + return parse(value); + else if (which.isFloat64()) + return parse(value); + else if (which.isDate()) + return UInt16{LocalDate{std::string(value)}.getDayNum()}; + else if (which.isDate32()) + return Int32{LocalDate{std::string(value)}.getExtenedDayNum()}; + else if (which.isDateTime64()) + { + ReadBufferFromString in(value); + DateTime64 time = 0; + readDateTime64Text(time, 6, in, assert_cast(data_type.get())->getTimeZone()); + return time; + } + + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unsupported DeltaLake type for {}", check_type->getColumnType()); + } + + DataTypePtr getSimpleTypeByName(const String & type_name) + { + /// https://github.com/delta-io/delta/blob/master/PROTOCOL.md#primitive-types + + if (type_name == "string" || type_name == "binary") + return std::make_shared(); + if (type_name == "long") + return std::make_shared(); + if (type_name == "integer") + return std::make_shared(); + if (type_name == "short") + return std::make_shared(); + if (type_name == "byte") + return std::make_shared(); + if (type_name == "float") + return std::make_shared(); + if (type_name == "double") + return std::make_shared(); + if (type_name == "boolean") + return DataTypeFactory::instance().get("Bool"); + if (type_name == "date") + return std::make_shared(); + if (type_name == "timestamp") + return std::make_shared(6); + if (type_name.starts_with("decimal(") && type_name.ends_with(')')) + { + ReadBufferFromString buf(std::string_view(type_name.begin() + 8, type_name.end() - 1)); + size_t precision; + size_t scale; + readIntText(precision, buf); + skipWhitespaceIfAny(buf); + assertChar(',', buf); + skipWhitespaceIfAny(buf); + tryReadIntText(scale, buf); + return createDecimal(precision, scale); + } + + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unsupported DeltaLake type: {}", type_name); + } + + DataTypePtr getComplexTypeFromObject(const Poco::JSON::Object::Ptr & type) + { + String type_name = type->getValue("type"); + + if (type_name == "struct") + { + DataTypes element_types; + Names element_names; + auto fields = type->get("fields").extract(); + element_types.reserve(fields->size()); + element_names.reserve(fields->size()); + for (size_t i = 0; i != fields->size(); ++i) + { + auto field = fields->getObject(static_cast(i)); + element_names.push_back(field->getValue("name")); + auto required = field->getValue("required"); + element_types.push_back(getFieldType(field, "type", required)); + } + + return std::make_shared(element_types, element_names); + } + + if (type_name == "array") + { + bool is_nullable = type->getValue("containsNull"); + auto element_type = getFieldType(type, "elementType", is_nullable); + return std::make_shared(element_type); + } + + if (type_name == "map") + { + bool is_nullable = type->getValue("containsNull"); + auto key_type = getFieldType(type, "keyType", /* is_nullable */false); + auto value_type = getFieldType(type, "valueType", is_nullable); + return std::make_shared(key_type, value_type); + } + + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unsupported DeltaLake type: {}", type_name); + } + /** * Checkpoints in delta-lake are created each 10 commits by default. * Latest checkpoint is written in _last_checkpoint file: _delta_log/_last_checkpoint @@ -242,7 +518,10 @@ struct DeltaLakeMetadata::Impl throw Exception(ErrorCodes::BAD_ARGUMENTS, "Arrow error: {}", _s.ToString()); \ } while (false) - size_t getCheckpointIfExists(std::set & result) + size_t getCheckpointIfExists( + std::set & result, + NamesAndTypesList & file_schema, + DataLakePartitionColumns & file_partition_columns) { const auto version = readLastCheckpointIfExists(); if (!version) @@ -263,7 +542,8 @@ struct DeltaLakeMetadata::Impl auto columns = ParquetSchemaReader(*buf, format_settings).readSchema(); /// Read only columns that we need. - columns.filterColumns(NameSet{"add", "remove"}); + auto filter_column_names = NameSet{"add", "metaData"}; + columns.filterColumns(filter_column_names); Block header; for (const auto & column : columns) header.insert({column.type->createColumn(), column.type, column.name}); @@ -277,9 +557,6 @@ struct DeltaLakeMetadata::Impl ArrowMemoryPool::instance(), &reader)); - std::shared_ptr schema; - THROW_ARROW_NOT_OK(reader->GetSchema(&schema)); - ArrowColumnToCHColumn column_reader( header, "Parquet", format_settings.parquet.allow_missing_columns, @@ -290,29 +567,85 @@ struct DeltaLakeMetadata::Impl std::shared_ptr table; THROW_ARROW_NOT_OK(reader->ReadTable(&table)); - Chunk res = column_reader.arrowTableToCHChunk(table, reader->parquet_reader()->metadata()->num_rows()); - const auto & res_columns = res.getColumns(); + Chunk chunk = column_reader.arrowTableToCHChunk(table, reader->parquet_reader()->metadata()->num_rows()); + auto res_block = header.cloneWithColumns(chunk.detachColumns()); + res_block = Nested::flatten(res_block); - if (res_columns.size() != 2) - { - throw Exception( - ErrorCodes::INCORRECT_DATA, - "Unexpected number of columns: {} (having: {}, expected: {})", - res_columns.size(), res.dumpStructure(), header.dumpStructure()); - } + const auto * nullable_path_column = assert_cast(res_block.getByName("add.path").column.get()); + const auto & path_column = assert_cast(nullable_path_column->getNestedColumn()); + + const auto * nullable_schema_column = assert_cast(res_block.getByName("metaData.schemaString").column.get()); + const auto & schema_column = assert_cast(nullable_schema_column->getNestedColumn()); + + auto partition_values_column_raw = res_block.getByName("add.partitionValues").column; + const auto & partition_values_column = assert_cast(*partition_values_column_raw); - const auto * tuple_column = assert_cast(res_columns[0].get()); - const auto & nullable_column = assert_cast(tuple_column->getColumn(0)); - const auto & path_column = assert_cast(nullable_column.getNestedColumn()); for (size_t i = 0; i < path_column.size(); ++i) { - const auto filename = String(path_column.getDataAt(i)); - if (filename.empty()) + const auto metadata = String(schema_column.getDataAt(i)); + if (!metadata.empty()) + { + Poco::JSON::Parser parser; + Poco::Dynamic::Var json = parser.parse(metadata); + const Poco::JSON::Object::Ptr & object = json.extract(); + + auto current_schema = parseMetadata(object); + if (file_schema.empty()) + { + file_schema = current_schema; + LOG_TEST(log, "Processed schema from checkpoint: {}", file_schema.toString()); + } + else if (file_schema != current_schema) + { + throw Exception(ErrorCodes::NOT_IMPLEMENTED, + "Reading from files with different schema is not possible " + "({} is different from {})", + file_schema.toString(), current_schema.toString()); + } + } + } + + for (size_t i = 0; i < path_column.size(); ++i) + { + const auto path = String(path_column.getDataAt(i)); + if (path.empty()) continue; - LOG_TEST(log, "Adding {}", filename); - const auto [_, inserted] = result.insert(std::filesystem::path(configuration->getPath()) / filename); + + auto filename = fs::path(path).filename().string(); + auto it = file_partition_columns.find(filename); + if (it == file_partition_columns.end()) + { + Field map; + partition_values_column.get(i, map); + auto partition_values_map = map.safeGet(); + if (!partition_values_map.empty()) + { + auto & current_partition_columns = file_partition_columns[filename]; + for (const auto & map_value : partition_values_map) + { + const auto tuple = map_value.safeGet(); + const auto partition_name = tuple[0].safeGet(); + auto name_and_type = file_schema.tryGetByName(partition_name); + if (!name_and_type) + { + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "No such column in schema: {} (schema: {})", + partition_name, file_schema.toString()); + } + const auto value = tuple[1].safeGet(); + auto field = getFieldValue(value, name_and_type->type); + current_partition_columns.emplace_back(std::move(name_and_type.value()), std::move(field)); + + LOG_TEST(log, "Partition {} value is {} (for {})", partition_name, value, filename); + } + } + } + + LOG_TEST(log, "Adding {}", path); + const auto [_, inserted] = result.insert(std::filesystem::path(configuration->getPath()) / path); if (!inserted) - throw Exception(ErrorCodes::INCORRECT_DATA, "File already exists {}", filename); + throw Exception(ErrorCodes::INCORRECT_DATA, "File already exists {}", path); } return version; @@ -325,18 +658,15 @@ DeltaLakeMetadata::DeltaLakeMetadata( ObjectStoragePtr object_storage_, ConfigurationPtr configuration_, ContextPtr context_) - : impl(std::make_unique(object_storage_, configuration_, context_)) { -} + auto impl = DeltaLakeMetadataImpl(object_storage_, configuration_, context_); + auto result = impl.processMetadataFiles(); + data_files = result.data_files; + schema = result.schema; + partition_columns = result.partition_columns; -Strings DeltaLakeMetadata::getDataFiles() const -{ - if (!data_files.empty()) - return data_files; - - auto result = impl->processMetadataFiles(); - data_files = Strings(result.begin(), result.end()); - return data_files; + LOG_TRACE(impl.log, "Found {} data files, {} partition files, schema: {}", + data_files.size(), partition_columns.size(), schema.toString()); } } diff --git a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h index e527721b29e..a479a3dd293 100644 --- a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h @@ -20,9 +20,13 @@ public: ConfigurationPtr configuration_, ContextPtr context_); - Strings getDataFiles() const override; + Strings getDataFiles() const override { return data_files; } - NamesAndTypesList getTableSchema() const override { return {}; } + NamesAndTypesList getTableSchema() const override { return schema; } + + const DataLakePartitionColumns & getPartitionColumns() const override { return partition_columns; } + + const std::unordered_map & getColumnNameToPhysicalNameMapping() const override { return column_name_to_physical_name; } bool operator ==(const IDataLakeMetadata & other) const override { @@ -41,9 +45,10 @@ public: } private: - struct Impl; - const std::shared_ptr impl; mutable Strings data_files; + NamesAndTypesList schema; + std::unordered_map column_name_to_physical_name; + DataLakePartitionColumns partition_columns; }; } diff --git a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h index 3ab274b1fbf..b060b1b0d39 100644 --- a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h @@ -26,6 +26,10 @@ public: NamesAndTypesList getTableSchema() const override { return {}; } + const DataLakePartitionColumns & getPartitionColumns() const override { return partition_columns; } + + const std::unordered_map & getColumnNameToPhysicalNameMapping() const override { return column_name_to_physical_name; } + bool operator ==(const IDataLakeMetadata & other) const override { const auto * hudi_metadata = dynamic_cast(&other); @@ -46,6 +50,8 @@ private: const ObjectStoragePtr object_storage; const ConfigurationPtr configuration; mutable Strings data_files; + std::unordered_map column_name_to_physical_name; + DataLakePartitionColumns partition_columns; Strings getDataFilesImpl() const; }; diff --git a/src/Storages/ObjectStorage/DataLakes/IDataLakeMetadata.h b/src/Storages/ObjectStorage/DataLakes/IDataLakeMetadata.h index a2bd5adb947..2954d50db91 100644 --- a/src/Storages/ObjectStorage/DataLakes/IDataLakeMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/IDataLakeMetadata.h @@ -2,6 +2,7 @@ #include #include #include +#include "PartitionColumns.h" namespace DB { @@ -13,6 +14,8 @@ public: virtual Strings getDataFiles() const = 0; virtual NamesAndTypesList getTableSchema() const = 0; virtual bool operator==(const IDataLakeMetadata & other) const = 0; + virtual const DataLakePartitionColumns & getPartitionColumns() const = 0; + virtual const std::unordered_map & getColumnNameToPhysicalNameMapping() const = 0; }; using DataLakeMetadataPtr = std::unique_ptr; diff --git a/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h b/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h index 83865c47eb8..c8603fccb86 100644 --- a/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h +++ b/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h @@ -41,6 +41,7 @@ public: auto object_storage = base_configuration->createObjectStorage(context, /* is_readonly */true); DataLakeMetadataPtr metadata; NamesAndTypesList schema_from_metadata; + const bool use_schema_from_metadata = columns_.empty(); if (base_configuration->format == "auto") base_configuration->format = "Parquet"; @@ -50,8 +51,9 @@ public: try { metadata = DataLakeMetadata::create(object_storage, base_configuration, context); - schema_from_metadata = metadata->getTableSchema(); configuration->setPaths(metadata->getDataFiles()); + if (use_schema_from_metadata) + schema_from_metadata = metadata->getTableSchema(); } catch (...) { @@ -66,7 +68,7 @@ public: return std::make_shared>( base_configuration, std::move(metadata), configuration, object_storage, context, table_id_, - columns_.empty() ? ColumnsDescription(schema_from_metadata) : columns_, + use_schema_from_metadata ? ColumnsDescription(schema_from_metadata) : columns_, constraints_, comment_, format_settings_); } @@ -81,7 +83,7 @@ public: auto metadata = DataLakeMetadata::create(object_storage_, base_configuration, local_context); auto schema_from_metadata = metadata->getTableSchema(); - if (schema_from_metadata != NamesAndTypesList{}) + if (!schema_from_metadata.empty()) { return ColumnsDescription(std::move(schema_from_metadata)); } @@ -99,13 +101,13 @@ public: Storage::updateConfiguration(local_context); auto new_metadata = DataLakeMetadata::create(Storage::object_storage, base_configuration, local_context); - if (current_metadata && *current_metadata == *new_metadata) return; current_metadata = std::move(new_metadata); auto updated_configuration = base_configuration->clone(); updated_configuration->setPaths(current_metadata->getDataFiles()); + updated_configuration->setPartitionColumns(current_metadata->getPartitionColumns()); Storage::configuration = updated_configuration; } @@ -123,11 +125,42 @@ public: { base_configuration->format = Storage::configuration->format; } + + if (current_metadata) + { + const auto & columns = current_metadata->getPartitionColumns(); + base_configuration->setPartitionColumns(columns); + Storage::configuration->setPartitionColumns(columns); + } } private: ConfigurationPtr base_configuration; DataLakeMetadataPtr current_metadata; + + ReadFromFormatInfo prepareReadingFromFormat( + const Strings & requested_columns, + const StorageSnapshotPtr & storage_snapshot, + bool supports_subset_of_columns, + ContextPtr local_context) override + { + auto info = DB::prepareReadingFromFormat(requested_columns, storage_snapshot, supports_subset_of_columns); + if (!current_metadata) + { + Storage::updateConfiguration(local_context); + current_metadata = DataLakeMetadata::create(Storage::object_storage, base_configuration, local_context); + } + auto column_mapping = current_metadata->getColumnNameToPhysicalNameMapping(); + if (!column_mapping.empty()) + { + for (const auto & [column_name, physical_name] : column_mapping) + { + auto & column = info.format_header.getByName(column_name); + column.name = physical_name; + } + } + return info; + } }; using StorageIceberg = IStorageDataLake; diff --git a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp index 0484f86542c..6d18b13df01 100644 --- a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp @@ -3,7 +3,7 @@ #if USE_AWS_S3 && USE_AVRO #include - +#include #include #include #include diff --git a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h index 06dbd373bf9..9476ac6e7d9 100644 --- a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h @@ -82,6 +82,10 @@ public: /// Get table schema parsed from metadata. NamesAndTypesList getTableSchema() const override { return schema; } + const std::unordered_map & getColumnNameToPhysicalNameMapping() const override { return column_name_to_physical_name; } + + const DataLakePartitionColumns & getPartitionColumns() const override { return partition_columns; } + bool operator ==(const IDataLakeMetadata & other) const override { const auto * iceberg_metadata = dynamic_cast(&other); @@ -104,6 +108,8 @@ private: Int32 current_schema_id; NamesAndTypesList schema; mutable Strings data_files; + std::unordered_map column_name_to_physical_name; + DataLakePartitionColumns partition_columns; LoggerPtr log; }; diff --git a/src/Storages/ObjectStorage/DataLakes/PartitionColumns.h b/src/Storages/ObjectStorage/DataLakes/PartitionColumns.h new file mode 100644 index 00000000000..eb605559145 --- /dev/null +++ b/src/Storages/ObjectStorage/DataLakes/PartitionColumns.h @@ -0,0 +1,19 @@ +#pragma once +#include +#include + +namespace DB +{ + +struct DataLakePartitionColumn +{ + NameAndTypePair name_and_type; + Field value; + + bool operator ==(const DataLakePartitionColumn & other) const = default; +}; + +/// Data file -> partition columns +using DataLakePartitionColumns = std::unordered_map>; + +} diff --git a/src/Storages/ObjectStorage/HDFS/Configuration.cpp b/src/Storages/ObjectStorage/HDFS/Configuration.cpp index 155f51adf61..e8071be6f02 100644 --- a/src/Storages/ObjectStorage/HDFS/Configuration.cpp +++ b/src/Storages/ObjectStorage/HDFS/Configuration.cpp @@ -2,6 +2,7 @@ #if USE_HDFS #include +#include #include #include #include diff --git a/src/Storages/ObjectStorage/ReadBufferIterator.cpp b/src/Storages/ObjectStorage/ReadBufferIterator.cpp index 78cdc442f64..df78f128c80 100644 --- a/src/Storages/ObjectStorage/ReadBufferIterator.cpp +++ b/src/Storages/ObjectStorage/ReadBufferIterator.cpp @@ -1,6 +1,7 @@ #include #include #include +#include #include @@ -162,7 +163,9 @@ ReadBufferIterator::Data ReadBufferIterator::next() { for (const auto & object_info : read_keys) { - if (auto format_from_file_name = FormatFactory::instance().tryGetFormatFromFileName(object_info->getFileName())) + auto format_from_file_name = FormatFactory::instance().tryGetFormatFromFileName(object_info->getFileName()); + /// Use this format only if we have a schema reader for it. + if (format_from_file_name && FormatFactory::instance().checkIfFormatHasAnySchemaReader(*format_from_file_name)) { format = format_from_file_name; break; @@ -220,7 +223,9 @@ ReadBufferIterator::Data ReadBufferIterator::next() { for (auto it = read_keys.begin() + prev_read_keys_size; it != read_keys.end(); ++it) { - if (auto format_from_file_name = FormatFactory::instance().tryGetFormatFromFileName((*it)->getFileName())) + auto format_from_file_name = FormatFactory::instance().tryGetFormatFromFileName((*it)->getFileName()); + /// Use this format only if we have a schema reader for it. + if (format_from_file_name && FormatFactory::instance().checkIfFormatHasAnySchemaReader(*format_from_file_name)) { format = format_from_file_name; break; diff --git a/src/Storages/ObjectStorage/S3/Configuration.cpp b/src/Storages/ObjectStorage/S3/Configuration.cpp index b33d55105e9..094ca069e7a 100644 --- a/src/Storages/ObjectStorage/S3/Configuration.cpp +++ b/src/Storages/ObjectStorage/S3/Configuration.cpp @@ -1,6 +1,7 @@ #include #if USE_AWS_S3 +#include #include #include #include diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index 90a97a9ea62..fe675a17d54 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include @@ -203,6 +204,15 @@ private: }; } +ReadFromFormatInfo StorageObjectStorage::prepareReadingFromFormat( + const Strings & requested_columns, + const StorageSnapshotPtr & storage_snapshot, + bool supports_subset_of_columns, + ContextPtr /* local_context */) +{ + return DB::prepareReadingFromFormat(requested_columns, storage_snapshot, supports_subset_of_columns); +} + void StorageObjectStorage::read( QueryPlan & query_plan, const Names & column_names, @@ -222,7 +232,7 @@ void StorageObjectStorage::read( } const auto read_from_format_info = prepareReadingFromFormat( - column_names, storage_snapshot, supportsSubsetOfColumns(local_context)); + column_names, storage_snapshot, supportsSubsetOfColumns(local_context), local_context); const bool need_only_count = (query_info.optimize_trivial_count || read_from_format_info.requested_columns.empty()) && local_context->getSettingsRef().optimize_count_from_files; @@ -451,6 +461,7 @@ StorageObjectStorage::Configuration::Configuration(const Configuration & other) format = other.format; compression_method = other.compression_method; structure = other.structure; + partition_columns = other.partition_columns; } bool StorageObjectStorage::Configuration::withPartitionWildcard() const diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.h b/src/Storages/ObjectStorage/StorageObjectStorage.h index cf8ec113653..818ce055c77 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.h +++ b/src/Storages/ObjectStorage/StorageObjectStorage.h @@ -1,10 +1,12 @@ #pragma once #include #include +#include #include #include #include #include +#include namespace DB { @@ -117,6 +119,12 @@ public: protected: virtual void updateConfiguration(ContextPtr local_context); + virtual ReadFromFormatInfo prepareReadingFromFormat( + const Strings & requested_columns, + const StorageSnapshotPtr & storage_snapshot, + bool supports_subset_of_columns, + ContextPtr local_context); + static std::unique_ptr createReadBufferIterator( const ObjectStoragePtr & object_storage, const ConfigurationPtr & configuration, @@ -188,6 +196,9 @@ public: virtual ConfigurationPtr clone() = 0; virtual bool isStaticConfiguration() const { return true; } + void setPartitionColumns(const DataLakePartitionColumns & columns) { partition_columns = columns; } + const DataLakePartitionColumns & getPartitionColumns() const { return partition_columns; } + String format = "auto"; String compression_method = "auto"; String structure = "auto"; @@ -199,6 +210,7 @@ protected: void assertInitialized() const; bool initialized = false; + DataLakePartitionColumns partition_columns; }; } diff --git a/src/Storages/ObjectStorage/StorageObjectStorageCluster.h b/src/Storages/ObjectStorage/StorageObjectStorageCluster.h index 69fec2b3c77..108aa109616 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageCluster.h +++ b/src/Storages/ObjectStorage/StorageObjectStorageCluster.h @@ -24,12 +24,6 @@ public: std::string getName() const override; - bool supportsSubcolumns() const override { return true; } - - bool supportsDynamicSubcolumns() const override { return true; } - - bool supportsTrivialCountOptimization(const StorageSnapshotPtr &, ContextPtr) const override { return true; } - RemoteQueryExecutor::Extension getTaskIteratorExtension( const ActionsDAG::Node * predicate, const ContextPtr & context) const override; diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSink.cpp b/src/Storages/ObjectStorage/StorageObjectStorageSink.cpp index f2f6eac333c..c864fb9a7e2 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSink.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageSink.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include namespace DB @@ -39,12 +40,12 @@ StorageObjectStorageSink::StorageObjectStorageSink( configuration->format, *write_buf, sample_block, context, format_settings_); } -void StorageObjectStorageSink::consume(Chunk chunk) +void StorageObjectStorageSink::consume(Chunk & chunk) { std::lock_guard lock(cancel_mutex); if (cancelled) return; - writer->write(getHeader().cloneWithColumns(chunk.detachColumns())); + writer->write(getHeader().cloneWithColumns(chunk.getColumns())); } void StorageObjectStorageSink::onCancel() diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSink.h b/src/Storages/ObjectStorage/StorageObjectStorageSink.h index e0081193686..6ab531bb21a 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSink.h +++ b/src/Storages/ObjectStorage/StorageObjectStorageSink.h @@ -20,7 +20,7 @@ public: String getName() const override { return "StorageObjectStorageSink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onCancel() override; diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp index aef783fc3c4..086482e330a 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp @@ -13,6 +13,7 @@ #include #include #include +#include namespace fs = std::filesystem; @@ -65,7 +66,6 @@ StorageObjectStorageSource::StorageObjectStorageSource( CurrentMetrics::StorageObjectStorageThreadsActive, CurrentMetrics::StorageObjectStorageThreadsScheduled, 1/* max_threads */)) - , columns_desc(info.columns_description) , file_iterator(file_iterator_) , schema_cache(StorageObjectStorage::getSchemaCache(context_, configuration->getTypeName())) , create_reader_scheduler(threadPoolCallbackRunnerUnsafe(*create_reader_pool, "Reader")) @@ -156,20 +156,20 @@ std::shared_ptr StorageObjectStorageSourc return iterator; } -void StorageObjectStorageSource::lazyInitialize(size_t processor) +void StorageObjectStorageSource::lazyInitialize() { if (initialized) return; - reader = createReader(processor); + reader = createReader(); if (reader) - reader_future = createReaderAsync(processor); + reader_future = createReaderAsync(); initialized = true; } Chunk StorageObjectStorageSource::generate() { - lazyInitialize(0); + lazyInitialize(); while (true) { @@ -196,13 +196,38 @@ Chunk StorageObjectStorageSource::generate() const auto & filename = object_info->getFileName(); chassert(object_info->metadata); VirtualColumnUtils::addRequestedFileLikeStorageVirtualsToChunk( - chunk, read_from_format_info.requested_virtual_columns, + chunk, + read_from_format_info.requested_virtual_columns, + {.path = getUniqueStoragePathIdentifier(*configuration, *object_info, false), + .size = object_info->isArchive() ? object_info->fileSizeInArchive() : object_info->metadata->size_bytes, + .filename = &filename, + .last_modified = object_info->metadata->last_modified}); + + const auto & partition_columns = configuration->getPartitionColumns(); + if (!partition_columns.empty() && chunk_size && chunk.hasColumns()) + { + auto partition_values = partition_columns.find(filename); + if (partition_values != partition_columns.end()) { - .path = getUniqueStoragePathIdentifier(*configuration, *object_info, false), - .size = object_info->metadata->size_bytes, - .filename = &filename, - .last_modified = object_info->metadata->last_modified - }); + for (const auto & [name_and_type, value] : partition_values->second) + { + if (!read_from_format_info.source_header.has(name_and_type.name)) + continue; + + const auto column_pos = read_from_format_info.source_header.getPositionByName(name_and_type.name); + auto partition_column = name_and_type.type->createColumnConst(chunk.getNumRows(), value)->convertToFullColumnIfConst(); + + /// This column is filled with default value now, remove it. + chunk.erase(column_pos); + + /// Add correct values. + if (column_pos < chunk.getNumColumns()) + chunk.addColumn(column_pos, std::move(partition_column)); + else + chunk.addColumn(std::move(partition_column)); + } + } + } return chunk; } @@ -236,27 +261,30 @@ void StorageObjectStorageSource::addNumRowsToCache(const ObjectInfo & object_inf schema_cache.addNumRows(cache_key, num_rows); } -std::optional StorageObjectStorageSource::tryGetNumRowsFromCache(const ObjectInfo & object_info) +StorageObjectStorageSource::ReaderHolder StorageObjectStorageSource::createReader() { - const auto cache_key = getKeyForSchemaCache( - getUniqueStoragePathIdentifier(*configuration, object_info), - configuration->format, - format_settings, - getContext()); - - auto get_last_mod_time = [&]() -> std::optional - { - return object_info.metadata - ? std::optional(object_info.metadata->last_modified.epochTime()) - : std::nullopt; - }; - return schema_cache.tryGetNumRows(cache_key, get_last_mod_time); + return createReader( + 0, file_iterator, configuration, object_storage, read_from_format_info, format_settings, + key_condition, getContext(), &schema_cache, log, max_block_size, max_parsing_threads, need_only_count); } -StorageObjectStorageSource::ReaderHolder StorageObjectStorageSource::createReader(size_t processor) +StorageObjectStorageSource::ReaderHolder StorageObjectStorageSource::createReader( + size_t processor, + const std::shared_ptr & file_iterator, + const ConfigurationPtr & configuration, + const ObjectStoragePtr & object_storage, + const ReadFromFormatInfo & read_from_format_info, + const std::optional & format_settings, + const std::shared_ptr & key_condition_, + const ContextPtr & context_, + SchemaCache * schema_cache, + const LoggerPtr & log, + size_t max_block_size, + size_t max_parsing_threads, + bool need_only_count) { ObjectInfoPtr object_info; - auto query_settings = configuration->getQuerySettings(getContext()); + auto query_settings = configuration->getQuerySettings(context_); do { @@ -277,9 +305,29 @@ StorageObjectStorageSource::ReaderHolder StorageObjectStorageSource::createReade std::shared_ptr source; std::unique_ptr read_buf; + auto try_get_num_rows_from_cache = [&]() -> std::optional + { + if (!schema_cache) + return std::nullopt; + + const auto cache_key = getKeyForSchemaCache( + getUniqueStoragePathIdentifier(*configuration, *object_info), + configuration->format, + format_settings, + context_); + + auto get_last_mod_time = [&]() -> std::optional + { + return object_info->metadata + ? std::optional(object_info->metadata->last_modified.epochTime()) + : std::nullopt; + }; + return schema_cache->tryGetNumRows(cache_key, get_last_mod_time); + }; + std::optional num_rows_from_cache = need_only_count - && getContext()->getSettingsRef().use_cache_for_count_from_files - ? tryGetNumRowsFromCache(*object_info) + && context_->getSettingsRef().use_cache_for_count_from_files + ? try_get_num_rows_from_cache() : std::nullopt; if (num_rows_from_cache) @@ -304,14 +352,14 @@ StorageObjectStorageSource::ReaderHolder StorageObjectStorageSource::createReade else { compression_method = chooseCompressionMethod(object_info->getFileName(), configuration->compression_method); - read_buf = createReadBuffer(*object_info); + read_buf = createReadBuffer(*object_info, object_storage, context_, log); } auto input_format = FormatFactory::instance().getInput( configuration->format, *read_buf, read_from_format_info.format_header, - getContext(), + context_, max_block_size, format_settings, need_only_count ? 1 : max_parsing_threads, @@ -320,20 +368,20 @@ StorageObjectStorageSource::ReaderHolder StorageObjectStorageSource::createReade compression_method, need_only_count); - if (key_condition) - input_format->setKeyCondition(key_condition); + if (key_condition_) + input_format->setKeyCondition(key_condition_); if (need_only_count) input_format->needOnlyCount(); builder.init(Pipe(input_format)); - if (columns_desc.hasDefaults()) + if (read_from_format_info.columns_description.hasDefaults()) { builder.addSimpleTransform( [&](const Block & header) { - return std::make_shared(header, columns_desc, *input_format, getContext()); + return std::make_shared(header, read_from_format_info.columns_description, *input_format, context_); }); } @@ -356,21 +404,25 @@ StorageObjectStorageSource::ReaderHolder StorageObjectStorageSource::createReade object_info, std::move(read_buf), std::move(source), std::move(pipeline), std::move(current_reader)); } -std::future StorageObjectStorageSource::createReaderAsync(size_t processor) +std::future StorageObjectStorageSource::createReaderAsync() { - return create_reader_scheduler([=, this] { return createReader(processor); }, Priority{}); + return create_reader_scheduler([=, this] { return createReader(); }, Priority{}); } -std::unique_ptr StorageObjectStorageSource::createReadBuffer(const ObjectInfo & object_info) +std::unique_ptr StorageObjectStorageSource::createReadBuffer( + const ObjectInfo & object_info, + const ObjectStoragePtr & object_storage, + const ContextPtr & context_, + const LoggerPtr & log) { const auto & object_size = object_info.metadata->size_bytes; - auto read_settings = getContext()->getReadSettings().adjustBufferSize(object_size); + auto read_settings = context_->getReadSettings().adjustBufferSize(object_size); read_settings.enable_filesystem_cache = false; /// FIXME: Changing this setting to default value breaks something around parquet reading read_settings.remote_read_min_bytes_for_seek = read_settings.remote_fs_buffer_size; - const bool object_too_small = object_size <= 2 * getContext()->getSettings().max_download_buffer_size; + const bool object_too_small = object_size <= 2 * context_->getSettingsRef().max_download_buffer_size; const bool use_prefetch = object_too_small && read_settings.remote_fs_method == RemoteFSReadMethod::threadpool; read_settings.remote_fs_method = use_prefetch ? RemoteFSReadMethod::threadpool : RemoteFSReadMethod::read; /// User's object may change, don't cache it. @@ -690,10 +742,9 @@ static IArchiveReader::NameFilter createArchivePathFilter(const std::string & ar StorageObjectStorageSource::ArchiveIterator::ObjectInfoInArchive::ObjectInfoInArchive( ObjectInfoPtr archive_object_, const std::string & path_in_archive_, - std::shared_ptr archive_reader_) - : archive_object(archive_object_) - , path_in_archive(path_in_archive_) - , archive_reader(archive_reader_) + std::shared_ptr archive_reader_, + IArchiveReader::FileInfo && file_info_) + : archive_object(archive_object_), path_in_archive(path_in_archive_), archive_reader(archive_reader_), file_info(file_info_) { } @@ -732,6 +783,7 @@ StorageObjectStorageSource::ObjectInfoPtr StorageObjectStorageSource::ArchiveIterator::nextImpl(size_t processor) { std::unique_lock lock{next_mutex}; + IArchiveReader::FileInfo current_file_info{}; while (true) { if (filter) @@ -756,6 +808,8 @@ StorageObjectStorageSource::ArchiveIterator::nextImpl(size_t processor) path_in_archive = file_enumerator->getFileName(); if (!filter(path_in_archive)) continue; + else + current_file_info = file_enumerator->getFileInfo(); } else { @@ -769,15 +823,19 @@ StorageObjectStorageSource::ArchiveIterator::nextImpl(size_t processor) archive_reader = createArchiveReader(archive_object); if (!archive_reader->fileExists(path_in_archive)) continue; + else + current_file_info = archive_reader->getFileInfo(path_in_archive); } - - auto object_in_archive = std::make_shared(archive_object, path_in_archive, archive_reader); - - if (read_keys != nullptr) - read_keys->push_back(object_in_archive); - - return object_in_archive; + break; } + + auto object_in_archive + = std::make_shared(archive_object, path_in_archive, archive_reader, std::move(current_file_info)); + + if (read_keys != nullptr) + read_keys->push_back(object_in_archive); + + return object_in_archive; } size_t StorageObjectStorageSource::ArchiveIterator::estimatedKeysCount() diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSource.h b/src/Storages/ObjectStorage/StorageObjectStorageSource.h index 1b7c3b9be3f..271b38fa75c 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSource.h +++ b/src/Storages/ObjectStorage/StorageObjectStorageSource.h @@ -6,6 +6,7 @@ #include #include #include +#include namespace DB @@ -15,7 +16,7 @@ class SchemaCache; class StorageObjectStorageSource : public SourceWithKeyCondition, WithContext { - friend class StorageS3QueueSource; + friend class ObjectStorageQueueSource; public: using Configuration = StorageObjectStorage::Configuration; using ConfigurationPtr = StorageObjectStorage::ConfigurationPtr; @@ -75,7 +76,6 @@ protected: const ReadFromFormatInfo read_from_format_info; const std::shared_ptr create_reader_pool; - ColumnsDescription columns_desc; std::shared_ptr file_iterator; SchemaCache & schema_cache; bool initialized = false; @@ -116,13 +116,32 @@ protected: std::future reader_future; /// Recreate ReadBuffer and Pipeline for each file. - ReaderHolder createReader(size_t processor = 0); - std::future createReaderAsync(size_t processor = 0); - std::unique_ptr createReadBuffer(const ObjectInfo & object_info); + static ReaderHolder createReader( + size_t processor, + const std::shared_ptr & file_iterator, + const ConfigurationPtr & configuration, + const ObjectStoragePtr & object_storage, + const ReadFromFormatInfo & read_from_format_info, + const std::optional & format_settings, + const std::shared_ptr & key_condition_, + const ContextPtr & context_, + SchemaCache * schema_cache, + const LoggerPtr & log, + size_t max_block_size, + size_t max_parsing_threads, + bool need_only_count); + + ReaderHolder createReader(); + + std::future createReaderAsync(); + static std::unique_ptr createReadBuffer( + const ObjectInfo & object_info, + const ObjectStoragePtr & object_storage, + const ContextPtr & context_, + const LoggerPtr & log); void addNumRowsToCache(const ObjectInfo & object_info, size_t num_rows); - std::optional tryGetNumRowsFromCache(const ObjectInfo & object_info); - void lazyInitialize(size_t processor); + void lazyInitialize(); }; class StorageObjectStorageSource::IIterator @@ -259,7 +278,8 @@ public: ObjectInfoInArchive( ObjectInfoPtr archive_object_, const std::string & path_in_archive_, - std::shared_ptr archive_reader_); + std::shared_ptr archive_reader_, + IArchiveReader::FileInfo && file_info_); std::string getFileName() const override { @@ -278,9 +298,12 @@ public: bool isArchive() const override { return true; } + size_t fileSizeInArchive() const override { return file_info.uncompressed_size; } + const ObjectInfoPtr archive_object; const std::string path_in_archive; const std::shared_ptr archive_reader; + const IArchiveReader::FileInfo file_info; }; private: diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index bf595b2f5d4..b5f4cf5bb54 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -1,9 +1,10 @@ +#include +#include #include #include #include #include #include -#include namespace DB { diff --git a/src/Storages/S3Queue/S3QueueIFileMetadata.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueIFileMetadata.cpp similarity index 84% rename from src/Storages/S3Queue/S3QueueIFileMetadata.cpp rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueIFileMetadata.cpp index 6d550571f22..6fac519849d 100644 --- a/src/Storages/S3Queue/S3QueueIFileMetadata.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueIFileMetadata.cpp @@ -1,4 +1,4 @@ -#include +#include #include #include #include @@ -11,8 +11,8 @@ namespace ProfileEvents { - extern const Event S3QueueProcessedFiles; - extern const Event S3QueueFailedFiles; + extern const Event ObjectStorageQueueProcessedFiles; + extern const Event ObjectStorageQueueFailedFiles; }; namespace DB @@ -35,25 +35,25 @@ namespace } } -void S3QueueIFileMetadata::FileStatus::setProcessingEndTime() +void ObjectStorageQueueIFileMetadata::FileStatus::setProcessingEndTime() { processing_end_time = now(); } -void S3QueueIFileMetadata::FileStatus::onProcessing() +void ObjectStorageQueueIFileMetadata::FileStatus::onProcessing() { state = FileStatus::State::Processing; processing_start_time = now(); } -void S3QueueIFileMetadata::FileStatus::onProcessed() +void ObjectStorageQueueIFileMetadata::FileStatus::onProcessed() { state = FileStatus::State::Processed; if (!processing_end_time) setProcessingEndTime(); } -void S3QueueIFileMetadata::FileStatus::onFailed(const std::string & exception) +void ObjectStorageQueueIFileMetadata::FileStatus::onFailed(const std::string & exception) { state = FileStatus::State::Failed; if (!processing_end_time) @@ -62,13 +62,18 @@ void S3QueueIFileMetadata::FileStatus::onFailed(const std::string & exception) last_exception = exception; } -std::string S3QueueIFileMetadata::FileStatus::getException() const +void ObjectStorageQueueIFileMetadata::FileStatus::updateState(State state_) +{ + state = state_; +} + +std::string ObjectStorageQueueIFileMetadata::FileStatus::getException() const { std::lock_guard lock(last_exception_mutex); return last_exception; } -std::string S3QueueIFileMetadata::NodeMetadata::toString() const +std::string ObjectStorageQueueIFileMetadata::NodeMetadata::toString() const { Poco::JSON::Object json; json.set("file_path", file_path); @@ -83,7 +88,7 @@ std::string S3QueueIFileMetadata::NodeMetadata::toString() const return oss.str(); } -S3QueueIFileMetadata::NodeMetadata S3QueueIFileMetadata::NodeMetadata::fromString(const std::string & metadata_str) +ObjectStorageQueueIFileMetadata::NodeMetadata ObjectStorageQueueIFileMetadata::NodeMetadata::fromString(const std::string & metadata_str) { Poco::JSON::Parser parser; auto json = parser.parse(metadata_str).extract(); @@ -98,7 +103,7 @@ S3QueueIFileMetadata::NodeMetadata S3QueueIFileMetadata::NodeMetadata::fromStrin return metadata; } -S3QueueIFileMetadata::S3QueueIFileMetadata( +ObjectStorageQueueIFileMetadata::ObjectStorageQueueIFileMetadata( const std::string & path_, const std::string & processing_node_path_, const std::string & processed_node_path_, @@ -123,7 +128,7 @@ S3QueueIFileMetadata::S3QueueIFileMetadata( processed_node_path, processing_node_path, failed_node_path); } -S3QueueIFileMetadata::~S3QueueIFileMetadata() +ObjectStorageQueueIFileMetadata::~ObjectStorageQueueIFileMetadata() { if (processing_id_version.has_value()) { @@ -162,9 +167,9 @@ S3QueueIFileMetadata::~S3QueueIFileMetadata() } } -std::string S3QueueIFileMetadata::getNodeName(const std::string & path) +std::string ObjectStorageQueueIFileMetadata::getNodeName(const std::string & path) { - /// Since with are dealing with paths in s3 which can have "/", + /// Since with are dealing with paths in object storage which can have "/", /// we cannot create a zookeeper node with the name equal to path. /// Therefore we use a hash of the path as a node name. @@ -173,7 +178,7 @@ std::string S3QueueIFileMetadata::getNodeName(const std::string & path) return toString(path_hash.get64()); } -S3QueueIFileMetadata::NodeMetadata S3QueueIFileMetadata::createNodeMetadata( +ObjectStorageQueueIFileMetadata::NodeMetadata ObjectStorageQueueIFileMetadata::createNodeMetadata( const std::string & path, const std::string & exception, size_t retries) @@ -182,9 +187,9 @@ S3QueueIFileMetadata::NodeMetadata S3QueueIFileMetadata::createNodeMetadata( /// Since node name is just a hash we want to know to which file it corresponds, /// so we keep "file_path" in nodes data. - /// "last_processed_timestamp" is needed for TTL metadata nodes enabled by s3queue_tracked_file_ttl_sec. - /// "last_exception" is kept for introspection, should also be visible in system.s3queue_log if it is enabled. - /// "retries" is kept for retrying the processing enabled by s3queue_loading_retries. + /// "last_processed_timestamp" is needed for TTL metadata nodes enabled by tracked_file_ttl_sec. + /// "last_exception" is kept for introspection, should also be visible in system.s3(azure)queue_log if it is enabled. + /// "retries" is kept for retrying the processing enabled by loading_retries. NodeMetadata metadata; metadata.file_path = path; metadata.last_processed_timestamp = now(); @@ -193,7 +198,7 @@ S3QueueIFileMetadata::NodeMetadata S3QueueIFileMetadata::createNodeMetadata( return metadata; } -std::string S3QueueIFileMetadata::getProcessorInfo(const std::string & processor_id) +std::string ObjectStorageQueueIFileMetadata::getProcessorInfo(const std::string & processor_id) { /// Add information which will be useful for debugging just in case. Poco::JSON::Object json; @@ -206,7 +211,7 @@ std::string S3QueueIFileMetadata::getProcessorInfo(const std::string & processor return oss.str(); } -bool S3QueueIFileMetadata::setProcessing() +bool ObjectStorageQueueIFileMetadata::setProcessing() { auto state = file_status->state.load(); if (state == FileStatus::State::Processing @@ -224,9 +229,14 @@ bool S3QueueIFileMetadata::setProcessing() auto [success, file_state] = setProcessingImpl(); if (success) + { file_status->onProcessing(); + } else + { + LOG_TEST(log, "Updating state of {} from {} to {}", path, file_status->state.load(), file_state); file_status->updateState(file_state); + } LOG_TEST(log, "File {} has state `{}`: will {}process (processing id version: {})", path, file_state, success ? "" : "not ", @@ -235,11 +245,11 @@ bool S3QueueIFileMetadata::setProcessing() return success; } -void S3QueueIFileMetadata::setProcessed() +void ObjectStorageQueueIFileMetadata::setProcessed() { LOG_TRACE(log, "Setting file {} as processed (path: {})", path, processed_node_path); - ProfileEvents::increment(ProfileEvents::S3QueueProcessedFiles); + ProfileEvents::increment(ProfileEvents::ObjectStorageQueueProcessedFiles); file_status->onProcessed(); try @@ -258,12 +268,12 @@ void S3QueueIFileMetadata::setProcessed() LOG_TRACE(log, "Set file {} as processed (rows: {})", path, file_status->processed_rows); } -void S3QueueIFileMetadata::setFailed(const std::string & exception_message, bool reduce_retry_count, bool overwrite_status) +void ObjectStorageQueueIFileMetadata::setFailed(const std::string & exception_message, bool reduce_retry_count, bool overwrite_status) { LOG_TRACE(log, "Setting file {} as failed (path: {}, reduce retry count: {}, exception: {})", path, failed_node_path, reduce_retry_count, exception_message); - ProfileEvents::increment(ProfileEvents::S3QueueFailedFiles); + ProfileEvents::increment(ProfileEvents::ObjectStorageQueueFailedFiles); if (overwrite_status || file_status->state != FileStatus::State::Failed) file_status->onFailed(exception_message); @@ -295,7 +305,7 @@ void S3QueueIFileMetadata::setFailed(const std::string & exception_message, bool LOG_TRACE(log, "Set file {} as failed (rows: {})", path, file_status->processed_rows); } -void S3QueueIFileMetadata::setFailedNonRetriable() +void ObjectStorageQueueIFileMetadata::setFailedNonRetriable() { auto zk_client = getZooKeeper(); Coordination::Requests requests; @@ -326,7 +336,7 @@ void S3QueueIFileMetadata::setFailedNonRetriable() throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected error while setting file as failed: {}", code); } -void S3QueueIFileMetadata::setFailedRetriable() +void ObjectStorageQueueIFileMetadata::setFailedRetriable() { /// Instead of creating a persistent /failed/node_hash node /// we create a persistent /failed/node_hash.retriable node. diff --git a/src/Storages/S3Queue/S3QueueIFileMetadata.h b/src/Storages/ObjectStorageQueue/ObjectStorageQueueIFileMetadata.h similarity index 93% rename from src/Storages/S3Queue/S3QueueIFileMetadata.h rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueIFileMetadata.h index 6f40338a421..9bab9e7f075 100644 --- a/src/Storages/S3Queue/S3QueueIFileMetadata.h +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueIFileMetadata.h @@ -6,7 +6,7 @@ namespace DB { -class S3QueueIFileMetadata +class ObjectStorageQueueIFileMetadata { public: struct FileStatus @@ -23,7 +23,7 @@ public: void onProcessing(); void onProcessed(); void onFailed(const std::string & exception); - void updateState(State state_) { state = state_; } + void updateState(State state_); std::string getException() const; @@ -34,7 +34,6 @@ public: std::atomic processing_start_time = 0; std::atomic processing_end_time = 0; std::atomic retries = 0; - ProfileEvents::Counters profile_counters; private: mutable std::mutex last_exception_mutex; @@ -42,7 +41,7 @@ public: }; using FileStatusPtr = std::shared_ptr; - explicit S3QueueIFileMetadata( + explicit ObjectStorageQueueIFileMetadata( const std::string & path_, const std::string & processing_node_path_, const std::string & processed_node_path_, @@ -51,7 +50,7 @@ public: size_t max_loading_retries_, LoggerPtr log_); - virtual ~S3QueueIFileMetadata(); + virtual ~ObjectStorageQueueIFileMetadata(); bool setProcessing(); void setProcessed(); @@ -95,7 +94,7 @@ protected: LoggerPtr log; /// processing node is ephemeral, so we cannot verify with it if - /// this node was created by a certain processor on a previous s3 queue processing stage, + /// this node was created by a certain processor on a previous processing stage, /// because we could get a session expired in between the stages /// and someone else could just create this processing node. /// Therefore we also create a persistent processing node diff --git a/src/Storages/S3Queue/S3QueueMetadata.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadata.cpp similarity index 71% rename from src/Storages/S3Queue/S3QueueMetadata.cpp rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadata.cpp index 734d0eed625..23ac92b667a 100644 --- a/src/Storages/S3Queue/S3QueueMetadata.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadata.cpp @@ -4,13 +4,12 @@ #include #include #include -#include -#include -#include -#include -#include -#include -#include +#include +#include +#include +#include +#include +#include #include #include #include @@ -22,13 +21,8 @@ namespace ProfileEvents { - extern const Event S3QueueSetFileProcessingMicroseconds; - extern const Event S3QueueSetFileProcessedMicroseconds; - extern const Event S3QueueSetFileFailedMicroseconds; - extern const Event S3QueueFailedFiles; - extern const Event S3QueueProcessedFiles; - extern const Event S3QueueCleanupMaxSetSizeOrTTLMicroseconds; - extern const Event S3QueueLockLocalFileStatusesMicroseconds; + extern const Event ObjectStorageQueueCleanupMaxSetSizeOrTTLMicroseconds; + extern const Event ObjectStorageQueueLockLocalFileStatusesMicroseconds; }; namespace DB @@ -63,7 +57,7 @@ namespace } } -class S3QueueMetadata::LocalFileStatuses +class ObjectStorageQueueMetadata::LocalFileStatuses { public: LocalFileStatuses() = default; @@ -109,98 +103,89 @@ private: std::unique_lock lock() const { - auto timer = DB::CurrentThread::getProfileEvents().timer(ProfileEvents::S3QueueLockLocalFileStatusesMicroseconds); + auto timer = DB::CurrentThread::getProfileEvents().timer(ProfileEvents::ObjectStorageQueueLockLocalFileStatusesMicroseconds); return std::unique_lock(mutex); } }; -S3QueueMetadata::S3QueueMetadata(const fs::path & zookeeper_path_, const S3QueueSettings & settings_) +ObjectStorageQueueMetadata::ObjectStorageQueueMetadata(const fs::path & zookeeper_path_, const ObjectStorageQueueSettings & settings_) : settings(settings_) , zookeeper_path(zookeeper_path_) , buckets_num(getBucketsNum(settings_)) - , log(getLogger("StorageS3Queue(" + zookeeper_path_.string() + ")")) + , log(getLogger("StorageObjectStorageQueue(" + zookeeper_path_.string() + ")")) , local_file_statuses(std::make_shared()) { - if (settings.mode == S3QueueMode::UNORDERED - && (settings.s3queue_tracked_files_limit || settings.s3queue_tracked_file_ttl_sec)) + if (settings.mode == ObjectStorageQueueMode::UNORDERED + && (settings.tracked_files_limit || settings.tracked_file_ttl_sec)) { task = Context::getGlobalContextInstance()->getSchedulePool().createTask( - "S3QueueCleanupFunc", + "ObjectStorageQueueCleanupFunc", [this] { cleanupThreadFunc(); }); task->activate(); task->scheduleAfter( generateRescheduleInterval( - settings.s3queue_cleanup_interval_min_ms, settings.s3queue_cleanup_interval_max_ms)); + settings.cleanup_interval_min_ms, settings.cleanup_interval_max_ms)); } LOG_TRACE(log, "Mode: {}, buckets: {}, processing threads: {}, result buckets num: {}", - settings.mode.toString(), settings.s3queue_buckets, settings.s3queue_processing_threads_num, buckets_num); + settings.mode.toString(), settings.buckets, settings.processing_threads_num, buckets_num); } -S3QueueMetadata::~S3QueueMetadata() +ObjectStorageQueueMetadata::~ObjectStorageQueueMetadata() { shutdown(); } -void S3QueueMetadata::shutdown() +void ObjectStorageQueueMetadata::shutdown() { shutdown_called = true; if (task) task->deactivate(); } -void S3QueueMetadata::checkSettings(const S3QueueSettings & settings_) const +void ObjectStorageQueueMetadata::checkSettings(const ObjectStorageQueueSettings & settings_) const { - S3QueueTableMetadata::checkEquals(settings, settings_); + ObjectStorageQueueTableMetadata::checkEquals(settings, settings_); } -S3QueueMetadata::FileStatusPtr S3QueueMetadata::getFileStatus(const std::string & path) +ObjectStorageQueueMetadata::FileStatusPtr ObjectStorageQueueMetadata::getFileStatus(const std::string & path) { return local_file_statuses->get(path, /* create */false); } -S3QueueMetadata::FileStatuses S3QueueMetadata::getFileStatuses() const +ObjectStorageQueueMetadata::FileStatuses ObjectStorageQueueMetadata::getFileStatuses() const { return local_file_statuses->getAll(); } -S3QueueMetadata::FileMetadataPtr S3QueueMetadata::getFileMetadata( +ObjectStorageQueueMetadata::FileMetadataPtr ObjectStorageQueueMetadata::getFileMetadata( const std::string & path, - S3QueueOrderedFileMetadata::BucketInfoPtr bucket_info) + ObjectStorageQueueOrderedFileMetadata::BucketInfoPtr bucket_info) { auto file_status = local_file_statuses->get(path, /* create */true); switch (settings.mode.value) { - case S3QueueMode::ORDERED: - return std::make_shared( + case ObjectStorageQueueMode::ORDERED: + return std::make_shared( zookeeper_path, path, file_status, bucket_info, buckets_num, - settings.s3queue_loading_retries, + settings.loading_retries, log); - case S3QueueMode::UNORDERED: - return std::make_shared( + case ObjectStorageQueueMode::UNORDERED: + return std::make_shared( zookeeper_path, path, file_status, - settings.s3queue_loading_retries, + settings.loading_retries, log); } } -size_t S3QueueMetadata::getBucketsNum(const S3QueueSettings & settings) -{ - if (settings.s3queue_buckets) - return settings.s3queue_buckets; - if (settings.s3queue_processing_threads_num) - return settings.s3queue_processing_threads_num; - return 0; -} - -size_t S3QueueMetadata::getBucketsNum(const S3QueueTableMetadata & settings) +size_t ObjectStorageQueueMetadata::getBucketsNum(const ObjectStorageQueueSettings & settings) { if (settings.buckets) return settings.buckets; @@ -209,32 +194,41 @@ size_t S3QueueMetadata::getBucketsNum(const S3QueueTableMetadata & settings) return 0; } -bool S3QueueMetadata::useBucketsForProcessing() const +size_t ObjectStorageQueueMetadata::getBucketsNum(const ObjectStorageQueueTableMetadata & settings) { - return settings.mode == S3QueueMode::ORDERED && (buckets_num > 1); + if (settings.buckets) + return settings.buckets; + if (settings.processing_threads_num) + return settings.processing_threads_num; + return 0; } -S3QueueMetadata::Bucket S3QueueMetadata::getBucketForPath(const std::string & path) const +bool ObjectStorageQueueMetadata::useBucketsForProcessing() const { - return S3QueueOrderedFileMetadata::getBucketForPath(path, buckets_num); + return settings.mode == ObjectStorageQueueMode::ORDERED && (buckets_num > 1); } -S3QueueOrderedFileMetadata::BucketHolderPtr -S3QueueMetadata::tryAcquireBucket(const Bucket & bucket, const Processor & processor) +ObjectStorageQueueMetadata::Bucket ObjectStorageQueueMetadata::getBucketForPath(const std::string & path) const { - return S3QueueOrderedFileMetadata::tryAcquireBucket(zookeeper_path, bucket, processor, log); + return ObjectStorageQueueOrderedFileMetadata::getBucketForPath(path, buckets_num); } -void S3QueueMetadata::initialize( +ObjectStorageQueueOrderedFileMetadata::BucketHolderPtr +ObjectStorageQueueMetadata::tryAcquireBucket(const Bucket & bucket, const Processor & processor) +{ + return ObjectStorageQueueOrderedFileMetadata::tryAcquireBucket(zookeeper_path, bucket, processor, log); +} + +void ObjectStorageQueueMetadata::initialize( const ConfigurationPtr & configuration, const StorageInMemoryMetadata & storage_metadata) { - const auto metadata_from_table = S3QueueTableMetadata(*configuration, settings, storage_metadata); + const auto metadata_from_table = ObjectStorageQueueTableMetadata(*configuration, settings, storage_metadata); const auto & columns_from_table = storage_metadata.getColumns(); const auto table_metadata_path = zookeeper_path / "metadata"; - const auto metadata_paths = settings.mode == S3QueueMode::ORDERED - ? S3QueueOrderedFileMetadata::getMetadataPaths(buckets_num) - : S3QueueUnorderedFileMetadata::getMetadataPaths(); + const auto metadata_paths = settings.mode == ObjectStorageQueueMode::ORDERED + ? ObjectStorageQueueOrderedFileMetadata::getMetadataPaths(buckets_num) + : ObjectStorageQueueUnorderedFileMetadata::getMetadataPaths(); auto zookeeper = getZooKeeper(); zookeeper->createAncestors(zookeeper_path); @@ -243,7 +237,7 @@ void S3QueueMetadata::initialize( { if (zookeeper->exists(table_metadata_path)) { - const auto metadata_from_zk = S3QueueTableMetadata::parse(zookeeper->get(fs::path(zookeeper_path) / "metadata")); + const auto metadata_from_zk = ObjectStorageQueueTableMetadata::parse(zookeeper->get(fs::path(zookeeper_path) / "metadata")); const auto columns_from_zk = ColumnsDescription::parse(metadata_from_zk.columns); metadata_from_table.checkEquals(metadata_from_zk); @@ -268,8 +262,8 @@ void S3QueueMetadata::initialize( requests.emplace_back(zkutil::makeCreateRequest(zk_path, "", zkutil::CreateMode::Persistent)); } - if (!settings.s3queue_last_processed_path.value.empty()) - getFileMetadata(settings.s3queue_last_processed_path)->setProcessedAtStartRequests(requests, zookeeper); + if (!settings.last_processed_path.value.empty()) + getFileMetadata(settings.last_processed_path)->setProcessedAtStartRequests(requests, zookeeper); Coordination::Responses responses; auto code = zookeeper->tryMulti(requests, responses); @@ -293,10 +287,10 @@ void S3QueueMetadata::initialize( "of wrong zookeeper path or because of logical error"); } -void S3QueueMetadata::cleanupThreadFunc() +void ObjectStorageQueueMetadata::cleanupThreadFunc() { /// A background task is responsible for maintaining - /// settings.s3queue_tracked_files_limit and max_set_age settings for `unordered` processing mode. + /// settings.tracked_files_limit and max_set_age settings for `unordered` processing mode. if (shutdown_called) return; @@ -315,12 +309,12 @@ void S3QueueMetadata::cleanupThreadFunc() task->scheduleAfter( generateRescheduleInterval( - settings.s3queue_cleanup_interval_min_ms, settings.s3queue_cleanup_interval_max_ms)); + settings.cleanup_interval_min_ms, settings.cleanup_interval_max_ms)); } -void S3QueueMetadata::cleanupThreadFuncImpl() +void ObjectStorageQueueMetadata::cleanupThreadFuncImpl() { - auto timer = DB::CurrentThread::getProfileEvents().timer(ProfileEvents::S3QueueCleanupMaxSetSizeOrTTLMicroseconds); + auto timer = DB::CurrentThread::getProfileEvents().timer(ProfileEvents::ObjectStorageQueueCleanupMaxSetSizeOrTTLMicroseconds); const auto zk_client = getZooKeeper(); const fs::path zookeeper_processed_path = zookeeper_path / "processed"; const fs::path zookeeper_failed_path = zookeeper_path / "failed"; @@ -358,11 +352,11 @@ void S3QueueMetadata::cleanupThreadFuncImpl() return; } - chassert(settings.s3queue_tracked_files_limit || settings.s3queue_tracked_file_ttl_sec); - const bool check_nodes_limit = settings.s3queue_tracked_files_limit > 0; - const bool check_nodes_ttl = settings.s3queue_tracked_file_ttl_sec > 0; + chassert(settings.tracked_files_limit || settings.tracked_file_ttl_sec); + const bool check_nodes_limit = settings.tracked_files_limit > 0; + const bool check_nodes_ttl = settings.tracked_file_ttl_sec > 0; - const bool nodes_limit_exceeded = nodes_num > settings.s3queue_tracked_files_limit; + const bool nodes_limit_exceeded = nodes_num > settings.tracked_files_limit; if ((!nodes_limit_exceeded || !check_nodes_limit) && !check_nodes_ttl) { LOG_TEST(log, "No limit exceeded"); @@ -384,7 +378,7 @@ void S3QueueMetadata::cleanupThreadFuncImpl() struct Node { std::string zk_path; - S3QueueIFileMetadata::NodeMetadata metadata; + ObjectStorageQueueIFileMetadata::NodeMetadata metadata; }; auto node_cmp = [](const Node & a, const Node & b) { @@ -405,7 +399,7 @@ void S3QueueMetadata::cleanupThreadFuncImpl() std::string metadata_str; if (zk_client->tryGet(path, metadata_str)) { - sorted_nodes.emplace(path, S3QueueIFileMetadata::NodeMetadata::fromString(metadata_str)); + sorted_nodes.emplace(path, ObjectStorageQueueIFileMetadata::NodeMetadata::fromString(metadata_str)); LOG_TEST(log, "Fetched metadata for node {}", path); } else @@ -435,9 +429,9 @@ void S3QueueMetadata::cleanupThreadFuncImpl() wb << fmt::format("Node: {}, path: {}, timestamp: {};\n", node, metadata.file_path, metadata.last_processed_timestamp); return wb.str(); }; - LOG_TEST(log, "Checking node limits (max size: {}, max age: {}) for {}", settings.s3queue_tracked_files_limit, settings.s3queue_tracked_file_ttl_sec, get_nodes_str()); + LOG_TEST(log, "Checking node limits (max size: {}, max age: {}) for {}", settings.tracked_files_limit, settings.tracked_file_ttl_sec, get_nodes_str()); - size_t nodes_to_remove = check_nodes_limit && nodes_limit_exceeded ? nodes_num - settings.s3queue_tracked_files_limit : 0; + size_t nodes_to_remove = check_nodes_limit && nodes_limit_exceeded ? nodes_num - settings.tracked_files_limit : 0; for (const auto & node : sorted_nodes) { if (nodes_to_remove) @@ -456,7 +450,7 @@ void S3QueueMetadata::cleanupThreadFuncImpl() else if (check_nodes_ttl) { UInt64 node_age = getCurrentTime() - node.metadata.last_processed_timestamp; - if (node_age >= settings.s3queue_tracked_file_ttl_sec) + if (node_age >= settings.tracked_file_ttl_sec) { LOG_TRACE(log, "Removing node at path {} ({}) because file ttl is reached", node.metadata.file_path, node.zk_path); diff --git a/src/Storages/S3Queue/S3QueueMetadata.h b/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadata.h similarity index 62% rename from src/Storages/S3Queue/S3QueueMetadata.h rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadata.h index 25d01fb52b9..e5fae047ac5 100644 --- a/src/Storages/S3Queue/S3QueueMetadata.h +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadata.h @@ -2,28 +2,26 @@ #include "config.h" #include -#include -#include #include -#include +#include #include -#include -#include -#include +#include +#include +#include +#include namespace fs = std::filesystem; namespace Poco { class Logger; } namespace DB { -struct S3QueueSettings; -class StorageS3Queue; -struct S3QueueTableMetadata; +class StorageObjectStorageQueue; +struct ObjectStorageQueueTableMetadata; struct StorageInMemoryMetadata; using ConfigurationPtr = StorageObjectStorage::ConfigurationPtr; /** - * A class for managing S3Queue metadata in zookeeper, e.g. + * A class for managing ObjectStorageQueue metadata in zookeeper, e.g. * the following folders: * - /processed * - /processing @@ -35,7 +33,7 @@ using ConfigurationPtr = StorageObjectStorage::ConfigurationPtr; * - /processing * - /failed * - * Depending on S3Queue processing mode (ordered or unordered) + * Depending on ObjectStorageQueue processing mode (ordered or unordered) * we can differently store metadata in /processed node. * * Implements caching of zookeeper metadata for faster responses. @@ -44,24 +42,24 @@ using ConfigurationPtr = StorageObjectStorage::ConfigurationPtr; * In case of Unordered mode - if files TTL is enabled or maximum tracked files limit is set * starts a background cleanup thread which is responsible for maintaining them. */ -class S3QueueMetadata +class ObjectStorageQueueMetadata { public: - using FileStatus = S3QueueIFileMetadata::FileStatus; - using FileMetadataPtr = std::shared_ptr; + using FileStatus = ObjectStorageQueueIFileMetadata::FileStatus; + using FileMetadataPtr = std::shared_ptr; using FileStatusPtr = std::shared_ptr; using FileStatuses = std::unordered_map; using Bucket = size_t; using Processor = std::string; - S3QueueMetadata(const fs::path & zookeeper_path_, const S3QueueSettings & settings_); - ~S3QueueMetadata(); + ObjectStorageQueueMetadata(const fs::path & zookeeper_path_, const ObjectStorageQueueSettings & settings_); + ~ObjectStorageQueueMetadata(); void initialize(const ConfigurationPtr & configuration, const StorageInMemoryMetadata & storage_metadata); - void checkSettings(const S3QueueSettings & settings) const; + void checkSettings(const ObjectStorageQueueSettings & settings) const; void shutdown(); - FileMetadataPtr getFileMetadata(const std::string & path, S3QueueOrderedFileMetadata::BucketInfoPtr bucket_info = {}); + FileMetadataPtr getFileMetadata(const std::string & path, ObjectStorageQueueOrderedFileMetadata::BucketInfoPtr bucket_info = {}); FileStatusPtr getFileStatus(const std::string & path); FileStatuses getFileStatuses() const; @@ -69,16 +67,16 @@ public: /// Method of Ordered mode parallel processing. bool useBucketsForProcessing() const; Bucket getBucketForPath(const std::string & path) const; - S3QueueOrderedFileMetadata::BucketHolderPtr tryAcquireBucket(const Bucket & bucket, const Processor & processor); + ObjectStorageQueueOrderedFileMetadata::BucketHolderPtr tryAcquireBucket(const Bucket & bucket, const Processor & processor); - static size_t getBucketsNum(const S3QueueSettings & settings); - static size_t getBucketsNum(const S3QueueTableMetadata & settings); + static size_t getBucketsNum(const ObjectStorageQueueSettings & settings); + static size_t getBucketsNum(const ObjectStorageQueueTableMetadata & settings); private: void cleanupThreadFunc(); void cleanupThreadFuncImpl(); - const S3QueueSettings settings; + const ObjectStorageQueueSettings settings; const fs::path zookeeper_path; const size_t buckets_num; diff --git a/src/Storages/S3Queue/S3QueueMetadataFactory.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadataFactory.cpp similarity index 62% rename from src/Storages/S3Queue/S3QueueMetadataFactory.cpp rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadataFactory.cpp index a319b21ca3e..ffae33d6f41 100644 --- a/src/Storages/S3Queue/S3QueueMetadataFactory.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadataFactory.cpp @@ -1,4 +1,4 @@ -#include +#include #include namespace DB @@ -8,20 +8,20 @@ namespace ErrorCodes extern const int BAD_ARGUMENTS; } -S3QueueMetadataFactory & S3QueueMetadataFactory::instance() +ObjectStorageQueueMetadataFactory & ObjectStorageQueueMetadataFactory::instance() { - static S3QueueMetadataFactory ret; + static ObjectStorageQueueMetadataFactory ret; return ret; } -S3QueueMetadataFactory::FilesMetadataPtr -S3QueueMetadataFactory::getOrCreate(const std::string & zookeeper_path, const S3QueueSettings & settings) +ObjectStorageQueueMetadataFactory::FilesMetadataPtr +ObjectStorageQueueMetadataFactory::getOrCreate(const std::string & zookeeper_path, const ObjectStorageQueueSettings & settings) { std::lock_guard lock(mutex); auto it = metadata_by_path.find(zookeeper_path); if (it == metadata_by_path.end()) { - auto files_metadata = std::make_shared(zookeeper_path, settings); + auto files_metadata = std::make_shared(zookeeper_path, settings); it = metadata_by_path.emplace(zookeeper_path, std::move(files_metadata)).first; } else @@ -32,7 +32,7 @@ S3QueueMetadataFactory::getOrCreate(const std::string & zookeeper_path, const S3 return it->second.metadata; } -void S3QueueMetadataFactory::remove(const std::string & zookeeper_path) +void ObjectStorageQueueMetadataFactory::remove(const std::string & zookeeper_path) { std::lock_guard lock(mutex); auto it = metadata_by_path.find(zookeeper_path); @@ -57,9 +57,9 @@ void S3QueueMetadataFactory::remove(const std::string & zookeeper_path) } } -std::unordered_map S3QueueMetadataFactory::getAll() +std::unordered_map ObjectStorageQueueMetadataFactory::getAll() { - std::unordered_map result; + std::unordered_map result; for (const auto & [zk_path, metadata_and_ref_count] : metadata_by_path) result.emplace(zk_path, metadata_and_ref_count.metadata); return result; diff --git a/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadataFactory.h b/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadataFactory.h new file mode 100644 index 00000000000..a93f5ee3d83 --- /dev/null +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueMetadataFactory.h @@ -0,0 +1,37 @@ +#pragma once +#include +#include +#include + +namespace DB +{ + +class ObjectStorageQueueMetadataFactory final : private boost::noncopyable +{ +public: + using FilesMetadataPtr = std::shared_ptr; + + static ObjectStorageQueueMetadataFactory & instance(); + + FilesMetadataPtr getOrCreate(const std::string & zookeeper_path, const ObjectStorageQueueSettings & settings); + + void remove(const std::string & zookeeper_path); + + std::unordered_map getAll(); + +private: + struct Metadata + { + explicit Metadata(std::shared_ptr metadata_) : metadata(metadata_), ref_count(1) {} + + std::shared_ptr metadata; + /// TODO: the ref count should be kept in keeper, because of the case with distributed processing. + size_t ref_count = 0; + }; + using MetadataByPath = std::unordered_map; + + MetadataByPath metadata_by_path; + std::mutex mutex; +}; + +} diff --git a/src/Storages/S3Queue/S3QueueOrderedFileMetadata.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueOrderedFileMetadata.cpp similarity index 91% rename from src/Storages/S3Queue/S3QueueOrderedFileMetadata.cpp rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueOrderedFileMetadata.cpp index da1e394ef82..3b711a892c9 100644 --- a/src/Storages/S3Queue/S3QueueOrderedFileMetadata.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueOrderedFileMetadata.cpp @@ -1,4 +1,4 @@ -#include +#include #include #include #include @@ -16,7 +16,7 @@ namespace ErrorCodes namespace { - S3QueueOrderedFileMetadata::Bucket getBucketForPathImpl(const std::string & path, size_t buckets_num) + ObjectStorageQueueOrderedFileMetadata::Bucket getBucketForPathImpl(const std::string & path, size_t buckets_num) { return sipHash64(path) % buckets_num; } @@ -40,7 +40,7 @@ namespace } } -S3QueueOrderedFileMetadata::BucketHolder::BucketHolder( +ObjectStorageQueueOrderedFileMetadata::BucketHolder::BucketHolder( const Bucket & bucket_, int bucket_version_, const std::string & bucket_lock_path_, @@ -57,7 +57,7 @@ S3QueueOrderedFileMetadata::BucketHolder::BucketHolder( { } -void S3QueueOrderedFileMetadata::BucketHolder::release() +void ObjectStorageQueueOrderedFileMetadata::BucketHolder::release() { if (released) return; @@ -89,7 +89,7 @@ void S3QueueOrderedFileMetadata::BucketHolder::release() zkutil::KeeperMultiException::check(code, requests, responses); } -S3QueueOrderedFileMetadata::BucketHolder::~BucketHolder() +ObjectStorageQueueOrderedFileMetadata::BucketHolder::~BucketHolder() { if (!released) LOG_TEST(log, "Releasing bucket ({}) holder in destructor", bucket_info->bucket); @@ -104,7 +104,7 @@ S3QueueOrderedFileMetadata::BucketHolder::~BucketHolder() } } -S3QueueOrderedFileMetadata::S3QueueOrderedFileMetadata( +ObjectStorageQueueOrderedFileMetadata::ObjectStorageQueueOrderedFileMetadata( const std::filesystem::path & zk_path_, const std::string & path_, FileStatusPtr file_status_, @@ -112,7 +112,7 @@ S3QueueOrderedFileMetadata::S3QueueOrderedFileMetadata( size_t buckets_num_, size_t max_loading_retries_, LoggerPtr log_) - : S3QueueIFileMetadata( + : ObjectStorageQueueIFileMetadata( path_, /* processing_node_path */zk_path_ / "processing" / getNodeName(path_), /* processed_node_path */getProcessedPath(zk_path_, path_, buckets_num_), @@ -126,7 +126,7 @@ S3QueueOrderedFileMetadata::S3QueueOrderedFileMetadata( { } -std::vector S3QueueOrderedFileMetadata::getMetadataPaths(size_t buckets_num) +std::vector ObjectStorageQueueOrderedFileMetadata::getMetadataPaths(size_t buckets_num) { if (buckets_num > 1) { @@ -139,7 +139,7 @@ std::vector S3QueueOrderedFileMetadata::getMetadataPaths(size_t buc return {"failed", "processing"}; } -bool S3QueueOrderedFileMetadata::getMaxProcessedFile( +bool ObjectStorageQueueOrderedFileMetadata::getMaxProcessedFile( NodeMetadata & result, Coordination::Stat * stat, const zkutil::ZooKeeperPtr & zk_client) @@ -147,7 +147,7 @@ bool S3QueueOrderedFileMetadata::getMaxProcessedFile( return getMaxProcessedFile(result, stat, processed_node_path, zk_client); } -bool S3QueueOrderedFileMetadata::getMaxProcessedFile( +bool ObjectStorageQueueOrderedFileMetadata::getMaxProcessedFile( NodeMetadata & result, Coordination::Stat * stat, const std::string & processed_node_path_, @@ -163,12 +163,12 @@ bool S3QueueOrderedFileMetadata::getMaxProcessedFile( return false; } -S3QueueOrderedFileMetadata::Bucket S3QueueOrderedFileMetadata::getBucketForPath(const std::string & path_, size_t buckets_num) +ObjectStorageQueueOrderedFileMetadata::Bucket ObjectStorageQueueOrderedFileMetadata::getBucketForPath(const std::string & path_, size_t buckets_num) { return getBucketForPathImpl(path_, buckets_num); } -S3QueueOrderedFileMetadata::BucketHolderPtr S3QueueOrderedFileMetadata::tryAcquireBucket( +ObjectStorageQueueOrderedFileMetadata::BucketHolderPtr ObjectStorageQueueOrderedFileMetadata::tryAcquireBucket( const std::filesystem::path & zk_path, const Bucket & bucket, const Processor & processor, @@ -190,7 +190,7 @@ S3QueueOrderedFileMetadata::BucketHolderPtr S3QueueOrderedFileMetadata::tryAcqui bucket_lock_id_path, processor_info, zkutil::CreateMode::Persistent, /* ignore_if_exists */true)); /// Update bucket lock id path. We use its version as a version of ephemeral bucket lock node. - /// (See comment near S3QueueIFileMetadata::processing_node_version). + /// (See comment near ObjectStorageQueueIFileMetadata::processing_node_version). requests.push_back(zkutil::makeSetRequest(bucket_lock_id_path, processor_info, -1)); Coordination::Responses responses; @@ -223,7 +223,7 @@ S3QueueOrderedFileMetadata::BucketHolderPtr S3QueueOrderedFileMetadata::tryAcqui throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected error: {}", code); } -std::pair S3QueueOrderedFileMetadata::setProcessingImpl() +std::pair ObjectStorageQueueOrderedFileMetadata::setProcessingImpl() { /// In one zookeeper transaction do the following: enum RequestType @@ -319,7 +319,7 @@ std::pair S3QueueOrderedFileMetad } } -void S3QueueOrderedFileMetadata::setProcessedAtStartRequests( +void ObjectStorageQueueOrderedFileMetadata::setProcessedAtStartRequests( Coordination::Requests & requests, const zkutil::ZooKeeperPtr & zk_client) { @@ -337,7 +337,7 @@ void S3QueueOrderedFileMetadata::setProcessedAtStartRequests( } } -void S3QueueOrderedFileMetadata::setProcessedRequests( +void ObjectStorageQueueOrderedFileMetadata::setProcessedRequests( Coordination::Requests & requests, const zkutil::ZooKeeperPtr & zk_client, const std::string & processed_node_path_, @@ -378,7 +378,7 @@ void S3QueueOrderedFileMetadata::setProcessedRequests( } } -void S3QueueOrderedFileMetadata::setProcessedImpl() +void ObjectStorageQueueOrderedFileMetadata::setProcessedImpl() { /// In one zookeeper transaction do the following: enum RequestType diff --git a/src/Storages/S3Queue/S3QueueOrderedFileMetadata.h b/src/Storages/ObjectStorageQueue/ObjectStorageQueueOrderedFileMetadata.h similarity index 86% rename from src/Storages/S3Queue/S3QueueOrderedFileMetadata.h rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueOrderedFileMetadata.h index 82ca87e3251..9a997838f4d 100644 --- a/src/Storages/S3Queue/S3QueueOrderedFileMetadata.h +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueOrderedFileMetadata.h @@ -1,5 +1,5 @@ #pragma once -#include +#include #include #include #include @@ -7,7 +7,7 @@ namespace DB { -class S3QueueOrderedFileMetadata : public S3QueueIFileMetadata +class ObjectStorageQueueOrderedFileMetadata : public ObjectStorageQueueIFileMetadata { public: using Processor = std::string; @@ -21,7 +21,7 @@ public: }; using BucketInfoPtr = std::shared_ptr; - explicit S3QueueOrderedFileMetadata( + explicit ObjectStorageQueueOrderedFileMetadata( const std::filesystem::path & zk_path_, const std::string & path_, FileStatusPtr file_status_, @@ -39,7 +39,7 @@ public: const Processor & processor, LoggerPtr log_); - static S3QueueOrderedFileMetadata::Bucket getBucketForPath(const std::string & path, size_t buckets_num); + static ObjectStorageQueueOrderedFileMetadata::Bucket getBucketForPath(const std::string & path, size_t buckets_num); static std::vector getMetadataPaths(size_t buckets_num); @@ -73,7 +73,7 @@ private: bool ignore_if_exists); }; -struct S3QueueOrderedFileMetadata::BucketHolder : private boost::noncopyable +struct ObjectStorageQueueOrderedFileMetadata::BucketHolder : private boost::noncopyable { BucketHolder( const Bucket & bucket_, diff --git a/src/Storages/S3Queue/S3QueueSettings.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSettings.cpp similarity index 52% rename from src/Storages/S3Queue/S3QueueSettings.cpp rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueSettings.cpp index cb312adc5d9..67743db6197 100644 --- a/src/Storages/S3Queue/S3QueueSettings.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSettings.cpp @@ -1,4 +1,4 @@ -#include +#include #include #include #include @@ -13,14 +13,23 @@ namespace ErrorCodes extern const int UNKNOWN_SETTING; } -IMPLEMENT_SETTINGS_TRAITS(S3QueueSettingsTraits, LIST_OF_S3QUEUE_SETTINGS) +IMPLEMENT_SETTINGS_TRAITS(ObjectStorageQueueSettingsTraits, LIST_OF_OBJECT_STORAGE_QUEUE_SETTINGS) -void S3QueueSettings::loadFromQuery(ASTStorage & storage_def) +void ObjectStorageQueueSettings::loadFromQuery(ASTStorage & storage_def) { if (storage_def.settings) { try { + /// We support settings starting with s3_ for compatibility. + for (auto & change : storage_def.settings->changes) + { + if (change.name.starts_with("s3queue_")) + change.name = change.name.substr(std::strlen("s3queue_")); + if (change.name == "enable_logging_to_s3queue_log") + change.name = "enable_logging_to_queue_log"; + } + applyChanges(storage_def.settings->changes); } catch (Exception & e) diff --git a/src/Storages/ObjectStorageQueue/ObjectStorageQueueSettings.h b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSettings.h new file mode 100644 index 00000000000..ea008c2334e --- /dev/null +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSettings.h @@ -0,0 +1,51 @@ +#pragma once + +#include +#include +#include + + +namespace DB +{ +class ASTStorage; + + +#define OBJECT_STORAGE_QUEUE_RELATED_SETTINGS(M, ALIAS) \ + M(ObjectStorageQueueMode, \ + mode, \ + ObjectStorageQueueMode::ORDERED, \ + "With unordered mode, the set of all already processed files is tracked with persistent nodes in ZooKepeer." \ + "With ordered mode, only the max name of the successfully consumed file stored.", \ + 0) \ + M(ObjectStorageQueueAction, after_processing, ObjectStorageQueueAction::KEEP, "Delete or keep file in after successful processing", 0) \ + M(String, keeper_path, "", "Zookeeper node path", 0) \ + M(UInt32, loading_retries, 10, "Retry loading up to specified number of times", 0) \ + M(UInt32, processing_threads_num, 1, "Number of processing threads", 0) \ + M(UInt32, enable_logging_to_queue_log, 1, "Enable logging to system table system.(s3/azure_)queue_log", 0) \ + M(String, last_processed_path, "", "For Ordered mode. Files that have lexicographically smaller file name are considered already processed", 0) \ + M(UInt32, tracked_file_ttl_sec, 0, "Maximum number of seconds to store processed files in ZooKeeper node (store forever by default)", 0) \ + M(UInt32, polling_min_timeout_ms, 1000, "Minimal timeout before next polling", 0) \ + M(UInt32, polling_max_timeout_ms, 10000, "Maximum timeout before next polling", 0) \ + M(UInt32, polling_backoff_ms, 1000, "Polling backoff", 0) \ + M(UInt32, tracked_files_limit, 1000, "For unordered mode. Max set size for tracking processed files in ZooKeeper", 0) \ + M(UInt32, cleanup_interval_min_ms, 60000, "For unordered mode. Polling backoff min for cleanup", 0) \ + M(UInt32, cleanup_interval_max_ms, 60000, "For unordered mode. Polling backoff max for cleanup", 0) \ + M(UInt32, buckets, 0, "Number of buckets for Ordered mode parallel processing", 0) \ + M(UInt32, max_processed_files_before_commit, 100, "Number of files which can be processed before being committed to keeper", 0) \ + M(UInt32, max_processed_rows_before_commit, 0, "Number of rows which can be processed before being committed to keeper", 0) \ + M(UInt32, max_processed_bytes_before_commit, 0, "Number of bytes which can be processed before being committed to keeper", 0) \ + M(UInt32, max_processing_time_sec_before_commit, 0, "Timeout in seconds after which to commit files committed to keeper", 0) \ + +#define LIST_OF_OBJECT_STORAGE_QUEUE_SETTINGS(M, ALIAS) \ + OBJECT_STORAGE_QUEUE_RELATED_SETTINGS(M, ALIAS) \ + LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS) + +DECLARE_SETTINGS_TRAITS(ObjectStorageQueueSettingsTraits, LIST_OF_OBJECT_STORAGE_QUEUE_SETTINGS) + + +struct ObjectStorageQueueSettings : public BaseSettings +{ + void loadFromQuery(ASTStorage & storage_def); +}; + +} diff --git a/src/Storages/S3Queue/S3QueueSource.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp similarity index 75% rename from src/Storages/S3Queue/S3QueueSource.cpp rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp index 3a611ece51b..4d921003e04 100644 --- a/src/Storages/S3Queue/S3QueueSource.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp @@ -5,20 +5,14 @@ #include #include #include -#include +#include #include #include -namespace CurrentMetrics -{ - extern const Metric StorageS3Threads; - extern const Metric StorageS3ThreadsActive; -} - namespace ProfileEvents { - extern const Event S3QueuePullMicroseconds; + extern const Event ObjectStorageQueuePullMicroseconds; } namespace DB @@ -26,25 +20,24 @@ namespace DB namespace ErrorCodes { - extern const int S3_ERROR; extern const int NOT_IMPLEMENTED; extern const int LOGICAL_ERROR; } -StorageS3QueueSource::S3QueueObjectInfo::S3QueueObjectInfo( +ObjectStorageQueueSource::ObjectStorageQueueObjectInfo::ObjectStorageQueueObjectInfo( const Source::ObjectInfo & object_info, - S3QueueMetadata::FileMetadataPtr file_metadata_) + ObjectStorageQueueMetadata::FileMetadataPtr file_metadata_) : Source::ObjectInfo(object_info.relative_path, object_info.metadata) , file_metadata(file_metadata_) { } -StorageS3QueueSource::FileIterator::FileIterator( - std::shared_ptr metadata_, +ObjectStorageQueueSource::FileIterator::FileIterator( + std::shared_ptr metadata_, std::unique_ptr glob_iterator_, std::atomic & shutdown_called_, LoggerPtr logger_) - : StorageObjectStorageSource::IIterator("S3QueueIterator") + : StorageObjectStorageSource::IIterator("ObjectStorageQueueIterator") , metadata(metadata_) , glob_iterator(std::move(glob_iterator_)) , shutdown_called(shutdown_called_) @@ -52,7 +45,7 @@ StorageS3QueueSource::FileIterator::FileIterator( { } -bool StorageS3QueueSource::FileIterator::isFinished() const +bool ObjectStorageQueueSource::FileIterator::isFinished() const { LOG_TEST(log, "Iterator finished: {}, objects to retry: {}", iterator_finished, objects_to_retry.size()); return iterator_finished @@ -60,15 +53,15 @@ bool StorageS3QueueSource::FileIterator::isFinished() const && objects_to_retry.empty(); } -size_t StorageS3QueueSource::FileIterator::estimatedKeysCount() +size_t ObjectStorageQueueSource::FileIterator::estimatedKeysCount() { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method estimateKeysCount is not implemented"); } -StorageS3QueueSource::Source::ObjectInfoPtr StorageS3QueueSource::FileIterator::nextImpl(size_t processor) +ObjectStorageQueueSource::Source::ObjectInfoPtr ObjectStorageQueueSource::FileIterator::nextImpl(size_t processor) { Source::ObjectInfoPtr object_info; - S3QueueOrderedFileMetadata::BucketInfoPtr bucket_info; + ObjectStorageQueueOrderedFileMetadata::BucketInfoPtr bucket_info; while (!shutdown_called) { @@ -107,26 +100,28 @@ StorageS3QueueSource::Source::ObjectInfoPtr StorageS3QueueSource::FileIterator:: auto file_metadata = metadata->getFileMetadata(object_info->relative_path, bucket_info); if (file_metadata->setProcessing()) - return std::make_shared(*object_info, file_metadata); + return std::make_shared(*object_info, file_metadata); } return {}; } -void StorageS3QueueSource::FileIterator::returnForRetry(Source::ObjectInfoPtr object_info) +void ObjectStorageQueueSource::FileIterator::returnForRetry(Source::ObjectInfoPtr object_info) { chassert(object_info); if (metadata->useBucketsForProcessing()) { const auto bucket = metadata->getBucketForPath(object_info->relative_path); + std::lock_guard lock(mutex); listed_keys_cache[bucket].keys.emplace_front(object_info); } else { + std::lock_guard lock(mutex); objects_to_retry.push_back(object_info); } } -void StorageS3QueueSource::FileIterator::releaseFinishedBuckets() +void ObjectStorageQueueSource::FileIterator::releaseFinishedBuckets() { for (const auto & [processor, holders] : bucket_holders) { @@ -157,8 +152,8 @@ void StorageS3QueueSource::FileIterator::releaseFinishedBuckets() } } -std::pair -StorageS3QueueSource::FileIterator::getNextKeyFromAcquiredBucket(size_t processor) +std::pair +ObjectStorageQueueSource::FileIterator::getNextKeyFromAcquiredBucket(size_t processor) { auto bucket_holder_it = bucket_holders.emplace(processor, std::vector{}).first; BucketHolder * current_bucket_holder = bucket_holder_it->second.empty() || bucket_holder_it->second.back()->isFinished() @@ -234,7 +229,7 @@ StorageS3QueueSource::FileIterator::getNextKeyFromAcquiredBucket(size_t processo } } /// If processing thread has already acquired some bucket - /// and while listing s3 directory gets a key which is in a different bucket, + /// and while listing object storage directory gets a key which is in a different bucket, /// it puts the key into listed_keys_cache to allow others to process it, /// because one processing thread can acquire only one bucket at a time. /// Once a thread is finished with its acquired bucket, it checks listed_keys_cache @@ -361,69 +356,51 @@ StorageS3QueueSource::FileIterator::getNextKeyFromAcquiredBucket(size_t processo } } -StorageS3QueueSource::StorageS3QueueSource( +ObjectStorageQueueSource::ObjectStorageQueueSource( String name_, size_t processor_id_, - const Block & header_, - std::unique_ptr internal_source_, - std::shared_ptr files_metadata_, - const S3QueueAction & action_, - RemoveFileFunc remove_file_func_, - const NamesAndTypesList & requested_virtual_columns_, + std::shared_ptr file_iterator_, + ConfigurationPtr configuration_, + ObjectStoragePtr object_storage_, + const ReadFromFormatInfo & read_from_format_info_, + const std::optional & format_settings_, + const ObjectStorageQueueSettings & queue_settings_, + std::shared_ptr files_metadata_, ContextPtr context_, + size_t max_block_size_, const std::atomic & shutdown_called_, const std::atomic & table_is_being_dropped_, - std::shared_ptr s3_queue_log_, + std::shared_ptr system_queue_log_, const StorageID & storage_id_, LoggerPtr log_, - size_t max_processed_files_before_commit_, - size_t max_processed_rows_before_commit_, - size_t max_processed_bytes_before_commit_, - size_t max_processing_time_sec_before_commit_, bool commit_once_processed_) - : ISource(header_) + : ISource(read_from_format_info_.source_header) , WithContext(context_) , name(std::move(name_)) , processor_id(processor_id_) - , action(action_) + , file_iterator(file_iterator_) + , configuration(configuration_) + , object_storage(object_storage_) + , read_from_format_info(read_from_format_info_) + , format_settings(format_settings_) + , queue_settings(queue_settings_) , files_metadata(files_metadata_) - , internal_source(std::move(internal_source_)) - , requested_virtual_columns(requested_virtual_columns_) + , max_block_size(max_block_size_) , shutdown_called(shutdown_called_) , table_is_being_dropped(table_is_being_dropped_) - , s3_queue_log(s3_queue_log_) + , system_queue_log(system_queue_log_) , storage_id(storage_id_) - , max_processed_files_before_commit(max_processed_files_before_commit_) - , max_processed_rows_before_commit(max_processed_rows_before_commit_) - , max_processed_bytes_before_commit(max_processed_bytes_before_commit_) - , max_processing_time_sec_before_commit(max_processing_time_sec_before_commit_) , commit_once_processed(commit_once_processed_) - , remove_file_func(remove_file_func_) , log(log_) { } -String StorageS3QueueSource::getName() const +String ObjectStorageQueueSource::getName() const { return name; } -void StorageS3QueueSource::lazyInitialize(size_t processor) -{ - if (initialized) - return; - - LOG_TEST(log, "Initializing a new reader"); - - internal_source->lazyInitialize(processor); - reader = std::move(internal_source->reader); - if (reader) - reader_future = std::move(internal_source->reader_future); - - initialized = true; -} - -Chunk StorageS3QueueSource::generate() +Chunk ObjectStorageQueueSource::generate() { Chunk chunk; try @@ -445,21 +422,35 @@ Chunk StorageS3QueueSource::generate() return chunk; } -Chunk StorageS3QueueSource::generateImpl() +Chunk ObjectStorageQueueSource::generateImpl() { - lazyInitialize(processor_id); - while (true) { if (!reader) { - LOG_TEST(log, "No reader"); - break; + if (shutdown_called) + { + LOG_TEST(log, "Shutdown called"); + break; + } + + const auto context = getContext(); + reader = StorageObjectStorageSource::createReader( + processor_id, file_iterator, configuration, object_storage, read_from_format_info, + format_settings, nullptr, context, nullptr, log, max_block_size, + context->getSettingsRef().max_parsing_threads.value, /* need_only_count */false); + + if (!reader) + { + LOG_TEST(log, "No reader"); + break; + } } - const auto * object_info = dynamic_cast(reader.getObjectInfo().get()); + const auto * object_info = dynamic_cast(reader.getObjectInfo().get()); auto file_metadata = object_info->file_metadata; auto file_status = file_metadata->getFileStatus(); + const auto & path = reader.getObjectInfo()->getPath(); if (isCancelled()) { @@ -476,16 +467,12 @@ Chunk StorageS3QueueSource::generateImpl() LOG_ERROR(log, "Failed to set file {} as failed: {}", object_info->relative_path, getCurrentExceptionMessage(true)); } - - appendLogElement(reader.getObjectInfo()->getPath(), *file_status, processed_rows_from_file, false); } LOG_TEST(log, "Query is cancelled"); break; } - const auto & path = reader.getObjectInfo()->getPath(); - if (shutdown_called) { LOG_TEST(log, "Shutdown called"); @@ -509,8 +496,6 @@ Chunk StorageS3QueueSource::generateImpl() object_info->relative_path, getCurrentExceptionMessage(true)); } - appendLogElement(path, *file_status, processed_rows_from_file, false); - /// Leave the file half processed. Table is being dropped, so we do not care. break; } @@ -520,13 +505,9 @@ Chunk StorageS3QueueSource::generateImpl() path, processed_rows_from_file); } - auto * prev_scope = CurrentThread::get().attachProfileCountersScope(&file_status->profile_counters); - SCOPE_EXIT({ CurrentThread::get().attachProfileCountersScope(prev_scope); }); - /// FIXME: if files are compressed, profile counters update does not work fully (s3 related counters are not saved). Why? - try { - auto timer = DB::CurrentThread::getProfileEvents().timer(ProfileEvents::S3QueuePullMicroseconds); + auto timer = DB::CurrentThread::getProfileEvents().timer(ProfileEvents::ObjectStorageQueuePullMicroseconds); Chunk chunk; if (reader->pull(chunk)) @@ -539,7 +520,7 @@ Chunk StorageS3QueueSource::generateImpl() total_processed_bytes += chunk.bytes(); VirtualColumnUtils::addRequestedFileLikeStorageVirtualsToChunk( - chunk, requested_virtual_columns, + chunk, read_from_format_info.requested_virtual_columns, { .path = path, .size = reader.getObjectInfo()->metadata->size_bytes @@ -555,13 +536,9 @@ Chunk StorageS3QueueSource::generateImpl() failed_during_read_files.push_back(file_metadata); file_status->onFailed(getCurrentExceptionMessage(true)); - appendLogElement(path, *file_status, processed_rows_from_file, false); if (processed_rows_from_file == 0) { - auto * file_iterator = dynamic_cast(internal_source->file_iterator.get()); - chassert(file_iterator); - if (file_status->retries < file_metadata->getMaxTries()) file_iterator->returnForRetry(reader.getObjectInfo()); @@ -574,15 +551,15 @@ Chunk StorageS3QueueSource::generateImpl() throw; } - appendLogElement(path, *file_status, processed_rows_from_file, true); - file_status->setProcessingEndTime(); file_status.reset(); + reader = {}; processed_rows_from_file = 0; processed_files.push_back(file_metadata); - if (processed_files.size() == max_processed_files_before_commit) + if (queue_settings.max_processed_files_before_commit + && processed_files.size() == queue_settings.max_processed_files_before_commit) { LOG_TRACE(log, "Number of max processed files before commit reached " "(rows: {}, bytes: {}, files: {})", @@ -590,74 +567,36 @@ Chunk StorageS3QueueSource::generateImpl() break; } - bool rows_or_bytes_or_time_limit_reached = false; - if (max_processed_rows_before_commit - && total_processed_rows == max_processed_rows_before_commit) + if (queue_settings.max_processed_rows_before_commit + && total_processed_rows == queue_settings.max_processed_rows_before_commit) { LOG_TRACE(log, "Number of max processed rows before commit reached " "(rows: {}, bytes: {}, files: {})", total_processed_rows, total_processed_bytes, processed_files.size()); - - rows_or_bytes_or_time_limit_reached = true; + break; } - else if (max_processed_bytes_before_commit - && total_processed_bytes == max_processed_bytes_before_commit) + else if (queue_settings.max_processed_bytes_before_commit + && total_processed_bytes == queue_settings.max_processed_bytes_before_commit) { LOG_TRACE(log, "Number of max processed bytes before commit reached " "(rows: {}, bytes: {}, files: {})", total_processed_rows, total_processed_bytes, processed_files.size()); - - rows_or_bytes_or_time_limit_reached = true; + break; } - else if (max_processing_time_sec_before_commit - && total_stopwatch.elapsedSeconds() >= max_processing_time_sec_before_commit) + else if (queue_settings.max_processing_time_sec_before_commit + && total_stopwatch.elapsedSeconds() >= queue_settings.max_processing_time_sec_before_commit) { LOG_TRACE(log, "Max processing time before commit reached " "(rows: {}, bytes: {}, files: {})", total_processed_rows, total_processed_bytes, processed_files.size()); - - rows_or_bytes_or_time_limit_reached = true; - } - - if (rows_or_bytes_or_time_limit_reached) - { - if (!reader_future.valid()) - break; - - LOG_TRACE(log, "Rows or bytes limit reached, but we have one more file scheduled already, " - "will process it despite the limit"); - } - - if (shutdown_called) - { - LOG_TRACE(log, "Shutdown was called, stopping sync"); break; } - - chassert(reader_future.valid()); - reader = reader_future.get(); - - if (!reader) - { - LOG_TEST(log, "Reader finished"); - break; - } - - file_status = files_metadata->getFileStatus(reader.getObjectInfo()->getPath()); - - if (!rows_or_bytes_or_time_limit_reached && processed_files.size() + 1 < max_processed_files_before_commit) - { - /// Even if task is finished the thread may be not freed in pool. - /// So wait until it will be freed before scheduling a new task. - internal_source->create_reader_pool->wait(); - reader_future = internal_source->createReaderAsync(processor_id); - } } return {}; } -void StorageS3QueueSource::commit(bool success, const std::string & exception_message) +void ObjectStorageQueueSource::commit(bool success, const std::string & exception_message) { LOG_TEST(log, "Having {} files to set as {}, failed files: {}", processed_files.size(), success ? "Processed" : "Failed", failed_during_read_files.size()); @@ -670,10 +609,14 @@ void StorageS3QueueSource::commit(bool success, const std::string & exception_me applyActionAfterProcessing(file_metadata->getPath()); } else + { file_metadata->setFailed( exception_message, /* reduce_retry_count */false, /* overwrite_status */true); + + } + appendLogElement(file_metadata->getPath(), *file_metadata->getFileStatus(), processed_rows_from_file, /* processed */success); } for (const auto & file_metadata : failed_during_read_files) @@ -684,36 +627,37 @@ void StorageS3QueueSource::commit(bool success, const std::string & exception_me file_metadata->getFileStatus()->getException(), /* reduce_retry_count */true, /* overwrite_status */false); + + appendLogElement(file_metadata->getPath(), *file_metadata->getFileStatus(), processed_rows_from_file, /* processed */false); } } -void StorageS3QueueSource::applyActionAfterProcessing(const String & path) +void ObjectStorageQueueSource::applyActionAfterProcessing(const String & path) { - switch (action) + switch (queue_settings.after_processing.value) { - case S3QueueAction::DELETE: + case ObjectStorageQueueAction::DELETE: { - assert(remove_file_func); - remove_file_func(path); + object_storage->removeObject(StoredObject(path)); break; } - case S3QueueAction::KEEP: + case ObjectStorageQueueAction::KEEP: break; } } -void StorageS3QueueSource::appendLogElement( +void ObjectStorageQueueSource::appendLogElement( const std::string & filename, - S3QueueMetadata::FileStatus & file_status_, + ObjectStorageQueueMetadata::FileStatus & file_status_, size_t processed_rows, bool processed) { - if (!s3_queue_log) + if (!system_queue_log) return; - S3QueueLogElement elem{}; + ObjectStorageQueueLogElement elem{}; { - elem = S3QueueLogElement + elem = ObjectStorageQueueLogElement { .event_time = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now()), .database = storage_id.database_name, @@ -721,14 +665,13 @@ void StorageS3QueueSource::appendLogElement( .uuid = toString(storage_id.uuid), .file_name = filename, .rows_processed = processed_rows, - .status = processed ? S3QueueLogElement::S3QueueStatus::Processed : S3QueueLogElement::S3QueueStatus::Failed, - .counters_snapshot = file_status_.profile_counters.getPartiallyAtomicSnapshot(), + .status = processed ? ObjectStorageQueueLogElement::ObjectStorageQueueStatus::Processed : ObjectStorageQueueLogElement::ObjectStorageQueueStatus::Failed, .processing_start_time = file_status_.processing_start_time, .processing_end_time = file_status_.processing_end_time, .exception = file_status_.getException(), }; } - s3_queue_log->add(std::move(elem)); + system_queue_log->add(std::move(elem)); } } diff --git a/src/Storages/S3Queue/S3QueueSource.h b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.h similarity index 60% rename from src/Storages/S3Queue/S3QueueSource.h rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.h index bfa1c358fa9..0f3d0ab2e92 100644 --- a/src/Storages/S3Queue/S3QueueSource.h +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.h @@ -3,10 +3,10 @@ #include #include -#include +#include #include #include -#include +#include namespace Poco { class Logger; } @@ -16,29 +16,28 @@ namespace DB struct ObjectMetadata; -class StorageS3QueueSource : public ISource, WithContext +class ObjectStorageQueueSource : public ISource, WithContext { public: using Storage = StorageObjectStorage; using Source = StorageObjectStorageSource; - using RemoveFileFunc = std::function; - using BucketHolderPtr = S3QueueOrderedFileMetadata::BucketHolderPtr; - using BucketHolder = S3QueueOrderedFileMetadata::BucketHolder; + using BucketHolderPtr = ObjectStorageQueueOrderedFileMetadata::BucketHolderPtr; + using BucketHolder = ObjectStorageQueueOrderedFileMetadata::BucketHolder; - struct S3QueueObjectInfo : public Source::ObjectInfo + struct ObjectStorageQueueObjectInfo : public Source::ObjectInfo { - S3QueueObjectInfo( + ObjectStorageQueueObjectInfo( const Source::ObjectInfo & object_info, - S3QueueMetadata::FileMetadataPtr file_metadata_); + ObjectStorageQueueMetadata::FileMetadataPtr file_metadata_); - S3QueueMetadata::FileMetadataPtr file_metadata; + ObjectStorageQueueMetadata::FileMetadataPtr file_metadata; }; class FileIterator : public StorageObjectStorageSource::IIterator { public: FileIterator( - std::shared_ptr metadata_, + std::shared_ptr metadata_, std::unique_ptr glob_iterator_, std::atomic & shutdown_called_, LoggerPtr logger_); @@ -63,10 +62,10 @@ public: void releaseFinishedBuckets(); private: - using Bucket = S3QueueMetadata::Bucket; - using Processor = S3QueueMetadata::Processor; + using Bucket = ObjectStorageQueueMetadata::Bucket; + using Processor = ObjectStorageQueueMetadata::Processor; - const std::shared_ptr metadata; + const std::shared_ptr metadata; const std::unique_ptr glob_iterator; std::atomic & shutdown_called; @@ -90,29 +89,27 @@ public: /// Only for processing without buckets. std::deque objects_to_retry; - std::pair getNextKeyFromAcquiredBucket(size_t processor); + std::pair getNextKeyFromAcquiredBucket(size_t processor); bool hasKeysForProcessor(const Processor & processor) const; }; - StorageS3QueueSource( + ObjectStorageQueueSource( String name_, size_t processor_id_, - const Block & header_, - std::unique_ptr internal_source_, - std::shared_ptr files_metadata_, - const S3QueueAction & action_, - RemoveFileFunc remove_file_func_, - const NamesAndTypesList & requested_virtual_columns_, + std::shared_ptr file_iterator_, + ConfigurationPtr configuration_, + ObjectStoragePtr object_storage_, + const ReadFromFormatInfo & read_from_format_info_, + const std::optional & format_settings_, + const ObjectStorageQueueSettings & queue_settings_, + std::shared_ptr files_metadata_, ContextPtr context_, + size_t max_block_size_, const std::atomic & shutdown_called_, const std::atomic & table_is_being_dropped_, - std::shared_ptr s3_queue_log_, + std::shared_ptr system_queue_log_, const StorageID & storage_id_, LoggerPtr log_, - size_t max_processed_files_before_commit_, - size_t max_processed_rows_before_commit_, - size_t max_processed_bytes_before_commit_, - size_t max_processing_time_sec_before_commit_, bool commit_once_processed_); static Block getHeader(Block sample_block, const std::vector & requested_virtual_columns); @@ -128,29 +125,27 @@ public: private: const String name; const size_t processor_id; - const S3QueueAction action; - const std::shared_ptr files_metadata; - const std::shared_ptr internal_source; - const NamesAndTypesList requested_virtual_columns; + const std::shared_ptr file_iterator; + const ConfigurationPtr configuration; + const ObjectStoragePtr object_storage; + const ReadFromFormatInfo read_from_format_info; + const std::optional format_settings; + const ObjectStorageQueueSettings queue_settings; + const std::shared_ptr files_metadata; + const size_t max_block_size; + const std::atomic & shutdown_called; const std::atomic & table_is_being_dropped; - const std::shared_ptr s3_queue_log; + const std::shared_ptr system_queue_log; const StorageID storage_id; - const size_t max_processed_files_before_commit; - const size_t max_processed_rows_before_commit; - const size_t max_processed_bytes_before_commit; - const size_t max_processing_time_sec_before_commit; const bool commit_once_processed; - RemoveFileFunc remove_file_func; LoggerPtr log; - std::vector processed_files; - std::vector failed_during_read_files; + std::vector processed_files; + std::vector failed_during_read_files; Source::ReaderHolder reader; - std::future reader_future; - std::atomic initialized{false}; size_t processed_rows_from_file = 0; size_t total_processed_rows = 0; @@ -160,8 +155,11 @@ private: Chunk generateImpl(); void applyActionAfterProcessing(const String & path); - void appendLogElement(const std::string & filename, S3QueueMetadata::FileStatus & file_status_, size_t processed_rows, bool processed); - void lazyInitialize(size_t processor); + void appendLogElement( + const std::string & filename, + ObjectStorageQueueMetadata::FileStatus & file_status_, + size_t processed_rows, + bool processed); }; } diff --git a/src/Storages/S3Queue/S3QueueTableMetadata.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueTableMetadata.cpp similarity index 72% rename from src/Storages/S3Queue/S3QueueTableMetadata.cpp rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueTableMetadata.cpp index ecaa7ad57cc..cb9cdf8e186 100644 --- a/src/Storages/S3Queue/S3QueueTableMetadata.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueTableMetadata.cpp @@ -3,9 +3,9 @@ #include #include #include -#include -#include -#include +#include +#include +#include #include @@ -20,33 +20,33 @@ namespace ErrorCodes namespace { - S3QueueMode modeFromString(const std::string & mode) + ObjectStorageQueueMode modeFromString(const std::string & mode) { if (mode == "ordered") - return S3QueueMode::ORDERED; + return ObjectStorageQueueMode::ORDERED; if (mode == "unordered") - return S3QueueMode::UNORDERED; - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected S3Queue mode: {}", mode); + return ObjectStorageQueueMode::UNORDERED; + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected ObjectStorageQueue mode: {}", mode); } } -S3QueueTableMetadata::S3QueueTableMetadata( +ObjectStorageQueueTableMetadata::ObjectStorageQueueTableMetadata( const StorageObjectStorage::Configuration & configuration, - const S3QueueSettings & engine_settings, + const ObjectStorageQueueSettings & engine_settings, const StorageInMemoryMetadata & storage_metadata) { format_name = configuration.format; after_processing = engine_settings.after_processing.toString(); mode = engine_settings.mode.toString(); - tracked_files_limit = engine_settings.s3queue_tracked_files_limit; - tracked_file_ttl_sec = engine_settings.s3queue_tracked_file_ttl_sec; - buckets = engine_settings.s3queue_buckets; - processing_threads_num = engine_settings.s3queue_processing_threads_num; + tracked_files_limit = engine_settings.tracked_files_limit; + tracked_file_ttl_sec = engine_settings.tracked_file_ttl_sec; + buckets = engine_settings.buckets; + processing_threads_num = engine_settings.processing_threads_num; columns = storage_metadata.getColumns().toString(); } -String S3QueueTableMetadata::toString() const +String ObjectStorageQueueTableMetadata::toString() const { Poco::JSON::Object json; json.set("after_processing", after_processing); @@ -65,7 +65,7 @@ String S3QueueTableMetadata::toString() const return oss.str(); } -void S3QueueTableMetadata::read(const String & metadata_str) +void ObjectStorageQueueTableMetadata::read(const String & metadata_str) { Poco::JSON::Parser parser; auto json = parser.parse(metadata_str).extract(); @@ -102,19 +102,19 @@ void S3QueueTableMetadata::read(const String & metadata_str) buckets = json->getValue("buckets"); } -S3QueueTableMetadata S3QueueTableMetadata::parse(const String & metadata_str) +ObjectStorageQueueTableMetadata ObjectStorageQueueTableMetadata::parse(const String & metadata_str) { - S3QueueTableMetadata metadata; + ObjectStorageQueueTableMetadata metadata; metadata.read(metadata_str); return metadata; } -void S3QueueTableMetadata::checkEquals(const S3QueueTableMetadata & from_zk) const +void ObjectStorageQueueTableMetadata::checkEquals(const ObjectStorageQueueTableMetadata & from_zk) const { checkImmutableFieldsEquals(from_zk); } -void S3QueueTableMetadata::checkImmutableFieldsEquals(const S3QueueTableMetadata & from_zk) const +void ObjectStorageQueueTableMetadata::checkImmutableFieldsEquals(const ObjectStorageQueueTableMetadata & from_zk) const { if (after_processing != from_zk.after_processing) throw Exception( @@ -164,29 +164,29 @@ void S3QueueTableMetadata::checkImmutableFieldsEquals(const S3QueueTableMetadata from_zk.last_processed_path, last_processed_path); - if (modeFromString(mode) == S3QueueMode::ORDERED) + if (modeFromString(mode) == ObjectStorageQueueMode::ORDERED) { if (buckets != from_zk.buckets) { throw Exception( ErrorCodes::METADATA_MISMATCH, - "Existing table metadata in ZooKeeper differs in s3queue_buckets setting. " + "Existing table metadata in ZooKeeper differs in buckets setting. " "Stored in ZooKeeper: {}, local: {}", from_zk.buckets, buckets); } - if (S3QueueMetadata::getBucketsNum(*this) != S3QueueMetadata::getBucketsNum(from_zk)) + if (ObjectStorageQueueMetadata::getBucketsNum(*this) != ObjectStorageQueueMetadata::getBucketsNum(from_zk)) { throw Exception( ErrorCodes::METADATA_MISMATCH, "Existing table metadata in ZooKeeper differs in processing buckets. " "Stored in ZooKeeper: {}, local: {}", - S3QueueMetadata::getBucketsNum(*this), S3QueueMetadata::getBucketsNum(from_zk)); + ObjectStorageQueueMetadata::getBucketsNum(*this), ObjectStorageQueueMetadata::getBucketsNum(from_zk)); } } } -void S3QueueTableMetadata::checkEquals(const S3QueueSettings & current, const S3QueueSettings & expected) +void ObjectStorageQueueTableMetadata::checkEquals(const ObjectStorageQueueSettings & current, const ObjectStorageQueueSettings & expected) { if (current.after_processing != expected.after_processing) throw Exception( @@ -204,48 +204,48 @@ void S3QueueTableMetadata::checkEquals(const S3QueueSettings & current, const S3 expected.mode.toString(), current.mode.toString()); - if (current.s3queue_tracked_files_limit != expected.s3queue_tracked_files_limit) + if (current.tracked_files_limit != expected.tracked_files_limit) throw Exception( ErrorCodes::METADATA_MISMATCH, "Existing table metadata in ZooKeeper differs in max set size. " "Stored in ZooKeeper: {}, local: {}", - expected.s3queue_tracked_files_limit, - current.s3queue_tracked_files_limit); + expected.tracked_files_limit, + current.tracked_files_limit); - if (current.s3queue_tracked_file_ttl_sec != expected.s3queue_tracked_file_ttl_sec) + if (current.tracked_file_ttl_sec != expected.tracked_file_ttl_sec) throw Exception( ErrorCodes::METADATA_MISMATCH, "Existing table metadata in ZooKeeper differs in max set age. " "Stored in ZooKeeper: {}, local: {}", - expected.s3queue_tracked_file_ttl_sec, - current.s3queue_tracked_file_ttl_sec); + expected.tracked_file_ttl_sec, + current.tracked_file_ttl_sec); - if (current.s3queue_last_processed_path.value != expected.s3queue_last_processed_path.value) + if (current.last_processed_path.value != expected.last_processed_path.value) throw Exception( ErrorCodes::METADATA_MISMATCH, "Existing table metadata in ZooKeeper differs in last_processed_path. " "Stored in ZooKeeper: {}, local: {}", - expected.s3queue_last_processed_path.value, - current.s3queue_last_processed_path.value); + expected.last_processed_path.value, + current.last_processed_path.value); - if (current.mode == S3QueueMode::ORDERED) + if (current.mode == ObjectStorageQueueMode::ORDERED) { - if (current.s3queue_buckets != expected.s3queue_buckets) + if (current.buckets != expected.buckets) { throw Exception( ErrorCodes::METADATA_MISMATCH, - "Existing table metadata in ZooKeeper differs in s3queue_buckets setting. " + "Existing table metadata in ZooKeeper differs in buckets setting. " "Stored in ZooKeeper: {}, local: {}", - expected.s3queue_buckets, current.s3queue_buckets); + expected.buckets, current.buckets); } - if (S3QueueMetadata::getBucketsNum(current) != S3QueueMetadata::getBucketsNum(expected)) + if (ObjectStorageQueueMetadata::getBucketsNum(current) != ObjectStorageQueueMetadata::getBucketsNum(expected)) { throw Exception( ErrorCodes::METADATA_MISMATCH, "Existing table metadata in ZooKeeper differs in processing buckets. " "Stored in ZooKeeper: {}, local: {}", - S3QueueMetadata::getBucketsNum(current), S3QueueMetadata::getBucketsNum(expected)); + ObjectStorageQueueMetadata::getBucketsNum(current), ObjectStorageQueueMetadata::getBucketsNum(expected)); } } } diff --git a/src/Storages/S3Queue/S3QueueTableMetadata.h b/src/Storages/ObjectStorageQueue/ObjectStorageQueueTableMetadata.h similarity index 50% rename from src/Storages/S3Queue/S3QueueTableMetadata.h rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueTableMetadata.h index d53b60570ae..bbae06b66c6 100644 --- a/src/Storages/S3Queue/S3QueueTableMetadata.h +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueTableMetadata.h @@ -1,6 +1,6 @@ #pragma once -#include +#include #include #include #include @@ -11,10 +11,10 @@ namespace DB class WriteBuffer; class ReadBuffer; -/** The basic parameters of S3Queue table engine for saving in ZooKeeper. +/** The basic parameters of ObjectStorageQueue table engine for saving in ZooKeeper. * Lets you verify that they match local ones. */ -struct S3QueueTableMetadata +struct ObjectStorageQueueTableMetadata { String format_name; String columns; @@ -26,22 +26,22 @@ struct S3QueueTableMetadata UInt64 processing_threads_num = 1; String last_processed_path; - S3QueueTableMetadata() = default; - S3QueueTableMetadata( + ObjectStorageQueueTableMetadata() = default; + ObjectStorageQueueTableMetadata( const StorageObjectStorage::Configuration & configuration, - const S3QueueSettings & engine_settings, + const ObjectStorageQueueSettings & engine_settings, const StorageInMemoryMetadata & storage_metadata); void read(const String & metadata_str); - static S3QueueTableMetadata parse(const String & metadata_str); + static ObjectStorageQueueTableMetadata parse(const String & metadata_str); String toString() const; - void checkEquals(const S3QueueTableMetadata & from_zk) const; - static void checkEquals(const S3QueueSettings & current, const S3QueueSettings & expected); + void checkEquals(const ObjectStorageQueueTableMetadata & from_zk) const; + static void checkEquals(const ObjectStorageQueueSettings & current, const ObjectStorageQueueSettings & expected); private: - void checkImmutableFieldsEquals(const S3QueueTableMetadata & from_zk) const; + void checkImmutableFieldsEquals(const ObjectStorageQueueTableMetadata & from_zk) const; }; diff --git a/src/Storages/S3Queue/S3QueueUnorderedFileMetadata.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueUnorderedFileMetadata.cpp similarity index 92% rename from src/Storages/S3Queue/S3QueueUnorderedFileMetadata.cpp rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueUnorderedFileMetadata.cpp index 2f7c238cd4d..40751d9c332 100644 --- a/src/Storages/S3Queue/S3QueueUnorderedFileMetadata.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueUnorderedFileMetadata.cpp @@ -1,4 +1,4 @@ -#include +#include #include #include #include @@ -18,13 +18,13 @@ namespace } } -S3QueueUnorderedFileMetadata::S3QueueUnorderedFileMetadata( +ObjectStorageQueueUnorderedFileMetadata::ObjectStorageQueueUnorderedFileMetadata( const std::filesystem::path & zk_path, const std::string & path_, FileStatusPtr file_status_, size_t max_loading_retries_, LoggerPtr log_) - : S3QueueIFileMetadata( + : ObjectStorageQueueIFileMetadata( path_, /* processing_node_path */zk_path / "processing" / getNodeName(path_), /* processed_node_path */zk_path / "processed" / getNodeName(path_), @@ -35,7 +35,7 @@ S3QueueUnorderedFileMetadata::S3QueueUnorderedFileMetadata( { } -std::pair S3QueueUnorderedFileMetadata::setProcessingImpl() +std::pair ObjectStorageQueueUnorderedFileMetadata::setProcessingImpl() { /// In one zookeeper transaction do the following: enum RequestType @@ -89,7 +89,7 @@ std::pair S3QueueUnorderedFileMet throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected state of zookeeper transaction: {}", magic_enum::enum_name(code)); } -void S3QueueUnorderedFileMetadata::setProcessedAtStartRequests( +void ObjectStorageQueueUnorderedFileMetadata::setProcessedAtStartRequests( Coordination::Requests & requests, const zkutil::ZooKeeperPtr &) { @@ -98,7 +98,7 @@ void S3QueueUnorderedFileMetadata::setProcessedAtStartRequests( processed_node_path, node_metadata.toString(), zkutil::CreateMode::Persistent)); } -void S3QueueUnorderedFileMetadata::setProcessedImpl() +void ObjectStorageQueueUnorderedFileMetadata::setProcessedImpl() { /// In one zookeeper transaction do the following: enum RequestType diff --git a/src/Storages/S3Queue/S3QueueUnorderedFileMetadata.h b/src/Storages/ObjectStorageQueue/ObjectStorageQueueUnorderedFileMetadata.h similarity index 75% rename from src/Storages/S3Queue/S3QueueUnorderedFileMetadata.h rename to src/Storages/ObjectStorageQueue/ObjectStorageQueueUnorderedFileMetadata.h index 24c2765bf3a..cc5d8a09ec9 100644 --- a/src/Storages/S3Queue/S3QueueUnorderedFileMetadata.h +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueUnorderedFileMetadata.h @@ -1,17 +1,17 @@ #pragma once -#include +#include #include #include namespace DB { -class S3QueueUnorderedFileMetadata : public S3QueueIFileMetadata +class ObjectStorageQueueUnorderedFileMetadata : public ObjectStorageQueueIFileMetadata { public: using Bucket = size_t; - explicit S3QueueUnorderedFileMetadata( + explicit ObjectStorageQueueUnorderedFileMetadata( const std::filesystem::path & zk_path, const std::string & path_, FileStatusPtr file_status_, diff --git a/src/Storages/S3Queue/StorageS3Queue.cpp b/src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.cpp similarity index 56% rename from src/Storages/S3Queue/StorageS3Queue.cpp rename to src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.cpp index b1253516f17..14b828e7268 100644 --- a/src/Storages/S3Queue/StorageS3Queue.cpp +++ b/src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.cpp @@ -1,10 +1,7 @@ #include -#include "config.h" #include -#include #include -#include #include #include #include @@ -15,16 +12,15 @@ #include #include #include -#include -#include -#include -#include -#include +#include +#include +#include +#include +#include #include #include #include #include -#include #include #include #include @@ -33,12 +29,6 @@ namespace fs = std::filesystem; -namespace ProfileEvents -{ - extern const Event S3DeleteObjects; - extern const Event S3ListObjects; -} - namespace DB { @@ -46,23 +36,22 @@ namespace ErrorCodes { extern const int LOGICAL_ERROR; extern const int BAD_ARGUMENTS; - extern const int S3_ERROR; extern const int QUERY_NOT_ALLOWED; } namespace { - std::string chooseZooKeeperPath(const StorageID & table_id, const Settings & settings, const S3QueueSettings & s3queue_settings) + std::string chooseZooKeeperPath(const StorageID & table_id, const Settings & settings, const ObjectStorageQueueSettings & queue_settings) { std::string zk_path_prefix = settings.s3queue_default_zookeeper_path.value; if (zk_path_prefix.empty()) zk_path_prefix = "/"; std::string result_zk_path; - if (s3queue_settings.keeper_path.changed) + if (queue_settings.keeper_path.changed) { /// We do not add table uuid here on purpose. - result_zk_path = fs::path(zk_path_prefix) / s3queue_settings.keeper_path.value; + result_zk_path = fs::path(zk_path_prefix) / queue_settings.keeper_path.value; } else { @@ -73,49 +62,66 @@ namespace } void checkAndAdjustSettings( - S3QueueSettings & s3queue_settings, - const Settings & settings, + ObjectStorageQueueSettings & queue_settings, + ASTStorage * engine_args, bool is_attach, - const LoggerPtr & log, - ASTStorage * engine_args) + const LoggerPtr & log) { - if (!is_attach && !s3queue_settings.mode.changed) + if (!is_attach && !queue_settings.mode.changed) { throw Exception(ErrorCodes::BAD_ARGUMENTS, "Setting `mode` (Unordered/Ordered) is not specified, but is required."); } /// In case !is_attach, we leave Ordered mode as default for compatibility. - if (!s3queue_settings.s3queue_enable_logging_to_s3queue_log.changed) + if (!queue_settings.processing_threads_num) { - s3queue_settings.s3queue_enable_logging_to_s3queue_log = settings.s3queue_enable_logging_to_s3queue_log; + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Setting `processing_threads_num` cannot be set to zero"); } - if (s3queue_settings.s3queue_cleanup_interval_min_ms > s3queue_settings.s3queue_cleanup_interval_max_ms) + if (queue_settings.cleanup_interval_min_ms > queue_settings.cleanup_interval_max_ms) { throw Exception(ErrorCodes::BAD_ARGUMENTS, - "Setting `s3queue_cleanup_interval_min_ms` ({}) must be less or equal to `s3queue_cleanup_interval_max_ms` ({})", - s3queue_settings.s3queue_cleanup_interval_min_ms, s3queue_settings.s3queue_cleanup_interval_max_ms); + "Setting `cleanup_interval_min_ms` ({}) must be less or equal to `cleanup_interval_max_ms` ({})", + queue_settings.cleanup_interval_min_ms, queue_settings.cleanup_interval_max_ms); } - if (!s3queue_settings.s3queue_processing_threads_num) + if (!is_attach && !queue_settings.processing_threads_num.changed) { - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Setting `s3queue_processing_threads_num` cannot be set to zero"); - } - - if (!is_attach && !s3queue_settings.s3queue_processing_threads_num.changed) - { - s3queue_settings.s3queue_processing_threads_num = std::max(getNumberOfPhysicalCPUCores(), 16); + queue_settings.processing_threads_num = std::max(getNumberOfPhysicalCPUCores(), 16); engine_args->settings->as()->changes.insertSetting( - "s3queue_processing_threads_num", - s3queue_settings.s3queue_processing_threads_num.value); + "processing_threads_num", + queue_settings.processing_threads_num.value); - LOG_TRACE(log, "Set `processing_threads_num` to {}", s3queue_settings.s3queue_processing_threads_num); + LOG_TRACE(log, "Set `processing_threads_num` to {}", queue_settings.processing_threads_num); } } + + std::shared_ptr getQueueLog(const ObjectStoragePtr & storage, const ContextPtr & context, const ObjectStorageQueueSettings & table_settings) + { + const auto & settings = context->getSettingsRef(); + switch (storage->getType()) + { + case DB::ObjectStorageType::S3: + { + if (table_settings.enable_logging_to_queue_log || settings.s3queue_enable_logging_to_s3queue_log) + return context->getS3QueueLog(); + return nullptr; + } + case DB::ObjectStorageType::Azure: + { + if (table_settings.enable_logging_to_queue_log) + return context->getAzureQueueLog(); + return nullptr; + } + default: + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected object storage type: {}", storage->getType()); + } + + } } -StorageS3Queue::StorageS3Queue( - std::unique_ptr s3queue_settings_, +StorageObjectStorageQueue::StorageObjectStorageQueue( + std::unique_ptr queue_settings_, const ConfigurationPtr configuration_, const StorageID & table_id_, const ColumnsDescription & columns_, @@ -127,12 +133,12 @@ StorageS3Queue::StorageS3Queue( LoadingStrictnessLevel mode) : IStorage(table_id_) , WithContext(context_) - , s3queue_settings(std::move(s3queue_settings_)) - , zk_path(chooseZooKeeperPath(table_id_, context_->getSettingsRef(), *s3queue_settings)) + , queue_settings(std::move(queue_settings_)) + , zk_path(chooseZooKeeperPath(table_id_, context_->getSettingsRef(), *queue_settings)) , configuration{configuration_} , format_settings(format_settings_) - , reschedule_processing_interval_ms(s3queue_settings->s3queue_polling_min_timeout_ms) - , log(getLogger("StorageS3Queue (" + table_id_.getFullTableName() + ")")) + , reschedule_processing_interval_ms(queue_settings->polling_min_timeout_ms) + , log(getLogger(fmt::format("Storage{}Queue ({})", configuration->getEngineName(), table_id_.getFullTableName()))) { if (configuration->getPath().empty()) { @@ -144,10 +150,10 @@ StorageS3Queue::StorageS3Queue( } else if (!configuration->isPathWithGlobs()) { - throw Exception(ErrorCodes::QUERY_NOT_ALLOWED, "S3Queue url must either end with '/' or contain globs"); + throw Exception(ErrorCodes::QUERY_NOT_ALLOWED, "ObjectStorageQueue url must either end with '/' or contain globs"); } - checkAndAdjustSettings(*s3queue_settings, context_->getSettingsRef(), mode > LoadingStrictnessLevel::CREATE, log, engine_args); + checkAndAdjustSettings(*queue_settings, engine_args, mode > LoadingStrictnessLevel::CREATE, log); object_storage = configuration->createObjectStorage(context_, /* is_readonly */true); FormatFactory::instance().checkFormatName(configuration->format); @@ -165,30 +171,30 @@ StorageS3Queue::StorageS3Queue( setInMemoryMetadata(storage_metadata); LOG_INFO(log, "Using zookeeper path: {}", zk_path.string()); - task = getContext()->getSchedulePool().createTask("S3QueueStreamingTask", [this] { threadFunc(); }); + task = getContext()->getSchedulePool().createTask("ObjectStorageQueueStreamingTask", [this] { threadFunc(); }); - /// Get metadata manager from S3QueueMetadataFactory, + /// Get metadata manager from ObjectStorageQueueMetadataFactory, /// it will increase the ref count for the metadata object. - /// The ref count is decreased when StorageS3Queue::drop() method is called. - files_metadata = S3QueueMetadataFactory::instance().getOrCreate(zk_path, *s3queue_settings); + /// The ref count is decreased when StorageObjectStorageQueue::drop() method is called. + files_metadata = ObjectStorageQueueMetadataFactory::instance().getOrCreate(zk_path, *queue_settings); try { files_metadata->initialize(configuration_, storage_metadata); } catch (...) { - S3QueueMetadataFactory::instance().remove(zk_path); + ObjectStorageQueueMetadataFactory::instance().remove(zk_path); throw; } } -void StorageS3Queue::startup() +void StorageObjectStorageQueue::startup() { if (task) task->activateAndSchedule(); } -void StorageS3Queue::shutdown(bool is_drop) +void StorageObjectStorageQueue::shutdown(bool is_drop) { table_is_being_dropped = is_drop; shutdown_called = true; @@ -207,31 +213,31 @@ void StorageS3Queue::shutdown(bool is_drop) LOG_TRACE(log, "Shut down storage"); } -void StorageS3Queue::drop() +void StorageObjectStorageQueue::drop() { - S3QueueMetadataFactory::instance().remove(zk_path); + ObjectStorageQueueMetadataFactory::instance().remove(zk_path); } -bool StorageS3Queue::supportsSubsetOfColumns(const ContextPtr & context_) const +bool StorageObjectStorageQueue::supportsSubsetOfColumns(const ContextPtr & context_) const { return FormatFactory::instance().checkIfFormatSupportsSubsetOfColumns(configuration->format, context_, format_settings); } -class ReadFromS3Queue : public SourceStepWithFilter +class ReadFromObjectStorageQueue : public SourceStepWithFilter { public: - std::string getName() const override { return "ReadFromS3Queue"; } + std::string getName() const override { return "ReadFromObjectStorageQueue"; } void initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) override; void applyFilters(ActionDAGNodes added_filter_nodes) override; - ReadFromS3Queue( + ReadFromObjectStorageQueue( const Names & column_names_, const SelectQueryInfo & query_info_, const StorageSnapshotPtr & storage_snapshot_, const ContextPtr & context_, Block sample_block, ReadFromFormatInfo info_, - std::shared_ptr storage_, + std::shared_ptr storage_, size_t max_block_size_) : SourceStepWithFilter( DataStream{.header = std::move(sample_block)}, @@ -247,15 +253,15 @@ public: private: ReadFromFormatInfo info; - std::shared_ptr storage; + std::shared_ptr storage; size_t max_block_size; - std::shared_ptr iterator; + std::shared_ptr iterator; void createIterator(const ActionsDAG::Node * predicate); }; -void ReadFromS3Queue::createIterator(const ActionsDAG::Node * predicate) +void ReadFromObjectStorageQueue::createIterator(const ActionsDAG::Node * predicate) { if (iterator) return; @@ -264,7 +270,7 @@ void ReadFromS3Queue::createIterator(const ActionsDAG::Node * predicate) } -void ReadFromS3Queue::applyFilters(ActionDAGNodes added_filter_nodes) +void ReadFromObjectStorageQueue::applyFilters(ActionDAGNodes added_filter_nodes) { SourceStepWithFilter::applyFilters(std::move(added_filter_nodes)); @@ -275,7 +281,7 @@ void ReadFromS3Queue::applyFilters(ActionDAGNodes added_filter_nodes) createIterator(predicate); } -void StorageS3Queue::read( +void StorageObjectStorageQueue::read( QueryPlan & query_plan, const Names & column_names, const StorageSnapshotPtr & storage_snapshot, @@ -297,10 +303,10 @@ void StorageS3Queue::read( "Cannot read from {} with attached materialized views", getName()); } - auto this_ptr = std::static_pointer_cast(shared_from_this()); + auto this_ptr = std::static_pointer_cast(shared_from_this()); auto read_from_format_info = prepareReadingFromFormat(column_names, storage_snapshot, supportsSubsetOfColumns(local_context)); - auto reading = std::make_unique( + auto reading = std::make_unique( column_names, query_info, storage_snapshot, @@ -313,10 +319,10 @@ void StorageS3Queue::read( query_plan.addStep(std::move(reading)); } -void ReadFromS3Queue::initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) +void ReadFromObjectStorageQueue::initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) { Pipes pipes; - const size_t adjusted_num_streams = storage->s3queue_settings->s3queue_processing_threads_num; + const size_t adjusted_num_streams = storage->queue_settings->processing_threads_num; createIterator(nullptr); for (size_t i = 0; i < adjusted_num_streams; ++i) @@ -338,54 +344,25 @@ void ReadFromS3Queue::initializePipeline(QueryPipelineBuilder & pipeline, const pipeline.init(std::move(pipe)); } -std::shared_ptr StorageS3Queue::createSource( +std::shared_ptr StorageObjectStorageQueue::createSource( size_t processor_id, const ReadFromFormatInfo & info, - std::shared_ptr file_iterator, + std::shared_ptr file_iterator, size_t max_block_size, ContextPtr local_context, bool commit_once_processed) { - auto internal_source = std::make_unique( - getName(), - object_storage, - configuration, - info, - format_settings, - local_context, - max_block_size, - file_iterator, - local_context->getSettingsRef().max_download_threads, - false); - - auto file_deleter = [=, this](const std::string & path) mutable - { - object_storage->removeObject(StoredObject(path)); - }; - auto s3_queue_log = s3queue_settings->s3queue_enable_logging_to_s3queue_log ? local_context->getS3QueueLog() : nullptr; - return std::make_shared( - getName(), - processor_id, - info.source_header, - std::move(internal_source), - files_metadata, - s3queue_settings->after_processing, - file_deleter, - info.requested_virtual_columns, - local_context, - shutdown_called, - table_is_being_dropped, - s3_queue_log, - getStorageID(), - log, - s3queue_settings->s3queue_max_processed_files_before_commit, - s3queue_settings->s3queue_max_processed_rows_before_commit, - s3queue_settings->s3queue_max_processed_bytes_before_commit, - s3queue_settings->s3queue_max_processing_time_sec_before_commit, - commit_once_processed); + return std::make_shared( + getName(), processor_id, + file_iterator, configuration, object_storage, + info, format_settings, + *queue_settings, files_metadata, + local_context, max_block_size, shutdown_called, table_is_being_dropped, + getQueueLog(object_storage, local_context, *queue_settings), + getStorageID(), log, commit_once_processed); } -bool StorageS3Queue::hasDependencies(const StorageID & table_id) +bool StorageObjectStorageQueue::hasDependencies(const StorageID & table_id) { // Check if all dependencies are attached auto view_ids = DatabaseCatalog::instance().getDependentViews(table_id); @@ -410,7 +387,7 @@ bool StorageS3Queue::hasDependencies(const StorageID & table_id) return true; } -void StorageS3Queue::threadFunc() +void StorageObjectStorageQueue::threadFunc() { if (shutdown_called) return; @@ -428,12 +405,12 @@ void StorageS3Queue::threadFunc() if (streamToViews()) { /// Reset the reschedule interval. - reschedule_processing_interval_ms = s3queue_settings->s3queue_polling_min_timeout_ms; + reschedule_processing_interval_ms = queue_settings->polling_min_timeout_ms; } else { /// Increase the reschedule interval. - reschedule_processing_interval_ms += s3queue_settings->s3queue_polling_backoff_ms; + reschedule_processing_interval_ms += queue_settings->polling_backoff_ms; } LOG_DEBUG(log, "Stopped streaming to {} attached views", dependencies_count); @@ -450,12 +427,12 @@ void StorageS3Queue::threadFunc() if (!shutdown_called) { - LOG_TRACE(log, "Reschedule S3 Queue processing thread in {} ms", reschedule_processing_interval_ms); + LOG_TRACE(log, "Reschedule processing thread in {} ms", reschedule_processing_interval_ms); task->scheduleAfter(reschedule_processing_interval_ms); } } -bool StorageS3Queue::streamToViews() +bool StorageObjectStorageQueue::streamToViews() { // Create a stream for each consumer and join them in a union stream // Only insert into dependent views and expect that input blocks contain virtual columns @@ -469,35 +446,41 @@ bool StorageS3Queue::streamToViews() insert->table_id = table_id; auto storage_snapshot = getStorageSnapshot(getInMemoryMetadataPtr(), getContext()); - auto s3queue_context = Context::createCopy(getContext()); - s3queue_context->makeQueryContext(); + auto queue_context = Context::createCopy(getContext()); + queue_context->makeQueryContext(); - auto file_iterator = createFileIterator(s3queue_context, nullptr); + auto file_iterator = createFileIterator(queue_context, nullptr); size_t total_rows = 0; while (!shutdown_called && !file_iterator->isFinished()) { - InterpreterInsertQuery interpreter(insert, s3queue_context, false, true, true); + InterpreterInsertQuery interpreter( + insert, + queue_context, + /* allow_materialized */ false, + /* no_squash */ true, + /* no_destination */ true, + /* async_isnert */ false); auto block_io = interpreter.execute(); auto read_from_format_info = prepareReadingFromFormat( block_io.pipeline.getHeader().getNames(), storage_snapshot, - supportsSubsetOfColumns(s3queue_context)); + supportsSubsetOfColumns(queue_context)); Pipes pipes; - std::vector> sources; + std::vector> sources; - pipes.reserve(s3queue_settings->s3queue_processing_threads_num); - sources.reserve(s3queue_settings->s3queue_processing_threads_num); + pipes.reserve(queue_settings->processing_threads_num); + sources.reserve(queue_settings->processing_threads_num); - for (size_t i = 0; i < s3queue_settings->s3queue_processing_threads_num; ++i) + for (size_t i = 0; i < queue_settings->processing_threads_num; ++i) { auto source = createSource( i/* processor_id */, read_from_format_info, file_iterator, DBMS_DEFAULT_BUFFER_SIZE, - s3queue_context, + queue_context, false/* commit_once_processed */); pipes.emplace_back(source); @@ -506,8 +489,8 @@ bool StorageS3Queue::streamToViews() auto pipe = Pipe::unitePipes(std::move(pipes)); block_io.pipeline.complete(std::move(pipe)); - block_io.pipeline.setNumThreads(s3queue_settings->s3queue_processing_threads_num); - block_io.pipeline.setConcurrencyControl(s3queue_context->getSettingsRef().use_concurrency_control); + block_io.pipeline.setNumThreads(queue_settings->processing_threads_num); + block_io.pipeline.setConcurrencyControl(queue_context->getSettingsRef().use_concurrency_control); std::atomic_size_t rows = 0; block_io.pipeline.setProgressCallback([&](const Progress & progress) { rows += progress.read_rows.load(); }); @@ -536,12 +519,12 @@ bool StorageS3Queue::streamToViews() return total_rows > 0; } -zkutil::ZooKeeperPtr StorageS3Queue::getZooKeeper() const +zkutil::ZooKeeperPtr StorageObjectStorageQueue::getZooKeeper() const { return getContext()->getZooKeeper(); } -std::shared_ptr StorageS3Queue::createFileIterator(ContextPtr local_context, const ActionsDAG::Node * predicate) +std::shared_ptr StorageObjectStorageQueue::createFileIterator(ContextPtr local_context, const ActionsDAG::Node * predicate) { auto settings = configuration->getQuerySettings(local_context); auto glob_iterator = std::make_unique( @@ -550,73 +533,4 @@ std::shared_ptr StorageS3Queue::createFileIterator return std::make_shared(files_metadata, std::move(glob_iterator), shutdown_called, log); } -#if USE_AWS_S3 -void registerStorageS3Queue(StorageFactory & factory) -{ - factory.registerStorage( - "S3Queue", - [](const StorageFactory::Arguments & args) - { - auto & engine_args = args.engine_args; - if (engine_args.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "External data source must have arguments"); - - auto configuration = std::make_shared(); - StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getContext(), false); - - // Use format settings from global server context + settings from - // the SETTINGS clause of the create query. Settings from current - // session and user are ignored. - std::optional format_settings; - - auto s3queue_settings = std::make_unique(); - if (args.storage_def->settings) - { - s3queue_settings->loadFromQuery(*args.storage_def); - FormatFactorySettings user_format_settings; - - // Apply changed settings from global context, but ignore the - // unknown ones, because we only have the format settings here. - const auto & changes = args.getContext()->getSettingsRef().changes(); - for (const auto & change : changes) - { - if (user_format_settings.has(change.name)) - user_format_settings.set(change.name, change.value); - else - LOG_TRACE(getLogger("StorageS3"), "Remove: {}", change.name); - args.storage_def->settings->changes.removeSetting(change.name); - } - - for (const auto & change : args.storage_def->settings->changes) - { - if (user_format_settings.has(change.name)) - user_format_settings.applyChange(change); - } - format_settings = getFormatSettings(args.getContext(), user_format_settings); - } - else - { - format_settings = getFormatSettings(args.getContext()); - } - - return std::make_shared( - std::move(s3queue_settings), - std::move(configuration), - args.table_id, - args.columns, - args.constraints, - args.comment, - args.getContext(), - format_settings, - args.storage_def, - args.mode); - }, - { - .supports_settings = true, - .supports_schema_inference = true, - .source_access_type = AccessType::S3, - }); -} -#endif - } diff --git a/src/Storages/S3Queue/StorageS3Queue.h b/src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.h similarity index 74% rename from src/Storages/S3Queue/StorageS3Queue.h rename to src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.h index f465fa92d1a..fc459c45f74 100644 --- a/src/Storages/S3Queue/StorageS3Queue.h +++ b/src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.h @@ -5,25 +5,24 @@ #include #include #include -#include -#include +#include +#include #include #include -#include #include namespace DB { -class S3QueueMetadata; +class ObjectStorageQueueMetadata; -class StorageS3Queue : public IStorage, WithContext +class StorageObjectStorageQueue : public IStorage, WithContext { public: using ConfigurationPtr = StorageObjectStorage::ConfigurationPtr; - StorageS3Queue( - std::unique_ptr s3queue_settings_, + StorageObjectStorageQueue( + std::unique_ptr queue_settings_, ConfigurationPtr configuration_, const StorageID & table_id_, const ColumnsDescription & columns_, @@ -34,7 +33,7 @@ public: ASTStorage * engine_args, LoadingStrictnessLevel mode); - String getName() const override { return "S3Queue"; } + String getName() const override { return "ObjectStorageQueue"; } void read( QueryPlan & query_plan, @@ -53,13 +52,13 @@ public: zkutil::ZooKeeperPtr getZooKeeper() const; private: - friend class ReadFromS3Queue; - using FileIterator = StorageS3QueueSource::FileIterator; + friend class ReadFromObjectStorageQueue; + using FileIterator = ObjectStorageQueueSource::FileIterator; - const std::unique_ptr s3queue_settings; + const std::unique_ptr queue_settings; const fs::path zk_path; - std::shared_ptr files_metadata; + std::shared_ptr files_metadata; ConfigurationPtr configuration; ObjectStoragePtr object_storage; @@ -80,13 +79,14 @@ private: void drop() override; bool supportsSubsetOfColumns(const ContextPtr & context_) const; bool supportsSubcolumns() const override { return true; } + bool supportsOptimizationToSubcolumns() const override { return false; } bool supportsDynamicSubcolumns() const override { return true; } std::shared_ptr createFileIterator(ContextPtr local_context, const ActionsDAG::Node * predicate); - std::shared_ptr createSource( + std::shared_ptr createSource( size_t processor_id, const ReadFromFormatInfo & info, - std::shared_ptr file_iterator, + std::shared_ptr file_iterator, size_t max_block_size, ContextPtr local_context, bool commit_once_processed); diff --git a/src/Storages/ObjectStorageQueue/registerQueueStorage.cpp b/src/Storages/ObjectStorageQueue/registerQueueStorage.cpp new file mode 100644 index 00000000000..20968143627 --- /dev/null +++ b/src/Storages/ObjectStorageQueue/registerQueueStorage.cpp @@ -0,0 +1,115 @@ +#include "config.h" + +#include +#include +#include +#include + +#if USE_AWS_S3 +#include +#include +#endif + +#if USE_AZURE_BLOB_STORAGE +#include +#endif + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; +} + +template +StoragePtr createQueueStorage(const StorageFactory::Arguments & args) +{ + auto & engine_args = args.engine_args; + if (engine_args.empty()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "External data source must have arguments"); + + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getContext(), false); + + // Use format settings from global server context + settings from + // the SETTINGS clause of the create query. Settings from current + // session and user are ignored. + std::optional format_settings; + + auto queue_settings = std::make_unique(); + if (args.storage_def->settings) + { + queue_settings->loadFromQuery(*args.storage_def); + FormatFactorySettings user_format_settings; + + // Apply changed settings from global context, but ignore the + // unknown ones, because we only have the format settings here. + const auto & changes = args.getContext()->getSettingsRef().changes(); + for (const auto & change : changes) + { + if (user_format_settings.has(change.name)) + user_format_settings.set(change.name, change.value); + + args.storage_def->settings->changes.removeSetting(change.name); + } + + for (const auto & change : args.storage_def->settings->changes) + { + if (user_format_settings.has(change.name)) + user_format_settings.applyChange(change); + } + format_settings = getFormatSettings(args.getContext(), user_format_settings); + } + else + { + format_settings = getFormatSettings(args.getContext()); + } + + return std::make_shared( + std::move(queue_settings), + std::move(configuration), + args.table_id, + args.columns, + args.constraints, + args.comment, + args.getContext(), + format_settings, + args.storage_def, + args.mode); +} + +#if USE_AWS_S3 +void registerStorageS3Queue(StorageFactory & factory) +{ + factory.registerStorage( + "S3Queue", + [](const StorageFactory::Arguments & args) + { + return createQueueStorage(args); + }, + { + .supports_settings = true, + .supports_schema_inference = true, + .source_access_type = AccessType::S3, + }); +} +#endif + +#if USE_AZURE_BLOB_STORAGE +void registerStorageAzureQueue(StorageFactory & factory) +{ + factory.registerStorage( + "AzureQueue", + [](const StorageFactory::Arguments & args) + { + return createQueueStorage(args); + }, + { + .supports_settings = true, + .supports_schema_inference = true, + .source_access_type = AccessType::AZURE, + }); +} +#endif +} diff --git a/src/Storages/PartitionedSink.cpp b/src/Storages/PartitionedSink.cpp index 09b009b26d8..ee2570756ed 100644 --- a/src/Storages/PartitionedSink.cpp +++ b/src/Storages/PartitionedSink.cpp @@ -51,7 +51,7 @@ SinkPtr PartitionedSink::getSinkForPartitionKey(StringRef partition_key) return it->second; } -void PartitionedSink::consume(Chunk chunk) +void PartitionedSink::consume(Chunk & chunk) { const auto & columns = chunk.getColumns(); @@ -104,7 +104,7 @@ void PartitionedSink::consume(Chunk chunk) for (const auto & [partition_key, partition_index] : partition_id_to_chunk_index) { auto sink = getSinkForPartitionKey(partition_key); - sink->consume(std::move(partition_index_to_chunk[partition_index])); + sink->consume(partition_index_to_chunk[partition_index]); } } diff --git a/src/Storages/PartitionedSink.h b/src/Storages/PartitionedSink.h index 68edeb6fd73..fcd67556dc9 100644 --- a/src/Storages/PartitionedSink.h +++ b/src/Storages/PartitionedSink.h @@ -20,7 +20,7 @@ public: String getName() const override { return "PartitionedSink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onException(std::exception_ptr exception) override; diff --git a/src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp b/src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp index ba3cc6f58d0..44479bd01e2 100644 --- a/src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp +++ b/src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp @@ -697,7 +697,13 @@ void MaterializedPostgreSQLConsumer::syncTables() insert->table_id = storage->getStorageID(); insert->columns = std::make_shared(buffer->columns_ast); - InterpreterInsertQuery interpreter(insert, insert_context, true); + InterpreterInsertQuery interpreter( + insert, + insert_context, + /* allow_materialized */ true, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); auto io = interpreter.execute(); auto input = std::make_shared( result_rows.cloneEmpty(), Chunk(result_rows.getColumns(), result_rows.rows())); diff --git a/src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp b/src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp index 2bb1e2dde0d..f632e553a0d 100644 --- a/src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp +++ b/src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp @@ -437,7 +437,13 @@ StorageInfo PostgreSQLReplicationHandler::loadFromSnapshot(postgres::Connection auto insert_context = materialized_storage->getNestedTableContext(); - InterpreterInsertQuery interpreter(insert, insert_context); + InterpreterInsertQuery interpreter( + insert, + insert_context, + /* allow_materialized */ false, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); auto block_io = interpreter.execute(); const StorageInMemoryMetadata & storage_metadata = nested_storage->getInMemoryMetadata(); diff --git a/src/Storages/PostgreSQL/StorageMaterializedPostgreSQL.cpp b/src/Storages/PostgreSQL/StorageMaterializedPostgreSQL.cpp index f686fbda664..f4c38a52a3f 100644 --- a/src/Storages/PostgreSQL/StorageMaterializedPostgreSQL.cpp +++ b/src/Storages/PostgreSQL/StorageMaterializedPostgreSQL.cpp @@ -21,7 +21,6 @@ #include #include -#include #include #include #include @@ -572,6 +571,7 @@ void registerStorageMaterializedPostgreSQL(StorageFactory & factory) StorageInMemoryMetadata metadata; metadata.setColumns(args.columns); metadata.setConstraints(args.constraints); + metadata.setComment(args.comment); if (args.mode <= LoadingStrictnessLevel::CREATE && !args.getLocalContext()->getSettingsRef().allow_experimental_materialized_postgresql_table) @@ -592,7 +592,8 @@ void registerStorageMaterializedPostgreSQL(StorageFactory & factory) auto configuration = StoragePostgreSQL::getConfiguration(args.engine_args, args.getContext()); auto connection_info = postgres::formatConnectionString( configuration.database, configuration.host, configuration.port, - configuration.username, configuration.password); + configuration.username, configuration.password, + args.getContext()->getSettingsRef().postgresql_connection_attempt_timeout); bool has_settings = args.storage_def->settings; auto postgresql_replication_settings = std::make_unique(); diff --git a/src/Storages/RabbitMQ/StorageRabbitMQ.cpp b/src/Storages/RabbitMQ/StorageRabbitMQ.cpp index e4b19992151..9e3c40071b5 100644 --- a/src/Storages/RabbitMQ/StorageRabbitMQ.cpp +++ b/src/Storages/RabbitMQ/StorageRabbitMQ.cpp @@ -70,6 +70,7 @@ StorageRabbitMQ::StorageRabbitMQ( const StorageID & table_id_, ContextPtr context_, const ColumnsDescription & columns_, + const String & comment, std::unique_ptr rabbitmq_settings_, LoadingStrictnessLevel mode) : IStorage(table_id_) @@ -145,6 +146,7 @@ StorageRabbitMQ::StorageRabbitMQ( StorageInMemoryMetadata storage_metadata; storage_metadata.setColumns(columns_); + storage_metadata.setComment(comment); setInMemoryMetadata(storage_metadata); setVirtuals(createVirtuals(rabbitmq_settings->rabbitmq_handle_error_mode)); @@ -1129,7 +1131,13 @@ bool StorageRabbitMQ::tryStreamToViews() } // Only insert into dependent views and expect that input blocks contain virtual columns - InterpreterInsertQuery interpreter(insert, rabbitmq_context, /* allow_materialized_ */ false, /* no_squash_ */ true, /* no_destination_ */ true); + InterpreterInsertQuery interpreter( + insert, + rabbitmq_context, + /* allow_materialized */ false, + /* no_squash */ true, + /* no_destination */ true, + /* async_isnert */ false); auto block_io = interpreter.execute(); block_io.pipeline.complete(Pipe::unitePipes(std::move(pipes))); @@ -1282,7 +1290,7 @@ void registerStorageRabbitMQ(StorageFactory & factory) if (!rabbitmq_settings->rabbitmq_format.changed) throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "You must specify `rabbitmq_format` setting"); - return std::make_shared(args.table_id, args.getContext(), args.columns, std::move(rabbitmq_settings), args.mode); + return std::make_shared(args.table_id, args.getContext(), args.columns, args.comment, std::move(rabbitmq_settings), args.mode); }; factory.registerStorage("RabbitMQ", creator_fn, StorageFactory::StorageFeatures{ .supports_settings = true, }); diff --git a/src/Storages/RabbitMQ/StorageRabbitMQ.h b/src/Storages/RabbitMQ/StorageRabbitMQ.h index b8fab5825e4..fed80a4357b 100644 --- a/src/Storages/RabbitMQ/StorageRabbitMQ.h +++ b/src/Storages/RabbitMQ/StorageRabbitMQ.h @@ -26,6 +26,7 @@ public: const StorageID & table_id_, ContextPtr context_, const ColumnsDescription & columns_, + const String & comment, std::unique_ptr rabbitmq_settings_, LoadingStrictnessLevel mode); diff --git a/src/Storages/ReadFinalForExternalReplicaStorage.cpp b/src/Storages/ReadFinalForExternalReplicaStorage.cpp index e1d52eefc20..5070b4a17d8 100644 --- a/src/Storages/ReadFinalForExternalReplicaStorage.cpp +++ b/src/Storages/ReadFinalForExternalReplicaStorage.cpp @@ -2,6 +2,7 @@ #if USE_MYSQL || USE_LIBPQXX +#include #include #include #include diff --git a/src/Storages/ReadInOrderOptimizer.cpp b/src/Storages/ReadInOrderOptimizer.cpp index e0ef967d491..9c8c4c2fe79 100644 --- a/src/Storages/ReadInOrderOptimizer.cpp +++ b/src/Storages/ReadInOrderOptimizer.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include diff --git a/src/Storages/RocksDB/EmbeddedRocksDBBulkSink.cpp b/src/Storages/RocksDB/EmbeddedRocksDBBulkSink.cpp index 31812406d34..f6127f3ed8a 100644 --- a/src/Storages/RocksDB/EmbeddedRocksDBBulkSink.cpp +++ b/src/Storages/RocksDB/EmbeddedRocksDBBulkSink.cpp @@ -1,23 +1,22 @@ -#include #include #include #include #include #include -#include -#include #include #include #include #include #include +#include #include #include #include #include #include #include +#include #include #include #include @@ -35,6 +34,7 @@ namespace DB namespace ErrorCodes { extern const int ROCKSDB_ERROR; + extern const int LOGICAL_ERROR; } static const IColumn::Permutation & getAscendingPermutation(const IColumn & column, IColumn::Permutation & perm) @@ -156,6 +156,7 @@ std::vector EmbeddedRocksDBBulkSink::squash(Chunk chunk) return {}; } +template std::pair EmbeddedRocksDBBulkSink::serializeChunks(std::vector && input_chunks) const { auto serialized_key_column = ColumnString::create(); @@ -169,14 +170,40 @@ std::pair EmbeddedRocksDBBulkSink::seriali WriteBufferFromVector writer_key(serialized_key_data); WriteBufferFromVector writer_value(serialized_value_data); FormatSettings format_settings; /// Format settings is 1.5KB, so it's not wise to create it for each row + + /// TTL handling + [[maybe_unused]] auto get_rocksdb_ts = [this](String & ts_string) + { + Int64 curtime = -1; + auto * system_clock = storage.rocksdb_ptr->GetEnv()->GetSystemClock().get(); + rocksdb::Status st = system_clock->GetCurrentTime(&curtime); + if (!st.ok()) + throw Exception(ErrorCodes::ROCKSDB_ERROR, "RocksDB error: {}", st.ToString()); + WriteBufferFromString buf(ts_string); + writeBinaryLittleEndian(static_cast(curtime), buf); + }; + for (auto && chunk : input_chunks) { + [[maybe_unused]] String ts_string; + if constexpr (with_timestamp) + get_rocksdb_ts(ts_string); + const auto & columns = chunk.getColumns(); auto rows = chunk.getNumRows(); for (size_t i = 0; i < rows; ++i) { for (size_t idx = 0; idx < columns.size(); ++idx) serializations[idx]->serializeBinary(*columns[idx], i, idx == primary_key_pos ? writer_key : writer_value, format_settings); + + /// Append timestamp to end of value, see rocksdb::DBWithTTLImpl::AppendTS + if constexpr (with_timestamp) + { + if (ts_string.size() != sizeof(Int32)) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Invalid timestamp size: expect 4, got {}", ts_string.size()); + writeString(ts_string, writer_value); + } + /// String in ColumnString must be null-terminated writeChar('\0', writer_key); writeChar('\0', writer_value); @@ -192,16 +219,18 @@ std::pair EmbeddedRocksDBBulkSink::seriali return {std::move(serialized_key_column), std::move(serialized_value_column)}; } -void EmbeddedRocksDBBulkSink::consume(Chunk chunk_) +void EmbeddedRocksDBBulkSink::consume(Chunk & chunk_) { std::vector chunks_to_write = squash(std::move(chunk_)); if (chunks_to_write.empty()) return; - auto [serialized_key_column, serialized_value_column] = serializeChunks(std::move(chunks_to_write)); + size_t num_chunks = chunks_to_write.size(); + auto [serialized_key_column, serialized_value_column] + = storage.ttl > 0 ? serializeChunks(std::move(chunks_to_write)) : serializeChunks(std::move(chunks_to_write)); auto sst_file_path = getTemporarySSTFilePath(); - LOG_DEBUG(getLogger("EmbeddedRocksDBBulkSink"), "Writing {} rows to SST file {}", serialized_key_column->size(), sst_file_path); + LOG_DEBUG(getLogger("EmbeddedRocksDBBulkSink"), "Writing {} rows from {} chunks to SST file {}", serialized_key_column->size(), num_chunks, sst_file_path); if (auto status = buildSSTFile(sst_file_path, *serialized_key_column, *serialized_value_column); !status.ok()) throw Exception(ErrorCodes::ROCKSDB_ERROR, "RocksDB write error: {}", status.ToString()); @@ -220,7 +249,10 @@ void EmbeddedRocksDBBulkSink::onFinish() { /// If there is any data left, write it. if (!chunks.empty()) - consume({}); + { + Chunk empty; + consume(empty); + } } String EmbeddedRocksDBBulkSink::getTemporarySSTFilePath() diff --git a/src/Storages/RocksDB/EmbeddedRocksDBBulkSink.h b/src/Storages/RocksDB/EmbeddedRocksDBBulkSink.h index 46193b152ca..64190c8c86f 100644 --- a/src/Storages/RocksDB/EmbeddedRocksDBBulkSink.h +++ b/src/Storages/RocksDB/EmbeddedRocksDBBulkSink.h @@ -1,7 +1,5 @@ #pragma once -#include -#include #include #include #include @@ -34,7 +32,7 @@ public: ~EmbeddedRocksDBBulkSink() override; - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onFinish() override; @@ -49,6 +47,7 @@ private: bool isEnoughSize(const std::vector & input_chunks) const; bool isEnoughSize(const Chunk & chunk) const; /// Serialize chunks to rocksdb key-value pairs + template std::pair serializeChunks(std::vector && input_chunks) const; StorageEmbeddedRocksDB & storage; diff --git a/src/Storages/RocksDB/EmbeddedRocksDBSink.cpp b/src/Storages/RocksDB/EmbeddedRocksDBSink.cpp index c451cfd1bf5..1f7f6939f40 100644 --- a/src/Storages/RocksDB/EmbeddedRocksDBSink.cpp +++ b/src/Storages/RocksDB/EmbeddedRocksDBSink.cpp @@ -29,7 +29,7 @@ EmbeddedRocksDBSink::EmbeddedRocksDBSink( serializations = getHeader().getSerializations(); } -void EmbeddedRocksDBSink::consume(Chunk chunk) +void EmbeddedRocksDBSink::consume(Chunk & chunk) { auto rows = chunk.getNumRows(); const auto & columns = chunk.getColumns(); diff --git a/src/Storages/RocksDB/EmbeddedRocksDBSink.h b/src/Storages/RocksDB/EmbeddedRocksDBSink.h index 011322df829..2e1e0c7b429 100644 --- a/src/Storages/RocksDB/EmbeddedRocksDBSink.h +++ b/src/Storages/RocksDB/EmbeddedRocksDBSink.h @@ -17,7 +17,7 @@ public: StorageEmbeddedRocksDB & storage_, const StorageMetadataPtr & metadata_snapshot_); - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; String getName() const override { return "EmbeddedRocksDBSink"; } private: diff --git a/src/Storages/RocksDB/StorageEmbeddedRocksDB.cpp b/src/Storages/RocksDB/StorageEmbeddedRocksDB.cpp index b9d3e071b6c..fafc72da04e 100644 --- a/src/Storages/RocksDB/StorageEmbeddedRocksDB.cpp +++ b/src/Storages/RocksDB/StorageEmbeddedRocksDB.cpp @@ -28,6 +28,7 @@ #include #include #include +#include #include #include #include @@ -313,7 +314,8 @@ void StorageEmbeddedRocksDB::mutate(const MutationCommands & commands, ContextPt Block block; while (executor.pull(block)) { - sink->consume(Chunk{block.getColumns(), block.rows()}); + auto chunk = Chunk(block.getColumns(), block.rows()); + sink->consume(chunk); } } @@ -689,6 +691,7 @@ static StoragePtr create(const StorageFactory::Arguments & args) StorageInMemoryMetadata metadata; metadata.setColumns(args.columns); metadata.setConstraints(args.constraints); + metadata.setComment(args.comment); if (!args.storage_def->primary_key) throw Exception(ErrorCodes::BAD_ARGUMENTS, "StorageEmbeddedRocksDB must require one column in primary key"); diff --git a/src/Storages/RocksDB/StorageSystemRocksDB.cpp b/src/Storages/RocksDB/StorageSystemRocksDB.cpp index 5105b190fd9..744d43e1c77 100644 --- a/src/Storages/RocksDB/StorageSystemRocksDB.cpp +++ b/src/Storages/RocksDB/StorageSystemRocksDB.cpp @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/S3Queue/S3QueueMetadataFactory.h b/src/Storages/S3Queue/S3QueueMetadataFactory.h deleted file mode 100644 index 80e96f8aa7e..00000000000 --- a/src/Storages/S3Queue/S3QueueMetadataFactory.h +++ /dev/null @@ -1,37 +0,0 @@ -#pragma once -#include -#include -#include - -namespace DB -{ - -class S3QueueMetadataFactory final : private boost::noncopyable -{ -public: - using FilesMetadataPtr = std::shared_ptr; - - static S3QueueMetadataFactory & instance(); - - FilesMetadataPtr getOrCreate(const std::string & zookeeper_path, const S3QueueSettings & settings); - - void remove(const std::string & zookeeper_path); - - std::unordered_map getAll(); - -private: - struct Metadata - { - explicit Metadata(std::shared_ptr metadata_) : metadata(metadata_), ref_count(1) {} - - std::shared_ptr metadata; - /// TODO: the ref count should be kept in keeper, because of the case with distributed processing. - size_t ref_count = 0; - }; - using MetadataByPath = std::unordered_map; - - MetadataByPath metadata_by_path; - std::mutex mutex; -}; - -} diff --git a/src/Storages/S3Queue/S3QueueSettings.h b/src/Storages/S3Queue/S3QueueSettings.h deleted file mode 100644 index 681713e8378..00000000000 --- a/src/Storages/S3Queue/S3QueueSettings.h +++ /dev/null @@ -1,51 +0,0 @@ -#pragma once - -#include -#include -#include - - -namespace DB -{ -class ASTStorage; - - -#define S3QUEUE_RELATED_SETTINGS(M, ALIAS) \ - M(S3QueueMode, \ - mode, \ - S3QueueMode::ORDERED, \ - "With unordered mode, the set of all already processed files is tracked with persistent nodes in ZooKepeer." \ - "With ordered mode, only the max name of the successfully consumed file stored.", \ - 0) \ - M(S3QueueAction, after_processing, S3QueueAction::KEEP, "Delete or keep file in S3 after successful processing", 0) \ - M(String, keeper_path, "", "Zookeeper node path", 0) \ - M(UInt32, s3queue_loading_retries, 10, "Retry loading up to specified number of times", 0) \ - M(UInt32, s3queue_processing_threads_num, 1, "Number of processing threads", 0) \ - M(UInt32, s3queue_enable_logging_to_s3queue_log, 1, "Enable logging to system table system.s3queue_log", 0) \ - M(String, s3queue_last_processed_path, "", "For Ordered mode. Files that have lexicographically smaller file name are considered already processed", 0) \ - M(UInt32, s3queue_tracked_file_ttl_sec, 0, "Maximum number of seconds to store processed files in ZooKeeper node (store forever by default)", 0) \ - M(UInt32, s3queue_polling_min_timeout_ms, 1000, "Minimal timeout before next polling", 0) \ - M(UInt32, s3queue_polling_max_timeout_ms, 10000, "Maximum timeout before next polling", 0) \ - M(UInt32, s3queue_polling_backoff_ms, 1000, "Polling backoff", 0) \ - M(UInt32, s3queue_tracked_files_limit, 1000, "For unordered mode. Max set size for tracking processed files in ZooKeeper", 0) \ - M(UInt32, s3queue_cleanup_interval_min_ms, 60000, "For unordered mode. Polling backoff min for cleanup", 0) \ - M(UInt32, s3queue_cleanup_interval_max_ms, 60000, "For unordered mode. Polling backoff max for cleanup", 0) \ - M(UInt32, s3queue_buckets, 0, "Number of buckets for Ordered mode parallel processing", 0) \ - M(UInt32, s3queue_max_processed_files_before_commit, 100, "Number of files which can be processed before being committed to keeper", 0) \ - M(UInt32, s3queue_max_processed_rows_before_commit, 0, "Number of rows which can be processed before being committed to keeper", 0) \ - M(UInt32, s3queue_max_processed_bytes_before_commit, 0, "Number of bytes which can be processed before being committed to keeper", 0) \ - M(UInt32, s3queue_max_processing_time_sec_before_commit, 0, "Timeout in seconds after which to commit files committed to keeper", 0) \ - -#define LIST_OF_S3QUEUE_SETTINGS(M, ALIAS) \ - S3QUEUE_RELATED_SETTINGS(M, ALIAS) \ - LIST_OF_ALL_FORMAT_SETTINGS(M, ALIAS) - -DECLARE_SETTINGS_TRAITS(S3QueueSettingsTraits, LIST_OF_S3QUEUE_SETTINGS) - - -struct S3QueueSettings : public BaseSettings -{ - void loadFromQuery(ASTStorage & storage_def); -}; - -} diff --git a/src/Storages/SelectQueryInfo.h b/src/Storages/SelectQueryInfo.h index 65c90fc1e6c..848aeb59dad 100644 --- a/src/Storages/SelectQueryInfo.h +++ b/src/Storages/SelectQueryInfo.h @@ -166,7 +166,7 @@ struct SelectQueryInfo /// It's guaranteed to be present in JOIN TREE of `query_tree` QueryTreeNodePtr table_expression; - bool analyzer_can_use_parallel_replicas_on_follower = false; + bool current_table_chosen_for_reading_with_parallel_replicas = false; /// Table expression modifiers for storage std::optional table_expression_modifiers; @@ -229,8 +229,8 @@ struct SelectQueryInfo bool is_parameterized_view = false; bool optimize_trivial_count = false; - // If limit is not 0, that means it's a trivial limit query. - UInt64 limit = 0; + // If not 0, that means it's a trivial limit query. + UInt64 trivial_limit = 0; /// For IStorageSystemOneBlock std::vector columns_mask; diff --git a/src/Storages/SetSettings.cpp b/src/Storages/SetSettings.cpp index baa3d231067..4e6dd6a0519 100644 --- a/src/Storages/SetSettings.cpp +++ b/src/Storages/SetSettings.cpp @@ -3,6 +3,7 @@ #include #include #include +#include namespace DB diff --git a/src/Storages/Statistics/ConditionSelectivityEstimator.cpp b/src/Storages/Statistics/ConditionSelectivityEstimator.cpp index 757136fdf42..57dff958b9a 100644 --- a/src/Storages/Statistics/ConditionSelectivityEstimator.cpp +++ b/src/Storages/Statistics/ConditionSelectivityEstimator.cpp @@ -16,7 +16,7 @@ void ConditionSelectivityEstimator::ColumnSelectivityEstimator::merge(String par part_statistics[part_name] = stats; } -Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateLess(Float64 val, Float64 rows) const +Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateLess(const Field & val, Float64 rows) const { if (part_statistics.empty()) return default_normal_cond_factor * rows; @@ -30,16 +30,19 @@ Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateLess( return result * rows / part_rows; } -Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateGreater(Float64 val, Float64 rows) const +Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateGreater(const Field & val, Float64 rows) const { return rows - estimateLess(val, rows); } -Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateEqual(Float64 val, Float64 rows) const +Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateEqual(const Field & val, Float64 rows) const { if (part_statistics.empty()) { - if (val < - threshold || val > threshold) + auto float_val = StatisticsUtils::tryConvertToFloat64(val); + if (!float_val) + return default_unknown_cond_factor * rows; + else if (float_val.value() < - threshold || float_val.value() > threshold) return default_normal_cond_factor * rows; else return default_good_cond_factor * rows; @@ -87,7 +90,7 @@ static std::pair tryToExtractSingleColumn(const RPNBuilderTreeNod return result; } -std::pair ConditionSelectivityEstimator::extractBinaryOp(const RPNBuilderTreeNode & node, const String & column_name) const +std::pair ConditionSelectivityEstimator::extractBinaryOp(const RPNBuilderTreeNode & node, const String & column_name) const { if (!node.isFunction()) return {}; @@ -123,48 +126,35 @@ std::pair ConditionSelectivityEstimator::extractBinaryOp(const DataTypePtr output_type; if (!constant_node->tryGetConstant(output_value, output_type)) return {}; - - const auto type = output_value.getType(); - Float64 value; - if (type == Field::Types::Int64) - value = output_value.get(); - else if (type == Field::Types::UInt64) - value = output_value.get(); - else if (type == Field::Types::Float64) - value = output_value.get(); - else - return {}; - return std::make_pair(function_name, value); + return std::make_pair(function_name, output_value); } Float64 ConditionSelectivityEstimator::estimateRowCount(const RPNBuilderTreeNode & node) const { auto result = tryToExtractSingleColumn(node); if (result.second != 1) - { - return default_unknown_cond_factor; - } + return default_unknown_cond_factor * total_rows; + String col = result.first; auto it = column_estimators.find(col); /// If there the estimator of the column is not found or there are no data at all, /// we use dummy estimation. - bool dummy = total_rows == 0; + bool dummy = false; ColumnSelectivityEstimator estimator; if (it != column_estimators.end()) - { estimator = it->second; - } else - { dummy = true; - } + auto [op, val] = extractBinaryOp(node, col); + if (op == "equals") { if (dummy) { - if (val < - threshold || val > threshold) + auto float_val = StatisticsUtils::tryConvertToFloat64(val); + if (!float_val || (float_val < - threshold || float_val > threshold)) return default_normal_cond_factor * total_rows; else return default_good_cond_factor * total_rows; @@ -187,13 +177,8 @@ Float64 ConditionSelectivityEstimator::estimateRowCount(const RPNBuilderTreeNode return default_unknown_cond_factor * total_rows; } -void ConditionSelectivityEstimator::merge(String part_name, UInt64 part_rows, ColumnStatisticsPtr column_stat) +void ConditionSelectivityEstimator::merge(String part_name, ColumnStatisticsPtr column_stat) { - if (!part_names.contains(part_name)) - { - total_rows += part_rows; - part_names.insert(part_name); - } if (column_stat != nullptr) column_estimators[column_stat->columnName()].merge(part_name, column_stat); } diff --git a/src/Storages/Statistics/ConditionSelectivityEstimator.h b/src/Storages/Statistics/ConditionSelectivityEstimator.h index f0599742276..ce7fdd12e92 100644 --- a/src/Storages/Statistics/ConditionSelectivityEstimator.h +++ b/src/Storages/Statistics/ConditionSelectivityEstimator.h @@ -1,6 +1,7 @@ #pragma once #include +#include namespace DB { @@ -10,6 +11,14 @@ class RPNBuilderTreeNode; /// It estimates the selectivity of a condition. class ConditionSelectivityEstimator { +public: + /// TODO: Support the condition consists of CNF/DNF like (cond1 and cond2) or (cond3) ... + /// Right now we only support simple condition like col = val / col < val + Float64 estimateRowCount(const RPNBuilderTreeNode & node) const; + + void merge(String part_name, ColumnStatisticsPtr column_stat); + void addRows(UInt64 part_rows) { total_rows += part_rows; } + private: friend class ColumnStatistics; struct ColumnSelectivityEstimator @@ -20,13 +29,15 @@ private: void merge(String part_name, ColumnStatisticsPtr stats); - Float64 estimateLess(Float64 val, Float64 rows) const; + Float64 estimateLess(const Field & val, Float64 rows) const; - Float64 estimateGreater(Float64 val, Float64 rows) const; + Float64 estimateGreater(const Field & val, Float64 rows) const; - Float64 estimateEqual(Float64 val, Float64 rows) const; + Float64 estimateEqual(const Field & val, Float64 rows) const; }; + std::pair extractBinaryOp(const RPNBuilderTreeNode & node, const String & column_name) const; + static constexpr auto default_good_cond_factor = 0.1; static constexpr auto default_normal_cond_factor = 0.5; static constexpr auto default_unknown_cond_factor = 1.0; @@ -35,16 +46,7 @@ private: static constexpr auto threshold = 2; UInt64 total_rows = 0; - std::set part_names; std::map column_estimators; - std::pair extractBinaryOp(const RPNBuilderTreeNode & node, const String & column_name) const; - -public: - /// TODO: Support the condition consists of CNF/DNF like (cond1 and cond2) or (cond3) ... - /// Right now we only support simple condition like col = val / col < val - Float64 estimateRowCount(const RPNBuilderTreeNode & node) const; - - void merge(String part_name, UInt64 part_rows, ColumnStatisticsPtr column_stat); }; } diff --git a/src/Storages/Statistics/Statistics.cpp b/src/Storages/Statistics/Statistics.cpp index fed0bd61c03..ade3326288a 100644 --- a/src/Storages/Statistics/Statistics.cpp +++ b/src/Storages/Statistics/Statistics.cpp @@ -1,15 +1,17 @@ -#include -#include - #include -#include -#include -#include -#include -#include #include #include +#include +#include +#include +#include +#include +#include #include +#include + + +#include "config.h" /// USE_DATASKETCHES namespace DB { @@ -20,70 +22,130 @@ namespace ErrorCodes extern const int INCORRECT_QUERY; } -/// Version / bitmask of statistics / data of statistics / enum StatisticsFileVersion : UInt16 { V0 = 0, }; -IStatistics::IStatistics(const SingleStatisticsDescription & stat_) : stat(stat_) {} +std::optional StatisticsUtils::tryConvertToFloat64(const Field & field) +{ + switch (field.getType()) + { + case Field::Types::Int64: + return field.get(); + case Field::Types::UInt64: + return field.get(); + case Field::Types::Float64: + return field.get(); + case Field::Types::Int128: + return field.get(); + case Field::Types::UInt128: + return field.get(); + case Field::Types::Int256: + return field.get(); + case Field::Types::UInt256: + return field.get(); + default: + return {}; + } +} + +std::optional StatisticsUtils::tryConvertToString(const DB::Field & field) +{ + if (field.getType() == Field::Types::String) + return field.get(); + return {}; +} + +IStatistics::IStatistics(const SingleStatisticsDescription & stat_) + : stat(stat_) +{ +} ColumnStatistics::ColumnStatistics(const ColumnStatisticsDescription & stats_desc_) - : stats_desc(stats_desc_), rows(0) + : stats_desc(stats_desc_) { } void ColumnStatistics::update(const ColumnPtr & column) { rows += column->size(); - for (const auto & iter : stats) - { - iter.second->update(column); - } + for (const auto & stat : stats) + stat.second->update(column); } -Float64 ColumnStatistics::estimateLess(Float64 val) const +UInt64 IStatistics::estimateCardinality() const +{ + throw Exception(ErrorCodes::LOGICAL_ERROR, "Cardinality estimation is not implemented for this type of statistics"); +} + +Float64 IStatistics::estimateEqual(const Field & /*val*/) const +{ + throw Exception(ErrorCodes::LOGICAL_ERROR, "Equality estimation is not implemented for this type of statistics"); +} + +Float64 IStatistics::estimateLess(const Field & /*val*/) const +{ + throw Exception(ErrorCodes::LOGICAL_ERROR, "Less-than estimation is not implemented for this type of statistics"); +} + +/// ------------------------------------- +/// Implementation of the estimation: +/// Note: Each statistics object supports certain types predicates natively, e.g. +/// - TDigest: '< X' (less-than predicates) +/// - Count-min sketches: '= X' (equal predicates) +/// - Uniq (HyperLogLog): 'count distinct(*)' (column cardinality) +/// If multiple statistics objects are available per column, it is sometimes also possible to combine them in a clever way. +/// For that reason, all estimation are performed in a central place (here), and we don't simply pass the predicate to the first statistics +/// object that supports it natively. + +Float64 ColumnStatistics::estimateLess(const Field & val) const { if (stats.contains(StatisticsType::TDigest)) - return std::static_pointer_cast(stats.at(StatisticsType::TDigest))->estimateLess(val); + return stats.at(StatisticsType::TDigest)->estimateLess(val); return rows * ConditionSelectivityEstimator::default_normal_cond_factor; } -Float64 ColumnStatistics::estimateGreater(Float64 val) const +Float64 ColumnStatistics::estimateGreater(const Field & val) const { return rows - estimateLess(val); } -Float64 ColumnStatistics::estimateEqual(Float64 val) const +Float64 ColumnStatistics::estimateEqual(const Field & val) const { - if (stats.contains(StatisticsType::Uniq) && stats.contains(StatisticsType::TDigest)) + auto float_val = StatisticsUtils::tryConvertToFloat64(val); + if (float_val.has_value() && stats.contains(StatisticsType::Uniq) && stats.contains(StatisticsType::TDigest)) { - auto uniq_static = std::static_pointer_cast(stats.at(StatisticsType::Uniq)); - /// 2048 is the default number of buckets in TDigest. In this case, TDigest stores exactly one value (with many rows) - /// for every bucket. - if (uniq_static->getCardinality() < 2048) - { - auto tdigest_static = std::static_pointer_cast(stats.at(StatisticsType::TDigest)); - return tdigest_static->estimateEqual(val); - } + /// 2048 is the default number of buckets in TDigest. In this case, TDigest stores exactly one value (with many rows) for every bucket. + if (stats.at(StatisticsType::Uniq)->estimateCardinality() < 2048) + return stats.at(StatisticsType::TDigest)->estimateEqual(val); } - if (val < - ConditionSelectivityEstimator::threshold || val > ConditionSelectivityEstimator::threshold) +#if USE_DATASKETCHES + if (stats.contains(StatisticsType::CountMinSketch)) + return stats.at(StatisticsType::CountMinSketch)->estimateEqual(val); +#endif + if (!float_val.has_value() && (float_val < - ConditionSelectivityEstimator::threshold || float_val > ConditionSelectivityEstimator::threshold)) return rows * ConditionSelectivityEstimator::default_normal_cond_factor; else return rows * ConditionSelectivityEstimator::default_good_cond_factor; } +/// ------------------------------------- + void ColumnStatistics::serialize(WriteBuffer & buf) { writeIntBinary(V0, buf); + UInt64 stat_types_mask = 0; for (const auto & [type, _]: stats) stat_types_mask |= 1 << UInt8(type); writeIntBinary(stat_types_mask, buf); - /// We write some basic statistics + + /// as the column row count is always useful, save it in any case writeIntBinary(rows, buf); - /// We write complex statistics - for (const auto & [type, stat_ptr]: stats) + + /// write the actual statistics object + for (const auto & [type, stat_ptr] : stats) stat_ptr->serialize(buf); } @@ -96,7 +158,9 @@ void ColumnStatistics::deserialize(ReadBuffer &buf) UInt64 stat_types_mask = 0; readIntBinary(stat_types_mask, buf); + readIntBinary(rows, buf); + for (auto it = stats.begin(); it != stats.end();) { if (!(stat_types_mask & 1 << UInt8(it->first))) @@ -136,15 +200,20 @@ void MergeTreeStatisticsFactory::registerValidator(StatisticsType stats_type, Va { if (!validators.emplace(stats_type, std::move(validator)).second) throw Exception(ErrorCodes::LOGICAL_ERROR, "MergeTreeStatisticsFactory: the statistics validator type {} is not unique", stats_type); - } MergeTreeStatisticsFactory::MergeTreeStatisticsFactory() { - registerCreator(StatisticsType::TDigest, TDigestCreator); - registerCreator(StatisticsType::Uniq, UniqCreator); - registerValidator(StatisticsType::TDigest, TDigestValidator); - registerValidator(StatisticsType::Uniq, UniqValidator); + registerValidator(StatisticsType::TDigest, tdigestValidator); + registerCreator(StatisticsType::TDigest, tdigestCreator); + + registerValidator(StatisticsType::Uniq, uniqValidator); + registerCreator(StatisticsType::Uniq, uniqCreator); + +#if USE_DATASKETCHES + registerValidator(StatisticsType::CountMinSketch, countMinSketchValidator); + registerCreator(StatisticsType::CountMinSketch, countMinSketchCreator); +#endif } MergeTreeStatisticsFactory & MergeTreeStatisticsFactory::instance() @@ -159,9 +228,7 @@ void MergeTreeStatisticsFactory::validate(const ColumnStatisticsDescription & st { auto it = validators.find(type); if (it == validators.end()) - { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown Statistic type '{}'", type); - } + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown statistic type '{}'", type); it->second(desc, data_type); } } @@ -173,10 +240,7 @@ ColumnStatisticsPtr MergeTreeStatisticsFactory::get(const ColumnStatisticsDescri { auto it = creators.find(type); if (it == creators.end()) - { - throw Exception(ErrorCodes::INCORRECT_QUERY, - "Unknown Statistic type '{}'. Available types: tdigest, uniq", type); - } + throw Exception(ErrorCodes::INCORRECT_QUERY, "Unknown statistic type '{}'. Available types: 'tdigest' 'uniq' and 'count_min'", type); auto stat_ptr = (it->second)(desc, stats.data_type); column_stat->stats[type] = stat_ptr; } diff --git a/src/Storages/Statistics/Statistics.h b/src/Storages/Statistics/Statistics.h index 2ab1337af02..16f0c67eabd 100644 --- a/src/Storages/Statistics/Statistics.h +++ b/src/Storages/Statistics/Statistics.h @@ -1,22 +1,27 @@ #pragma once -#include -#include - #include -#include +#include #include #include #include +#include namespace DB { -/// this is for user-defined statistic. constexpr auto STATS_FILE_PREFIX = "statistics_"; constexpr auto STATS_FILE_SUFFIX = ".stats"; + +struct StatisticsUtils +{ + /// Returns std::nullopt if input Field cannot be converted to a concrete value + static std::optional tryConvertToFloat64(const Field & field); + static std::optional tryConvertToString(const Field & field); +}; + /// Statistics describe properties of the values in the column, /// e.g. how many unique values exist, /// what are the N most frequent values, @@ -25,14 +30,21 @@ class IStatistics { public: explicit IStatistics(const SingleStatisticsDescription & stat_); - virtual ~IStatistics() = default; - virtual void serialize(WriteBuffer & buf) = 0; + virtual void update(const ColumnPtr & column) = 0; + virtual void serialize(WriteBuffer & buf) = 0; virtual void deserialize(ReadBuffer & buf) = 0; - virtual void update(const ColumnPtr & column) = 0; + /// Estimate the cardinality of the column. + /// Throws if the statistics object is not able to do a meaningful estimation. + virtual UInt64 estimateCardinality() const; + + /// Per-value estimations. + /// Throws if the statistics object is not able to do a meaningful estimation. + virtual Float64 estimateEqual(const Field & val) const; /// cardinality of val in the column + virtual Float64 estimateLess(const Field & val) const; /// summarized cardinality of values < val in the column protected: SingleStatisticsDescription stat; @@ -43,29 +55,27 @@ using StatisticsPtr = std::shared_ptr; class ColumnStatistics { public: - explicit ColumnStatistics(const ColumnStatisticsDescription & stats_); + explicit ColumnStatistics(const ColumnStatisticsDescription & stats_desc_); + void serialize(WriteBuffer & buf); void deserialize(ReadBuffer & buf); - String getFileName() const; + String getFileName() const; const String & columnName() const; UInt64 rowCount() const; void update(const ColumnPtr & column); - Float64 estimateLess(Float64 val) const; - - Float64 estimateGreater(Float64 val) const; - - Float64 estimateEqual(Float64 val) const; + Float64 estimateLess(const Field & val) const; + Float64 estimateGreater(const Field & val) const; + Float64 estimateEqual(const Field & val) const; private: - friend class MergeTreeStatisticsFactory; ColumnStatisticsDescription stats_desc; std::map stats; - UInt64 rows; /// the number of rows of the column + UInt64 rows = 0; /// the number of rows in the column }; class ColumnsDescription; @@ -79,25 +89,23 @@ public: void validate(const ColumnStatisticsDescription & stats, DataTypePtr data_type) const; + using Validator = std::function; using Creator = std::function; - using Validator = std::function; - ColumnStatisticsPtr get(const ColumnStatisticsDescription & stats) const; - ColumnsStatistics getMany(const ColumnsDescription & columns) const; - void registerCreator(StatisticsType type, Creator creator); void registerValidator(StatisticsType type, Validator validator); + void registerCreator(StatisticsType type, Creator creator); protected: MergeTreeStatisticsFactory(); private: - using Creators = std::unordered_map; using Validators = std::unordered_map; - Creators creators; + using Creators = std::unordered_map; Validators validators; + Creators creators; }; } diff --git a/src/Storages/Statistics/StatisticsCountMinSketch.cpp b/src/Storages/Statistics/StatisticsCountMinSketch.cpp new file mode 100644 index 00000000000..e69bbc1515b --- /dev/null +++ b/src/Storages/Statistics/StatisticsCountMinSketch.cpp @@ -0,0 +1,102 @@ + +#include +#include +#include +#include +#include +#include + +#if USE_DATASKETCHES + +namespace DB +{ + +namespace ErrorCodes +{ +extern const int LOGICAL_ERROR; +extern const int ILLEGAL_STATISTICS; +} + +/// Constants chosen based on rolling dices. +/// The values provides: +/// 1. an error tolerance of 0.1% (ε = 0.001) +/// 2. a confidence level of 99.9% (δ = 0.001). +/// And sketch the size is 152kb. +static constexpr auto num_hashes = 7uz; +static constexpr auto num_buckets = 2718uz; + +StatisticsCountMinSketch::StatisticsCountMinSketch(const SingleStatisticsDescription & stat_, DataTypePtr data_type_) + : IStatistics(stat_) + , sketch(num_hashes, num_buckets) + , data_type(data_type_) +{ +} + +Float64 StatisticsCountMinSketch::estimateEqual(const Field & val) const +{ + /// Try to convert field to data_type. Converting string to proper data types such as: number, date, datetime, IPv4, Decimal etc. + /// Return null if val larger than the range of data_type + /// + /// For example: if data_type is Int32: + /// 1. For 1.0, 1, '1', return Field(1) + /// 2. For 1.1, max_value_int64, return null + Field val_converted = convertFieldToType(val, *data_type); + if (val_converted.isNull()) + return 0; + + if (data_type->isValueRepresentedByNumber()) + return sketch.get_estimate(&val_converted, data_type->getSizeOfValueInMemory()); + + if (isStringOrFixedString(data_type)) + return sketch.get_estimate(val.get()); + + throw Exception(ErrorCodes::LOGICAL_ERROR, "Statistics 'count_min' does not support estimate data type of {}", data_type->getName()); +} + +void StatisticsCountMinSketch::update(const ColumnPtr & column) +{ + for (size_t row = 0; row < column->size(); ++row) + { + if (column->isNullAt(row)) + continue; + auto data = column->getDataAt(row); + sketch.update(data.data, data.size, 1); + } +} + +void StatisticsCountMinSketch::serialize(WriteBuffer & buf) +{ + Sketch::vector_bytes bytes = sketch.serialize(); + writeIntBinary(static_cast(bytes.size()), buf); + buf.write(reinterpret_cast(bytes.data()), bytes.size()); +} + +void StatisticsCountMinSketch::deserialize(ReadBuffer & buf) +{ + UInt64 size; + readIntBinary(size, buf); + + Sketch::vector_bytes bytes; + bytes.resize(size); /// To avoid 'container-overflow' in AddressSanitizer checking + buf.readStrict(reinterpret_cast(bytes.data()), size); + + sketch = Sketch::deserialize(bytes.data(), size); +} + + +void countMinSketchValidator(const SingleStatisticsDescription &, DataTypePtr data_type) +{ + data_type = removeNullable(data_type); + data_type = removeLowCardinalityAndNullable(data_type); + if (!data_type->isValueRepresentedByNumber() && !isStringOrFixedString(data_type)) + throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Statistics of type 'count_min' does not support type {}", data_type->getName()); +} + +StatisticsPtr countMinSketchCreator(const SingleStatisticsDescription & stat, DataTypePtr data_type) +{ + return std::make_shared(stat, data_type); +} + +} + +#endif diff --git a/src/Storages/Statistics/StatisticsCountMinSketch.h b/src/Storages/Statistics/StatisticsCountMinSketch.h new file mode 100644 index 00000000000..6c8b74f8c35 --- /dev/null +++ b/src/Storages/Statistics/StatisticsCountMinSketch.h @@ -0,0 +1,39 @@ +#pragma once + +#include + +#include "config.h" + +#if USE_DATASKETCHES + +#include + +namespace DB +{ + +class StatisticsCountMinSketch : public IStatistics +{ +public: + StatisticsCountMinSketch(const SingleStatisticsDescription & stat_, DataTypePtr data_type_); + + Float64 estimateEqual(const Field & val) const override; + + void update(const ColumnPtr & column) override; + + void serialize(WriteBuffer & buf) override; + void deserialize(ReadBuffer & buf) override; + +private: + using Sketch = datasketches::count_min_sketch; + Sketch sketch; + + DataTypePtr data_type; +}; + + +void countMinSketchValidator(const SingleStatisticsDescription &, DataTypePtr data_type); +StatisticsPtr countMinSketchCreator(const SingleStatisticsDescription & stat, DataTypePtr); + +} + +#endif diff --git a/src/Storages/Statistics/StatisticsTDigest.cpp b/src/Storages/Statistics/StatisticsTDigest.cpp new file mode 100644 index 00000000000..66150e00fdb --- /dev/null +++ b/src/Storages/Statistics/StatisticsTDigest.cpp @@ -0,0 +1,73 @@ +#include +#include +#include + +namespace DB +{ +namespace ErrorCodes +{ +extern const int ILLEGAL_STATISTICS; +extern const int LOGICAL_ERROR; +} + +StatisticsTDigest::StatisticsTDigest(const SingleStatisticsDescription & stat_) + : IStatistics(stat_) +{ +} + +void StatisticsTDigest::update(const ColumnPtr & column) +{ + size_t rows = column->size(); + for (size_t row = 0; row < rows; ++row) + { + Field field; + column->get(row, field); + + if (field.isNull()) + continue; + + if (auto field_as_float = StatisticsUtils::tryConvertToFloat64(field)) + t_digest.add(*field_as_float, 1); + } +} + +void StatisticsTDigest::serialize(WriteBuffer & buf) +{ + t_digest.serialize(buf); +} + +void StatisticsTDigest::deserialize(ReadBuffer & buf) +{ + t_digest.deserialize(buf); +} + +Float64 StatisticsTDigest::estimateLess(const Field & val) const +{ + auto val_as_float = StatisticsUtils::tryConvertToFloat64(val); + if (val_as_float) + return t_digest.getCountLessThan(*val_as_float); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Statistics 'tdigest' does not support estimating value of type {}", val.getTypeName()); +} + +Float64 StatisticsTDigest::estimateEqual(const Field & val) const +{ + auto val_as_float = StatisticsUtils::tryConvertToFloat64(val); + if (val_as_float) + return t_digest.getCountEqual(*val_as_float); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Statistics 'tdigest' does not support estimating value of type {}", val.getTypeName()); +} + +void tdigestValidator(const SingleStatisticsDescription &, DataTypePtr data_type) +{ + data_type = removeNullable(data_type); + data_type = removeLowCardinalityAndNullable(data_type); + if (!data_type->isValueRepresentedByNumber()) + throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Statistics of type 'tdigest' do not support type {}", data_type->getName()); +} + +StatisticsPtr tdigestCreator(const SingleStatisticsDescription & stat, DataTypePtr) +{ + return std::make_shared(stat); +} + +} diff --git a/src/Storages/Statistics/StatisticsTDigest.h b/src/Storages/Statistics/StatisticsTDigest.h new file mode 100644 index 00000000000..614973e5d8b --- /dev/null +++ b/src/Storages/Statistics/StatisticsTDigest.h @@ -0,0 +1,29 @@ +#pragma once + +#include +#include + +namespace DB +{ + +class StatisticsTDigest : public IStatistics +{ +public: + explicit StatisticsTDigest(const SingleStatisticsDescription & stat_); + + void update(const ColumnPtr & column) override; + + void serialize(WriteBuffer & buf) override; + void deserialize(ReadBuffer & buf) override; + + Float64 estimateLess(const Field & val) const override; + Float64 estimateEqual(const Field & val) const override; + +private: + QuantileTDigest t_digest; +}; + +void tdigestValidator(const SingleStatisticsDescription &, DataTypePtr data_type); +StatisticsPtr tdigestCreator(const SingleStatisticsDescription & stat, DataTypePtr); + +} diff --git a/src/Storages/Statistics/UniqStatistics.cpp b/src/Storages/Statistics/StatisticsUniq.cpp similarity index 66% rename from src/Storages/Statistics/UniqStatistics.cpp rename to src/Storages/Statistics/StatisticsUniq.cpp index fc748e769ca..8f60ffcf0b5 100644 --- a/src/Storages/Statistics/UniqStatistics.cpp +++ b/src/Storages/Statistics/StatisticsUniq.cpp @@ -1,6 +1,7 @@ -#include +#include #include #include +#include namespace DB { @@ -10,7 +11,7 @@ namespace ErrorCodes extern const int ILLEGAL_STATISTICS; } -UniqStatistics::UniqStatistics(const SingleStatisticsDescription & stat_, const DataTypePtr & data_type) +StatisticsUniq::StatisticsUniq(const SingleStatisticsDescription & stat_, const DataTypePtr & data_type) : IStatistics(stat_) { arena = std::make_unique(); @@ -20,29 +21,12 @@ UniqStatistics::UniqStatistics(const SingleStatisticsDescription & stat_, const collector->create(data); } -UniqStatistics::~UniqStatistics() +StatisticsUniq::~StatisticsUniq() { collector->destroy(data); } -UInt64 UniqStatistics::getCardinality() -{ - auto column = DataTypeUInt64().createColumn(); - collector->insertResultInto(data, *column, nullptr); - return column->getUInt(0); -} - -void UniqStatistics::serialize(WriteBuffer & buf) -{ - collector->serialize(data, buf); -} - -void UniqStatistics::deserialize(ReadBuffer & buf) -{ - collector->deserialize(data, buf); -} - -void UniqStatistics::update(const ColumnPtr & column) +void StatisticsUniq::update(const ColumnPtr & column) { /// TODO(hanfei): For low cardinality, it's very slow to convert to full column. We can read the dictionary directly. /// Here we intend to avoid crash in CI. @@ -51,16 +35,34 @@ void UniqStatistics::update(const ColumnPtr & column) collector->addBatchSinglePlace(0, column->size(), data, &(raw_ptr), nullptr); } -void UniqValidator(const SingleStatisticsDescription &, DataTypePtr data_type) +void StatisticsUniq::serialize(WriteBuffer & buf) +{ + collector->serialize(data, buf); +} + +void StatisticsUniq::deserialize(ReadBuffer & buf) +{ + collector->deserialize(data, buf); +} + +UInt64 StatisticsUniq::estimateCardinality() const +{ + auto column = DataTypeUInt64().createColumn(); + collector->insertResultInto(data, *column, nullptr); + return column->getUInt(0); +} + +void uniqValidator(const SingleStatisticsDescription &, DataTypePtr data_type) { data_type = removeNullable(data_type); + data_type = removeLowCardinalityAndNullable(data_type); if (!data_type->isValueRepresentedByNumber()) - throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Statistics of type 'uniq' does not support type {}", data_type->getName()); + throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Statistics of type 'uniq' do not support type {}", data_type->getName()); } -StatisticsPtr UniqCreator(const SingleStatisticsDescription & stat, DataTypePtr data_type) +StatisticsPtr uniqCreator(const SingleStatisticsDescription & stat, DataTypePtr data_type) { - return std::make_shared(stat, data_type); + return std::make_shared(stat, data_type); } } diff --git a/src/Storages/Statistics/UniqStatistics.h b/src/Storages/Statistics/StatisticsUniq.h similarity index 61% rename from src/Storages/Statistics/UniqStatistics.h rename to src/Storages/Statistics/StatisticsUniq.h index 0d86a6e458a..faabde8d47c 100644 --- a/src/Storages/Statistics/UniqStatistics.h +++ b/src/Storages/Statistics/StatisticsUniq.h @@ -7,30 +7,27 @@ namespace DB { -class UniqStatistics : public IStatistics +class StatisticsUniq : public IStatistics { public: - UniqStatistics(const SingleStatisticsDescription & stat_, const DataTypePtr & data_type); - - ~UniqStatistics() override; - - UInt64 getCardinality(); - - void serialize(WriteBuffer & buf) override; - - void deserialize(ReadBuffer & buf) override; + StatisticsUniq(const SingleStatisticsDescription & stat_, const DataTypePtr & data_type); + ~StatisticsUniq() override; void update(const ColumnPtr & column) override; -private: + void serialize(WriteBuffer & buf) override; + void deserialize(ReadBuffer & buf) override; + UInt64 estimateCardinality() const override; + +private: std::unique_ptr arena; AggregateFunctionPtr collector; AggregateDataPtr data; }; -StatisticsPtr UniqCreator(const SingleStatisticsDescription & stat, DataTypePtr data_type); -void UniqValidator(const SingleStatisticsDescription &, DataTypePtr data_type); +void uniqValidator(const SingleStatisticsDescription &, DataTypePtr data_type); +StatisticsPtr uniqCreator(const SingleStatisticsDescription & stat, DataTypePtr data_type); } diff --git a/src/Storages/Statistics/TDigestStatistics.cpp b/src/Storages/Statistics/TDigestStatistics.cpp deleted file mode 100644 index aa5662c979d..00000000000 --- a/src/Storages/Statistics/TDigestStatistics.cpp +++ /dev/null @@ -1,60 +0,0 @@ -#include -#include - -namespace DB -{ -namespace ErrorCodes -{ - extern const int ILLEGAL_STATISTICS; -} - -TDigestStatistics::TDigestStatistics(const SingleStatisticsDescription & stat_): - IStatistics(stat_) -{ -} - -Float64 TDigestStatistics::estimateLess(Float64 val) const -{ - return data.getCountLessThan(val); -} - -Float64 TDigestStatistics::estimateEqual(Float64 val) const -{ - return data.getCountEqual(val); -} - -void TDigestStatistics::serialize(WriteBuffer & buf) -{ - data.serialize(buf); -} - -void TDigestStatistics::deserialize(ReadBuffer & buf) -{ - data.deserialize(buf); -} - -void TDigestStatistics::update(const ColumnPtr & column) -{ - size_t size = column->size(); - - for (size_t i = 0; i < size; ++i) - { - /// TODO: support more types. - Float64 value = column->getFloat64(i); - data.add(value, 1); - } -} - -StatisticsPtr TDigestCreator(const SingleStatisticsDescription & stat, DataTypePtr) -{ - return std::make_shared(stat); -} - -void TDigestValidator(const SingleStatisticsDescription &, DataTypePtr data_type) -{ - data_type = removeNullable(data_type); - if (!data_type->isValueRepresentedByNumber()) - throw Exception(ErrorCodes::ILLEGAL_STATISTICS, "Statistics of type 'tdigest' does not support type {}", data_type->getName()); -} - -} diff --git a/src/Storages/Statistics/TDigestStatistics.h b/src/Storages/Statistics/TDigestStatistics.h deleted file mode 100644 index 7c361b8751f..00000000000 --- a/src/Storages/Statistics/TDigestStatistics.h +++ /dev/null @@ -1,32 +0,0 @@ -#pragma once - -#include -#include - -namespace DB -{ - - -/// TDigestStatistic is a kind of histogram. -class TDigestStatistics : public IStatistics -{ -public: - explicit TDigestStatistics(const SingleStatisticsDescription & stat_); - - Float64 estimateLess(Float64 val) const; - - Float64 estimateEqual(Float64 val) const; - - void serialize(WriteBuffer & buf) override; - - void deserialize(ReadBuffer & buf) override; - - void update(const ColumnPtr & column) override; -private: - QuantileTDigest data; -}; - -StatisticsPtr TDigestCreator(const SingleStatisticsDescription & stat, DataTypePtr); -void TDigestValidator(const SingleStatisticsDescription &, DataTypePtr data_type); - -} diff --git a/src/Storages/Statistics/tests/gtest_stats.cpp b/src/Storages/Statistics/tests/gtest_stats.cpp index f94f310be56..e55c52c49f3 100644 --- a/src/Storages/Statistics/tests/gtest_stats.cpp +++ b/src/Storages/Statistics/tests/gtest_stats.cpp @@ -1,6 +1,10 @@ #include -#include +#include +#include +#include + +using namespace DB; TEST(Statistics, TDigestLessThan) { @@ -39,6 +43,4 @@ TEST(Statistics, TDigestLessThan) std::reverse(data.begin(), data.end()); test_less_than(data, {-1, 1e9, 50000.0, 3000.0, 30.0}, {0, 100000, 50000, 3000, 30}, {0, 0, 0.001, 0.001, 0.001}); - - } diff --git a/src/Storages/StatisticsDescription.cpp b/src/Storages/StatisticsDescription.cpp index f10fb78f933..9c5fd3604b2 100644 --- a/src/Storages/StatisticsDescription.cpp +++ b/src/Storages/StatisticsDescription.cpp @@ -1,19 +1,14 @@ #include -#include #include #include #include #include -#include -#include #include #include #include -#include #include -#include namespace DB { @@ -54,7 +49,9 @@ static StatisticsType stringToStatisticsType(String type) return StatisticsType::TDigest; if (type == "uniq") return StatisticsType::Uniq; - throw Exception(ErrorCodes::INCORRECT_QUERY, "Unknown statistics type: {}. Supported statistics types are `tdigest` and `uniq`.", type); + if (type == "count_min") + return StatisticsType::CountMinSketch; + throw Exception(ErrorCodes::INCORRECT_QUERY, "Unknown statistics type: {}. Supported statistics types are 'tdigest', 'uniq' and 'count_min'.", type); } String SingleStatisticsDescription::getTypeName() const @@ -65,8 +62,10 @@ String SingleStatisticsDescription::getTypeName() const return "TDigest"; case StatisticsType::Uniq: return "Uniq"; + case StatisticsType::CountMinSketch: + return "count_min"; default: - throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown statistics type: {}. Supported statistics types are `tdigest` and `uniq`.", type); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown statistics type: {}. Supported statistics types are 'tdigest', 'uniq' and 'count_min'.", type); } } @@ -99,10 +98,9 @@ void ColumnStatisticsDescription::merge(const ColumnStatisticsDescription & othe chassert(merging_column_type); if (column_name.empty()) - { column_name = merging_column_name; - data_type = merging_column_type; - } + + data_type = merging_column_type; for (const auto & [stats_type, stats_desc]: other.types_to_desc) { @@ -121,6 +119,7 @@ void ColumnStatisticsDescription::assign(const ColumnStatisticsDescription & oth throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot assign statistics from column {} to {}", column_name, other.column_name); types_to_desc = other.types_to_desc; + data_type = other.data_type; } void ColumnStatisticsDescription::clear() @@ -159,6 +158,7 @@ std::vector ColumnStatisticsDescription::fromAST(co const auto & column = columns.getPhysical(physical_column_name); stats.column_name = column.name; + stats.data_type = column.type; stats.types_to_desc = statistics_types; result.push_back(stats); } diff --git a/src/Storages/StatisticsDescription.h b/src/Storages/StatisticsDescription.h index 4862fb79d45..03b8fb0d583 100644 --- a/src/Storages/StatisticsDescription.h +++ b/src/Storages/StatisticsDescription.h @@ -13,6 +13,7 @@ enum class StatisticsType : UInt8 { TDigest = 0, Uniq = 1, + CountMinSketch = 2, Max = 63, }; diff --git a/src/Storages/StorageBuffer.cpp b/src/Storages/StorageBuffer.cpp index a3f6b6afc5d..1bcc602f168 100644 --- a/src/Storages/StorageBuffer.cpp +++ b/src/Storages/StorageBuffer.cpp @@ -38,6 +38,7 @@ #include #include #include +#include namespace ProfileEvents @@ -607,7 +608,7 @@ public: String getName() const override { return "BufferSink"; } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { size_t rows = chunk.getNumRows(); if (!rows) @@ -1020,7 +1021,13 @@ void StorageBuffer::writeBlockToDestination(const Block & block, StoragePtr tabl auto insert_context = Context::createCopy(getContext()); insert_context->makeQueryContext(); - InterpreterInsertQuery interpreter{insert, insert_context, allow_materialized}; + InterpreterInsertQuery interpreter( + insert, + insert_context, + allow_materialized, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); auto block_io = interpreter.execute(); PushingPipelineExecutor executor(block_io.pipeline); diff --git a/src/Storages/StorageDictionary.cpp b/src/Storages/StorageDictionary.cpp index 447fd87cdc9..b43cc4fa426 100644 --- a/src/Storages/StorageDictionary.cpp +++ b/src/Storages/StorageDictionary.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -9,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -25,13 +27,14 @@ namespace ErrorCodes extern const int CANNOT_DETACH_DICTIONARY_AS_TABLE; extern const int DICTIONARY_ALREADY_EXISTS; extern const int NOT_IMPLEMENTED; + extern const int BAD_ARGUMENTS; } namespace { void checkNamesAndTypesCompatibleWithDictionary(const String & dictionary_name, const ColumnsDescription & columns, const DictionaryStructure & dictionary_structure) { - auto dictionary_names_and_types = StorageDictionary::getNamesAndTypes(dictionary_structure); + auto dictionary_names_and_types = StorageDictionary::getNamesAndTypes(dictionary_structure, false); std::set names_and_types_set(dictionary_names_and_types.begin(), dictionary_names_and_types.end()); for (const auto & column : columns.getOrdinary()) @@ -47,13 +50,17 @@ namespace } -NamesAndTypesList StorageDictionary::getNamesAndTypes(const DictionaryStructure & dictionary_structure) +NamesAndTypesList StorageDictionary::getNamesAndTypes(const DictionaryStructure & dictionary_structure, bool validate_id_type) { NamesAndTypesList dictionary_names_and_types; if (dictionary_structure.id) - dictionary_names_and_types.emplace_back(dictionary_structure.id->name, std::make_shared()); + { + if (validate_id_type && dictionary_structure.id->type->getTypeId() != TypeIndex::UInt64) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Incorrect type of ID column: must be UInt64, but it is {}", dictionary_structure.id->type->getFamilyName()); + dictionary_names_and_types.emplace_back(dictionary_structure.id->name, std::make_shared()); + } /// In old-style (XML) configuration we don't have this attributes in the /// main attribute list, so we have to add them to columns list explicitly. /// In the new configuration (DDL) we have them both in range_* nodes and @@ -105,7 +112,7 @@ StorageDictionary::StorageDictionary( Location location_, ContextPtr context_) : StorageDictionary( - table_id_, dictionary_name_, ColumnsDescription{getNamesAndTypes(dictionary_structure_)}, comment, location_, context_) + table_id_, dictionary_name_, ColumnsDescription{getNamesAndTypes(dictionary_structure_, context_->getSettingsRef().dictionary_validate_primary_key_type)}, comment, location_, context_) { } @@ -162,6 +169,7 @@ Pipe StorageDictionary::read( { auto registered_dictionary_name = location == Location::SameDatabaseAndNameAsDictionary ? getStorageID().getInternalDictionaryName() : dictionary_name; auto dictionary = getContext()->getExternalDictionariesLoader().getDictionary(registered_dictionary_name, local_context); + local_context->checkAccess(AccessType::dictGet, dictionary->getDatabaseOrNoDatabaseTag(), dictionary->getDictionaryID().getTableName()); return dictionary->read(column_names, max_block_size, threads); } diff --git a/src/Storages/StorageDictionary.h b/src/Storages/StorageDictionary.h index 17e4efda2cd..44a274cf97c 100644 --- a/src/Storages/StorageDictionary.h +++ b/src/Storages/StorageDictionary.h @@ -80,7 +80,7 @@ public: std::shared_ptr getDictionary() const; - static NamesAndTypesList getNamesAndTypes(const DictionaryStructure & dictionary_structure); + static NamesAndTypesList getNamesAndTypes(const DictionaryStructure & dictionary_structure, bool validate_id_type); bool isDictionary() const override { return true; } void shutdown(bool is_drop) override; diff --git a/src/Storages/StorageDistributed.cpp b/src/Storages/StorageDistributed.cpp index 849fa5dbe0b..07892971ec2 100644 --- a/src/Storages/StorageDistributed.cpp +++ b/src/Storages/StorageDistributed.cpp @@ -43,7 +43,6 @@ #include #include -#include #include #include #include @@ -61,26 +60,20 @@ #include #include #include -#include #include #include #include #include #include -#include #include #include #include #include #include #include -#include #include -#include #include -#include -#include #include #include @@ -90,7 +83,6 @@ #include #include #include -#include #include #include #include @@ -426,7 +418,7 @@ QueryProcessingStage::Enum StorageDistributed::getQueryProcessingStage( query_info.cluster = cluster; - if (!local_context->canUseParallelReplicasCustomKey(*cluster)) + if (!local_context->canUseParallelReplicasCustomKeyForCluster(*cluster)) { if (nodes > 1 && settings.optimize_skip_unused_shards) { @@ -496,7 +488,7 @@ QueryProcessingStage::Enum StorageDistributed::getQueryProcessingStage( } std::optional optimized_stage; - if (settings.allow_experimental_analyzer) + if (query_info.query_tree) optimized_stage = getOptimizedQueryProcessingStageAnalyzer(query_info, settings); else optimized_stage = getOptimizedQueryProcessingStage(query_info, settings); @@ -839,7 +831,9 @@ void StorageDistributed::read( SelectQueryInfo modified_query_info = query_info; - if (local_context->getSettingsRef().allow_experimental_analyzer) + const auto & settings = local_context->getSettingsRef(); + + if (settings.allow_experimental_analyzer) { StorageID remote_storage_id = StorageID::createEmpty(); if (!remote_table_function_ptr) @@ -858,31 +852,28 @@ void StorageDistributed::read( modified_query_info.query = queryNodeToDistributedSelectQuery(query_tree_distributed); modified_query_info.query_tree = std::move(query_tree_distributed); + + /// Return directly (with correct header) if no shard to query. + if (modified_query_info.getCluster()->getShardsInfo().empty()) + return; } else { header = InterpreterSelectQuery(modified_query_info.query, local_context, SelectQueryOptions(processed_stage).analyze()).getSampleBlock(); - } - if (!local_context->getSettingsRef().allow_experimental_analyzer) - { modified_query_info.query = ClusterProxy::rewriteSelectQuery( local_context, modified_query_info.query, remote_database, remote_table, remote_table_function_ptr); - } - /// Return directly (with correct header) if no shard to query. - if (modified_query_info.getCluster()->getShardsInfo().empty()) - { - if (local_context->getSettingsRef().allow_experimental_analyzer) + if (modified_query_info.getCluster()->getShardsInfo().empty()) + { + Pipe pipe(std::make_shared(header)); + auto read_from_pipe = std::make_unique(std::move(pipe)); + read_from_pipe->setStepDescription("Read from NullSource (Distributed)"); + query_plan.addStep(std::move(read_from_pipe)); + return; - - Pipe pipe(std::make_shared(header)); - auto read_from_pipe = std::make_unique(std::move(pipe)); - read_from_pipe->setStepDescription("Read from NullSource (Distributed)"); - query_plan.addStep(std::move(read_from_pipe)); - - return; + } } const auto & snapshot_data = assert_cast(*storage_snapshot->data); @@ -893,27 +884,8 @@ void StorageDistributed::read( storage_snapshot, processed_stage); - const auto & settings = local_context->getSettingsRef(); - - ClusterProxy::AdditionalShardFilterGenerator additional_shard_filter_generator; - if (local_context->canUseParallelReplicasCustomKey(*modified_query_info.getCluster())) - { - if (auto custom_key_ast = parseCustomKeyForTable(settings.parallel_replicas_custom_key, *local_context)) - { - additional_shard_filter_generator = - [my_custom_key_ast = std::move(custom_key_ast), - column_description = this->getInMemoryMetadataPtr()->columns, - custom_key_type = settings.parallel_replicas_custom_key_filter_type.value, - custom_key_range_lower = settings.parallel_replicas_custom_key_range_lower.value, - custom_key_range_upper = settings.parallel_replicas_custom_key_range_upper.value, - context = local_context, - replica_count = modified_query_info.getCluster()->getShardsInfo().front().per_replica_pools.size()](uint64_t replica_num) -> ASTPtr - { - return getCustomKeyFilterForParallelReplica( - replica_count, replica_num - 1, my_custom_key_ast, {custom_key_type, custom_key_range_lower, custom_key_range_upper}, column_description, context); - }; - } - } + auto shard_filter_generator = ClusterProxy::getShardFilterGeneratorForCustomKey( + *modified_query_info.getCluster(), local_context, getInMemoryMetadataPtr()->columns); ClusterProxy::executeQuery( query_plan, @@ -928,7 +900,7 @@ void StorageDistributed::read( sharding_key_expr, sharding_key_column_name, distributed_settings, - additional_shard_filter_generator, + shard_filter_generator, /* is_remote_function= */ static_cast(owned_cluster)); /// This is a bug, it is possible only when there is no shards to query, and this is handled earlier. @@ -1050,7 +1022,13 @@ std::optional StorageDistributed::distributedWriteBetweenDistribu const auto & shard_info = shards_info[shard_index]; if (shard_info.isLocal()) { - InterpreterInsertQuery interpreter(new_query, query_context); + InterpreterInsertQuery interpreter( + new_query, + query_context, + /* allow_materialized */ false, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); pipeline.addCompletedPipeline(interpreter.execute().pipeline); } else diff --git a/src/Storages/StorageExecutable.cpp b/src/Storages/StorageExecutable.cpp index e475211deb3..0094723e3fd 100644 --- a/src/Storages/StorageExecutable.cpp +++ b/src/Storages/StorageExecutable.cpp @@ -8,6 +8,7 @@ #include #include +#include #include #include @@ -76,7 +77,8 @@ StorageExecutable::StorageExecutable( const ExecutableSettings & settings_, const std::vector & input_queries_, const ColumnsDescription & columns, - const ConstraintsDescription & constraints) + const ConstraintsDescription & constraints, + const String & comment) : IStorage(table_id_) , settings(settings_) , input_queries(input_queries_) @@ -85,6 +87,7 @@ StorageExecutable::StorageExecutable( StorageInMemoryMetadata storage_metadata; storage_metadata.setColumns(columns); storage_metadata.setConstraints(constraints); + storage_metadata.setComment(comment); setInMemoryMetadata(storage_metadata); ShellCommandSourceCoordinator::Configuration configuration @@ -147,7 +150,7 @@ void StorageExecutable::read( for (auto & input_query : input_queries) { QueryPipelineBuilder builder; - if (context->getSettings().allow_experimental_analyzer) + if (context->getSettingsRef().allow_experimental_analyzer) builder = InterpreterSelectQueryAnalyzer(input_query, context, {}).buildQueryPipeline(); else builder = InterpreterSelectWithUnionQuery(input_query, context, {}).buildQueryPipeline(); @@ -225,7 +228,7 @@ void registerStorageExecutable(StorageFactory & factory) { size_t max_command_execution_time = 10; - size_t max_execution_time_seconds = static_cast(args.getContext()->getSettings().max_execution_time.totalSeconds()); + size_t max_execution_time_seconds = static_cast(args.getContext()->getSettingsRef().max_execution_time.totalSeconds()); if (max_execution_time_seconds != 0 && max_command_execution_time > max_execution_time_seconds) max_command_execution_time = max_execution_time_seconds; @@ -236,7 +239,7 @@ void registerStorageExecutable(StorageFactory & factory) settings.loadFromQuery(*args.storage_def); auto global_context = args.getContext()->getGlobalContext(); - return std::make_shared(args.table_id, format, settings, input_queries, columns, constraints); + return std::make_shared(args.table_id, format, settings, input_queries, columns, constraints, args.comment); }; StorageFactory::StorageFeatures storage_features; @@ -254,4 +257,3 @@ void registerStorageExecutable(StorageFactory & factory) } } - diff --git a/src/Storages/StorageExecutable.h b/src/Storages/StorageExecutable.h index 2be2a84ab49..6748bb3223e 100644 --- a/src/Storages/StorageExecutable.h +++ b/src/Storages/StorageExecutable.h @@ -22,7 +22,8 @@ public: const ExecutableSettings & settings, const std::vector & input_queries, const ColumnsDescription & columns, - const ConstraintsDescription & constraints); + const ConstraintsDescription & constraints, + const String & comment); String getName() const override { diff --git a/src/Storages/StorageExternalDistributed.cpp b/src/Storages/StorageExternalDistributed.cpp index beb93afc972..951c87807bb 100644 --- a/src/Storages/StorageExternalDistributed.cpp +++ b/src/Storages/StorageExternalDistributed.cpp @@ -1,6 +1,6 @@ #include "StorageExternalDistributed.h" - +#include #include #include #include @@ -167,8 +167,9 @@ void registerStorageExternalDistributed(StorageFactory & factory) current_configuration, settings.postgresql_connection_pool_size, settings.postgresql_connection_pool_wait_timeout, - POSTGRESQL_POOL_WITH_FAILOVER_DEFAULT_MAX_TRIES, - settings.postgresql_connection_pool_auto_close_connection); + settings.postgresql_connection_pool_retries, + settings.postgresql_connection_pool_auto_close_connection, + settings.postgresql_connection_attempt_timeout); shards.insert(std::make_shared( args.table_id, std::move(pool), configuration.table, args.columns, args.constraints, String{}, context)); } diff --git a/src/Storages/StorageFactory.cpp b/src/Storages/StorageFactory.cpp index 9d12a1569d8..060b271d8f4 100644 --- a/src/Storages/StorageFactory.cpp +++ b/src/Storages/StorageFactory.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include diff --git a/src/Storages/StorageFile.cpp b/src/Storages/StorageFile.cpp index 7f39ff615f0..8a828938f7a 100644 --- a/src/Storages/StorageFile.cpp +++ b/src/Storages/StorageFile.cpp @@ -53,6 +53,8 @@ #include #include +#include + #include #include @@ -366,12 +368,21 @@ Strings StorageFile::getPathsList(const String & table_path, const String & user } else if (path.find_first_of("*?{") == std::string::npos) { - std::error_code error; - size_t size = fs::file_size(path, error); - if (!error) - total_bytes_to_read += size; + if (!fs::is_directory(path)) + { + std::error_code error; + size_t size = fs::file_size(path, error); + if (!error) + total_bytes_to_read += size; - paths.push_back(path); + paths.push_back(path); + } + else + { + /// We list non-directory files under that directory. + paths = listFilesWithRegexpMatching(path / fs::path("*"), total_bytes_to_read); + can_be_directory = false; + } } else { @@ -416,7 +427,9 @@ namespace { for (const auto & path : paths) { - if (auto format_from_path = FormatFactory::instance().tryGetFormatFromFileName(path)) + auto format_from_path = FormatFactory::instance().tryGetFormatFromFileName(path); + /// Use this format only if we have a schema reader for it. + if (format_from_path && FormatFactory::instance().checkIfFormatHasAnySchemaReader(*format_from_path)) { format = format_from_path; break; @@ -705,7 +718,9 @@ namespace /// If format is unknown we can try to determine it by the file name. if (!format) { - if (auto format_from_file = FormatFactory::instance().tryGetFormatFromFileName(*filename)) + auto format_from_file = FormatFactory::instance().tryGetFormatFromFileName(*filename); + /// Use this format only if we have a schema reader for it. + if (format_from_file && FormatFactory::instance().checkIfFormatHasAnySchemaReader(*format_from_file)) format = format_from_file; } @@ -1778,12 +1793,12 @@ public: String getName() const override { return "StorageFileSink"; } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { std::lock_guard cancel_lock(cancel_mutex); if (cancelled) return; - writer->write(getHeader().cloneWithColumns(chunk.detachColumns())); + writer->write(getHeader().cloneWithColumns(chunk.getColumns())); } void onCancel() override diff --git a/src/Storages/StorageFile.h b/src/Storages/StorageFile.h index ac094aeb489..f955889185c 100644 --- a/src/Storages/StorageFile.h +++ b/src/Storages/StorageFile.h @@ -89,6 +89,7 @@ public: bool supportsSubsetOfColumns(const ContextPtr & context) const; bool supportsSubcolumns() const override { return true; } + bool supportsOptimizationToSubcolumns() const override { return false; } bool supportsDynamicSubcolumns() const override { return true; } diff --git a/src/Storages/StorageFileCluster.h b/src/Storages/StorageFileCluster.h index f5a4362901e..9549f3a035c 100644 --- a/src/Storages/StorageFileCluster.h +++ b/src/Storages/StorageFileCluster.h @@ -27,15 +27,8 @@ public: const ConstraintsDescription & constraints_); std::string getName() const override { return "FileCluster"; } - RemoteQueryExecutor::Extension getTaskIteratorExtension(const ActionsDAG::Node * predicate, const ContextPtr & context) const override; - bool supportsSubcolumns() const override { return true; } - - bool supportsDynamicSubcolumns() const override { return true; } - - bool supportsTrivialCountOptimization(const StorageSnapshotPtr &, ContextPtr) const override { return true; } - private: void updateQueryToSendIfNeeded(ASTPtr & query, const StorageSnapshotPtr & storage_snapshot, const ContextPtr & context) override; diff --git a/src/Storages/StorageFromMergeTreeDataPart.cpp b/src/Storages/StorageFromMergeTreeDataPart.cpp new file mode 100644 index 00000000000..481d9e66901 --- /dev/null +++ b/src/Storages/StorageFromMergeTreeDataPart.cpp @@ -0,0 +1,59 @@ +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ +extern const int LOGICAL_ERROR; +} + +bool StorageFromMergeTreeDataPart::materializeTTLRecalculateOnly() const +{ + if (parts.empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "parts must not be empty for materializeTTLRecalculateOnly"); + return parts.front()->storage.getSettings()->materialize_ttl_recalculate_only; +} + +void StorageFromMergeTreeDataPart::read( + QueryPlan & query_plan, + const Names & column_names, + const StorageSnapshotPtr & storage_snapshot, + SelectQueryInfo & query_info, + ContextPtr context, + QueryProcessingStage::Enum /*processed_stage*/, + size_t max_block_size, + size_t num_streams) +{ + query_plan.addStep(MergeTreeDataSelectExecutor(storage).readFromParts( + parts, + alter_conversions, + column_names, + storage_snapshot, + query_info, + context, + max_block_size, + num_streams, + nullptr, + analysis_result_ptr)); +} + +StorageSnapshotPtr +StorageFromMergeTreeDataPart::getStorageSnapshot(const StorageMetadataPtr & metadata_snapshot, ContextPtr /*query_context*/) const +{ + const auto & storage_columns = metadata_snapshot->getColumns(); + if (!hasDynamicSubcolumns(storage_columns)) + return std::make_shared(*this, metadata_snapshot); + + auto data_parts = storage.getDataPartsVectorForInternalUsage(); + + auto object_columns = getConcreteObjectColumns( + data_parts.begin(), data_parts.end(), storage_columns, [](const auto & part) -> const auto & { return part->getColumns(); }); + + return std::make_shared(*this, metadata_snapshot, std::move(object_columns)); +} + +} diff --git a/src/Storages/StorageGenerateRandom.cpp b/src/Storages/StorageGenerateRandom.cpp index 2f850c76465..f3eb88f3207 100644 --- a/src/Storages/StorageGenerateRandom.cpp +++ b/src/Storages/StorageGenerateRandom.cpp @@ -30,6 +30,7 @@ #include #include +#include #include #include @@ -705,7 +706,7 @@ Pipe StorageGenerateRandom::read( } } - UInt64 query_limit = query_info.limit; + UInt64 query_limit = query_info.trivial_limit; if (query_limit && num_streams * max_block_size > query_limit) { /// We want to avoid spawning more streams than necessary @@ -717,7 +718,7 @@ Pipe StorageGenerateRandom::read( /// Will create more seed values for each source from initial seed. pcg64 generate(random_seed); - auto shared_state = std::make_shared(query_info.limit); + auto shared_state = std::make_shared(query_info.trivial_limit); for (UInt64 i = 0; i < num_streams; ++i) { diff --git a/src/Storages/StorageInMemoryMetadata.cpp b/src/Storages/StorageInMemoryMetadata.cpp index a5bae0acce5..2226de3e64f 100644 --- a/src/Storages/StorageInMemoryMetadata.cpp +++ b/src/Storages/StorageInMemoryMetadata.cpp @@ -3,6 +3,8 @@ #include #include +#include + #include #include #include diff --git a/src/Storages/StorageJoin.cpp b/src/Storages/StorageJoin.cpp index d12e5b1a20b..a0d6cf11b64 100644 --- a/src/Storages/StorageJoin.cpp +++ b/src/Storages/StorageJoin.cpp @@ -2,7 +2,7 @@ #include #include #include -#include +#include #include #include #include @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -395,11 +396,14 @@ void registerStorageJoin(StorageFactory & factory) else if (kind_str == "full") { if (strictness == JoinStrictness::Any) - strictness = JoinStrictness::RightAny; + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "ANY FULL JOINs are not implemented"); kind = JoinKind::Full; } } + if ((strictness == JoinStrictness::Semi || strictness == JoinStrictness::Anti) && (kind != JoinKind::Left && kind != JoinKind::Right)) + throw Exception(ErrorCodes::BAD_ARGUMENTS, " SEMI|ANTI JOIN should be LEFT or RIGHT"); + if (kind == JoinKind::Comma) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Second parameter of storage Join must be LEFT or INNER or RIGHT or FULL (without quotes)."); diff --git a/src/Storages/StorageKeeperMap.cpp b/src/Storages/StorageKeeperMap.cpp index 20f99070000..ef157239e26 100644 --- a/src/Storages/StorageKeeperMap.cpp +++ b/src/Storages/StorageKeeperMap.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include @@ -36,6 +37,7 @@ #include #include +#include #include #include #include @@ -63,6 +65,11 @@ namespace DB { +namespace FailPoints +{ + extern const char keepermap_fail_drop_data[]; +} + namespace ErrorCodes { extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; @@ -119,10 +126,10 @@ public: std::string getName() const override { return "StorageKeeperMapSink"; } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { auto rows = chunk.getNumRows(); - auto block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto block = getHeader().cloneWithColumns(chunk.getColumns()); WriteBufferFromOwnString wb_key; WriteBufferFromOwnString wb_value; @@ -410,18 +417,16 @@ StorageKeeperMap::StorageKeeperMap( auto code = client->tryCreate(zk_table_path, "", zkutil::CreateMode::Persistent); - // tables_path was removed with drop - if (code == Coordination::Error::ZNONODE) - { - LOG_INFO(log, "Metadata nodes were removed by another server, will retry"); - continue; - } - else if (code != Coordination::Error::ZOK) - { - throw zkutil::KeeperException(code, "Failed to create table on path {} because a table with same UUID already exists", zk_root_path); - } + /// A table on the same Keeper path already exists, we just appended our table id to subscribe as a new replica + /// We still don't know if the table matches the expected metadata so table_is_valid is not changed + /// It will be checked lazily on the first operation + if (code == Coordination::Error::ZOK) + return; - return; + if (code != Coordination::Error::ZNONODE) + throw zkutil::KeeperException(code, "Failed to create table on path {} because a table with same UUID already exists", zk_root_path); + + /// ZNONODE means we dropped zk_tables_path but didn't finish drop completely } if (client->exists(zk_dropped_path)) @@ -472,6 +477,7 @@ StorageKeeperMap::StorageKeeperMap( table_is_valid = true; + /// we are the first table created for the specified Keeper path, i.e. we are the first replica return; } @@ -560,6 +566,10 @@ void StorageKeeperMap::truncate(const ASTPtr &, const StorageMetadataPtr &, Cont bool StorageKeeperMap::dropTable(zkutil::ZooKeeperPtr zookeeper, const zkutil::EphemeralNodeHolder::Ptr & metadata_drop_lock) { + fiu_do_on(FailPoints::keepermap_fail_drop_data, + { + throw zkutil::KeeperException(Coordination::Error::ZOPERATIONTIMEOUT, "Manually triggered operation timeout"); + }); zookeeper->removeChildrenRecursive(zk_data_path); bool completely_removed = false; @@ -1248,7 +1258,10 @@ void StorageKeeperMap::mutate(const MutationCommands & commands, ContextPtr loca Block block; while (executor.pull(block)) - sink->consume(Chunk{block.getColumns(), block.rows()}); + { + auto chunk = Chunk(block.getColumns(), block.rows()); + sink->consume(chunk); + } sink->finalize(strict); } @@ -1276,6 +1289,7 @@ StoragePtr create(const StorageFactory::Arguments & args) StorageInMemoryMetadata metadata; metadata.setColumns(args.columns); metadata.setConstraints(args.constraints); + metadata.setComment(args.comment); if (!args.storage_def->primary_key) throw Exception(ErrorCodes::BAD_ARGUMENTS, "StorageKeeperMap requires one column in primary key"); diff --git a/src/Storages/StorageLog.cpp b/src/Storages/StorageLog.cpp index de0324d7998..303532dfeca 100644 --- a/src/Storages/StorageLog.cpp +++ b/src/Storages/StorageLog.cpp @@ -1,9 +1,11 @@ #include #include +#include #include #include #include +#include #include @@ -21,7 +23,6 @@ #include #include -#include "StorageLogSettings.h" #include #include #include @@ -341,7 +342,7 @@ public: } } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onFinish() override; private: @@ -398,9 +399,9 @@ private: }; -void LogSink::consume(Chunk chunk) +void LogSink::consume(Chunk & chunk) { - auto block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto block = getHeader().cloneWithColumns(chunk.getColumns()); metadata_snapshot->check(block, true); for (auto & stream : streams | boost::adaptors::map_values) diff --git a/src/Storages/StorageMaterializedView.cpp b/src/Storages/StorageMaterializedView.cpp index 316f398b476..b603d0ecf87 100644 --- a/src/Storages/StorageMaterializedView.cpp +++ b/src/Storages/StorageMaterializedView.cpp @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -92,11 +93,6 @@ StorageMaterializedView::StorageMaterializedView( { StorageInMemoryMetadata storage_metadata; storage_metadata.setColumns(columns_); - auto * storage_def = query.storage; - if (storage_def && storage_def->primary_key) - storage_metadata.primary_key = KeyDescription::getKeyFromAST(storage_def->primary_key->ptr(), - storage_metadata.columns, - local_context->getGlobalContext()); if (query.sql_security) storage_metadata.setSQLSecurity(query.sql_security->as()); @@ -109,12 +105,21 @@ StorageMaterializedView::StorageMaterializedView( throw Exception(ErrorCodes::INCORRECT_QUERY, "SELECT query is not specified for {}", getName()); /// If the destination table is not set, use inner table - has_inner_table = query.to_table_id.empty(); - if (has_inner_table && !query.storage) + auto to_table_id = query.getTargetTableID(ViewTarget::To); + has_inner_table = to_table_id.empty(); + auto to_inner_uuid = query.getTargetInnerUUID(ViewTarget::To); + auto to_table_engine = query.getTargetInnerEngine(ViewTarget::To); + + if (has_inner_table && !to_table_engine) throw Exception(ErrorCodes::INCORRECT_QUERY, "You must specify where to save results of a MaterializedView query: " "either ENGINE or an existing table in a TO clause"); + if (to_table_engine && to_table_engine->primary_key) + storage_metadata.primary_key = KeyDescription::getKeyFromAST(to_table_engine->primary_key->ptr(), + storage_metadata.columns, + local_context->getGlobalContext()); + auto select = SelectQueryDescription::getSelectQueryFromASTForMatView(query.select->clone(), query.refresh_strategy != nullptr, local_context); if (select.select_table_id) { @@ -134,25 +139,25 @@ StorageMaterializedView::StorageMaterializedView( setInMemoryMetadata(storage_metadata); - bool point_to_itself_by_uuid = has_inner_table && query.to_inner_uuid != UUIDHelpers::Nil - && query.to_inner_uuid == table_id_.uuid; - bool point_to_itself_by_name = !has_inner_table && query.to_table_id.database_name == table_id_.database_name - && query.to_table_id.table_name == table_id_.table_name; + bool point_to_itself_by_uuid = has_inner_table && to_inner_uuid != UUIDHelpers::Nil + && to_inner_uuid == table_id_.uuid; + bool point_to_itself_by_name = !has_inner_table && to_table_id.database_name == table_id_.database_name + && to_table_id.table_name == table_id_.table_name; if (point_to_itself_by_uuid || point_to_itself_by_name) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Materialized view {} cannot point to itself", table_id_.getFullTableName()); if (!has_inner_table) { - target_table_id = query.to_table_id; + target_table_id = to_table_id; } else if (LoadingStrictnessLevel::ATTACH <= mode) { /// If there is an ATTACH request, then the internal table must already be created. - target_table_id = StorageID(getStorageID().database_name, generateInnerTableName(getStorageID()), query.to_inner_uuid); + target_table_id = StorageID(getStorageID().database_name, generateInnerTableName(getStorageID()), to_inner_uuid); } else { - const String & engine = query.storage->engine->name; + const String & engine = to_table_engine->engine->name; const auto & storage_features = StorageFactory::instance().getStorageFeatures(engine); /// We will create a query to create an internal table. @@ -160,7 +165,8 @@ StorageMaterializedView::StorageMaterializedView( auto manual_create_query = std::make_shared(); manual_create_query->setDatabase(getStorageID().database_name); manual_create_query->setTable(generateInnerTableName(getStorageID())); - manual_create_query->uuid = query.to_inner_uuid; + manual_create_query->uuid = to_inner_uuid; + manual_create_query->has_uuid = to_inner_uuid != UUIDHelpers::Nil; auto new_columns_list = std::make_shared(); new_columns_list->set(new_columns_list->columns, query.columns_list->columns->ptr()); @@ -182,7 +188,9 @@ StorageMaterializedView::StorageMaterializedView( } manual_create_query->set(manual_create_query->columns_list, new_columns_list); - manual_create_query->set(manual_create_query->storage, query.storage->ptr()); + + if (to_table_engine) + manual_create_query->set(manual_create_query->storage, to_table_engine); InterpreterCreateQuery create_interpreter(manual_create_query, create_context); create_interpreter.setInternal(true); diff --git a/src/Storages/StorageMemory.cpp b/src/Storages/StorageMemory.cpp index f69c4adb552..42bac783618 100644 --- a/src/Storages/StorageMemory.cpp +++ b/src/Storages/StorageMemory.cpp @@ -1,5 +1,5 @@ -#include #include +#include #include #include @@ -37,7 +37,6 @@ #include #include - namespace DB { @@ -63,7 +62,7 @@ public: String getName() const override { return "MemorySink"; } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { auto block = getHeader().cloneWithColumns(chunk.getColumns()); storage_snapshot->metadata->check(block, true); @@ -408,7 +407,7 @@ namespace auto data_file_path = temp_dir / fs::path{file_paths[data_bin_pos]}.filename(); auto data_out_compressed = temp_disk->writeFile(data_file_path); auto data_out = std::make_unique(*data_out_compressed, CompressionCodecFactory::instance().getDefaultCodec(), max_compress_block_size); - NativeWriter block_out{*data_out, 0, metadata_snapshot->getSampleBlock(), false, &index}; + NativeWriter block_out{*data_out, 0, metadata_snapshot->getSampleBlock(), std::nullopt, false, &index}; for (const auto & block : *blocks) block_out.write(block); data_out->finalize(); diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp index 2dbe82c92d8..f5bc183931f 100644 --- a/src/Storages/StorageMerge.cpp +++ b/src/Storages/StorageMerge.cpp @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/StorageMergeTree.cpp b/src/Storages/StorageMergeTree.cpp index 9255ee00340..527872d701e 100644 --- a/src/Storages/StorageMergeTree.cpp +++ b/src/Storages/StorageMergeTree.cpp @@ -1,51 +1,43 @@ -#include "StorageMergeTree.h" -#include "Core/QueryProcessingStage.h" -#include "Storages/MergeTree/IMergeTreeDataPart.h" +#include #include #include -#include -#include #include +#include +#include #include -#include "Common/Exception.h" -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include #include +#include +#include +#include +#include +#include +#include #include #include -#include #include -#include -#include #include #include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include #include #include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include #include +#include +#include +#include +#include namespace DB @@ -220,24 +212,50 @@ void StorageMergeTree::read( { ClusterProxy::executeQueryWithParallelReplicas( query_plan, getStorageID(), processed_stage, query_info.query, local_context, query_info.storage_limits); + return; } - else - { - const bool enable_parallel_reading = local_context->canUseParallelReplicasOnFollower() - && local_context->getSettingsRef().parallel_replicas_for_non_replicated_merge_tree - && (!local_context->getSettingsRef().allow_experimental_analyzer || query_info.analyzer_can_use_parallel_replicas_on_follower); - if (auto plan = reader.read( - column_names, + if (local_context->canUseParallelReplicasCustomKey() && settings.parallel_replicas_for_non_replicated_merge_tree + && !settings.allow_experimental_analyzer && local_context->getClientInfo().distributed_depth == 0) + { + if (auto cluster = local_context->getClusterForParallelReplicas(); + local_context->canUseParallelReplicasCustomKeyForCluster(*cluster)) + { + auto modified_query_info = query_info; + modified_query_info.cluster = std::move(cluster); + ClusterProxy::executeQueryWithParallelReplicasCustomKey( + query_plan, + getStorageID(), + std::move(modified_query_info), + getInMemoryMetadataPtr()->getColumns(), storage_snapshot, - query_info, - local_context, - max_block_size, - num_streams, - nullptr, - enable_parallel_reading)) - query_plan = std::move(*plan); + processed_stage, + query_info.query, + local_context); + return; + } + else + LOG_WARNING( + log, + "Parallel replicas with custom key will not be used because cluster defined by 'cluster_for_parallel_replicas' ('{}') has " + "multiple shards", + cluster->getName()); } + + const bool enable_parallel_reading = local_context->canUseParallelReplicasOnFollower() + && local_context->getSettingsRef().parallel_replicas_for_non_replicated_merge_tree + && (!local_context->getSettingsRef().allow_experimental_analyzer || query_info.current_table_chosen_for_reading_with_parallel_replicas); + + if (auto plan = reader.read( + column_names, + storage_snapshot, + query_info, + local_context, + max_block_size, + num_streams, + nullptr, + enable_parallel_reading)) + query_plan = std::move(*plan); } std::optional StorageMergeTree::totalRows(const Settings &) const @@ -487,18 +505,18 @@ Int64 StorageMergeTree::startMutation(const MutationCommands & commands, Context additional_info = fmt::format(" (TID: {}; TIDH: {})", current_tid, current_tid.getHash()); } - Int64 version; + MergeTreeMutationEntry entry(commands, disk, relative_data_path, insert_increment.get(), current_tid, getContext()->getWriteSettings()); + Int64 version = increment.get(); + entry.commit(version); + String mutation_id = entry.file_name; + if (txn) + txn->addMutation(shared_from_this(), mutation_id); + + bool alter_conversions_mutations_updated = updateAlterConversionsMutations(entry.commands, alter_conversions_mutations, /* remove= */ false); + { std::lock_guard lock(currently_processing_in_background_mutex); - MergeTreeMutationEntry entry(commands, disk, relative_data_path, insert_increment.get(), current_tid, getContext()->getWriteSettings()); - version = increment.get(); - entry.commit(version); - String mutation_id = entry.file_name; - if (txn) - txn->addMutation(shared_from_this(), mutation_id); - - bool alter_conversions_mutations_updated = updateAlterConversionsMutations(entry.commands, alter_conversions_mutations, /* remove= */ false); bool inserted = current_mutations_by_version.try_emplace(version, std::move(entry)).second; if (!inserted) { @@ -509,9 +527,9 @@ Int64 StorageMergeTree::startMutation(const MutationCommands & commands, Context } throw Exception(ErrorCodes::LOGICAL_ERROR, "Mutation {} already exists, it's a bug", version); } - - LOG_INFO(log, "Added mutation: {}{}", mutation_id, additional_info); } + + LOG_INFO(log, "Added mutation: {}{}", mutation_id, additional_info); background_operations_assignee.trigger(); return version; } @@ -630,12 +648,13 @@ void StorageMergeTree::mutate(const MutationCommands & commands, ContextPtr quer { /// It's important to serialize order of mutations with alter queries because /// they can depend on each other. - if (auto alter_lock = tryLockForAlter(query_context->getSettings().lock_acquire_timeout); alter_lock == std::nullopt) + if (auto alter_lock = tryLockForAlter(query_context->getSettingsRef().lock_acquire_timeout); alter_lock == std::nullopt) { - throw Exception(ErrorCodes::TIMEOUT_EXCEEDED, - "Cannot start mutation in {}ms because some metadata-changing ALTER (MODIFY|RENAME|ADD|DROP) is currently executing. " - "You can change this timeout with `lock_acquire_timeout` setting", - query_context->getSettings().lock_acquire_timeout.totalMilliseconds()); + throw Exception( + ErrorCodes::TIMEOUT_EXCEEDED, + "Cannot start mutation in {}ms because some metadata-changing ALTER (MODIFY|RENAME|ADD|DROP) is currently executing. " + "You can change this timeout with `lock_acquire_timeout` setting", + query_context->getSettingsRef().lock_acquire_timeout.totalMilliseconds()); } version = startMutation(commands, query_context); } @@ -1168,7 +1187,6 @@ bool StorageMergeTree::merge( task->setCurrentTransaction(MergeTreeTransactionHolder{}, MergeTreeTransactionPtr{txn}); executeHere(task); - return true; } @@ -1457,13 +1475,13 @@ bool StorageMergeTree::scheduleDataProcessingJob(BackgroundJobsAssignee & assign cleared_count += clearOldPartsFromFilesystem(); cleared_count += clearOldMutations(); cleared_count += clearEmptyParts(); + cleared_count += unloadPrimaryKeysOfOutdatedParts(); return cleared_count; /// TODO maybe take into account number of cleared objects when calculating backoff }, common_assignee_trigger, getStorageID()), /* need_trigger */ false); scheduled = true; } - return scheduled; } @@ -1550,6 +1568,12 @@ bool StorageMergeTree::optimize( { assertNotReadonly(); + if (deduplicate && getInMemoryMetadataPtr()->hasProjections()) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, + "OPTIMIZE DEDUPLICATE query is not supported for table {} as it has projections. " + "User should drop all the projections manually before running the query", + getStorageID().getTableName()); + if (deduplicate) { if (deduplicate_by_columns.empty()) @@ -1564,9 +1588,7 @@ bool StorageMergeTree::optimize( if (!partition && final) { if (cleanup && this->merging_params.mode != MergingParams::Mode::Replacing) - { throw Exception(ErrorCodes::CANNOT_ASSIGN_OPTIMIZE, "Cannot OPTIMIZE with CLEANUP table: only ReplacingMergeTree can be CLEANUP"); - } if (cleanup && !getSettings()->allow_experimental_replacing_merge_with_cleanup) throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "Experimental merges with CLEANUP are not allowed"); @@ -1580,15 +1602,15 @@ bool StorageMergeTree::optimize( for (const String & partition_id : partition_ids) { if (!merge( - true, - partition_id, - true, - deduplicate, - deduplicate_by_columns, - cleanup, - txn, - disable_reason, - local_context->getSettingsRef().optimize_skip_merged_partitions)) + true, + partition_id, + true, + deduplicate, + deduplicate_by_columns, + cleanup, + txn, + disable_reason, + local_context->getSettingsRef().optimize_skip_merged_partitions)) { constexpr auto message = "Cannot OPTIMIZE table: {}"; if (disable_reason.text.empty()) @@ -1608,15 +1630,15 @@ bool StorageMergeTree::optimize( partition_id = getPartitionIDFromQuery(partition, local_context); if (!merge( - true, - partition_id, - final, - deduplicate, - deduplicate_by_columns, - cleanup, - txn, - disable_reason, - local_context->getSettingsRef().optimize_skip_merged_partitions)) + true, + partition_id, + final, + deduplicate, + deduplicate_by_columns, + cleanup, + txn, + disable_reason, + local_context->getSettingsRef().optimize_skip_merged_partitions)) { constexpr auto message = "Cannot OPTIMIZE table: {}"; if (disable_reason.text.empty()) diff --git a/src/Storages/StorageMergeTree.h b/src/Storages/StorageMergeTree.h index 4d819508934..064b51739bd 100644 --- a/src/Storages/StorageMergeTree.h +++ b/src/Storages/StorageMergeTree.h @@ -211,7 +211,6 @@ private: bool optimize_skip_merged_partitions = false, SelectPartsDecision * select_decision_out = nullptr); - MergeMutateSelectedEntryPtr selectPartsToMutate( const StorageMetadataPtr & metadata_snapshot, PreformattedMessage & disable_reason, TableLockHolder & table_lock_holder, std::unique_lock & currently_processing_in_background_mutex_lock); @@ -310,6 +309,7 @@ private: }; protected: + /// Collect mutations that have to be applied on the fly: currently they are only RENAME COLUMN. MutationCommands getAlterMutationCommandsForPart(const DataPartPtr & part) const override; }; diff --git a/src/Storages/StorageMySQL.cpp b/src/Storages/StorageMySQL.cpp index da391909dff..1d1a0ffdeaf 100644 --- a/src/Storages/StorageMySQL.cpp +++ b/src/Storages/StorageMySQL.cpp @@ -21,6 +21,7 @@ #include #include #include +#include #include #include @@ -151,9 +152,9 @@ public: String getName() const override { return "StorageMySQLSink"; } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { - auto block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto block = getHeader().cloneWithColumns(chunk.getColumns()); auto blocks = splitBlocks(block, max_batch_rows); mysqlxx::Transaction trans(entry); try diff --git a/src/Storages/StoragePostgreSQL.cpp b/src/Storages/StoragePostgreSQL.cpp index a8713c61e4d..d3214e7ed13 100644 --- a/src/Storages/StoragePostgreSQL.cpp +++ b/src/Storages/StoragePostgreSQL.cpp @@ -227,9 +227,9 @@ public: String getName() const override { return "PostgreSQLSink"; } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { - auto block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto block = getHeader().cloneWithColumns(chunk.getColumns()); if (!inserter) { if (on_conflict.empty()) @@ -613,8 +613,9 @@ void registerStoragePostgreSQL(StorageFactory & factory) auto pool = std::make_shared(configuration, settings.postgresql_connection_pool_size, settings.postgresql_connection_pool_wait_timeout, - POSTGRESQL_POOL_WITH_FAILOVER_DEFAULT_MAX_TRIES, - settings.postgresql_connection_pool_auto_close_connection); + settings.postgresql_connection_pool_retries, + settings.postgresql_connection_pool_auto_close_connection, + settings.postgresql_connection_attempt_timeout); return std::make_shared( args.table_id, diff --git a/src/Storages/StorageRedis.cpp b/src/Storages/StorageRedis.cpp index 83bb3c606c9..1a275320f43 100644 --- a/src/Storages/StorageRedis.cpp +++ b/src/Storages/StorageRedis.cpp @@ -147,7 +147,7 @@ class RedisSink : public SinkToStorage public: RedisSink(StorageRedis & storage_, const StorageMetadataPtr & metadata_snapshot_); - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; String getName() const override { return "RedisSink"; } private: @@ -169,10 +169,10 @@ RedisSink::RedisSink(StorageRedis & storage_, const StorageMetadataPtr & metadat } } -void RedisSink::consume(Chunk chunk) +void RedisSink::consume(Chunk & chunk) { auto rows = chunk.getNumRows(); - auto block = getHeader().cloneWithColumns(chunk.detachColumns()); + auto block = getHeader().cloneWithColumns(chunk.getColumns()); WriteBufferFromOwnString wb_key; WriteBufferFromOwnString wb_value; @@ -567,7 +567,8 @@ void StorageRedis::mutate(const MutationCommands & commands, ContextPtr context_ Block block; while (executor.pull(block)) { - sink->consume(Chunk{block.getColumns(), block.rows()}); + Chunk chunk(block.getColumns(), block.rows()); + sink->consume(chunk); } } diff --git a/src/Storages/StorageReplicatedMergeTree.cpp b/src/Storages/StorageReplicatedMergeTree.cpp index a127384c03c..da379a466af 100644 --- a/src/Storages/StorageReplicatedMergeTree.cpp +++ b/src/Storages/StorageReplicatedMergeTree.cpp @@ -5,22 +5,25 @@ #include #include +#include #include #include #include #include +#include #include #include #include #include +#include #include +#include #include #include -#include -#include -#include +#include #include +#include #include @@ -39,6 +42,7 @@ #include #include #include +#include #include #include #include @@ -1512,7 +1516,7 @@ static time_t tryGetPartCreateTime(zkutil::ZooKeeperPtr & zookeeper, const Strin void StorageReplicatedMergeTree::paranoidCheckForCoveredPartsInZooKeeperOnStart(const Strings & parts_in_zk, const Strings & parts_to_fetch) const { -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD constexpr bool paranoid_check_for_covered_parts_default = true; #else constexpr bool paranoid_check_for_covered_parts_default = false; @@ -2379,7 +2383,7 @@ static void paranoidCheckForCoveredPartsInZooKeeper( const String & covering_part_name, const StorageReplicatedMergeTree & storage) { -#ifdef ABORT_ON_LOGICAL_ERROR +#ifdef DEBUG_OR_SANITIZER_BUILD constexpr bool paranoid_check_for_covered_parts_default = true; #else constexpr bool paranoid_check_for_covered_parts_default = false; @@ -3936,7 +3940,7 @@ void StorageReplicatedMergeTree::mergeSelectingTask() merge_selecting_task->schedule(); else { - LOG_TRACE(log, "Scheduling next merge selecting task after {}ms", merge_selecting_sleep_ms); + LOG_TRACE(log, "Scheduling next merge selecting task after {}ms, current attempt status: {}", merge_selecting_sleep_ms, result); merge_selecting_task->scheduleAfter(merge_selecting_sleep_ms); } } @@ -5272,6 +5276,8 @@ void StorageReplicatedMergeTree::flushAndPrepareForShutdown() if (shutdown_prepared_called.exchange(true)) return; + LOG_TRACE(log, "Start preparing for shutdown"); + try { auto settings_ptr = getSettings(); @@ -5282,7 +5288,11 @@ void StorageReplicatedMergeTree::flushAndPrepareForShutdown() stopBeingLeader(); if (attach_thread) + { attach_thread->shutdown(); + LOG_TRACE(log, "The attach thread is shutdown"); + } + restarting_thread.shutdown(/* part_of_full_shutdown */true); /// Explicitly set the event, because the restarting thread will not set it again @@ -5295,6 +5305,8 @@ void StorageReplicatedMergeTree::flushAndPrepareForShutdown() shutdown_deadline.emplace(std::chrono::system_clock::now()); throw; } + + LOG_TRACE(log, "Finished preparing for shutdown"); } void StorageReplicatedMergeTree::partialShutdown() @@ -5332,6 +5344,8 @@ void StorageReplicatedMergeTree::shutdown(bool) if (shutdown_called.exchange(true)) return; + LOG_TRACE(log, "Shutdown started"); + flushAndPrepareForShutdown(); if (!shutdown_deadline.has_value()) @@ -5374,6 +5388,7 @@ void StorageReplicatedMergeTree::shutdown(bool) /// Wait for all of them std::lock_guard lock(data_parts_exchange_ptr->rwlock); } + LOG_TRACE(log, "Shutdown finished"); } @@ -5460,13 +5475,45 @@ void StorageReplicatedMergeTree::read( /// 2. Do not read parts that have not yet been written to the quorum of the replicas. /// For this you have to synchronously go to ZooKeeper. if (settings.select_sequential_consistency) + { readLocalSequentialConsistencyImpl(query_plan, column_names, storage_snapshot, query_info, local_context, max_block_size, num_streams); + return; + } /// reading step for parallel replicas with new analyzer is built in Planner, so don't do it here - else if (local_context->canUseParallelReplicasOnInitiator() && !settings.allow_experimental_analyzer) + if (local_context->canUseParallelReplicasOnInitiator() && !settings.allow_experimental_analyzer) + { readParallelReplicasImpl(query_plan, column_names, query_info, local_context, processed_stage); - else - readLocalImpl(query_plan, column_names, storage_snapshot, query_info, local_context, max_block_size, num_streams); -} + return; + } + + if (local_context->canUseParallelReplicasCustomKey() && !settings.allow_experimental_analyzer + && local_context->getClientInfo().distributed_depth == 0) + { + if (auto cluster = local_context->getClusterForParallelReplicas(); + local_context->canUseParallelReplicasCustomKeyForCluster(*cluster)) + { + auto modified_query_info = query_info; + modified_query_info.cluster = std::move(cluster); + ClusterProxy::executeQueryWithParallelReplicasCustomKey( + query_plan, + getStorageID(), + std::move(modified_query_info), + getInMemoryMetadataPtr()->getColumns(), + storage_snapshot, + processed_stage, + query_info.query, + local_context); + return; + } + else + LOG_WARNING( + log, + "Parallel replicas with custom key will not be used because cluster defined by 'cluster_for_parallel_replicas' ('{}') has " + "multiple shards", + cluster->getName()); + } + + readLocalImpl(query_plan, column_names, storage_snapshot, query_info, local_context, max_block_size, num_streams); } void StorageReplicatedMergeTree::readLocalSequentialConsistencyImpl( QueryPlan & query_plan, @@ -5508,7 +5555,8 @@ void StorageReplicatedMergeTree::readLocalImpl( const size_t num_streams) { const bool enable_parallel_reading = local_context->canUseParallelReplicasOnFollower() - && (!local_context->getSettingsRef().allow_experimental_analyzer || query_info.analyzer_can_use_parallel_replicas_on_follower); + && (!local_context->getSettingsRef().allow_experimental_analyzer + || query_info.current_table_chosen_for_reading_with_parallel_replicas); auto plan = reader.read( column_names, storage_snapshot, query_info, @@ -5746,6 +5794,12 @@ bool StorageReplicatedMergeTree::optimize( if (!is_leader) throw Exception(ErrorCodes::NOT_A_LEADER, "OPTIMIZE cannot be done on this replica because it is not a leader"); + if (deduplicate && getInMemoryMetadataPtr()->hasProjections()) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, + "OPTIMIZE DEDUPLICATE query is not supported for table {} as it has projections. " + "User should drop all the projections manually before running the query", + getStorageID().getTableName()); + if (cleanup) { if (!getSettings()->allow_experimental_replacing_merge_with_cleanup) @@ -7309,7 +7363,7 @@ void StorageReplicatedMergeTree::fetchPartition( if (try_no) LOG_INFO(log, "Some of parts ({}) are missing. Will try to fetch covering parts.", missing_parts.size()); - if (try_no >= query_context->getSettings().max_fetch_partition_retries_count) + if (try_no >= query_context->getSettingsRef().max_fetch_partition_retries_count) throw Exception(ErrorCodes::TOO_MANY_RETRIES_TO_FETCH_PARTS, "Too many retries to fetch parts from {}:{}", from_zookeeper_name, best_replica_path); diff --git a/src/Storages/StorageSQLite.cpp b/src/Storages/StorageSQLite.cpp index 179e4cee199..b90b15f3b99 100644 --- a/src/Storages/StorageSQLite.cpp +++ b/src/Storages/StorageSQLite.cpp @@ -50,6 +50,7 @@ StorageSQLite::StorageSQLite( const String & remote_table_name_, const ColumnsDescription & columns_, const ConstraintsDescription & constraints_, + const String & comment, ContextPtr context_) : IStorage(table_id_) , WithContext(context_->getGlobalContext()) @@ -71,6 +72,7 @@ StorageSQLite::StorageSQLite( storage_metadata.setConstraints(constraints_); setInMemoryMetadata(storage_metadata); + storage_metadata.setComment(comment); } @@ -141,7 +143,7 @@ public: String getName() const override { return "SQLiteSink"; } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { auto block = getHeader().cloneWithColumns(chunk.getColumns()); WriteBufferFromOwnString sqlbuf; @@ -211,7 +213,7 @@ void registerStorageSQLite(StorageFactory & factory) auto sqlite_db = openSQLiteDB(database_path, args.getContext(), /* throw_on_error */ args.mode <= LoadingStrictnessLevel::CREATE); return std::make_shared(args.table_id, sqlite_db, database_path, - table_name, args.columns, args.constraints, args.getContext()); + table_name, args.columns, args.constraints, args.comment, args.getContext()); }, { .supports_schema_inference = true, diff --git a/src/Storages/StorageSQLite.h b/src/Storages/StorageSQLite.h index ed673123fe0..97638ac04cb 100644 --- a/src/Storages/StorageSQLite.h +++ b/src/Storages/StorageSQLite.h @@ -27,6 +27,7 @@ public: const String & remote_table_name_, const ColumnsDescription & columns_, const ConstraintsDescription & constraints_, + const String & comment, ContextPtr context_); std::string getName() const override { return "SQLite"; } diff --git a/src/Storages/StorageSet.cpp b/src/Storages/StorageSet.cpp index 5b7f9fc0ac2..5692da4e454 100644 --- a/src/Storages/StorageSet.cpp +++ b/src/Storages/StorageSet.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -44,7 +45,7 @@ public: const String & backup_file_name_, bool persistent_); String getName() const override { return "SetOrJoinSink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onFinish() override; private: @@ -82,9 +83,9 @@ SetOrJoinSink::SetOrJoinSink( { } -void SetOrJoinSink::consume(Chunk chunk) +void SetOrJoinSink::consume(Chunk & chunk) { - Block block = getHeader().cloneWithColumns(chunk.detachColumns()); + Block block = getHeader().cloneWithColumns(chunk.getColumns()); table.insertBlock(block, getContext()); if (persistent) diff --git a/src/Storages/StorageSet.h b/src/Storages/StorageSet.h index 67a9528ff5e..5b6d899b48d 100644 --- a/src/Storages/StorageSet.h +++ b/src/Storages/StorageSet.h @@ -2,7 +2,6 @@ #include #include -#include namespace DB diff --git a/src/Storages/StorageStripeLog.cpp b/src/Storages/StorageStripeLog.cpp index 8df87d6290f..c892cca4523 100644 --- a/src/Storages/StorageStripeLog.cpp +++ b/src/Storages/StorageStripeLog.cpp @@ -7,6 +7,8 @@ #include #include +#include + #include #include #include @@ -193,7 +195,7 @@ public: storage.saveFileSizes(lock); size_t initial_data_size = storage.file_checker.getFileSize(storage.data_file_path); - block_out = std::make_unique(*data_out, 0, metadata_snapshot->getSampleBlock(), false, &storage.indices, initial_data_size); + block_out = std::make_unique(*data_out, 0, metadata_snapshot->getSampleBlock(), std::nullopt, false, &storage.indices, initial_data_size); } String getName() const override { return "StripeLogSink"; } @@ -226,9 +228,9 @@ public: } } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { - block_out->write(getHeader().cloneWithColumns(chunk.detachColumns())); + block_out->write(getHeader().cloneWithColumns(chunk.getColumns())); } void onFinish() override @@ -289,7 +291,7 @@ StorageStripeLog::StorageStripeLog( , data_file_path(table_path + "data.bin") , index_file_path(table_path + "index.mrk") , file_checker(disk, table_path + "sizes.json") - , max_compress_block_size(context_->getSettings().max_compress_block_size) + , max_compress_block_size(context_->getSettingsRef().max_compress_block_size) , log(getLogger("StorageStripeLog")) { StorageInMemoryMetadata storage_metadata; diff --git a/src/Storages/StorageURL.cpp b/src/Storages/StorageURL.cpp index 895da028fc2..efee6151afd 100644 --- a/src/Storages/StorageURL.cpp +++ b/src/Storages/StorageURL.cpp @@ -36,6 +36,8 @@ #include #include #include +#include +#include #include #include @@ -565,12 +567,12 @@ StorageURLSink::StorageURLSink( } -void StorageURLSink::consume(Chunk chunk) +void StorageURLSink::consume(Chunk & chunk) { std::lock_guard lock(cancel_mutex); if (cancelled) return; - writer->write(getHeader().cloneWithColumns(chunk.detachColumns())); + writer->write(getHeader().cloneWithColumns(chunk.getColumns())); } void StorageURLSink::onCancel() @@ -735,7 +737,9 @@ namespace { for (const auto & url : options) { - if (auto format_from_file_name = FormatFactory::instance().tryGetFormatFromFileName(url)) + auto format_from_file_name = FormatFactory::instance().tryGetFormatFromFileName(url); + /// Use this format only if we have a schema reader for it. + if (format_from_file_name && FormatFactory::instance().checkIfFormatHasAnySchemaReader(*format_from_file_name)) { format = format_from_file_name; break; diff --git a/src/Storages/StorageURL.h b/src/Storages/StorageURL.h index 3090f8db12e..1804079e75f 100644 --- a/src/Storages/StorageURL.h +++ b/src/Storages/StorageURL.h @@ -251,7 +251,7 @@ public: const String & method = Poco::Net::HTTPRequest::HTTP_POST); std::string getName() const override { return "StorageURLSink"; } - void consume(Chunk chunk) override; + void consume(Chunk & chunk) override; void onCancel() override; void onException(std::exception_ptr exception) override; void onFinish() override; @@ -296,6 +296,7 @@ public: } bool supportsSubcolumns() const override { return true; } + bool supportsOptimizationToSubcolumns() const override { return false; } bool supportsDynamicSubcolumns() const override { return true; } diff --git a/src/Storages/StorageURLCluster.cpp b/src/Storages/StorageURLCluster.cpp index 2e7c63d0097..69041482fd8 100644 --- a/src/Storages/StorageURLCluster.cpp +++ b/src/Storages/StorageURLCluster.cpp @@ -3,6 +3,7 @@ #include #include +#include #include #include #include diff --git a/src/Storages/StorageURLCluster.h b/src/Storages/StorageURLCluster.h index a6334e7430d..31bffa06210 100644 --- a/src/Storages/StorageURLCluster.h +++ b/src/Storages/StorageURLCluster.h @@ -30,15 +30,8 @@ public: const StorageURL::Configuration & configuration_); std::string getName() const override { return "URLCluster"; } - RemoteQueryExecutor::Extension getTaskIteratorExtension(const ActionsDAG::Node * predicate, const ContextPtr & context) const override; - bool supportsSubcolumns() const override { return true; } - - bool supportsDynamicSubcolumns() const override { return true; } - - bool supportsTrivialCountOptimization(const StorageSnapshotPtr &, ContextPtr) const override { return true; } - private: void updateQueryToSendIfNeeded(ASTPtr & query, const StorageSnapshotPtr & storage_snapshot, const ContextPtr & context) override; diff --git a/src/Storages/StorageView.cpp b/src/Storages/StorageView.cpp index 016de94c17c..5f768bce978 100644 --- a/src/Storages/StorageView.cpp +++ b/src/Storages/StorageView.cpp @@ -19,6 +19,8 @@ #include +#include + #include #include #include diff --git a/src/Storages/StorageXDBC.cpp b/src/Storages/StorageXDBC.cpp index fb8fa2d6da4..fa23a574ade 100644 --- a/src/Storages/StorageXDBC.cpp +++ b/src/Storages/StorageXDBC.cpp @@ -4,6 +4,8 @@ #include #include +#include +#include #include #include #include diff --git a/src/Storages/System/StorageSystemColumns.cpp b/src/Storages/System/StorageSystemColumns.cpp index 8dd8d3b6154..6a7810b97f9 100644 --- a/src/Storages/System/StorageSystemColumns.cpp +++ b/src/Storages/System/StorageSystemColumns.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/System/StorageSystemContributors.generated.cpp b/src/Storages/System/StorageSystemContributors.generated.cpp index 9201eef185f..35b9c0008c6 100644 --- a/src/Storages/System/StorageSystemContributors.generated.cpp +++ b/src/Storages/System/StorageSystemContributors.generated.cpp @@ -1,5 +1,6 @@ // autogenerated by tests/ci/version_helper.py const char * auto_contributors[] { + "0x01f", "0xflotus", "13DaGGeR", "1lann", @@ -167,6 +168,7 @@ const char * auto_contributors[] { "AnneClickHouse", "Anselmo D. Adams", "Anthony N. Simon", + "AntiTopQuark", "Anton Ivashkin", "Anton Kobzev", "Anton Kozlov", @@ -299,6 +301,7 @@ const char * auto_contributors[] { "Dan Wu", "DanRoscigno", "Dani Pozo", + "Daniel Anugerah", "Daniel Bershatsky", "Daniel Byta", "Daniel Dao", @@ -370,6 +373,7 @@ const char * auto_contributors[] { "Elena", "Elena Baskakova", "Elena Torró", + "Elena Torró Martínez", "Elghazal Ahmed", "Eliot Hautefeuille", "Elizaveta Mironyuk", @@ -415,6 +419,7 @@ const char * auto_contributors[] { "FgoDt", "Filatenkov Artur", "Filipe Caixeta", + "Filipp Bakanov", "Filipp Ozinov", "Filippov Denis", "Fille", @@ -451,6 +456,7 @@ const char * auto_contributors[] { "Gleb Novikov", "Gleb-Tretyakov", "GoGoWen2021", + "Gosha Letov", "Gregory", "Grigorii Sokolik", "Grigory", @@ -461,6 +467,7 @@ const char * auto_contributors[] { "Guillaume Tassery", "Guo Wangyang", "Guo Wei (William)", + "Guspan Tanadi", "Haavard Kvaalen", "Habibullah Oladepo", "HaiBo Li", @@ -474,6 +481,7 @@ const char * auto_contributors[] { "HarryLeeIBM", "Hasitha Kanchana", "Hasnat", + "Haydn", "Heena Bansal", "HeenaBansal2009", "Hendrik M", @@ -606,6 +614,7 @@ const char * auto_contributors[] { "Kevin Chiang", "Kevin Michel", "Kevin Mingtarja", + "Kevin Song", "Kevin Zhang", "KevinyhZou", "KinderRiven", @@ -661,6 +670,7 @@ const char * auto_contributors[] { "Lewinma", "Li Shuai", "Li Yin", + "Linh Giang", "Lino Uruñuela", "Lirikl", "Liu Cong", @@ -690,6 +700,7 @@ const char * auto_contributors[] { "Maksim Alekseev", "Maksim Buren", "Maksim Fedotov", + "Maksim Galkin", "Maksim Kita", "Maksym Sobolyev", "Mal Curtis", @@ -724,6 +735,7 @@ const char * auto_contributors[] { "Max Akhmedov", "Max Bruce", "Max K", + "Max K.", "Max Kainov", "Max Vetrov", "MaxTheHuman", @@ -811,6 +823,7 @@ const char * auto_contributors[] { "Nataly Merezhuk", "Natalya Chizhonkova", "Natasha Murashkina", + "Nathan Clevenger", "NeZeD [Mac Pro]", "Neeke Gao", "Neng Liu", @@ -946,6 +959,7 @@ const char * auto_contributors[] { "Robert Coelho", "Robert Hodges", "Robert Schulze", + "Rodolphe Dugé de Bernonville", "RogerYK", "Rohit Agarwal", "Romain Neutron", @@ -1107,6 +1121,7 @@ const char * auto_contributors[] { "Timur Solodovnikov", "TiunovNN", "Tobias Adamson", + "Tobias Florek", "Tobias Lins", "Tom Bombadil", "Tom Risse", @@ -1231,11 +1246,13 @@ const char * auto_contributors[] { "Yingchun Lai", "Yingfan Chen", "Yinzheng-Sun", + "Yinzuo Jiang", "Yiğit Konur", "Yohann Jardin", "Yong Wang", "Yong-Hao Zou", "Youenn Lebras", + "Your Name", "Yu, Peng", "Yuko Takagi", "Yuntao Wu", @@ -1250,6 +1267,7 @@ const char * auto_contributors[] { "Yury Stankevich", "Yusuke Tanaka", "Zach Naimon", + "Zawa-II", "Zheng Miao", "ZhiHong Zhang", "ZhiYong Wang", @@ -1380,6 +1398,7 @@ const char * auto_contributors[] { "conicliu", "copperybean", "coraxster", + "cw5121", "cwkyaoyao", "d.v.semenov", "dalei2019", @@ -1460,12 +1479,14 @@ const char * auto_contributors[] { "fuzzERot", "fyu", "g-arslan", + "gabrielmcg44", "ggerogery", "giordyb", "glockbender", "glushkovds", "grantovsky", "gulige", + "gun9nir", "guoleiyi", "guomaolin", "guov100", @@ -1527,6 +1548,7 @@ const char * auto_contributors[] { "jferroal", "jiahui-97", "jianmei zhang", + "jiaosenvip", "jinjunzh", "jiyoungyoooo", "jktng", @@ -1541,6 +1563,7 @@ const char * auto_contributors[] { "jun won", "jus1096", "justindeguzman", + "jwoodhead", "jyz0309", "karnevil13", "kashwy", @@ -1633,10 +1656,12 @@ const char * auto_contributors[] { "mateng0915", "mateng915", "mauidude", + "max-vostrikov", "maxim", "maxim-babenko", "maxkuzn", "maxulan", + "maxvostrikov", "mayamika", "mehanizm", "melin", @@ -1677,6 +1702,7 @@ const char * auto_contributors[] { "nathanbegbie", "nauta", "nautaa", + "nauu", "ndchikin", "nellicus", "nemonlou", @@ -1975,6 +2001,7 @@ const char * auto_contributors[] { "张健", "张风啸", "徐炘", + "忒休斯~Theseus", "曲正鹏", "木木夕120", "未来星___费", diff --git a/src/Storages/System/StorageSystemDDLWorkerQueue.cpp b/src/Storages/System/StorageSystemDDLWorkerQueue.cpp index f96b839a322..ae1a233ea81 100644 --- a/src/Storages/System/StorageSystemDDLWorkerQueue.cpp +++ b/src/Storages/System/StorageSystemDDLWorkerQueue.cpp @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/System/StorageSystemDetachedParts.cpp b/src/Storages/System/StorageSystemDetachedParts.cpp index f48a8c67971..7e4c1de1c65 100644 --- a/src/Storages/System/StorageSystemDetachedParts.cpp +++ b/src/Storages/System/StorageSystemDetachedParts.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include @@ -328,7 +329,7 @@ void ReadFromSystemDetachedParts::applyFilters(ActionDAGNodes added_filter_nodes filter = VirtualColumnUtils::splitFilterDagForAllowedInputs(predicate, &block); if (filter) - VirtualColumnUtils::buildSetsForDAG(filter, context); + VirtualColumnUtils::buildSetsForDAG(*filter, context); } } diff --git a/src/Storages/System/StorageSystemDetachedTables.cpp b/src/Storages/System/StorageSystemDetachedTables.cpp new file mode 100644 index 00000000000..56c5e49b467 --- /dev/null +++ b/src/Storages/System/StorageSystemDetachedTables.cpp @@ -0,0 +1,241 @@ +#include "StorageSystemDetachedTables.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + + +namespace DB +{ + +namespace +{ + +class DetachedTablesBlockSource : public ISource +{ +public: + DetachedTablesBlockSource( + std::vector columns_mask_, + Block header_, + UInt64 max_block_size_, + ColumnPtr databases_, + ColumnPtr detached_tables_, + ContextPtr context_) + : ISource(std::move(header_)) + , columns_mask(std::move(columns_mask_)) + , max_block_size(max_block_size_) + , databases(std::move(databases_)) + , context(Context::createCopy(context_)) + { + size_t size = detached_tables_->size(); + detached_tables.reserve(size); + for (size_t idx = 0; idx < size; ++idx) + { + detached_tables.insert(detached_tables_->getDataAt(idx).toString()); + } + } + + String getName() const override { return "DetachedTables"; } + +protected: + Chunk generate() override + { + if (done) + return {}; + + MutableColumns result_columns = getPort().getHeader().cloneEmptyColumns(); + + const auto access = context->getAccess(); + const bool need_to_check_access_for_databases = !access->isGranted(AccessType::SHOW_TABLES); + + size_t database_idx = 0; + size_t rows_count = 0; + for (; database_idx < databases->size() && rows_count < max_block_size; ++database_idx) + { + database_name = databases->getDataAt(database_idx).toString(); + database = DatabaseCatalog::instance().tryGetDatabase(database_name); + + if (!database) + continue; + + const bool need_to_check_access_for_tables + = need_to_check_access_for_databases && !access->isGranted(AccessType::SHOW_TABLES, database_name); + + if (!detached_tables_it || !detached_tables_it->isValid()) + detached_tables_it = database->getDetachedTablesIterator(context, {}, false); + + for (; rows_count < max_block_size && detached_tables_it->isValid(); detached_tables_it->next()) + { + const auto detached_table_name = detached_tables_it->table(); + + if (!detached_tables.contains(detached_table_name)) + continue; + + if (need_to_check_access_for_tables && !access->isGranted(AccessType::SHOW_TABLES, database_name, detached_table_name)) + continue; + + fillResultColumnsByDetachedTableIterator(result_columns); + ++rows_count; + } + } + + if (databases->size() == database_idx && (!detached_tables_it || !detached_tables_it->isValid())) + { + done = true; + } + const UInt64 num_rows = result_columns.at(0)->size(); + return Chunk(std::move(result_columns), num_rows); + } + +private: + const std::vector columns_mask; + const UInt64 max_block_size; + const ColumnPtr databases; + NameSet detached_tables; + DatabaseDetachedTablesSnapshotIteratorPtr detached_tables_it; + ContextPtr context; + bool done = false; + DatabasePtr database; + std::string database_name; + + void fillResultColumnsByDetachedTableIterator(MutableColumns & result_columns) const + { + size_t src_index = 0; + size_t res_index = 0; + + if (columns_mask[src_index++]) + result_columns[res_index++]->insert(detached_tables_it->database()); + + if (columns_mask[src_index++]) + result_columns[res_index++]->insert(detached_tables_it->table()); + + + if (columns_mask[src_index++]) + result_columns[res_index++]->insert(detached_tables_it->uuid()); + + if (columns_mask[src_index++]) + result_columns[res_index++]->insert(detached_tables_it->metadataPath()); + + if (columns_mask[src_index++]) + result_columns[res_index++]->insert(detached_tables_it->isPermanently()); + } +}; + +} + +class ReadFromSystemDetachedTables : public SourceStepWithFilter +{ +public: + std::string getName() const override { return "ReadFromSystemDetachedTables"; } + void initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) override; + + ReadFromSystemDetachedTables( + const Names & column_names_, + const SelectQueryInfo & query_info_, + const StorageSnapshotPtr & storage_snapshot_, + const ContextPtr & context_, + Block sample_block, + std::vector columns_mask_, + size_t max_block_size_); + + void applyFilters(ActionDAGNodes added_filter_nodes) override; + +private: + std::vector columns_mask; + size_t max_block_size; + + ColumnPtr filtered_databases_column; + ColumnPtr filtered_tables_column; +}; + +StorageSystemDetachedTables::StorageSystemDetachedTables(const StorageID & table_id_) : IStorage(table_id_) +{ + StorageInMemoryMetadata storage_metadata; + + auto description = ColumnsDescription{ + ColumnDescription{"database", std::make_shared(), "The name of the database the table is in."}, + ColumnDescription{"table", std::make_shared(), "Table name."}, + ColumnDescription{"uuid", std::make_shared(), "Table uuid (Atomic database)."}, + ColumnDescription{"metadata_path", std::make_shared(), "Path to the table metadata in the file system."}, + ColumnDescription{"is_permanently", std::make_shared(), "Table was detached permanently."}, + }; + + storage_metadata.setColumns(std::move(description)); + + setInMemoryMetadata(storage_metadata); +} + +void StorageSystemDetachedTables::read( + QueryPlan & query_plan, + const Names & column_names, + const StorageSnapshotPtr & storage_snapshot, + SelectQueryInfo & query_info, + ContextPtr context, + QueryProcessingStage::Enum /*processed_stage*/, + const size_t max_block_size, + size_t /*num_streams*/) +{ + storage_snapshot->check(column_names); + auto sample_block = storage_snapshot->metadata->getSampleBlock(); + + auto [columns_mask, res_block] = getQueriedColumnsMaskAndHeader(sample_block, column_names); + + auto reading = std::make_unique( + column_names, query_info, storage_snapshot, context, std::move(res_block), std::move(columns_mask), max_block_size); + + query_plan.addStep(std::move(reading)); +} + +ReadFromSystemDetachedTables::ReadFromSystemDetachedTables( + const Names & column_names_, + const SelectQueryInfo & query_info_, + const StorageSnapshotPtr & storage_snapshot_, + const ContextPtr & context_, + Block sample_block, + std::vector columns_mask_, + size_t max_block_size_) + : SourceStepWithFilter(DataStream{.header = std::move(sample_block)}, column_names_, query_info_, storage_snapshot_, context_) + , columns_mask(std::move(columns_mask_)) + , max_block_size(max_block_size_) +{ +} + +void ReadFromSystemDetachedTables::applyFilters(ActionDAGNodes added_filter_nodes) +{ + SourceStepWithFilter::applyFilters(std::move(added_filter_nodes)); + + const ActionsDAG::Node * predicate = nullptr; + if (filter_actions_dag) + predicate = filter_actions_dag->getOutputs().at(0); + + filtered_databases_column = detail::getFilteredDatabases(predicate, context); + filtered_tables_column = detail::getFilteredTables(predicate, filtered_databases_column, context, true); +} + +void ReadFromSystemDetachedTables::initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) +{ + auto pipe = Pipe(std::make_shared( + std::move(columns_mask), + getOutputStream().header, + max_block_size, + std::move(filtered_databases_column), + std::move(filtered_tables_column), + context)); + pipeline.init(std::move(pipe)); +} +} diff --git a/src/Storages/System/StorageSystemDetachedTables.h b/src/Storages/System/StorageSystemDetachedTables.h new file mode 100644 index 00000000000..cd042f51eaa --- /dev/null +++ b/src/Storages/System/StorageSystemDetachedTables.h @@ -0,0 +1,32 @@ +#pragma once + +#include + + +namespace DB +{ + +class Context; + +/** Implements the system table `detached_tables`, which allows you to get information about detached tables. + */ +class StorageSystemDetachedTables final : public IStorage +{ +public: + explicit StorageSystemDetachedTables(const StorageID & table_id_); + + std::string getName() const override { return "SystemDetachedTables"; } + + void read( + QueryPlan & query_plan, + const Names & column_names, + const StorageSnapshotPtr & storage_snapshot, + SelectQueryInfo & /*query_info*/, + ContextPtr context, + QueryProcessingStage::Enum processed_stage, + size_t max_block_size, + size_t num_streams) override; + + bool isSystemStorage() const override { return true; } +}; +} diff --git a/src/Storages/System/StorageSystemErrors.cpp b/src/Storages/System/StorageSystemErrors.cpp index 2da268305f8..2955aebeb7e 100644 --- a/src/Storages/System/StorageSystemErrors.cpp +++ b/src/Storages/System/StorageSystemErrors.cpp @@ -5,6 +5,7 @@ #include #include #include +#include namespace DB diff --git a/src/Storages/System/StorageSystemEvents.cpp b/src/Storages/System/StorageSystemEvents.cpp index 822d5c77788..f74c316e3fc 100644 --- a/src/Storages/System/StorageSystemEvents.cpp +++ b/src/Storages/System/StorageSystemEvents.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include diff --git a/src/Storages/System/StorageSystemFunctions.cpp b/src/Storages/System/StorageSystemFunctions.cpp index 6e4ac8b2747..f10ce9e3987 100644 --- a/src/Storages/System/StorageSystemFunctions.cpp +++ b/src/Storages/System/StorageSystemFunctions.cpp @@ -16,16 +16,6 @@ namespace DB { -namespace ErrorCodes -{ - extern const int DICTIONARIES_WAS_NOT_LOADED; - extern const int FUNCTION_NOT_ALLOWED; - extern const int NOT_IMPLEMENTED; - extern const int SUPPORT_IS_DISABLED; - extern const int ACCESS_DENIED; - extern const int DEPRECATED_FUNCTION; -}; - enum class FunctionOrigin : int8_t { SYSTEM = 0, @@ -40,7 +30,6 @@ namespace MutableColumns & res_columns, const String & name, UInt64 is_aggregate, - std::optional is_deterministic, const String & create_query, FunctionOrigin function_origin, const Factory & factory) @@ -48,58 +37,53 @@ namespace res_columns[0]->insert(name); res_columns[1]->insert(is_aggregate); - if (!is_deterministic.has_value()) - res_columns[2]->insertDefault(); - else - res_columns[2]->insert(*is_deterministic); - if constexpr (std::is_same_v || std::is_same_v) { - res_columns[3]->insert(false); - res_columns[4]->insertDefault(); + res_columns[2]->insert(false); + res_columns[3]->insertDefault(); } else { - res_columns[3]->insert(factory.isCaseInsensitive(name)); + res_columns[2]->insert(factory.isCaseInsensitive(name)); if (factory.isAlias(name)) - res_columns[4]->insert(factory.aliasTo(name)); + res_columns[3]->insert(factory.aliasTo(name)); else - res_columns[4]->insertDefault(); + res_columns[3]->insertDefault(); } - res_columns[5]->insert(create_query); - res_columns[6]->insert(static_cast(function_origin)); + res_columns[4]->insert(create_query); + res_columns[5]->insert(static_cast(function_origin)); if constexpr (std::is_same_v) { if (factory.isAlias(name)) { + res_columns[6]->insertDefault(); res_columns[7]->insertDefault(); res_columns[8]->insertDefault(); res_columns[9]->insertDefault(); res_columns[10]->insertDefault(); res_columns[11]->insertDefault(); - res_columns[12]->insertDefault(); } else { auto documentation = factory.getDocumentation(name); - res_columns[7]->insert(documentation.description); - res_columns[8]->insert(documentation.syntax); - res_columns[9]->insert(documentation.argumentsAsString()); - res_columns[10]->insert(documentation.returned_value); - res_columns[11]->insert(documentation.examplesAsString()); - res_columns[12]->insert(documentation.categoriesAsString()); + res_columns[6]->insert(documentation.description); + res_columns[7]->insert(documentation.syntax); + res_columns[8]->insert(documentation.argumentsAsString()); + res_columns[9]->insert(documentation.returned_value); + res_columns[10]->insert(documentation.examplesAsString()); + res_columns[11]->insert(documentation.categoriesAsString()); } } else { + res_columns[6]->insertDefault(); res_columns[7]->insertDefault(); res_columns[8]->insertDefault(); res_columns[9]->insertDefault(); res_columns[10]->insertDefault(); res_columns[11]->insertDefault(); - res_columns[12]->insertDefault(); } } } @@ -120,7 +104,6 @@ ColumnsDescription StorageSystemFunctions::getColumnsDescription() { {"name", std::make_shared(), "The name of the function."}, {"is_aggregate", std::make_shared(), "Whether the function is an aggregate function."}, - {"is_deterministic", std::make_shared(std::make_shared()), "Whether the function is deterministic."}, {"case_insensitive", std::make_shared(), "Whether the function name can be used case-insensitively."}, {"alias_to", std::make_shared(), "The original function name, if the function name is an alias."}, {"create_query", std::make_shared(), "Obsolete."}, @@ -140,36 +123,14 @@ void StorageSystemFunctions::fillData(MutableColumns & res_columns, ContextPtr c const auto & function_names = functions_factory.getAllRegisteredNames(); for (const auto & function_name : function_names) { - std::optional is_deterministic; - try - { - DO_NOT_UPDATE_ERROR_STATISTICS(); - is_deterministic = functions_factory.tryGet(function_name, context)->isDeterministic(); - } - catch (const Exception & e) - { - /// Some functions throw because they need special configuration or setup before use. - if (e.code() == ErrorCodes::DICTIONARIES_WAS_NOT_LOADED - || e.code() == ErrorCodes::FUNCTION_NOT_ALLOWED - || e.code() == ErrorCodes::NOT_IMPLEMENTED - || e.code() == ErrorCodes::SUPPORT_IS_DISABLED - || e.code() == ErrorCodes::ACCESS_DENIED - || e.code() == ErrorCodes::DEPRECATED_FUNCTION) - { - /// Ignore exception, show is_deterministic = NULL. - } - else - throw; - } - - fillRow(res_columns, function_name, 0, is_deterministic, "", FunctionOrigin::SYSTEM, functions_factory); + fillRow(res_columns, function_name, 0, "", FunctionOrigin::SYSTEM, functions_factory); } const auto & aggregate_functions_factory = AggregateFunctionFactory::instance(); const auto & aggregate_function_names = aggregate_functions_factory.getAllRegisteredNames(); for (const auto & function_name : aggregate_function_names) { - fillRow(res_columns, function_name, 1, {1}, "", FunctionOrigin::SYSTEM, aggregate_functions_factory); + fillRow(res_columns, function_name, 1, "", FunctionOrigin::SYSTEM, aggregate_functions_factory); } const auto & user_defined_sql_functions_factory = UserDefinedSQLFunctionFactory::instance(); @@ -177,14 +138,14 @@ void StorageSystemFunctions::fillData(MutableColumns & res_columns, ContextPtr c for (const auto & function_name : user_defined_sql_functions_names) { auto create_query = queryToString(user_defined_sql_functions_factory.get(function_name)); - fillRow(res_columns, function_name, 0, {0}, create_query, FunctionOrigin::SQL_USER_DEFINED, user_defined_sql_functions_factory); + fillRow(res_columns, function_name, 0, create_query, FunctionOrigin::SQL_USER_DEFINED, user_defined_sql_functions_factory); } const auto & user_defined_executable_functions_factory = UserDefinedExecutableFunctionFactory::instance(); const auto & user_defined_executable_functions_names = user_defined_executable_functions_factory.getRegisteredNames(context); /// NOLINT(readability-static-accessed-through-instance) for (const auto & function_name : user_defined_executable_functions_names) { - fillRow(res_columns, function_name, 0, {0}, "", FunctionOrigin::EXECUTABLE_USER_DEFINED, user_defined_executable_functions_factory); + fillRow(res_columns, function_name, 0, "", FunctionOrigin::EXECUTABLE_USER_DEFINED, user_defined_executable_functions_factory); } } diff --git a/src/Storages/System/StorageSystemMergeTreeSettings.cpp b/src/Storages/System/StorageSystemMergeTreeSettings.cpp index 7781e3789a4..18730010e53 100644 --- a/src/Storages/System/StorageSystemMergeTreeSettings.cpp +++ b/src/Storages/System/StorageSystemMergeTreeSettings.cpp @@ -1,7 +1,9 @@ +#include #include #include #include #include +#include #include #include diff --git a/src/Storages/System/StorageSystemMergeTreeSettings.h b/src/Storages/System/StorageSystemMergeTreeSettings.h index e2913a7e55b..4d4150dcec2 100644 --- a/src/Storages/System/StorageSystemMergeTreeSettings.h +++ b/src/Storages/System/StorageSystemMergeTreeSettings.h @@ -1,6 +1,5 @@ #pragma once -#include #include diff --git a/src/Storages/System/StorageSystemPartsBase.cpp b/src/Storages/System/StorageSystemPartsBase.cpp index 175c0834bcb..f7d1c1b3eb8 100644 --- a/src/Storages/System/StorageSystemPartsBase.cpp +++ b/src/Storages/System/StorageSystemPartsBase.cpp @@ -274,7 +274,7 @@ void ReadFromSystemPartsBase::applyFilters(ActionDAGNodes added_filter_nodes) filter_by_database = VirtualColumnUtils::splitFilterDagForAllowedInputs(predicate, &block); if (filter_by_database) - VirtualColumnUtils::buildSetsForDAG(filter_by_database, context); + VirtualColumnUtils::buildSetsForDAG(*filter_by_database, context); block.insert(ColumnWithTypeAndName({}, std::make_shared(), table_column_name)); block.insert(ColumnWithTypeAndName({}, std::make_shared(), engine_column_name)); @@ -283,7 +283,7 @@ void ReadFromSystemPartsBase::applyFilters(ActionDAGNodes added_filter_nodes) filter_by_other_columns = VirtualColumnUtils::splitFilterDagForAllowedInputs(predicate, &block); if (filter_by_other_columns) - VirtualColumnUtils::buildSetsForDAG(filter_by_other_columns, context); + VirtualColumnUtils::buildSetsForDAG(*filter_by_other_columns, context); } } diff --git a/src/Storages/System/StorageSystemPartsBase.h b/src/Storages/System/StorageSystemPartsBase.h index be945162b39..8671fd850f8 100644 --- a/src/Storages/System/StorageSystemPartsBase.h +++ b/src/Storages/System/StorageSystemPartsBase.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include #include diff --git a/src/Storages/System/StorageSystemRemoteDataPaths.cpp b/src/Storages/System/StorageSystemRemoteDataPaths.cpp index a5c496db7e7..00b4824cc3d 100644 --- a/src/Storages/System/StorageSystemRemoteDataPaths.cpp +++ b/src/Storages/System/StorageSystemRemoteDataPaths.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/System/StorageSystemS3Queue.cpp b/src/Storages/System/StorageSystemS3Queue.cpp index 131570709d0..a1c9380f616 100644 --- a/src/Storages/System/StorageSystemS3Queue.cpp +++ b/src/Storages/System/StorageSystemS3Queue.cpp @@ -11,9 +11,9 @@ #include #include #include -#include -#include -#include +#include +#include +#include #include @@ -32,7 +32,6 @@ ColumnsDescription StorageSystemS3Queue::getColumnsDescription() {"status", std::make_shared(), "Status of processing: Processed, Processing, Failed"}, {"processing_start_time", std::make_shared(std::make_shared()), "Time at which processing of the file started"}, {"processing_end_time", std::make_shared(std::make_shared()), "Time at which processing of the file ended"}, - {"ProfileEvents", std::make_shared(std::make_shared(), std::make_shared()), "Profile events collected during processing of the file"}, {"exception", std::make_shared(), "Exception which happened during processing"}, }; } @@ -44,7 +43,7 @@ StorageSystemS3Queue::StorageSystemS3Queue(const StorageID & table_id_) void StorageSystemS3Queue::fillData(MutableColumns & res_columns, ContextPtr, const ActionsDAG::Node *, std::vector) const { - for (const auto & [zookeeper_path, metadata] : S3QueueMetadataFactory::instance().getAll()) + for (const auto & [zookeeper_path, metadata] : ObjectStorageQueueMetadataFactory::instance().getAll()) { for (const auto & [file_path, file_status] : metadata->getFileStatuses()) { @@ -65,8 +64,6 @@ void StorageSystemS3Queue::fillData(MutableColumns & res_columns, ContextPtr, co else res_columns[i++]->insertDefault(); - ProfileEvents::dumpToMapColumn(file_status->profile_counters.getPartiallyAtomicSnapshot(), res_columns[i++].get(), true); - res_columns[i++]->insert(file_status->getException()); } } diff --git a/src/Storages/System/StorageSystemSettings.cpp b/src/Storages/System/StorageSystemSettings.cpp index b437108b00e..f1e9d542fec 100644 --- a/src/Storages/System/StorageSystemSettings.cpp +++ b/src/Storages/System/StorageSystemSettings.cpp @@ -1,4 +1,6 @@ #include + +#include #include #include #include diff --git a/src/Storages/System/StorageSystemSettingsChanges.cpp b/src/Storages/System/StorageSystemSettingsChanges.cpp index de47ec52031..d6c83870741 100644 --- a/src/Storages/System/StorageSystemSettingsChanges.cpp +++ b/src/Storages/System/StorageSystemSettingsChanges.cpp @@ -26,6 +26,7 @@ ColumnsDescription StorageSystemSettingsChanges::getColumnsDescription() void StorageSystemSettingsChanges::fillData(MutableColumns & res_columns, ContextPtr, const ActionsDAG::Node *, std::vector) const { + const auto & settings_changes_history = getSettingsChangesHistory(); for (auto it = settings_changes_history.rbegin(); it != settings_changes_history.rend(); ++it) { res_columns[0]->insert(it->first.toString()); diff --git a/src/Storages/System/StorageSystemSettingsProfileElements.cpp b/src/Storages/System/StorageSystemSettingsProfileElements.cpp index 2af3e6dfd05..40a669351b1 100644 --- a/src/Storages/System/StorageSystemSettingsProfileElements.cpp +++ b/src/Storages/System/StorageSystemSettingsProfileElements.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/System/StorageSystemStackTrace.cpp b/src/Storages/System/StorageSystemStackTrace.cpp index ba7433fb9ae..cc221bbea69 100644 --- a/src/Storages/System/StorageSystemStackTrace.cpp +++ b/src/Storages/System/StorageSystemStackTrace.cpp @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include diff --git a/src/Storages/System/StorageSystemTables.cpp b/src/Storages/System/StorageSystemTables.cpp index 783b899c978..049a442d494 100644 --- a/src/Storages/System/StorageSystemTables.cpp +++ b/src/Storages/System/StorageSystemTables.cpp @@ -1,29 +1,30 @@ +#include + +#include #include -#include +#include #include #include -#include -#include -#include -#include -#include -#include +#include +#include +#include +#include #include #include #include #include #include -#include -#include -#include -#include -#include #include #include #include #include #include -#include +#include +#include +#include +#include +#include +#include #include @@ -31,6 +32,105 @@ namespace DB { +namespace +{ + +/// Avoid heavy operation on tables if we only queried columns that we can get without table object. +/// Otherwise it will require table initialization for Lazy database. +bool needTable(const DatabasePtr & database, const Block & header) +{ + if (database->getEngineName() != "Lazy") + return true; + + static const std::set columns_without_table = {"database", "name", "uuid", "metadata_modification_time"}; + for (const auto & column : header.getColumnsWithTypeAndName()) + { + if (columns_without_table.find(column.name) == columns_without_table.end()) + return true; + } + return false; +} +} + +namespace detail +{ +ColumnPtr getFilteredDatabases(const ActionsDAG::Node * predicate, ContextPtr context) +{ + MutableColumnPtr column = ColumnString::create(); + + const auto databases = DatabaseCatalog::instance().getDatabases(); + for (const auto & database_name : databases | boost::adaptors::map_keys) + { + if (database_name == DatabaseCatalog::TEMPORARY_DATABASE) + continue; /// We don't want to show the internal database for temporary tables in system.tables + + column->insert(database_name); + } + + Block block{ColumnWithTypeAndName(std::move(column), std::make_shared(), "database")}; + VirtualColumnUtils::filterBlockWithPredicate(predicate, block, context); + return block.getByPosition(0).column; +} + +ColumnPtr getFilteredTables( + const ActionsDAG::Node * predicate, const ColumnPtr & filtered_databases_column, ContextPtr context, const bool is_detached) +{ + Block sample{ + ColumnWithTypeAndName(nullptr, std::make_shared(), "name"), + ColumnWithTypeAndName(nullptr, std::make_shared(), "engine")}; + + MutableColumnPtr database_column = ColumnString::create(); + MutableColumnPtr engine_column; + + auto dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(predicate, &sample); + if (dag) + { + bool filter_by_engine = false; + for (const auto * input : dag->getInputs()) + if (input->result_name == "engine") + filter_by_engine = true; + + if (filter_by_engine) + engine_column = ColumnString::create(); + } + + for (size_t database_idx = 0; database_idx < filtered_databases_column->size(); ++database_idx) + { + const auto & database_name = filtered_databases_column->getDataAt(database_idx).toString(); + DatabasePtr database = DatabaseCatalog::instance().tryGetDatabase(database_name); + if (!database) + continue; + + if (is_detached) + { + auto table_it = database->getDetachedTablesIterator(context, {}, false); + for (; table_it->isValid(); table_it->next()) + { + database_column->insert(table_it->table()); + } + } + else + { + for (auto table_it = database->getTablesIterator(context); table_it->isValid(); table_it->next()) + { + database_column->insert(table_it->name()); + if (engine_column) + engine_column->insert(table_it->table()->getName()); + } + } + } + + Block block{ColumnWithTypeAndName(std::move(database_column), std::make_shared(), "name")}; + if (engine_column) + block.insert(ColumnWithTypeAndName(std::move(engine_column), std::make_shared(), "engine")); + + if (dag) + VirtualColumnUtils::filterBlockWithDAG(dag, block, context); + + return block.getByPosition(0).column; +} + +} StorageSystemTables::StorageSystemTables(const StorageID & table_id_) : IStorage(table_id_) @@ -105,92 +205,6 @@ StorageSystemTables::StorageSystemTables(const StorageID & table_id_) setInMemoryMetadata(storage_metadata); } - -namespace -{ - -ColumnPtr getFilteredDatabases(const ActionsDAG::Node * predicate, ContextPtr context) -{ - MutableColumnPtr column = ColumnString::create(); - - const auto databases = DatabaseCatalog::instance().getDatabases(); - for (const auto & database_name : databases | boost::adaptors::map_keys) - { - if (database_name == DatabaseCatalog::TEMPORARY_DATABASE) - continue; /// We don't want to show the internal database for temporary tables in system.tables - - column->insert(database_name); - } - - Block block { ColumnWithTypeAndName(std::move(column), std::make_shared(), "database") }; - VirtualColumnUtils::filterBlockWithPredicate(predicate, block, context); - return block.getByPosition(0).column; -} - -ColumnPtr getFilteredTables(const ActionsDAG::Node * predicate, const ColumnPtr & filtered_databases_column, ContextPtr context) -{ - Block sample { - ColumnWithTypeAndName(nullptr, std::make_shared(), "name"), - ColumnWithTypeAndName(nullptr, std::make_shared(), "engine") - }; - - MutableColumnPtr database_column = ColumnString::create(); - MutableColumnPtr engine_column; - - auto dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(predicate, &sample); - if (dag) - { - bool filter_by_engine = false; - for (const auto * input : dag->getInputs()) - if (input->result_name == "engine") - filter_by_engine = true; - - if (filter_by_engine) - engine_column = ColumnString::create(); - } - - for (size_t database_idx = 0; database_idx < filtered_databases_column->size(); ++database_idx) - { - const auto & database_name = filtered_databases_column->getDataAt(database_idx).toString(); - DatabasePtr database = DatabaseCatalog::instance().tryGetDatabase(database_name); - if (!database) - continue; - - for (auto table_it = database->getTablesIterator(context); table_it->isValid(); table_it->next()) - { - database_column->insert(table_it->name()); - if (engine_column) - engine_column->insert(table_it->table()->getName()); - } - } - - Block block {ColumnWithTypeAndName(std::move(database_column), std::make_shared(), "name")}; - if (engine_column) - block.insert(ColumnWithTypeAndName(std::move(engine_column), std::make_shared(), "engine")); - - if (dag) - VirtualColumnUtils::filterBlockWithDAG(dag, block, context); - - return block.getByPosition(0).column; -} - -/// Avoid heavy operation on tables if we only queried columns that we can get without table object. -/// Otherwise it will require table initialization for Lazy database. -bool needTable(const DatabasePtr & database, const Block & header) -{ - if (database->getEngineName() != "Lazy") - return true; - - static const std::set columns_without_table = { "database", "name", "uuid", "metadata_modification_time" }; - for (const auto & column : header.getColumnsWithTypeAndName()) - { - if (columns_without_table.find(column.name) == columns_without_table.end()) - return true; - } - return false; -} - - class TablesBlockSource : public ISource { public: @@ -470,7 +484,8 @@ protected: if (ast_create && !context->getSettingsRef().show_table_uuid_in_table_create_query_if_not_nil) { ast_create->uuid = UUIDHelpers::Nil; - ast_create->to_inner_uuid = UUIDHelpers::Nil; + if (ast_create->targets) + ast_create->targets->resetInnerUUIDs(); } if (columns_mask[src_index++]) @@ -690,8 +705,6 @@ private: std::string database_name; }; -} - class ReadFromSystemTables : public SourceStepWithFilter { public: @@ -756,8 +769,8 @@ void ReadFromSystemTables::applyFilters(ActionDAGNodes added_filter_nodes) if (filter_actions_dag) predicate = filter_actions_dag->getOutputs().at(0); - filtered_databases_column = getFilteredDatabases(predicate, context); - filtered_tables_column = getFilteredTables(predicate, filtered_databases_column, context); + filtered_databases_column = detail::getFilteredDatabases(predicate, context); + filtered_tables_column = detail::getFilteredTables(predicate, filtered_databases_column, context, false); } void ReadFromSystemTables::initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) diff --git a/src/Storages/System/StorageSystemTables.h b/src/Storages/System/StorageSystemTables.h index 87cdf1b6a8e..d6e3996b8e3 100644 --- a/src/Storages/System/StorageSystemTables.h +++ b/src/Storages/System/StorageSystemTables.h @@ -8,6 +8,15 @@ namespace DB class Context; +namespace detail +{ + +ColumnPtr getFilteredDatabases(const ActionsDAG::Node * predicate, ContextPtr context); +ColumnPtr +getFilteredTables(const ActionsDAG::Node * predicate, const ColumnPtr & filtered_databases_column, ContextPtr context, bool is_detached); + +} + /** Implements the system table `tables`, which allows you to get information about all tables. */ diff --git a/src/Storages/System/StorageSystemUsers.cpp b/src/Storages/System/StorageSystemUsers.cpp index 0c34f04844d..b4d83058c82 100644 --- a/src/Storages/System/StorageSystemUsers.cpp +++ b/src/Storages/System/StorageSystemUsers.cpp @@ -17,6 +17,13 @@ #include #include #include +#include + +#include + +#include +#include + #include @@ -142,10 +149,19 @@ void StorageSystemUsers::fillData(MutableColumns & res_columns, ContextPtr conte } else if (auth_data.getType() == AuthenticationType::SSL_CERTIFICATE) { - Poco::JSON::Array::Ptr arr = new Poco::JSON::Array(); - for (const auto & common_name : auth_data.getSSLCertificateCommonNames()) - arr->add(common_name); - auth_params_json.set("common_names", arr); + Poco::JSON::Array::Ptr common_names = new Poco::JSON::Array(); + Poco::JSON::Array::Ptr subject_alt_names = new Poco::JSON::Array(); + + const auto & subjects = auth_data.getSSLCertificateSubjects(); + for (const String & subject : subjects.at(SSLCertificateSubjects::Type::CN)) + common_names->add(subject); + for (const String & subject : subjects.at(SSLCertificateSubjects::Type::SAN)) + subject_alt_names->add(subject); + + if (common_names->size() > 0) + auth_params_json.set("common_names", common_names); + if (subject_alt_names->size() > 0) + auth_params_json.set("subject_alt_names", subject_alt_names); } std::ostringstream oss; // STYLE_CHECK_ALLOW_STD_STRING_STREAM diff --git a/src/Storages/System/StorageSystemZeros.cpp b/src/Storages/System/StorageSystemZeros.cpp index 09a2bb5d963..0720a2f24d9 100644 --- a/src/Storages/System/StorageSystemZeros.cpp +++ b/src/Storages/System/StorageSystemZeros.cpp @@ -109,8 +109,8 @@ Pipe StorageSystemZeros::read( storage_snapshot->check(column_names); UInt64 query_limit = limit ? *limit : 0; - if (query_info.limit) - query_limit = query_limit ? std::min(query_limit, query_info.limit) : query_info.limit; + if (query_info.trivial_limit) + query_limit = query_limit ? std::min(query_limit, query_info.trivial_limit) : query_info.trivial_limit; if (query_limit && query_limit < max_block_size) max_block_size = query_limit; diff --git a/src/Storages/System/StorageSystemZooKeeper.cpp b/src/Storages/System/StorageSystemZooKeeper.cpp index cb46cd19517..81d3b0fe659 100644 --- a/src/Storages/System/StorageSystemZooKeeper.cpp +++ b/src/Storages/System/StorageSystemZooKeeper.cpp @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -119,7 +120,7 @@ public: ZooKeeperSink(const Block & header, ContextPtr context) : SinkToStorage(header), zookeeper(context->getZooKeeper()) { } String getName() const override { return "ZooKeeperSink"; } - void consume(Chunk chunk) override + void consume(Chunk & chunk) override { auto block = getHeader().cloneWithColumns(chunk.getColumns()); size_t rows = block.rows(); diff --git a/src/Storages/System/StorageSystemZooKeeperConnection.cpp b/src/Storages/System/StorageSystemZooKeeperConnection.cpp index 950e20512c0..ec29b84dac3 100644 --- a/src/Storages/System/StorageSystemZooKeeperConnection.cpp +++ b/src/Storages/System/StorageSystemZooKeeperConnection.cpp @@ -36,7 +36,8 @@ ColumnsDescription StorageSystemZooKeeperConnection::getColumnsDescription() /* 9 */ {"xid", std::make_shared(), "XID of the current session."}, /* 10*/ {"enabled_feature_flags", std::make_shared(std::move(feature_flags_enum)), "Feature flags which are enabled. Only applicable to ClickHouse Keeper." - } + }, + /* 11*/ {"availability_zone", std::make_shared(), "Availability zone"}, }; } @@ -85,6 +86,7 @@ void StorageSystemZooKeeperConnection::fillData(MutableColumns & res_columns, Co columns[8]->insert(zookeeper->getClientID()); columns[9]->insert(zookeeper->getConnectionXid()); add_enabled_feature_flags(zookeeper); + columns[11]->insert(zookeeper->getConnectedHostAvailabilityZone()); } }; diff --git a/src/Storages/System/attachSystemTables.cpp b/src/Storages/System/attachSystemTables.cpp index 6ff86b26ca9..97eda1db3fa 100644 --- a/src/Storages/System/attachSystemTables.cpp +++ b/src/Storages/System/attachSystemTables.cpp @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -129,6 +130,7 @@ void attachSystemTablesServer(ContextPtr context, IDatabase & system_database, b attachNoDescription(context, system_database, "zeros_mt", "Multithreaded version of system.zeros.", true); attach(context, system_database, "databases", "Lists all databases of the current server."); attachNoDescription(context, system_database, "tables", "Lists all tables of the current server."); + attachNoDescription(context, system_database, "detached_tables", "Lists all detached tables of the current server."); attachNoDescription(context, system_database, "columns", "Lists all columns from all tables of the current server."); attach(context, system_database, "functions", "Contains a list of all available ordinary and aggregate functions with their descriptions."); attach(context, system_database, "events", "Contains profiling events and their current value."); diff --git a/src/Storages/TTLDescription.cpp b/src/Storages/TTLDescription.cpp index f831465277d..92f6f17583d 100644 --- a/src/Storages/TTLDescription.cpp +++ b/src/Storages/TTLDescription.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include @@ -171,7 +172,7 @@ static ExpressionAndSets buildExpressionAndSets(ASTPtr & ast, const NamesAndType /// with subqueries it's possible that new analyzer will be enabled in ::read method /// of underlying storage when all other parts of infra are not ready for it /// (built with old analyzer). - context_copy->setSetting("allow_experimental_analyzer", Field{0}); + context_copy->setSetting("allow_experimental_analyzer", false); auto syntax_analyzer_result = TreeRewriter(context_copy).analyze(ast, columns); ExpressionAnalyzer analyzer(ast, syntax_analyzer_result, context_copy); auto dag = analyzer.getActionsDAG(false); diff --git a/src/Storages/VirtualColumnUtils.cpp b/src/Storages/VirtualColumnUtils.cpp index 778c9e13adb..151079154b1 100644 --- a/src/Storages/VirtualColumnUtils.cpp +++ b/src/Storages/VirtualColumnUtils.cpp @@ -54,9 +54,9 @@ namespace DB namespace VirtualColumnUtils { -void buildSetsForDAG(const ActionsDAGPtr & dag, const ContextPtr & context) +void buildSetsForDAG(const ActionsDAG & dag, const ContextPtr & context) { - for (const auto & node : dag->getNodes()) + for (const auto & node : dag.getNodes()) { if (node.type == ActionsDAG::ActionType::COLUMN) { @@ -79,7 +79,7 @@ void buildSetsForDAG(const ActionsDAGPtr & dag, const ContextPtr & context) void filterBlockWithDAG(ActionsDAGPtr dag, Block & block, ContextPtr context) { - buildSetsForDAG(dag, context); + buildSetsForDAG(*dag, context); auto actions = std::make_shared(dag); Block block_with_filter = block; actions->execute(block_with_filter, /*dry_run=*/ false, /*allow_duplicates_in_input=*/ true); @@ -271,7 +271,8 @@ bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node) static const ActionsDAG::Node * splitFilterNodeForAllowedInputs( const ActionsDAG::Node * node, const Block * allowed_inputs, - ActionsDAG::Nodes & additional_nodes) + ActionsDAG::Nodes & additional_nodes, + bool allow_non_deterministic_functions) { if (node->type == ActionsDAG::ActionType::FUNCTION) { @@ -280,8 +281,14 @@ static const ActionsDAG::Node * splitFilterNodeForAllowedInputs( auto & node_copy = additional_nodes.emplace_back(*node); node_copy.children.clear(); for (const auto * child : node->children) - if (const auto * child_copy = splitFilterNodeForAllowedInputs(child, allowed_inputs, additional_nodes)) + if (const auto * child_copy = splitFilterNodeForAllowedInputs(child, allowed_inputs, additional_nodes, allow_non_deterministic_functions)) node_copy.children.push_back(child_copy); + /// Expression like (now_allowed AND allowed) is not allowed if allow_non_deterministic_functions = true. This is important for + /// trivial count optimization, otherwise we can get incorrect results. For example, if the query is + /// SELECT count() FROM table WHERE _partition_id = '0' AND rowNumberInBlock() = 1, we cannot apply + /// trivial count. + else if (!allow_non_deterministic_functions) + return nullptr; if (node_copy.children.empty()) return nullptr; @@ -307,7 +314,7 @@ static const ActionsDAG::Node * splitFilterNodeForAllowedInputs( { auto & node_copy = additional_nodes.emplace_back(*node); for (auto & child : node_copy.children) - if (child = splitFilterNodeForAllowedInputs(child, allowed_inputs, additional_nodes); !child) + if (child = splitFilterNodeForAllowedInputs(child, allowed_inputs, additional_nodes, allow_non_deterministic_functions); !child) return nullptr; return &node_copy; @@ -321,7 +328,7 @@ static const ActionsDAG::Node * splitFilterNodeForAllowedInputs( auto index_hint_dag = index_hint->getActions()->clone(); ActionsDAG::NodeRawConstPtrs atoms; for (const auto & output : index_hint_dag->getOutputs()) - if (const auto * child_copy = splitFilterNodeForAllowedInputs(output, allowed_inputs, additional_nodes)) + if (const auto * child_copy = splitFilterNodeForAllowedInputs(output, allowed_inputs, additional_nodes, allow_non_deterministic_functions)) atoms.push_back(child_copy); if (!atoms.empty()) @@ -355,13 +362,13 @@ static const ActionsDAG::Node * splitFilterNodeForAllowedInputs( return node; } -ActionsDAGPtr splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs) +ActionsDAGPtr splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs, bool allow_non_deterministic_functions) { if (!predicate) return nullptr; ActionsDAG::Nodes additional_nodes; - const auto * res = splitFilterNodeForAllowedInputs(predicate, allowed_inputs, additional_nodes); + const auto * res = splitFilterNodeForAllowedInputs(predicate, allowed_inputs, additional_nodes, allow_non_deterministic_functions); if (!res) return nullptr; @@ -370,7 +377,7 @@ ActionsDAGPtr splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, void filterBlockWithPredicate(const ActionsDAG::Node * predicate, Block & block, ContextPtr context) { - auto dag = splitFilterDagForAllowedInputs(predicate, &block); + auto dag = splitFilterDagForAllowedInputs(predicate, &block, /*allow_non_deterministic_functions=*/ false); if (dag) filterBlockWithDAG(dag, block, context); } diff --git a/src/Storages/VirtualColumnUtils.h b/src/Storages/VirtualColumnUtils.h index fbfbdd6c6cc..e5cfa47c8f6 100644 --- a/src/Storages/VirtualColumnUtils.h +++ b/src/Storages/VirtualColumnUtils.h @@ -26,13 +26,21 @@ void filterBlockWithPredicate(const ActionsDAG::Node * predicate, Block & block, void filterBlockWithDAG(ActionsDAGPtr dag, Block & block, ContextPtr context); /// Builds sets used by ActionsDAG inplace. -void buildSetsForDAG(const ActionsDAGPtr & dag, const ContextPtr & context); +void buildSetsForDAG(const ActionsDAG & dag, const ContextPtr & context); /// Recursively checks if all functions used in DAG are deterministic in scope of query. bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node); /// Extract a part of predicate that can be evaluated using only columns from input_names. -ActionsDAGPtr splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs); +/// When allow_non_deterministic_functions is true then even if the predicate contains non-deterministic +/// functions, we still allow to extract a part of the predicate, otherwise we return nullptr. +/// allow_non_deterministic_functions must be false when we are going to use the result to filter parts in +/// MergeTreeData::totalRowsByPartitionPredicateImp. For example, if the query is +/// `SELECT count() FROM table WHERE _partition_id = '0' AND rowNumberInBlock() = 1` +/// The predicate will be `_partition_id = '0' AND rowNumberInBlock() = 1`, and `rowNumberInBlock()` is +/// non-deterministic. If we still extract the part `_partition_id = '0'` for filtering parts, then trivial +/// count optimization will be mistakenly applied to the query. +ActionsDAGPtr splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs, bool allow_non_deterministic_functions = true); /// Extract from the input stream a set of `name` column values template diff --git a/src/Storages/WindowView/StorageWindowView.cpp b/src/Storages/WindowView/StorageWindowView.cpp index 77e6ee9cb24..8975f9327b1 100644 --- a/src/Storages/WindowView/StorageWindowView.cpp +++ b/src/Storages/WindowView/StorageWindowView.cpp @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include @@ -48,6 +49,7 @@ #include #include #include +#include #include #include @@ -304,7 +306,7 @@ namespace public: explicit AddingAggregatedChunkInfoTransform(Block header) : ISimpleTransform(header, header, false) { } - void transform(Chunk & chunk) override { chunk.setChunkInfo(std::make_shared()); } + void transform(Chunk & chunk) override { chunk.getChunkInfos().add(std::make_shared()); } String getName() const override { return "AddingAggregatedChunkInfoTransform"; } }; @@ -689,7 +691,13 @@ inline void StorageWindowView::fire(UInt32 watermark) StoragePtr target_table = getTargetTable(); auto insert = std::make_shared(); insert->table_id = target_table->getStorageID(); - InterpreterInsertQuery interpreter(insert, getContext()); + InterpreterInsertQuery interpreter( + insert, + getContext(), + /* allow_materialized */ false, + /* no_squash */ false, + /* no_destination */ false, + /* async_isnert */ false); auto block_io = interpreter.execute(); auto pipe = Pipe(std::make_shared(blocks, header)); @@ -1180,6 +1188,7 @@ StorageWindowView::StorageWindowView( ContextPtr context_, const ASTCreateQuery & query, const ColumnsDescription & columns_, + const String & comment, LoadingStrictnessLevel mode) : IStorage(table_id_) , WithContext(context_->getGlobalContext()) @@ -1198,11 +1207,15 @@ StorageWindowView::StorageWindowView( StorageInMemoryMetadata storage_metadata; storage_metadata.setColumns(columns_); + storage_metadata.setComment(comment); setInMemoryMetadata(storage_metadata); /// If the target table is not set, use inner target table - has_inner_target_table = query.to_table_id.empty(); - if (has_inner_target_table && !query.storage) + auto to_table_id = query.getTargetTableID(ViewTarget::To); + has_inner_target_table = to_table_id.empty(); + auto to_table_engine = query.getTargetInnerEngine(ViewTarget::To); + + if (has_inner_target_table && !to_table_engine) throw Exception(ErrorCodes::INCORRECT_QUERY, "You must specify where to save results of a WindowView query: " "either ENGINE or an existing table in a TO clause"); @@ -1217,12 +1230,12 @@ StorageWindowView::StorageWindowView( auto inner_query = initInnerQuery(query.select->list_of_selects->children.at(0)->as(), context_); - if (query.inner_storage) - inner_table_engine = query.inner_storage->clone(); + if (auto inner_storage = query.getTargetInnerEngine(ViewTarget::Inner)) + inner_table_engine = inner_storage->clone(); inner_table_id = StorageID(getStorageID().database_name, generateInnerTableName(getStorageID())); inner_fetch_query = generateInnerFetchQuery(inner_table_id); - target_table_id = has_inner_target_table ? StorageID(table_id_.database_name, generateTargetTableName(table_id_)) : query.to_table_id; + target_table_id = has_inner_target_table ? StorageID(table_id_.database_name, generateTargetTableName(table_id_)) : to_table_id; if (is_proctime) next_fire_signal = getWindowUpperBound(now()); @@ -1247,7 +1260,7 @@ StorageWindowView::StorageWindowView( new_columns_list->set(new_columns_list->columns, query.columns_list->columns->ptr()); target_create_query->set(target_create_query->columns_list, new_columns_list); - target_create_query->set(target_create_query->storage, query.storage->ptr()); + target_create_query->set(target_create_query->storage, to_table_engine); InterpreterCreateQuery create_interpreter_(target_create_query, create_context_); create_interpreter_.setInternal(true); @@ -1413,7 +1426,7 @@ void StorageWindowView::eventTimeParser(const ASTCreateQuery & query) } void StorageWindowView::writeIntoWindowView( - StorageWindowView & window_view, const Block & block, ContextPtr local_context) + StorageWindowView & window_view, Block && block, Chunk::ChunkInfoCollection && chunk_infos, ContextPtr local_context) { window_view.throwIfWindowViewIsDisabled(local_context); while (window_view.modifying_query) @@ -1428,7 +1441,7 @@ void StorageWindowView::writeIntoWindowView( window_view.max_watermark = window_view.getWindowUpperBound(first_record_timestamp); } - Pipe pipe(std::make_shared(block.cloneEmpty(), Chunk(block.getColumns(), block.rows()))); + Pipe pipe(std::make_shared(block)); UInt32 lateness_bound = 0; UInt32 t_max_watermark = 0; @@ -1473,10 +1486,10 @@ void StorageWindowView::writeIntoWindowView( auto syntax_result = TreeRewriter(local_context).analyze(query, columns); auto filter_expression = ExpressionAnalyzer(filter_function, syntax_result, local_context).getActionsDAG(false); - pipe.addSimpleTransform([&](const Block & header) + pipe.addSimpleTransform([&](const Block & header_) { return std::make_shared( - header, std::make_shared(filter_expression), + header_, std::make_shared(filter_expression), filter_function->getColumnName(), true); }); } @@ -1531,6 +1544,30 @@ void StorageWindowView::writeIntoWindowView( QueryProcessingStage::WithMergeableState); builder = select_block.buildQueryPipeline(); + + builder.addSimpleTransform([&](const Block & stream_header) + { + // Can't move chunk_infos here, that function could be called several times + return std::make_shared(chunk_infos.clone(), stream_header); + }); + + String window_view_id = window_view.getStorageID().hasUUID() ? toString(window_view.getStorageID().uuid) : window_view.getStorageID().getFullNameNotQuoted(); + builder.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared(window_view_id, stream_header); + }); + builder.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared(stream_header); + }); + +#ifdef ABORT_ON_LOGICAL_ERROR + builder.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared("StorageWindowView: Afrer tmp table before squashing", stream_header); + }); +#endif + builder.addSimpleTransform([&](const Block & current_header) { return std::make_shared( @@ -1570,6 +1607,13 @@ void StorageWindowView::writeIntoWindowView( lateness_upper_bound); }); +#ifdef ABORT_ON_LOGICAL_ERROR + builder.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared("StorageWindowView: Afrer WatermarkTransform", stream_header); + }); +#endif + auto inner_table = window_view.getInnerTable(); auto lock = inner_table->lockForShare( local_context->getCurrentQueryId(), local_context->getSettingsRef().lock_acquire_timeout); @@ -1586,9 +1630,16 @@ void StorageWindowView::writeIntoWindowView( auto convert_actions = std::make_shared( convert_actions_dag, ExpressionActionsSettings::fromContext(local_context, CompileExpressions::yes)); - builder.addSimpleTransform([&](const Block & header) { return std::make_shared(header, convert_actions); }); + builder.addSimpleTransform([&](const Block & header_) { return std::make_shared(header_, convert_actions); }); } +#ifdef ABORT_ON_LOGICAL_ERROR + builder.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared("StorageWindowView: Before out", stream_header); + }); +#endif + builder.addChain(Chain(std::move(output))); builder.setSinks([&](const Block & cur_header, Pipe::StreamType) { @@ -1715,7 +1766,7 @@ void registerStorageWindowView(StorageFactory & factory) "Experimental WINDOW VIEW feature " "is not enabled (the setting 'allow_experimental_window_view')"); - return std::make_shared(args.table_id, args.getLocalContext(), args.query, args.columns, args.mode); + return std::make_shared(args.table_id, args.getLocalContext(), args.query, args.columns, args.comment, args.mode); }); } diff --git a/src/Storages/WindowView/StorageWindowView.h b/src/Storages/WindowView/StorageWindowView.h index f79867df424..38fca512ed9 100644 --- a/src/Storages/WindowView/StorageWindowView.h +++ b/src/Storages/WindowView/StorageWindowView.h @@ -111,6 +111,7 @@ public: ContextPtr context_, const ASTCreateQuery & query, const ColumnsDescription & columns_, + const String & comment, LoadingStrictnessLevel mode); String getName() const override { return "WindowView"; } @@ -166,7 +167,7 @@ public: BlockIO populate(); - static void writeIntoWindowView(StorageWindowView & window_view, const Block & block, ContextPtr context); + static void writeIntoWindowView(StorageWindowView & window_view, Block && block, Chunk::ChunkInfoCollection && chunk_infos, ContextPtr context); ASTPtr getMergeableQuery() const { return mergeable_query->clone(); } diff --git a/src/Storages/buildQueryTreeForShard.cpp b/src/Storages/buildQueryTreeForShard.cpp index ed378169381..69e8f0a7880 100644 --- a/src/Storages/buildQueryTreeForShard.cpp +++ b/src/Storages/buildQueryTreeForShard.cpp @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -290,7 +291,7 @@ TableNodePtr executeSubqueryNode(const QueryTreeNodePtr & subquery_node, size_t min_block_size_rows = mutable_context->getSettingsRef().min_external_table_block_size_rows; size_t min_block_size_bytes = mutable_context->getSettingsRef().min_external_table_block_size_bytes; - auto squashing = std::make_shared(builder->getHeader(), min_block_size_rows, min_block_size_bytes); + auto squashing = std::make_shared(builder->getHeader(), min_block_size_rows, min_block_size_bytes); builder->resize(1); builder->addTransform(std::move(squashing)); diff --git a/src/Storages/examples/CMakeLists.txt b/src/Storages/examples/CMakeLists.txt index b4786b7313b..4f221efbd2b 100644 --- a/src/Storages/examples/CMakeLists.txt +++ b/src/Storages/examples/CMakeLists.txt @@ -5,4 +5,4 @@ clickhouse_add_executable (merge_selector2 merge_selector2.cpp) target_link_libraries (merge_selector2 PRIVATE dbms) clickhouse_add_executable (get_current_inserts_in_replicated get_current_inserts_in_replicated.cpp) -target_link_libraries (get_current_inserts_in_replicated PRIVATE dbms clickhouse_common_config clickhouse_common_zookeeper) +target_link_libraries (get_current_inserts_in_replicated PRIVATE dbms clickhouse_common_config clickhouse_common_zookeeper clickhouse_functions) diff --git a/src/Storages/getStructureOfRemoteTable.cpp b/src/Storages/getStructureOfRemoteTable.cpp index 6ea7bdc312d..56071abaa95 100644 --- a/src/Storages/getStructureOfRemoteTable.cpp +++ b/src/Storages/getStructureOfRemoteTable.cpp @@ -1,4 +1,5 @@ #include "getStructureOfRemoteTable.h" +#include #include #include #include diff --git a/src/Storages/registerStorages.cpp b/src/Storages/registerStorages.cpp index daf13cbf180..1b2b1e51317 100644 --- a/src/Storages/registerStorages.cpp +++ b/src/Storages/registerStorages.cpp @@ -43,6 +43,10 @@ void registerStorageIceberg(StorageFactory & factory); #endif #endif +#if USE_AZURE_BLOB_STORAGE +void registerStorageAzureQueue(StorageFactory & factory); +#endif + #if USE_HDFS #if USE_HIVE void registerStorageHive(StorageFactory & factory); @@ -130,6 +134,10 @@ void registerStorages(bool use_legacy_mongodb_integration [[maybe_unused]]) registerStorageFuzzJSON(factory); #endif +#if USE_AZURE_BLOB_STORAGE + registerStorageAzureQueue(factory); +#endif + #if USE_AWS_S3 registerStorageHudi(factory); registerStorageS3Queue(factory); diff --git a/src/Storages/transformQueryForExternalDatabase.cpp b/src/Storages/transformQueryForExternalDatabase.cpp index fc85bde11d9..251470a17a8 100644 --- a/src/Storages/transformQueryForExternalDatabase.cpp +++ b/src/Storages/transformQueryForExternalDatabase.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include @@ -419,7 +420,7 @@ String transformQueryForExternalDatabase( throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "No column names for query '{}' to external table '{}.{}'", query_info.query_tree->formatASTForErrorMessage(), database, table); - auto clone_query = getASTForExternalDatabaseFromQueryTree(query_info.query_tree); + auto clone_query = getASTForExternalDatabaseFromQueryTree(query_info.query_tree, query_info.table_expression); return transformQueryForExternalDatabaseImpl( clone_query, diff --git a/src/Storages/transformQueryForExternalDatabaseAnalyzer.cpp b/src/Storages/transformQueryForExternalDatabaseAnalyzer.cpp index 5e0bfdd5f2a..a7f56999d73 100644 --- a/src/Storages/transformQueryForExternalDatabaseAnalyzer.cpp +++ b/src/Storages/transformQueryForExternalDatabaseAnalyzer.cpp @@ -3,6 +3,7 @@ #include #include +#include #include #include @@ -10,7 +11,7 @@ #include #include #include - +#include #include @@ -20,6 +21,7 @@ namespace DB namespace ErrorCodes { extern const int UNSUPPORTED_METHOD; + extern const int LOGICAL_ERROR; } namespace @@ -55,7 +57,7 @@ public: } -ASTPtr getASTForExternalDatabaseFromQueryTree(const QueryTreeNodePtr & query_tree) +ASTPtr getASTForExternalDatabaseFromQueryTree(const QueryTreeNodePtr & query_tree, const QueryTreeNodePtr & table_expression) { auto new_tree = query_tree->clone(); @@ -63,6 +65,18 @@ ASTPtr getASTForExternalDatabaseFromQueryTree(const QueryTreeNodePtr & query_tre visitor.visit(new_tree); const auto * query_node = new_tree->as(); + const auto & join_tree = query_node->getJoinTree(); + bool allow_where = true; + if (const auto * join_node = join_tree->as()) + { + if (join_node->getKind() == JoinKind::Left) + allow_where = join_node->getLeftTableExpression()->isEqual(*table_expression); + else if (join_node->getKind() == JoinKind::Right) + allow_where = join_node->getRightTableExpression()->isEqual(*table_expression); + else + allow_where = (join_node->getKind() == JoinKind::Inner); + } + auto query_node_ast = query_node->toAST({ .add_cast_for_constants = false, .fully_qualified_identifiers = false }); const IAST * ast = query_node_ast.get(); @@ -76,7 +90,13 @@ ASTPtr getASTForExternalDatabaseFromQueryTree(const QueryTreeNodePtr & query_tre if (union_ast->list_of_selects->children.size() != 1) throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "QueryNode AST is not a single ASTSelectQuery, got {}", union_ast->list_of_selects->children.size()); - return union_ast->list_of_selects->children.at(0); + ASTPtr select_query = union_ast->list_of_selects->children.at(0); + auto * select_query_typed = select_query->as(); + if (!select_query_typed) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected ASTSelectQuery, got {}", select_query ? select_query->formatForErrorMessage() : "nullptr"); + if (!allow_where) + select_query_typed->setExpression(ASTSelectQuery::Expression::WHERE, nullptr); + return select_query; } } diff --git a/src/Storages/transformQueryForExternalDatabaseAnalyzer.h b/src/Storages/transformQueryForExternalDatabaseAnalyzer.h index f8983619d1f..7d8bf99646b 100644 --- a/src/Storages/transformQueryForExternalDatabaseAnalyzer.h +++ b/src/Storages/transformQueryForExternalDatabaseAnalyzer.h @@ -6,6 +6,6 @@ namespace DB { -ASTPtr getASTForExternalDatabaseFromQueryTree(const QueryTreeNodePtr & query_tree); +ASTPtr getASTForExternalDatabaseFromQueryTree(const QueryTreeNodePtr & query_tree, const QueryTreeNodePtr & table_expression); } diff --git a/src/TableFunctions/ITableFunction.cpp b/src/TableFunctions/ITableFunction.cpp index 137e1dc27fe..e5676c5c25d 100644 --- a/src/TableFunctions/ITableFunction.cpp +++ b/src/TableFunctions/ITableFunction.cpp @@ -36,7 +36,7 @@ StoragePtr ITableFunction::execute(const ASTPtr & ast_function, ContextPtr conte if (cached_columns.empty()) return executeImpl(ast_function, context, table_name, std::move(cached_columns), is_insert_query); - if (hasStaticStructure() && cached_columns == getActualTableStructure(context,is_insert_query)) + if (hasStaticStructure() && cached_columns == getActualTableStructure(context, is_insert_query)) return executeImpl(ast_function, context_to_use, table_name, std::move(cached_columns), is_insert_query); auto this_table_function = shared_from_this(); diff --git a/src/TableFunctions/ITableFunctionXDBC.cpp b/src/TableFunctions/ITableFunctionXDBC.cpp index a5c16b3a5aa..83d0ac4ef41 100644 --- a/src/TableFunctions/ITableFunctionXDBC.cpp +++ b/src/TableFunctions/ITableFunctionXDBC.cpp @@ -1,8 +1,10 @@ +#include +#include #include -#include -#include -#include #include +#include +#include +#include #include #include #include diff --git a/src/TableFunctions/TableFunctionDictionary.cpp b/src/TableFunctions/TableFunctionDictionary.cpp index 867fbf5b11e..9f0a7ac91b6 100644 --- a/src/TableFunctions/TableFunctionDictionary.cpp +++ b/src/TableFunctions/TableFunctionDictionary.cpp @@ -74,13 +74,12 @@ ColumnsDescription TableFunctionDictionary::getActualTableStructure(ContextPtr c /// otherwise, we get table structure by dictionary structure. auto dictionary_structure = external_loader.getDictionaryStructure(dictionary_name, context); - return ColumnsDescription(StorageDictionary::getNamesAndTypes(dictionary_structure)); + return ColumnsDescription(StorageDictionary::getNamesAndTypes(dictionary_structure, false)); } StoragePtr TableFunctionDictionary::executeImpl( const ASTPtr &, ContextPtr context, const std::string & table_name, ColumnsDescription, bool is_insert_query) const { - context->checkAccess(AccessType::dictGet, getDatabaseName(), table_name); StorageID dict_id(getDatabaseName(), table_name); auto dictionary_table_structure = getActualTableStructure(context, is_insert_query); diff --git a/src/TableFunctions/TableFunctionExecutable.cpp b/src/TableFunctions/TableFunctionExecutable.cpp index 2c3802e8667..cccd3587bc7 100644 --- a/src/TableFunctions/TableFunctionExecutable.cpp +++ b/src/TableFunctions/TableFunctionExecutable.cpp @@ -170,7 +170,14 @@ StoragePtr TableFunctionExecutable::executeImpl(const ASTPtr & /*ast_function*/, if (settings_query != nullptr) settings.applyChanges(settings_query->as()->changes); - auto storage = std::make_shared(storage_id, format, settings, input_queries, getActualTableStructure(context, is_insert_query), ConstraintsDescription{}); + auto storage = std::make_shared( + storage_id, + format, + settings, + input_queries, + getActualTableStructure(context, is_insert_query), + ConstraintsDescription{}, + /* comment = */ ""); storage->startup(); return storage; } diff --git a/src/TableFunctions/TableFunctionExplain.cpp b/src/TableFunctions/TableFunctionExplain.cpp index df2835dd630..552b9fde986 100644 --- a/src/TableFunctions/TableFunctionExplain.cpp +++ b/src/TableFunctions/TableFunctionExplain.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/TableFunctions/TableFunctionFactory.cpp b/src/TableFunctions/TableFunctionFactory.cpp index ce3daff0785..e505535ae76 100644 --- a/src/TableFunctions/TableFunctionFactory.cpp +++ b/src/TableFunctions/TableFunctionFactory.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include #include @@ -18,17 +19,17 @@ namespace ErrorCodes } void TableFunctionFactory::registerFunction( - const std::string & name, Value value, CaseSensitiveness case_sensitiveness) + const std::string & name, Value value, Case case_sensitiveness) { if (!table_functions.emplace(name, value).second) throw Exception(ErrorCodes::LOGICAL_ERROR, "TableFunctionFactory: the table function name '{}' is not unique", name); - if (case_sensitiveness == CaseInsensitive + if (case_sensitiveness == Case::Insensitive && !case_insensitive_table_functions.emplace(Poco::toLower(name), value).second) throw Exception(ErrorCodes::LOGICAL_ERROR, "TableFunctionFactory: " "the case insensitive table function name '{}' is not unique", name); - KnownTableFunctionNames::instance().add(name, (case_sensitiveness == CaseInsensitive)); + KnownTableFunctionNames::instance().add(name, (case_sensitiveness == Case::Insensitive)); } TableFunctionPtr TableFunctionFactory::get( diff --git a/src/TableFunctions/TableFunctionFactory.h b/src/TableFunctions/TableFunctionFactory.h index 2cc648ba181..adc74c2e735 100644 --- a/src/TableFunctions/TableFunctionFactory.h +++ b/src/TableFunctions/TableFunctionFactory.h @@ -48,10 +48,10 @@ public: void registerFunction( const std::string & name, Value value, - CaseSensitiveness case_sensitiveness = CaseSensitive); + Case case_sensitiveness = Case::Sensitive); template - void registerFunction(TableFunctionProperties properties = {}, CaseSensitiveness case_sensitiveness = CaseSensitive) + void registerFunction(TableFunctionProperties properties = {}, Case case_sensitiveness = Case::Sensitive) { auto creator = []() -> TableFunctionPtr { return std::make_shared(); }; registerFunction(Function::name, diff --git a/src/TableFunctions/TableFunctionFile.cpp b/src/TableFunctions/TableFunctionFile.cpp index 28bf72e07fb..1b6d86f8fa5 100644 --- a/src/TableFunctions/TableFunctionFile.cpp +++ b/src/TableFunctions/TableFunctionFile.cpp @@ -4,6 +4,7 @@ #include "registerTableFunctions.h" #include +#include #include #include #include diff --git a/src/TableFunctions/TableFunctionFileCluster.cpp b/src/TableFunctions/TableFunctionFileCluster.cpp index 3e53349b022..514fa3143d0 100644 --- a/src/TableFunctions/TableFunctionFileCluster.cpp +++ b/src/TableFunctions/TableFunctionFileCluster.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/TableFunctions/TableFunctionFormat.cpp b/src/TableFunctions/TableFunctionFormat.cpp index ad2a142a140..7e4fdea1ff3 100644 --- a/src/TableFunctions/TableFunctionFormat.cpp +++ b/src/TableFunctions/TableFunctionFormat.cpp @@ -1,5 +1,7 @@ #include +#include + #include #include @@ -217,7 +219,7 @@ Result: void registerTableFunctionFormat(TableFunctionFactory & factory) { - factory.registerFunction({format_table_function_documentation, false}, TableFunctionFactory::CaseInsensitive); + factory.registerFunction({format_table_function_documentation, false}, TableFunctionFactory::Case::Insensitive); } } diff --git a/src/TableFunctions/TableFunctionMySQL.cpp b/src/TableFunctions/TableFunctionMySQL.cpp index b027cf42880..ffd318f36f7 100644 --- a/src/TableFunctions/TableFunctionMySQL.cpp +++ b/src/TableFunctions/TableFunctionMySQL.cpp @@ -3,6 +3,8 @@ #if USE_MYSQL #include + +#include #include #include #include diff --git a/src/TableFunctions/TableFunctionPostgreSQL.cpp b/src/TableFunctions/TableFunctionPostgreSQL.cpp index 8d94988cd65..d690da9949a 100644 --- a/src/TableFunctions/TableFunctionPostgreSQL.cpp +++ b/src/TableFunctions/TableFunctionPostgreSQL.cpp @@ -4,6 +4,7 @@ #include #include +#include #include #include #include @@ -80,8 +81,9 @@ void TableFunctionPostgreSQL::parseArguments(const ASTPtr & ast_function, Contex *configuration, settings.postgresql_connection_pool_size, settings.postgresql_connection_pool_wait_timeout, - POSTGRESQL_POOL_WITH_FAILOVER_DEFAULT_MAX_TRIES, - settings.postgresql_connection_pool_auto_close_connection); + settings.postgresql_connection_pool_retries, + settings.postgresql_connection_pool_auto_close_connection, + settings.postgresql_connection_attempt_timeout); } } diff --git a/src/TableFunctions/TableFunctionRemote.cpp b/src/TableFunctions/TableFunctionRemote.cpp index d7774bb6478..e60c31b2d77 100644 --- a/src/TableFunctions/TableFunctionRemote.cpp +++ b/src/TableFunctions/TableFunctionRemote.cpp @@ -17,6 +17,7 @@ #include #include #include +#include #include #include "registerTableFunctions.h" diff --git a/src/TableFunctions/TableFunctionSQLite.cpp b/src/TableFunctions/TableFunctionSQLite.cpp index e367e05bf73..87353025d1d 100644 --- a/src/TableFunctions/TableFunctionSQLite.cpp +++ b/src/TableFunctions/TableFunctionSQLite.cpp @@ -57,7 +57,7 @@ StoragePtr TableFunctionSQLite::executeImpl(const ASTPtr & /*ast_function*/, sqlite_db, database_path, remote_table_name, - cached_columns, ConstraintsDescription{}, context); + cached_columns, ConstraintsDescription{}, /* comment = */ "", context); storage->startup(); return storage; diff --git a/src/TableFunctions/TableFunctionValues.cpp b/src/TableFunctions/TableFunctionValues.cpp index 4b56fa57091..95c531f8a3b 100644 --- a/src/TableFunctions/TableFunctionValues.cpp +++ b/src/TableFunctions/TableFunctionValues.cpp @@ -174,7 +174,7 @@ StoragePtr TableFunctionValues::executeImpl(const ASTPtr & ast_function, Context void registerTableFunctionValues(TableFunctionFactory & factory) { - factory.registerFunction({.documentation = {}, .allow_readonly = true}, TableFunctionFactory::CaseInsensitive); + factory.registerFunction({.documentation = {}, .allow_readonly = true}, TableFunctionFactory::Case::Insensitive); } } diff --git a/src/TableFunctions/TableFunctionView.cpp b/src/TableFunctions/TableFunctionView.cpp index 2a50fb2d006..57501df6d4d 100644 --- a/src/TableFunctions/TableFunctionView.cpp +++ b/src/TableFunctions/TableFunctionView.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/TableFunctions/TableFunctionViewIfPermitted.cpp b/src/TableFunctions/TableFunctionViewIfPermitted.cpp index b1691803988..935be6c1987 100644 --- a/src/TableFunctions/TableFunctionViewIfPermitted.cpp +++ b/src/TableFunctions/TableFunctionViewIfPermitted.cpp @@ -1,3 +1,4 @@ +#include #include #include #include diff --git a/src/configure_config.cmake b/src/configure_config.cmake index 78d13b49417..d04cfbe946e 100644 --- a/src/configure_config.cmake +++ b/src/configure_config.cmake @@ -170,6 +170,9 @@ endif() if (TARGET ch_contrib::pocketfft) set(USE_POCKETFFT 1) endif() +if (TARGET ch_contrib::prometheus_protobufs) + set(USE_PROMETHEUS_PROTOBUFS 1) +endif() if (ch_contrib::mongo-cxx-driver) set(USE_MONGODB 1) endif() diff --git a/tests/ci/.gitignore b/tests/ci/.gitignore index bc407a1f961..b5e58611040 100644 --- a/tests/ci/.gitignore +++ b/tests/ci/.gitignore @@ -1,4 +1 @@ -*_lambda/lambda-venv -*_lambda/lambda-package -*_lambda/lambda-package.zip gh_cache \ No newline at end of file diff --git a/tests/ci/.mypy.ini b/tests/ci/.mypy.ini index 9bc44025826..f12d27979ce 100644 --- a/tests/ci/.mypy.ini +++ b/tests/ci/.mypy.ini @@ -15,3 +15,4 @@ warn_return_any = True no_implicit_reexport = True strict_equality = True extra_checks = True +ignore_missing_imports = True \ No newline at end of file diff --git a/tests/ci/artifactory.py b/tests/ci/artifactory.py new file mode 100644 index 00000000000..86dcaf79854 --- /dev/null +++ b/tests/ci/artifactory.py @@ -0,0 +1,385 @@ +import argparse +import time +from pathlib import Path +from typing import Optional +from shutil import copy2 +from create_release import ( + PackageDownloader, + ReleaseInfo, + ReleaseContextManager, + ReleaseProgress, +) +from ci_utils import WithIter, Shell + + +class MountPointApp(metaclass=WithIter): + RCLONE = "rclone" + S3FS = "s3fs" + + +class R2MountPoint: + _TEST_BUCKET_NAME = "repo-test" + _PROD_BUCKET_NAME = "packages" + _CACHE_MAX_SIZE_GB = 20 + MOUNT_POINT = "/home/ubuntu/mountpoint" + API_ENDPOINT = "https://d4fd593eebab2e3a58a599400c4cd64d.r2.cloudflarestorage.com" + LOG_FILE = "/home/ubuntu/fuse_mount.log" + # mod time is not required by reprepro and createrepo - disable to simplify bucket's mount sync (applicable fro rclone) + NOMODTIME = True + # enable debug messages in mount log + DEBUG = True + # enable cache for mountpoint + CACHE_ENABLED = False + # TODO: which mode is better: minimal/writes/full/off + _RCLONE_CACHE_MODE = "minimal" + UMASK = "0000" + + def __init__(self, app: str, dry_run: bool) -> None: + assert app in MountPointApp + self.app = app + if dry_run: + self.bucket_name = self._TEST_BUCKET_NAME + else: + self.bucket_name = self._PROD_BUCKET_NAME + + self.aux_mount_options = "" + if self.app == MountPointApp.S3FS: + self.cache_dir = "/home/ubuntu/s3fs_cache" + # self.aux_mount_options += "-o nomodtime " if self.NOMODTIME else "" not for s3fs + self.aux_mount_options += "--debug " if self.DEBUG else "" + self.aux_mount_options += ( + f"-o use_cache={self.cache_dir} -o cache_size_mb={self._CACHE_MAX_SIZE_GB * 1024} " + if self.CACHE_ENABLED + else "" + ) + # without -o nomultipart there are errors like "Error 5 writing to /home/ubuntu/***.deb: Input/output error" + self.mount_cmd = f"s3fs {self.bucket_name} {self.MOUNT_POINT} -o url={self.API_ENDPOINT} -o use_path_request_style -o umask=0000 -o nomultipart -o logfile={self.LOG_FILE} {self.aux_mount_options}" + elif self.app == MountPointApp.RCLONE: + # run rclone mount process asynchronously, otherwise subprocess.run(daemonized command) will not return + self.cache_dir = "/home/ubuntu/rclone_cache" + self.aux_mount_options += "--no-modtime " if self.NOMODTIME else "" + self.aux_mount_options += "-v " if self.DEBUG else "" # -vv too verbose + self.aux_mount_options += ( + f"--vfs-cache-mode {self._RCLONE_CACHE_MODE} --vfs-cache-max-size {self._CACHE_MAX_SIZE_GB}G" + if self.CACHE_ENABLED + else "--vfs-cache-mode off" + ) + # Use --no-modtime to try to avoid: ERROR : rpm/lts/clickhouse-client-24.3.6.5.x86_64.rpm: Failed to apply pending mod time + self.mount_cmd = f"rclone mount remote:{self.bucket_name} {self.MOUNT_POINT} --daemon --cache-dir {self.cache_dir} --umask 0000 --log-file {self.LOG_FILE} {self.aux_mount_options}" + else: + assert False + + def init(self): + print(f"Mount bucket [{self.bucket_name}] to [{self.MOUNT_POINT}]") + _CLEAN_LOG_FILE_CMD = f"tail -n 1000 {self.LOG_FILE} > {self.LOG_FILE}_tmp && mv {self.LOG_FILE}_tmp {self.LOG_FILE} ||:" + _MKDIR_CMD = f"mkdir -p {self.MOUNT_POINT}" + _MKDIR_FOR_CACHE = f"mkdir -p {self.cache_dir}" + _UNMOUNT_CMD = ( + f"mount | grep -q {self.MOUNT_POINT} && umount {self.MOUNT_POINT} ||:" + ) + + _TEST_MOUNT_CMD = f"mount | grep -q {self.MOUNT_POINT}" + Shell.run(_CLEAN_LOG_FILE_CMD) + Shell.run(_UNMOUNT_CMD) + Shell.run(_MKDIR_CMD) + Shell.run(_MKDIR_FOR_CACHE) + if self.app == MountPointApp.S3FS: + Shell.run(self.mount_cmd, check=True) + else: + # didn't manage to use simple run() and without blocking or failure + Shell.run_as_daemon(self.mount_cmd) + time.sleep(3) + Shell.run(_TEST_MOUNT_CMD, check=True) + + @classmethod + def teardown(cls): + print(f"Unmount [{cls.MOUNT_POINT}]") + Shell.run(f"umount {cls.MOUNT_POINT}") + + +class RepoCodenames(metaclass=WithIter): + LTS = "lts" + STABLE = "stable" + + +class DebianArtifactory: + _TEST_REPO_URL = "https://pub-73dd1910f4284a81a02a67018967e028.r2.dev/deb" + _PROD_REPO_URL = "https://packages.clickhouse.com/deb" + + def __init__(self, release_info: ReleaseInfo, dry_run: bool): + self.release_info = release_info + self.codename = release_info.codename + self.version = release_info.version + if dry_run: + self.repo_url = self._TEST_REPO_URL + else: + self.repo_url = self._PROD_REPO_URL + assert self.codename in RepoCodenames + self.pd = PackageDownloader( + release=release_info.release_branch, + commit_sha=release_info.commit_sha, + version=release_info.version, + ) + + def export_packages(self): + assert self.pd.local_deb_packages_ready(), "BUG: Packages are not downloaded" + print("Start adding packages") + paths = [ + self.pd.LOCAL_DIR + "/" + file for file in self.pd.get_deb_packages_files() + ] + REPREPRO_CMD_PREFIX = f"reprepro --basedir {R2MountPoint.MOUNT_POINT}/configs/deb --outdir {R2MountPoint.MOUNT_POINT}/deb --verbose" + cmd = f"{REPREPRO_CMD_PREFIX} includedeb {self.codename} {' '.join(paths)}" + print("Running export command:") + print(f" {cmd}") + Shell.run(cmd, check=True) + Shell.run("sync") + + if self.codename == RepoCodenames.LTS: + packages_with_version = [ + package + "=" + self.version for package in self.pd.get_packages_names() + ] + print( + f"Copy packages from {RepoCodenames.LTS} to {RepoCodenames.STABLE} repository" + ) + cmd = f"{REPREPRO_CMD_PREFIX} copy {RepoCodenames.STABLE} {RepoCodenames.LTS} {' '.join(packages_with_version)}" + print("Running copy command:") + print(f" {cmd}") + Shell.run(cmd, check=True) + Shell.run("sync") + + def test_packages(self): + Shell.run("docker pull ubuntu:latest") + print(f"Test packages installation, version [{self.version}]") + debian_command = f"echo 'deb {self.repo_url} stable main' | tee /etc/apt/sources.list.d/clickhouse.list; apt update -y; apt-get install -y clickhouse-common-static={self.version} clickhouse-client={self.version}" + cmd = f'docker run --rm ubuntu:latest bash -c "apt update -y; apt install -y sudo gnupg ca-certificates; apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 8919F6BD2B48D754; {debian_command}"' + print("Running test command:") + print(f" {cmd}") + Shell.run(cmd, check=True) + self.release_info.debian_command = debian_command + self.release_info.dump() + + +def _copy_if_not_exists(src: Path, dst: Path) -> Path: + if dst.is_dir(): + dst = dst / src.name + if not dst.exists(): + return copy2(src, dst) # type: ignore + if src.stat().st_size == dst.stat().st_size: + return dst + return copy2(src, dst) # type: ignore + + +class RpmArtifactory: + _TEST_REPO_URL = ( + "https://pub-73dd1910f4284a81a02a67018967e028.r2.dev/rpm/clickhouse.repo" + ) + _PROD_REPO_URL = "https://packages.clickhouse.com/rpm/clickhouse.repo" + _SIGN_KEY = "885E2BDCF96B0B45ABF058453E4AD4719DDE9A38" + + def __init__(self, release_info: ReleaseInfo, dry_run: bool): + self.release_info = release_info + self.codename = release_info.codename + self.version = release_info.version + if dry_run: + self.repo_url = self._TEST_REPO_URL + else: + self.repo_url = self._PROD_REPO_URL + assert self.codename in RepoCodenames + self.pd = PackageDownloader( + release=release_info.release_branch, + commit_sha=release_info.commit_sha, + version=release_info.version, + ) + + def export_packages(self, codename: Optional[str] = None) -> None: + assert self.pd.local_rpm_packages_ready(), "BUG: Packages are not downloaded" + codename = codename or self.codename + print(f"Start adding packages to [{codename}]") + paths = [ + self.pd.LOCAL_DIR + "/" + file for file in self.pd.get_rpm_packages_files() + ] + + dest_dir = Path(R2MountPoint.MOUNT_POINT) / "rpm" / codename + + for package in paths: + _copy_if_not_exists(Path(package), dest_dir) + + commands = ( + f"createrepo_c --local-sqlite --workers=2 --update --verbose {dest_dir}", + f"gpg --sign-with {self._SIGN_KEY} --detach-sign --batch --yes --armor {dest_dir / 'repodata' / 'repomd.xml'}", + ) + print(f"Exporting RPM packages into [{codename}]") + + for command in commands: + print("Running command:") + print(f" {command}") + Shell.run(command, check=True) + + update_public_key = f"gpg --armor --export {self._SIGN_KEY}" + pub_key_path = dest_dir / "repodata" / "repomd.xml.key" + print("Updating repomd.xml.key") + pub_key_path.write_text(Shell.run(update_public_key, check=True)) + if codename == RepoCodenames.LTS: + self.export_packages(RepoCodenames.STABLE) + Shell.run("sync") + + def test_packages(self): + Shell.run("docker pull fedora:latest") + print(f"Test package installation, version [{self.version}]") + rpm_command = f"dnf config-manager --add-repo={self.repo_url} && dnf makecache && dnf -y install clickhouse-client-{self.version}-1" + cmd = f'docker run --rm fedora:latest /bin/bash -c "dnf -y install dnf-plugins-core && dnf config-manager --add-repo={self.repo_url} && {rpm_command}"' + print("Running test command:") + print(f" {cmd}") + Shell.run(cmd, check=True) + self.release_info.rpm_command = rpm_command + self.release_info.dump() + + +class TgzArtifactory: + _TEST_REPO_URL = "https://pub-73dd1910f4284a81a02a67018967e028.r2.dev/tgz" + _PROD_REPO_URL = "https://packages.clickhouse.com/tgz" + + def __init__(self, release_info: ReleaseInfo, dry_run: bool): + self.release_info = release_info + self.codename = release_info.codename + self.version = release_info.version + if dry_run: + self.repo_url = self._TEST_REPO_URL + else: + self.repo_url = self._PROD_REPO_URL + assert self.codename in RepoCodenames + self.pd = PackageDownloader( + release=release_info.release_branch, + commit_sha=release_info.commit_sha, + version=release_info.version, + ) + + def export_packages(self, codename: Optional[str] = None) -> None: + assert self.pd.local_tgz_packages_ready(), "BUG: Packages are not downloaded" + codename = codename or self.codename + + paths = [ + self.pd.LOCAL_DIR + "/" + file for file in self.pd.get_tgz_packages_files() + ] + + dest_dir = Path(R2MountPoint.MOUNT_POINT) / "tgz" / codename + + print(f"Exporting TGZ packages into [{codename}]") + + for package in paths: + _copy_if_not_exists(Path(package), dest_dir) + + if codename == RepoCodenames.LTS: + self.export_packages(RepoCodenames.STABLE) + Shell.run("sync") + + def test_packages(self): + tgz_file = "/tmp/tmp.tgz" + tgz_sha_file = "/tmp/tmp.tgz.sha512" + cmd = f"curl -o {tgz_file} -f0 {self.repo_url}/stable/clickhouse-client-{self.version}-arm64.tgz" + Shell.run( + cmd, + check=True, + ) + Shell.run( + f"curl -o {tgz_sha_file} -f0 {self.repo_url}/stable/clickhouse-client-{self.version}-arm64.tgz.sha512", + check=True, + ) + expected_checksum = Shell.run(f"cut -d ' ' -f 1 {tgz_sha_file}", check=True) + actual_checksum = Shell.run(f"sha512sum {tgz_file} | cut -d ' ' -f 1") + assert ( + expected_checksum == actual_checksum + ), f"[{actual_checksum} != {expected_checksum}]" + Shell.run("rm /tmp/tmp.tgz*") + self.release_info.tgz_command = cmd + self.release_info.dump() + + +def parse_args() -> argparse.Namespace: + parser = argparse.ArgumentParser( + formatter_class=argparse.ArgumentDefaultsHelpFormatter, + description="Adds release packages to the repository", + ) + parser.add_argument( + "--export-debian", + action="store_true", + help="Export debian packages to repository", + ) + parser.add_argument( + "--export-rpm", + action="store_true", + help="Export rpm packages to repository", + ) + parser.add_argument( + "--export-tgz", + action="store_true", + help="Export tgz packages to repository", + ) + parser.add_argument( + "--test-debian", + action="store_true", + help="Test debian packages installation", + ) + parser.add_argument( + "--test-rpm", + action="store_true", + help="Test rpm packages installation", + ) + parser.add_argument( + "--test-tgz", + action="store_true", + help="Test tgz packages installation", + ) + parser.add_argument( + "--dry-run", + action="store_true", + help="Dry run mode", + ) + return parser.parse_args() + + +if __name__ == "__main__": + args = parse_args() + + """ + Use S3FS. RCLONE has some errors with r2 remote which I didn't figure out how to resolve: + ERROR : IO error: NotImplemented: versionId not implemented + Failed to copy: NotImplemented: versionId not implemented + """ + mp = R2MountPoint(MountPointApp.S3FS, dry_run=args.dry_run) + if args.export_debian: + with ReleaseContextManager( + release_progress=ReleaseProgress.EXPORT_DEB + ) as release_info: + mp.init() + DebianArtifactory(release_info, dry_run=args.dry_run).export_packages() + mp.teardown() + if args.export_rpm: + with ReleaseContextManager( + release_progress=ReleaseProgress.EXPORT_RPM + ) as release_info: + mp.init() + RpmArtifactory(release_info, dry_run=args.dry_run).export_packages() + mp.teardown() + if args.export_tgz: + with ReleaseContextManager( + release_progress=ReleaseProgress.EXPORT_TGZ + ) as release_info: + mp.init() + TgzArtifactory(release_info, dry_run=args.dry_run).export_packages() + mp.teardown() + if args.test_debian: + with ReleaseContextManager( + release_progress=ReleaseProgress.TEST_DEB + ) as release_info: + DebianArtifactory(release_info, dry_run=args.dry_run).test_packages() + if args.test_tgz: + with ReleaseContextManager( + release_progress=ReleaseProgress.TEST_TGZ + ) as release_info: + TgzArtifactory(release_info, dry_run=args.dry_run).test_packages() + if args.test_rpm: + with ReleaseContextManager( + release_progress=ReleaseProgress.TEST_RPM + ) as release_info: + RpmArtifactory(release_info, dry_run=args.dry_run).test_packages() diff --git a/tests/ci/auto_release.py b/tests/ci/auto_release.py index bfd19b11e6d..f2386fe207f 100644 --- a/tests/ci/auto_release.py +++ b/tests/ci/auto_release.py @@ -1,17 +1,17 @@ import argparse -from datetime import timedelta, datetime -import logging +import dataclasses +import json import os -from commit_status_helper import get_commit_filtered_statuses +import sys +from typing import List + from get_robot_token import get_best_robot_token from github_helper import GitHub -from release import Release, Repo as ReleaseRepo, RELEASE_READY_STATUS +from ci_utils import Shell +from env_helper import GITHUB_REPOSITORY from report import SUCCESS -from ssh import SSHKey - -LOGGER_NAME = __name__ -HELPER_LOGGERS = ["github_helper", LOGGER_NAME] -logger = logging.getLogger(LOGGER_NAME) +from ci_buddy import CIBuddy +from ci_config import CI def parse_args(): @@ -21,120 +21,198 @@ def parse_args(): ) parser.add_argument("--token", help="GitHub token, if not set, used from smm") parser.add_argument( - "--repo", default="ClickHouse/ClickHouse", help="Repo owner/name" - ) - parser.add_argument("--dry-run", action="store_true", help="Do not create anything") - parser.add_argument( - "--release-after-days", - type=int, - default=3, - help="Do automatic release on the latest green commit after the latest " - "release if the newest release is older than the specified days", - ) - parser.add_argument( - "--debug-helpers", + "--post-status", action="store_true", - help="Add debug logging for this script and github_helper", + help="Post release branch statuses", ) parser.add_argument( - "--remote-protocol", - "-p", - default="ssh", - choices=ReleaseRepo.VALID, - help="repo protocol for git commands remote, 'origin' is a special case and " - "uses 'origin' as a remote", + "--post-auto-release-complete", + action="store_true", + help="Post autorelease completion status", ) + parser.add_argument( + "--prepare", + action="store_true", + help="Prepare autorelease info", + ) + parser.add_argument( + "--wf-status", + type=str, + default="", + help="overall workflow status [success|failure]", + ) + return parser.parse_args(), parser - return parser.parse_args() + +MAX_NUMBER_OF_COMMITS_TO_CONSIDER_FOR_RELEASE = 5 +AUTORELEASE_INFO_FILE = "/tmp/autorelease_info.json" + + +@dataclasses.dataclass +class ReleaseParams: + ready: bool + ci_status: str + num_patches: int + release_branch: str + commit_sha: str + commits_to_branch_head: int + latest: bool + + def to_dict(self): + return dataclasses.asdict(self) + + +@dataclasses.dataclass +class AutoReleaseInfo: + releases: List[ReleaseParams] + + def add_release(self, release_params: ReleaseParams) -> None: + self.releases.append(release_params) + + def dump(self): + print(f"Dump release info into [{AUTORELEASE_INFO_FILE}]") + with open(AUTORELEASE_INFO_FILE, "w", encoding="utf-8") as f: + print(json.dumps(dataclasses.asdict(self), indent=2), file=f) + + @staticmethod + def from_file() -> "AutoReleaseInfo": + with open(AUTORELEASE_INFO_FILE, "r", encoding="utf-8") as json_file: + res = json.load(json_file) + releases = [ReleaseParams(**release) for release in res["releases"]] + return AutoReleaseInfo(releases=releases) + + +def _prepare(token): + assert len(token) > 10 + os.environ["GH_TOKEN"] = token + Shell.run("gh auth status", check=True) + + gh = GitHub(token) + prs = gh.get_release_pulls(GITHUB_REPOSITORY) + prs.sort(key=lambda x: x.head.ref) + branch_names = [pr.head.ref for pr in prs] + print(f"Found release branches [{branch_names}]") + + repo = gh.get_repo(GITHUB_REPOSITORY) + autoRelease_info = AutoReleaseInfo(releases=[]) + + for pr in prs: + print(f"\nChecking PR [{pr.head.ref}]") + + refs = list(repo.get_git_matching_refs(f"tags/v{pr.head.ref}")) + assert refs + + refs.sort(key=lambda ref: ref.ref) + latest_release_tag_ref = refs[-1] + latest_release_tag = repo.get_git_tag(latest_release_tag_ref.object.sha) + + commits = Shell.run( + f"git rev-list --first-parent {latest_release_tag.tag}..origin/{pr.head.ref}", + check=True, + ).split("\n") + commit_num = len(commits) + print( + f"Previous release [{latest_release_tag.tag}] was [{commit_num}] commits ago, date [{latest_release_tag.tagger.date}]" + ) + + commits_to_check = commits[:-1] # Exclude the version bump commit + commit_sha = "" + commit_ci_status = "" + commits_to_branch_head = 0 + + for idx, commit in enumerate( + commits_to_check[:MAX_NUMBER_OF_COMMITS_TO_CONSIDER_FOR_RELEASE] + ): + print( + f"Check commit [{commit}] [{pr.head.ref}~{idx+1}] as release candidate" + ) + commit_num -= 1 + + is_completed = CI.GHActions.check_wf_completed( + token=token, commit_sha=commit + ) + if not is_completed: + print(f"CI is in progress for [{commit}] - check previous commit") + commits_to_branch_head += 1 + continue + + commit_ci_status = CI.GHActions.get_commit_status_by_name( + token=token, + commit_sha=commit, + status_name=(CI.JobNames.BUILD_CHECK, "ClickHouse build check"), + ) + commit_sha = commit + if commit_ci_status == SUCCESS: + break + + print(f"CI status [{commit_ci_status}] - skip") + commits_to_branch_head += 1 + + ready = False + if commit_ci_status == SUCCESS and commit_sha: + print( + f"Add release ready info for commit [{commit_sha}] and release branch [{pr.head.ref}]" + ) + ready = True + else: + print(f"WARNING: No ready commits found for release branch [{pr.head.ref}]") + + autoRelease_info.add_release( + ReleaseParams( + release_branch=pr.head.ref, + commit_sha=commit_sha, + ready=ready, + ci_status=commit_ci_status, + num_patches=commit_num, + commits_to_branch_head=commits_to_branch_head, + latest=False, + ) + ) + + if autoRelease_info.releases: + autoRelease_info.releases[-1].latest = True + + autoRelease_info.dump() def main(): - args = parse_args() - logging.basicConfig(level=logging.INFO) - if args.debug_helpers: - for logger_name in HELPER_LOGGERS: - logging.getLogger(logger_name).setLevel(logging.DEBUG) + args, parser = parse_args() - token = args.token or get_best_robot_token() - days_as_timedelta = timedelta(days=args.release_after_days) - now = datetime.now() - - gh = GitHub(token) - prs = gh.get_release_pulls(args.repo) - branch_names = [pr.head.ref for pr in prs] - - logger.info("Found release branches: %s\n ", " \n".join(branch_names)) - repo = gh.get_repo(args.repo) - - # In general there is no guarantee on which order the refs/commits are - # returned from the API, so we have to order them. - for pr in prs: - logger.info("Checking PR %s", pr.head.ref) - - refs = list(repo.get_git_matching_refs(f"tags/v{pr.head.ref}")) - refs.sort(key=lambda ref: ref.ref) - - latest_release_tag_ref = refs[-1] - latest_release_tag = repo.get_git_tag(latest_release_tag_ref.object.sha) - logger.info("That last release was done at %s", latest_release_tag.tagger.date) - - if latest_release_tag.tagger.date + days_as_timedelta > now: - logger.info( - "Not enough days since the last release %s," - " no automatic release can be done", - latest_release_tag.tag, + if args.post_status: + info = AutoReleaseInfo.from_file() + for release_info in info.releases: + if release_info.ready: + CIBuddy(dry_run=False).post_info( + title=f"Auto Release Status for {release_info.release_branch}", + body=release_info.to_dict(), + ) + else: + CIBuddy(dry_run=False).post_warning( + title=f"Auto Release Status for {release_info.release_branch}", + body=release_info.to_dict(), + ) + elif args.post_auto_release_complete: + assert args.wf_status, "--wf-status Required with --post-auto-release-complete" + if args.wf_status != SUCCESS: + CIBuddy(dry_run=False).post_job_error( + error_description="Autorelease workflow failed", + job_name="Autorelease", + with_instance_info=False, + with_wf_link=True, + critical=True, ) - continue - - unreleased_commits = list( - repo.get_commits(sha=pr.head.ref, since=latest_release_tag.tagger.date) - ) - unreleased_commits.sort( - key=lambda commit: commit.commit.committer.date, reverse=True - ) - - for commit in unreleased_commits: - logger.info("Checking statuses of commit %s", commit.sha) - statuses = get_commit_filtered_statuses(commit) - all_success = all(st.state == SUCCESS for st in statuses) - passed_ready_for_release_check = any( - st.context == RELEASE_READY_STATUS and st.state == SUCCESS - for st in statuses + else: + CIBuddy(dry_run=False).post_info( + title=f"Autorelease completed", + body="", + with_wf_link=True, ) - if not (all_success and passed_ready_for_release_check): - logger.info("Commit is not green, thus not suitable for release") - continue - - logger.info("Commit is ready for release, let's release!") - - release = Release( - ReleaseRepo(args.repo, args.remote_protocol), - commit.sha, - "patch", - args.dry_run, - True, - ) - try: - release.do(True, True, True) - except: - if release.has_rollback: - logging.error( - "!!The release process finished with error, read the output carefully!!" - ) - logging.error( - "Probably, rollback finished with error. " - "If you don't see any of the following commands in the output, " - "execute them manually:" - ) - release.log_rollback() - raise - logging.info("New release is done!") - break + elif args.prepare: + _prepare(token=args.token or get_best_robot_token()) + else: + parser.print_help() + sys.exit(2) if __name__ == "__main__": - if os.getenv("ROBOT_CLICKHOUSE_SSH_KEY", ""): - with SSHKey("ROBOT_CLICKHOUSE_SSH_KEY"): - main() - else: - main() + main() diff --git a/tests/ci/autoscale_runners_lambda/app.py b/tests/ci/autoscale_runners_lambda/app.py deleted file mode 100644 index 520abcb41bb..00000000000 --- a/tests/ci/autoscale_runners_lambda/app.py +++ /dev/null @@ -1,232 +0,0 @@ -#!/usr/bin/env python3 - -"""The lambda to decrease/increase ASG desired capacity based on current queue""" - -import logging -from dataclasses import dataclass -from pprint import pformat -from typing import Any, List, Literal, Optional, Tuple - -import boto3 # type: ignore -from lambda_shared import ( - RUNNER_TYPE_LABELS, - CHException, - ClickHouseHelper, - get_parameter_from_ssm, -) - -### Update comment on the change ### -# 4 HOUR - is a balance to get the most precise values -# - Our longest possible running check is around 5h on the worst scenario -# - The long queue won't be wiped out and replaced, so the measurmenet is fine -# - If the data is spoiled by something, we are from the bills perspective -# Changed it to 3 HOUR: in average we have 1h tasks, but p90 is around 2h. -# With 4h we have too much wasted computing time in case of issues with DB -QUEUE_QUERY = f"""SELECT - last_status AS status, - toUInt32(count()) AS length, - labels -FROM -( - SELECT - arraySort(groupArray(status))[-1] AS last_status, - labels, - id, - html_url - FROM default.workflow_jobs - WHERE has(labels, 'self-hosted') - AND hasAny({RUNNER_TYPE_LABELS}, labels) - AND started_at > now() - INTERVAL 3 HOUR - GROUP BY ALL - HAVING last_status IN ('in_progress', 'queued') -) -GROUP BY ALL -ORDER BY labels, last_status""" - - -@dataclass -class Queue: - status: Literal["in_progress", "queued"] - lentgh: int - label: str - - -def get_scales(runner_type: str) -> Tuple[int, int]: - "returns the multipliers for scaling down and up ASG by types" - # Scaling down is quicker on the lack of running jobs than scaling up on - # queue - - # The ASG should deflate almost instantly - scale_down = 1 - # the style checkers have so many noise, so it scales up too quickly - # The 5 was too quick, there are complainings regarding too slow with - # 10. I am trying 7 now. - # 7 still looks a bit slow, so I try 6 - # Let's have it the same as the other ASG - # - # All type of style-checkers should be added very quickly to not block the workflows - # UPDATE THE COMMENT ON CHANGES - scale_up = 3 - if "style" in runner_type: - scale_up = 1 - return scale_down, scale_up - - -CH_CLIENT = None # type: Optional[ClickHouseHelper] - - -def set_capacity( - runner_type: str, queues: List[Queue], client: Any, dry_run: bool = True -) -> None: - assert len(queues) in (1, 2) - assert all(q.label == runner_type for q in queues) - as_groups = client.describe_auto_scaling_groups( - Filters=[ - {"Name": "tag-key", "Values": ["github:runner-type"]}, - {"Name": "tag-value", "Values": [runner_type]}, - ] - )["AutoScalingGroups"] - assert len(as_groups) == 1 - asg = as_groups[0] - running = 0 - queued = 0 - for q in queues: - if q.status == "in_progress": - running = q.lentgh - continue - if q.status == "queued": - queued = q.lentgh - continue - raise ValueError("Queue status is not in ['in_progress', 'queued']") - - scale_down, scale_up = get_scales(runner_type) - # With lyfecycle hooks some instances are actually free because some of - # them are in 'Terminating:Wait' state - effective_capacity = max( - asg["DesiredCapacity"], - len([ins for ins in asg["Instances"] if ins["HealthStatus"] == "Healthy"]), - ) - - # How much nodes are free (positive) or need to be added (negative) - capacity_reserve = effective_capacity - running - queued - stop = False - if capacity_reserve < 0: - # This part is about scaling up - capacity_deficit = -capacity_reserve - # It looks that we are still OK, since no queued jobs exist - stop = stop or queued == 0 - # Are we already at the capacity limits - stop = stop or asg["MaxSize"] <= asg["DesiredCapacity"] - # Let's calculate a new desired capacity - # (capacity_deficit + scale_up - 1) // scale_up : will increase min by 1 - # if there is any capacity_deficit - new_capacity = ( - asg["DesiredCapacity"] + (capacity_deficit + scale_up - 1) // scale_up - ) - new_capacity = max(new_capacity, asg["MinSize"]) - new_capacity = min(new_capacity, asg["MaxSize"]) - # Finally, should the capacity be even changed - stop = stop or asg["DesiredCapacity"] == new_capacity - if stop: - logging.info( - "Do not increase ASG %s capacity, current capacity=%s, effective " - "capacity=%s, maximum capacity=%s, running jobs=%s, queue size=%s", - asg["AutoScalingGroupName"], - asg["DesiredCapacity"], - effective_capacity, - asg["MaxSize"], - running, - queued, - ) - return - - logging.info( - "The ASG %s capacity will be increased to %s, current capacity=%s, " - "effective capacity=%s, maximum capacity=%s, running jobs=%s, queue size=%s", - asg["AutoScalingGroupName"], - new_capacity, - asg["DesiredCapacity"], - effective_capacity, - asg["MaxSize"], - running, - queued, - ) - if not dry_run: - client.set_desired_capacity( - AutoScalingGroupName=asg["AutoScalingGroupName"], - DesiredCapacity=new_capacity, - ) - return - - # Now we will calculate if we need to scale down - stop = stop or asg["DesiredCapacity"] == asg["MinSize"] - new_capacity = asg["DesiredCapacity"] - (capacity_reserve // scale_down) - new_capacity = max(new_capacity, asg["MinSize"]) - new_capacity = min(new_capacity, asg["MaxSize"]) - stop = stop or asg["DesiredCapacity"] == new_capacity - if stop: - logging.info( - "Do not decrease ASG %s capacity, current capacity=%s, effective " - "capacity=%s, minimum capacity=%s, running jobs=%s, queue size=%s", - asg["AutoScalingGroupName"], - asg["DesiredCapacity"], - effective_capacity, - asg["MinSize"], - running, - queued, - ) - return - - logging.info( - "The ASG %s capacity will be decreased to %s, current capacity=%s, effective " - "capacity=%s, minimum capacity=%s, running jobs=%s, queue size=%s", - asg["AutoScalingGroupName"], - new_capacity, - asg["DesiredCapacity"], - effective_capacity, - asg["MinSize"], - running, - queued, - ) - if not dry_run: - client.set_desired_capacity( - AutoScalingGroupName=asg["AutoScalingGroupName"], - DesiredCapacity=new_capacity, - ) - - -def main(dry_run: bool = True) -> None: - logging.getLogger().setLevel(logging.INFO) - asg_client = boto3.client("autoscaling") - try: - global CH_CLIENT - CH_CLIENT = CH_CLIENT or ClickHouseHelper( - get_parameter_from_ssm("clickhouse-test-stat-url"), "play" - ) - queues = CH_CLIENT.select_json_each_row("default", QUEUE_QUERY) - except CHException as ex: - logging.exception( - "Got an exception on insert, tryuing to update the client " - "credentials and repeat", - exc_info=ex, - ) - CH_CLIENT = ClickHouseHelper( - get_parameter_from_ssm("clickhouse-test-stat-url"), "play" - ) - queues = CH_CLIENT.select_json_each_row("default", QUEUE_QUERY) - - logging.info("Received queue data:\n%s", pformat(queues, width=120)) - for runner_type in RUNNER_TYPE_LABELS: - runner_queues = [ - Queue(queue["status"], queue["length"], runner_type) - for queue in queues - if runner_type in queue["labels"] - ] - runner_queues = runner_queues or [Queue("in_progress", 0, runner_type)] - set_capacity(runner_type, runner_queues, asg_client, dry_run) - - -def handler(event: dict, context: Any) -> None: - _ = event - _ = context - return main(False) diff --git a/tests/ci/autoscale_runners_lambda/build_and_deploy_archive.sh b/tests/ci/autoscale_runners_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/autoscale_runners_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/autoscale_runners_lambda/lambda_shared b/tests/ci/autoscale_runners_lambda/lambda_shared deleted file mode 120000 index ba86e090f6c..00000000000 --- a/tests/ci/autoscale_runners_lambda/lambda_shared +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package/lambda_shared \ No newline at end of file diff --git a/tests/ci/autoscale_runners_lambda/requirements.txt b/tests/ci/autoscale_runners_lambda/requirements.txt deleted file mode 100644 index 098e04a9798..00000000000 --- a/tests/ci/autoscale_runners_lambda/requirements.txt +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package diff --git a/tests/ci/autoscale_runners_lambda/test_autoscale.py b/tests/ci/autoscale_runners_lambda/test_autoscale.py deleted file mode 100644 index 3aeab49ffc7..00000000000 --- a/tests/ci/autoscale_runners_lambda/test_autoscale.py +++ /dev/null @@ -1,181 +0,0 @@ -#!/usr/bin/env python - -import unittest -from dataclasses import dataclass -from typing import Any, List - -from app import Queue, set_capacity - - -@dataclass -class TestCase: - name: str - min_size: int - desired_capacity: int - max_size: int - queues: List[Queue] - expected_capacity: int - - -class TestSetCapacity(unittest.TestCase): - class FakeClient: - def __init__(self): - self._expected_data = {} # type: dict - self._expected_capacity = -1 - - @property - def expected_data(self) -> dict: - """a one-time property""" - data, self._expected_data = self._expected_data, {} - return data - - @expected_data.setter - def expected_data(self, value: dict) -> None: - self._expected_data = value - - @property - def expected_capacity(self) -> int: - """a one-time property""" - capacity, self._expected_capacity = self._expected_capacity, -1 - return capacity - - def describe_auto_scaling_groups(self, **kwargs: Any) -> dict: - _ = kwargs - return self.expected_data - - def set_desired_capacity(self, **kwargs: Any) -> None: - self._expected_capacity = kwargs["DesiredCapacity"] - - def data_helper( - self, name: str, min_size: int, desired_capacity: int, max_size: int - ) -> None: - self.expected_data = { - "AutoScalingGroups": [ - { - "AutoScalingGroupName": name, - "DesiredCapacity": desired_capacity, - "MinSize": min_size, - "MaxSize": max_size, - "Instances": [], # necessary for ins["HealthStatus"] check - } - ] - } - - def setUp(self): - self.client = self.FakeClient() - - def test_normal_cases(self): - test_cases = ( - # Do not change capacity - TestCase("noqueue", 1, 13, 20, [Queue("in_progress", 155, "noqueue")], -1), - TestCase("reserve", 1, 13, 20, [Queue("queued", 13, "reserve")], -1), - # Increase capacity - TestCase( - "increase-always", - 1, - 13, - 20, - [Queue("queued", 14, "increase-always")], - 14, - ), - TestCase("increase-1", 1, 13, 20, [Queue("queued", 23, "increase-1")], 17), - TestCase( - "style-checker", 1, 13, 20, [Queue("queued", 19, "style-checker")], 19 - ), - TestCase("increase-2", 1, 13, 20, [Queue("queued", 18, "increase-2")], 15), - TestCase("increase-3", 1, 13, 20, [Queue("queued", 183, "increase-3")], 20), - TestCase( - "increase-w/o reserve", - 1, - 13, - 20, - [ - Queue("in_progress", 11, "increase-w/o reserve"), - Queue("queued", 12, "increase-w/o reserve"), - ], - 17, - ), - TestCase("lower-min", 10, 5, 20, [Queue("queued", 5, "lower-min")], 10), - # Decrease capacity - TestCase("w/reserve", 1, 13, 20, [Queue("queued", 5, "w/reserve")], 5), - TestCase( - "style-checker", 1, 13, 20, [Queue("queued", 5, "style-checker")], 5 - ), - TestCase("w/reserve", 1, 23, 20, [Queue("queued", 17, "w/reserve")], 17), - TestCase("decrease", 1, 13, 20, [Queue("in_progress", 3, "decrease")], 3), - TestCase( - "style-checker", - 1, - 13, - 20, - [Queue("in_progress", 5, "style-checker")], - 5, - ), - ) - for t in test_cases: - self.client.data_helper(t.name, t.min_size, t.desired_capacity, t.max_size) - set_capacity(t.name, t.queues, self.client, False) - self.assertEqual(t.expected_capacity, self.client.expected_capacity, t.name) - - def test_effective_capacity(self): - """Normal cases test increasing w/o considering - effective_capacity much lower than DesiredCapacity""" - test_cases = ( - TestCase( - "desired-overwritten", - 1, - 20, # DesiredCapacity, overwritten by effective_capacity - 50, - [ - Queue("in_progress", 30, "desired-overwritten"), - Queue("queued", 60, "desired-overwritten"), - ], - 40, - ), - ) - for t in test_cases: - self.client.data_helper(t.name, t.min_size, t.desired_capacity, t.max_size) - # we test that effective_capacity is 30 (a half of 60) - data_with_instances = self.client.expected_data - data_with_instances["AutoScalingGroups"][0]["Instances"] = [ - {"HealthStatus": "Healthy" if i % 2 else "Unhealthy"} for i in range(60) - ] - self.client.expected_data = data_with_instances - set_capacity(t.name, t.queues, self.client, False) - self.assertEqual(t.expected_capacity, self.client.expected_capacity, t.name) - - def test_exceptions(self): - test_cases = ( - ( - TestCase( - "different names", - 1, - 1, - 1, - [Queue("queued", 5, "another name")], - -1, - ), - AssertionError, - ), - (TestCase("wrong queue len", 1, 1, 1, [], -1), AssertionError), - ( - TestCase( - "wrong queue", 1, 1, 1, [Queue("wrong", 1, "wrong queue")], -1 # type: ignore - ), - ValueError, - ), - ) - for t, error in test_cases: - with self.assertRaises(error): - self.client.data_helper( - t.name, t.min_size, t.desired_capacity, t.max_size - ) - set_capacity(t.name, t.queues, self.client, False) - - with self.assertRaises(AssertionError): - self.client.expected_data = {"AutoScalingGroups": [1, 2]} - set_capacity( - "wrong number of ASGs", - [Queue("queued", 1, "wrong number of ASGs")], - self.client, - ) diff --git a/tests/ci/build_check.py b/tests/ci/build_check.py index 39f34ed9ccf..77d91c8400b 100644 --- a/tests/ci/build_check.py +++ b/tests/ci/build_check.py @@ -12,7 +12,6 @@ import docker_images_helper from ci_config import CI from env_helper import REPO_COPY, S3_BUILDS_BUCKET, TEMP_PATH from git_helper import Git -from lambda_shared_package.lambda_shared.pr import Labels from pr_info import PRInfo from report import FAILURE, SUCCESS, JobReport, StatusType from stopwatch import Stopwatch @@ -108,7 +107,9 @@ def build_clickhouse( def is_release_pr(pr_info: PRInfo) -> bool: - return Labels.RELEASE in pr_info.labels or Labels.RELEASE_LTS in pr_info.labels + return ( + CI.Labels.RELEASE in pr_info.labels or CI.Labels.RELEASE_LTS in pr_info.labels + ) def get_release_or_pr(pr_info: PRInfo, version: ClickHouseVersion) -> Tuple[str, str]: diff --git a/tests/ci/cancel_and_rerun_workflow_lambda/app.py b/tests/ci/cancel_and_rerun_workflow_lambda/app.py deleted file mode 100644 index 578ade5c8a0..00000000000 --- a/tests/ci/cancel_and_rerun_workflow_lambda/app.py +++ /dev/null @@ -1,376 +0,0 @@ -#!/usr/bin/env python3 - -import json -import time -from base64 import b64decode -from collections import namedtuple -from queue import Queue -from threading import Thread -from typing import Any, Dict, List, Optional - -import requests -from lambda_shared.pr import Labels -from lambda_shared.token import get_cached_access_token - -NEED_RERUN_OR_CANCELL_WORKFLOWS = { - "BackportPR", - "DocsCheck", - "MasterCI", - "PullRequestCI", -} - -MAX_RETRY = 5 - -DEBUG_INFO = {} # type: Dict[str, Any] - - -class Worker(Thread): - def __init__( - self, request_queue: Queue, token: str, ignore_exception: bool = False - ): - Thread.__init__(self) - self.queue = request_queue - self.token = token - self.ignore_exception = ignore_exception - self.response = {} # type: Dict - - def run(self): - m = self.queue.get() - try: - self.response = _exec_get_with_retry(m, self.token) - except Exception as e: - if not self.ignore_exception: - raise - print(f"Exception occured, still continue: {e}") - self.queue.task_done() - - -def _exec_get_with_retry(url: str, token: str) -> dict: - headers = {"Authorization": f"token {token}"} - e = Exception() - for i in range(MAX_RETRY): - try: - response = requests.get(url, headers=headers, timeout=30) - response.raise_for_status() - return response.json() # type: ignore - except Exception as ex: - print("Got exception executing request", ex) - e = ex - time.sleep(i + 1) - - raise requests.HTTPError("Cannot execute GET request with retries") from e - - -WorkflowDescription = namedtuple( - "WorkflowDescription", - [ - "url", - "run_id", - "name", - "head_sha", - "status", - "rerun_url", - "cancel_url", - "conclusion", - ], -) - - -def get_workflows_description_for_pull_request( - pull_request_event: dict, token: str -) -> List[WorkflowDescription]: - head_repo = pull_request_event["head"]["repo"]["full_name"] - head_branch = pull_request_event["head"]["ref"] - print("PR", pull_request_event["number"], "has head ref", head_branch) - - workflows_data = [] - repo_url = pull_request_event["base"]["repo"]["url"] - request_url = f"{repo_url}/actions/runs?per_page=100" - # Get all workflows for the current branch - for i in range(1, 11): - workflows = _exec_get_with_retry( - f"{request_url}&event=pull_request&branch={head_branch}&page={i}", token - ) - if not workflows["workflow_runs"]: - break - workflows_data += workflows["workflow_runs"] - if i == 10: - print("Too many workflows found") - - if not workflows_data: - print("No workflows found by filter") - return [] - - print(f"Total workflows for the branch {head_branch} found: {len(workflows_data)}") - - DEBUG_INFO["workflows"] = [] - workflow_descriptions = [] - for workflow in workflows_data: - # Some time workflow["head_repository"]["full_name"] is None - if workflow["head_repository"] is None: - continue - DEBUG_INFO["workflows"].append( - { - "full_name": workflow["head_repository"]["full_name"], - "name": workflow["name"], - "branch": workflow["head_branch"], - } - ) - # unfortunately we cannot filter workflows from forks in request to API - # so doing it manually - if ( - workflow["head_repository"]["full_name"] == head_repo - and workflow["name"] in NEED_RERUN_OR_CANCELL_WORKFLOWS - ): - workflow_descriptions.append( - WorkflowDescription( - url=workflow["url"], - run_id=workflow["id"], - name=workflow["name"], - head_sha=workflow["head_sha"], - status=workflow["status"], - rerun_url=workflow["rerun_url"], - cancel_url=workflow["cancel_url"], - conclusion=workflow["conclusion"], - ) - ) - - return workflow_descriptions - - -def get_workflow_description_fallback( - pull_request_event: dict, token: str -) -> List[WorkflowDescription]: - head_repo = pull_request_event["head"]["repo"]["full_name"] - head_branch = pull_request_event["head"]["ref"] - print("Get last 500 workflows from API to search related there") - # Fallback for a case of an already deleted branch and no workflows received - repo_url = pull_request_event["base"]["repo"]["url"] - request_url = f"{repo_url}/actions/runs?per_page=100" - q = Queue() # type: Queue - workers = [] - workflows_data = [] - i = 1 - for i in range(1, 6): - q.put(f"{request_url}&page={i}") - worker = Worker(q, token, True) - worker.start() - workers.append(worker) - - for worker in workers: - worker.join() - if not worker.response: - # We ignore get errors, so response can be empty - continue - # Prefilter workflows - workflows_data += [ - wf - for wf in worker.response["workflow_runs"] - if wf["head_repository"] is not None - and wf["head_repository"]["full_name"] == head_repo - and wf["head_branch"] == head_branch - and wf["name"] in NEED_RERUN_OR_CANCELL_WORKFLOWS - ] - - print(f"Total workflows in last 500 actions matches: {len(workflows_data)}") - - DEBUG_INFO["workflows"] = [ - { - "full_name": wf["head_repository"]["full_name"], - "name": wf["name"], - "branch": wf["head_branch"], - } - for wf in workflows_data - ] - - workflow_descriptions = [ - WorkflowDescription( - url=wf["url"], - run_id=wf["id"], - name=wf["name"], - head_sha=wf["head_sha"], - status=wf["status"], - rerun_url=wf["rerun_url"], - cancel_url=wf["cancel_url"], - conclusion=wf["conclusion"], - ) - for wf in workflows_data - ] - - return workflow_descriptions - - -def get_workflow_description(workflow_url: str, token: str) -> WorkflowDescription: - workflow = _exec_get_with_retry(workflow_url, token) - return WorkflowDescription( - url=workflow["url"], - run_id=workflow["id"], - name=workflow["name"], - head_sha=workflow["head_sha"], - status=workflow["status"], - rerun_url=workflow["rerun_url"], - cancel_url=workflow["cancel_url"], - conclusion=workflow["conclusion"], - ) - - -def _exec_post_with_retry(url: str, token: str, json: Optional[Any] = None) -> Any: - headers = {"Authorization": f"token {token}"} - e = Exception() - for i in range(MAX_RETRY): - try: - response = requests.post(url, headers=headers, json=json, timeout=30) - response.raise_for_status() - return response.json() - except Exception as ex: - print("Got exception executing request", ex) - e = ex - time.sleep(i + 1) - - raise requests.HTTPError("Cannot execute POST request with retry") from e - - -def exec_workflow_url(urls_to_post, token): - for url in urls_to_post: - print("Post for workflow workflow using url", url) - _exec_post_with_retry(url, token) - print("Workflow post finished") - - -def main(event): - token = get_cached_access_token() - DEBUG_INFO["event"] = event - if event["isBase64Encoded"]: - event_data = json.loads(b64decode(event["body"])) - else: - event_data = json.loads(event["body"]) - - print("Got event for PR", event_data["number"]) - action = event_data["action"] - print("Got action", event_data["action"]) - pull_request = event_data["pull_request"] - label = "" - if action == "labeled": - label = event_data["label"]["name"] - print("Added label:", label) - - print("PR has labels", {label["name"] for label in pull_request["labels"]}) - if action == "opened" or ( - action == "labeled" and pull_request["created_at"] == pull_request["updated_at"] - ): - print("Freshly opened PR, nothing to do") - return - - if action == "closed" or label == Labels.DO_NOT_TEST: - print("PR merged/closed or manually labeled 'do not test', will kill workflows") - workflow_descriptions = get_workflows_description_for_pull_request( - pull_request, token - ) - workflow_descriptions = ( - workflow_descriptions - or get_workflow_description_fallback(pull_request, token) - ) - urls_to_cancel = [] - for workflow_description in workflow_descriptions: - if ( - workflow_description.status != "completed" - and workflow_description.conclusion != "cancelled" - ): - urls_to_cancel.append(workflow_description.cancel_url) - print(f"Found {len(urls_to_cancel)} workflows to cancel") - exec_workflow_url(urls_to_cancel, token) - return - - if label == Labels.CAN_BE_TESTED: - print("PR marked with can be tested label, rerun workflow") - workflow_descriptions = get_workflows_description_for_pull_request( - pull_request, token - ) - workflow_descriptions = ( - workflow_descriptions - or get_workflow_description_fallback(pull_request, token) - ) - if not workflow_descriptions: - print("Not found any workflows") - return - - workflow_descriptions.sort(key=lambda x: x.run_id) # type: ignore - most_recent_workflow = workflow_descriptions[-1] - print("Latest workflow", most_recent_workflow) - if ( - most_recent_workflow.status != "completed" - and most_recent_workflow.conclusion != "cancelled" - ): - print("Latest workflow is not completed, cancelling") - exec_workflow_url([most_recent_workflow.cancel_url], token) - print("Cancelled") - - for _ in range(45): - # If the number of retries is changed: tune the lambda limits accordingly - latest_workflow_desc = get_workflow_description( - most_recent_workflow.url, token - ) - print("Checking latest workflow", latest_workflow_desc) - if latest_workflow_desc.status in ("completed", "cancelled"): - print("Finally latest workflow done, going to rerun") - exec_workflow_url([most_recent_workflow.rerun_url], token) - print("Rerun finished, exiting") - break - print("Still have strange status") - time.sleep(3) - return - - if action == "edited": - print("PR is edited - do nothing") - # error, _ = check_pr_description( - # pull_request["body"], pull_request["base"]["repo"]["full_name"] - # ) - # if error: - # print( - # f"The PR's body is wrong, is going to comment it. The error is: {error}" - # ) - # post_json = { - # "body": "This is an automatic comment. The PR descriptions does not " - # f"match the [template]({pull_request['base']['repo']['html_url']}/" - # "blob/master/.github/PULL_REQUEST_TEMPLATE.md?plain=1).\n\n" - # f"Please, edit it accordingly.\n\nThe error is: {error}" - # } - # _exec_post_with_retry(pull_request["comments_url"], token, json=post_json) - return - - if action == "synchronize": - print("PR is synchronized, going to stop old actions") - workflow_descriptions = get_workflows_description_for_pull_request( - pull_request, token - ) - workflow_descriptions = ( - workflow_descriptions - or get_workflow_description_fallback(pull_request, token) - ) - urls_to_cancel = [] - for workflow_description in workflow_descriptions: - if ( - workflow_description.status != "completed" - and workflow_description.conclusion != "cancelled" - and workflow_description.head_sha != pull_request["head"]["sha"] - ): - urls_to_cancel.append(workflow_description.cancel_url) - print(f"Found {len(urls_to_cancel)} workflows to cancel") - exec_workflow_url(urls_to_cancel, token) - return - - print("Nothing to do") - - -def handler(event, _): - try: - main(event) - - return { - "statusCode": 200, - "headers": {"Content-Type": "application/json"}, - "body": '{"status": "OK"}', - } - finally: - for name, value in DEBUG_INFO.items(): - print(f"Value of {name}: ", value) diff --git a/tests/ci/cancel_and_rerun_workflow_lambda/build_and_deploy_archive.sh b/tests/ci/cancel_and_rerun_workflow_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/cancel_and_rerun_workflow_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/cancel_and_rerun_workflow_lambda/lambda_shared b/tests/ci/cancel_and_rerun_workflow_lambda/lambda_shared deleted file mode 120000 index ba86e090f6c..00000000000 --- a/tests/ci/cancel_and_rerun_workflow_lambda/lambda_shared +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package/lambda_shared \ No newline at end of file diff --git a/tests/ci/cancel_and_rerun_workflow_lambda/requirements.txt b/tests/ci/cancel_and_rerun_workflow_lambda/requirements.txt deleted file mode 100644 index 4cb3fba0f7b..00000000000 --- a/tests/ci/cancel_and_rerun_workflow_lambda/requirements.txt +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package[token] diff --git a/tests/ci/cherry_pick.py b/tests/ci/cherry_pick.py index 459be12ada0..a7fc6d02853 100644 --- a/tests/ci/cherry_pick.py +++ b/tests/ci/cherry_pick.py @@ -38,7 +38,7 @@ from env_helper import TEMP_PATH from get_robot_token import get_best_robot_token from git_helper import GIT_PREFIX, git_runner, is_shallow from github_helper import GitHub, PullRequest, PullRequests, Repository -from lambda_shared_package.lambda_shared.pr import Labels +from ci_config import Labels from ssh import SSHKey diff --git a/tests/ci/ci.py b/tests/ci/ci.py index af2f4c0a1fc..e30062c32ff 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -6,6 +6,7 @@ import os import re import subprocess import sys +import time from dataclasses import dataclass from pathlib import Path from typing import Any, Dict, List, Optional @@ -15,7 +16,7 @@ import upload_result_helper from build_check import get_release_or_pr from ci_config import CI from ci_metadata import CiMetadata -from ci_utils import GHActions, normalize_string +from ci_utils import GHActions, normalize_string, Utils from clickhouse_helper import ( CiLogsCredentials, ClickHouseHelper, @@ -47,12 +48,25 @@ from git_helper import GIT_PREFIX, Git from git_helper import Runner as GitRunner from github_helper import GitHub from pr_info import PRInfo -from report import ERROR, FAILURE, PENDING, SUCCESS, BuildResult, JobReport, TestResult +from report import ( + ERROR, + FAILURE, + PENDING, + SUCCESS, + BuildResult, + JobReport, + TestResult, + OK, + JOB_STARTED_TEST_NAME, + JOB_FINISHED_TEST_NAME, + FAIL, +) from s3_helper import S3Helper from stopwatch import Stopwatch from tee_popen import TeePopen from ci_cache import CiCache from ci_settings import CiSettings +from ci_buddy import CIBuddy from version_helper import get_version_from_repo # pylint: disable=too-many-lines @@ -81,6 +95,12 @@ def parse_args(parser: argparse.ArgumentParser) -> argparse.Namespace: action="store_true", help="Action that configures ci run. Calculates digests, checks job to be executed, generates json output", ) + parser.add_argument( + "--workflow", + default="", + type=str, + help="Workflow Name, to be provided with --configure for workflow-specific CI runs", + ) parser.add_argument( "--update-gh-statuses", action="store_true", @@ -261,7 +281,10 @@ def check_missing_images_on_dockerhub( return result -def _pre_action(s3, indata, pr_info): +def _pre_action(s3, job_name, batch, indata, pr_info): + no_cache = CiSettings.create_from_run_config(indata).no_ci_cache + print("Clear dmesg") + Utils.clear_dmesg() CommitStatusData.cleanup() JobReport.cleanup() BuildResult.cleanup() @@ -270,7 +293,10 @@ def _pre_action(s3, indata, pr_info): # for release/master branches reports must be from the same branch report_prefix = "" if pr_info.is_master or pr_info.is_release: - report_prefix = normalize_string(pr_info.head_ref) + # do not set report prefix for scheduled or dispatched wf (in case it started from feature branch while + # testing), otherwise reports won't be found + if not (pr_info.is_scheduled or pr_info.is_dispatched): + report_prefix = normalize_string(pr_info.head_ref) print( f"Use report prefix [{report_prefix}], pr_num [{pr_info.number}], head_ref [{pr_info.head_ref}]" ) @@ -278,6 +304,90 @@ def _pre_action(s3, indata, pr_info): ci_cache.dump_run_config(indata) + to_be_skipped = False + skip_status = SUCCESS + # check if job was run already + if CI.is_build_job(job_name): + # this is a build job - check if a build report is present + build_result = ( + BuildResult.load_any(job_name, pr_info.number, pr_info.head_ref) + if not no_cache + else None + ) + if build_result: + if build_result.status == SUCCESS: + to_be_skipped = True + else: + print( + "Build report found but status is unsuccessful - will try to rerun" + ) + print("::group::Build Report") + print(build_result.as_json()) + print("::endgroup::") + else: + # this is a test job - check if GH commit status or cache record is present + commit = get_commit(GitHub(get_best_robot_token(), per_page=100), pr_info.sha) + # rerun helper check + # FIXME: Find a way to identify if job restarted manually (by developer) or by automatic workflow restart (died spot-instance) + # disable rerun check for the former + if job_name not in ( + CI.JobNames.BUILD_CHECK, + ): # we might want to rerun build report job + rerun_helper = RerunHelper(commit, _get_ext_check_name(job_name)) + if rerun_helper.is_already_finished_by_status(): + print("WARNING: Rerunning job with GH status ") + status = rerun_helper.get_finished_status() + assert status + print("::group::Commit Status") + print(status) + print("::endgroup::") + to_be_skipped = True + skip_status = status.state + + # ci cache check + if not to_be_skipped and not no_cache: + ci_cache = CiCache(s3, indata["jobs_data"]["digests"]).update() + job_config = CI.get_job_config(job_name) + if ci_cache.is_successful( + job_name, + batch, + job_config.num_batches, + job_config.required_on_release_branch, + ): + print("CICache record has be found - job will be skipped") + job_status = ci_cache.get_successful( + job_name, batch, job_config.num_batches + ) + assert job_status, "BUG" + _create_gh_status( + commit, + job_name, + batch, + job_config.num_batches, + job_status, + ) + to_be_skipped = True + # skip_status = SUCCESS already there + GHActions.print_in_group("Commit Status Data", job_status) + + # create pre report + jr = JobReport.create_pre_report(status=skip_status, job_skipped=to_be_skipped) + jr.dump() + + if not to_be_skipped: + print("push start record to ci db") + prepared_events = prepare_tests_results_for_clickhouse( + pr_info, + [TestResult(JOB_STARTED_TEST_NAME, OK)], + SUCCESS, + 0.0, + JobReport.get_start_time_from_current(), + "", + _get_ext_check_name(job_name), + ) + ClickHouseHelper().insert_events_into( + db="default", table="checks", events=prepared_events + ) print(f"Pre action done. Report files [{reports_files}] have been downloaded") @@ -322,8 +432,8 @@ def _mark_success_action( # do nothing, exit without failure print(f"ERROR: no status file for job [{job}]") - if job_config.run_always or job_config.run_by_label: - print(f"Job [{job}] runs always or by label in CI - do not cache") + if job_config.run_by_label or not job_config.has_digest(): + print(f"Job [{job}] has no digest or run by label in CI - do not cache") else: if pr_info.is_master: pass @@ -419,6 +529,7 @@ def _configure_jobs( pr_info: PRInfo, ci_settings: CiSettings, skip_jobs: bool, + workflow_name: str = "", dry_run: bool = False, ) -> CiCache: """ @@ -436,18 +547,27 @@ def _configure_jobs( is_docs_only=pr_info.has_changes_in_documentation_only(), is_master=pr_info.is_master, is_pr=pr_info.is_pr, + workflow_name=workflow_name, ) else: job_configs = {} - # filter jobs in accordance with ci settings - job_configs = ci_settings.apply( - job_configs, - pr_info.is_release, - is_pr=pr_info.is_pr, - is_mq=pr_info.is_merge_queue, - labels=pr_info.labels, - ) + if not workflow_name: + # filter jobs in accordance with ci settings + job_configs = ci_settings.apply( + job_configs, + pr_info.is_release, + is_pr=pr_info.is_pr, + is_mq=pr_info.is_merge_queue, + labels=pr_info.labels, + ) + + # add all job batches to job's to_do batches + for _job, job_config in job_configs.items(): + batches = [] + for batch in range(job_config.num_batches): + batches.append(batch) + job_config.batches = batches # check jobs in ci cache ci_cache = CiCache.calc_digests_and_create( @@ -547,7 +667,17 @@ def _update_gh_statuses_action(indata: Dict, s3: S3Helper) -> None: except Exception as e: raise e print("Going to update overall CI report") - set_status_comment(commit, pr_info) + for retry in range(2): + try: + set_status_comment(commit, pr_info) + break + except Exception as e: + print( + f"WARNING: Failed to update CI Running status, attempt [{retry + 1}], exception [{e}]" + ) + time.sleep(1) + else: + print("ERROR: All retry attempts failed.") print("... CI report update - done") @@ -991,17 +1121,23 @@ def main() -> int: pr_info, ci_settings, args.skip_jobs, + args.workflow, ) + ci_cache.print_status() + if IS_CI and pr_info.is_pr and not ci_settings.no_ci_cache: + ci_cache.filter_out_not_affected_jobs() + ci_cache.print_status() if IS_CI and not pr_info.is_merge_queue: - # wait for pending jobs to be finished, await_jobs is a long blocking call - ci_cache.await_pending_jobs(pr_info.is_release) - if pr_info.is_release: + if pr_info.is_release and pr_info.is_push_event: print("Release/master: CI Cache add pending records for all todo jobs") ci_cache.push_pending_all(pr_info.is_release) + # wait for pending jobs to be finished, await_jobs is a long blocking call + ci_cache.await_pending_jobs(pr_info.is_release) + # conclude results result["git_ref"] = git_ref result["version"] = version @@ -1027,99 +1163,45 @@ def main() -> int: ### PRE action: start elif args.pre: assert indata, "Run config must be provided via --infile" - _pre_action(s3, indata, pr_info) + _pre_action(s3, args.job_name, args.batch, indata, pr_info) ### RUN action: start elif args.run: assert indata - ci_settings = CiSettings.create_from_run_config(indata) + job_report = JobReport.load() check_name = args.job_name check_name_with_group = _get_ext_check_name(check_name) print( f"Check if rerun for name: [{check_name}], extended name [{check_name_with_group}]" ) - previous_status = None - if CI.is_build_job(check_name): - # this is a build job - check if a build report is present - build_result = ( - BuildResult.load_any(check_name, pr_info.number, pr_info.head_ref) - if not ci_settings.no_ci_cache - else None - ) - if build_result: - if build_result.status == SUCCESS: - previous_status = build_result.status - else: - # FIXME: Consider reusing failures for build jobs. - # Just remove this if/else - that makes build job starting and failing immediately - print( - "Build report found but status is unsuccessful - will try to rerun" - ) - print("::group::Build Report") - print(build_result.as_json()) - print("::endgroup::") - else: - # this is a test job - check if GH commit status or cache record is present - commit = get_commit( - GitHub(get_best_robot_token(), per_page=100), pr_info.sha - ) - # rerun helper check - # FIXME: Find a way to identify if job restarted manually (by developer) or by automatic workflow restart (died spot-instance) - # disable rerun check for the former - if check_name not in ( - CI.JobNames.BUILD_CHECK, - ): # we might want to rerun build report job - rerun_helper = RerunHelper(commit, check_name_with_group) - if rerun_helper.is_already_finished_by_status(): - print("WARNING: Rerunning job with GH status ") - status = rerun_helper.get_finished_status() - assert status - print("::group::Commit Status") - print(status) - print("::endgroup::") - previous_status = status.state - - # ci cache check - if not previous_status and not ci_settings.no_ci_cache: - ci_cache = CiCache(s3, indata["jobs_data"]["digests"]).update() - job_config = CI.get_job_config(check_name) - if ci_cache.is_successful( - check_name, - args.batch, - job_config.num_batches, - job_config.required_on_release_branch, - ): - job_status = ci_cache.get_successful( - check_name, args.batch, job_config.num_batches - ) - assert job_status, "BUG" - _create_gh_status( - commit, - check_name, - args.batch, - job_config.num_batches, - job_status, - ) - previous_status = job_status.status - GHActions.print_in_group("Commit Status Data", job_status) - - if previous_status and not args.force: + if job_report.job_skipped and not args.force: print( - f"Commit status or Build Report is already present - job will be skipped with status: [{previous_status}]" + f"Commit status or Build Report is already present - job will be skipped with status: [{job_report.status}]" ) - if previous_status == SUCCESS: + if job_report.status == SUCCESS: exit_code = 0 else: exit_code = 1 else: exit_code = _run_test(check_name, args.run_command) + job_report = JobReport.load() if JobReport.exist() else None + assert ( + job_report + ), "BUG. There must be job report either real report, or pre-report if job was killed" + job_report.exit_code = exit_code + job_report.dump() ### RUN action: end ### POST action: start elif args.post: job_report = JobReport.load() if JobReport.exist() else None - if job_report: + assert ( + job_report + ), "BUG. There must be job report either real report, or pre-report if job was killed" + error_description = "" + if not job_report.pre_report: + # it's a real job report ch_helper = ClickHouseHelper() check_url = "" @@ -1196,7 +1278,6 @@ def main() -> int: pr_info, dump_to_file=True, ) - print(f"Job report url: [{check_url}]") prepared_events = prepare_tests_results_for_clickhouse( pr_info, @@ -1219,9 +1300,60 @@ def main() -> int: indata["build"], ch_helper, ) + elif job_report.job_skipped: + print(f"Skipped after rerun check {[args.job_name]} - do nothing") else: - # no job report - print(f"No job report for {[args.job_name]} - do nothing") + print(f"ERROR: Job was killed - generate evidence") + job_report.update_duration() + ret_code = os.getenv("JOB_EXIT_CODE", "") + if ret_code: + try: + job_report.exit_code = int(ret_code) + except ValueError: + pass + if Utils.is_killed_with_oom(): + print("WARNING: OOM while job execution") + print(subprocess.run("sudo dmesg -T", check=False)) + error_description = f"Out Of Memory, exit_code {job_report.exit_code}" + else: + error_description = f"Unknown, exit_code {job_report.exit_code}" + CIBuddy().post_job_error( + error_description + f" after {int(job_report.duration)}s", + job_name=_get_ext_check_name(args.job_name), + ) + if CI.is_test_job(args.job_name): + gh = GitHub(get_best_robot_token(), per_page=100) + commit = get_commit(gh, pr_info.sha) + post_commit_status( + commit, + ERROR, + "", + "Error: " + error_description, + _get_ext_check_name(args.job_name), + pr_info, + dump_to_file=True, + ) + + if not job_report.job_skipped: + print("push finish record to ci db") + prepared_events = prepare_tests_results_for_clickhouse( + pr_info, + [ + TestResult( + JOB_FINISHED_TEST_NAME, + FAIL if error_description else OK, + raw_logs=error_description or None, + ) + ], + SUCCESS if not error_description else ERROR, + 0.0, + JobReport.get_start_time_from_current(), + "", + _get_ext_check_name(args.job_name), + ) + ClickHouseHelper().insert_events_into( + db="default", table="checks", events=prepared_events + ) ### POST action: end ### MARK SUCCESS action: start diff --git a/tests/ci/ci_buddy.py b/tests/ci/ci_buddy.py new file mode 100644 index 00000000000..dfb5885270a --- /dev/null +++ b/tests/ci/ci_buddy.py @@ -0,0 +1,175 @@ +import argparse +import json +import os +from typing import Union, Dict + +import boto3 +import requests +from botocore.exceptions import ClientError + +from pr_info import PRInfo +from ci_utils import Shell, GHActions + + +class CIBuddy: + _HEADERS = {"Content-Type": "application/json"} + + def __init__(self, dry_run=False): + self.repo = os.getenv("GITHUB_REPOSITORY", "") + self.dry_run = dry_run + res = self._get_webhooks() + self.test_channel = "" + self.dev_ci_channel = "" + if res: + self.test_channel = json.loads(res)["test_channel"] + self.dev_ci_channel = json.loads(res)["ci_channel"] + self.job_name = os.getenv("CHECK_NAME", "unknown") + pr_info = PRInfo() + self.pr_number = pr_info.number + self.head_ref = pr_info.head_ref + self.commit_url = pr_info.commit_html_url + self.sha = pr_info.sha[:10] + + def check_workflow(self): + GHActions.print_workflow_results() + res = GHActions.get_workflow_job_result(GHActions.ActionsNames.RunConfig) + if res != GHActions.ActionStatuses.SUCCESS: + self.post_job_error("Workflow Configuration Failed", critical=True) + + @staticmethod + def _get_webhooks(): + name = "ci_buddy_web_hooks" + + session = boto3.Session(region_name="us-east-1") # Replace with your region + ssm_client = session.client("ssm") + json_string = None + try: + response = ssm_client.get_parameter( + Name=name, + WithDecryption=True, # Set to True if the parameter is a SecureString + ) + json_string = response["Parameter"]["Value"] + except ClientError as e: + print(f"An error occurred: {e}") + + return json_string + + def post(self, message, dry_run=None): + if dry_run is None: + dry_run = self.dry_run + print(f"Posting slack message, dry_run [{dry_run}]") + if dry_run: + url = self.test_channel + else: + url = self.dev_ci_channel + data = {"text": message} + try: + requests.post(url, headers=self._HEADERS, data=json.dumps(data), timeout=10) + except Exception as e: + print(f"ERROR: Failed to post message, ex {e}") + + def _post_formatted( + self, title: str, body: Union[Dict, str], with_wf_link: bool + ) -> None: + message = title + if isinstance(body, dict): + for name, value in body.items(): + if "commit_sha" in name: + value = ( + f"" + ) + message += f" *{name}*: {value}\n" + else: + message += body + "\n" + run_id = os.getenv("GITHUB_RUN_ID", "") + if with_wf_link and run_id: + message += f" *workflow*: \n" + self.post(message) + + def post_info( + self, title: str, body: Union[Dict, str], with_wf_link: bool = True + ) -> None: + title_extended = f":white_circle: *{title}*\n\n" + self._post_formatted(title_extended, body, with_wf_link) + + def post_done( + self, title: str, body: Union[Dict, str], with_wf_link: bool = True + ) -> None: + title_extended = f":white_check_mark: *{title}*\n\n" + self._post_formatted(title_extended, body, with_wf_link) + + def post_warning( + self, title: str, body: Union[Dict, str], with_wf_link: bool = True + ) -> None: + title_extended = f":warning: *{title}*\n\n" + self._post_formatted(title_extended, body, with_wf_link) + + def post_critical( + self, title: str, body: Union[Dict, str], with_wf_link: bool = True + ) -> None: + title_extended = f":black_circle: *{title}*\n\n" + self._post_formatted(title_extended, body, with_wf_link) + + def post_job_error( + self, + error_description: str, + job_name: str = "", + with_instance_info: bool = True, + with_wf_link: bool = True, + critical: bool = False, + ) -> None: + instance_id, instance_type = "unknown", "unknown" + if with_instance_info: + instance_id = Shell.run("ec2metadata --instance-id") or instance_id + instance_type = Shell.run("ec2metadata --instance-type") or instance_type + if not job_name: + job_name = os.getenv("CHECK_NAME", "unknown") + sign = ":red_circle:" if not critical else ":black_circle:" + line_err = f"{sign} *Error: {error_description}*\n\n" + line_ghr = f" *Runner:* `{instance_type}`, `{instance_id}`\n" + line_job = f" *Job:* `{job_name}`\n" + line_pr_ = f" *PR:* , <{self.commit_url}|{self.sha}>\n" + line_br_ = ( + f" *Branch:* `{self.head_ref}`, <{self.commit_url}|{self.sha}>\n" + ) + message = line_err + message += line_job + if with_instance_info: + message += line_ghr + if self.pr_number > 0: + message += line_pr_ + else: + message += line_br_ + run_id = os.getenv("GITHUB_RUN_ID", "") + if with_wf_link and run_id: + message += f" *workflow*: \n" + self.post(message) + + +def parse_args(): + parser = argparse.ArgumentParser("CI Buddy bot notifies about CI events") + parser.add_argument( + "--check-wf-status", + action="store_true", + help="Checks workflow status", + ) + parser.add_argument( + "--test", + action="store_true", + help="for test and debug", + ) + parser.add_argument( + "--dry-run", + action="store_true", + help="dry run mode", + ) + return parser.parse_args(), parser + + +if __name__ == "__main__": + args, parser = parse_args() + + if args.test: + CIBuddy(dry_run=True).post_job_error("TEst") + elif args.check_wf_status: + CIBuddy(dry_run=args.dry_run).check_workflow() diff --git a/tests/ci/ci_cache.py b/tests/ci/ci_cache.py index 8ee0ae54385..85eabb84f9f 100644 --- a/tests/ci/ci_cache.py +++ b/tests/ci/ci_cache.py @@ -520,6 +520,35 @@ class CiCache: self.RecordType.SUCCESSFUL, job, batch, num_batches, release_branch ) + def has_evidence(self, job: str, job_config: CI.JobConfig) -> bool: + """ + checks if the job has been seen in master/release CI + function is to be used to check if change did not affect the job + :param job_config: + :param job: + :return: + """ + return ( + self.is_successful( + job=job, + batch=0, + num_batches=job_config.num_batches, + release_branch=True, + ) + or self.is_pending( + job=job, + batch=0, + num_batches=job_config.num_batches, + release_branch=True, + ) + or self.is_failed( + job=job, + batch=0, + num_batches=job_config.num_batches, + release_branch=True, + ) + ) + def is_failed( self, job: str, batch: int, num_batches: int, release_branch: bool ) -> bool: @@ -609,7 +638,14 @@ class CiCache: pushes pending records for all jobs that supposed to be run """ for job, job_config in self.jobs_to_do.items(): - if job_config.run_always: + if ( + job in self.jobs_to_wait + or not job_config.has_digest() + or job_config.disable_await + ): + # 1. "job in self.jobs_to_wait" - this job already has a pending record in cache + # 2. "not job_config.has_digest()" - cache is not used for these jobs + # 3. "job_config.disable_await" - await is explicitly disabled continue pending_state = PendingState(time.time(), run_url=GITHUB_RUN_URL) assert job_config.batches @@ -674,6 +710,62 @@ class CiCache: bucket=S3_BUILDS_BUCKET, file_path=result_json_path, s3_path=s3_path ) + def filter_out_not_affected_jobs(self): + """ + Filter is to be applied in PRs to remove jobs that are not affected by the change + :return: + """ + remove_from_workflow = [] + required_builds = [] + has_test_jobs_to_skip = False + for job_name, job_config in self.jobs_to_do.items(): + if CI.is_test_job(job_name) and job_name != CI.JobNames.BUILD_CHECK: + if job_config.reference_job_name: + reference_name = job_config.reference_job_name + reference_config = CI.JOB_CONFIGS[reference_name] + else: + reference_name = job_name + reference_config = job_config + if self.has_evidence( + job=reference_name, + job_config=reference_config, + ): + remove_from_workflow.append(job_name) + has_test_jobs_to_skip = True + else: + required_builds += ( + job_config.required_builds if job_config.required_builds else [] + ) + if has_test_jobs_to_skip: + # If there are tests to skip, it means builds are not affected as well. + # No need to test builds. Let's keep all builds required for test jobs and skip the others + for job_name, job_config in self.jobs_to_do.items(): + if CI.is_build_job(job_name): + if job_name not in required_builds: + remove_from_workflow.append(job_name) + + for job in remove_from_workflow: + print(f"Filter job [{job}] - not affected by the change") + if job in self.jobs_to_do: + del self.jobs_to_do[job] + if job in self.jobs_to_wait: + del self.jobs_to_wait[job] + if job in self.jobs_to_skip: + self.jobs_to_skip.remove(job) + + # special handling for the special job: BUILD_CHECK + has_builds = False + for job in list(self.jobs_to_do) + self.jobs_to_skip: + if CI.is_build_job(job): + has_builds = True + break + if not has_builds: + if CI.JobNames.BUILD_CHECK in self.jobs_to_do: + print( + f"Filter job [{CI.JobNames.BUILD_CHECK}] - no builds are required in the workflow" + ) + del self.jobs_to_do[CI.JobNames.BUILD_CHECK] + def await_pending_jobs(self, is_release: bool, dry_run: bool = False) -> None: """ await pending jobs to be finished @@ -688,17 +780,13 @@ class CiCache: # TIMEOUT * MAX_ROUNDS_TO_WAIT must be less than 6h (GH job timeout) with a room for rest RunConfig work TIMEOUT = 3000 # 50 min MAX_ROUNDS_TO_WAIT = 6 - MAX_JOB_NUM_TO_WAIT = 3 round_cnt = 0 - # FIXME: temporary experiment: lets enable await for PR' workflows but for a shorter time if not is_release: - MAX_ROUNDS_TO_WAIT = 3 + # in PRs we can wait only for builds, TIMEOUT*MAX_ROUNDS_TO_WAIT=100min is enough + MAX_ROUNDS_TO_WAIT = 2 - while ( - len(self.jobs_to_wait) > MAX_JOB_NUM_TO_WAIT - and round_cnt < MAX_ROUNDS_TO_WAIT - ): + while round_cnt < MAX_ROUNDS_TO_WAIT: round_cnt += 1 GHActions.print_in_group( f"Wait pending jobs, round [{round_cnt}/{MAX_ROUNDS_TO_WAIT}]:", @@ -740,6 +828,10 @@ class CiCache: f"Job [{job_name}_[{batch}/{num_batches}]] is not pending anymore" ) job_config.batches.remove(batch) + if not job_config.batches: + print(f"Remove job [{job_name}] from jobs_to_do") + self.jobs_to_skip.append(job_name) + del self.jobs_to_do[job_name] else: print( f"NOTE: Job [{job_name}:{batch}] finished failed - do not add to ready" @@ -750,9 +842,7 @@ class CiCache: await_finished.add(job_name) for job in await_finished: - self.jobs_to_skip.append(job) del self.jobs_to_wait[job] - del self.jobs_to_do[job] if not dry_run: expired_sec = int(time.time()) - start_at @@ -816,3 +906,87 @@ class CiCache: self.jobs_to_wait[job] = job_config return self + + +if __name__ == "__main__": + # for testing + job_digest = { + "package_release": "bbbd3519d1", + "package_aarch64": "bbbd3519d1", + "package_asan": "bbbd3519d1", + "package_ubsan": "bbbd3519d1", + "package_tsan": "bbbd3519d1", + "package_msan": "bbbd3519d1", + "package_debug": "bbbd3519d1", + "package_release_coverage": "bbbd3519d1", + "binary_release": "bbbd3519d1", + "binary_tidy": "bbbd3519d1", + "binary_darwin": "bbbd3519d1", + "binary_aarch64": "bbbd3519d1", + "binary_aarch64_v80compat": "bbbd3519d1", + "binary_freebsd": "bbbd3519d1", + "binary_darwin_aarch64": "bbbd3519d1", + "binary_ppc64le": "bbbd3519d1", + "binary_amd64_compat": "bbbd3519d1", + "binary_amd64_musl": "bbbd3519d1", + "binary_riscv64": "bbbd3519d1", + "binary_s390x": "bbbd3519d1", + "binary_loongarch64": "bbbd3519d1", + "Builds": "f5dffeecb8", + "Install packages (release)": "ba0c89660e", + "Install packages (aarch64)": "ba0c89660e", + "Stateful tests (asan)": "32a9a1aba9", + "Stateful tests (tsan)": "32a9a1aba9", + "Stateful tests (msan)": "32a9a1aba9", + "Stateful tests (ubsan)": "32a9a1aba9", + "Stateful tests (debug)": "32a9a1aba9", + "Stateful tests (release)": "32a9a1aba9", + "Stateful tests (coverage)": "32a9a1aba9", + "Stateful tests (aarch64)": "32a9a1aba9", + "Stateful tests (release, ParallelReplicas)": "32a9a1aba9", + "Stateful tests (debug, ParallelReplicas)": "32a9a1aba9", + "Stateless tests (asan)": "deb6778b88", + "Stateless tests (tsan)": "deb6778b88", + "Stateless tests (msan)": "deb6778b88", + "Stateless tests (ubsan)": "deb6778b88", + "Stateless tests (debug)": "deb6778b88", + "Stateless tests (release)": "deb6778b88", + "Stateless tests (coverage)": "deb6778b88", + "Stateless tests (aarch64)": "deb6778b88", + "Stateless tests (release, old analyzer, s3, DatabaseReplicated)": "deb6778b88", + "Stateless tests (debug, s3 storage)": "deb6778b88", + "Stateless tests (tsan, s3 storage)": "deb6778b88", + "Stress test (debug)": "aa298abf10", + "Stress test (tsan)": "aa298abf10", + "Upgrade check (debug)": "5ce4d3ee02", + "Integration tests (asan, old analyzer)": "42e58be3aa", + "Integration tests (tsan)": "42e58be3aa", + "Integration tests (aarch64)": "42e58be3aa", + "Integration tests flaky check (asan)": "42e58be3aa", + "Compatibility check (release)": "ecb69d8c4b", + "Compatibility check (aarch64)": "ecb69d8c4b", + "Unit tests (release)": "09d00b702e", + "Unit tests (asan)": "09d00b702e", + "Unit tests (msan)": "09d00b702e", + "Unit tests (tsan)": "09d00b702e", + "Unit tests (ubsan)": "09d00b702e", + "AST fuzzer (debug)": "c38ebf947f", + "AST fuzzer (asan)": "c38ebf947f", + "AST fuzzer (msan)": "c38ebf947f", + "AST fuzzer (tsan)": "c38ebf947f", + "AST fuzzer (ubsan)": "c38ebf947f", + "Stateless tests flaky check (asan)": "deb6778b88", + "Performance Comparison (release)": "a8a7179258", + "ClickBench (release)": "45c07c4aa6", + "ClickBench (aarch64)": "45c07c4aa6", + "Docker server image": "6a24d5b187", + "Docker keeper image": "6a24d5b187", + "Docs check": "4764154c62", + "Fast test": "cb269133f2", + "Style check": "ffffffffff", + "Stateful tests (ubsan, ParallelReplicas)": "32a9a1aba9", + "Stress test (msan)": "aa298abf10", + "Upgrade check (asan)": "5ce4d3ee02", + } + ci_cache = CiCache(job_digests=job_digest, cache_enabled=True, s3=S3Helper()) + ci_cache.update() diff --git a/tests/ci/ci_config.py b/tests/ci/ci_config.py index bef43083a35..a9bdb639835 100644 --- a/tests/ci/ci_config.py +++ b/tests/ci/ci_config.py @@ -13,6 +13,9 @@ class CI: each config item in the below dicts should be an instance of JobConfig class or inherited from it """ + MAX_TOTAL_FAILURES_BEFORE_BLOCKING_CI = 5 + MAX_TOTAL_FAILURES_PER_JOB_BEFORE_BLOCKING_CI = 2 + # reimport types to CI class so that they visible as CI.* and mypy is happy # pylint:disable=useless-import-alias,reimported,import-outside-toplevel from ci_definitions import BuildConfig as BuildConfig @@ -29,10 +32,27 @@ class CI: from ci_definitions import MQ_JOBS as MQ_JOBS from ci_definitions import WorkflowStages as WorkflowStages from ci_definitions import Runners as Runners + from ci_utils import Envs as Envs + from ci_utils import Utils as Utils + from ci_utils import GHActions as GHActions + from ci_definitions import Labels as Labels + from ci_definitions import TRUSTED_CONTRIBUTORS as TRUSTED_CONTRIBUTORS + from ci_definitions import WorkFlowNames as WorkFlowNames + from ci_utils import CATEGORY_TO_LABEL as CATEGORY_TO_LABEL # Jobs that run for doc related updates _DOCS_CHECK_JOBS = [JobNames.DOCS_CHECK, JobNames.STYLE_CHECK] + WORKFLOW_CONFIGS = { + WorkFlowNames.JEPSEN: LabelConfig( + run_jobs=[ + BuildNames.BINARY_RELEASE, + JobNames.JEPSEN_KEEPER, + JobNames.JEPSEN_SERVER, + ] + ) + } # type: Dict[str, LabelConfig] + TAG_CONFIGS = { Tags.DO_NOT_TEST_LABEL: LabelConfig(run_jobs=[JobNames.STYLE_CHECK]), Tags.CI_SET_ARM: LabelConfig( @@ -42,24 +62,14 @@ class CI: JobNames.INTEGRATION_TEST_ARM, ] ), - Tags.CI_SET_REQUIRED: LabelConfig(run_jobs=REQUIRED_CHECKS), + Tags.CI_SET_REQUIRED: LabelConfig( + run_jobs=REQUIRED_CHECKS + + [build for build in BuildNames if build != BuildNames.FUZZERS] + ), Tags.CI_SET_BUILDS: LabelConfig( run_jobs=[JobNames.STYLE_CHECK, JobNames.BUILD_CHECK] + [build for build in BuildNames if build != BuildNames.FUZZERS] ), - Tags.CI_SET_NON_REQUIRED: LabelConfig( - run_jobs=[job for job in JobNames if job not in REQUIRED_CHECKS] - ), - Tags.CI_SET_OLD_ANALYZER: LabelConfig( - run_jobs=[ - JobNames.STYLE_CHECK, - JobNames.FAST_TEST, - BuildNames.PACKAGE_RELEASE, - BuildNames.PACKAGE_ASAN, - JobNames.STATELESS_TEST_OLD_ANALYZER_S3_REPLICATED_RELEASE, - JobNames.INTEGRATION_TEST_ASAN_OLD_ANALYZER, - ] - ), Tags.CI_SET_SYNC: LabelConfig( run_jobs=[ BuildNames.PACKAGE_ASAN, @@ -69,7 +79,7 @@ class CI: JobNames.STATEFUL_TEST_ASAN, ] ), - } + } # type: Dict[str, LabelConfig] JOB_CONFIGS: Dict[str, JobConfig] = { BuildNames.PACKAGE_RELEASE: CommonJobConfigs.BUILD.with_properties( @@ -311,19 +321,19 @@ class CI: random_bucket="parrepl_with_sanitizer", ), JobNames.STATELESS_TEST_ASAN: CommonJobConfigs.STATELESS_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_ASAN], num_batches=4 + required_builds=[BuildNames.PACKAGE_ASAN], num_batches=2 ), JobNames.STATELESS_TEST_TSAN: CommonJobConfigs.STATELESS_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_TSAN], num_batches=5 + required_builds=[BuildNames.PACKAGE_TSAN], num_batches=4 ), JobNames.STATELESS_TEST_MSAN: CommonJobConfigs.STATELESS_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_MSAN], num_batches=6 + required_builds=[BuildNames.PACKAGE_MSAN], num_batches=4 ), JobNames.STATELESS_TEST_UBSAN: CommonJobConfigs.STATELESS_TEST.with_properties( required_builds=[BuildNames.PACKAGE_UBSAN], num_batches=2 ), JobNames.STATELESS_TEST_DEBUG: CommonJobConfigs.STATELESS_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_DEBUG], num_batches=5 + required_builds=[BuildNames.PACKAGE_DEBUG], num_batches=2 ), JobNames.STATELESS_TEST_RELEASE: CommonJobConfigs.STATELESS_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE], @@ -339,14 +349,14 @@ class CI: required_builds=[BuildNames.PACKAGE_RELEASE], num_batches=4 ), JobNames.STATELESS_TEST_S3_DEBUG: CommonJobConfigs.STATELESS_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_DEBUG], num_batches=6 + required_builds=[BuildNames.PACKAGE_DEBUG], num_batches=2 ), JobNames.STATELESS_TEST_AZURE_ASAN: CommonJobConfigs.STATELESS_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_ASAN], num_batches=4, release_only=True + required_builds=[BuildNames.PACKAGE_ASAN], num_batches=3, release_only=True ), JobNames.STATELESS_TEST_S3_TSAN: CommonJobConfigs.STATELESS_TEST.with_properties( required_builds=[BuildNames.PACKAGE_TSAN], - num_batches=5, + num_batches=4, ), JobNames.STRESS_TEST_DEBUG: CommonJobConfigs.STRESS_TEST.with_properties( required_builds=[BuildNames.PACKAGE_DEBUG], @@ -410,7 +420,10 @@ class CI: release_only=True, ), JobNames.INTEGRATION_TEST_FLAKY: CommonJobConfigs.INTEGRATION_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_ASAN], pr_only=True + required_builds=[BuildNames.PACKAGE_ASAN], + pr_only=True, + # TODO: approach with reference job names does not work because digest may not be calculated if job skipped in wf + # reference_job_name=JobNames.INTEGRATION_TEST_TSAN, ), JobNames.COMPATIBILITY_TEST: CommonJobConfigs.COMPATIBILITY_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE], @@ -452,7 +465,11 @@ class CI: required_builds=[BuildNames.PACKAGE_UBSAN], ), JobNames.STATELESS_TEST_FLAKY_ASAN: CommonJobConfigs.STATELESS_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_ASAN], pr_only=True, timeout=3600 + required_builds=[BuildNames.PACKAGE_ASAN], + pr_only=True, + timeout=3600, + # TODO: approach with reference job names does not work because digest may not be calculated if job skipped in wf + # reference_job_name=JobNames.STATELESS_TEST_RELEASE, ), JobNames.JEPSEN_KEEPER: JobConfig( required_builds=[BuildNames.BINARY_RELEASE], @@ -593,16 +610,25 @@ class CI: @classmethod def get_workflow_jobs_with_configs( - cls, is_mq: bool, is_docs_only: bool, is_master: bool, is_pr: bool + cls, + is_mq: bool, + is_docs_only: bool, + is_master: bool, + is_pr: bool, + workflow_name: str, ) -> Dict[str, JobConfig]: """ get a list of all jobs for a workflow with configs """ - jobs = [] if is_mq: jobs = MQ_JOBS elif is_docs_only: jobs = cls._DOCS_CHECK_JOBS + elif workflow_name: + assert ( + workflow_name in cls.WORKFLOW_CONFIGS + ), "Workflow name if provided must be configured in WORKFLOW_CONFIGS" + jobs = list(cls.WORKFLOW_CONFIGS[workflow_name].run_jobs) else: # add all jobs jobs = list(cls.JOB_CONFIGS) @@ -637,7 +663,7 @@ class CI: @classmethod def is_test_job(cls, job: str) -> bool: - return not cls.is_build_job(job) and job != cls.JobNames.STYLE_CHECK + return not cls.is_build_job(job) @classmethod def is_docs_job(cls, job: str) -> bool: diff --git a/tests/ci/ci_definitions.py b/tests/ci/ci_definitions.py index 48e1280d939..149177ecba5 100644 --- a/tests/ci/ci_definitions.py +++ b/tests/ci/ci_definitions.py @@ -7,6 +7,53 @@ from ci_utils import WithIter from integration_test_images import IMAGES +class Labels: + PR_BUGFIX = "pr-bugfix" + PR_CRITICAL_BUGFIX = "pr-critical-bugfix" + CAN_BE_TESTED = "can be tested" + DO_NOT_TEST = "do not test" + MUST_BACKPORT = "pr-must-backport" + MUST_BACKPORT_CLOUD = "pr-must-backport-cloud" + JEPSEN_TEST = "jepsen-test" + SKIP_MERGEABLE_CHECK = "skip mergeable check" + PR_BACKPORT = "pr-backport" + PR_BACKPORTS_CREATED = "pr-backports-created" + PR_BACKPORTS_CREATED_CLOUD = "pr-backports-created-cloud" + PR_CHERRYPICK = "pr-cherrypick" + PR_CI = "pr-ci" + PR_FEATURE = "pr-feature" + PR_SYNCED_TO_CLOUD = "pr-synced-to-cloud" + PR_SYNC_UPSTREAM = "pr-sync-upstream" + RELEASE = "release" + RELEASE_LTS = "release-lts" + SUBMODULE_CHANGED = "submodule changed" + + # automatic backport for critical bug fixes + AUTO_BACKPORT = {"pr-critical-bugfix"} + + +TRUSTED_CONTRIBUTORS = { + e.lower() + for e in [ + "amosbird", + "azat", # SEMRush + "bharatnc", # Many contributions. + "cwurm", # ClickHouse, Inc + "den-crane", # Documentation contributor + "ildus", # adjust, ex-pgpro + "nvartolomei", # Seasoned contributor, CloudFlare + "taiyang-li", + "ucasFL", # Amos Bird's friend + "thomoco", # ClickHouse, Inc + "tonickkozlov", # Cloudflare + "tylerhannan", # ClickHouse, Inc + "tsolodov", # ClickHouse, Inc + "justindeguzman", # ClickHouse, Inc + "XuJia0210", # ClickHouse, Inc + ] +} + + class WorkflowStages(metaclass=WithIter): """ Stages of GitHUb actions workflow @@ -55,12 +102,18 @@ class Tags(metaclass=WithIter): CI_SET_ARM = "ci_set_arm" CI_SET_REQUIRED = "ci_set_required" CI_SET_BUILDS = "ci_set_builds" - CI_SET_NON_REQUIRED = "ci_set_non_required" - CI_SET_OLD_ANALYZER = "ci_set_old_analyzer" libFuzzer = "libFuzzer" +class WorkFlowNames(metaclass=WithIter): + """ + CI WorkFlow Names for custom CI runs + """ + + JEPSEN = "JepsenWorkflow" + + class BuildNames(metaclass=WithIter): """ Build' job names @@ -284,8 +337,12 @@ class JobConfig: # GH Runner type (tag from @Runners) runner_type: str - # used for config validation in ci unittests + # used in ci unittests for config validation job_name_keyword: str = "" + # name of another job that (if provided) should be used to check if job was affected by the change or not (in CiCache.has_evidence(job=@reference_job_name) call) + # for example: "Stateless flaky check" can use reference_job_name="Stateless tests (release)". "Stateless flaky check" does not run on master + # and there cannot be an evidence for it, so instead "Stateless tests (release)" job name can be used to check the evidence + reference_job_name: str = "" # builds required for the job (applicable for test jobs) required_builds: Optional[List[str]] = None # build config for the build job (applicable for builds) @@ -302,6 +359,8 @@ class JobConfig: run_by_label: str = "" # to run always regardless of the job digest or/and label run_always: bool = False + # disables CI await for a given job + disable_await: bool = False # if the job needs to be run on the release branch, including master (building packages, docker server). # NOTE: Subsequent runs on the same branch with the similar digest are still considered skip-able. required_on_release_branch: bool = False @@ -327,6 +386,9 @@ class JobConfig: assert self.required_builds return self.required_builds[0] + def has_digest(self) -> bool: + return self.digest != DigestConfig() + class CommonJobConfigs: """ @@ -343,6 +405,7 @@ class CommonJobConfigs: ], ), runner_type=Runners.STYLE_CHECKER_ARM, + disable_await=True, ) COMPATIBILITY_TEST = JobConfig( job_name_keyword="compatibility", @@ -378,7 +441,7 @@ class CommonJobConfigs: ), run_command='functional_test_check.py "$CHECK_NAME"', runner_type=Runners.FUNC_TESTER, - timeout=10800, + timeout=9000, ) STATEFUL_TEST = JobConfig( job_name_keyword="stateful", @@ -440,7 +503,12 @@ class CommonJobConfigs: ) ASTFUZZER_TEST = JobConfig( job_name_keyword="ast", - digest=DigestConfig(), + digest=DigestConfig( + include_paths=[ + "./tests/ci/ast_fuzzer_check.py", + ], + docker=["clickhouse/fuzzer"], + ), run_command="ast_fuzzer_check.py", run_always=True, runner_type=Runners.FUZZER_UNIT_TESTER, diff --git a/tests/ci/ci_runners_metrics_lambda/app.py b/tests/ci/ci_runners_metrics_lambda/app.py deleted file mode 100644 index 47161215a97..00000000000 --- a/tests/ci/ci_runners_metrics_lambda/app.py +++ /dev/null @@ -1,164 +0,0 @@ -#!/usr/bin/env python3 -""" -Lambda function to: - - calculate number of running runners - - cleaning dead runners from GitHub - - terminating stale lost runners in EC2 -""" - -import argparse -import sys -from typing import Dict - -import boto3 # type: ignore -from lambda_shared import RUNNER_TYPE_LABELS, RunnerDescriptions, list_runners -from lambda_shared.token import ( - get_access_token_by_key_app, - get_cached_access_token, - get_key_and_app_from_aws, -) - -UNIVERSAL_LABEL = "universal" - - -def handler(event, context): - _ = event - _ = context - main(get_cached_access_token(), True) - - -def group_runners_by_tag( - listed_runners: RunnerDescriptions, -) -> Dict[str, RunnerDescriptions]: - result = {} # type: Dict[str, RunnerDescriptions] - - def add_to_result(tag, runner): - if tag not in result: - result[tag] = [] - result[tag].append(runner) - - for runner in listed_runners: - if UNIVERSAL_LABEL in runner.tags: - # Do not proceed other labels if UNIVERSAL_LABEL is included - add_to_result(UNIVERSAL_LABEL, runner) - continue - - for tag in runner.tags: - if tag in RUNNER_TYPE_LABELS: - add_to_result(tag, runner) - break - else: - add_to_result("unlabeled", runner) - return result - - -def push_metrics_to_cloudwatch( - listed_runners: RunnerDescriptions, group_name: str -) -> None: - client = boto3.client("cloudwatch") - namespace = "RunnersMetrics" - metrics_data = [] - busy_runners = sum( - 1 for runner in listed_runners if runner.busy and not runner.offline - ) - dimensions = [{"Name": "group", "Value": group_name}] - metrics_data.append( - { - "MetricName": "BusyRunners", - "Value": busy_runners, - "Unit": "Count", - "Dimensions": dimensions, - } - ) - total_active_runners = sum(1 for runner in listed_runners if not runner.offline) - metrics_data.append( - { - "MetricName": "ActiveRunners", - "Value": total_active_runners, - "Unit": "Count", - "Dimensions": dimensions, - } - ) - total_runners = len(listed_runners) - metrics_data.append( - { - "MetricName": "TotalRunners", - "Value": total_runners, - "Unit": "Count", - "Dimensions": dimensions, - } - ) - if total_active_runners == 0: - busy_ratio = 100.0 - else: - busy_ratio = busy_runners / total_active_runners * 100 - - metrics_data.append( - { - "MetricName": "BusyRunnersRatio", - "Value": busy_ratio, - "Unit": "Percent", - "Dimensions": dimensions, - } - ) - - client.put_metric_data(Namespace=namespace, MetricData=metrics_data) - - -def main( - access_token: str, - push_to_cloudwatch: bool, -) -> None: - gh_runners = list_runners(access_token) - grouped_runners = group_runners_by_tag(gh_runners) - for group, group_runners in grouped_runners.items(): - if push_to_cloudwatch: - print(f"Pushing metrics for group '{group}'") - push_metrics_to_cloudwatch(group_runners, group) - else: - print(group, f"({len(group_runners)})") - for runner in group_runners: - print("\t", runner) - - -if __name__ == "__main__": - parser = argparse.ArgumentParser(description="Get list of runners and their states") - parser.add_argument( - "-p", "--private-key-path", help="Path to file with private key" - ) - parser.add_argument("-k", "--private-key", help="Private key") - parser.add_argument( - "-a", "--app-id", type=int, help="GitHub application ID", required=True - ) - parser.add_argument( - "--push-to-cloudwatch", - action="store_true", - help="Push metrics for active and busy runners to cloudwatch", - ) - - args = parser.parse_args() - - if not args.private_key_path and not args.private_key: - print( - "Either --private-key-path or --private-key must be specified", - file=sys.stderr, - ) - - if args.private_key_path and args.private_key: - print( - "Either --private-key-path or --private-key must be specified", - file=sys.stderr, - ) - - if args.private_key: - private_key = args.private_key - elif args.private_key_path: - with open(args.private_key_path, "r", encoding="utf-8") as key_file: - private_key = key_file.read() - else: - print("Attempt to get key and id from AWS secret manager") - private_key, args.app_id = get_key_and_app_from_aws() - - token = get_access_token_by_key_app(private_key, args.app_id) - - main(token, args.push_to_cloudwatch) diff --git a/tests/ci/ci_runners_metrics_lambda/build_and_deploy_archive.sh b/tests/ci/ci_runners_metrics_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/ci_runners_metrics_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/ci_runners_metrics_lambda/lambda_shared b/tests/ci/ci_runners_metrics_lambda/lambda_shared deleted file mode 120000 index ba86e090f6c..00000000000 --- a/tests/ci/ci_runners_metrics_lambda/lambda_shared +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package/lambda_shared \ No newline at end of file diff --git a/tests/ci/ci_runners_metrics_lambda/requirements.txt b/tests/ci/ci_runners_metrics_lambda/requirements.txt deleted file mode 100644 index e2b16067a93..00000000000 --- a/tests/ci/ci_runners_metrics_lambda/requirements.txt +++ /dev/null @@ -1,2 +0,0 @@ -../lambda_shared_package -../lambda_shared_package[token] diff --git a/tests/ci/ci_settings.py b/tests/ci/ci_settings.py index 7b2dd12c310..d6e9765ceb7 100644 --- a/tests/ci/ci_settings.py +++ b/tests/ci/ci_settings.py @@ -160,7 +160,8 @@ class CiSettings: else: return False - if self.exclude_keywords: + # do not exclude builds + if self.exclude_keywords and not CI.is_build_job(job): for keyword in self.exclude_keywords: if keyword in normalize_string(job): print(f"Job [{job}] matches Exclude keyword [{keyword}] - deny") @@ -168,7 +169,8 @@ class CiSettings: to_deny = False if self.include_keywords: - if job == CI.JobNames.STYLE_CHECK: + # do not exclude builds + if job == CI.JobNames.STYLE_CHECK or CI.is_build_job(job): # never exclude Style Check by include keywords return True for keyword in self.include_keywords: @@ -232,11 +234,4 @@ class CiSettings: for job in add_parents: res[job] = job_configs[job] - for job, job_config in res.items(): - batches = [] - for batch in range(job_config.num_batches): - if not self.job_batches or batch in self.job_batches: - batches.append(batch) - job_config.batches = batches - return res diff --git a/tests/ci/ci_utils.py b/tests/ci/ci_utils.py index e7034d0b104..d42091fb0da 100644 --- a/tests/ci/ci_utils.py +++ b/tests/ci/ci_utils.py @@ -1,7 +1,55 @@ +import json import os +import re +import subprocess +import time from contextlib import contextmanager from pathlib import Path -from typing import Any, Iterator, List, Union +from typing import Any, Iterator, List, Union, Optional, Sequence, Tuple + +import requests + + +class Envs: + GITHUB_REPOSITORY = os.getenv("GITHUB_REPOSITORY", "ClickHouse/ClickHouse") + WORKFLOW_RESULT_FILE = os.getenv( + "WORKFLOW_RESULT_FILE", "/tmp/workflow_results.json" + ) + + +LABEL_CATEGORIES = { + "pr-backward-incompatible": ["Backward Incompatible Change"], + "pr-bugfix": [ + "Bug Fix", + "Bug Fix (user-visible misbehavior in an official stable release)", + "Bug Fix (user-visible misbehaviour in official stable or prestable release)", + "Bug Fix (user-visible misbehavior in official stable or prestable release)", + ], + "pr-critical-bugfix": ["Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC)"], + "pr-build": [ + "Build/Testing/Packaging Improvement", + "Build Improvement", + "Build/Testing Improvement", + "Build", + "Packaging Improvement", + ], + "pr-documentation": [ + "Documentation (changelog entry is not required)", + "Documentation", + ], + "pr-feature": ["New Feature"], + "pr-improvement": ["Improvement"], + "pr-not-for-changelog": [ + "Not for changelog (changelog entry is not required)", + "Not for changelog", + ], + "pr-performance": ["Performance Improvement"], + "pr-ci": ["CI Fix or Improvement (changelog entry is not required)"], +} + +CATEGORY_TO_LABEL = { + c: lb for lb, categories in LABEL_CATEGORIES.items() for c in categories +} class WithIter(type): @@ -35,6 +83,47 @@ def normalize_string(string: str) -> str: class GHActions: + class ActionsNames: + RunConfig = "RunConfig" + + class ActionStatuses: + ERROR = "error" + FAILURE = "failure" + PENDING = "pending" + SUCCESS = "success" + + @classmethod + def _get_workflow_results(cls): + if not Path(Envs.WORKFLOW_RESULT_FILE).exists(): + print( + f"ERROR: Failed to get workflow results from file [{Envs.WORKFLOW_RESULT_FILE}]" + ) + return {} + with open(Envs.WORKFLOW_RESULT_FILE, "r", encoding="utf-8") as json_file: + try: + res = json.load(json_file) + except json.JSONDecodeError as e: + print(f"ERROR: json decoder exception {e}") + json_file.seek(0) + print(" File content:") + print(json_file.read()) + return {} + return res + + @classmethod + def print_workflow_results(cls): + res = cls._get_workflow_results() + results = [f"{job}: {data['result']}" for job, data in res.items()] + cls.print_in_group("Workflow results", results) + + @classmethod + def get_workflow_job_result(cls, wf_job_name: str) -> Optional[str]: + res = cls._get_workflow_results() + if wf_job_name in res: + return res[wf_job_name]["result"] # type: ignore + else: + return None + @staticmethod def print_in_group(group_name: str, lines: Union[Any, List[Any]]) -> None: lines = list(lines) @@ -42,3 +131,228 @@ class GHActions: for line in lines: print(line) print("::endgroup::") + + @staticmethod + def get_commit_status_by_name( + token: str, commit_sha: str, status_name: Union[str, Sequence] + ) -> str: + assert len(token) == 40 + assert len(commit_sha) == 40 + assert is_hex(commit_sha) + assert not is_hex(token) + url = f"https://api.github.com/repos/{Envs.GITHUB_REPOSITORY}/commits/{commit_sha}/statuses?per_page={200}" + headers = { + "Authorization": f"token {token}", + "Accept": "application/vnd.github.v3+json", + } + response = requests.get(url, headers=headers, timeout=5) + + if isinstance(status_name, str): + status_name = (status_name,) + if response.status_code == 200: + assert "next" not in response.links, "Response truncated" + statuses = response.json() + for status in statuses: + if status["context"] in status_name: + return status["state"] # type: ignore + return "" + + @staticmethod + def check_wf_completed(token: str, commit_sha: str) -> bool: + headers = { + "Authorization": f"token {token}", + "Accept": "application/vnd.github.v3+json", + } + url = f"https://api.github.com/repos/{Envs.GITHUB_REPOSITORY}/commits/{commit_sha}/check-runs?per_page={100}" + + for i in range(3): + try: + response = requests.get(url, headers=headers, timeout=5) + response.raise_for_status() + # assert "next" not in response.links, "Response truncated" + + data = response.json() + assert data["check_runs"], "?" + + for check in data["check_runs"]: + if check["status"] != "completed": + print( + f" Check workflow status: Check not completed [{check['name']}]" + ) + return False + return True + except Exception as e: + print(f"ERROR: exception after attempt [{i}]: {e}") + time.sleep(1) + + return False + + @staticmethod + def get_pr_url_by_branch(repo, branch): + get_url_cmd = ( + f"gh pr list --repo {repo} --head {branch} --json url --jq '.[0].url'" + ) + url = Shell.run(get_url_cmd) + if not url: + print(f"ERROR: PR nor found, branch [{branch}]") + return url + + +class Shell: + @classmethod + def run_strict(cls, command): + res = subprocess.run( + command, + shell=True, + stdout=subprocess.PIPE, + stderr=subprocess.PIPE, + text=True, + check=True, + ) + return res.stdout.strip() + + @classmethod + def run(cls, command, check=False, dry_run=False): + if dry_run: + print(f"Dry-ryn. Would run command [{command}]") + return "" + print(f"Run command [{command}]") + res = "" + result = subprocess.run( + command, + shell=True, + stdout=subprocess.PIPE, + stderr=subprocess.PIPE, + text=True, + check=False, + ) + if result.returncode == 0: + print(f"stdout: {result.stdout.strip()}") + res = result.stdout + else: + print( + f"ERROR: stdout: {result.stdout.strip()}, stderr: {result.stderr.strip()}" + ) + if check: + assert result.returncode == 0 + return res.strip() + + @classmethod + def run_as_daemon(cls, command): + print(f"Run daemon command [{command}]") + subprocess.Popen(command.split(" ")) # pylint:disable=consider-using-with + return 0, "" + + @classmethod + def check(cls, command): + result = subprocess.run( + command, + shell=True, + stdout=subprocess.PIPE, + stderr=subprocess.PIPE, + text=True, + check=False, + ) + return result.returncode == 0 + + +class Utils: + @staticmethod + def get_failed_tests_number(description: str) -> Optional[int]: + description = description.lower() + + pattern = r"fail:\s*(\d+)\s*(?=,|$)" + match = re.search(pattern, description) + if match: + return int(match.group(1)) + return None + + @staticmethod + def is_killed_with_oom(): + if Shell.check( + "sudo dmesg -T | grep -q -e 'Out of memory: Killed process' -e 'oom_reaper: reaped process' -e 'oom-kill:constraint=CONSTRAINT_NONE'" + ): + return True + return False + + @staticmethod + def clear_dmesg(): + Shell.run("sudo dmesg --clear ||:") + + @staticmethod + def check_pr_description(pr_body: str, repo_name: str) -> Tuple[str, str]: + """The function checks the body to being properly formatted according to + .github/PULL_REQUEST_TEMPLATE.md, if the first returned string is not empty, + then there is an error.""" + lines = list(map(lambda x: x.strip(), pr_body.split("\n") if pr_body else [])) + lines = [re.sub(r"\s+", " ", line) for line in lines] + + # Check if body contains "Reverts ClickHouse/ClickHouse#36337" + if [ + True for line in lines if re.match(rf"\AReverts {repo_name}#[\d]+\Z", line) + ]: + return "", LABEL_CATEGORIES["pr-not-for-changelog"][0] + + category = "" + entry = "" + description_error = "" + + i = 0 + while i < len(lines): + if re.match(r"(?i)^[#>*_ ]*change\s*log\s*category", lines[i]): + i += 1 + if i >= len(lines): + break + # Can have one empty line between header and the category + # itself. Filter it out. + if not lines[i]: + i += 1 + if i >= len(lines): + break + category = re.sub(r"^[-*\s]*", "", lines[i]) + i += 1 + + # Should not have more than one category. Require empty line + # after the first found category. + if i >= len(lines): + break + if lines[i]: + second_category = re.sub(r"^[-*\s]*", "", lines[i]) + description_error = ( + "More than one changelog category specified: " + f"'{category}', '{second_category}'" + ) + return description_error, category + + elif re.match( + r"(?i)^[#>*_ ]*(short\s*description|change\s*log\s*entry)", lines[i] + ): + i += 1 + # Can have one empty line between header and the entry itself. + # Filter it out. + if i < len(lines) and not lines[i]: + i += 1 + # All following lines until empty one are the changelog entry. + entry_lines = [] + while i < len(lines) and lines[i]: + entry_lines.append(lines[i]) + i += 1 + entry = " ".join(entry_lines) + # Don't accept changelog entries like '...'. + entry = re.sub(r"[#>*_.\- ]", "", entry) + # Don't accept changelog entries like 'Close #12345'. + entry = re.sub(r"^[\w\-\s]{0,10}#?\d{5,6}\.?$", "", entry) + else: + i += 1 + + if not category: + description_error = "Changelog category is empty" + # Filter out the PR categories that are not for changelog. + elif "(changelog entry is not required)" in category: + pass # to not check the rest of the conditions + elif category not in CATEGORY_TO_LABEL: + description_error, category = f"Category '{category}' is not valid", "" + elif not entry: + description_error = f"Changelog entry required for category '{category}'" + + return description_error, category diff --git a/tests/ci/clean_lost_instances_lambda/app.py b/tests/ci/clean_lost_instances_lambda/app.py deleted file mode 100644 index 4accc14f7ae..00000000000 --- a/tests/ci/clean_lost_instances_lambda/app.py +++ /dev/null @@ -1,336 +0,0 @@ -#!/usr/bin/env python3 -""" -Lambda function to: - - calculate number of running runners - - cleaning dead runners from GitHub - - terminating stale lost runners in EC2 -""" - -import argparse -import sys -from dataclasses import dataclass -from datetime import datetime -from typing import Dict, List - -import boto3 # type: ignore -import requests -from botocore.exceptions import ClientError # type: ignore -from lambda_shared import ( - RUNNER_TYPE_LABELS, - RunnerDescription, - RunnerDescriptions, - list_runners, -) -from lambda_shared.token import ( - get_access_token_by_key_app, - get_cached_access_token, - get_key_and_app_from_aws, -) - -UNIVERSAL_LABEL = "universal" - - -@dataclass -class LostInstance: - counter: int - seen: datetime - - def set_offline(self) -> None: - now = datetime.now() - if now.timestamp() <= self.seen.timestamp() + 120: - # the instance is offline for more than 2 minutes, so we increase - # the counter - self.counter += 1 - else: - self.counter = 1 - self.seen = now - - @property - def recently_offline(self) -> bool: - """Returns True if the instance has been seen less than 5 minutes ago""" - return datetime.now().timestamp() <= self.seen.timestamp() + 300 - - @property - def stable_offline(self) -> bool: - return self.counter >= 3 - - -LOST_INSTANCES = {} # type: Dict["str", LostInstance] - - -def get_dead_runners_in_ec2(runners: RunnerDescriptions) -> RunnerDescriptions: - """Returns instances that are offline/dead in EC2, or not found in EC2""" - ids = { - runner.name: runner - for runner in runners - # Only `i-deadbead123` are valid names for an instance ID - if runner.name.startswith("i-") and runner.offline and not runner.busy - } - if not ids: - return [] - - # Delete all offline runners with wrong name - result_to_delete = [ - runner - for runner in runners - if not ids.get(runner.name) and runner.offline and not runner.busy - ] - - client = boto3.client("ec2") - - i = 0 - inc = 100 - - print("Checking ids: ", " ".join(ids.keys())) - instances_statuses = [] - while i < len(ids.keys()): - try: - instances_statuses.append( - client.describe_instance_status( - InstanceIds=list(ids.keys())[i : i + inc] - ) - ) - # It applied only if all ids exist in EC2 - i += inc - except ClientError as e: - # The list of non-existent instances is in the message: - # The instance IDs 'i-069b1c256c06cf4e3, i-0f26430432b044035, - # i-0faa2ff44edbc147e, i-0eccf2514585045ec, i-0ee4ee53e0daa7d4a, - # i-07928f15acd473bad, i-0eaddda81298f9a85' do not exist - message = e.response["Error"]["Message"] - if message.startswith("The instance IDs '") and message.endswith( - "' do not exist" - ): - non_existent = message[18:-14].split(", ") - for n in non_existent: - result_to_delete.append(ids.pop(n)) - else: - raise - - found_instances = set([]) - print("Response", instances_statuses) - for instances_status in instances_statuses: - for instance_status in instances_status["InstanceStatuses"]: - if instance_status["InstanceState"]["Name"] in ("pending", "running"): - found_instances.add(instance_status["InstanceId"]) - - print("Found instances", found_instances) - for runner in result_to_delete: - print("Instance", runner.name, "is not alive, going to remove it") - for instance_id, runner in ids.items(): - if instance_id not in found_instances: - print("Instance", instance_id, "is not found in EC2, going to remove it") - result_to_delete.append(runner) - return result_to_delete - - -def handler(event, context): - _ = event - _ = context - main(get_cached_access_token(), True) - - -def delete_runner(access_token: str, runner: RunnerDescription) -> bool: - headers = { - "Authorization": f"token {access_token}", - "Accept": "application/vnd.github.v3+json", - } - - response = requests.delete( - f"https://api.github.com/orgs/ClickHouse/actions/runners/{runner.id}", - headers=headers, - timeout=30, - ) - response.raise_for_status() - print(f"Response code deleting {runner.name} is {response.status_code}") - return bool(response.status_code == 204) - - -def get_lost_ec2_instances(runners: RunnerDescriptions) -> List[str]: - global LOST_INSTANCES - now = datetime.now() - client = boto3.client("ec2") - reservations = client.describe_instances( - Filters=[ - {"Name": "tag-key", "Values": ["github:runner-type"]}, - {"Name": "instance-state-name", "Values": ["pending", "running"]}, - ], - )["Reservations"] - # flatten the reservation into instances - instances = [ - instance - for reservation in reservations - for instance in reservation["Instances"] - ] - offline_runner_names = { - runner.name for runner in runners if runner.offline and not runner.busy - } - runner_names = {runner.name for runner in runners} - - def offline_instance(iid: str) -> None: - if iid in LOST_INSTANCES: - LOST_INSTANCES[iid].set_offline() - return - LOST_INSTANCES[iid] = LostInstance(1, now) - - for instance in instances: - # Do not consider instances started 20 minutes ago as problematic - if now.timestamp() - instance["LaunchTime"].timestamp() < 1200: - continue - - runner_type = [ - tag["Value"] - for tag in instance["Tags"] - if tag["Key"] == "github:runner-type" - ][0] - # If there's no necessary labels in runner type it's fine - if not (UNIVERSAL_LABEL in runner_type or runner_type in RUNNER_TYPE_LABELS): - continue - - if instance["InstanceId"] in offline_runner_names: - offline_instance(instance["InstanceId"]) - continue - - if ( - instance["State"]["Name"] == "running" - and not instance["InstanceId"] in runner_names - ): - offline_instance(instance["InstanceId"]) - - instance_ids = [instance["InstanceId"] for instance in instances] - # clean out long unseen instances - LOST_INSTANCES = { - instance_id: stats - for instance_id, stats in LOST_INSTANCES.items() - if stats.recently_offline and instance_id in instance_ids - } - print("The remained LOST_INSTANCES: ", LOST_INSTANCES) - - return [ - instance_id - for instance_id, stats in LOST_INSTANCES.items() - if stats.stable_offline - ] - - -def continue_lifecycle_hooks(delete_offline_runners: bool) -> None: - """The function to trigger CONTINUE for instances' lifectycle hooks""" - client = boto3.client("ec2") - reservations = client.describe_instances( - Filters=[ - {"Name": "tag-key", "Values": ["github:runner-type"]}, - {"Name": "instance-state-name", "Values": ["shutting-down", "terminated"]}, - ], - )["Reservations"] - # flatten the reservation into instances - terminated_instances = [ - instance["InstanceId"] - for reservation in reservations - for instance in reservation["Instances"] - ] - - asg_client = boto3.client("autoscaling") - as_groups = asg_client.describe_auto_scaling_groups( - Filters=[{"Name": "tag-key", "Values": ["github:runner-type"]}] - )["AutoScalingGroups"] - for asg in as_groups: - lifecycle_hooks = [ - lch - for lch in asg_client.describe_lifecycle_hooks( - AutoScalingGroupName=asg["AutoScalingGroupName"] - )["LifecycleHooks"] - if lch["LifecycleTransition"] == "autoscaling:EC2_INSTANCE_TERMINATING" - ] - if not lifecycle_hooks: - continue - for instance in asg["Instances"]: - continue_instance = False - if instance["LifecycleState"] == "Terminating:Wait": - if instance["HealthStatus"] == "Unhealthy": - print(f"The instance {instance['InstanceId']} is Unhealthy") - continue_instance = True - elif ( - instance["HealthStatus"] == "Healthy" - and instance["InstanceId"] in terminated_instances - ): - print( - f"The instance {instance['InstanceId']} is already terminated" - ) - continue_instance = True - if continue_instance: - if delete_offline_runners: - for lch in lifecycle_hooks: - print(f"Continue lifecycle hook {lch['LifecycleHookName']}") - asg_client.complete_lifecycle_action( - LifecycleHookName=lch["LifecycleHookName"], - AutoScalingGroupName=asg["AutoScalingGroupName"], - LifecycleActionResult="CONTINUE", - InstanceId=instance["InstanceId"], - ) - - -def main( - access_token: str, - delete_offline_runners: bool, -) -> None: - gh_runners = list_runners(access_token) - - dead_runners = get_dead_runners_in_ec2(gh_runners) - print("Runners in GH API to terminate: ", [runner.name for runner in dead_runners]) - if delete_offline_runners and dead_runners: - print("Going to delete offline runners") - for runner in dead_runners: - print("Deleting runner", runner) - delete_runner(access_token, runner) - elif dead_runners: - print("Would delete dead runners: ", dead_runners) - - lost_instances = get_lost_ec2_instances(gh_runners) - print("Instances to terminate: ", lost_instances) - if delete_offline_runners: - if lost_instances: - print("Going to terminate lost instances") - boto3.client("ec2").terminate_instances(InstanceIds=lost_instances) - - continue_lifecycle_hooks(delete_offline_runners) - - -if __name__ == "__main__": - parser = argparse.ArgumentParser(description="Get list of runners and their states") - parser.add_argument( - "-p", "--private-key-path", help="Path to file with private key" - ) - parser.add_argument("-k", "--private-key", help="Private key") - parser.add_argument( - "-a", "--app-id", type=int, help="GitHub application ID", required=True - ) - parser.add_argument( - "--delete-offline", action="store_true", help="Remove offline runners" - ) - - args = parser.parse_args() - - if not args.private_key_path and not args.private_key: - print( - "Either --private-key-path or --private-key must be specified", - file=sys.stderr, - ) - - if args.private_key_path and args.private_key: - print( - "Either --private-key-path or --private-key must be specified", - file=sys.stderr, - ) - - if args.private_key: - private_key = args.private_key - elif args.private_key_path: - with open(args.private_key_path, "r", encoding="utf-8") as key_file: - private_key = key_file.read() - else: - print("Attempt to get key and id from AWS secret manager") - private_key, args.app_id = get_key_and_app_from_aws() - - token = get_access_token_by_key_app(private_key, args.app_id) - - main(token, args.delete_offline) diff --git a/tests/ci/clean_lost_instances_lambda/build_and_deploy_archive.sh b/tests/ci/clean_lost_instances_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/clean_lost_instances_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/clean_lost_instances_lambda/lambda_shared b/tests/ci/clean_lost_instances_lambda/lambda_shared deleted file mode 120000 index ba86e090f6c..00000000000 --- a/tests/ci/clean_lost_instances_lambda/lambda_shared +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package/lambda_shared \ No newline at end of file diff --git a/tests/ci/clean_lost_instances_lambda/requirements.txt b/tests/ci/clean_lost_instances_lambda/requirements.txt deleted file mode 100644 index e2b16067a93..00000000000 --- a/tests/ci/clean_lost_instances_lambda/requirements.txt +++ /dev/null @@ -1,2 +0,0 @@ -../lambda_shared_package -../lambda_shared_package[token] diff --git a/tests/ci/clickhouse_helper.py b/tests/ci/clickhouse_helper.py index 91ea5c6d5d3..287970cce9a 100644 --- a/tests/ci/clickhouse_helper.py +++ b/tests/ci/clickhouse_helper.py @@ -197,6 +197,10 @@ def get_instance_id(): return _query_imds("latest/meta-data/instance-id") +def get_instance_lifecycle(): + return _query_imds("latest/meta-data/instance-life-cycle") + + def prepare_tests_results_for_clickhouse( pr_info: PRInfo, test_results: TestResults, @@ -233,7 +237,7 @@ def prepare_tests_results_for_clickhouse( "head_ref": head_ref, "head_repo": head_repo, "task_url": pr_info.task_url, - "instance_type": get_instance_type(), + "instance_type": ",".join([get_instance_type(), get_instance_lifecycle()]), "instance_id": get_instance_id(), } diff --git a/tests/ci/compress_files.py b/tests/ci/compress_files.py index 70c0aca6a36..f49c872b5a0 100644 --- a/tests/ci/compress_files.py +++ b/tests/ci/compress_files.py @@ -58,14 +58,14 @@ def decompress_fast(archive_path: Path, result_path: Optional[Path] = None) -> N archive_path, result_path, ) - program_part = "--use-compress-program='zstd --threads=0'" + program_part = "--use-compress-program='zstd --threads=0 -d'" elif PIGZ.exists(): logging.info( "pigz found, will compress and decompress faster ('%s' -> '%s')", archive_path, result_path, ) - program_part = "--use-compress-program='pigz'" + program_part = "--use-compress-program='pigz -d'" else: program_part = "-z" logging.info( diff --git a/tests/ci/create_release.py b/tests/ci/create_release.py new file mode 100755 index 00000000000..a0b4083b673 --- /dev/null +++ b/tests/ci/create_release.py @@ -0,0 +1,860 @@ +import argparse +import dataclasses +import json +import os + +from contextlib import contextmanager +from copy import copy +from pathlib import Path +from typing import Iterator, List + +from git_helper import Git, GIT_PREFIX +from ssh import SSHAgent +from env_helper import GITHUB_REPOSITORY, S3_BUILDS_BUCKET +from s3_helper import S3Helper +from ci_utils import Shell, GHActions +from ci_buddy import CIBuddy +from version_helper import ( + FILE_WITH_VERSION_PATH, + GENERATED_CONTRIBUTORS, + get_abs_path, + get_version_from_repo, + update_cmake_version, + update_contributors, + VersionType, +) +from ci_config import CI + +CMAKE_PATH = get_abs_path(FILE_WITH_VERSION_PATH) +CONTRIBUTORS_PATH = get_abs_path(GENERATED_CONTRIBUTORS) +RELEASE_INFO_FILE = "/tmp/release_info.json" + + +class ReleaseProgress: + STARTED = "started" + DOWNLOAD_PACKAGES = "download packages" + PUSH_RELEASE_TAG = "push release tag" + PUSH_NEW_RELEASE_BRANCH = "push new release branch" + BUMP_VERSION = "bump version" + CREATE_GH_RELEASE = "create GH release" + EXPORT_TGZ = "export TGZ packages" + EXPORT_RPM = "export RPM packages" + EXPORT_DEB = "export DEB packages" + TEST_TGZ = "test TGZ packages" + TEST_RPM = "test RPM packages" + TEST_DEB = "test DEB packages" + COMPLETED = "completed" + + +class ReleaseProgressDescription: + OK = "OK" + FAILED = "FAILED" + + +class ReleaseContextManager: + def __init__(self, release_progress): + self.release_progress = release_progress + self.release_info = None + + def __enter__(self): + if self.release_progress == ReleaseProgress.STARTED: + # create initial release info + self.release_info = ReleaseInfo( + release_branch="NA", + commit_sha=args.ref, + release_tag="NA", + version="NA", + codename="NA", + previous_release_tag="NA", + previous_release_sha="NA", + release_progress=ReleaseProgress.STARTED, + ).dump() + else: + # fetch release info from fs and update + self.release_info = ReleaseInfo.from_file() + assert self.release_info + assert ( + self.release_info.progress_description == ReleaseProgressDescription.OK + ), "Must be OK on the start of new context" + self.release_info.release_progress = self.release_progress + self.release_info.dump() + return self.release_info + + def __exit__(self, exc_type, exc_value, traceback): + assert self.release_info + if exc_type is not None: + self.release_info.progress_description = ReleaseProgressDescription.FAILED + else: + self.release_info.progress_description = ReleaseProgressDescription.OK + self.release_info.dump() + + +@dataclasses.dataclass +class ReleaseInfo: + version: str + release_tag: str + release_branch: str + commit_sha: str + # lts or stable + codename: str + previous_release_tag: str + previous_release_sha: str + changelog_pr: str = "" + version_bump_pr: str = "" + release_url: str = "" + debian_command: str = "" + rpm_command: str = "" + tgz_command: str = "" + docker_command: str = "" + release_progress: str = "" + progress_description: str = "" + + def is_patch(self): + return self.release_branch != "master" + + def is_new_release_branch(self): + return self.release_branch == "master" + + @staticmethod + def from_file() -> "ReleaseInfo": + with open(RELEASE_INFO_FILE, "r", encoding="utf-8") as json_file: + res = json.load(json_file) + return ReleaseInfo(**res) + + def dump(self): + print(f"Dump release info into [{RELEASE_INFO_FILE}]") + with open(RELEASE_INFO_FILE, "w", encoding="utf-8") as f: + print(json.dumps(dataclasses.asdict(self), indent=2), file=f) + return self + + def prepare(self, commit_ref: str, release_type: str) -> "ReleaseInfo": + version = None + release_branch = None + release_tag = None + previous_release_tag = None + previous_release_sha = None + codename = "" + assert release_type in ("patch", "new") + if release_type == "new": + # check commit_ref is right and on a right branch + Shell.run( + f"git merge-base --is-ancestor {commit_ref} origin/master", + check=True, + ) + with checkout(commit_ref): + commit_sha = Shell.run(f"git rev-parse {commit_ref}", check=True) + # Git() must be inside "with checkout" contextmanager + git = Git() + version = get_version_from_repo(git=git) + release_branch = "master" + expected_prev_tag = f"v{version.major}.{version.minor}.1.1-new" + version.bump().with_description(VersionType.NEW) + assert ( + git.latest_tag == expected_prev_tag + ), f"BUG: latest tag [{git.latest_tag}], expected [{expected_prev_tag}]" + release_tag = version.describe + previous_release_tag = expected_prev_tag + previous_release_sha = Shell.run_strict( + f"git rev-parse {previous_release_tag}" + ) + assert previous_release_sha + if release_type == "patch": + with checkout(commit_ref): + commit_sha = Shell.run(f"git rev-parse {commit_ref}", check=True) + # Git() must be inside "with checkout" contextmanager + git = Git() + version = get_version_from_repo(git=git) + codename = version.get_stable_release_type() + version.with_description(codename) + release_branch = f"{version.major}.{version.minor}" + release_tag = version.describe + Shell.run(f"{GIT_PREFIX} fetch origin {release_branch} --tags", check=True) + # check commit is right and on a right branch + Shell.run( + f"git merge-base --is-ancestor {commit_ref} origin/{release_branch}", + check=True, + ) + if version.patch == 1: + expected_version = copy(version) + previous_release_tag = f"v{version.major}.{version.minor}.1.1-new" + expected_version.bump() + expected_tag_prefix = ( + f"v{expected_version.major}.{expected_version.minor}." + ) + expected_tag_suffix = "-new" + else: + expected_tag_prefix = ( + f"v{version.major}.{version.minor}.{version.patch-1}." + ) + expected_tag_suffix = f"-{version.get_stable_release_type()}" + previous_release_tag = git.latest_tag + if git.latest_tag.startswith( + expected_tag_prefix + ) and git.latest_tag.endswith(expected_tag_suffix): + pass + else: + assert ( + False + ), f"BUG: Unexpected latest tag [{git.latest_tag}] expected [{expected_tag_prefix}*{expected_tag_suffix}]" + + previous_release_sha = Shell.run_strict( + f"git rev-parse {previous_release_tag}" + ) + assert previous_release_sha + + assert ( + release_branch + and previous_release_tag + and previous_release_sha + and commit_sha + and release_tag + and version + and (codename in ("lts", "stable") or release_type == "new") + ) + + self.release_branch = release_branch + self.commit_sha = commit_sha + self.release_tag = release_tag + self.version = version.string + self.codename = codename + self.previous_release_tag = previous_release_tag + self.previous_release_sha = previous_release_sha + self.release_progress = ReleaseProgress.STARTED + self.progress_description = ReleaseProgressDescription.OK + return self + + def push_release_tag(self, dry_run: bool) -> None: + if dry_run: + # remove locally created tag from prev run + Shell.run( + f"{GIT_PREFIX} tag -l | grep -q {self.release_tag} && git tag -d {self.release_tag} ||:" + ) + # Create release tag + print( + f"Create and push release tag [{self.release_tag}], commit [{self.commit_sha}]" + ) + tag_message = f"Release {self.release_tag}" + Shell.run( + f"{GIT_PREFIX} tag -a -m '{tag_message}' {self.release_tag} {self.commit_sha}", + check=True, + ) + cmd_push_tag = f"{GIT_PREFIX} push origin {self.release_tag}:{self.release_tag}" + Shell.run(cmd_push_tag, dry_run=dry_run, check=True) + + @staticmethod + def _create_gh_label(label: str, color_hex: str, dry_run: bool) -> None: + cmd = f"gh api repos/{GITHUB_REPOSITORY}/labels -f name={label} -f color={color_hex}" + Shell.run(cmd, dry_run=dry_run, check=True) + + def push_new_release_branch(self, dry_run: bool) -> None: + assert ( + self.release_branch == "master" + ), "New release branch can be created only for release type [new]" + git = Git() + version = get_version_from_repo(git=git) + new_release_branch = f"{version.major}.{version.minor}" + stable_release_type = version.get_stable_release_type() + version_after_release = copy(version) + version_after_release.bump() + assert ( + version_after_release.string == self.version + ), f"Unexpected current version in git, must precede [{self.version}] by one step, actual [{version.string}]" + if dry_run: + # remove locally created branch from prev run + Shell.run( + f"{GIT_PREFIX} branch -l | grep -q {new_release_branch} && git branch -d {new_release_branch}" + ) + print( + f"Create and push new release branch [{new_release_branch}], commit [{self.commit_sha}]" + ) + with checkout(self.release_branch): + with checkout_new(new_release_branch): + pr_labels = f"--label {CI.Labels.RELEASE}" + if stable_release_type == VersionType.LTS: + pr_labels += f" --label {CI.Labels.RELEASE_LTS}" + cmd_push_branch = ( + f"{GIT_PREFIX} push --set-upstream origin {new_release_branch}" + ) + Shell.run(cmd_push_branch, dry_run=dry_run, check=True) + + print("Create and push backport tags for new release branch") + ReleaseInfo._create_gh_label( + f"v{new_release_branch}-must-backport", "10dbed", dry_run=dry_run + ) + ReleaseInfo._create_gh_label( + f"v{new_release_branch}-affected", "c2bfff", dry_run=dry_run + ) + Shell.run( + f"""gh pr create --repo {GITHUB_REPOSITORY} --title 'Release pull request for branch {new_release_branch}' + --head {new_release_branch} {pr_labels} + --body 'This PullRequest is a part of ClickHouse release cycle. It is used by CI system only. Do not perform any changes with it.' + """, + dry_run=dry_run, + check=True, + ) + + def update_version_and_contributors_list(self, dry_run: bool) -> None: + # Bump version, update contributors list, create PR + branch_upd_version_contributors = f"bump_version_{self.version}" + with checkout(self.commit_sha): + git = Git() + version = get_version_from_repo(git=git) + if self.release_branch == "master": + version.bump() + version.with_description(VersionType.TESTING) + else: + version.with_description(version.get_stable_release_type()) + assert ( + version.string == self.version + ), f"BUG: version in release info does not match version in git commit, expected [{self.version}], got [{version.string}]" + with checkout(self.release_branch): + with checkout_new(branch_upd_version_contributors): + update_cmake_version(version) + update_contributors(raise_error=True) + cmd_commit_version_upd = f"{GIT_PREFIX} commit '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}' -m 'Update autogenerated version to {self.version} and contributors'" + cmd_push_branch = f"{GIT_PREFIX} push --set-upstream origin {branch_upd_version_contributors}" + body_file = get_abs_path(".github/PULL_REQUEST_TEMPLATE.md") + actor = os.getenv("GITHUB_ACTOR", "") or "me" + cmd_create_pr = f"gh pr create --repo {GITHUB_REPOSITORY} --title 'Update version after release' --head {branch_upd_version_contributors} --base {self.release_branch} --body-file '{body_file} --label 'do not test' --assignee @{actor}" + Shell.run(cmd_commit_version_upd, check=True, dry_run=dry_run) + Shell.run(cmd_push_branch, check=True, dry_run=dry_run) + Shell.run(cmd_create_pr, check=True, dry_run=dry_run) + if dry_run: + Shell.run(f"{GIT_PREFIX} diff '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}'") + Shell.run( + f"{GIT_PREFIX} checkout '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}'" + ) + self.version_bump_pr = "dry-run" + else: + self.version_bump_pr = GHActions.get_pr_url_by_branch( + repo=GITHUB_REPOSITORY, branch=branch_upd_version_contributors + ) + + def update_release_info(self, dry_run: bool) -> "ReleaseInfo": + if self.release_branch != "master": + branch = f"auto/{release_info.release_tag}" + if not dry_run: + url = GHActions.get_pr_url_by_branch( + repo=GITHUB_REPOSITORY, branch=branch + ) + else: + url = "dry-run" + print(f"ChangeLog PR url [{url}]") + self.changelog_pr = url + print(f"Release url [{url}]") + self.release_url = f"https://github.com/{GITHUB_REPOSITORY}/releases/tag/{self.release_tag}" + if self.release_progress == ReleaseProgress.COMPLETED: + self.docker_command = f"docker run --rm clickhouse/clickhouse:{self.version} clickhouse --version" + self.dump() + return self + + def create_gh_release(self, packages_files: List[str], dry_run: bool) -> None: + repo = os.getenv("GITHUB_REPOSITORY") + assert repo + cmds = [ + f"gh release create --repo {repo} --title 'Release {self.release_tag}' {self.release_tag}" + ] + for file in packages_files: + cmds.append(f"gh release upload {self.release_tag} {file}") + if not dry_run: + for cmd in cmds: + Shell.run(cmd, check=True) + self.release_url = f"https://github.com/{GITHUB_REPOSITORY}/releases/tag/{self.release_tag}" + else: + print("Dry-run, would run commands:") + print("\n * ".join(cmds)) + self.release_url = f"dry-run" + self.dump() + + +class RepoTypes: + RPM = "rpm" + DEBIAN = "deb" + TGZ = "tgz" + + +class PackageDownloader: + PACKAGES = ( + "clickhouse-client", + "clickhouse-common-static", + "clickhouse-common-static-dbg", + "clickhouse-keeper", + "clickhouse-keeper-dbg", + "clickhouse-server", + ) + + EXTRA_PACKAGES = ( + "clickhouse-library-bridge", + "clickhouse-odbc-bridge", + ) + PACKAGE_TYPES = (CI.BuildNames.PACKAGE_RELEASE, CI.BuildNames.PACKAGE_AARCH64) + MACOS_PACKAGE_TO_BIN_SUFFIX = { + CI.BuildNames.BINARY_DARWIN: "macos", + CI.BuildNames.BINARY_DARWIN_AARCH64: "macos-aarch64", + } + LOCAL_DIR = "/tmp/packages" + + @classmethod + def _get_arch_suffix(cls, package_arch, repo_type): + if package_arch == CI.BuildNames.PACKAGE_RELEASE: + return ( + "amd64" if repo_type in (RepoTypes.DEBIAN, RepoTypes.TGZ) else "x86_64" + ) + elif package_arch == CI.BuildNames.PACKAGE_AARCH64: + return ( + "arm64" if repo_type in (RepoTypes.DEBIAN, RepoTypes.TGZ) else "aarch64" + ) + else: + assert False, "BUG" + + def __init__(self, release, commit_sha, version): + assert version.startswith(release), "Invalid release branch or version" + major, minor = map(int, release.split(".")) + self.package_names = list(self.PACKAGES) + if major > 24 or (major == 24 and minor > 3): + self.package_names += list(self.EXTRA_PACKAGES) + self.release = release + self.commit_sha = commit_sha + self.version = version + self.s3 = S3Helper() + self.deb_package_files = [] + self.rpm_package_files = [] + self.tgz_package_files = [] + # just binaries for macos + self.macos_package_files = ["clickhouse-macos", "clickhouse-macos-aarch64"] + self.file_to_type = {} + + Shell.run(f"mkdir -p {self.LOCAL_DIR}") + + for package_type in self.PACKAGE_TYPES: + for package in self.package_names: + deb_package_file_name = f"{package}_{self.version}_{self._get_arch_suffix(package_type, RepoTypes.DEBIAN)}.deb" + self.deb_package_files.append(deb_package_file_name) + self.file_to_type[deb_package_file_name] = package_type + + rpm_package_file_name = f"{package}-{self.version}.{self._get_arch_suffix(package_type, RepoTypes.RPM)}.rpm" + self.rpm_package_files.append(rpm_package_file_name) + self.file_to_type[rpm_package_file_name] = package_type + + tgz_package_file_name = f"{package}-{self.version}-{self._get_arch_suffix(package_type, RepoTypes.TGZ)}.tgz" + self.tgz_package_files.append(tgz_package_file_name) + self.file_to_type[tgz_package_file_name] = package_type + tgz_package_file_name += ".sha512" + self.tgz_package_files.append(tgz_package_file_name) + self.file_to_type[tgz_package_file_name] = package_type + + def get_deb_packages_files(self): + return self.deb_package_files + + def get_rpm_packages_files(self): + return self.rpm_package_files + + def get_tgz_packages_files(self): + return self.tgz_package_files + + def get_macos_packages_files(self): + return self.macos_package_files + + def get_packages_names(self): + return self.package_names + + def get_all_packages_files(self): + assert self.local_tgz_packages_ready() + assert self.local_deb_packages_ready() + assert self.local_rpm_packages_ready() + assert self.local_macos_packages_ready() + res = [] + for package_file in ( + self.deb_package_files + + self.rpm_package_files + + self.tgz_package_files + + self.macos_package_files + ): + res.append(self.LOCAL_DIR + "/" + package_file) + return res + + def run(self): + Shell.run(f"rm -rf {self.LOCAL_DIR}/*") + for package_file in ( + self.deb_package_files + self.rpm_package_files + self.tgz_package_files + ): + print(f"Downloading: [{package_file}]") + s3_path = "/".join( + [ + self.release, + self.commit_sha, + self.file_to_type[package_file], + package_file, + ] + ) + self.s3.download_file( + bucket=S3_BUILDS_BUCKET, + s3_path=s3_path, + local_file_path="/".join([self.LOCAL_DIR, package_file]), + ) + + for macos_package, bin_suffix in self.MACOS_PACKAGE_TO_BIN_SUFFIX.items(): + binary_name = "clickhouse" + destination_binary_name = f"{binary_name}-{bin_suffix}" + assert destination_binary_name in self.macos_package_files + print( + f"Downloading: [{macos_package}] binary to [{destination_binary_name}]" + ) + s3_path = "/".join( + [ + self.release, + self.commit_sha, + macos_package, + binary_name, + ] + ) + self.s3.download_file( + bucket=S3_BUILDS_BUCKET, + s3_path=s3_path, + local_file_path="/".join([self.LOCAL_DIR, destination_binary_name]), + ) + + def local_deb_packages_ready(self) -> bool: + assert self.deb_package_files + for package_file in self.deb_package_files: + print(f"Check package is downloaded [{package_file}]") + if not Path(self.LOCAL_DIR + "/" + package_file).is_file(): + return False + return True + + def local_rpm_packages_ready(self) -> bool: + assert self.rpm_package_files + for package_file in self.rpm_package_files: + print(f"Check package is downloaded [{package_file}]") + if not Path(self.LOCAL_DIR + "/" + package_file).is_file(): + return False + return True + + def local_tgz_packages_ready(self) -> bool: + assert self.tgz_package_files + for package_file in self.tgz_package_files: + print(f"Check package is downloaded [{package_file}]") + if not Path(self.LOCAL_DIR + "/" + package_file).is_file(): + return False + return True + + def local_macos_packages_ready(self) -> bool: + assert self.macos_package_files + for package_file in self.macos_package_files: + print(f"Check package is downloaded [{package_file}]") + if not Path(self.LOCAL_DIR + "/" + package_file).is_file(): + return False + return True + + +@contextmanager +def checkout(ref: str) -> Iterator[None]: + orig_ref = Shell.run(f"{GIT_PREFIX} symbolic-ref --short HEAD", check=True) + rollback_cmd = f"{GIT_PREFIX} checkout {orig_ref}" + assert orig_ref + if ref not in (orig_ref,): + Shell.run(f"{GIT_PREFIX} checkout {ref}") + try: + yield + except (Exception, KeyboardInterrupt) as e: + print(f"ERROR: Exception [{e}]") + Shell.run(rollback_cmd) + raise + Shell.run(rollback_cmd) + + +@contextmanager +def checkout_new(ref: str) -> Iterator[None]: + orig_ref = Shell.run(f"{GIT_PREFIX} symbolic-ref --short HEAD", check=True) + rollback_cmd = f"{GIT_PREFIX} checkout {orig_ref}" + assert orig_ref + Shell.run(f"{GIT_PREFIX} checkout -b {ref}", check=True) + try: + yield + except (Exception, KeyboardInterrupt) as e: + print(f"ERROR: Exception [{e}]") + Shell.run(rollback_cmd) + raise + Shell.run(rollback_cmd) + + +def parse_args() -> argparse.Namespace: + parser = argparse.ArgumentParser( + formatter_class=argparse.ArgumentDefaultsHelpFormatter, + description="Creates release", + ) + parser.add_argument( + "--prepare-release-info", + action="store_true", + help="Initial step to prepare info like release branch, release tag, etc.", + ) + parser.add_argument( + "--push-release-tag", + action="store_true", + help="Creates and pushes git tag", + ) + parser.add_argument( + "--push-new-release-branch", + action="store_true", + help="Creates and pushes new release branch and corresponding service gh tags for backports", + ) + parser.add_argument( + "--create-bump-version-pr", + action="store_true", + help="Updates version, contributors' list and creates PR", + ) + parser.add_argument( + "--download-packages", + action="store_true", + help="Downloads all required packages from s3", + ) + parser.add_argument( + "--create-gh-release", + action="store_true", + help="Create GH Release object and attach all packages", + ) + parser.add_argument( + "--post-status", + action="store_true", + help="Post release status into Slack", + ) + parser.add_argument( + "--ref", + type=str, + help="the commit hash or branch", + ) + parser.add_argument( + "--release-type", + choices=("new", "patch"), + # dest="release_type", + help="a release type to bump the major.minor.patch version part, " + "new branch is created only for the value 'new'", + ) + parser.add_argument( + "--dry-run", + action="store_true", + help="do not make any actual changes in the repo, just show what will be done", + ) + parser.add_argument( + "--set-progress-started", + action="store_true", + help="Set new progress step, --progress must be set", + ) + parser.add_argument( + "--progress", + type=str, + help="Progress step name, see @ReleaseProgress", + ) + parser.add_argument( + "--set-progress-completed", + action="store_true", + help="Set current progress step to OK (completed)", + ) + return parser.parse_args() + + +if __name__ == "__main__": + args = parse_args() + + # prepare ssh for git if needed + _ssh_agent = None + _key_pub = None + if os.getenv("ROBOT_CLICKHOUSE_SSH_KEY", ""): + _key = os.getenv("ROBOT_CLICKHOUSE_SSH_KEY") + _ssh_agent = SSHAgent() + _key_pub = _ssh_agent.add(_key) + _ssh_agent.print_keys() + + if args.prepare_release_info: + with ReleaseContextManager( + release_progress=ReleaseProgress.STARTED + ) as release_info: + assert ( + args.ref and args.release_type + ), "--ref and --release-type must be provided with --prepare-release-info" + release_info.prepare(commit_ref=args.ref, release_type=args.release_type) + + if args.download_packages: + with ReleaseContextManager( + release_progress=ReleaseProgress.DOWNLOAD_PACKAGES + ) as release_info: + p = PackageDownloader( + release=release_info.release_branch, + commit_sha=release_info.commit_sha, + version=release_info.version, + ) + p.run() + + if args.push_release_tag: + with ReleaseContextManager( + release_progress=ReleaseProgress.PUSH_RELEASE_TAG + ) as release_info: + release_info.push_release_tag(dry_run=args.dry_run) + + if args.push_new_release_branch: + with ReleaseContextManager( + release_progress=ReleaseProgress.PUSH_NEW_RELEASE_BRANCH + ) as release_info: + release_info.push_new_release_branch(dry_run=args.dry_run) + + if args.create_bump_version_pr: + with ReleaseContextManager( + release_progress=ReleaseProgress.BUMP_VERSION + ) as release_info: + release_info.update_version_and_contributors_list(dry_run=args.dry_run) + + if args.create_gh_release: + with ReleaseContextManager( + release_progress=ReleaseProgress.CREATE_GH_RELEASE + ) as release_info: + p = PackageDownloader( + release=release_info.release_branch, + commit_sha=release_info.commit_sha, + version=release_info.version, + ) + release_info.create_gh_release( + packages_files=p.get_all_packages_files(), dry_run=args.dry_run + ) + + if args.post_status: + release_info = ReleaseInfo.from_file() + release_info.update_release_info(dry_run=args.dry_run) + if release_info.is_new_release_branch(): + title = "New release branch" + else: + title = "New release" + if ( + release_info.progress_description == ReleaseProgressDescription.OK + and release_info.release_progress == ReleaseProgress.COMPLETED + ): + title = "Completed: " + title + CIBuddy(dry_run=args.dry_run).post_done( + title, dataclasses.asdict(release_info) + ) + else: + title = "Failed: " + title + CIBuddy(dry_run=args.dry_run).post_critical( + title, dataclasses.asdict(release_info) + ) + + if args.set_progress_started: + ri = ReleaseInfo.from_file() + ri.release_progress = args.progress + ri.progress_description = ReleaseProgressDescription.FAILED + ri.dump() + assert args.progress, "Progress step name must be provided" + + if args.set_progress_completed: + ri = ReleaseInfo.from_file() + assert ( + ri.progress_description == ReleaseProgressDescription.FAILED + ), "Must be FAILED before set to OK" + ri.progress_description = ReleaseProgressDescription.OK + ri.dump() + + # tear down ssh + if _ssh_agent and _key_pub: + _ssh_agent.remove(_key_pub) + + +""" +Prepare release machine: + +### INSTALL PACKAGES +sudo apt update +sudo apt install --yes --no-install-recommends python3-dev python3-pip gh unzip +sudo apt install --yes python3-boto3 +sudo apt install --yes python3-github +sudo apt install --yes python3-unidiff +sudo apt install --yes python3-tqdm # cloud changelog +sudo apt install --yes python3-thefuzz # cloud changelog +sudo apt install --yes s3fs + +### INSTALL AWS CLI +cd /tmp +curl "https://awscli.amazonaws.com/awscli-exe-linux-$(uname -m).zip" -o "awscliv2.zip" +unzip awscliv2.zip +sudo ./aws/install +rm -rf aws* +cd - + +### INSTALL GH ACTIONS RUNNER: +# Create a folder +RUNNER_VERSION=2.317.0 +cd ~ +mkdir actions-runner && cd actions-runner +# Download the latest runner package +runner_arch() { + case $(uname -m) in + x86_64 ) + echo x64;; + aarch64 ) + echo arm64;; + esac +} +curl -O -L https://github.com/actions/runner/releases/download/v$RUNNER_VERSION/actions-runner-linux-$(runner_arch)-$RUNNER_VERSION.tar.gz +# Extract the installer +tar xzf ./actions-runner-linux-$(runner_arch)-$RUNNER_VERSION.tar.gz +rm ./actions-runner-linux-$(runner_arch)-$RUNNER_VERSION.tar.gz + +### Install reprepro: +cd ~ +sudo apt install dpkg-dev libgpgme-dev libdb-dev libbz2-dev liblzma-dev libarchive-dev shunit2 db-util debhelper +git clone https://salsa.debian.org/debian/reprepro.git +cd reprepro +dpkg-buildpackage -b --no-sign && sudo dpkg -i ../reprepro_$(dpkg-parsechangelog --show-field Version)_$(dpkg-architecture -q DEB_HOST_ARCH).deb + +### Install createrepo-c: +sudo apt install createrepo-c +createrepo_c --version +#Version: 0.17.3 (Features: DeltaRPM LegacyWeakdeps ) + +### Import gpg sign key +gpg --import key.pgp +gpg --list-secret-keys + +### Install docker +sudo su; cd ~ + +deb_arch() { + case $(uname -m) in + x86_64 ) + echo amd64;; + aarch64 ) + echo arm64;; + esac +} +curl -fsSL https://download.docker.com/linux/ubuntu/gpg | gpg --dearmor -o /usr/share/keyrings/docker-archive-keyring.gpg + +echo "deb [arch=$(deb_arch) signed-by=/usr/share/keyrings/docker-archive-keyring.gpg] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null + +sudo apt-get update +sudo apt-get install --yes --no-install-recommends docker-ce docker-buildx-plugin docker-ce-cli containerd.io + +sudo usermod -aG docker ubuntu + +# enable ipv6 in containers (fixed-cidr-v6 is some random network mask) +cat < /etc/docker/daemon.json +{ + "ipv6": true, + "fixed-cidr-v6": "2001:db8:1::/64", + "log-driver": "json-file", + "log-opts": { + "max-file": "5", + "max-size": "1000m" + }, + "insecure-registries" : ["dockerhub-proxy.dockerhub-proxy-zone:5000"], + "registry-mirrors" : ["http://dockerhub-proxy.dockerhub-proxy-zone:5000"] +} +EOT + +# if docker build does not work: + sudo systemctl restart docker + docker buildx rm mybuilder + docker buildx create --name mybuilder --driver docker-container --use + docker buildx inspect mybuilder --bootstrap + +### Install tailscale + +### Configure GH runner +""" diff --git a/tests/ci/docker_server.py b/tests/ci/docker_server.py index 151cc5a4c02..413c35cbebe 100644 --- a/tests/ci/docker_server.py +++ b/tests/ci/docker_server.py @@ -11,7 +11,6 @@ from os import path as p from pathlib import Path from typing import Dict, List -from build_check import get_release_or_pr from build_download_helper import read_build_urls from docker_images_helper import DockerImageData, docker_login from env_helper import ( @@ -63,6 +62,12 @@ def parse_args() -> argparse.Namespace: help="a version to build, automaticaly got from version_helper, accepts either " "tag ('refs/tags/' is removed automatically) or a normal 22.2.2.2 format", ) + parser.add_argument( + "--sha", + type=str, + default="", + help="sha of the commit to use packages from", + ) parser.add_argument( "--release-type", type=str, @@ -122,7 +127,7 @@ def parse_args() -> argparse.Namespace: def retry_popen(cmd: str, log_file: Path) -> int: - max_retries = 5 + max_retries = 2 for retry in range(max_retries): # From time to time docker build may failed. Curl issues, or even push # It will sleep progressively 5, 15, 30 and 50 seconds between retries @@ -370,16 +375,23 @@ def main(): tags = gen_tags(args.version, args.release_type) repo_urls = {} direct_urls: Dict[str, List[str]] = {} - release_or_pr, _ = get_release_or_pr(pr_info, args.version) for arch, build_name in zip(ARCH, ("package_release", "package_aarch64")): - if not args.bucket_prefix: + if args.bucket_prefix: + assert not args.allow_build_reuse + repo_urls[arch] = f"{args.bucket_prefix}/{build_name}" + elif args.sha: + # CreateRelease workflow only. TODO + version = args.version repo_urls[arch] = ( f"{S3_DOWNLOAD}/{S3_BUILDS_BUCKET}/" - f"{release_or_pr}/{pr_info.sha}/{build_name}" + f"{version.major}.{version.minor}/{args.sha}/{build_name}" ) else: - repo_urls[arch] = f"{args.bucket_prefix}/{build_name}" + # In all other cases urls must be fetched from build reports. TODO: script needs refactoring + repo_urls[arch] = "" + assert args.allow_build_reuse + if args.allow_build_reuse: # read s3 urls from pre-downloaded build reports if "clickhouse-server" in image_repo: @@ -417,7 +429,6 @@ def main(): ) if test_results[-1].status != "OK": status = FAILURE - pr_info = pr_info or PRInfo() description = f"Processed tags: {', '.join(tags)}" JobReport( diff --git a/tests/ci/download_release_packages.py b/tests/ci/download_release_packages.py index 8f3a2190ae8..e8260cf68f1 100755 --- a/tests/ci/download_release_packages.py +++ b/tests/ci/download_release_packages.py @@ -13,26 +13,28 @@ from get_previous_release_tag import ( PACKAGES_DIR = Path("previous_release_package_folder") -def download_packages(release: ReleaseInfo, dest_path: Path = PACKAGES_DIR) -> None: +def download_packages( + release: ReleaseInfo, dest_path: Path = PACKAGES_DIR, debug: bool = False +) -> None: dest_path.mkdir(parents=True, exist_ok=True) logging.info("Will download %s", release) for pkg, url in release.assets.items(): - if not pkg.endswith("_amd64.deb") or "-dbg_" in pkg: + if not pkg.endswith("_amd64.deb") or (not debug and "-dbg_" in pkg): continue pkg_name = dest_path / pkg download_build_with_progress(url, pkg_name) -def download_last_release(dest_path: Path) -> None: +def download_last_release(dest_path: Path, debug: bool = False) -> None: current_release = get_previous_release(None) if current_release is None: raise DownloadException("The current release is not found") - download_packages(current_release, dest_path=dest_path) + download_packages(current_release, dest_path=dest_path, debug=debug) if __name__ == "__main__": logging.basicConfig(level=logging.INFO) release = get_release_by_tag(input()) - download_packages(release) + download_packages(release, debug=True) diff --git a/tests/ci/functional_test_check.py b/tests/ci/functional_test_check.py index ee459ce35a0..66db082677f 100644 --- a/tests/ci/functional_test_check.py +++ b/tests/ci/functional_test_check.py @@ -17,13 +17,35 @@ from download_release_packages import download_last_release from env_helper import REPO_COPY, REPORT_PATH, TEMP_PATH from get_robot_token import get_parameter_from_ssm from pr_info import PRInfo -from report import ERROR, SUCCESS, JobReport, StatusType, TestResults, read_test_results +from report import ( + ERROR, + SUCCESS, + JobReport, + StatusType, + TestResults, + read_test_results, + FAILURE, +) from stopwatch import Stopwatch from tee_popen import TeePopen +from ci_config import CI +from ci_utils import Utils NO_CHANGES_MSG = "Nothing to run" +class SensitiveFormatter(logging.Formatter): + @staticmethod + def _filter(s): + return re.sub( + r"(.*)(AZURE_CONNECTION_STRING.*\')(.*)", r"\1AZURE_CONNECTION_STRING\3", s + ) + + def format(self, record): + original = logging.Formatter.format(self, record) + return self._filter(original) + + def get_additional_envs( check_name: str, run_by_hash_num: int, run_by_hash_total: int ) -> List[str]: @@ -90,8 +112,8 @@ def get_run_command( ] if flaky_check: - envs.append("-e NUM_TRIES=100") - envs.append("-e MAX_RUN_TIME=1800") + envs.append("-e NUM_TRIES=50") + envs.append("-e MAX_RUN_TIME=2800") envs += [f"-e {e}" for e in additional_envs] @@ -104,8 +126,11 @@ def get_run_command( return ( f"docker run --volume={builds_path}:/package_folder " + # For dmesg and sysctl + "--privileged " f"{ci_logs_args}" f"--volume={repo_path}/tests:/usr/share/clickhouse-test " + f"--volume={repo_path}/utils/grpc-client:/usr/share/clickhouse-utils/grpc-client " f"{volume_with_broken_test}" f"--volume={result_path}:/test_output " f"--volume={server_log_path}:/var/log/clickhouse-server " @@ -210,6 +235,9 @@ def parse_args(): def main(): logging.basicConfig(level=logging.INFO) + for handler in logging.root.handlers: + # pylint: disable=protected-access + handler.setFormatter(SensitiveFormatter(handler.formatter._fmt)) # type: ignore stopwatch = Stopwatch() @@ -253,7 +281,7 @@ def main(): packages_path.mkdir(parents=True, exist_ok=True) if validate_bugfix_check: - download_last_release(packages_path) + download_last_release(packages_path, debug=True) else: download_all_deb_packages(check_name, reports_path, packages_path) @@ -333,7 +361,23 @@ def main(): additional_files=additional_logs, ).dump(to_file=args.report_to_file if args.report_to_file else None) + should_block_ci = False if state != SUCCESS: + should_block_ci = True + + if state == FAILURE and CI.is_required(check_name): + failed_cnt = Utils.get_failed_tests_number(description) + print( + f"Job status is [{state}] with [{failed_cnt}] failed test cases. status description [{description}]" + ) + if ( + failed_cnt + and failed_cnt <= CI.MAX_TOTAL_FAILURES_PER_JOB_BEFORE_BLOCKING_CI + ): + print(f"Won't block the CI workflow") + should_block_ci = False + + if should_block_ci: sys.exit(1) diff --git a/tests/ci/integration_test_check.py b/tests/ci/integration_test_check.py index 7c74f52b610..6245f0490fc 100644 --- a/tests/ci/integration_test_check.py +++ b/tests/ci/integration_test_check.py @@ -23,10 +23,13 @@ from report import ( TestResult, TestResults, read_test_results, + FAILURE, ) from stopwatch import Stopwatch import integration_tests_runner as runner +from ci_config import CI +from ci_utils import Utils def get_json_params_dict( @@ -185,7 +188,7 @@ def main(): build_path.mkdir(parents=True, exist_ok=True) if validate_bugfix_check: - download_last_release(build_path) + download_last_release(build_path, debug=True) else: download_all_deb_packages(check_name, reports_path, build_path) @@ -233,7 +236,23 @@ def main(): additional_files=additional_logs, ).dump(to_file=args.report_to_file if args.report_to_file else None) + should_block_ci = False if state != SUCCESS: + should_block_ci = True + + if state == FAILURE and CI.is_required(check_name): + failed_cnt = Utils.get_failed_tests_number(description) + print( + f"Job status is [{state}] with [{failed_cnt}] failed test cases. status description [{description}]" + ) + if ( + failed_cnt + and failed_cnt <= CI.MAX_TOTAL_FAILURES_PER_JOB_BEFORE_BLOCKING_CI + ): + print(f"Won't block the CI workflow") + should_block_ci = False + + if should_block_ci: sys.exit(1) diff --git a/tests/ci/integration_tests_runner.py b/tests/ci/integration_tests_runner.py index 87f721cfde7..2b348be8b51 100755 --- a/tests/ci/integration_tests_runner.py +++ b/tests/ci/integration_tests_runner.py @@ -29,7 +29,8 @@ CLICKHOUSE_BINARY_PATH = "usr/bin/clickhouse" CLICKHOUSE_ODBC_BRIDGE_BINARY_PATH = "usr/bin/clickhouse-odbc-bridge" CLICKHOUSE_LIBRARY_BRIDGE_BINARY_PATH = "usr/bin/clickhouse-library-bridge" -FLAKY_TRIES_COUNT = 10 +FLAKY_TRIES_COUNT = 10 # run whole pytest several times +FLAKY_REPEAT_COUNT = 5 # runs test case in single module several times MAX_TIME_SECONDS = 3600 MAX_TIME_IN_SANDBOX = 20 * 60 # 20 minutes @@ -434,7 +435,14 @@ class ClickhouseIntegrationTestsRunner: "Getting all tests to the file %s with cmd: \n%s", out_file_full, cmd ) with open(out_file_full, "wb") as ofd: - subprocess.check_call(cmd, shell=True, stdout=ofd, stderr=ofd) + try: + subprocess.check_call(cmd, shell=True, stdout=ofd, stderr=ofd) + except subprocess.CalledProcessError as ex: + print("ERROR: Setting test plan failed. Output:") + with open(out_file_full, "r", encoding="utf-8") as file: + for line in file: + print(" " + line, end="") + raise ex all_tests = set() with open(out_file_full, "r", encoding="utf-8") as all_tests_fd: @@ -561,6 +569,7 @@ class ClickhouseIntegrationTestsRunner: tests_in_group, num_tries, num_workers, + repeat_count, ): try: return self.run_test_group( @@ -569,6 +578,7 @@ class ClickhouseIntegrationTestsRunner: tests_in_group, num_tries, num_workers, + repeat_count, ) except Exception as e: logging.info("Failed to run %s:\n%s", test_group, e) @@ -591,6 +601,7 @@ class ClickhouseIntegrationTestsRunner: tests_in_group, num_tries, num_workers, + repeat_count, ): counters = { "ERROR": [], @@ -632,6 +643,7 @@ class ClickhouseIntegrationTestsRunner: test_cmd = " ".join([shlex.quote(test) for test in sorted(test_names)]) parallel_cmd = f" --parallel {num_workers} " if num_workers > 0 else "" + repeat_cmd = f" --count {repeat_count} " if repeat_count > 0 else "" # -r -- show extra test summary: # -f -- (f)ailed # -E -- (E)rror @@ -640,7 +652,7 @@ class ClickhouseIntegrationTestsRunner: cmd = ( f"cd {repo_path}/tests/integration && " f"timeout --signal=KILL 1h ./runner {self._get_runner_opts()} " - f"{image_cmd} -t {test_cmd} {parallel_cmd} -- -rfEps --run-id={i} " + f"{image_cmd} -t {test_cmd} {parallel_cmd} {repeat_cmd} -- -rfEps --run-id={i} " f"--color=no --durations=0 {_get_deselect_option(self.should_skip_tests())} " f"| tee {info_path}" ) @@ -777,7 +789,12 @@ class ClickhouseIntegrationTestsRunner: final_retry += 1 logging.info("Running tests for the %s time", i) counters, tests_times, log_paths = self.try_run_test_group( - repo_path, "bugfix" if should_fail else "flaky", tests_to_run, 1, 1 + repo_path, + "bugfix" if should_fail else "flaky", + tests_to_run, + 1, + 1, + FLAKY_REPEAT_COUNT, ) logs += log_paths if counters["FAILED"]: @@ -912,7 +929,7 @@ class ClickhouseIntegrationTestsRunner: for group, tests in items_to_run: logging.info("Running test group %s containing %s tests", group, len(tests)) group_counters, group_test_times, log_paths = self.try_run_test_group( - repo_path, group, tests, MAX_RETRY, NUM_WORKERS + repo_path, group, tests, MAX_RETRY, NUM_WORKERS, 0 ) total_tests = 0 for counter, value in group_counters.items(): diff --git a/tests/ci/jepsen_check.py b/tests/ci/jepsen_check.py index f91a3f080c0..772467d4245 100644 --- a/tests/ci/jepsen_check.py +++ b/tests/ci/jepsen_check.py @@ -9,16 +9,13 @@ from pathlib import Path from typing import Any, List import boto3 # type: ignore -import requests from build_download_helper import ( - download_build_with_progress, read_build_urls, ) from compress_files import compress_fast -from env_helper import REPO_COPY, REPORT_PATH, S3_BUILDS_BUCKET, S3_URL, TEMP_PATH +from env_helper import REPO_COPY, REPORT_PATH, TEMP_PATH from get_robot_token import get_parameter_from_ssm -from git_helper import git_runner from pr_info import PRInfo from report import FAILURE, SUCCESS, JobReport, TestResult, TestResults from ssh import SSHKey @@ -32,11 +29,10 @@ KEEPER_DESIRED_INSTANCE_COUNT = 3 SERVER_DESIRED_INSTANCE_COUNT = 4 KEEPER_IMAGE_NAME = "clickhouse/keeper-jepsen-test" -KEEPER_CHECK_NAME = "ClickHouse Keeper Jepsen" +KEEPER_CHECK_NAME = CI.JobNames.JEPSEN_KEEPER SERVER_IMAGE_NAME = "clickhouse/server-jepsen-test" -SERVER_CHECK_NAME = "ClickHouse Server Jepsen" - +SERVER_CHECK_NAME = CI.JobNames.JEPSEN_SERVER SUCCESSFUL_TESTS_ANCHOR = "# Successful tests" INTERMINATE_TESTS_ANCHOR = "# Indeterminate tests" @@ -201,36 +197,14 @@ def main(): # always use latest docker_image = KEEPER_IMAGE_NAME if args.program == "keeper" else SERVER_IMAGE_NAME - if pr_info.is_scheduled or pr_info.is_dispatched: - # get latest clickhouse by the static link for latest master buit - get its version and provide permanent url for this version to the jepsen - build_url = f"{S3_URL}/{S3_BUILDS_BUCKET}/master/amd64/clickhouse" - download_build_with_progress(build_url, Path(TEMP_PATH) / "clickhouse") - git_runner.run(f"chmod +x {TEMP_PATH}/clickhouse") - sha = git_runner.run( - f"{TEMP_PATH}/clickhouse local -q \"select value from system.build_options where name='GIT_HASH'\"" - ) - version_full = git_runner.run( - f'{TEMP_PATH}/clickhouse local -q "select version()"' - ) - version = ".".join(version_full.split(".")[0:2]) - assert len(sha) == 40, f"failed to fetch sha from the binary. result: {sha}" - assert ( - version - ), f"failed to fetch version from the binary. result: {version_full}" - build_url = ( - f"{S3_URL}/{S3_BUILDS_BUCKET}/{version}/{sha}/binary_release/clickhouse" - ) - print(f"Clickhouse version: [{version_full}], sha: [{sha}], url: [{build_url}]") - head = requests.head(build_url, timeout=60) - assert head.status_code == 200, f"Clickhouse binary not found: {build_url}" - else: - build_name = CI.get_required_build_name(check_name) - urls = read_build_urls(build_name, REPORT_PATH) - build_url = None - for url in urls: - if url.endswith("clickhouse"): - build_url = url - assert build_url, "No build url found in the report" + # binary_release assumed to be always ready on the master as it's part of the merge queue workflow + build_name = CI.get_required_build_name(check_name) + urls = read_build_urls(build_name, REPORT_PATH) + build_url = None + for url in urls: + if url.endswith("clickhouse"): + build_url = url + assert build_url, "No build url found in the report" extra_args = "" if args.program == "server": diff --git a/tests/ci/lambda_shared_package/.gitignore b/tests/ci/lambda_shared_package/.gitignore deleted file mode 100644 index 59d52651e06..00000000000 --- a/tests/ci/lambda_shared_package/.gitignore +++ /dev/null @@ -1,2 +0,0 @@ -build -*.egg-info diff --git a/tests/ci/lambda_shared_package/lambda_shared/__init__.py b/tests/ci/lambda_shared_package/lambda_shared/__init__.py deleted file mode 100644 index 8b53f9dcb23..00000000000 --- a/tests/ci/lambda_shared_package/lambda_shared/__init__.py +++ /dev/null @@ -1,237 +0,0 @@ -"""The shared code and types for all our CI lambdas -It exists as __init__.py and lambda_shared/__init__.py to work both in local and venv""" - -import json -import logging -import time -from collections import namedtuple -from typing import Any, Dict, Iterable, List, Optional - -import boto3 # type: ignore -import requests - -RUNNER_TYPE_LABELS = [ - "builder", - "func-tester", - "func-tester-aarch64", - "fuzzer-unit-tester", - "limited-tester", - "stress-tester", - "style-checker", - "style-checker-aarch64", - # private runners - "private-builder", - "private-clickpipes", - "private-func-tester", - "private-fuzzer-unit-tester", - "private-stress-tester", - "private-style-checker", -] - - -### VENDORING -def get_parameter_from_ssm( - name: str, decrypt: bool = True, client: Optional[Any] = None -) -> str: - if not client: - client = boto3.client("ssm", region_name="us-east-1") - return client.get_parameter(Name=name, WithDecryption=decrypt)[ # type: ignore - "Parameter" - ]["Value"] - - -class CHException(Exception): - pass - - -class InsertException(CHException): - pass - - -class ClickHouseHelper: - def __init__( - self, - url: str, - user: Optional[str] = None, - password: Optional[str] = None, - ): - self.url = url - self.auth = {} - if user: - self.auth["X-ClickHouse-User"] = user - if password: - self.auth["X-ClickHouse-Key"] = password - - @staticmethod - def _insert_json_str_info_impl( - url: str, auth: Dict[str, str], db: str, table: str, json_str: str - ) -> None: - params = { - "database": db, - "query": f"INSERT INTO {table} FORMAT JSONEachRow", - "date_time_input_format": "best_effort", - "send_logs_level": "warning", - } - - for i in range(5): - try: - response = requests.post( - url, params=params, data=json_str, headers=auth - ) - except Exception as e: - error = f"Received exception while sending data to {url} on {i} attempt: {e}" - logging.warning(error) - continue - - logging.info("Response content '%s'", response.content) - - if response.ok: - break - - error = ( - "Cannot insert data into clickhouse at try " - + str(i) - + ": HTTP code " - + str(response.status_code) - + ": '" - + str(response.text) - + "'" - ) - - if response.status_code >= 500: - # A retriable error - time.sleep(1) - continue - - logging.info( - "Request headers '%s', body '%s'", - response.request.headers, - response.request.body, - ) - - raise InsertException(error) - else: - raise InsertException(error) - - def _insert_json_str_info(self, db: str, table: str, json_str: str) -> None: - self._insert_json_str_info_impl(self.url, self.auth, db, table, json_str) - - def insert_event_into( - self, db: str, table: str, event: object, safe: bool = True - ) -> None: - event_str = json.dumps(event) - try: - self._insert_json_str_info(db, table, event_str) - except InsertException as e: - logging.error( - "Exception happened during inserting data into clickhouse: %s", e - ) - if not safe: - raise - - def insert_events_into( - self, db: str, table: str, events: Iterable[object], safe: bool = True - ) -> None: - jsons = [] - for event in events: - jsons.append(json.dumps(event)) - - try: - self._insert_json_str_info(db, table, ",".join(jsons)) - except InsertException as e: - logging.error( - "Exception happened during inserting data into clickhouse: %s", e - ) - if not safe: - raise - - def _select_and_get_json_each_row(self, db: str, query: str) -> str: - params = { - "database": db, - "query": query, - "default_format": "JSONEachRow", - } - for i in range(5): - response = None - try: - response = requests.get(self.url, params=params, headers=self.auth) - response.raise_for_status() - return response.text # type: ignore - except Exception as ex: - logging.warning("Cannot fetch data with exception %s", str(ex)) - if response: - logging.warning("Reponse text %s", response.text) - time.sleep(0.1 * i) - - raise CHException("Cannot fetch data from clickhouse") - - def select_json_each_row(self, db: str, query: str) -> List[dict]: - text = self._select_and_get_json_each_row(db, query) - result = [] - for line in text.split("\n"): - if line: - result.append(json.loads(line)) - return result - - -### Runners - -RunnerDescription = namedtuple( - "RunnerDescription", ["id", "name", "tags", "offline", "busy"] -) -RunnerDescriptions = List[RunnerDescription] - - -def list_runners(access_token: str) -> RunnerDescriptions: - headers = { - "Authorization": f"token {access_token}", - "Accept": "application/vnd.github.v3+json", - } - per_page = 100 - response = requests.get( - f"https://api.github.com/orgs/ClickHouse/actions/runners?per_page={per_page}", - headers=headers, - ) - response.raise_for_status() - data = response.json() - total_runners = data["total_count"] - print("Expected total runners", total_runners) - runners = data["runners"] - - # round to 0 for 0, 1 for 1..100, but to 2 for 101..200 - total_pages = (total_runners - 1) // per_page + 1 - - print("Total pages", total_pages) - for i in range(2, total_pages + 1): - response = requests.get( - "https://api.github.com/orgs/ClickHouse/actions/runners" - f"?page={i}&per_page={per_page}", - headers=headers, - ) - response.raise_for_status() - data = response.json() - runners += data["runners"] - - print("Total runners", len(runners)) - result = [] - for runner in runners: - tags = [tag["name"] for tag in runner["labels"]] - desc = RunnerDescription( - id=runner["id"], - name=runner["name"], - tags=tags, - offline=runner["status"] == "offline", - busy=runner["busy"], - ) - result.append(desc) - - return result - - -def cached_value_is_valid(updated_at: float, ttl: float) -> bool: - "a common function to identify if cachable value is still valid" - if updated_at == 0: - return False - if time.time() - ttl < updated_at: - return True - return False diff --git a/tests/ci/lambda_shared_package/lambda_shared/pr.py b/tests/ci/lambda_shared_package/lambda_shared/pr.py deleted file mode 100644 index e981e28a454..00000000000 --- a/tests/ci/lambda_shared_package/lambda_shared/pr.py +++ /dev/null @@ -1,189 +0,0 @@ -#!/usr/bin/env python - -import re -from typing import Tuple - -# Individual trusted contirbutors who are not in any trusted organization. -# Can be changed in runtime: we will append users that we learned to be in -# a trusted org, to save GitHub API calls. -TRUSTED_CONTRIBUTORS = { - e.lower() - for e in [ - "achimbab", # Kakao corp - "Algunenano", # Raúl Marín, ClickHouse, Inc - "amosbird", - "azat", # SEMRush - "bharatnc", # Many contributions. - "bobrik", # Seasoned contributor, CloudFlare - "cwurm", # ClickHouse, Inc - "den-crane", # Documentation contributor - "hagen1778", # Roman Khavronenko, seasoned contributor - "hczhcz", - "hexiaoting", # Seasoned contributor - "ildus", # adjust, ex-pgpro - "javisantana", # a Spanish ClickHouse enthusiast, ex-Carto - "kreuzerkrieg", - "nikvas0", - "nvartolomei", # Seasoned contributor, CloudFlare - "spongedu", # Seasoned contributor - "taiyang-li", - "ucasFL", # Amos Bird's friend - "vdimir", # ClickHouse, Inc - "YiuRULE", - "zlobober", # Developer of YT - "ilejn", # Arenadata, responsible for Kerberized Kafka - "thomoco", # ClickHouse, Inc - "BoloniniD", # Seasoned contributor, HSE - "tonickkozlov", # Cloudflare - "tylerhannan", # ClickHouse, Inc - "myrrc", # Mike Kot, DoubleCloud - "thevar1able", # ClickHouse, Inc - "aalexfvk", - "MikhailBurdukov", - "tsolodov", # ClickHouse, Inc - "kitaisreal", - "k-morozov", # Konstantin Morozov, Yandex Cloud - "justindeguzman", # ClickHouse, Inc - "jrdi", # ClickHouse contributor, TinyBird - ] -} - - -class Labels: - PR_BUGFIX = "pr-bugfix" - PR_CRITICAL_BUGFIX = "pr-critical-bugfix" - CAN_BE_TESTED = "can be tested" - DO_NOT_TEST = "do not test" - MUST_BACKPORT = "pr-must-backport" - MUST_BACKPORT_CLOUD = "pr-must-backport-cloud" - JEPSEN_TEST = "jepsen-test" - SKIP_MERGEABLE_CHECK = "skip mergeable check" - PR_BACKPORT = "pr-backport" - PR_BACKPORTS_CREATED = "pr-backports-created" - PR_BACKPORTS_CREATED_CLOUD = "pr-backports-created-cloud" - PR_CHERRYPICK = "pr-cherrypick" - PR_CI = "pr-ci" - PR_FEATURE = "pr-feature" - PR_SYNCED_TO_CLOUD = "pr-synced-to-cloud" - PR_SYNC_UPSTREAM = "pr-sync-upstream" - RELEASE = "release" - RELEASE_LTS = "release-lts" - SUBMODULE_CHANGED = "submodule changed" - - # automatic backport for critical bug fixes - AUTO_BACKPORT = {"pr-critical-bugfix"} - - -# Descriptions are used in .github/PULL_REQUEST_TEMPLATE.md, keep comments there -# updated accordingly -# The following lists are append only, try to avoid editing them -# They still could be cleaned out after the decent time though. -LABEL_CATEGORIES = { - "pr-backward-incompatible": ["Backward Incompatible Change"], - "pr-bugfix": [ - "Bug Fix", - "Bug Fix (user-visible misbehavior in an official stable release)", - "Bug Fix (user-visible misbehaviour in official stable or prestable release)", - "Bug Fix (user-visible misbehavior in official stable or prestable release)", - ], - "pr-critical-bugfix": ["Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC)"], - "pr-build": [ - "Build/Testing/Packaging Improvement", - "Build Improvement", - "Build/Testing Improvement", - "Build", - "Packaging Improvement", - ], - "pr-documentation": [ - "Documentation (changelog entry is not required)", - "Documentation", - ], - "pr-feature": ["New Feature"], - "pr-improvement": ["Improvement"], - "pr-not-for-changelog": [ - "Not for changelog (changelog entry is not required)", - "Not for changelog", - ], - "pr-performance": ["Performance Improvement"], - "pr-ci": ["CI Fix or Improvement (changelog entry is not required)"], -} - -CATEGORY_TO_LABEL = { - c: lb for lb, categories in LABEL_CATEGORIES.items() for c in categories -} - - -def check_pr_description(pr_body: str, repo_name: str) -> Tuple[str, str]: - """The function checks the body to being properly formatted according to - .github/PULL_REQUEST_TEMPLATE.md, if the first returned string is not empty, - then there is an error.""" - lines = list(map(lambda x: x.strip(), pr_body.split("\n") if pr_body else [])) - lines = [re.sub(r"\s+", " ", line) for line in lines] - - # Check if body contains "Reverts ClickHouse/ClickHouse#36337" - if [True for line in lines if re.match(rf"\AReverts {repo_name}#[\d]+\Z", line)]: - return "", LABEL_CATEGORIES["pr-not-for-changelog"][0] - - category = "" - entry = "" - description_error = "" - - i = 0 - while i < len(lines): - if re.match(r"(?i)^[#>*_ ]*change\s*log\s*category", lines[i]): - i += 1 - if i >= len(lines): - break - # Can have one empty line between header and the category - # itself. Filter it out. - if not lines[i]: - i += 1 - if i >= len(lines): - break - category = re.sub(r"^[-*\s]*", "", lines[i]) - i += 1 - - # Should not have more than one category. Require empty line - # after the first found category. - if i >= len(lines): - break - if lines[i]: - second_category = re.sub(r"^[-*\s]*", "", lines[i]) - description_error = ( - "More than one changelog category specified: " - f"'{category}', '{second_category}'" - ) - return description_error, category - - elif re.match( - r"(?i)^[#>*_ ]*(short\s*description|change\s*log\s*entry)", lines[i] - ): - i += 1 - # Can have one empty line between header and the entry itself. - # Filter it out. - if i < len(lines) and not lines[i]: - i += 1 - # All following lines until empty one are the changelog entry. - entry_lines = [] - while i < len(lines) and lines[i]: - entry_lines.append(lines[i]) - i += 1 - entry = " ".join(entry_lines) - # Don't accept changelog entries like '...'. - entry = re.sub(r"[#>*_.\- ]", "", entry) - # Don't accept changelog entries like 'Close #12345'. - entry = re.sub(r"^[\w\-\s]{0,10}#?\d{5,6}\.?$", "", entry) - else: - i += 1 - - if not category: - description_error = "Changelog category is empty" - # Filter out the PR categories that are not for changelog. - elif "(changelog entry is not required)" in category: - pass # to not check the rest of the conditions - elif category not in CATEGORY_TO_LABEL: - description_error, category = f"Category '{category}' is not valid", "" - elif not entry: - description_error = f"Changelog entry required for category '{category}'" - - return description_error, category diff --git a/tests/ci/lambda_shared_package/lambda_shared/token.py b/tests/ci/lambda_shared_package/lambda_shared/token.py deleted file mode 100644 index 9749122bd39..00000000000 --- a/tests/ci/lambda_shared_package/lambda_shared/token.py +++ /dev/null @@ -1,94 +0,0 @@ -"""Module to get the token for GitHub""" -from dataclasses import dataclass -import json -import time -from typing import Tuple - -import boto3 # type: ignore -import jwt -import requests - -from . import cached_value_is_valid - - -def get_key_and_app_from_aws() -> Tuple[str, int]: - secret_name = "clickhouse_github_secret_key" - session = boto3.session.Session() - client = session.client( - service_name="secretsmanager", - ) - get_secret_value_response = client.get_secret_value(SecretId=secret_name) - data = json.loads(get_secret_value_response["SecretString"]) - return data["clickhouse-app-key"], int(data["clickhouse-app-id"]) - - -def get_installation_id(jwt_token: str) -> int: - headers = { - "Authorization": f"Bearer {jwt_token}", - "Accept": "application/vnd.github.v3+json", - } - response = requests.get("https://api.github.com/app/installations", headers=headers) - response.raise_for_status() - data = response.json() - for installation in data: - if installation["account"]["login"] == "ClickHouse": - installation_id = installation["id"] - - return installation_id # type: ignore - - -def get_access_token_by_jwt(jwt_token: str, installation_id: int) -> str: - headers = { - "Authorization": f"Bearer {jwt_token}", - "Accept": "application/vnd.github.v3+json", - } - response = requests.post( - f"https://api.github.com/app/installations/{installation_id}/access_tokens", - headers=headers, - ) - response.raise_for_status() - data = response.json() - return data["token"] # type: ignore - - -def get_token_from_aws() -> str: - private_key, app_id = get_key_and_app_from_aws() - return get_access_token_by_key_app(private_key, app_id) - - -def get_access_token_by_key_app(private_key: str, app_id: int) -> str: - payload = { - "iat": int(time.time()) - 60, - "exp": int(time.time()) + (10 * 60), - "iss": app_id, - } - - # FIXME: apparently should be switched to this so that mypy is happy - # jwt_instance = JWT() - # encoded_jwt = jwt_instance.encode(payload, private_key, algorithm="RS256") - encoded_jwt = jwt.encode(payload, private_key, algorithm="RS256") # type: ignore - installation_id = get_installation_id(encoded_jwt) - return get_access_token_by_jwt(encoded_jwt, installation_id) - - -@dataclass -class CachedToken: - time: float - value: str - updating: bool = False - - -_cached_token = CachedToken(0, "") - - -def get_cached_access_token() -> str: - if time.time() - 550 < _cached_token.time or _cached_token.updating: - return _cached_token.value - # Indicate that the value is updating now, so the cached value can be - # used. The first setting and close-to-ttl are not counted as update - _cached_token.updating = cached_value_is_valid(_cached_token.time, 590) - private_key, app_id = get_key_and_app_from_aws() - _cached_token.time = time.time() - _cached_token.value = get_access_token_by_key_app(private_key, app_id) - _cached_token.updating = False - return _cached_token.value diff --git a/tests/ci/lambda_shared_package/pyproject.toml b/tests/ci/lambda_shared_package/pyproject.toml deleted file mode 100644 index dff36b89fbb..00000000000 --- a/tests/ci/lambda_shared_package/pyproject.toml +++ /dev/null @@ -1,24 +0,0 @@ -[build-system] -requires = ["setuptools"] -build-backend = "setuptools.build_meta" - -[project] -name = "lambda_shared" -version = "0.0.1" -dependencies = [ - "requests", - "urllib3 < 2" -] - -[project.optional-dependencies] -token = [ - "PyJWT", - "cryptography", -] -dev = [ - "boto3", - "lambda_shared[token]", -] - -[tool.distutils.bdist_wheel] -universal = true diff --git a/tests/ci/lambda_shared_package/setup.cfg b/tests/ci/lambda_shared_package/setup.cfg deleted file mode 100644 index 744280ae41b..00000000000 --- a/tests/ci/lambda_shared_package/setup.cfg +++ /dev/null @@ -1,8 +0,0 @@ -### This file exists for clear builds in docker ### -# without it the `build` directory wouldn't be # -# updated on the fly and will require manual clean # -[build] -build_base = /tmp/lambda_shared - -[egg_info] -egg_base = /tmp/ diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 1f5936c3fec..d9e33229932 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -74,7 +74,8 @@ def get_run_command( f"--volume={repo_path}/tests:/usr/share/clickhouse-test " f"--volume={result_path}:/test_output " "--security-opt seccomp=unconfined " # required to issue io_uring sys-calls - f"--cap-add=SYS_PTRACE {env_str} {additional_options_str} {image}" + f"--cap-add=SYS_PTRACE {env_str} {additional_options_str} {image} " + "python3 ./utils/runner.py" ) diff --git a/tests/ci/merge_pr.py b/tests/ci/merge_pr.py index 37c08fc4efe..13c7537a84b 100644 --- a/tests/ci/merge_pr.py +++ b/tests/ci/merge_pr.py @@ -26,6 +26,7 @@ from pr_info import PRInfo from report import SUCCESS, FAILURE from env_helper import GITHUB_UPSTREAM_REPOSITORY, GITHUB_REPOSITORY from synchronizer_utils import SYNC_BRANCH_PREFIX +from ci_config import CI # The team name for accepted approvals TEAM_NAME = getenv("GITHUB_TEAM_NAME", "core") @@ -247,29 +248,45 @@ def main(): repo = gh.get_repo(args.repo) if args.set_ci_status: + # set Mergeable check status and exit assert args.wf_status in (FAILURE, SUCCESS) - # set mergeable check status and exit commit = get_commit(gh, args.pr_info.sha) statuses = get_commit_filtered_statuses(commit) - state = trigger_mergeable_check( - commit, - statuses, - workflow_failed=(args.wf_status != "success"), - ) - # Process upstream StatusNames.SYNC - pr_info = PRInfo() - if ( - pr_info.head_ref.startswith(f"{SYNC_BRANCH_PREFIX}/pr/") - and GITHUB_REPOSITORY != GITHUB_UPSTREAM_REPOSITORY - ): - print("Updating upstream statuses") - update_upstream_sync_status(pr_info, state) + has_failed_statuses = False + has_native_failed_status = False + for status in statuses: + print(f"Check status [{status.context}], [{status.state}]") + if CI.is_required(status.context) and status.state != SUCCESS: + print(f"WARNING: Failed status [{status.context}], [{status.state}]") + has_failed_statuses = True + if status.context != CI.StatusNames.SYNC: + has_native_failed_status = True - if args.wf_status != "success": - # exit with 1 to rerun on workflow failed job restart + if args.wf_status == SUCCESS or has_failed_statuses: + # set Mergeable check if workflow is successful (green) + # or if we have GH statuses with failures (red) + # to avoid false-green on a died runner + state = trigger_mergeable_check( + commit, + statuses, + ) + # Process upstream StatusNames.SYNC + pr_info = PRInfo() + if ( + pr_info.head_ref.startswith(f"{SYNC_BRANCH_PREFIX}/pr/") + and GITHUB_REPOSITORY != GITHUB_UPSTREAM_REPOSITORY + ): + print("Updating upstream statuses") + update_upstream_sync_status(pr_info, state) + else: + print( + "Workflow failed but no failed statuses found (died runner?) - cannot set Mergeable Check status" + ) + if args.wf_status == SUCCESS and not has_native_failed_status: + sys.exit(0) + else: sys.exit(1) - sys.exit(0) # An ugly and not nice fix to patch the wrong organization URL, # see https://github.com/PyGithub/PyGithub/issues/2395#issuecomment-1378629710 diff --git a/tests/ci/pr_info.py b/tests/ci/pr_info.py index a411fc4e8f6..2c8ada7b983 100644 --- a/tests/ci/pr_info.py +++ b/tests/ci/pr_info.py @@ -15,7 +15,7 @@ from env_helper import ( GITHUB_SERVER_URL, GITHUB_UPSTREAM_REPOSITORY, ) -from lambda_shared_package.lambda_shared.pr import Labels +from ci_config import Labels from get_robot_token import get_best_robot_token from github_helper import GitHub @@ -296,13 +296,16 @@ class PRInfo: else: if "schedule" in github_event: self.event_type = EventType.SCHEDULE - else: + elif "inputs" in github_event: # assume this is a dispatch self.event_type = EventType.DISPATCH - logging.warning( - "event.json does not match pull_request or push:\n%s", - json.dumps(github_event, sort_keys=True, indent=4), - ) + print("PR Info:") + print(self) + else: + logging.warning( + "event.json does not match pull_request or push:\n%s", + json.dumps(github_event, sort_keys=True, indent=4), + ) self.sha = os.getenv( "GITHUB_SHA", "0000000000000000000000000000000000000000" ) @@ -334,6 +337,10 @@ class PRInfo: return True return False + @property + def is_push_event(self) -> bool: + return self.event_type == EventType.PUSH + @property def is_scheduled(self) -> bool: return self.event_type == EventType.SCHEDULE diff --git a/tests/ci/release.py b/tests/ci/release.py index 7ab1149310f..2de20d00a00 100755 --- a/tests/ci/release.py +++ b/tests/ci/release.py @@ -25,7 +25,7 @@ from contextlib import contextmanager from typing import Any, Final, Iterator, List, Optional, Tuple from git_helper import Git, commit, release_branch -from lambda_shared_package.lambda_shared.pr import Labels +from ci_config import Labels from report import SUCCESS from version_helper import ( FILE_WITH_VERSION_PATH, diff --git a/tests/ci/report.py b/tests/ci/report.py index bdaa2e15130..77043dfc8be 100644 --- a/tests/ci/report.py +++ b/tests/ci/report.py @@ -23,7 +23,7 @@ from typing import ( from build_download_helper import get_gh_api from ci_config import CI from ci_utils import normalize_string -from env_helper import REPORT_PATH, TEMP_PATH +from env_helper import REPORT_PATH, GITHUB_WORKSPACE logger = logging.getLogger(__name__) @@ -244,7 +244,11 @@ HTML_TEST_PART = """ """ BASE_HEADERS = ["Test name", "Test status"] -JOB_REPORT_FILE = Path(TEMP_PATH) / "job_report.json" +# should not be in TEMP directory or any directory that may be cleaned during the job execution +JOB_REPORT_FILE = Path(GITHUB_WORKSPACE) / "job_report.json" + +JOB_STARTED_TEST_NAME = "STARTED" +JOB_FINISHED_TEST_NAME = "COMPLETED" @dataclass @@ -296,6 +300,38 @@ class JobReport: build_dir_for_upload: Union[Path, str] = "" # if False no GH commit status will be created by CI need_commit_status: bool = True + # indicates that this is not real job report but report for the job that was skipped by rerun check + job_skipped: bool = False + # indicates that report generated by CI script in order to check later if job was killed before real report is generated + pre_report: bool = False + exit_code: int = -1 + + @staticmethod + def get_start_time_from_current(): + return datetime.datetime.utcnow().strftime("%Y-%m-%d %H:%M:%S") + + @classmethod + def create_pre_report(cls, status: str, job_skipped: bool) -> "JobReport": + return JobReport( + status=status, + description="", + test_results=[], + start_time=cls.get_start_time_from_current(), + duration=0.0, + additional_files=[], + job_skipped=job_skipped, + pre_report=True, + ) + + def update_duration(self): + if not self.start_time: + self.duration = 0.0 + else: + start_time = datetime.datetime.strptime( + self.start_time, "%Y-%m-%d %H:%M:%S" + ) + current_time = datetime.datetime.utcnow() + self.duration = (current_time - start_time).total_seconds() def __post_init__(self): assert self.status in (SUCCESS, ERROR, FAILURE, PENDING) diff --git a/tests/ci/run_check.py b/tests/ci/run_check.py index 00942352dde..0ad01e3accd 100644 --- a/tests/ci/run_check.py +++ b/tests/ci/run_check.py @@ -15,26 +15,22 @@ from commit_status_helper import ( ) from env_helper import GITHUB_REPOSITORY, GITHUB_SERVER_URL from get_robot_token import get_best_robot_token -from lambda_shared_package.lambda_shared.pr import ( - CATEGORY_TO_LABEL, - TRUSTED_CONTRIBUTORS, - Labels, - check_pr_description, -) +from ci_config import CI +from ci_utils import Utils from pr_info import PRInfo from report import FAILURE, PENDING, SUCCESS, StatusType -from ci_config import CI + TRUSTED_ORG_IDS = { 54801242, # clickhouse } -OK_SKIP_LABELS = {Labels.RELEASE, Labels.PR_BACKPORT, Labels.PR_CHERRYPICK} +OK_SKIP_LABELS = {CI.Labels.RELEASE, CI.Labels.PR_BACKPORT, CI.Labels.PR_CHERRYPICK} PR_CHECK = "PR Check" def pr_is_by_trusted_user(pr_user_login, pr_user_orgs): - if pr_user_login.lower() in TRUSTED_CONTRIBUTORS: + if pr_user_login.lower() in CI.TRUSTED_CONTRIBUTORS: logging.info("User '%s' is trusted", pr_user_login) return True @@ -63,13 +59,13 @@ def should_run_ci_for_pr(pr_info: PRInfo) -> Tuple[bool, str]: if OK_SKIP_LABELS.intersection(pr_info.labels): return True, "Don't try new checks for release/backports/cherry-picks" - if Labels.CAN_BE_TESTED not in pr_info.labels and not pr_is_by_trusted_user( + if CI.Labels.CAN_BE_TESTED not in pr_info.labels and not pr_is_by_trusted_user( pr_info.user_login, pr_info.user_orgs ): logging.info( "PRs by untrusted users need the '%s' label - " "please contact a member of the core team", - Labels.CAN_BE_TESTED, + CI.Labels.CAN_BE_TESTED, ) return False, "Needs 'can be tested' label" @@ -96,30 +92,32 @@ def main(): commit = get_commit(gh, pr_info.sha) status = SUCCESS # type: StatusType - description_error, category = check_pr_description(pr_info.body, GITHUB_REPOSITORY) + description_error, category = Utils.check_pr_description( + pr_info.body, GITHUB_REPOSITORY + ) pr_labels_to_add = [] pr_labels_to_remove = [] if ( - category in CATEGORY_TO_LABEL - and CATEGORY_TO_LABEL[category] not in pr_info.labels + category in CI.CATEGORY_TO_LABEL + and CI.CATEGORY_TO_LABEL[category] not in pr_info.labels ): - pr_labels_to_add.append(CATEGORY_TO_LABEL[category]) + pr_labels_to_add.append(CI.CATEGORY_TO_LABEL[category]) for label in pr_info.labels: if ( - label in CATEGORY_TO_LABEL.values() - and category in CATEGORY_TO_LABEL - and label != CATEGORY_TO_LABEL[category] + label in CI.CATEGORY_TO_LABEL.values() + and category in CI.CATEGORY_TO_LABEL + and label != CI.CATEGORY_TO_LABEL[category] ): pr_labels_to_remove.append(label) if pr_info.has_changes_in_submodules(): - pr_labels_to_add.append(Labels.SUBMODULE_CHANGED) - elif Labels.SUBMODULE_CHANGED in pr_info.labels: - pr_labels_to_remove.append(Labels.SUBMODULE_CHANGED) + pr_labels_to_add.append(CI.Labels.SUBMODULE_CHANGED) + elif CI.Labels.SUBMODULE_CHANGED in pr_info.labels: + pr_labels_to_remove.append(CI.Labels.SUBMODULE_CHANGED) - if any(label in Labels.AUTO_BACKPORT for label in pr_labels_to_add): - backport_labels = [Labels.MUST_BACKPORT, Labels.MUST_BACKPORT_CLOUD] + if any(label in CI.Labels.AUTO_BACKPORT for label in pr_labels_to_add): + backport_labels = [CI.Labels.MUST_BACKPORT, CI.Labels.MUST_BACKPORT_CLOUD] pr_labels_to_add += [ label for label in backport_labels if label not in pr_info.labels ] @@ -164,15 +162,15 @@ def main(): # 2. Then we check if the documentation is not created to fail the Mergeable check if ( - Labels.PR_FEATURE in pr_info.labels + CI.Labels.PR_FEATURE in pr_info.labels and not pr_info.has_changes_in_documentation() ): print( - f"::error ::The '{Labels.PR_FEATURE}' in the labels, " + f"::error ::The '{CI.Labels.PR_FEATURE}' in the labels, " "but there's no changed documentation" ) status = FAILURE - description = f"expect adding docs for {Labels.PR_FEATURE}" + description = f"expect adding docs for {CI.Labels.PR_FEATURE}" # 3. But we allow the workflow to continue # 4. And post only a single commit status on a failure diff --git a/tests/ci/runner_token_rotation_lambda/app.py b/tests/ci/runner_token_rotation_lambda/app.py deleted file mode 100644 index 445704648a9..00000000000 --- a/tests/ci/runner_token_rotation_lambda/app.py +++ /dev/null @@ -1,93 +0,0 @@ -#!/usr/bin/env python3 - -import argparse -import sys - -import boto3 # type: ignore -import requests -from lambda_shared.token import get_access_token_by_key_app, get_cached_access_token - - -def get_runner_registration_token(access_token): - headers = { - "Authorization": f"token {access_token}", - "Accept": "application/vnd.github.v3+json", - } - response = requests.post( - "https://api.github.com/orgs/ClickHouse/actions/runners/registration-token", - headers=headers, - timeout=30, - ) - response.raise_for_status() - data = response.json() - return data["token"] - - -def main(access_token, push_to_ssm, ssm_parameter_name): - runner_registration_token = get_runner_registration_token(access_token) - - if push_to_ssm: - print("Trying to put params into ssm manager") - client = boto3.client("ssm") - client.put_parameter( - Name=ssm_parameter_name, - Value=runner_registration_token, - Type="SecureString", - Overwrite=True, - ) - else: - print( - "Not push token to AWS Parameter Store, just print:", - runner_registration_token, - ) - - -def handler(event, context): - _, _ = event, context - main(get_cached_access_token(), True, "github_runner_registration_token") - - -if __name__ == "__main__": - parser = argparse.ArgumentParser( - description="Get new token from github to add runners" - ) - parser.add_argument( - "-p", "--private-key-path", help="Path to file with private key" - ) - parser.add_argument("-k", "--private-key", help="Private key") - parser.add_argument( - "-a", "--app-id", type=int, help="GitHub application ID", required=True - ) - parser.add_argument( - "--push-to-ssm", - action="store_true", - help="Store received token in parameter store", - ) - parser.add_argument( - "--ssm-parameter-name", - default="github_runner_registration_token", - help="AWS paramater store parameter name", - ) - - args = parser.parse_args() - - if not args.private_key_path and not args.private_key: - print( - "Either --private-key-path or --private-key must be specified", - file=sys.stderr, - ) - - if args.private_key_path and args.private_key: - print( - "Either --private-key-path or --private-key must be specified", - file=sys.stderr, - ) - - if args.private_key: - private_key = args.private_key - else: - with open(args.private_key_path, "r", encoding="utf-8") as key_file: - private_key = key_file.read() - - token = get_access_token_by_key_app(private_key, args.app_id) - main(token, args.push_to_ssm, args.ssm_parameter_name) diff --git a/tests/ci/runner_token_rotation_lambda/build_and_deploy_archive.sh b/tests/ci/runner_token_rotation_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/runner_token_rotation_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/runner_token_rotation_lambda/lambda_shared b/tests/ci/runner_token_rotation_lambda/lambda_shared deleted file mode 120000 index ba86e090f6c..00000000000 --- a/tests/ci/runner_token_rotation_lambda/lambda_shared +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package/lambda_shared \ No newline at end of file diff --git a/tests/ci/runner_token_rotation_lambda/requirements.txt b/tests/ci/runner_token_rotation_lambda/requirements.txt deleted file mode 100644 index 4cb3fba0f7b..00000000000 --- a/tests/ci/runner_token_rotation_lambda/requirements.txt +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package[token] diff --git a/tests/ci/slack_bot_ci_lambda/app.py b/tests/ci/slack_bot_ci_lambda/app.py deleted file mode 100755 index 94b71724b1c..00000000000 --- a/tests/ci/slack_bot_ci_lambda/app.py +++ /dev/null @@ -1,323 +0,0 @@ -#!/usr/bin/env python3 - -""" -A trivial stateless slack bot that notifies about new broken tests in ClickHouse CI. -It checks what happened to our CI during the last check_period hours (1 hour) and - notifies us in slack if necessary. -This script should be executed once each check_period hours (1 hour). -It will post duplicate messages if you run it more often; it will lose some messages - if you run it less often. - -You can run it locally with no arguments, it will work in a dry-run mode. - Or you can set your own SLACK_URL_DEFAULT. -Feel free to add more checks, more details to messages, or better heuristics. - -It's deployed to slack-bot-ci-lambda in CI/CD account - -See also: https://aretestsgreenyet.com/ -""" - -import base64 -import json -import os -import random - -import requests - -DRY_RUN_MARK = "" - -MAX_FAILURES_DEFAULT = 30 -SLACK_URL_DEFAULT = DRY_RUN_MARK - -FLAKY_ALERT_PROBABILITY = 0.50 -REPORT_NO_FAILURES_PROBABILITY = 0.99 - -MAX_TESTS_TO_REPORT = 4 - -# Slack has a stupid limitation on message size, it splits long messages into multiple, -# ones breaking formatting -MESSAGE_LENGTH_LIMIT = 4000 - -# Find tests that failed in master during the last check_period * 24 hours, -# but did not fail during the last 2 weeks. Assuming these tests were broken recently. -# Counts number of failures in check_period and check_period * 24 time windows -# to distinguish rare flaky tests from completely broken tests -NEW_BROKEN_TESTS_QUERY = """ -WITH - 1 AS check_period, - check_period * 24 AS extended_check_period, - now() as now -SELECT - test_name, - any(report_url), - countIf((check_start_time + check_duration_ms / 1000) < now - INTERVAL check_period HOUR) AS count_prev_periods, - countIf((check_start_time + check_duration_ms / 1000) >= now - INTERVAL check_period HOUR) AS count -FROM checks -WHERE 1 - AND check_start_time BETWEEN now - INTERVAL 1 WEEK AND now - AND (check_start_time + check_duration_ms / 1000) >= now - INTERVAL extended_check_period HOUR - AND pull_request_number = 0 - AND test_status LIKE 'F%' - AND check_status != 'success' - AND test_name NOT IN ( - SELECT test_name FROM checks WHERE 1 - AND check_start_time >= now - INTERVAL 1 MONTH - AND (check_start_time + check_duration_ms / 1000) BETWEEN now - INTERVAL 2 WEEK AND now - INTERVAL extended_check_period HOUR - AND pull_request_number = 0 - AND check_status != 'success' - AND test_status LIKE 'F%') - AND test_context_raw NOT LIKE '%CannotSendRequest%' and test_context_raw NOT LIKE '%Server does not respond to health check%' -GROUP BY test_name -ORDER BY (count_prev_periods + count) DESC -""" - -# Returns total number of failed checks during the last 24 hours -# and previous value of that metric (check_period hours ago) -COUNT_FAILURES_QUERY = """ -WITH - 1 AS check_period, - '%' AS check_name_pattern, - now() as now -SELECT - countIf((check_start_time + check_duration_ms / 1000) >= now - INTERVAL 24 HOUR) AS new_val, - countIf((check_start_time + check_duration_ms / 1000) <= now - INTERVAL check_period HOUR) AS prev_val -FROM checks -WHERE 1 - AND check_start_time >= now - INTERVAL 1 WEEK - AND (check_start_time + check_duration_ms / 1000) >= now - INTERVAL 24 + check_period HOUR - AND pull_request_number = 0 - AND test_status LIKE 'F%' - AND check_status != 'success' - AND check_name ILIKE check_name_pattern -""" - -# Returns percentage of failed checks (once per day, at noon) -FAILED_CHECKS_PERCENTAGE_QUERY = """ -SELECT if(toHour(now('Europe/Amsterdam')) = 12, v, 0) -FROM -( - SELECT - countDistinctIf((commit_sha, check_name), (test_status LIKE 'F%') AND (check_status != 'success')) - / countDistinct((commit_sha, check_name)) AS v - FROM checks - WHERE 1 - AND (pull_request_number = 0) - AND (test_status != 'SKIPPED') - AND (check_start_time > (now() - toIntervalDay(1))) -) -""" - -# It shows all recent failures of the specified test (helps to find when it started) -ALL_RECENT_FAILURES_QUERY = """ -WITH - '{}' AS name_substr, - 90 AS interval_days, - ('Stateless tests (asan)', 'Stateless tests (address)', 'Stateless tests (address, actions)', 'Integration tests (asan) [1/3]', 'Stateless tests (tsan) [1/3]') AS backport_and_release_specific_checks -SELECT - toStartOfDay(check_start_time) AS d, - count(), - groupUniqArray(pull_request_number) AS prs, - any(report_url) -FROM checks -WHERE ((now() - toIntervalDay(interval_days)) <= check_start_time) AND (pull_request_number NOT IN ( - SELECT pull_request_number AS prn - FROM checks - WHERE (prn != 0) AND ((now() - toIntervalDay(interval_days)) <= check_start_time) AND (check_name IN (backport_and_release_specific_checks)) -)) AND (position(test_name, name_substr) > 0) AND (test_status IN ('FAIL', 'ERROR', 'FLAKY')) -GROUP BY d -ORDER BY d DESC -""" - -SLACK_MESSAGE_JSON = {"type": "mrkdwn", "text": None} - - -def get_play_url(query): - return ( - "https://play.clickhouse.com/play?user=play#" - + base64.b64encode(query.encode()).decode() - ) - - -def run_clickhouse_query(query): - url = "https://play.clickhouse.com/?user=play&query=" + requests.compat.quote(query) - res = requests.get(url, timeout=30) - if res.status_code != 200: - print("Failed to execute query: ", res.status_code, res.content) - res.raise_for_status() - - lines = res.text.strip().splitlines() - return [x.split("\t") for x in lines] - - -def split_broken_and_flaky_tests(failed_tests): - if not failed_tests: - return None - - broken_tests = [] - flaky_tests = [] - for name, report, count_prev_str, count_str in failed_tests: - count_prev, count = int(count_prev_str), int(count_str) - if (count_prev < 2 <= count) or (count_prev == count == 1): - # It failed 2 times or more within extended time window, it's definitely broken. - # 2 <= count means that it was not reported as broken on previous runs - broken_tests.append([name, report]) - elif 0 < count and count_prev == 0: - # It failed only once, can be a rare flaky test - flaky_tests.append([name, report]) - - return broken_tests, flaky_tests - - -def format_failed_tests_list(failed_tests, failure_type): - if len(failed_tests) == 1: - res = f"There is a new {failure_type} test:\n" - else: - res = f"There are {len(failed_tests)} new {failure_type} tests:\n" - - for name, report in failed_tests[:MAX_TESTS_TO_REPORT]: - cidb_url = get_play_url(ALL_RECENT_FAILURES_QUERY.format(name)) - res += f"- *{name}* - <{report}|Report> - <{cidb_url}|CI DB> \n" - - if MAX_TESTS_TO_REPORT < len(failed_tests): - res += ( - f"- and {len(failed_tests) - MAX_TESTS_TO_REPORT} other " - "tests... :this-is-fine-fire:" - ) - - return res - - -def get_new_broken_tests_message(failed_tests): - if not failed_tests: - return None - - broken_tests, flaky_tests = split_broken_and_flaky_tests(failed_tests) - if len(broken_tests) == 0 and len(flaky_tests) == 0: - return None - - msg = "" - if len(broken_tests) > 0: - msg += format_failed_tests_list(broken_tests, "*BROKEN*") - elif random.random() > FLAKY_ALERT_PROBABILITY: - looks_like_fuzzer = [x[0].count(" ") > 2 for x in flaky_tests] - if not any(looks_like_fuzzer): - print("Will not report flaky tests to avoid noise: ", flaky_tests) - return None - - if len(flaky_tests) > 0: - if len(msg) > 0: - msg += "\n" - msg += format_failed_tests_list(flaky_tests, "flaky") - - return msg - - -def get_too_many_failures_message_impl(failures_count): - MAX_FAILURES = int(os.environ.get("MAX_FAILURES", MAX_FAILURES_DEFAULT)) - curr_failures = int(failures_count[0][0]) - prev_failures = int(failures_count[0][1]) - if curr_failures == 0 and prev_failures != 0: - if random.random() < REPORT_NO_FAILURES_PROBABILITY: - return None - return "Wow, there are *no failures* at all... 0_o" - return_none = ( - curr_failures < MAX_FAILURES - or curr_failures < prev_failures - or (curr_failures - prev_failures) / prev_failures < 0.2 - ) - if return_none: - return None - if prev_failures < MAX_FAILURES: - return f":alert: *CI is broken: there are {curr_failures} failures during the last 24 hours*" - return "CI is broken and it's getting worse: there are {curr_failures} failures during the last 24 hours" - - -def get_too_many_failures_message(failures_count): - msg = get_too_many_failures_message_impl(failures_count) - if msg: - msg += "\nSee https://aretestsgreenyet.com/" - return msg - - -def get_failed_checks_percentage_message(percentage): - p = float(percentage[0][0]) * 100 - - # Always report more than 1% of failed checks - # For <= 1%: higher percentage of failures == higher probability - if p <= random.random(): - return None - - msg = ":alert: " if p > 1 else "Only " if p < 0.5 else "" - msg += f"*{p:.2f}%* of all checks in master have failed yesterday" - return msg - - -def split_slack_message(long_message): - lines = long_message.split("\n") - messages = [] - curr_msg = "" - for line in lines: - if len(curr_msg) + len(line) < MESSAGE_LENGTH_LIMIT: - curr_msg += "\n" - curr_msg += line - else: - messages.append(curr_msg) - curr_msg = line - messages.append(curr_msg) - return messages - - -def send_to_slack_impl(message): - SLACK_URL = os.environ.get("SLACK_URL", SLACK_URL_DEFAULT) - if SLACK_URL == DRY_RUN_MARK: - return - - payload = SLACK_MESSAGE_JSON.copy() - payload["text"] = message - res = requests.post(SLACK_URL, json.dumps(payload), timeout=30) - if res.status_code != 200: - print("Failed to send a message to Slack: ", res.status_code, res.content) - res.raise_for_status() - - -def send_to_slack(message): - messages = split_slack_message(message) - for msg in messages: - send_to_slack_impl(msg) - - -def query_and_alert_if_needed(query, get_message_func): - query_res = run_clickhouse_query(query) - print("Got result {} for query {}", query_res, query) - msg = get_message_func(query_res) - if msg is None: - return - - msg += f"\nCI DB query: <{get_play_url(query)}|link>" - print("Sending message to slack:", msg) - send_to_slack(msg) - - -def check_and_alert(): - query_and_alert_if_needed(NEW_BROKEN_TESTS_QUERY, get_new_broken_tests_message) - query_and_alert_if_needed(COUNT_FAILURES_QUERY, get_too_many_failures_message) - query_and_alert_if_needed( - FAILED_CHECKS_PERCENTAGE_QUERY, get_failed_checks_percentage_message - ) - - -def handler(event, context): - _, _ = event, context - try: - check_and_alert() - return {"statusCode": 200, "body": "OK"} - except Exception as e: - send_to_slack( - "I failed, please help me " - f"(see ClickHouse/ClickHouse/tests/ci/slack_bot_ci_lambda/app.py): {e}" - ) - return {"statusCode": 200, "body": "FAIL"} - - -if __name__ == "__main__": - check_and_alert() diff --git a/tests/ci/slack_bot_ci_lambda/build_and_deploy_archive.sh b/tests/ci/slack_bot_ci_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/slack_bot_ci_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/slack_bot_ci_lambda/requirements.txt b/tests/ci/slack_bot_ci_lambda/requirements.txt deleted file mode 100644 index 098e04a9798..00000000000 --- a/tests/ci/slack_bot_ci_lambda/requirements.txt +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package diff --git a/tests/ci/ssh.py b/tests/ci/ssh.py index 321826fcf44..89d90d724d2 100644 --- a/tests/ci/ssh.py +++ b/tests/ci/ssh.py @@ -37,9 +37,9 @@ class SSHAgent: ssh_options = ( "," + os.environ["SSH_OPTIONS"] if os.environ.get("SSH_OPTIONS") else "" ) - os.environ[ - "SSH_OPTIONS" - ] = f"{ssh_options}UserKnownHostsFile=/dev/null,StrictHostKeyChecking=no" + os.environ["SSH_OPTIONS"] = ( + f"{ssh_options}UserKnownHostsFile=/dev/null,StrictHostKeyChecking=no" + ) def add(self, key): key_pub = self._key_pub(key) diff --git a/tests/ci/stress_check.py b/tests/ci/stress_check.py index bf0281cae68..85da601e379 100644 --- a/tests/ci/stress_check.py +++ b/tests/ci/stress_check.py @@ -3,6 +3,7 @@ import csv import logging import os +import re import subprocess import sys from pathlib import Path @@ -19,6 +20,18 @@ from stopwatch import Stopwatch from tee_popen import TeePopen +class SensitiveFormatter(logging.Formatter): + @staticmethod + def _filter(s): + return re.sub( + r"(.*)(AZURE_CONNECTION_STRING.*\')(.*)", r"\1AZURE_CONNECTION_STRING\3", s + ) + + def format(self, record): + original = logging.Formatter.format(self, record) + return self._filter(original) + + def get_additional_envs(check_name: str) -> List[str]: result = [] azure_connection_string = get_parameter_from_ssm("azure_connection_string") @@ -30,6 +43,9 @@ def get_additional_envs(check_name: str) -> List[str]: if "azure" in check_name: result.append("USE_AZURE_STORAGE_FOR_MERGE_TREE=1") + if "s3" in check_name: + result.append("USE_S3_STORAGE_FOR_MERGE_TREE=1") + return result @@ -114,6 +130,9 @@ def process_results( def run_stress_test(docker_image_name: str) -> None: logging.basicConfig(level=logging.INFO) + for handler in logging.root.handlers: + # pylint: disable=protected-access + handler.setFormatter(SensitiveFormatter(handler.formatter._fmt)) # type: ignore stopwatch = Stopwatch() temp_path = Path(TEMP_PATH) diff --git a/tests/ci/style_check.py b/tests/ci/style_check.py index 9deae06d9f4..36620d44a2d 100644 --- a/tests/ci/style_check.py +++ b/tests/ci/style_check.py @@ -16,7 +16,15 @@ from docker_images_helper import get_docker_image, pull_image from env_helper import IS_CI, REPO_COPY, TEMP_PATH, GITHUB_EVENT_PATH from git_helper import GIT_PREFIX, git_runner from pr_info import PRInfo -from report import ERROR, FAILURE, SUCCESS, JobReport, TestResults, read_test_results +from report import ( + ERROR, + FAILURE, + SUCCESS, + JobReport, + TestResults, + read_test_results, + FAIL, +) from ssh import SSHKey from stopwatch import Stopwatch @@ -192,15 +200,6 @@ def main(): future = executor.submit(subprocess.run, cmd_shell, shell=True) _ = future.result() - autofix_description = "" - if args.push: - try: - commit_push_staged(pr_info) - except subprocess.SubprocessError: - # do not fail the whole script if the autofix didn't work out - logging.error("Unable to push the autofix. Continue.") - autofix_description = "Failed to push autofix to the PR. " - subprocess.check_call( f"python3 ../../utils/check-style/process_style_check_result.py --in-results-dir {temp_path} " f"--out-results-file {temp_path}/test_results.tsv --out-status-file {temp_path}/check_status.tsv || " @@ -210,6 +209,21 @@ def main(): state, description, test_results, additional_files = process_result(temp_path) + autofix_description = "" + fail_cnt = 0 + for result in test_results: + if result.status in (FAILURE, FAIL): + # do not autofix if not only black failed + fail_cnt += 1 + + if args.push and fail_cnt == 1: + try: + commit_push_staged(pr_info) + except subprocess.SubprocessError: + # do not fail the whole script if the autofix didn't work out + logging.error("Unable to push the autofix. Continue.") + autofix_description = "Failed to push autofix to the PR. " + JobReport( description=f"{autofix_description}{description}", test_results=test_results, diff --git a/tests/ci/sync_pr.py b/tests/ci/sync_pr.py index 8251ccbaf38..1b71231f820 100644 --- a/tests/ci/sync_pr.py +++ b/tests/ci/sync_pr.py @@ -101,23 +101,20 @@ def main(): assert pr_info.merged_pr, "BUG. merged PR number could not been determined" prs = gh.get_pulls_from_search( - query=f"head:sync-upstream/pr/{pr_info.merged_pr} org:ClickHouse type:pr", + query=f"head:sync-upstream/pr/{pr_info.merged_pr} org:ClickHouse type:pr is:open", repo="ClickHouse/clickhouse-private", ) - sync_pr = None - if len(prs) > 1: print(f"WARNING: More than one PR found [{prs}] - exiting") elif len(prs) == 0: print("WARNING: No Sync PR found") else: sync_pr = prs[0] - - if args.merge: - merge_sync_pr(gh, sync_pr) - elif args.status: - set_sync_status(gh, pr_info, sync_pr) + if args.merge: + merge_sync_pr(gh, sync_pr) + elif args.status: + set_sync_status(gh, pr_info, sync_pr) if __name__ == "__main__": diff --git a/tests/ci/team_keys_lambda/app.py b/tests/ci/team_keys_lambda/app.py deleted file mode 100644 index c056808a9b4..00000000000 --- a/tests/ci/team_keys_lambda/app.py +++ /dev/null @@ -1,136 +0,0 @@ -#!/usr/bin/env python3 - -import argparse -import json -from datetime import datetime -from queue import Queue -from threading import Thread - -import boto3 # type: ignore -import requests - - -class Keys(set): - def __init__(self, *args, **kwargs): - super().__init__(*args, **kwargs) - self.updated_at = 0.0 - - def update_now(self): - self.updated_at = datetime.now().timestamp() - - -keys = Keys() - - -class Worker(Thread): - def __init__(self, request_queue): - Thread.__init__(self) - self.queue = request_queue - self.results = set() - - def run(self): - while True: - m = self.queue.get() - if m == "": - break - response = requests.get(f"https://github.com/{m}.keys", timeout=30) - self.results.add(f"# {m}\n{response.text}\n") - self.queue.task_done() - - -def get_org_team_members(token: str, org: str, team_slug: str) -> set: - headers = { - "Authorization": f"token {token}", - "Accept": "application/vnd.github.v3+json", - } - response = requests.get( - f"https://api.github.com/orgs/{org}/teams/{team_slug}/members", - headers=headers, - timeout=30, - ) - response.raise_for_status() - data = response.json() - return set(m["login"] for m in data) - - -def get_cached_members_keys(members: set) -> Keys: - if (datetime.now().timestamp() - 3600) <= keys.updated_at: - return keys - - q = Queue() # type: Queue - workers = [] - for m in members: - q.put(m) - # Create workers and add to the queue - worker = Worker(q) - worker.start() - workers.append(worker) - - # Workers keep working till they receive an empty string - for _ in workers: - q.put("") - - # Join workers to wait till they finished - for worker in workers: - worker.join() - - keys.clear() - for worker in workers: - keys.update(worker.results) - keys.update_now() - return keys - - -def get_token_from_aws() -> str: - # We need a separate token, since the clickhouse-ci app does not have - # access to the organization members' endpoint - secret_name = "clickhouse_robot_token" - session = boto3.session.Session() - client = session.client( - service_name="secretsmanager", - ) - get_secret_value_response = client.get_secret_value(SecretId=secret_name) - data = json.loads(get_secret_value_response["SecretString"]) - return data["clickhouse_robot_token"] # type: ignore - - -def main(token: str, org: str, team_slug: str) -> str: - members = get_org_team_members(token, org, team_slug) - keys = get_cached_members_keys(members) - - return "".join(sorted(keys)) - - -def handler(event, context): - _ = context - _ = event - if keys.updated_at < (datetime.now().timestamp() - 3600): - token = get_token_from_aws() - body = main(token, "ClickHouse", "core") - else: - body = "".join(sorted(keys)) - - result = { - "statusCode": 200, - "headers": { - "Content-Type": "text/html", - }, - "body": body, - } - return result - - -if __name__ == "__main__": - parser = argparse.ArgumentParser( - description="Get the public SSH keys for members of given org and team" - ) - parser.add_argument("--token", required=True, help="Github PAT") - parser.add_argument( - "--organization", help="GitHub organization name", default="ClickHouse" - ) - parser.add_argument("--team", help="GitHub team name", default="core") - - args = parser.parse_args() - output = main(args.token, args.organization, args.team) - - print(f"# Just showing off the keys:\n{output}") diff --git a/tests/ci/team_keys_lambda/build_and_deploy_archive.sh b/tests/ci/team_keys_lambda/build_and_deploy_archive.sh deleted file mode 100644 index b72bce4a677..00000000000 --- a/tests/ci/team_keys_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1,76 +0,0 @@ -#!/usr/bin/env bash -set -xeo pipefail - -WORKDIR=$(dirname "$0") -WORKDIR=$(readlink -f "${WORKDIR}") -DIR_NAME=$(basename "$WORKDIR") -cd "$WORKDIR" - -# Do not deploy the lambda to AWS -DRY_RUN=${DRY_RUN:-} -# Python runtime to install dependencies -PY_VERSION=${PY_VERSION:-3.10} -PY_EXEC="python${PY_VERSION}" -# Image to build the lambda zip package -DOCKER_IMAGE="public.ecr.aws/lambda/python:${PY_VERSION}" -# Rename the_lambda_name directory to the-lambda-name lambda in AWS -LAMBDA_NAME=${DIR_NAME//_/-} -# The name of directory with lambda code -PACKAGE=lambda-package - -# Do not rebuild and deploy the archive if it's newer than sources -if [ -e "$PACKAGE.zip" ] && [ -z "$FORCE" ]; then - REBUILD="" - for src in app.py build_and_deploy_archive.sh requirements.txt lambda_shared/*; do - if [ "$src" -nt "$PACKAGE.zip" ]; then - REBUILD=1 - fi - done - [ -n "$REBUILD" ] || exit 0 -fi - -docker_cmd=( - docker run -i --net=host --rm --user="${UID}" -e HOME=/tmp --entrypoint=/bin/bash - --volume="${WORKDIR}/..:/ci" --workdir="/ci/${DIR_NAME}" "${DOCKER_IMAGE}" -) -rm -rf "$PACKAGE" "$PACKAGE".zip -mkdir "$PACKAGE" -cp app.py "$PACKAGE" -if [ -f requirements.txt ]; then - VENV=lambda-venv - rm -rf "$VENV" - "${docker_cmd[@]}" -ex < dict: - """return cached instances description with updating it once per five minutes""" - if time.time() - 250 < cached_instances.time or cached_instances.updating: - return cached_instances.value - cached_instances.updating = cached_value_is_valid(cached_instances.time, 300) - ec2_client = boto3.client("ec2") - instances_response = ec2_client.describe_instances( - Filters=[{"Name": "instance-state-name", "Values": ["running"]}] - ) - cached_instances.time = time.time() - cached_instances.value = { - instance["InstanceId"]: instance - for reservation in instances_response["Reservations"] - for instance in reservation["Instances"] - } - cached_instances.updating = False - return cached_instances.value - - -@dataclass -class CachedRunners: - time: float - value: RunnerDescriptions - updating: bool = False - - -cached_runners = CachedRunners(0, []) - - -def get_cached_runners(access_token: str) -> RunnerDescriptions: - """From time to time request to GH api costs up to 3 seconds, and - it's a disaster from the termination lambda perspective""" - if time.time() - 5 < cached_runners.time or cached_instances.updating: - return cached_runners.value - cached_runners.updating = cached_value_is_valid(cached_runners.time, 15) - cached_runners.value = list_runners(access_token) - cached_runners.time = time.time() - cached_runners.updating = False - return cached_runners.value - - -def how_many_instances_to_kill(event_data: dict) -> Dict[str, int]: - data_array = event_data["CapacityToTerminate"] - to_kill_by_zone = {} # type: Dict[str, int] - for av_zone in data_array: - zone_name = av_zone["AvailabilityZone"] - to_kill = av_zone["Capacity"] - if zone_name not in to_kill_by_zone: - to_kill_by_zone[zone_name] = 0 - - to_kill_by_zone[zone_name] += to_kill - - return to_kill_by_zone - - -def get_candidates_to_be_killed(event_data: dict) -> Dict[str, List[str]]: - data_array = event_data["Instances"] - instances_by_zone = {} # type: Dict[str, List[str]] - for instance in data_array: - zone_name = instance["AvailabilityZone"] - instance_id = instance["InstanceId"] # type: str - if zone_name not in instances_by_zone: - instances_by_zone[zone_name] = [] - instances_by_zone[zone_name].append(instance_id) - - return instances_by_zone - - -def main(access_token: str, event: dict) -> Dict[str, List[str]]: - start = time.time() - print("Got event", json.dumps(event, sort_keys=True).replace("\n", "")) - to_kill_by_zone = how_many_instances_to_kill(event) - instances_by_zone = get_candidates_to_be_killed(event) - # Getting ASG and instances' descriptions from the API - # We don't kill instances that alive for less than 10 minutes, since they - # could be not in the GH active runners yet - print(f"Check other hosts from the same ASG {event['AutoScalingGroupName']}") - asg_client = boto3.client("autoscaling") - as_groups_response = asg_client.describe_auto_scaling_groups( - AutoScalingGroupNames=[event["AutoScalingGroupName"]] - ) - assert len(as_groups_response["AutoScalingGroups"]) == 1 - asg = as_groups_response["AutoScalingGroups"][0] - asg_instance_ids = [instance["InstanceId"] for instance in asg["Instances"]] - instance_descriptions = get_cached_instances() - # The instances launched less than 10 minutes ago - immune_ids = [ - instance["InstanceId"] - for instance in instance_descriptions.values() - if start - instance["LaunchTime"].timestamp() < 600 - ] - # if the ASG's instance ID not in instance_descriptions, it's most probably - # is not cached yet, so we must mark it as immuned - immune_ids.extend( - iid for iid in asg_instance_ids if iid not in instance_descriptions - ) - print("Time spent on the requests to AWS: ", time.time() - start) - - runners = get_cached_runners(access_token) - runner_ids = set(runner.name for runner in runners) - # We used to delete potential hosts to terminate from GitHub runners pool, - # but the documentation states: - # --- Returning an instance first in the response data does not guarantee its termination - # so they will be cleaned out by ci_runners_metrics_lambda eventually - - instances_to_kill = [] - total_to_kill = 0 - for zone, num_to_kill in to_kill_by_zone.items(): - candidates = instances_by_zone[zone] - total_to_kill += num_to_kill - if num_to_kill > len(candidates): - raise RuntimeError( - f"Required to kill {num_to_kill}, but have only {len(candidates)}" - f" candidates in AV {zone}" - ) - - delete_for_av = [] # type: RunnerDescriptions - for candidate in candidates: - if candidate in immune_ids: - print( - f"Candidate {candidate} started less than 10 minutes ago, won't touch a child" - ) - break - if candidate not in runner_ids: - print( - f"Candidate {candidate} was not in runners list, simply delete it" - ) - instances_to_kill.append(candidate) - break - if len(delete_for_av) + len(instances_to_kill) == num_to_kill: - break - if candidate in instances_to_kill: - continue - - for runner in runners: - if runner.name == candidate: - if not runner.busy: - print( - f"Runner {runner.name} is not busy and can be deleted from AV {zone}" - ) - delete_for_av.append(runner) - else: - print(f"Runner {runner.name} is busy, not going to delete it") - break - - if len(delete_for_av) < num_to_kill: - print( - f"Checked all candidates for av {zone}, get to delete " - f"{len(delete_for_av)}, but still cannot get required {num_to_kill}" - ) - - instances_to_kill += [runner.name for runner in delete_for_av] - - if len(instances_to_kill) < total_to_kill: - for instance in asg_instance_ids: - if instance in immune_ids: - continue - for runner in runners: - if runner.name == instance and not runner.busy: - print(f"Runner {runner.name} is not busy and can be deleted") - instances_to_kill.append(runner.name) - - if total_to_kill <= len(instances_to_kill): - print("Got enough instances to kill") - break - - response = {"InstanceIDs": instances_to_kill} - print("Got instances to kill: ", response) - print("Time spent on the request: ", time.time() - start) - return response - - -def handler(event: dict, context: Any) -> Dict[str, List[str]]: - _ = context - return main(get_cached_access_token(), event) - - -if __name__ == "__main__": - parser = argparse.ArgumentParser(description="Get list of runners and their states") - parser.add_argument( - "-p", "--private-key-path", help="Path to file with private key" - ) - parser.add_argument("-k", "--private-key", help="Private key") - parser.add_argument( - "-a", "--app-id", type=int, help="GitHub application ID", required=True - ) - - args = parser.parse_args() - - if not args.private_key_path and not args.private_key: - print( - "Either --private-key-path or --private-key must be specified", - file=sys.stderr, - ) - - if args.private_key_path and args.private_key: - print( - "Either --private-key-path or --private-key must be specified", - file=sys.stderr, - ) - - if args.private_key: - private_key = args.private_key - else: - with open(args.private_key_path, "r", encoding="utf-8") as key_file: - private_key = key_file.read() - - token = get_access_token_by_key_app(private_key, args.app_id) - - sample_event = { - "AutoScalingGroupARN": "arn:aws:autoscaling:us-east-1::autoScalingGroup:d4738357-2d40-4038-ae7e-b00ae0227003:autoScalingGroupName/my-asg", - "AutoScalingGroupName": "my-asg", - "CapacityToTerminate": [ - { - "AvailabilityZone": "us-east-1b", - "Capacity": 1, - "InstanceMarketOption": "OnDemand", - }, - { - "AvailabilityZone": "us-east-1c", - "Capacity": 2, - "InstanceMarketOption": "OnDemand", - }, - ], - "Instances": [ - { - "AvailabilityZone": "us-east-1b", - "InstanceId": "i-08d0b3c1a137e02a5", - "InstanceType": "t2.nano", - "InstanceMarketOption": "OnDemand", - }, - { - "AvailabilityZone": "us-east-1c", - "InstanceId": "ip-172-31-45-253.eu-west-1.compute.internal", - "InstanceType": "t2.nano", - "InstanceMarketOption": "OnDemand", - }, - { - "AvailabilityZone": "us-east-1c", - "InstanceId": "ip-172-31-27-227.eu-west-1.compute.internal", - "InstanceType": "t2.nano", - "InstanceMarketOption": "OnDemand", - }, - { - "AvailabilityZone": "us-east-1c", - "InstanceId": "ip-172-31-45-253.eu-west-1.compute.internal", - "InstanceType": "t2.nano", - "InstanceMarketOption": "OnDemand", - }, - ], - "Cause": "SCALE_IN", - } - - main(token, sample_event) diff --git a/tests/ci/terminate_runner_lambda/build_and_deploy_archive.sh b/tests/ci/terminate_runner_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/terminate_runner_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/terminate_runner_lambda/lambda_shared b/tests/ci/terminate_runner_lambda/lambda_shared deleted file mode 120000 index ba86e090f6c..00000000000 --- a/tests/ci/terminate_runner_lambda/lambda_shared +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package/lambda_shared \ No newline at end of file diff --git a/tests/ci/terminate_runner_lambda/requirements.txt b/tests/ci/terminate_runner_lambda/requirements.txt deleted file mode 100644 index 4cb3fba0f7b..00000000000 --- a/tests/ci/terminate_runner_lambda/requirements.txt +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package[token] diff --git a/tests/ci/test_ci_config.py b/tests/ci/test_ci_config.py index 47247b91858..4a2bd606d0e 100644 --- a/tests/ci/test_ci_config.py +++ b/tests/ci/test_ci_config.py @@ -1,6 +1,8 @@ #!/usr/bin/env python3 - +import copy import unittest +import random + from ci_config import CI import ci as CIPY from ci_settings import CiSettings @@ -57,6 +59,18 @@ class TestCIConfig(unittest.TestCase): f"Job [{job}] apparently uses wrong common config with job keyword [{CI.JOB_CONFIGS[job].job_name_keyword}]", ) + def test_job_config_has_proper_values(self): + for job in CI.JobNames: + if CI.JOB_CONFIGS[job].reference_job_name: + reference_job_config = CI.JOB_CONFIGS[ + CI.JOB_CONFIGS[job].reference_job_name + ] + # reference job must run in all workflows and has digest + self.assertTrue(reference_job_config.pr_only == False) + self.assertTrue(reference_job_config.release_only == False) + self.assertTrue(reference_job_config.run_always == False) + self.assertTrue(reference_job_config.digest != CI.DigestConfig()) + def test_required_checks(self): for job in CI.REQUIRED_CHECKS: if job in (CI.StatusNames.PR_CHECK, CI.StatusNames.SYNC): @@ -269,6 +283,7 @@ class TestCIConfig(unittest.TestCase): ci_cache = CIPY._configure_jobs( S3Helper(), pr_info, settings, skip_jobs=False, dry_run=True ) + ci_cache.filter_out_not_affected_jobs() actual_jobs_to_do = list(ci_cache.jobs_to_do) expected_jobs_to_do = [] for set_ in settings.ci_sets: @@ -280,6 +295,13 @@ class TestCIConfig(unittest.TestCase): continue expected_jobs_to_do.append(job) for job, config in CI.JOB_CONFIGS.items(): + if ( + CI.is_build_job(job) + and not config.run_by_label + and job not in expected_jobs_to_do + ): + # expected to run all builds jobs + expected_jobs_to_do.append(job) if not any( keyword in normalize_string(job) for keyword in settings.include_keywords @@ -397,10 +419,60 @@ class TestCIConfig(unittest.TestCase): ] self.assertCountEqual(expected_jobs_to_do, actual_jobs_to_do) + def test_ci_py_for_specific_workflow(self): + """ + checks ci.py job configuration + """ + settings = CiSettings() + settings.no_ci_cache = True + pr_info = PRInfo(github_event=_TEST_EVENT_JSON) + # make it merge_queue + pr_info.event_type = EventType.SCHEDULE + assert pr_info.number == 0 and not pr_info.is_merge_queue and not pr_info.is_pr + ci_cache = CIPY._configure_jobs( + S3Helper(), + pr_info, + settings, + skip_jobs=False, + dry_run=True, + workflow_name=CI.WorkFlowNames.JEPSEN, + ) + actual_jobs_to_do = list(ci_cache.jobs_to_do) + expected_jobs_to_do = [ + CI.BuildNames.BINARY_RELEASE, + CI.JobNames.JEPSEN_KEEPER, + CI.JobNames.JEPSEN_SERVER, + ] + self.assertCountEqual(expected_jobs_to_do, actual_jobs_to_do) + def test_ci_py_await(self): """ checks ci.py job configuration """ + + def _reset_ci_cache_to_wait_all_jobs(ci_cache): + # pretend there are pending jobs that we need to wait + ci_cache.jobs_to_wait = dict(ci_cache.jobs_to_do) + for job, config in ci_cache.jobs_to_wait.items(): + assert config.batches + config.pending_batches = list(config.batches) + + for batch in range(config.num_batches): + record = CiCache.Record( + record_type=CiCache.RecordType.PENDING, + job_name=job, + job_digest=ci_cache.job_digests[job], + batch=batch, + num_batches=config.num_batches, + release_branch=True, + ) + for record_t_, records_ in ci_cache.records.items(): + if record_t_.value == CiCache.RecordType.PENDING.value: + records_[record.to_str_key()] = record + assert not ci_cache.jobs_to_skip + assert ci_cache.jobs_to_wait + ci_cache.jobs_to_skip = [] + settings = CiSettings() settings.no_ci_cache = True pr_info = PRInfo(github_event=_TEST_EVENT_JSON) @@ -417,26 +489,6 @@ class TestCIConfig(unittest.TestCase): assert not ci_cache.jobs_to_skip assert not ci_cache.jobs_to_wait - # pretend there are pending jobs that we neet to wait - ci_cache.jobs_to_wait = dict(ci_cache.jobs_to_do) - for job, config in ci_cache.jobs_to_wait.items(): - assert not config.pending_batches - assert config.batches - config.pending_batches = list(config.batches) - for job, config in ci_cache.jobs_to_wait.items(): - for batch in range(config.num_batches): - record = CiCache.Record( - record_type=CiCache.RecordType.PENDING, - job_name=job, - job_digest=ci_cache.job_digests[job], - batch=batch, - num_batches=config.num_batches, - release_branch=True, - ) - for record_t_, records_ in ci_cache.records.items(): - if record_t_.value == CiCache.RecordType.PENDING.value: - records_[record.to_str_key()] = record - def _test_await_for_batch( ci_cache: CiCache, record_type: CiCache.RecordType, batch: int ) -> None: @@ -462,30 +514,212 @@ class TestCIConfig(unittest.TestCase): and batch < config_.num_batches ): assert batch not in config_.pending_batches - else: - assert batch in config_.pending_batches for _, config_ in ci_cache.jobs_to_do.items(): # jobs to do must have batches to run before/after await # if it's an empty list after await - apparently job has not been removed after await assert config_.batches - _test_await_for_batch(ci_cache, CiCache.RecordType.SUCCESSFUL, 0) - # check all one-batch jobs are in jobs_to_skip - for job in all_jobs_in_wf: - config = CI.JOB_CONFIGS[job] - if config.num_batches == 1: - self.assertTrue(job in ci_cache.jobs_to_skip) - self.assertTrue(job not in ci_cache.jobs_to_do) - else: - self.assertTrue(job not in ci_cache.jobs_to_skip) - self.assertTrue(job in ci_cache.jobs_to_do) - - _test_await_for_batch(ci_cache, CiCache.RecordType.FAILED, 1) - _test_await_for_batch(ci_cache, CiCache.RecordType.SUCCESSFUL, 2) - - self.assertTrue(len(ci_cache.jobs_to_skip) > 0) - self.assertTrue(len(ci_cache.jobs_to_do) > 0) + _reset_ci_cache_to_wait_all_jobs(ci_cache) + _test_await_for_batch(ci_cache, CiCache.RecordType.FAILED, 0) + tested = False + for job, config in ci_cache.jobs_to_do.items(): + if config.batches == [0]: + tested = True + self.assertTrue( + job not in ci_cache.jobs_to_wait, + "Job must be removed from @jobs_to_wait, because its only batch has FAILED cache record", + ) self.assertCountEqual( - list(ci_cache.jobs_to_do) + ci_cache.jobs_to_skip, all_jobs_in_wf + ci_cache.jobs_to_skip, + [], + "No jobs must be skipped, since all cache records are of type FAILED", ) + assert tested + + # reset jobs_to_wait after previous test + _reset_ci_cache_to_wait_all_jobs(ci_cache) + assert not ci_cache.jobs_to_skip + + # set batch 0 as SUCCESSFUL in ci cache + jobs_to_do_prev = list(ci_cache.jobs_to_do) + jobs_to_skip_prev = [] + jobs_to_wait_prev = list(ci_cache.jobs_to_wait) + _test_await_for_batch(ci_cache, CiCache.RecordType.SUCCESSFUL, 0) + self.assertTrue(len(jobs_to_skip_prev) != len(ci_cache.jobs_to_skip)) + self.assertTrue(len(jobs_to_wait_prev) > len(ci_cache.jobs_to_wait)) + self.assertCountEqual( + list(ci_cache.jobs_to_do) + ci_cache.jobs_to_skip, + jobs_to_do_prev + jobs_to_skip_prev, + ) + + # set batch 1 as SUCCESSFUL in ci cache + jobs_to_do_prev = list(ci_cache.jobs_to_do) + jobs_to_skip_prev = list(ci_cache.jobs_to_skip) + jobs_to_wait_prev = list(ci_cache.jobs_to_wait) + _test_await_for_batch(ci_cache, CiCache.RecordType.SUCCESSFUL, 1) + self.assertTrue(len(jobs_to_skip_prev) != len(ci_cache.jobs_to_skip)) + self.assertTrue(len(jobs_to_wait_prev) > len(ci_cache.jobs_to_wait)) + self.assertCountEqual( + list(ci_cache.jobs_to_do) + ci_cache.jobs_to_skip, + jobs_to_do_prev + jobs_to_skip_prev, + ) + + # set batch 3, 4, 5, 6 as SUCCESSFUL in ci cache + jobs_to_do_prev = list(ci_cache.jobs_to_do) + jobs_to_skip_prev = list(ci_cache.jobs_to_skip) + jobs_to_wait_prev = list(ci_cache.jobs_to_wait) + _test_await_for_batch(ci_cache, CiCache.RecordType.SUCCESSFUL, 2) + self.assertTrue(ci_cache.jobs_to_do) + _test_await_for_batch(ci_cache, CiCache.RecordType.SUCCESSFUL, 3) + self.assertTrue(ci_cache.jobs_to_do) + _test_await_for_batch(ci_cache, CiCache.RecordType.SUCCESSFUL, 4) + self.assertTrue(ci_cache.jobs_to_do) + _test_await_for_batch(ci_cache, CiCache.RecordType.SUCCESSFUL, 5) + self.assertTrue( + not ci_cache.jobs_to_do + ) # by this moment there must be no jobs left as batch 5 is currently the maximum + self.assertTrue(len(jobs_to_skip_prev) != len(ci_cache.jobs_to_skip)) + self.assertTrue(len(jobs_to_wait_prev) > len(ci_cache.jobs_to_wait)) + self.assertCountEqual( + list(ci_cache.jobs_to_do) + ci_cache.jobs_to_skip, + jobs_to_do_prev + jobs_to_skip_prev, + ) + + def test_ci_py_filters_not_affected_jobs_in_prs(self): + """ + checks ci.py filters not affected jobs in PRs + """ + settings = CiSettings() + settings.no_ci_cache = True + pr_info = PRInfo(github_event=_TEST_EVENT_JSON) + pr_info.event_type = EventType.PULL_REQUEST + pr_info.number = 123 + assert pr_info.is_pr + ci_cache = CIPY._configure_jobs( + S3Helper(), pr_info, settings, skip_jobs=False, dry_run=True + ) + self.assertTrue(not ci_cache.jobs_to_skip, "Must be no jobs in skip list") + assert not ci_cache.jobs_to_wait + assert not ci_cache.jobs_to_skip + + MOCK_AFFECTED_JOBS = [ + CI.JobNames.STATELESS_TEST_S3_DEBUG, + CI.JobNames.STRESS_TEST_TSAN, + ] + MOCK_REQUIRED_BUILDS = [] + + # pretend there are pending jobs that we need to wait + for job, job_config in ci_cache.jobs_to_do.items(): + if job in MOCK_AFFECTED_JOBS: + MOCK_REQUIRED_BUILDS += job_config.required_builds + elif job not in MOCK_AFFECTED_JOBS and not job_config.disable_await: + ci_cache.jobs_to_wait[job] = job_config + + for job, job_config in ci_cache.jobs_to_do.items(): + if job_config.reference_job_name or job_config.disable_await: + # jobs with reference_job_name in config are not supposed to have records in the cache - continue + continue + if job in MOCK_AFFECTED_JOBS: + continue + for batch in range(job_config.num_batches): + # add any record into cache + record = CiCache.Record( + record_type=random.choice( + [ + CiCache.RecordType.FAILED, + CiCache.RecordType.PENDING, + CiCache.RecordType.SUCCESSFUL, + ] + ), + job_name=job, + job_digest=ci_cache.job_digests[job], + batch=batch, + num_batches=job_config.num_batches, + release_branch=True, + ) + for record_t_, records_ in ci_cache.records.items(): + if record_t_.value == CiCache.RecordType.FAILED.value: + records_[record.to_str_key()] = record + + ci_cache.filter_out_not_affected_jobs() + expected_to_do = ( + [ + CI.JobNames.BUILD_CHECK, + ] + + MOCK_AFFECTED_JOBS + + MOCK_REQUIRED_BUILDS + ) + self.assertTrue( + CI.JobNames.BUILD_CHECK not in ci_cache.jobs_to_wait, + "We must never await on Builds Report", + ) + self.assertCountEqual( + list(ci_cache.jobs_to_wait), + MOCK_REQUIRED_BUILDS, + ) + self.assertCountEqual(list(ci_cache.jobs_to_do), expected_to_do) + + def test_ci_py_filters_not_affected_jobs_in_prs_no_builds(self): + """ + checks ci.py filters not affected jobs in PRs, no builds required + """ + settings = CiSettings() + settings.no_ci_cache = True + pr_info = PRInfo(github_event=_TEST_EVENT_JSON) + pr_info.event_type = EventType.PULL_REQUEST + pr_info.number = 123 + assert pr_info.is_pr + ci_cache = CIPY._configure_jobs( + S3Helper(), pr_info, settings, skip_jobs=False, dry_run=True + ) + self.assertTrue(not ci_cache.jobs_to_skip, "Must be no jobs in skip list") + assert not ci_cache.jobs_to_wait + assert not ci_cache.jobs_to_skip + + MOCK_AFFECTED_JOBS = [ + CI.JobNames.FAST_TEST, + ] + MOCK_REQUIRED_BUILDS = [] + + # pretend there are pending jobs that we need to wait + for job, job_config in ci_cache.jobs_to_do.items(): + if job in MOCK_AFFECTED_JOBS: + if job_config.required_builds: + MOCK_REQUIRED_BUILDS += job_config.required_builds + elif job not in MOCK_AFFECTED_JOBS and not job_config.disable_await: + ci_cache.jobs_to_wait[job] = job_config + + for job, job_config in ci_cache.jobs_to_do.items(): + if job_config.reference_job_name or job_config.disable_await: + # jobs with reference_job_name in config are not supposed to have records in the cache - continue + continue + if job in MOCK_AFFECTED_JOBS: + continue + for batch in range(job_config.num_batches): + # add any record into cache + record = CiCache.Record( + record_type=random.choice( + [ + CiCache.RecordType.FAILED, + CiCache.RecordType.PENDING, + CiCache.RecordType.SUCCESSFUL, + ] + ), + job_name=job, + job_digest=ci_cache.job_digests[job], + batch=batch, + num_batches=job_config.num_batches, + release_branch=True, + ) + for record_t_, records_ in ci_cache.records.items(): + if record_t_.value == CiCache.RecordType.FAILED.value: + records_[record.to_str_key()] = record + + ci_cache.filter_out_not_affected_jobs() + expected_to_do = MOCK_AFFECTED_JOBS + MOCK_REQUIRED_BUILDS + self.assertCountEqual( + list(ci_cache.jobs_to_wait), + MOCK_REQUIRED_BUILDS, + ) + self.assertCountEqual(list(ci_cache.jobs_to_do), expected_to_do) diff --git a/tests/ci/test_ci_options.py b/tests/ci/test_ci_options.py index 3f158e79f30..534885dda2b 100644 --- a/tests/ci/test_ci_options.py +++ b/tests/ci/test_ci_options.py @@ -8,8 +8,8 @@ from ci_config import CI _TEST_BODY_1 = """ #### Run only: -- [x] Non required -- [ ] Integration tests (arm64) +- [ ] Some Set +- [x] Integration tests (arm64) - [x] Integration tests - [x] Integration tests - [ ] Integration tests @@ -149,7 +149,7 @@ class TestCIOptions(unittest.TestCase): self.assertFalse(ci_options.no_ci_cache) self.assertTrue(ci_options.no_merge_commit) self.assertTrue(ci_options.woolen_wolfdog) - self.assertEqual(ci_options.ci_sets, ["ci_set_non_required"]) + self.assertEqual(ci_options.ci_sets, ["ci_set_arm"]) self.assertCountEqual(ci_options.include_keywords, ["foo", "foo_bar"]) self.assertCountEqual(ci_options.exclude_keywords, ["foo", "foo_bar"]) @@ -172,14 +172,10 @@ class TestCIOptions(unittest.TestCase): job: CI.JobConfig(runner_type=CI.Runners.STYLE_CHECKER) for job in _TEST_JOB_LIST } - jobs_configs[ - "fuzzers" - ].run_by_label = ( + jobs_configs["fuzzers"].run_by_label = ( "TEST_LABEL" # check "fuzzers" appears in the result due to the label ) - jobs_configs[ - "Integration tests (asan)" - ].release_only = ( + jobs_configs["Integration tests (asan)"].release_only = ( True # still must be included as it's set with include keywords ) filtered_jobs = list( @@ -201,6 +197,10 @@ class TestCIOptions(unittest.TestCase): "package_debug", "package_msan", "package_ubsan", + "package_aarch64", + "package_release_coverage", + "package_tsan", + "binary_release", "Stateless tests (asan)", "Stateless tests (azure, asan)", "Stateless tests flaky check (asan)", @@ -280,6 +280,7 @@ class TestCIOptions(unittest.TestCase): filtered_jobs, [ "Style check", + "fuzzers", ], ) @@ -295,9 +296,7 @@ class TestCIOptions(unittest.TestCase): ) self.assertCountEqual( filtered_jobs, - [ - "Style check", - ], + ["Style check", "fuzzers"], ) def test_options_applied_4(self): @@ -311,9 +310,9 @@ class TestCIOptions(unittest.TestCase): job: CI.JobConfig(runner_type=CI.Runners.STYLE_CHECKER) for job in _TEST_JOB_LIST } - jobs_configs[ - "fuzzers" - ].run_by_label = "TEST_LABEL" # check "fuzzers" does not appears in the result + jobs_configs["fuzzers"].run_by_label = ( + "TEST_LABEL" # check "fuzzers" does not appears in the result + ) jobs_configs["Integration tests (asan)"].release_only = True filtered_jobs = list( ci_options.apply( @@ -333,5 +332,12 @@ class TestCIOptions(unittest.TestCase): "Stateless tests (release, old analyzer, s3, DatabaseReplicated)", "package_asan", "fuzzers", + "package_aarch64", + "package_release_coverage", + "package_debug", + "package_tsan", + "package_msan", + "package_ubsan", + "binary_release", ], ) diff --git a/tests/ci/version_helper.py b/tests/ci/version_helper.py index 50263f6ebb6..07a7a9601c0 100755 --- a/tests/ci/version_helper.py +++ b/tests/ci/version_helper.py @@ -72,6 +72,19 @@ class ClickHouseVersion: return self.patch_update() raise KeyError(f"wrong part {part} is used") + def bump(self) -> "ClickHouseVersion": + if self.minor < 12: + self._minor += 1 + self._revision += 1 + self._patch = 1 + self._tweak = 1 + else: + self._major += 1 + self._revision += 1 + self._patch = 1 + self._tweak = 1 + return self + def major_update(self) -> "ClickHouseVersion": if self._git is not None: self._git.update() @@ -148,6 +161,11 @@ class ClickHouseVersion: """our X.3 and X.8 are LTS""" return self.minor % 5 == 3 + def get_stable_release_type(self) -> str: + if self.is_lts: + return VersionType.LTS + return VersionType.STABLE + def as_dict(self) -> VERSIONS: return { "revision": self.revision, @@ -168,6 +186,7 @@ class ClickHouseVersion: raise ValueError(f"version type {version_type} not in {VersionType.VALID}") self._description = version_type self._describe = f"v{self.string}-{version_type}" + return self def copy(self) -> "ClickHouseVersion": copy = ClickHouseVersion( diff --git a/tests/ci/worker/.gitignore b/tests/ci/worker/.gitignore deleted file mode 100644 index 4ed18989e78..00000000000 --- a/tests/ci/worker/.gitignore +++ /dev/null @@ -1 +0,0 @@ -generated_*init_runner.sh diff --git a/tests/ci/worker/deploy-runner-init.sh b/tests/ci/worker/deploy-runner-init.sh deleted file mode 100755 index 06edede48fa..00000000000 --- a/tests/ci/worker/deploy-runner-init.sh +++ /dev/null @@ -1,85 +0,0 @@ -#!/usr/bin/env bash - -usage() { - echo "Usage: $0 ENVIRONMENT" >&2 - echo "Valid values for ENVIRONMENT: staging, production" >&2 - exit 1 -} - -case "$1" in - staging|production) - ENVIRONMENT="$1" ;; - --help) - usage ;; - *) - echo "Invalid argument" >&2 - usage ;; -esac - -cd "$(dirname "$0")" || exit 1 -SOURCE_SCRIPT='init_runner.sh' - -check_response() { - # Are we even in the interactive shell? - [ -t 1 ] || return 1 - local request - request="$1" - read -rp "$request (y/N): " response - case "$response" in - [Yy]) - return 0 - # Your code to continue goes here - ;; - *) - return 1 - ;; - esac -} - -check_dirty() { - if [ -n "$(git status --porcelain=v2 "$SOURCE_SCRIPT")" ]; then - echo "The $SOURCE_SCRIPT has uncommited changes, won't deploy it" >&2 - exit 1 - fi -} -GIT_HASH=$(git log -1 --format=format:%H) - -header() { - cat << EOF -#!/usr/bin/env bash - -echo 'The $ENVIRONMENT script is generated from $SOURCE_SCRIPT, commit $GIT_HASH' - -EOF -} - -body() { - local first_line - first_line=$(sed -n '/^# THE SCRIPT START$/{=;q}' "$SOURCE_SCRIPT") - if [ -z "$first_line" ]; then - echo "The pattern '# THE SCRIPT START' is not found in $SOURCE_SCRIPT" >&2 - exit 1 - fi - tail "+$first_line" "$SOURCE_SCRIPT" -} - -GENERATED_FILE="generated_${ENVIRONMENT}_${SOURCE_SCRIPT}" - -{ header && body; } > "$GENERATED_FILE" - -echo "The file $GENERATED_FILE is generated" - -if check_response "Display the content of $GENERATED_FILE?"; then - if [ -z "$PAGER" ]; then - less "$GENERATED_FILE" - else - $PAGER "$GENERATED_FILE" - fi -fi - -check_dirty - -S3_OBJECT=${S3_OBJECT:-s3://github-runners-data/cloud-init/${ENVIRONMENT}.sh} -if check_response "Deploy the generated script to $S3_OBJECT?"; then - aws s3 mv "$GENERATED_FILE" "$S3_OBJECT" -fi diff --git a/tests/ci/worker/init_runner.sh b/tests/ci/worker/init_runner.sh deleted file mode 100644 index ec7793db2aa..00000000000 --- a/tests/ci/worker/init_runner.sh +++ /dev/null @@ -1,397 +0,0 @@ -#!/usr/bin/env bash - -cat > /dev/null << 'EOF' -The following content is embedded into the s3 object via the script -deploy-runner-init.sh {staging,production} -with additional helping information - -In the `user data` you should define as the following text -between `### COPY BELOW` and `### COPY ABOVE` - -### COPY BELOW -Content-Type: multipart/mixed; boundary="//" -MIME-Version: 1.0 - ---// -Content-Type: text/cloud-config; charset="us-ascii" -MIME-Version: 1.0 -Content-Transfer-Encoding: 7bit -Content-Disposition: attachment; filename="cloud-config.txt" - -#cloud-config -cloud_final_modules: -- [scripts-user, always] - ---// -Content-Type: text/x-shellscript; charset="us-ascii" -MIME-Version: 1.0 -Content-Transfer-Encoding: 7bit -Content-Disposition: attachment; filename="userdata.txt" - -#!/bin/bash -INSTANCE_ID=$(ec2metadata --instance-id) -INIT_ENVIRONMENT=$(/usr/local/bin/aws ec2 describe-tags --filters "Name=resource-id,Values=$INSTANCE_ID" --query "Tags[?Key=='github:init-environment'].Value" --output text) -echo "Downloading and using $INIT_ENVIRONMENT cloud-init.sh" -aws s3 cp "s3://github-runners-data/cloud-init/${INIT_ENVIRONMENT:-production}.sh" /tmp/cloud-init.sh -chmod 0700 /tmp/cloud-init.sh -exec bash /tmp/cloud-init.sh ---// -### COPY ABOVE -EOF - -# THE SCRIPT START - -set -uo pipefail - -#################################### -# IMPORTANT! # -# EC2 instance should have # -# `github:runner-type` tag # -# set accordingly to a runner role # -#################################### - -echo "Running init script" -export DEBIAN_FRONTEND=noninteractive -export RUNNER_HOME=/home/ubuntu/actions-runner - -export RUNNER_ORG="ClickHouse" -export RUNNER_URL="https://github.com/${RUNNER_ORG}" -# Funny fact, but metadata service has fixed IP -INSTANCE_ID=$(ec2metadata --instance-id) -export INSTANCE_ID - -bash /usr/local/share/scripts/init-network.sh - -# combine labels -RUNNER_TYPE=$(/usr/local/bin/aws ec2 describe-tags --filters "Name=resource-id,Values=$INSTANCE_ID" --query "Tags[?Key=='github:runner-type'].Value" --output text) -LABELS="self-hosted,Linux,$(uname -m),$RUNNER_TYPE" -export LABELS - -# Refresh CloudWatch agent config -aws ssm get-parameter --region us-east-1 --name AmazonCloudWatch-github-runners --query 'Parameter.Value' --output text > /opt/aws/amazon-cloudwatch-agent/etc/amazon-cloudwatch-agent.json -systemctl restart amazon-cloudwatch-agent.service - -# Refresh teams ssh keys -TEAM_KEYS_URL=$(aws ssm get-parameter --region us-east-1 --name team-keys-url --query 'Parameter.Value' --output=text) -curl -s "${TEAM_KEYS_URL}" > /home/ubuntu/.ssh/authorized_keys2 -chown ubuntu: /home/ubuntu/.ssh -R - - -# Create a pre-run script that will provide diagnostics info -mkdir -p /tmp/actions-hooks -cat > /tmp/actions-hooks/common.sh << 'EOF' -#!/bin/bash -EOF - -terminate_delayed() { - # The function for post hook to gracefully finish the job and then tear down - # The very specific sleep time is used later to determine in the main loop if - # the instance is tearing down - # IF `sleep` IS CHANGED, CHANGE ANOTHER VALUE IN `pgrep` - sleep=13.14159265358979323846 - echo "Going to terminate the runner's instance in $sleep seconds" - INSTANCE_ID=$(ec2metadata --instance-id) - # We execute it with `at` to not have it as an orphan process, but launched independently - # GH Runners kill all remain processes - echo "sleep '$sleep'; aws ec2 terminate-instances --instance-ids $INSTANCE_ID" | at now || \ - aws ec2 terminate-instances --instance-ids "$INSTANCE_ID" # workaround for complete out of space or non-installed `at` - exit 0 -} - -detect_delayed_termination() { - # The function look for very specific sleep with pi - if pgrep 'sleep 13.14159265358979323846'; then - echo 'The instance has delayed termination, sleep the same time to wait if it goes down' - sleep 14 - fi -} - -declare -f terminate_delayed >> /tmp/actions-hooks/common.sh - -terminate_and_exit() { - # Terminate instance and exit from the script instantly - echo "Going to terminate the runner's instance" - INSTANCE_ID=$(ec2metadata --instance-id) - aws ec2 terminate-instances --instance-ids "$INSTANCE_ID" - exit 0 -} - -declare -f terminate_and_exit >> /tmp/actions-hooks/common.sh - -check_spot_instance_is_old() { - # This function should be executed ONLY BETWEEN runnings. - # It's unsafe to execute while the runner is working! - local LIFE_CYCLE - LIFE_CYCLE=$(curl -s --fail http://169.254.169.254/latest/meta-data/instance-life-cycle) - if [ "$LIFE_CYCLE" == "spot" ]; then - local UPTIME - UPTIME=$(< /proc/uptime) - UPTIME=${UPTIME%%.*} - if (( 3600 < UPTIME )); then - echo "The spot instance has uptime $UPTIME, it's time to shut it down" - return 0 - fi - fi - return 1 -} - -check_proceed_spot_termination() { - # The function checks and proceeds spot instance termination if exists - # The event for spot instance termination - local FORCE - FORCE=${1:-} - if TERMINATION_DATA=$(curl -s --fail http://169.254.169.254/latest/meta-data/spot/instance-action); then - # https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-instance-termination-notices.html#instance-action-metadata - _action=$(jq '.action' -r <<< "$TERMINATION_DATA") - _time=$(jq '.time | fromdate' <<< "$TERMINATION_DATA") - _until_action=$((_time - $(date +%s))) - echo "Received the '$_action' event that will be effective in $_until_action seconds" - if (( _until_action <= 30 )) || [ "$FORCE" == "force" ]; then - echo "The action $_action will be done in $_until_action, killing the runner and exit" - local runner_pid - runner_pid=$(pgrep Runner.Listener) - if [ -n "$runner_pid" ]; then - # Kill the runner to not allow it cancelling the job - # shellcheck disable=SC2046 - kill -9 "$runner_pid" $(list_children "$runner_pid") - fi - sudo -u ubuntu ./config.sh remove --token "$(get_runner_token)" - terminate_and_exit - fi - fi -} - -no_terminating_metadata() { - # The function check that instance could continue work - # Returns 1 if any of termination events are received - - # The event for rebalance recommendation. Not strict, so we have some room to make a decision here - if curl -s --fail http://169.254.169.254/latest/meta-data/events/recommendations/rebalance; then - echo 'Received recommendation to rebalance, checking the uptime' - local UPTIME - UPTIME=$(< /proc/uptime) - UPTIME=${UPTIME%%.*} - # We don't shutdown the instances younger than 30m - if (( 1800 < UPTIME )); then - # To not shutdown everything at once, use the 66% to survive - if (( $((RANDOM % 3)) == 0 )); then - echo 'The instance is older than 30m and won the roulette' - return 1 - fi - echo 'The instance is older than 30m, but is not chosen for rebalance' - else - echo 'The instance is younger than 30m, do not shut it down' - fi - fi - - # Checks if the ASG in a lifecycle hook state - local ASG_STATUS - ASG_STATUS=$(curl -s http://169.254.169.254/latest/meta-data/autoscaling/target-lifecycle-state) - if [ "$ASG_STATUS" == "Terminated" ]; then - echo 'The instance in ASG status Terminating:Wait' - return 1 - fi -} - -terminate_on_event() { - # If there is a rebalance event, then the instance could die soon - # Let's don't wait for it and terminate proactively - if curl -s --fail http://169.254.169.254/latest/meta-data/events/recommendations/rebalance; then - terminate_and_exit - fi - - # Here we check if the autoscaling group marked the instance for termination, and it's wait for the job to finish - ASG_STATUS=$(curl -s http://169.254.169.254/latest/meta-data/autoscaling/target-lifecycle-state) - if [ "$ASG_STATUS" == "Terminated" ]; then - INSTANCE_ID=$(ec2metadata --instance-id) - ASG_NAME=$(aws ec2 describe-tags --filters "Name=resource-id,Values=$INSTANCE_ID" --query "Tags[?Key=='aws:autoscaling:groupName'].Value" --output text) - LIFECYCLE_HOOKS=$(aws autoscaling describe-lifecycle-hooks --auto-scaling-group-name "$ASG_NAME" --query "LifecycleHooks[].LifecycleHookName" --output text) - for LCH in $LIFECYCLE_HOOKS; do - aws autoscaling complete-lifecycle-action --lifecycle-action-result CONTINUE \ - --lifecycle-hook-name "$LCH" --auto-scaling-group-name "$ASG_NAME" \ - --instance-id "$INSTANCE_ID" - true # autoformat issue - done - echo 'The runner is marked as "Terminated" by the autoscaling group, we are terminating' - terminate_and_exit - fi -} - -cat > /tmp/actions-hooks/pre-run.sh << EOF -#!/bin/bash -set -uo pipefail - -echo "Runner's public DNS: $(ec2metadata --public-hostname)" -echo "Runner's labels: ${LABELS}" -echo "Runner's instance type: $(ec2metadata --instance-type)" -EOF - -# Create a post-run script that will restart docker daemon before the job started -cat > /tmp/actions-hooks/post-run.sh << 'EOF' -#!/bin/bash -set -xuo pipefail - -source /tmp/actions-hooks/common.sh - -# Free KiB, free percents -ROOT_STAT=($(df / | awk '/\// {print $4 " " int($4/$2 * 100)}')) -if [[ ${ROOT_STAT[0]} -lt 3000000 ]] || [[ ${ROOT_STAT[1]} -lt 5 ]]; then - echo "The runner has ${ROOT_STAT[0]}KiB and ${ROOT_STAT[1]}% of free space on /" - terminate_delayed -fi - -# shellcheck disable=SC2046 -docker ps --quiet | xargs --no-run-if-empty docker kill ||: -# shellcheck disable=SC2046 -docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||: - -# If we have hanged containers after the previous commands, than we have a hanged one -# and should restart the daemon -if [ "$(docker ps --all --quiet)" ]; then - # Systemd service of docker has StartLimitBurst=3 and StartLimitInterval=60s, - # that's why we try restarting it for long - for i in {1..25}; - do - sudo systemctl restart docker && break || sleep 5 - done - - for i in {1..10} - do - docker info && break || sleep 2 - done - # Last chance, otherwise we have to terminate poor instance - docker info 1>/dev/null || { echo Docker unable to start; terminate_delayed ; } -fi -EOF - -get_runner_token() { - /usr/local/bin/aws ssm get-parameter --name github_runner_registration_token --with-decryption --output text --query Parameter.Value -} - -is_job_assigned() { - local runner_pid - runner_pid=$(pgrep Runner.Listener) - if [ -z "$runner_pid" ]; then - # if runner has finished, it's fine - return 0 - fi - local log_file - log_file=$(lsof -p "$runner_pid" 2>/dev/null | grep -o "$RUNNER_HOME/_diag/Runner.*log") - if [ -z "$log_file" ]; then - # assume, the process is over or just started - return 0 - fi - # So far it's the only solid way to determine that the job is starting - grep -q 'Terminal] .* Running job:' "$log_file" \ - && return 0 \ - || return 1 -} - -list_children () { - local children - children=$(ps --ppid "$1" -o pid=) - if [ -z "$children" ]; then - return - fi - - for pid in $children; do - list_children "$pid" - done - echo "$children" -} - -# There's possibility that it fails because the runner's version is outdated, -# so after the first failure we'll try to launch it with enabled autoupdate. -# -# We'll fail and terminate after 10 consequent failures. -ATTEMPT=0 -# In `kill` 0 means "all processes in process group", -1 is "all but PID 1" -# We use `-2` to get an error -RUNNER_PID=-2 - -while true; do - # Does not send signal, but checks that the process $RUNNER_PID is running - if kill -0 -- $RUNNER_PID; then - ATTEMPT=0 - echo "Runner is working with pid $RUNNER_PID, checking the metadata in background" - check_proceed_spot_termination - - if ! is_job_assigned; then - RUNNER_AGE=$(( $(date +%s) - $(stat -c +%Y /proc/"$RUNNER_PID" 2>/dev/null || date +%s) )) - echo "The runner is launched $RUNNER_AGE seconds ago and still hasn't received a job" - if (( 60 < RUNNER_AGE )); then - echo "Attempt to delete the runner for a graceful shutdown" - sudo -u ubuntu ./config.sh remove --token "$(get_runner_token)" \ - || continue - echo "Runner didn't launch or have assigned jobs after ${RUNNER_AGE} seconds, shutting down" - terminate_and_exit - fi - fi - else - if [ "$RUNNER_PID" != "-2" ]; then - wait $RUNNER_PID \ - && echo "Runner with PID $RUNNER_PID successfully finished" \ - || echo "Attempt $((++ATTEMPT)) to start the runner" - fi - if (( ATTEMPT > 10 )); then - echo "The runner has failed to start after $ATTEMPT attempt. Give up and terminate it" - terminate_and_exit - fi - - cd $RUNNER_HOME || terminate_and_exit - detect_delayed_termination - # If runner is not active, check that it needs to terminate itself - echo "Checking if the instance suppose to terminate" - no_terminating_metadata || terminate_on_event - check_spot_instance_is_old && terminate_and_exit - check_proceed_spot_termination force - - echo "Going to configure runner" - token_args=(--token "$(get_runner_token)") - config_args=( - "${token_args[@]}" --url "$RUNNER_URL" - --ephemeral --unattended --replace --runnergroup Default - --labels "$LABELS" --work _work --name "$INSTANCE_ID" - ) - if (( ATTEMPT > 1 )); then - echo 'The runner failed to start at least once. Removing it and then configuring with autoupdate enabled.' - sudo -u ubuntu ./config.sh remove "${token_args[@]}" - sudo -u ubuntu ./config.sh "${config_args[@]}" - else - echo "Configure runner with disabled autoupdate" - config_args+=("--disableupdate") - sudo -u ubuntu ./config.sh "${config_args[@]}" - fi - - echo "Another one check to avoid race between runner and infrastructure" - no_terminating_metadata || terminate_on_event - check_spot_instance_is_old && terminate_and_exit - check_proceed_spot_termination force - - # There were some failures to start the Job because of trash in _work - rm -rf _work - - # https://github.com/actions/runner/issues/3266 - # We're unable to know if the runner is failed to start. - echo 'Monkey-patching run helpers to get genuine exit code of the runner' - for script in run.sh run-helper.sh.template; do - # shellcheck disable=SC2016 - grep -q 'exit 0$' "$script" && \ - sed 's/exit 0/exit $returnCode/' -i "$script" && \ - echo "Script $script is patched" - done - - echo "Run" - sudo -u ubuntu \ - ACTIONS_RUNNER_HOOK_JOB_STARTED=/tmp/actions-hooks/pre-run.sh \ - ACTIONS_RUNNER_HOOK_JOB_COMPLETED=/tmp/actions-hooks/post-run.sh \ - ./run.sh & - RUNNER_PID=$! - - sleep 10 - fi - - sleep 5 -done - -# vim:ts=4:sw=4 diff --git a/tests/ci/worker/prepare-ci-ami.sh b/tests/ci/worker/prepare-ci-ami.sh index 3e2f33c89d1..eb410ddcb00 100644 --- a/tests/ci/worker/prepare-ci-ami.sh +++ b/tests/ci/worker/prepare-ci-ami.sh @@ -9,7 +9,7 @@ set -xeuo pipefail echo "Running prepare script" export DEBIAN_FRONTEND=noninteractive -export RUNNER_VERSION=2.316.1 +export RUNNER_VERSION=2.317.0 export RUNNER_HOME=/home/ubuntu/actions-runner deb_arch() { @@ -54,7 +54,8 @@ apt-get install --yes --no-install-recommends \ python3-dev \ python3-pip \ qemu-user-static \ - unzip + unzip \ + gh # Install docker curl -fsSL https://download.docker.com/linux/ubuntu/gpg | gpg --dearmor -o /usr/share/keyrings/docker-archive-keyring.gpg @@ -101,7 +102,7 @@ sudo -u ubuntu docker buildx version sudo -u ubuntu docker buildx rm default-builder || : # if it's the second attempt sudo -u ubuntu docker buildx create --use --name default-builder -pip install boto3 pygithub requests urllib3 unidiff dohq-artifactory +pip install boto3 pygithub requests urllib3 unidiff dohq-artifactory jwt rm -rf $RUNNER_HOME # if it's the second attempt mkdir -p $RUNNER_HOME && cd $RUNNER_HOME @@ -212,9 +213,9 @@ chmod +x /usr/local/share/scripts/init-network.sh touch /var/tmp/clickhouse-ci-ami.success # END OF THE SCRIPT -# TOE description +# TOE (Task Orchestrator and Executor) description # name: CIInfrastructurePrepare -# description: instals the infrastructure for ClickHouse CI runners +# description: installs the infrastructure for ClickHouse CI runners # schemaVersion: 1.0 # # phases: diff --git a/tests/ci/workflow_approve_rerun_lambda/app.py b/tests/ci/workflow_approve_rerun_lambda/app.py deleted file mode 100644 index 8ed47a54ab1..00000000000 --- a/tests/ci/workflow_approve_rerun_lambda/app.py +++ /dev/null @@ -1,404 +0,0 @@ -#!/usr/bin/env python3 - -import fnmatch -import json -import time -from collections import namedtuple -from urllib.parse import quote - -import requests -from lambda_shared.pr import TRUSTED_CONTRIBUTORS -from lambda_shared.token import get_cached_access_token - -SUSPICIOUS_CHANGED_FILES_NUMBER = 200 - -SUSPICIOUS_PATTERNS = [ - ".github/*", - "docker/*", - "docs/tools/*", - "packages/*", - "tests/ci/*", -] - -# Number of retries for API calls. -MAX_RETRY = 5 - -# Number of times a check can re-run as a whole. -# It is needed, because we are using AWS "spot" instances, that are terminated often -MAX_WORKFLOW_RERUN = 30 - -WorkflowDescription = namedtuple( - "WorkflowDescription", - [ - "name", - "action", - "run_id", - "event", - "workflow_id", - "conclusion", - "status", - "api_url", - "fork_owner_login", - "fork_branch", - "rerun_url", - "jobs_url", - "attempt", - "repo_url", - "url", - ], -) - -# See https://api.github.com/orgs/{name} -TRUSTED_ORG_IDS = { - 54801242, # clickhouse -} - -# See https://api.github.com/repos/ClickHouse/ClickHouse/actions/workflows -# Use ID to not inject a malicious workflow -TRUSTED_WORKFLOW_IDS = { - 14586616, # Cancel workflows, always trusted -} - -NEED_RERUN_WORKFLOWS = { - "BackportPR", - "DocsCheck", - "MasterCI", - "NightlyBuilds", - "PublishedReleaseCI", - "PullRequestCI", - "ReleaseBranchCI", -} - - -def is_trusted_contributor(pr_user_login, pr_user_orgs): - if pr_user_login.lower() in TRUSTED_CONTRIBUTORS: - print(f"User '{pr_user_login}' is trusted") - return True - - print(f"User '{pr_user_login}' is not trusted") - - for org_id in pr_user_orgs: - if org_id in TRUSTED_ORG_IDS: - print( - f"Org '{org_id}' is trusted; will mark user {pr_user_login} as trusted" - ) - return True - print(f"Org '{org_id}' is not trusted") - - return False - - -def _exec_get_with_retry(url, token): - headers = {"Authorization": f"token {token}"} - e = Exception() - for i in range(MAX_RETRY): - try: - response = requests.get(url, headers=headers, timeout=30) - response.raise_for_status() - return response.json() - except Exception as ex: - print("Got exception executing request", ex) - e = ex - time.sleep(i + 1) - - raise requests.HTTPError("Cannot execute GET request with retries") from e - - -def _exec_post_with_retry(url, token, data=None): - headers = {"Authorization": f"token {token}"} - e = Exception() - for i in range(MAX_RETRY): - try: - if data: - response = requests.post(url, headers=headers, json=data, timeout=30) - else: - response = requests.post(url, headers=headers, timeout=30) - if response.status_code == 403: - data = response.json() - if ( - "message" in data - and data["message"] - == "This workflow run is not waiting for approval" - ): - print("Workflow doesn't need approval") - return data - response.raise_for_status() - return response.json() - except Exception as ex: - print("Got exception executing request", ex) - e = ex - time.sleep(i + 1) - - raise requests.HTTPError("Cannot execute POST request with retry") from e - - -def _get_pull_requests_from(repo_url, owner, branch, token): - url = f"{repo_url}/pulls?head={quote(owner)}:{quote(branch)}" - return _exec_get_with_retry(url, token) - - -def get_workflow_description_from_event(event): - action = event["action"] - run_id = event["workflow_run"]["id"] - event_type = event["workflow_run"]["event"] - fork_owner = event["workflow_run"]["head_repository"]["owner"]["login"] - fork_branch = event["workflow_run"]["head_branch"] - name = event["workflow_run"]["name"] - workflow_id = event["workflow_run"]["workflow_id"] - conclusion = event["workflow_run"]["conclusion"] - attempt = event["workflow_run"]["run_attempt"] - status = event["workflow_run"]["status"] - jobs_url = event["workflow_run"]["jobs_url"] - rerun_url = event["workflow_run"]["rerun_url"] - url = event["workflow_run"]["html_url"] - api_url = event["workflow_run"]["url"] - repo_url = event["repository"]["url"] - return WorkflowDescription( - name=name, - action=action, - run_id=run_id, - event=event_type, - fork_owner_login=fork_owner, - fork_branch=fork_branch, - workflow_id=workflow_id, - conclusion=conclusion, - attempt=attempt, - status=status, - jobs_url=jobs_url, - rerun_url=rerun_url, - url=url, - repo_url=repo_url, - api_url=api_url, - ) - - -def get_pr_author_and_orgs(pull_request, token): - author = pull_request["user"]["login"] - orgs = _exec_get_with_retry(pull_request["user"]["organizations_url"], token) - return author, [org["id"] for org in orgs] - - -def get_changed_files_for_pull_request(pull_request, token): - url = pull_request["url"] - - changed_files = set([]) - for i in range(1, 31): - print("Requesting changed files page", i) - data = _exec_get_with_retry(f"{url}/files?page={i}&per_page=100", token) - print(f"Got {len(data)} changed files") - if len(data) == 0: - print("No more changed files") - break - - for change in data: - # print("Adding changed file", change['filename']) - changed_files.add(change["filename"]) - - if len(changed_files) >= SUSPICIOUS_CHANGED_FILES_NUMBER: - print( - f"More than {len(changed_files)} changed files. " - "Will stop fetching new files." - ) - break - - return changed_files - - -def check_suspicious_changed_files(changed_files): - if len(changed_files) >= SUSPICIOUS_CHANGED_FILES_NUMBER: - print(f"Too many files changed {len(changed_files)}, need manual approve") - return True - - for path in changed_files: - for pattern in SUSPICIOUS_PATTERNS: - if fnmatch.fnmatch(path, pattern): - print( - f"File {path} match suspicious pattern {pattern}, " - "will not approve automatically" - ) - return True - - print("No changed files match suspicious patterns, run could be approved") - return False - - -def approve_run(workflow_description: WorkflowDescription, token: str) -> None: - print("Approving run") - url = f"{workflow_description.api_url}/approve" - _exec_post_with_retry(url, token) - - -def label_manual_approve(pull_request, token): - url = f"{pull_request['issue_url']}/labels" - data = {"labels": ["manual approve"]} - - _exec_post_with_retry(url, token, data) - - -def get_workflow_jobs(workflow_description, token): - jobs_url = ( - workflow_description.api_url + f"/attempts/{workflow_description.attempt}/jobs" - ) - jobs = [] - i = 1 - while True: - got_jobs = _exec_get_with_retry(jobs_url + f"?page={i}", token) - if len(got_jobs["jobs"]) == 0: - break - - jobs += got_jobs["jobs"] - i += 1 - - return jobs - - -def check_need_to_rerun(workflow_description, token): - if workflow_description.attempt >= MAX_WORKFLOW_RERUN: - print( - "Not going to rerun workflow because it's already tried more than two times" - ) - return False - print("Going to check jobs") - - jobs = get_workflow_jobs(workflow_description, token) - print("Got jobs", len(jobs)) - for job in jobs: - print(f"Job {job['name']} has a conclusion '{job['conclusion']}'") - if job["conclusion"] not in ("success", "skipped"): - print("Job", job["name"], "failed, checking steps") - for step in job["steps"]: - # always the last job - if step["name"] == "Complete job": - print("Found Complete job step for job", job["name"]) - break - else: - print( - "Checked all steps and doesn't found Complete job, going to rerun" - ) - return True - - return False - - -def rerun_workflow(workflow_description, token): - print("Going to rerun workflow") - try: - _exec_post_with_retry(f"{workflow_description.rerun_url}-failed-jobs", token) - except Exception: - _exec_post_with_retry(workflow_description.rerun_url, token) - - -def check_workflow_completed( - event_data: dict, workflow_description: WorkflowDescription, token: str -) -> bool: - if workflow_description.action == "completed": - attempt = 0 - # Nice and reliable GH API sends from time to time such events, e.g: - # action='completed', conclusion=None, status='in_progress', - # So let's try receiving a real workflow data - while workflow_description.conclusion is None and attempt < MAX_RETRY: - progressive_sleep = 3 * sum(i + 1 for i in range(attempt)) - time.sleep(progressive_sleep) - event_data["workflow_run"] = _exec_get_with_retry( - workflow_description.api_url, token - ) - workflow_description = get_workflow_description_from_event(event_data) - attempt += 1 - - if workflow_description.conclusion != "failure": - print( - "Workflow finished with status " - f"{workflow_description.conclusion}, exiting" - ) - return True - - print( - "Workflow", - workflow_description.url, - "completed and failed, let's check for rerun", - ) - - if workflow_description.name not in NEED_RERUN_WORKFLOWS: - print( - "Workflow", - workflow_description.name, - "not in list of rerunable workflows", - ) - return True - - if check_need_to_rerun(workflow_description, token): - rerun_workflow(workflow_description, token) - return True - - return False - - -def main(event): - token = get_cached_access_token() - event_data = json.loads(event["body"]) - print("The body received:", event["body"]) - workflow_description = get_workflow_description_from_event(event_data) - - print("Got workflow description", workflow_description) - if check_workflow_completed(event_data, workflow_description, token): - return - - if workflow_description.action != "requested": - print("Exiting, event action is", workflow_description.action) - return - - if workflow_description.workflow_id in TRUSTED_WORKFLOW_IDS: - print("Workflow in trusted list, approving run") - approve_run(workflow_description, token) - return - - pull_requests = _get_pull_requests_from( - workflow_description.repo_url, - workflow_description.fork_owner_login, - workflow_description.fork_branch, - token, - ) - - print("Got pull requests for workflow", len(pull_requests)) - if len(pull_requests) != 1: - print(f"Can't continue with non-uniq PRs: {pull_requests}") - return - - pull_request = pull_requests[0] - print("Pull request for workflow number", pull_request["number"]) - - author, author_orgs = get_pr_author_and_orgs(pull_request, token) - if is_trusted_contributor(author, author_orgs): - print("Contributor is trusted, approving run") - approve_run(workflow_description, token) - return - - labels = {label["name"] for label in pull_request["labels"]} - if "can be tested" not in labels: - print("Label 'can be tested' is required for untrusted users") - return - - changed_files = get_changed_files_for_pull_request(pull_request, token) - print(f"Totally have {len(changed_files)} changed files in PR:", changed_files) - if check_suspicious_changed_files(changed_files): - print(f"Pull Request {pull_request['number']} has suspicious changes") - if "manual approve" not in labels: - print("Label the PR as needed for manuall approve") - label_manual_approve(pull_request, token) - else: - print(f"Pull Request {pull_request['number']} has no suspicious changes") - approve_run(workflow_description, token) - - -def handler(event, _): - try: - main(event) - - return { - "statusCode": 200, - "headers": {"Content-Type": "application/json"}, - "body": '{"status": "OK"}', - } - except Exception: - print("Received event: ", event) - raise diff --git a/tests/ci/workflow_approve_rerun_lambda/build_and_deploy_archive.sh b/tests/ci/workflow_approve_rerun_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/workflow_approve_rerun_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/workflow_approve_rerun_lambda/lambda_shared b/tests/ci/workflow_approve_rerun_lambda/lambda_shared deleted file mode 120000 index ba86e090f6c..00000000000 --- a/tests/ci/workflow_approve_rerun_lambda/lambda_shared +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package/lambda_shared \ No newline at end of file diff --git a/tests/ci/workflow_approve_rerun_lambda/requirements.txt b/tests/ci/workflow_approve_rerun_lambda/requirements.txt deleted file mode 100644 index 4cb3fba0f7b..00000000000 --- a/tests/ci/workflow_approve_rerun_lambda/requirements.txt +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package[token] diff --git a/tests/ci/workflow_jobs_lambda/app.py b/tests/ci/workflow_jobs_lambda/app.py deleted file mode 100644 index 4210ca22da9..00000000000 --- a/tests/ci/workflow_jobs_lambda/app.py +++ /dev/null @@ -1,202 +0,0 @@ -#!/usr/bin/env python - -""" -Lambda gets the workflow_job events, see -https://docs.github.com/en/developers/webhooks-and-events/webhooks/webhook-events-and-payloads#workflow_job - -Then it either posts it as is to the play.clickhouse.com, or anonymizes the sensitive -fields for private repositories -""" - -import json -import logging -from base64 import b64decode -from dataclasses import dataclass -from typing import Any, List, Optional - -from lambda_shared import ClickHouseHelper, InsertException, get_parameter_from_ssm - -logging.getLogger().setLevel(logging.INFO) - - -@dataclass -class WorkflowJob: - id: int - run_id: int - workflow_name: str - head_branch: str - run_url: str - run_attempt: int - node_id: str - head_sha: str - url: str - html_url: str - status: str - conclusion: str - started_at: str - completed_at: str - name: str - steps: int # just number of steps, we don't keep steps - check_run_url: str - labels: List[str] - runner_id: int - runner_name: str - runner_group_id: int - runner_group_name: str - repository: str - - def anonimyze_url(self, url: str) -> str: - return url.replace(self.repository, "ANONYMIZED_REPO") - - def anonimyze(self): - anm = "ANONYMIZED" - self.workflow_name = anm - self.head_branch = anm - self.run_url = self.anonimyze_url(self.run_url) - self.node_id = anm - self.url = self.anonimyze_url(self.url) - self.html_url = self.anonimyze_url(self.html_url) - self.name = anm - self.check_run_url = self.anonimyze_url(self.check_run_url) - self.repository = anm - - def as_dict(self) -> dict: - return self.__dict__ - - -CH_CLIENT = None # type: Optional[ClickHouseHelper] - - -def send_event_workflow_job(workflow_job: WorkflowJob) -> None: - # # SHOW CREATE TABLE default.workflow_jobs - # CREATE TABLE default.workflow_jobs UUID 'c0351924-8ccd-47a6-9db0-e28a9eee2fdf' - # ( - # `id` UInt64, - # `run_id` UInt64, - # `workflow_name` LowCardinality(String), - # `head_branch` LowCardinality(String), - # `run_url` String, - # `run_attempt` UInt16, - # `node_id` String, - # `head_sha` String, - # `url` String, - # `html_url` String, - # `status` Enum8('waiting' = 1, 'queued' = 2, 'in_progress' = 3, 'completed' = 4), - # `conclusion` LowCardinality(String), - # `started_at` DateTime, - # `completed_at` DateTime, - # `name` LowCardinality(String), - # `steps` UInt16, - # `check_run_url` String, - # `labels` Array(LowCardinality(String)), - # `runner_id` UInt64, - # `runner_name` String, - # `runner_group_id` UInt64, - # `runner_group_name` LowCardinality(String), - # `repository` LowCardinality(String), - # `updated_at` DateTime DEFAULT now() - # ) - # ENGINE = ReplicatedMergeTree('/clickhouse/tables/c0351924-8ccd-47a6-9db0-e28a9eee2fdf/{shard}', '{replica}') - # PARTITION BY toStartOfMonth(started_at) - # ORDER BY (id, updated_at) - # SETTINGS index_granularity = 8192 - global CH_CLIENT - CH_CLIENT = CH_CLIENT or ClickHouseHelper( - get_parameter_from_ssm("clickhouse-test-stat-url"), - get_parameter_from_ssm("clickhouse-test-stat-login"), - get_parameter_from_ssm("clickhouse-test-stat-password"), - ) - try: - CH_CLIENT.insert_event_into( - "default", "workflow_jobs", workflow_job.as_dict(), False - ) - except InsertException as ex: - logging.exception( - "Got an exception on insert, tryuing to update the client " - "credentials and repeat", - exc_info=ex, - ) - CH_CLIENT = ClickHouseHelper( - get_parameter_from_ssm("clickhouse-test-stat-url"), - get_parameter_from_ssm("clickhouse-test-stat-login"), - get_parameter_from_ssm("clickhouse-test-stat-password"), - ) - CH_CLIENT.insert_event_into( - "default", "workflow_jobs", workflow_job.as_dict(), False - ) - - -def killed_job(wf_job: dict) -> bool: - """a hack to identify the killed runner if "Complete job" is omit""" - if ( - wf_job.get("status", "") != "completed" - or wf_job.get("conclusion", "") != "failure" - ): - # The task either success or in progress - return False - return not any( - step["name"] == "Complete job" and step["conclusion"] is not None - for step in wf_job["steps"] - ) - - -def handler(event: dict, context: Any) -> dict: - if event["isBase64Encoded"]: - event_data = json.loads(b64decode(event["body"])) - else: - event_data = json.loads(event["body"]) - - logging.info("Got the next raw event from the github hook: %s", event_data) - repo = event_data["repository"] - try: - wf_job = event_data["workflow_job"] - except KeyError: - logging.error("The event does not contain valid workflow_jobs data") - logging.error("The event data: %s", event) - logging.error("The context data: %s", context) - - if killed_job(wf_job): - # for killed job we record 0 - steps = 0 - else: - # We record only finished steps - steps = sum(1 for st in wf_job["steps"] if st["conclusion"] is not None) - - workflow_job = WorkflowJob( - wf_job["id"], - wf_job["run_id"], - wf_job["workflow_name"] or "", # nullable - wf_job["head_branch"], - wf_job["run_url"], - wf_job["run_attempt"], - wf_job["node_id"], - wf_job["head_sha"], - wf_job["url"], - wf_job["html_url"], - wf_job["status"], - wf_job["conclusion"] or "", # nullable - wf_job["started_at"], - wf_job["completed_at"] or "1970-01-01T00:00:00", # nullable date - wf_job["name"], - steps, - wf_job["check_run_url"], - wf_job["labels"], - wf_job["runner_id"] or 0, # nullable - wf_job["runner_name"] or "", # nullable - wf_job["runner_group_id"] or 0, # nullable - wf_job["runner_group_name"] or "", # nullable - repo["full_name"], - ) - logging.info( - "Got the next event (private_repo=%s): %s", repo["private"], workflow_job - ) - if repo["private"]: - workflow_job.anonimyze() - - send_event_workflow_job(workflow_job) - - return { - "statusCode": 200, - "headers": {"Content-Type": "application/json"}, - "body": '{"status": "OK"}', - } diff --git a/tests/ci/workflow_jobs_lambda/build_and_deploy_archive.sh b/tests/ci/workflow_jobs_lambda/build_and_deploy_archive.sh deleted file mode 120000 index 96ba3fa024e..00000000000 --- a/tests/ci/workflow_jobs_lambda/build_and_deploy_archive.sh +++ /dev/null @@ -1 +0,0 @@ -../team_keys_lambda/build_and_deploy_archive.sh \ No newline at end of file diff --git a/tests/ci/workflow_jobs_lambda/lambda_shared b/tests/ci/workflow_jobs_lambda/lambda_shared deleted file mode 120000 index ba86e090f6c..00000000000 --- a/tests/ci/workflow_jobs_lambda/lambda_shared +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package/lambda_shared \ No newline at end of file diff --git a/tests/ci/workflow_jobs_lambda/requirements.txt b/tests/ci/workflow_jobs_lambda/requirements.txt deleted file mode 100644 index 098e04a9798..00000000000 --- a/tests/ci/workflow_jobs_lambda/requirements.txt +++ /dev/null @@ -1 +0,0 @@ -../lambda_shared_package diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 36870d59c3a..a29c786e998 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -34,10 +34,8 @@ import urllib.parse # for crc32 import zlib from argparse import ArgumentParser -from contextlib import closing from datetime import datetime, timedelta from errno import ESRCH -from queue import Full from subprocess import PIPE, Popen from time import sleep, time from typing import Dict, List, Optional, Set, Tuple, Union @@ -69,7 +67,7 @@ TEST_FILE_EXTENSIONS = [".sql", ".sql.j2", ".sh", ".py", ".expect"] VERSION_PATTERN = r"^((\d+\.)?(\d+\.)?(\d+\.)?\d+)$" -TEST_MAX_RUN_TIME_IN_SECONDS = 120 +TEST_MAX_RUN_TIME_IN_SECONDS = 180 class SharedEngineReplacer: @@ -360,37 +358,14 @@ def clickhouse_execute_json( return rows -class Terminated(KeyboardInterrupt): - pass - - -def signal_handler(sig, frame): - raise Terminated(f"Terminated with {sig} signal") - - def stop_tests(): - global stop_tests_triggered_lock - global stop_tests_triggered - global restarted_tests - - with stop_tests_triggered_lock: - print("Stopping tests") - if not stop_tests_triggered.is_set(): - stop_tests_triggered.set() - - # materialize multiprocessing.Manager().list() object before - # sending SIGTERM since this object is a proxy, that requires - # communicating with manager thread, but after SIGTERM will be - # send, this thread will die, and you will get - # ConnectionRefusedError error for any access to "restarted_tests" - # variable. - restarted_tests = [*restarted_tests] - - # send signal to all processes in group to avoid hung check triggering - # (to avoid terminating clickhouse-test itself, the signal should be ignored) - signal.signal(signal.SIGTERM, signal.SIG_IGN) - os.killpg(os.getpgid(os.getpid()), signal.SIGTERM) - signal.signal(signal.SIGTERM, signal.SIG_DFL) + # send signal to all processes in group to avoid hung check triggering + # (to avoid terminating clickhouse-test itself, the signal should be ignored) + print("Sending signals") + signal.signal(signal.SIGTERM, signal.SIG_IGN) + os.killpg(os.getpgid(os.getpid()), signal.SIGTERM) + signal.signal(signal.SIGTERM, signal.SIG_DFL) + print("Sending signals DONE") def get_db_engine(args, database_name): @@ -700,7 +675,9 @@ class FailureReason(enum.Enum): NO_LONG = "not running long tests" REPLICATED_DB = "replicated-database" NON_ATOMIC_DB = "database engine not Atomic" + OBJECT_STORAGE = "object-storage" S3_STORAGE = "s3-storage" + AZURE_BLOB_STORAGE = "azure-blob-storage" BUILD = "not running for current build" NO_PARALLEL_REPLICAS = "smth in not supported with parallel replicas" SHARED_MERGE_TREE = "no-shared-merge-tree" @@ -735,7 +712,7 @@ def get_localzone(): class SettingsRandomizer: settings = { "max_insert_threads": lambda: ( - 0 if random.random() < 0.5 else random.randint(1, 16) + 12 if random.random() < 0.03 else random.randint(1, 3) ), "group_by_two_level_threshold": threshold_generator(0.2, 0.2, 1, 1000000), "group_by_two_level_threshold_bytes": threshold_generator( @@ -752,7 +729,7 @@ class SettingsRandomizer: "prefer_localhost_replica": lambda: random.randint(0, 1), "max_block_size": lambda: random.randint(8000, 100000), "max_joined_block_size_rows": lambda: random.randint(8000, 100000), - "max_threads": lambda: random.randint(1, 64), + "max_threads": lambda: 32 if random.random() < 0.03 else random.randint(1, 3), "optimize_append_index": lambda: random.randint(0, 1), "optimize_if_chain_to_multiif": lambda: random.randint(0, 1), "optimize_if_transform_strings_to_enum": lambda: random.randint(0, 1), @@ -844,7 +821,10 @@ class SettingsRandomizer: get_localzone(), ] ), - "prefer_warmed_unmerged_parts_seconds": lambda: random.randint(0, 10), + # This setting affect part names and their content which can be read from tables in tests. + # We have a lot of tests which relies on part names, so it's very unsafe to enable randomization + # of this setting + # "prefer_warmed_unmerged_parts_seconds": lambda: random.randint(0, 10), "use_page_cache_for_disks_without_file_cache": lambda: random.random() < 0.7, "page_cache_inject_eviction": lambda: random.random() < 0.5, "merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability": lambda: round( @@ -855,6 +835,7 @@ class SettingsRandomizer: "cross_join_min_bytes_to_compress": lambda: random.choice([0, 1, 100000000]), "min_external_table_block_size_bytes": lambda: random.choice([0, 1, 100000000]), "max_parsing_threads": lambda: random.choice([0, 1, 10]), + "optimize_functions_to_subcolumns": lambda: random.randint(0, 1), } @staticmethod @@ -1226,13 +1207,26 @@ class TestCase: elif tags and ("no-s3-storage" in tags) and args.s3_storage: return FailureReason.S3_STORAGE + elif tags and ("no-azure-blob-storage" in tags) and args.azure_blob_storage: + return FailureReason.AZURE_BLOB_STORAGE elif ( tags - and "no-s3-storage-with-slow-build" in tags - and args.s3_storage + and ("no-object-storage" in tags) + and (args.azure_blob_storage or args.s3_storage) + ): + return FailureReason.OBJECT_STORAGE + elif ( + tags + and "no-object-storage-with-slow-build" in tags + and (args.s3_storage or args.azure_blob_storage) and BuildFlags.RELEASE not in args.build_flags ): - return FailureReason.S3_STORAGE + return FailureReason.OBJECT_STORAGE + + elif "no-batch" in tags and ( + args.run_by_hash_num is not None or args.run_by_hash_total is not None + ): + return FailureReason.SKIP elif tags: for build_flag in args.build_flags: @@ -1464,8 +1458,7 @@ class TestCase: description_full = messages[result.status] description_full += self.print_test_time(result.total_time) if result.reason is not None: - description_full += " - " - description_full += result.reason.value + description_full += f"\nReason: {result.reason.value} " description_full += result.description @@ -1536,6 +1529,15 @@ class TestCase: start_time = args.testcase_start_time database = args.testcase_database + if args.client_log: + log_opt = " --client_logs_file=" + args.client_log + " " + client_options += log_opt + os.environ["CLICKHOUSE_CLIENT_OPT"] = ( + os.environ["CLICKHOUSE_CLIENT_OPT"] + if "CLICKHOUSE_CLIENT_OPT" in os.environ + else "" + ) + log_opt + # This is for .sh tests os.environ["CLICKHOUSE_LOG_COMMENT"] = args.testcase_basename @@ -1592,10 +1594,11 @@ class TestCase: # pylint:disable-next=consider-using-with; TODO: fix proc = Popen(command, shell=True, env=os.environ, start_new_session=True) - while ( - datetime.now() - start_time - ).total_seconds() < args.timeout and proc.poll() is None: - sleep(0.01) + try: + proc.wait(args.timeout) + except subprocess.TimeoutExpired: + # Whether the test timed out will be decided later + pass debug_log = "" if os.path.exists(self.testcase_args.debug_log_file): @@ -1617,6 +1620,44 @@ class TestCase: # Normalize hostname in stdout file. replace_in_file(self.stdout_file, socket.gethostname(), "localhost") + if os.environ.get("CLICKHOUSE_PORT_TCP"): + replace_in_file( + self.stdout_file, + f"PORT {os.environ['CLICKHOUSE_PORT_TCP']}", + "PORT 9000", + ) + replace_in_file( + self.stdout_file, + f"localhost {os.environ['CLICKHOUSE_PORT_TCP']}", + "localhost 9000", + ) + + if os.environ.get("CLICKHOUSE_PORT_TCP_SECURE"): + replace_in_file( + self.stdout_file, + f"PORT {os.environ['CLICKHOUSE_PORT_TCP_SECURE']}", + "PORT 9440", + ) + replace_in_file( + self.stdout_file, + f"localhost {os.environ['CLICKHOUSE_PORT_TCP_SECURE']}", + "localhost 9440", + ) + + if os.environ.get("CLICKHOUSE_PATH"): + replace_in_file( + self.stdout_file, + os.environ["CLICKHOUSE_PATH"], + "/var/lib/clickhouse", + ) + + if os.environ.get("CLICKHOUSE_PORT_HTTPS"): + replace_in_file( + self.stdout_file, + f"https://localhost:{os.environ['CLICKHOUSE_PORT_HTTPS']}/", + "https://localhost:8443/", + ) + stdout = "" if os.path.exists(self.stdout_file): with open(self.stdout_file, "rb") as stdfd: @@ -1722,7 +1763,7 @@ class TestCase: return TestResult( self.name, TestStatus.FAIL, - FailureReason.INTERNAL_QUERY_FAIL, + FailureReason.TIMEOUT, total_time, self.add_info_about_settings( self.get_description_from_exception_info(sys.exc_info()) @@ -2061,15 +2102,25 @@ class TestSuite: stop_time = None exit_code = None server_died = None -stop_tests_triggered_lock = None -stop_tests_triggered = None -queue = None multiprocessing_manager = None restarted_tests = None -def run_tests_array(all_tests_with_params: Tuple[List[str], int, TestSuite]): - all_tests, num_tests, test_suite = all_tests_with_params +class ServerDied(Exception): + pass + + +class GlobalTimeout(Exception): + pass + + +def run_tests_array(all_tests_with_params: Tuple[List[str], int, TestSuite, bool]): + ( + all_tests, + num_tests, + test_suite, + is_concurrent, + ) = all_tests_with_params global stop_time global exit_code global server_died @@ -2112,56 +2163,68 @@ def run_tests_array(all_tests_with_params: Tuple[List[str], int, TestSuite]): failures_chain = 0 start_time = datetime.now() - is_concurrent = multiprocessing.current_process().name != "MainProcess" - client_options = get_additional_client_options(args) if num_tests > 0: about = "about " if is_concurrent else "" proc_name = multiprocessing.current_process().name - print(f"\nRunning {about}{num_tests} {test_suite.suite} tests ({proc_name}).\n") + print(f"Running {about}{num_tests} {test_suite.suite} tests ({proc_name}).") while True: - if is_concurrent: - case = queue.get(timeout=args.timeout * 1.1) - if not case: + if all_tests: + try: + case = all_tests.pop(0) + except IndexError: break else: - if all_tests: - case = all_tests.pop(0) - else: - break + break if server_died.is_set(): stop_tests() - break + raise ServerDied("Server died") if stop_time and time() > stop_time: print("\nStop tests run because global time limit is exceeded.\n") stop_tests() - break + raise GlobalTimeout("Stop tests run because global time limit is exceeded") test_case = TestCase(test_suite, case, args, is_concurrent) try: description = "" - test_cace_name = removesuffix(test_case.name, ".gen", ".sql") + ": " - if not is_concurrent: + test_case_name = removesuffix(test_case.name, ".gen", ".sql") + ": " + + if is_concurrent: + description = f"{test_case_name:72}" + else: sys.stdout.flush() - sys.stdout.write(f"{test_cace_name:72}") + sys.stdout.write(f"{test_case_name:72}") # This flush is needed so you can see the test name of the long # running test before it will finish. But don't do it in parallel # mode, so that the lines don't mix. sys.stdout.flush() - else: - description = f"{test_cace_name:72}" while True: - test_result = test_case.run( - args, test_suite, client_options, server_logs_level - ) - test_result = test_case.process_result(test_result, MESSAGES) - if not test_result.need_retry: + # This is the upper level timeout + # It helps with completely frozen processes, like in case of gdb errors + def timeout_handler(signum, frame): + raise TimeoutError("Test execution timed out") + + signal.signal(signal.SIGALRM, timeout_handler) + signal.alarm(int(args.timeout * 1.1)) + test_result = None + try: + test_result = test_case.run( + args, test_suite, client_options, server_logs_level + ) + test_result = test_case.process_result(test_result, MESSAGES) + break + except TimeoutError: + break + finally: + signal.alarm(0) + + if not test_result or not test_result.need_retry: break restarted_tests.append(test_result) @@ -2181,8 +2244,9 @@ def run_tests_array(all_tests_with_params: Tuple[List[str], int, TestSuite]): failures_total += 1 failures_chain += 1 if test_result.reason == FailureReason.SERVER_DIED: - server_died.set() stop_tests() + server_died.set() + raise ServerDied("Server died") elif test_result.status == TestStatus.SKIPPED: skipped_total += 1 @@ -2193,7 +2257,7 @@ def run_tests_array(all_tests_with_params: Tuple[List[str], int, TestSuite]): if failures_chain >= args.max_failures_chain: stop_tests() - break + raise ServerDied("Max failures chain") if failures_total > 0: print( @@ -2390,7 +2454,11 @@ def extract_key(key: str) -> str: )[1] -def do_run_tests(jobs, test_suite: TestSuite, parallel): +def run_tests_process(*args, **kwargs): + return run_tests_array(*args, **kwargs) + + +def do_run_tests(jobs, test_suite: TestSuite): if jobs > 1 and len(test_suite.parallel_tests) > 0: print( "Found", @@ -2399,19 +2467,8 @@ def do_run_tests(jobs, test_suite: TestSuite, parallel): len(test_suite.sequential_tests), "sequential tests", ) - run_n, run_total = parallel.split("/") - run_n = float(run_n) - run_total = float(run_total) tests_n = len(test_suite.parallel_tests) - run_total = min(run_total, tests_n) - jobs = min(jobs, tests_n) - run_total = max(jobs, run_total) - - batch_size = max(1, len(test_suite.parallel_tests) // jobs) - parallel_tests_array = [] - for _ in range(jobs): - parallel_tests_array.append((None, batch_size, test_suite)) # If we don't do random shuffling then there will be always # nearly the same groups of test suites running concurrently. @@ -2424,33 +2481,71 @@ def do_run_tests(jobs, test_suite: TestSuite, parallel): # of failures will be nearly the same for all tests from the group. random.shuffle(test_suite.parallel_tests) - try: - with closing(multiprocessing.Pool(processes=jobs)) as pool: - pool.map_async(run_tests_array, parallel_tests_array) + batch_size = len(test_suite.parallel_tests) // jobs + manager = multiprocessing.Manager() + parallel_tests = manager.list() + parallel_tests.extend(test_suite.parallel_tests) - for suit in test_suite.parallel_tests: - queue.put(suit, timeout=args.timeout * 1.1) - - for _ in range(jobs): - queue.put(None, timeout=args.timeout * 1.1) - - queue.close() - except Full: - print( - "Couldn't put test to the queue within timeout. Server probably hung." + processes = [] + for _ in range(jobs): + process = multiprocessing.Process( + target=run_tests_process, + args=((parallel_tests, batch_size, test_suite, True),), ) - print_stacktraces() - queue.close() + processes.append(process) + process.start() - pool.join() + while processes: + sys.stdout.flush() + # Periodically check the server for hangs + # and stop all processes in this case + try: + clickhouse_execute( + args, + query="SELECT 1 /*hang up check*/", + max_http_retries=5, + timeout=20, + ) + except Exception: + print("Hang up check failed") + server_died.set() + + if server_died.is_set(): + print("Server died, terminating all processes...") + kill_gdb_if_any() + # Wait for test results + sleep(args.timeout) + for p in processes: + if p.is_alive(): + p.terminate() + break + + for p in processes[:]: + if not p.is_alive(): + processes.remove(p) + + sleep(5) run_tests_array( - (test_suite.sequential_tests, len(test_suite.sequential_tests), test_suite) + ( + test_suite.sequential_tests, + len(test_suite.sequential_tests), + test_suite, + False, + ) ) + return len(test_suite.sequential_tests) + len(test_suite.parallel_tests) else: num_tests = len(test_suite.all_tests) - run_tests_array((test_suite.all_tests, num_tests, test_suite)) + run_tests_array( + ( + test_suite.all_tests, + num_tests, + test_suite, + False, + ) + ) return num_tests @@ -2755,6 +2850,7 @@ def main(args): f"{get_db_engine(args, db_name)}", settings=get_create_database_settings(args, None), ) + break except HTTPError as e: total_time = (datetime.now() - start_time).total_seconds() if not need_retry(args, e.message, e.message, total_time): @@ -2807,7 +2903,7 @@ def main(args): test_suite.cloud_skip_list = cloud_skip_list test_suite.private_skip_list = private_skip_list - total_tests_run += do_run_tests(args.jobs, test_suite, args.parallel) + total_tests_run += do_run_tests(args.jobs, test_suite) if server_died.is_set(): exit_code.value = 1 @@ -2865,6 +2961,15 @@ def main(args): else: print(colored("\nNo queries hung.", args, "green", attrs=["bold"])) + if args.client_log: + if os.path.exists(args.client_log): + with open(args.client_log, "rb") as stream: + content = stream.read().decode() + if len(content): + print("Has fatal logs from client:\n") + print(content) + os.remove(args.client_log) + if len(restarted_tests) > 0: print("\nSome tests were restarted:\n") @@ -3099,6 +3204,12 @@ def parse_args(): default=False, help="Run tests over s3 storage", ) + parser.add_argument( + "--azure-blob-storage", + action="store_true", + default=False, + help="Run tests over azure blob storage", + ) parser.add_argument( "--no-random-settings", action="store_true", @@ -3251,26 +3362,36 @@ def parse_args(): parser.add_argument( "--replace-replicated-with-shared", action="store_true", - default=os.environ.get("USE_META_IN_KEEPER_FOR_MERGE_TREE", False), + default=os.environ.get("REPLACE_RMT_WITH_SMT", False), help="Replace ReplicatedMergeTree engine with SharedMergeTree", ) parser.add_argument( "--replace-non-replicated-with-shared", action="store_true", - default=False, + default=os.environ.get("REPLACE_MT_WITH_SMT", False), help="Replace ordinary MergeTree engine with SharedMergeTree", ) + parser.add_argument( + "--client-log", + default="./client.fatal.log", + help="Path to file for fatal logs from client", + ) return parser.parse_args() +class Terminated(KeyboardInterrupt): + pass + + +def signal_handler(sig, frame): + raise Terminated(f"Terminated with {sig} signal") + + if __name__ == "__main__": stop_time = None exit_code = multiprocessing.Value("i", 0) server_died = multiprocessing.Event() - stop_tests_triggered_lock = multiprocessing.Lock() - stop_tests_triggered = multiprocessing.Event() - queue = multiprocessing.Queue(maxsize=1) multiprocessing_manager = multiprocessing.Manager() restarted_tests = multiprocessing_manager.list() diff --git a/tests/config/config.d/grpc_protocol.xml b/tests/config/config.d/grpc_protocol.xml new file mode 100644 index 00000000000..b957618120d --- /dev/null +++ b/tests/config/config.d/grpc_protocol.xml @@ -0,0 +1,3 @@ + + 9100 + diff --git a/tests/config/install.sh b/tests/config/install.sh index e04392d893b..1b0edc5fc16 100755 --- a/tests/config/install.sh +++ b/tests/config/install.sh @@ -27,6 +27,7 @@ ln -sf $SRC_PATH/config.d/secure_ports.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/clusters.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/graphite.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/graphite_alternative.xml $DEST_SERVER_PATH/config.d/ +ln -sf $SRC_PATH/config.d/grpc_protocol.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/database_atomic.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/max_concurrent_queries.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/merge_tree_settings.xml $DEST_SERVER_PATH/config.d/ @@ -57,7 +58,6 @@ ln -sf $SRC_PATH/config.d/forbidden_headers.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/enable_keeper_map.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/custom_disks_base_path.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/display_name.xml $DEST_SERVER_PATH/config.d/ -ln -sf $SRC_PATH/config.d/reverse_dns_query_function.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/compressed_marks_and_index.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/disable_s3_env_credentials.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/enable_wait_for_shutdown_replicated_tables.xml $DEST_SERVER_PATH/config.d/ @@ -183,13 +183,7 @@ elif [[ "$USE_AZURE_STORAGE_FOR_MERGE_TREE" == "1" ]]; then fi if [[ -n "$EXPORT_S3_STORAGE_POLICIES" ]]; then - if [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then - echo "Azure configuration will not be added" - else - echo "Adding azure configuration" - ln -sf $SRC_PATH/config.d/azure_storage_conf.xml $DEST_SERVER_PATH/config.d/ - fi - + ln -sf $SRC_PATH/config.d/azure_storage_conf.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/storage_conf.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/storage_conf_02944.xml $DEST_SERVER_PATH/config.d/ ln -sf $SRC_PATH/config.d/storage_conf_02963.xml $DEST_SERVER_PATH/config.d/ diff --git a/tests/config/users.d/insert_keeper_retries.xml b/tests/config/users.d/insert_keeper_retries.xml index f01157d40ac..0cf41473616 100644 --- a/tests/config/users.d/insert_keeper_retries.xml +++ b/tests/config/users.d/insert_keeper_retries.xml @@ -4,7 +4,7 @@ 100 0.01 1 - 1 + 10 diff --git a/tests/instructions/easy_tasks_sorted_ru.md b/tests/instructions/easy_tasks_sorted_ru.md index bc95e6b1c37..fbd86ebf08f 100644 --- a/tests/instructions/easy_tasks_sorted_ru.md +++ b/tests/instructions/easy_tasks_sorted_ru.md @@ -78,7 +78,7 @@ Upd: сделали по-другому: теперь всё безопасно. ## LEFT ONLY JOIN -## Функции makeDate, makeDateTime. +## + Функции makeDate, makeDateTime. `makeDate(year, month, day)` `makeDateTime(year, month, day, hour, minute, second, [timezone])` @@ -187,13 +187,13 @@ https://clickhouse.com/docs/en/operations/table_engines/external_data/ Не работает, если открыть clickhouse-client в интерактивном режиме и делать несколько запросов. -## + Настройка для возможности получить частичный результат при cancel-е. +## Настройка для возможности получить частичный результат при cancel-е. Хотим по Ctrl+C получить те данные, которые успели обработаться. ## Раскрытие кортежей в функциях высшего порядка. -## Табличная функция loop. +## + Табличная функция loop. `SELECT * FROM loop(database, table)` diff --git a/tests/integration/compose/docker_compose_ldap.yml b/tests/integration/compose/docker_compose_ldap.yml index f199516f315..1f50b34735d 100644 --- a/tests/integration/compose/docker_compose_ldap.yml +++ b/tests/integration/compose/docker_compose_ldap.yml @@ -15,7 +15,10 @@ services: ports: - ${LDAP_EXTERNAL_PORT:-1389}:${LDAP_INTERNAL_PORT:-1389} healthcheck: - test: "ldapsearch -x -b dc=example,dc=org cn > /dev/null" + test: > + ldapsearch -x -H ldap://localhost:$$LDAP_PORT_NUMBER -D $$LDAP_ADMIN_DN -w $$LDAP_ADMIN_PASSWORD -b $$LDAP_ROOT + | grep -c -E "member: cn=j(ohn|ane)doe" + | grep 2 >> /dev/null interval: 10s retries: 10 timeout: 2s diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index cda9abe8bce..400ee735916 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -52,6 +52,7 @@ from helpers.client import QueryRuntimeException import docker from .client import Client +from .retry_decorator import retry from .config_cluster import * @@ -73,7 +74,7 @@ CLICKHOUSE_ERROR_LOG_FILE = "/var/log/clickhouse-server/clickhouse-server.err.lo # Minimum version we use in integration tests to check compatibility with old releases # Keep in mind that we only support upgrading between releases that are at most 1 year different. # This means that this minimum need to be, at least, 1 year older than the current release -CLICKHOUSE_CI_MIN_TESTED_VERSION = "22.8" +CLICKHOUSE_CI_MIN_TESTED_VERSION = "23.3" # to create docker-compose env file @@ -1444,9 +1445,9 @@ class ClickHouseCluster: def setup_azurite_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_azurite = True env_variables["AZURITE_PORT"] = str(self.azurite_port) - env_variables[ - "AZURITE_STORAGE_ACCOUNT_URL" - ] = f"http://azurite1:{env_variables['AZURITE_PORT']}/devstoreaccount1" + env_variables["AZURITE_STORAGE_ACCOUNT_URL"] = ( + f"http://azurite1:{env_variables['AZURITE_PORT']}/devstoreaccount1" + ) env_variables["AZURITE_CONNECTION_STRING"] = ( f"DefaultEndpointsProtocol=http;AccountName=devstoreaccount1;" f"AccountKey=Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==;" @@ -1642,9 +1643,9 @@ class ClickHouseCluster: # Code coverage files will be placed in database directory # (affect only WITH_COVERAGE=1 build) - env_variables[ - "LLVM_PROFILE_FILE" - ] = "/var/lib/clickhouse/server_%h_%p_%m.profraw" + env_variables["LLVM_PROFILE_FILE"] = ( + "/var/lib/clickhouse/server_%h_%p_%m.profraw" + ) clickhouse_start_command = CLICKHOUSE_START_COMMAND if clickhouse_log_file: @@ -1657,9 +1658,9 @@ class ClickHouseCluster: cluster=self, base_path=self.base_dir, name=name, - base_config_dir=base_config_dir - if base_config_dir - else self.base_config_dir, + base_config_dir=( + base_config_dir if base_config_dir else self.base_config_dir + ), custom_main_configs=main_configs or [], custom_user_configs=user_configs or [], custom_dictionaries=dictionaries or [], @@ -2618,7 +2619,9 @@ class ClickHouseCluster: [ "bash", "-c", - f"/opt/bitnami/openldap/bin/ldapsearch -x -H ldap://{self.ldap_host}:{self.ldap_port} -D cn=admin,dc=example,dc=org -w clickhouse -b dc=example,dc=org", + f"/opt/bitnami/openldap/bin/ldapsearch -x -H ldap://{self.ldap_host}:{self.ldap_port} -D cn=admin,dc=example,dc=org -w clickhouse -b dc=example,dc=org" + f'| grep -c -E "member: cn=j(ohn|ane)doe"' + f"| grep 2 >> /dev/null", ], user="root", ) @@ -2666,15 +2669,12 @@ class ClickHouseCluster: images_pull_cmd = self.base_cmd + ["pull"] # sometimes dockerhub/proxy can be flaky - for i in range(5): - try: - run_and_check(images_pull_cmd) - break - except Exception as ex: - if i == 4: - raise ex - logging.info("Got exception pulling images: %s", ex) - time.sleep(i * 3) + + retry( + log_function=lambda exception: logging.info( + "Got exception pulling images: %s", exception + ), + )(run_and_check)(images_pull_cmd) if self.with_zookeeper_secure and self.base_zookeeper_cmd: logging.debug("Setup ZooKeeper Secure") @@ -2947,7 +2947,11 @@ class ClickHouseCluster: "Trying to create Azurite instance by command %s", " ".join(map(str, azurite_start_cmd)), ) - run_and_check(azurite_start_cmd) + retry( + log_function=lambda exception: logging.info( + f"Azurite initialization failed with error: {exception}" + ), + )(run_and_check)(azurite_start_cmd) self.up_called = True logging.info("Trying to connect to Azurite") self.wait_azurite_to_start() diff --git a/tests/integration/helpers/retry_decorator.py b/tests/integration/helpers/retry_decorator.py new file mode 100644 index 00000000000..aaa040464c2 --- /dev/null +++ b/tests/integration/helpers/retry_decorator.py @@ -0,0 +1,36 @@ +import time +import random +from typing import Type, List + + +def retry( + retries: int = 5, + delay: float = 1, + backoff: float = 1.5, + jitter: float = 2, + log_function=lambda *args, **kwargs: None, + retriable_expections_list: List[Type[BaseException]] = [Exception], +): + def inner(func): + def wrapper(*args, **kwargs): + current_delay = delay + for retry in range(retries): + try: + func(*args, **kwargs) + break + except Exception as e: + should_retry = False + for retriable_exception in retriable_expections_list: + if isinstance(e, retriable_exception): + should_retry = True + break + if not should_retry or (retry == retries - 1): + raise e + log_function(retry=retry, exception=e) + sleep_time = current_delay + random.uniform(0, jitter) + time.sleep(sleep_time) + current_delay *= backoff + + return wrapper + + return inner diff --git a/tests/integration/parallel_skip.json b/tests/integration/parallel_skip.json index 33dd85aceaf..3c3d1b6cc96 100644 --- a/tests/integration/parallel_skip.json +++ b/tests/integration/parallel_skip.json @@ -48,6 +48,7 @@ "test_system_metrics/test.py::test_readonly_metrics", "test_system_replicated_fetches/test.py::test_system_replicated_fetches", "test_zookeeper_config_load_balancing/test.py::test_round_robin", + "test_zookeeper_config_load_balancing/test.py::test_az", "test_zookeeper_fallback_session/test.py::test_fallback_session", "test_global_overcommit_tracker/test.py::test_global_overcommit", diff --git a/tests/integration/runner b/tests/integration/runner index a583d7fe897..0667541b196 100755 --- a/tests/integration/runner +++ b/tests/integration/runner @@ -243,6 +243,10 @@ if __name__ == "__main__": "-n", "--parallel", action="store", dest="parallel", help="Parallelism" ) + parser.add_argument( + "--count", action="store", type=int, dest="count", help="Repeat count" + ) + parser.add_argument( "--no-random", action="store", @@ -318,6 +322,10 @@ if __name__ == "__main__": parallel_args += "--dist=loadfile" parallel_args += f" -n {args.parallel}".format() + repeat_args = ( + f" --count {args.count}" if args.count is not None and args.count > 0 else "" + ) + rand_args = "" # if not args.no_random: # rand_args += f"--random-seed={os.getpid()}" @@ -409,7 +417,7 @@ if __name__ == "__main__": f"--volume={args.utils_dir}/grpc-client/pb2:/ClickHouse/utils/grpc-client/pb2 " f"--volume=/run:/run/host:ro {dockerd_internal_volume} {env_tags} {env_cleanup} " f"-e DOCKER_CLIENT_TIMEOUT=300 -e COMPOSE_HTTP_TIMEOUT=600 {use_old_analyzer} -e PYTHONUNBUFFERED=1 " - f'-e PYTEST_ADDOPTS="{parallel_args} {pytest_opts} {tests_list} {rand_args} -vvv"' + f'-e PYTEST_ADDOPTS="{parallel_args} {repeat_args} {pytest_opts} {tests_list} {rand_args} -vvv"' f" {DIND_INTEGRATION_TESTS_IMAGE_NAME}:{args.docker_image_version}" ) diff --git a/tests/ci/lambda_shared_package/__init__.py b/tests/integration/test_accept_invalid_certificate/__init__.py similarity index 100% rename from tests/ci/lambda_shared_package/__init__.py rename to tests/integration/test_accept_invalid_certificate/__init__.py diff --git a/tests/integration/test_accept_invalid_certificate/certs/ca-cert.pem b/tests/integration/test_accept_invalid_certificate/certs/ca-cert.pem new file mode 100644 index 00000000000..c4f88831126 --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/certs/ca-cert.pem @@ -0,0 +1,32 @@ +-----BEGIN CERTIFICATE----- +MIIFhTCCA22gAwIBAgIUZHeZ4ME7rL1RYGJZGPgOTHl7btMwDQYJKoZIhvcNAQEL +BQAwUjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUxITAfBgNVBAoM +GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDELMAkGA1UEAwwCY2EwHhcNMjQwNzAx +MDUxMTI0WhcNMzQwNjI5MDUxMTI0WjBSMQswCQYDVQQGEwJSVTETMBEGA1UECAwK +U29tZS1TdGF0ZTEhMB8GA1UECgwYSW50ZXJuZXQgV2lkZ2l0cyBQdHkgTHRkMQsw +CQYDVQQDDAJjYTCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBALf623IC +1r6Bzwa4dTufRPvFWQinT4uIBv7UvAk8tu8kmLSMOOC2COALNRCSu0s1SFcO6i5G +CgmaWdoTXuw+NYH9n5rU06N991NBfCoLbf0eSoie7bRdTCKkqWWKaw+dCvKL5vWj +LEZRmdh8ZN0lxa05eIsLxErvgWqE8hOOQ7jgRqYLAagp6XkMdB+YRkXFtbVcze83 +ZNxFqNjdgwYYybaqDZKGeOnRDRof7P/cyMkbGNUI7+JYA5eGmDgiOTGbaCZx5AUK +DKmkdmy5IJBWS0rYR8Ol9Sjs6NSYkmYiLFEWFNSvI88K99+Ab3DEi9JMEqMeNTJ5 +1sx48e2OfMG7ywoGj3D3zf7SUxAfwW+7Ls+v7c/kB9hvspA9iVsqjAKcSPAPnJXk +zR5r+zcPwjb/dvGGGB+D7PhvreEpj0n7PRFEB07jviA1E8zkwdk1L9FefYMG2oRl +8J7miWxoyqUOtA6AYhbN36+I0iJkjxIjXLLzB/dhOzerNTzRHcJtmgSyIpcPIV42 +IsJBttLOW0Mpkfw8z88xji+1VWHm+H8sOWCzbfjncScu/z5Xhb92C2T+AG+RpakB +LwyTsovf+xv1gGF/1gmmdKSoV/JZuHSczMmmDLJ9vXRnIpK0AgL16lfs6o0C2Pe2 +OHJv0oViokU5mW+Mge3U9kUrh6BVITk8UjUDAgMBAAGjUzBRMB0GA1UdDgQWBBRA +x0+tgav3Q2gRt34AtKZiGfFH7zAfBgNVHSMEGDAWgBRAx0+tgav3Q2gRt34AtKZi +GfFH7zAPBgNVHRMBAf8EBTADAQH/MA0GCSqGSIb3DQEBCwUAA4ICAQBcV7s7au6r +bEkHaW9RbGe/7iHesOcH2uyiALGeiCF+pfN9QbfkpJJbMRDhKr+sVqJH6lS+Iys5 +MpPaRHOInNs5ajiPQcIxtxAnw2HEsy6LUVvs6RUHrS7Bxlc406YQCY/sPyItAt8t +U/qhx5VNj6Mx1smyKZHFHZIZ0eYieriJkK6NB2E7Ji6EPGe7lkh3kEoVTfaSVCkP +p3d4U603kzCk8PGkYlhjWUCVMhDRwgEy6FXHmU156HivGpYurKcab7cMUfr50MyK +1Mx/NaHOYHhIIQ87t9wF2yQaMskEcyK9XOorp+HFtnTdci+6+J3u2Kmyckq81uy9 +uZreH8bUKgmntOhSqtl4tWGMWxX78c7rX2cGIbSJsZzzjmUdhvnpYeuhCCrUPzY3 +kIt9Q6q7DCKG+f+D4bHQ292emEfKYb9LgftyOX3pcAJ+d3c56Cs7c0RRVFAZyVnX +1O9ZvHrypmii5RARNuyVtttgDpuGBrRCa2kSNlXD9PJJitI2kfoWrCI/88Pw8H/y +bLO9/ZcCOsxBGmfrQDMUSC8rE8JLUSKV80otWwtHcxcnfpsobs64/Rz6aOLYEiS4 +/0M6xyUYdOsEGUh1uyX5qxto8iOWop4S9DLy1QVx+N8/fC6gjhMb4Vg2oTBnHlGr +JcxdYP8jVe6tUgvRKAyNlccPIq20udfGTw== +-----END CERTIFICATE----- diff --git a/tests/integration/test_accept_invalid_certificate/certs/ca-key.pem b/tests/integration/test_accept_invalid_certificate/certs/ca-key.pem new file mode 100644 index 00000000000..66d548d0cf5 --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/certs/ca-key.pem @@ -0,0 +1,52 @@ +-----BEGIN PRIVATE KEY----- +MIIJQwIBADANBgkqhkiG9w0BAQEFAASCCS0wggkpAgEAAoICAQC3+ttyAta+gc8G +uHU7n0T7xVkIp0+LiAb+1LwJPLbvJJi0jDjgtgjgCzUQkrtLNUhXDuouRgoJmlna +E17sPjWB/Z+a1NOjffdTQXwqC239HkqInu20XUwipKllimsPnQryi+b1oyxGUZnY +fGTdJcWtOXiLC8RK74FqhPITjkO44EamCwGoKel5DHQfmEZFxbW1XM3vN2TcRajY +3YMGGMm2qg2Shnjp0Q0aH+z/3MjJGxjVCO/iWAOXhpg4Ijkxm2gmceQFCgyppHZs +uSCQVktK2EfDpfUo7OjUmJJmIixRFhTUryPPCvffgG9wxIvSTBKjHjUyedbMePHt +jnzBu8sKBo9w983+0lMQH8Fvuy7Pr+3P5AfYb7KQPYlbKowCnEjwD5yV5M0ea/s3 +D8I2/3bxhhgfg+z4b63hKY9J+z0RRAdO474gNRPM5MHZNS/RXn2DBtqEZfCe5ols +aMqlDrQOgGIWzd+viNIiZI8SI1yy8wf3YTs3qzU80R3CbZoEsiKXDyFeNiLCQbbS +zltDKZH8PM/PMY4vtVVh5vh/LDlgs23453EnLv8+V4W/dgtk/gBvkaWpAS8Mk7KL +3/sb9YBhf9YJpnSkqFfyWbh0nMzJpgyyfb10ZyKStAIC9epX7OqNAtj3tjhyb9KF +YqJFOZlvjIHt1PZFK4egVSE5PFI1AwIDAQABAoICACPKkBVqevddznYqvVfHgMaQ +GmWGEwH8iQvbO+d6IcZ9Mp3xJ4GDuE9Jrt2WBgy4BWY2VxWAd6RtnCxCF3XTFlq/ +CeU3oSKNt2dBdpPBXHjhAzP34xtzCQQ2T2QbHP/RF9eeouRHha36YvkMHh0to3qz +EwHN3LaNSHp9+ZAXXZJiQyw7OOvLl4l+hDq/R3B8aYCqWL/AYYU+of3qffpaKSe5 +hmFqeqxWCHv8GQGfpIoHFPogvnYklDJOs188PS8q4gSUpHzceyEnhwhYrX6W4xfg +4nkr2HlJOXReaCCpZkCyEIRdvBQjuJwI4uUnv2G+feOw1T2IGWL25m3GCfwweeuZ +AWqqbj5Xnl1+VgvaNm68aHBVlfZ14rIJ/Hu1yJV/E+5m8+3xewAor0jshWceC3OS +lhrZp+EyrwxsKj17s2IBhk4EIOLjfosoDqdw3E2DrtqKrnI63MLUTaEqHWyXnF82 +Mn880DR/E3Byj2kKtdIVesAdO0ARtIRqedPunpcvhp1eMZ2wrLJXAHMDBJlUZF64 +ZAyO3ll79PhBn/QShGu+KQ682Ns1T3j3J14n2zqBKsU1GzSLBOZO8QXfFELsv6Xj +NGYPg2SbkRwbYHEPr6G9ehsxKaTpMAtPesdz1devKDI4nSpd2qE9ZfREH+CW0MkK +bpiL20d4blIG1DP84qodAoIBAQDEBXbn0IEeyL3j50/AUJY/K4seeDm1ltJmhoNU +HC0dwt2FOHrD1J/Cx6S19zuEVt1qMwJXELxOVzixykKPy73ETW+Qfcpo2J51THrl +fSEUpoheD37Cwjv9vzb6Qb4ZXDr4acLIB06mAUKfo0Pez2x6IybWqS/dZoD1bfh5 +wiSCITXQB4n6mI7chfDDyTqgf3C5U5b2UABLDdIKQZHY5kX55Vfr0Lz+Gt2t6uUy +hUiN4C+Qrx1LoHY4x7g1SG9lKASG5zv5dksc+a6uYsS4N8TM3ZQz7BnZrCQPNiWz +K70p5pMjdVTHmQqThoww9y+rRiHMU3qUSZ0tX4nrY99/MiFNAoIBAQDwRi0GzkeA +ECWfr2R+vh5lnxXBeAvM46v+8CzwJHNxtGAhB1ojod7GxOxHiMnWDXubRiNVlhx0 +OHeCq11hSQh2+a7s8xwxYbC3paYf1ZQq81YmTGd8v+/QX9b21Kq9dpx+G2C3eikZ +rHMuSifrSEwZMxdw/gKvQ+x66PTNeK8fIgAtZsE96VX3u7sCCNhbBElWWg5OwwMf +0RH5Pi/O8UXxkZs2SUKFaXUFMxjYn9r9636QI8LaLsCvcbPjrhshPtHWHqgrZzia +8gD00ty3mcydRU//xNfHruew2iRZUVVB8E0GQ80eQCsHrI0YkP876at55DKbyZLi +pGisswDdyoePAoIBAQCfOFUSDMGksKfVY8Da1FqGKydQc5Vkk4KJdr9b5Zsb7nmi +yXKmeKHgQ+vdRlmYsDofFhnTgKuD8sMBXuncBUE6Hy/8rMBpk3J6ksnMeGwTawWf +aXSsW7sA7H2KnzX7QXhA6R8m/RB8NM/Z/XKWm+Nf/geBG2Hes3ESaeYHeJPRA4rD +VCxWWe7I3x7m2R/+2CVG08fugKMMU8iFPidqh/jU994HlMC+ojHoH8FV3P9HBo98 +NVFVaZw9mLRey1x4TR4IOfD0O4rdmU2zJlTl0Z7QTTchSajhrzqHBJNwyplXXIut +pJNdoUj+9a4P4+hbdZbkcfZCVt0NyumrQq96kMyNAoIBAQC8Jcemd6tHnd1IYPip +wshcvSk1GYLsCzAo7vvRhEmeNSe9VVuBYYToU1XOYk/ivSYYneCXaXeYjosPGI+D +2cO2HqiaxXhDNwNBm4YtkXp385p3XQcAifAJVLMpD7kvqYy7t0aSBiffAcoj6k++ +Dxv/86EcituhUOgCCx9/Mky3bVGEO92AVJJL7oZu/mwlQYMICnb/t/dfV8DiUo7v +VsF1e5Z1+QINJFpNGwkOwIakpV6zEj3Hyip6d+fslu6Jpv570//OFQMfWffjvqa1 +Ysr2MEuHtiVeKRkmxHsIMgc4HirCiB872/WWCclm/kvPJ5huV4JFuZ+6uBF7gqRg +CPMDAoIBAGs0KCL3a/eXIPWhFcOgtpcNTb3o4gK/s38+GLIVbdP7MUNEkMeWq+Pf +PBUca2p1M8CkbpRi0OAvg1gewAR/+bCdyzoipl1+3v3goYC3J5TAs5NT/siBcjrm +QcTKEHmTeVu82Kpi1JVtcfS2TWuSmT6vSB2A2H3R4GEPQQXnXeS/FnRA0I85MFzq +zxMkar5VY7Eg0Z3/bN5dNqAYSzqjgMMhRMPN1W54gjL2ev65Mzt1jYf6xz7gpJOh +6Fw2EOrM23c4OckZx2dZ7Je6ns2U82D81ffhXRwWfQxoKXvMVMccBJmHTbkVEZlE +VAq9BWwNistc+YfRka1OOeq37MgpXq8= +-----END PRIVATE KEY----- diff --git a/tests/integration/test_accept_invalid_certificate/certs/client-cert.pem b/tests/integration/test_accept_invalid_certificate/certs/client-cert.pem new file mode 100644 index 00000000000..2dd497409c3 --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/certs/client-cert.pem @@ -0,0 +1,30 @@ +-----BEGIN CERTIFICATE----- +MIIFLzCCAxcCFA6o2DM1UGQ1NNCQzp0n4hb3iSd9MA0GCSqGSIb3DQEBCwUAMFIx +CzAJBgNVBAYTAlJVMRMwEQYDVQQIDApTb21lLVN0YXRlMSEwHwYDVQQKDBhJbnRl +cm5ldCBXaWRnaXRzIFB0eSBMdGQxCzAJBgNVBAMMAmNhMB4XDTI0MDcwMTA1MTEy +NloXDTM0MDYyOTA1MTEyNlowVjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUt +U3RhdGUxITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEPMA0GA1UE +AwwGY2xpZW50MIICIjANBgkqhkiG9w0BAQEFAAOCAg8AMIICCgKCAgEAl9i6heiD +2inNs0lKt7XjK0QyXKyIg63uVO39ZpLY5sj1noZnbFmRo5MYC4MMIrlD3wGnqUo/ +mdev2eg8qYry0CmzhLscVEha7LN+7wmdftYSGkXOl9E08KIv9Uw2bHap4KEIRvJ6 +298bdsuv1l4ljWo3VDkPCczXIaUi94/o1uNu0sbPRluBP8kjFLhwV8Ph/7mxsofu +V5tDYNOZ8VozA7aHAnJqXBCFZuW6igIH3XaROY9LOe2FD+qxtpqJHp7E+ZNsXXH5 +SXMQzE2/X7eKHA7zvdG4z5GLgmz3GEvq3ZBJUGUe6DCnrxqBXUeLTqyummohGE57 +LuqzrH9etFj/91NaZW9uwN42pxGJSkeZaJ5BDnzO3VYxqvvDnFs/2TFbJiZ065vI +NpA0PnuAkGjztaNAm6qXNTXsWo8bdi8TDgIwKxoo5KIlWsxnxKB82tfBovWd+JLa +xLlWNKbKgNb6xc9OmNCyhYkRXhqIsBrpLy6n2yppyYEiCBzPrqwJ2p3Y88yj3dsc +xiiu0ngFSroE2pLwCgi1rQY1x9VO4XQrrq6vBvPhi530+cdXdVTnnyBhaOMX4bWX +OOjI97xJmR/YbOvpgKw9v/na82t2fgKoqIs124iZ21QOiffOIfwwg2AKj8EaWJ9/ +Era8sZMsq/g56PWYwT1J85cCMvbNwGOXjusCAwEAATANBgkqhkiG9w0BAQsFAAOC +AgEAAPSps9k8HEZrfBbzH5SHASNTUc7F81h2EYhaEuQrXoan39u3syCKYMSJDt0P +AVCYbIvXmlHeaBba9YXtZVZJpCM53b+x01297x1bXzRklYxnpuD2IauuQgiaJBXV +VlCjEy7GHBAmRQwV7llkoxx+qHsA0b+oHKvTRc9FJzk/Fx4Eb496gNMvUk47f+15 +WBmWoQolqwbH1Ivhca7BpTvTSeNGkJ14cbmuhwryLstHJCPep+vp2TYS9bHFZIPY +g3APnWv1qShuA7Oj68GLpRVJmImgyPMK0i8LDGcLhrDiG+HYEr89MiRWDvBHun7O +Sp5y8zDnDZSrN+EtRp8DypHeB7R0Ukmn76NP6NsigWCPMlLaHNVsvVAwWwFNae9u +OxsoW07tJo3lZjDx+tFrUb7IRyn/9PLDIwEs4SvDP+O97o0JkCszwfRhkDnjY7Qj +XCgXoCBalIadgnGshbEjFLH0ZSwAoCIC/35X0Zo7cd1+Fz4Xq1tgnV5ec6ci/EHi +qsOmGfTqZz5HJlmfyxUOYicizV6SjNZSFWY07ebYRewx6ymdDFhqrbnJIEUGJYlU +5qDY1nemoqh9QRIhm1+73hAKHtmcaa1u72gj1QBtRD5eG3ToM3GHgdbUUPvFpenx +UPfzw/wbzITSqPoYr1/XIGq9e1muTqHbFNlGVByIW0ejeZQ= +-----END CERTIFICATE----- diff --git a/tests/integration/test_accept_invalid_certificate/certs/client-key.pem b/tests/integration/test_accept_invalid_certificate/certs/client-key.pem new file mode 100644 index 00000000000..09c04c1e815 --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/certs/client-key.pem @@ -0,0 +1,52 @@ +-----BEGIN PRIVATE KEY----- +MIIJQwIBADANBgkqhkiG9w0BAQEFAASCCS0wggkpAgEAAoICAQCX2LqF6IPaKc2z +SUq3teMrRDJcrIiDre5U7f1mktjmyPWehmdsWZGjkxgLgwwiuUPfAaepSj+Z16/Z +6DypivLQKbOEuxxUSFrss37vCZ1+1hIaRc6X0TTwoi/1TDZsdqngoQhG8nrb3xt2 +y6/WXiWNajdUOQ8JzNchpSL3j+jW427Sxs9GW4E/ySMUuHBXw+H/ubGyh+5Xm0Ng +05nxWjMDtocCcmpcEIVm5bqKAgfddpE5j0s57YUP6rG2mokensT5k2xdcflJcxDM +Tb9ft4ocDvO90bjPkYuCbPcYS+rdkElQZR7oMKevGoFdR4tOrK6aaiEYTnsu6rOs +f160WP/3U1plb27A3janEYlKR5lonkEOfM7dVjGq+8OcWz/ZMVsmJnTrm8g2kDQ+ +e4CQaPO1o0Cbqpc1Nexajxt2LxMOAjArGijkoiVazGfEoHza18Gi9Z34ktrEuVY0 +psqA1vrFz06Y0LKFiRFeGoiwGukvLqfbKmnJgSIIHM+urAnandjzzKPd2xzGKK7S +eAVKugTakvAKCLWtBjXH1U7hdCuurq8G8+GLnfT5x1d1VOefIGFo4xfhtZc46Mj3 +vEmZH9hs6+mArD2/+drza3Z+AqioizXbiJnbVA6J984h/DCDYAqPwRpYn38Stryx +kyyr+Dno9ZjBPUnzlwIy9s3AY5eO6wIDAQABAoICAAl8R6ICPpCcEXrUb5qn2Ujy +IJQdT+D0yJqGTD6QCDvsLDstGcWMcx1him5785IxJ9YpZBbnidO0euXJo8R3Ql8S +aGHK4ZtfFyeKAYNoJ0mSIx03LdiRAEOJ7QDf2LI017yJS9IkeW5ql0fTas0za2Jm +nCBZZrNNzvv7a/BEn8NHkfMQV5Fgy9ESvP3qcHhsNHly/aeD5ZEmbZcmWeb82orS +nxEnR0hUSnI0JpibsycT90DimAOh3LDXfd3E0Iz9VsA8urMpqtpDYzZ1tShed7TP +tyiaxS1+PUNIkEc9xRv2i2FQyhVmpmZdoCXfNi84nUtWhJYE6HkrmgRwnnLQp5rr +4stSQLo9X/5ML/bo2cj1eAiH+P6bAWaZhkMYPGkhBYwwgYqZF0Jn2Ay07BPF5gqq +rDY0mjF+Nm/NOqHfQjWHSj7KYYHm6qU1s5cAnDWLfznXGWZHpoI4UqAZtpAwGWEh +f/nS2LyepFW8RE6mbRDe+PkOM2455QdM8MuQc8drGInJ57A5e7y1LxOvFaUP+QtG +fqXok95XciMfttcR8nfYkUyC5QX+JmDZRf1U/DUxTJtbKXRKVG8WaFagkViiCniw +Q4RbrS4Utu6h+8HXdI9mK2pIHlJPp202b9gPry1j3kya8/jx5cTFWCcPj4IlgY51 +ZC1+vXJUezwq60gWJrRtAoIBAQDLsR9M+ZdBC5sl5fiVGsznAvfEVv/Cf7pmeXqt +q4BZT7kB8zfdfRRMzmZiQIX0zTbBn7Q7kWULcl8HgPbMEGZorc24CFuCq8k0n2ji +0aJsTJlfr7kUfTZ5c3HN1fGhDrnGoCL1eElt/Es4hZTIDnbn3i+RoCqwhCqhC0Lo +L7f6iHrVIcKyoaz+Cye9QKA89PVHMNM0f41zFJTVpSzGfWefG4vNoeU6nxqwkWVJ +Guhkt1EsJUScHueAa7hUYlELDy99tlqOPLdG7/GG6LmE6pM7R1E6DishnL1VgOll +KL6fouoz8nTve59eCDQECNzXN8lxPvx6TdufabGJTSZ3iNjPAoIBAQC+10DFlO3Y +XLHfX9LWK4rdkghx9AoeCFIVZO2G29H4zyDQn7hAMISWeVCqdAKhPQMJmAXyhQcc +PLEUdKVwifbPNO/IxsWL8SOfbI28XoZLYkQLual6JPa10jemFsinVRTPLU7RA1/T +7t/mWvyaAHmODWUPKSs1ltzh5dNwvnSNou6h72TytwlMIk5h8Z8ZtMIY2v5Ca2Mx +I7Wi8/5RebpszBbuLYy0GBqK5oqQov1YH0WK3ae4leaPWlZ49nyNYpx5Bvngm+d8 +4F+HvcxdpckRPfenl9nBW+0UHVXZORepQ+g0CNJxAQTomXy5VeEnrpiylrec4mE7 +YD15GKDxeHclAoIBAQC6efcr4zwDwnpRIc6HwrdfriniR7tx9rlRY1lacS3ZldDL +5jx4uy/AeRzmSifYy56beHfqKNuWtSgmmHxYqZL3tlYlaxFL+bUERgQh0sWd3Cp1 +nUwVY+RhH71KMdk6F18gZi5bB7pBxL2gizqARC27bH+kFDvAyd60/LhOFJGb/SJm +yLeUDctv8Ap7ihSoeaGrgRl6cGOwvWDcjpEFTn10/iTpJpPTLSGkEBfn64hh3Cml +/qBnRSuvN+7S4gdL1qUls7IHkbJAcR5u4dQlyRLtLAdC3QHDjdkP1CrX/4s0XlOH +JyrCrCxkS6JDJ1H2fg4fbQh7d5few0ZTapwSxuelAoIBAQCloou8MTuR6zoShsem +TRp0npen18ikMlKT8QpbXxTJ33Q6vjWbuMCLHXYOyyuxKBQn0oNu71ixgFetSmSA +3a2SP/7SAvtYL43cg8ossNncOciXaKgYtFZQjsjv5Zx351dtTYS30jbs6cn41AWj +4D+gNC3/T5sKIwtJIUE3/IWKnNyDPhUeuBBgvPHM/no5RHKmQgQmP7BS6QxwB1c2 +/fD5FtXcGTqi4LzAy8F+MYizYEjqcuFsSuZWi3C8AisuR2VJge2Eravn7ZHcbbYf +nZF63o7BLgG4WgHV0HQ/WdcYeTDjYLjNjbXl77mZ3ljSKS6s6jUo/C6p6yjPnrmp +ewXlAoIBAHX9aLtoj06zU5dtrMVjhYip6Z5LEtHsXCvOmnAl91uUXKwtMze6Y7xc +4ZggVWBR9lFyq9bAAuIXkRuE1jbvEwl5sLZY5Hf6GrIUR2uD7ZYFbogBTgmWCX9S +vYRpofEwquX0AUt+a4nO09wPBX1BJMViAEZSAWGfCfzVz8rXYIgcmu471wrpQHv4 +/whNA90JX0v6tuOASgR2dGorU//0ejGcm0bcYSuivKPxBfmVNeZNgIJXLM0ZHql1 +qwLnWXS0e0kADWhp/GAWVo0xzefgD3mDmyG2Qbz8MLaDkwPU/L073ZZNT1AHFZZE +xfVijyydtPqa/JDgqcUFBxLtiKGTbbI= +-----END PRIVATE KEY----- diff --git a/tests/integration/test_accept_invalid_certificate/certs/client-req.pem b/tests/integration/test_accept_invalid_certificate/certs/client-req.pem new file mode 100644 index 00000000000..507db7bae6c --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/certs/client-req.pem @@ -0,0 +1,27 @@ +-----BEGIN CERTIFICATE REQUEST----- +MIIEmzCCAoMCAQAwVjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUx +ITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEPMA0GA1UEAwwGY2xp +ZW50MIICIjANBgkqhkiG9w0BAQEFAAOCAg8AMIICCgKCAgEAl9i6heiD2inNs0lK +t7XjK0QyXKyIg63uVO39ZpLY5sj1noZnbFmRo5MYC4MMIrlD3wGnqUo/mdev2eg8 +qYry0CmzhLscVEha7LN+7wmdftYSGkXOl9E08KIv9Uw2bHap4KEIRvJ6298bdsuv +1l4ljWo3VDkPCczXIaUi94/o1uNu0sbPRluBP8kjFLhwV8Ph/7mxsofuV5tDYNOZ +8VozA7aHAnJqXBCFZuW6igIH3XaROY9LOe2FD+qxtpqJHp7E+ZNsXXH5SXMQzE2/ +X7eKHA7zvdG4z5GLgmz3GEvq3ZBJUGUe6DCnrxqBXUeLTqyummohGE57LuqzrH9e +tFj/91NaZW9uwN42pxGJSkeZaJ5BDnzO3VYxqvvDnFs/2TFbJiZ065vINpA0PnuA +kGjztaNAm6qXNTXsWo8bdi8TDgIwKxoo5KIlWsxnxKB82tfBovWd+JLaxLlWNKbK +gNb6xc9OmNCyhYkRXhqIsBrpLy6n2yppyYEiCBzPrqwJ2p3Y88yj3dscxiiu0ngF +SroE2pLwCgi1rQY1x9VO4XQrrq6vBvPhi530+cdXdVTnnyBhaOMX4bWXOOjI97xJ +mR/YbOvpgKw9v/na82t2fgKoqIs124iZ21QOiffOIfwwg2AKj8EaWJ9/Era8sZMs +q/g56PWYwT1J85cCMvbNwGOXjusCAwEAAaAAMA0GCSqGSIb3DQEBCwUAA4ICAQBr +ZWJGghyeCUESKu9v9dgQYD4Bq5YviYY7BAjdQsn5/Ti+tqkAKHEn/L+264QJMrTU +muTKg09o7n7r+g4sHmZhT2QEtwGFwaEGvz2zRd4po29WlGQAN5a8yzXIb8b9n+C4 +Z5APeAzjBV9ooqbJCURlgeDP247XTcU8yIvvX0YpLmXYRQ5Na8TTfdBdlD3Y23mD +aPe+2YGkK1Go4iNMoyTEoJ0cj+89WMA9o+BiGcptALYMzKBM26f/vz9hb0J49RDV +MkccL8TKNDJQrLFCIcwpLRw08uZZdqJ/cy01hrtkMkp+SSBQO9qSmalDrTLgs4Pp +3ml88OIsQHMkdBUjUAw5FyAJoik4iXab/sLoMmMJ/lqH0in3dT4KeEh8Xe2iAfNU +97/6lAuZOrn2EPvqBJI/gfjwPj9wTgjlvLpmcjhBvfqkuM/dd7zTmwRoM1klRMC2 +07x2YNoG7cGXjAYk1h5HDYVJqdphgQLTS1nRqmiQ8QnPryqlZcI7T/GgpOOulHce +lgMp6FFlGaQMysCbdYJQk4JMQXjwlj5ms/6rZSkgwmrv6WhCY7ImDj2E/7oE5XJd +BqXHFwOQdCSgQOVH+QgmxceYolI73cMNF3Frq8/IztAOtMyBF0bpRpI1U6a3ALf4 +cK2Dkd/TLg8ZwsVBjPaRrtXlntO6tB+tn3suukCKcg== +-----END CERTIFICATE REQUEST----- diff --git a/tests/integration/test_accept_invalid_certificate/certs/generate_certs.sh b/tests/integration/test_accept_invalid_certificate/certs/generate_certs.sh new file mode 100755 index 00000000000..610a1281f8b --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/certs/generate_certs.sh @@ -0,0 +1,13 @@ +#!/bin/bash + +# 1. Generate CA's private key and self-signed certificate +openssl req -newkey rsa:4096 -x509 -days 3650 -nodes -batch -keyout ca-key.pem -out ca-cert.pem -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=ca" + +# 2. Generate self-signed certificate and private key for using as wrong server certificate (because it's not signed by CA) +openssl req -newkey rsa:4096 -x509 -days 3650 -nodes -batch -keyout self-key.pem -out self-cert.pem -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=server" + +# 3. Generate client's private key and certificate signing request (CSR) +openssl req -newkey rsa:4096 -nodes -batch -keyout client-key.pem -out client-req.pem -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=client" + +# 4. Use CA's private key to sign client's CSR and get back the signed certificate +openssl x509 -req -days 3650 -in client-req.pem -CA ca-cert.pem -CAkey ca-key.pem -CAcreateserial -out client-cert.pem diff --git a/tests/integration/test_accept_invalid_certificate/certs/self-cert.pem b/tests/integration/test_accept_invalid_certificate/certs/self-cert.pem new file mode 100644 index 00000000000..7b3832a3b6b --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/certs/self-cert.pem @@ -0,0 +1,32 @@ +-----BEGIN CERTIFICATE----- +MIIFjTCCA3WgAwIBAgIUFpPVJytX13EWXwaLqg/rOTdu4UgwDQYJKoZIhvcNAQEL +BQAwVjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUxITAfBgNVBAoM +GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEPMA0GA1UEAwwGc2VydmVyMB4XDTI0 +MDcwMTA1MTEyNVoXDTM0MDYyOTA1MTEyNVowVjELMAkGA1UEBhMCUlUxEzARBgNV +BAgMClNvbWUtU3RhdGUxITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0 +ZDEPMA0GA1UEAwwGc2VydmVyMIICIjANBgkqhkiG9w0BAQEFAAOCAg8AMIICCgKC +AgEAsQO7f1Ar0ThZjh00KeuckyJDf/oPQ+67EX2Y82W4/dl8KmHksmIKp2+CJFwg +Y7/c1ti+DjxVXr4tZynzUeDYqujTrAidUMz0zrvjXNeeyoe2hgc2X6YFtVhyEtMQ +6q8VMaqomjuQDIyxUUdt2xHZWyxHPntDrzsmCCDH5Md27CZ4Z5jsSOdgnlrCkjd9 +NtuIzxTlrExe3whIlTL0DURTvAVV/tPoVpVSvHTWCdziaCWm1hG5T6ZCN2W0pBgV +k+UNNtXr1ksr/Ma0Z84ShKjpN+soFOIZ3GDDGY8Xr/74jZo2c4025TpeIPcPBE0Z +b2K24l0UNnQbgm3ryI7NRrZ7psNnhvnWZru4w5Ykr8RKBf3n1AZOt8+dXHNvJQeU +/CUA6pKpc+sIxzAtWtS2ABqdOTJTXj2GyYNsz38dI2uzecUzT7oEtHWq9sr376EO +xMv1wXsUBli8INhPricfojH2/ACnRPysFglAKTx0IpcwqYoV5pGkk0iK1DNkQMv4 +Kf7cD2hLWv9+Fgbrjfu/gNCL0bSJBjaTHngACePsLEQkmGomscDySYXa+PHeB8Ln +QCWsE6MkeDa5g3gUOPV/vC6qv0ZBvQj1p3sOeyIQATDBYX+KheCO+N0ZCCdlT5Sx +thKNDusultu6/lyjGum0c/5oGgoDKTIkibgljekKupBRAykCAwEAAaNTMFEwHQYD +VR0OBBYEFEnPwBjLDPzta3nBrUEH+Gz1nBJRMB8GA1UdIwQYMBaAFEnPwBjLDPzt +a3nBrUEH+Gz1nBJRMA8GA1UdEwEB/wQFMAMBAf8wDQYJKoZIhvcNAQELBQADggIB +AIW829ODEBvBp+vUwt3hbsJdBx9DzwcBBJ+Mew95YVveiosU+3p1HaWCkc4srgOi +hDNbpvfbqBIKVAZzesJbuTEeQ+Aj4M1lcFed9tvidVnmqDgHJApOI61GihT9MDjC +dH6vhcsorE8E5F58FcMJ0QDynVOuCA/gHkS3ucMxBsNKGfYNe0rOGMw23CzHdnz7 +UonOKhrhIQawI3ELLQxo2LWZEmMsluznE4MB+RGdn3k/UzOY9x4bauv4hrHjYvEm +x/zp2iWhZXvvTZ9u9xaHXEzbSU3WFCcXE/l9SDPM7ip9fabUrvaTJkmM190bD9sk +VcEFOAFBVxfx2fvmBIYqvSkVbcWdAT6x8k+MVNfq99/1aIdh+DTB3XUB7FkE/fnU +tXr/1sz0guwfNI/cHvhHisjmbgMhZWeeVV1YZHOJCsK7eq9rGq5Uul8nYrTdYKlN +u0OcrBXYBB6ny7ljCTahyPPmFZh/4Mgr0xROn2HyhTCpDHjnD1M5jT783fvWqu06 +cpOjRIPYNtA/UQAWUZWrTuJTD6Kju9tDI8i1NtMWIXd7w0LDXsXeHpRoDrVvOIPT +P0ivUpt4dFHnwR8Dos4NI/j9mkdGCblUxDsd/7or8gjedOoWjWfGCNlYa72aq0gf +KjKGPISC3BGTNtwRTe+E7k38r+Y6YO/Tnu9RvBdDM3vF +-----END CERTIFICATE----- diff --git a/tests/integration/test_accept_invalid_certificate/certs/self-key.pem b/tests/integration/test_accept_invalid_certificate/certs/self-key.pem new file mode 100644 index 00000000000..10b299a8c0c --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/certs/self-key.pem @@ -0,0 +1,52 @@ +-----BEGIN PRIVATE KEY----- +MIIJQwIBADANBgkqhkiG9w0BAQEFAASCCS0wggkpAgEAAoICAQCxA7t/UCvROFmO +HTQp65yTIkN/+g9D7rsRfZjzZbj92XwqYeSyYgqnb4IkXCBjv9zW2L4OPFVevi1n +KfNR4Niq6NOsCJ1QzPTOu+Nc157Kh7aGBzZfpgW1WHIS0xDqrxUxqqiaO5AMjLFR +R23bEdlbLEc+e0OvOyYIIMfkx3bsJnhnmOxI52CeWsKSN30224jPFOWsTF7fCEiV +MvQNRFO8BVX+0+hWlVK8dNYJ3OJoJabWEblPpkI3ZbSkGBWT5Q021evWSyv8xrRn +zhKEqOk36ygU4hncYMMZjxev/viNmjZzjTblOl4g9w8ETRlvYrbiXRQ2dBuCbevI +js1Gtnumw2eG+dZmu7jDliSvxEoF/efUBk63z51cc28lB5T8JQDqkqlz6wjHMC1a +1LYAGp05MlNePYbJg2zPfx0ja7N5xTNPugS0dar2yvfvoQ7Ey/XBexQGWLwg2E+u +Jx+iMfb8AKdE/KwWCUApPHQilzCpihXmkaSTSIrUM2RAy/gp/twPaEta/34WBuuN ++7+A0IvRtIkGNpMeeAAJ4+wsRCSYaiaxwPJJhdr48d4HwudAJawToyR4NrmDeBQ4 +9X+8Lqq/RkG9CPWnew57IhABMMFhf4qF4I743RkIJ2VPlLG2Eo0O6y6W27r+XKMa +6bRz/mgaCgMpMiSJuCWN6Qq6kFEDKQIDAQABAoICAA4BaZhifipeTFeo395hPKY3 +k8z7EAd5SrvQ4C/XNefVNAnLwmM1dqUAzEnsMB6JqQN8FVgzAIjagI/PtedzOYbB +nTMuC08VUwP4bGjay+mNC1gCF2nsojmoTH5IzOcvVdH0MBrj6OXVS+z5EIFrdsh2 +nGcegaB1DEagXAXcnQAvmwniA5qmaUfzjmmF58EAuwuyGj9OW5icoyw1jTHE3OCH +hjeqCeav3I1PVpAOi1YKL0UeFIzExCsglPEBfCjoz+mCy4IiYydA4Y1FChoM80bw +CDUpc+UH6D02y9JVA317XMuPuhmGS2RNp+vYiHv+Uide40vbq+w0nkSOvUHXC7lo +UXf26sZWHcQP8TxewjZFcajApk03lCAb/u4skt7OQxbDOcvrcdbCsSpZPS16s3CA +OWA9fS7WJDmFx/dhzm4xNaXpz9z5MMkUfnFqgceG3H1leamL66J+eOzlquQ8LdsJ +/wTvCi1SsR3JmqT+TpqlXTJci7A+I/b0Yt6a0s2HunQ5LA1Xomffhn/00IxUeqyi +4MKh7wnoCaRlwZVORgQfKnosSkbkixDabwr9BMnMryE+NdQI3JKCvchdB0idyk1z +igTgZYpaxZg9hwTlOITGYSsprN7tiwjNQG+Kivatn1NxnxhD8lpDoUqluR7mZs1e +CwOLrB+irCULr7tcRYtxAoIBAQDFV7E1Oz8j8NkvkLztIFJ0ej7zEyNp3HUJb4bd +SRvF2AL6sqSw0xVLK8HbrlfPo1Nex8nZUfY3v2DmPS4+S3mrtk+8XoRkvNn0VdL4 +FqLfygYsURI8xVp4zTsAmR+7grPJbb/u3XG7CBY3T4XGHSbMmYAu7+W3A2t8Q+0U +fptju7Gc9gEACcSfap45CzEnFbabxNq3FJJ9V92N/xVQcOzNLxPxgn6tVhPeQnXV +0WHJQdxB/8DivgOObiTCB7on9tNbSHisTYtGwn6Kz0961ewYU7Xj4epj7A35nWhI +E10FDAf3T7cQa9ubSFZwOiSe0voAbiWbN3/M+zgkZDp2+wlnAoIBAQDloTtzgQaA +//VPNhB2d7iOgAl5cUSWqY0HLJobw1ETsdbkGRTgZrFXe4XlXwHB4S+zVEMmyR3p +HY0sYoAdpmd3Ggo4efJEx9IxkD6UZ71we02szvt4MdV5mwjHOAQfEJffktVkeqEJ +gdekSmaHTeryBf3ESpefq+OH1qaj/IjivySLI4c8fPcxNg0nMmDsdMzRaDOHDDa2 +nUpRBeMFkwiiPJpALPPp/4be0d9mrahPd4zBoBieZwaP6Zhwksi0NaM4WKEfx4Fd +TzH0K7UfLDMJ13wuTsNPzjPlDA+R4E/O3qVZxtHzuqc7fCksE7inYZghIRSjQHmv +mmLPaFdaIWTvAoIBAQC50BQfVyXgpYsaamI5tFFvMxvKrJ6PsvDQbZ7DBW+NHhPQ +Eh1YEajfGthyDPrJm7OIbv30DmSToTOQFtNP/KkXsZXqe+SVJezjT1xSFJ5PsV2W +pM9TQVAqprODoKpsWGkIpLowI3pO7RJeChafsBs0UuSLTX8yj5nJa/vzbDLL+ZMH +g+r1Kj1lDXLlq1AgySUkDnsSH+C+/Ni0NxbOo0WU/Vr98C1u8LnMcS+aoIg+hDDs +Q/BOjzOyDKu3tqblslbYgKh/NJSLj96uvPzNYFfpoAiIBdQ5Oa7+9eOW2gSgF+/4 +QarRX31r/PTGvzh0A5s6hHwMZnmDfJgkS+tw1HbJAoIBAQDfauQBGYROdrsieqPB +3/H9E64kLoRhYk4IsOiwInT8CIfgHlcljkT1rQU1ih6XmJVkkZNUZVFowovcp2mn +b4QzaW6QTHXfXduwxXtdwSGcExwcCra41yOw0ZrU2jeIUgiySpa80GNmcKJzAsCG +3NS+7f1dwoWLoR4byEkSPF6i3v6bMdIelCeeBcz+WhHbfjjZddzMhKGnQ5pPAr9D +aS95SiV6U0JuZWO/38rQiB2+U77rlOzfEbZcxxa+UNJeI+HOrMz3spC3TJp5yxzK +W6K87J+yHogBWcs8C2QKtohv6khO9RbJYM3dHoygif+bO2Vctmbx/4VlyXxh+E8l +LmCZAoIBAFR+dd7cW0pDkOsPsGZbWwDPNRwo2lxmqf2B5w9zoGNFoc4qV5k9tJfu +kPcwol6h2P4s2olYTgYC6hjPmE17U1TyP2DFeOw5IdxzNqy7dZndmaTs4qFLnDv3 +HIbkX42p7v++EU9eKAhYKJzJpuianMZ+v5azV1wNS5TDuN2TFULyIi+1UEw0ONuJ +rA97idb2PrO2sT/eBj02uiOcPhZH6CHSMxDp7QG+S1AVZBgutTQQdeAsb6pE/iW0 +u3IzGNxTYjTmF+u7iNdOPiaTFpdnltDviADESMJGH7PIk7r7PXMI9ID+LbBFGJ7Q +duW6PGuZ9JMBbxbNNvYayVgPeAeUsUI= +-----END PRIVATE KEY----- diff --git a/tests/integration/test_accept_invalid_certificate/configs/ssl_config.xml b/tests/integration/test_accept_invalid_certificate/configs/ssl_config.xml new file mode 100644 index 00000000000..a1dd1a24693 --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/configs/ssl_config.xml @@ -0,0 +1,17 @@ + + + 9440 + + + + + /etc/clickhouse-server/config.d/self-cert.pem + /etc/clickhouse-server/config.d/self-key.pem + /etc/clickhouse-server/config.d/ca-cert.pem + none + + + diff --git a/tests/integration/test_accept_invalid_certificate/test.py b/tests/integration/test_accept_invalid_certificate/test.py new file mode 100644 index 00000000000..87229d75f90 --- /dev/null +++ b/tests/integration/test_accept_invalid_certificate/test.py @@ -0,0 +1,92 @@ +import pytest +from helpers.client import Client +from helpers.cluster import ClickHouseCluster +import os.path +from os import remove + + +SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__)) +MAX_RETRY = 5 + +cluster = ClickHouseCluster(__file__) +instance = cluster.add_instance( + "node", + main_configs=[ + "configs/ssl_config.xml", + "certs/self-key.pem", + "certs/self-cert.pem", + "certs/ca-cert.pem", + ], +) + + +@pytest.fixture(scope="module", autouse=True) +def started_cluster(): + try: + cluster.start() + yield cluster + + finally: + cluster.shutdown() + + +config_default = """ +""" + +config_accept = """ + 1 +""" + +config_connection_accept = """ + + + {ip_address} + 1 + + +""" + + +def execute_query_native(node, query, config): + config_path = f"{SCRIPT_DIR}/configs/client.xml" + + file = open(config_path, "w") + file.write(config) + file.close() + + client = Client( + node.ip_address, + 9440, + command=cluster.client_bin_path, + secure=True, + config=config_path, + ) + + try: + result = client.query(query) + remove(config_path) + return result + except: + remove(config_path) + raise + + +def test_default(): + with pytest.raises(Exception) as err: + execute_query_native(instance, "SELECT 1", config_default) + assert "certificate verify failed" in str(err.value) + + +def test_accept(): + assert execute_query_native(instance, "SELECT 1", config_accept) == "1\n" + + +def test_connection_accept(): + assert ( + execute_query_native( + instance, + "SELECT 1", + config_connection_accept.format(ip_address=f"{instance.ip_address}"), + ) + == "1\n" + ) diff --git a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/__init__.py b/tests/integration/test_analyzer_compatibility/__init__.py similarity index 100% rename from tests/integration/test_host_regexp_multiple_ptr_records_concurrent/__init__.py rename to tests/integration/test_analyzer_compatibility/__init__.py diff --git a/tests/integration/test_analyzer_compatibility/configs/remote_servers.xml b/tests/integration/test_analyzer_compatibility/configs/remote_servers.xml new file mode 100644 index 00000000000..0a50dab7fd3 --- /dev/null +++ b/tests/integration/test_analyzer_compatibility/configs/remote_servers.xml @@ -0,0 +1,17 @@ + + + + + true + + current + 9000 + + + backward + 9000 + + + + + diff --git a/tests/integration/test_analyzer_compatibility/test.py b/tests/integration/test_analyzer_compatibility/test.py new file mode 100644 index 00000000000..d4ded420c61 --- /dev/null +++ b/tests/integration/test_analyzer_compatibility/test.py @@ -0,0 +1,100 @@ +import uuid + +import pytest +from helpers.cluster import ClickHouseCluster +from helpers.test_tools import TSV + +CLICKHOUSE_MAX_VERSION_WITH_ANALYZER_DISABLED_BY_DEFAULT = "24.2" + +cluster = ClickHouseCluster(__file__) +# Here analyzer is enabled by default +current = cluster.add_instance( + "current", + main_configs=["configs/remote_servers.xml"], +) +# Here analyzer is disabled by default +backward = cluster.add_instance( + "backward", + use_old_analyzer=True, + main_configs=["configs/remote_servers.xml"], + image="clickhouse/clickhouse-server", + tag=CLICKHOUSE_MAX_VERSION_WITH_ANALYZER_DISABLED_BY_DEFAULT, + with_installed_binary=True, +) + + +@pytest.fixture(scope="module") +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +def test_two_new_versions(start_cluster): + # Two new versions (both know about the analyzer) + # One have it enabled by default, another one - disabled. + + current.query("SYSTEM FLUSH LOGS") + backward.query("SYSTEM FLUSH LOGS") + + query_id = str(uuid.uuid4()) + current.query( + "SELECT * FROM clusterAllReplicas('test_cluster_mixed', system.tables);", + query_id=query_id, + ) + + current.query("SYSTEM FLUSH LOGS") + backward.query("SYSTEM FLUSH LOGS") + + assert ( + current.query( + """ +SELECT hostname() AS h, getSetting('allow_experimental_analyzer') +FROM clusterAllReplicas('test_cluster_mixed', system.one) +ORDER BY h;""" + ) + == TSV([["backward", "true"], ["current", "true"]]) + ) + + # Should be enabled everywhere + analyzer_enabled = current.query( + f""" +SELECT +DISTINCT Settings['allow_experimental_analyzer'] +FROM clusterAllReplicas('test_cluster_mixed', system.query_log) +WHERE initial_query_id = '{query_id}';""" + ) + + assert TSV(analyzer_enabled) == TSV("1") + + query_id = str(uuid.uuid4()) + backward.query( + "SELECT * FROM clusterAllReplicas('test_cluster_mixed', system.tables)", + query_id=query_id, + ) + + current.query("SYSTEM FLUSH LOGS") + backward.query("SYSTEM FLUSH LOGS") + + assert ( + backward.query( + """ +SELECT hostname() AS h, getSetting('allow_experimental_analyzer') +FROM clusterAllReplicas('test_cluster_mixed', system.one) +ORDER BY h;""" + ) + == TSV([["backward", "false"], ["current", "false"]]) + ) + + # Should be disabled everywhere + analyzer_enabled = backward.query( + f""" +SELECT +DISTINCT Settings['allow_experimental_analyzer'] +FROM clusterAllReplicas('test_cluster_mixed', system.query_log) +WHERE initial_query_id = '{query_id}';""" + ) + + assert TSV(analyzer_enabled) == TSV("0") diff --git a/tests/integration/test_asynchronous_metric_jemalloc_profile_active/test.py b/tests/integration/test_asynchronous_metric_jemalloc_profile_active/test.py index a8f4ab05888..b3769a61b3f 100644 --- a/tests/integration/test_asynchronous_metric_jemalloc_profile_active/test.py +++ b/tests/integration/test_asynchronous_metric_jemalloc_profile_active/test.py @@ -7,7 +7,6 @@ cluster = ClickHouseCluster(__file__) node1 = cluster.add_instance( "node1", main_configs=["configs/asynchronous_metrics_update_period_s.xml"], - env_variables={"MALLOC_CONF": "background_thread:true,prof:true"}, ) @@ -29,26 +28,11 @@ def test_asynchronous_metric_jemalloc_profile_active(started_cluster): if node1.is_built_with_sanitizer(): pytest.skip("Disabled for sanitizers") - res_o = node1.query( + res = node1.query( "SELECT * FROM system.asynchronous_metrics WHERE metric ILIKE '%jemalloc.prof.active%' FORMAT Vertical;" ) assert ( - res_o - == """Row 1: -────── -metric: jemalloc.prof.active -value: 1 -description: An internal metric of the low-level memory allocator (jemalloc). See https://jemalloc.net/jemalloc.3.html -""" - ) - # disable - node1.query("SYSTEM JEMALLOC DISABLE PROFILE") - time.sleep(5) - res_t = node1.query( - "SELECT * FROM system.asynchronous_metrics WHERE metric ILIKE '%jemalloc.prof.active%' FORMAT Vertical;" - ) - assert ( - res_t + res == """Row 1: ────── metric: jemalloc.prof.active @@ -58,16 +42,31 @@ description: An internal metric of the low-level memory allocator (jemalloc). Se ) # enable node1.query("SYSTEM JEMALLOC ENABLE PROFILE") - time.sleep(5) - res_f = node1.query( + node1.query("SYSTEM RELOAD ASYNCHRONOUS METRICS") + res = node1.query( "SELECT * FROM system.asynchronous_metrics WHERE metric ILIKE '%jemalloc.prof.active%' FORMAT Vertical;" ) assert ( - res_f + res == """Row 1: ────── metric: jemalloc.prof.active value: 1 description: An internal metric of the low-level memory allocator (jemalloc). See https://jemalloc.net/jemalloc.3.html +""" + ) + # disable + node1.query("SYSTEM JEMALLOC DISABLE PROFILE") + node1.query("SYSTEM RELOAD ASYNCHRONOUS METRICS") + res = node1.query( + "SELECT * FROM system.asynchronous_metrics WHERE metric ILIKE '%jemalloc.prof.active%' FORMAT Vertical;" + ) + assert ( + res + == """Row 1: +────── +metric: jemalloc.prof.active +value: 0 +description: An internal metric of the low-level memory allocator (jemalloc). See https://jemalloc.net/jemalloc.3.html """ ) diff --git a/tests/integration/test_backup_restore_new/test.py b/tests/integration/test_backup_restore_new/test.py index d8662fad011..4806625f3f0 100644 --- a/tests/integration/test_backup_restore_new/test.py +++ b/tests/integration/test_backup_restore_new/test.py @@ -1,9 +1,10 @@ -import pytest import glob -import re -import random import os.path +import pytest +import random +import re import sys +import uuid from collections import namedtuple from helpers.cluster import ClickHouseCluster from helpers.test_tools import assert_eq_with_retry, TSV @@ -538,7 +539,8 @@ def test_backup_not_found_or_already_exists(): def test_file_engine(): - backup_name = f"File('/backups/file/')" + id = uuid.uuid4() + backup_name = f"File('/backups/file/{id}/')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -549,6 +551,7 @@ def test_file_engine(): instance.query(f"RESTORE TABLE test.table FROM {backup_name}") assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" + instance.query("DROP TABLE test.table") def test_database(): @@ -565,7 +568,8 @@ def test_database(): def test_zip_archive(): - backup_name = f"Disk('backups', 'archive.zip')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_{id}.zip')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -578,10 +582,12 @@ def test_zip_archive(): instance.query(f"RESTORE TABLE test.table FROM {backup_name}") assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" + instance.query("DROP TABLE test.table") def test_zip_archive_with_settings(): - backup_name = f"Disk('backups', 'archive_with_settings.zip')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_with_settings_{id}.zip')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -596,10 +602,12 @@ def test_zip_archive_with_settings(): f"RESTORE TABLE test.table FROM {backup_name} SETTINGS password='qwerty'" ) assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" + instance.query("DROP TABLE test.table") def test_zip_archive_with_bad_compression_method(): - backup_name = f"Disk('backups', 'archive_with_bad_compression_method.zip')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_with_bad_compression_method_{id}.zip')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -617,7 +625,8 @@ def test_zip_archive_with_bad_compression_method(): def test_tar_archive(): - backup_name = f"Disk('backups', 'archive.tar')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_{id}.tar')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -630,10 +639,12 @@ def test_tar_archive(): instance.query(f"RESTORE TABLE test.table FROM {backup_name}") assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" + instance.query("DROP TABLE test.table") def test_tar_bz2_archive(): - backup_name = f"Disk('backups', 'archive.tar.bz2')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_{id}.tar.bz2')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -649,7 +660,8 @@ def test_tar_bz2_archive(): def test_tar_gz_archive(): - backup_name = f"Disk('backups', 'archive.tar.gz')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_{id}.tar.gz')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -665,7 +677,8 @@ def test_tar_gz_archive(): def test_tar_lzma_archive(): - backup_name = f"Disk('backups', 'archive.tar.lzma')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_{id}.tar.lzma')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -681,7 +694,8 @@ def test_tar_lzma_archive(): def test_tar_zst_archive(): - backup_name = f"Disk('backups', 'archive.tar.zst')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_{id}.tar.zst')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -697,7 +711,8 @@ def test_tar_zst_archive(): def test_tar_xz_archive(): - backup_name = f"Disk('backups', 'archive.tar.xz')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_{id}.tar.xz')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -713,7 +728,8 @@ def test_tar_xz_archive(): def test_tar_archive_with_password(): - backup_name = f"Disk('backups', 'archive_with_password.tar')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_with_password_{id}.tar')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -731,7 +747,8 @@ def test_tar_archive_with_password(): def test_tar_archive_with_bad_compression_method(): - backup_name = f"Disk('backups', 'archive_with_bad_compression_method.tar')" + id = uuid.uuid4() + backup_name = f"Disk('backups', 'archive_with_bad_compression_method_{id}.tar')" create_and_fill_table() assert instance.query("SELECT count(), sum(x) FROM test.table") == "100\t4950\n" @@ -1220,6 +1237,10 @@ def test_system_users_required_privileges(): assert instance.query("SHOW CREATE ROLE r1") == "CREATE ROLE r1\n" assert instance.query("SHOW GRANTS FOR r1") == "" + instance.query("DROP USER u1") + instance.query("DROP ROLE r1") + instance.query("DROP USER u2") + def test_system_users_async(): instance.query("CREATE USER u1 IDENTIFIED BY 'qwe123' SETTINGS custom_c = 3") @@ -1412,6 +1433,8 @@ def test_system_functions(): assert instance.query("SELECT number, parity_str(number) FROM numbers(3)") == TSV( [[0, "even"], [1, "odd"], [2, "even"]] ) + instance.query("DROP FUNCTION linear_equation") + instance.query("DROP FUNCTION parity_str") def test_backup_partition(): diff --git a/tests/integration/test_backup_restore_on_cluster/test_concurrency.py b/tests/integration/test_backup_restore_on_cluster/test_concurrency.py index c08f3c9c242..0e381d48fb1 100644 --- a/tests/integration/test_backup_restore_on_cluster/test_concurrency.py +++ b/tests/integration/test_backup_restore_on_cluster/test_concurrency.py @@ -276,15 +276,10 @@ def test_create_or_drop_tables_during_backup(db_engine, table_engine): for node in nodes: assert_eq_with_retry( node, - f"SELECT status from system.backups WHERE id IN {ids_list} AND (status == 'CREATING_BACKUP')", - "", - ) - - for node in nodes: - assert_eq_with_retry( - node, - f"SELECT status, error from system.backups WHERE id IN {ids_list} AND (status == 'BACKUP_FAILED')", + f"SELECT status, error from system.backups " + f"WHERE id IN {ids_list} AND ((status == 'CREATING_BACKUP') OR (status == 'BACKUP_FAILED'))", "", + retry_count=100, ) backup_names = {} diff --git a/tests/integration/test_backup_restore_s3/configs/disk_s3.xml b/tests/integration/test_backup_restore_s3/configs/disk_s3.xml index 45a1e17b039..7ac1a052c30 100644 --- a/tests/integration/test_backup_restore_s3/configs/disk_s3.xml +++ b/tests/integration/test_backup_restore_s3/configs/disk_s3.xml @@ -21,6 +21,13 @@ minio123 33554432 + + s3_plain_rewritable + http://minio1:9001/root/data/disks/disk_s3_plain_rewritable/ + minio + minio123 + 33554432 + cache disk_s3 @@ -37,6 +44,20 @@ + + +

+ disk_s3_plain +
+ + + + +
+ disk_s3_plain_rewritable +
+
+
@@ -57,6 +78,7 @@ default disk_s3 disk_s3_plain + disk_s3_plain_rewritable disk_s3_cache disk_s3_other_bucket diff --git a/tests/integration/test_backup_restore_s3/test.py b/tests/integration/test_backup_restore_s3/test.py index d53335000a6..381268ce7fe 100644 --- a/tests/integration/test_backup_restore_s3/test.py +++ b/tests/integration/test_backup_restore_s3/test.py @@ -254,6 +254,7 @@ def check_system_tables(backup_query_id=None): ("disk_s3_cache", "ObjectStorage", "S3", "Local"), ("disk_s3_other_bucket", "ObjectStorage", "S3", "Local"), ("disk_s3_plain", "ObjectStorage", "S3", "Plain"), + ("disk_s3_plain_rewritable", "ObjectStorage", "S3", "PlainRewritable"), ("disk_s3_restricted_user", "ObjectStorage", "S3", "Local"), ) assert len(expected_disks) == len(disks) @@ -419,25 +420,15 @@ def test_backup_to_s3_multipart(): assert "ReadBufferFromS3RequestsErrors" not in restore_events -def test_backup_to_s3_native_copy(): - storage_policy = "policy_s3" - backup_name = new_backup_name() - backup_destination = ( - f"S3('http://minio1:9001/root/data/backups/{backup_name}', 'minio', 'minio123')" - ) - (backup_events, restore_events) = check_backup_and_restore( - storage_policy, backup_destination - ) - # single part upload - assert backup_events["S3CopyObject"] > 0 - assert restore_events["S3CopyObject"] > 0 - assert node.contains_in_log( - f"copyS3File: Single operation copy has completed. Bucket: root, Key: data/backups/{backup_name}" - ) - - -def test_backup_to_s3_native_copy_other_bucket(): - storage_policy = "policy_s3_other_bucket" +@pytest.mark.parametrize( + "storage_policy", + [ + "policy_s3", + "policy_s3_other_bucket", + "policy_s3_plain_rewritable", + ], +) +def test_backup_to_s3_native_copy(storage_policy): backup_name = new_backup_name() backup_destination = ( f"S3('http://minio1:9001/root/data/backups/{backup_name}', 'minio', 'minio123')" diff --git a/tests/integration/test_backward_compatibility/test_functions.py b/tests/integration/test_backward_compatibility/test_functions.py index 758dda655da..3231fb87f33 100644 --- a/tests/integration/test_backward_compatibility/test_functions.py +++ b/tests/integration/test_backward_compatibility/test_functions.py @@ -75,11 +75,15 @@ def test_aggregate_states(start_cluster): except QueryRuntimeException as e: error_message = str(e) allowed_errors = [ - "NUMBER_OF_ARGUMENTS_DOESNT_MATCH", "ILLEGAL_TYPE_OF_ARGUMENT", # sequenceNextNode() and friends "UNKNOWN_AGGREGATE_FUNCTION", # Function X takes exactly one parameter: + "NUMBER_OF_ARGUMENTS_DOESNT_MATCH", + # Function X takes at least one argument + "TOO_FEW_ARGUMENTS_FOR_FUNCTION", + # Function X accepts at most 3 arguments, Y given + "TOO_MANY_ARGUMENTS_FOR_FUNCTION", # The function 'X' can only be used as a window function "BAD_ARGUMENTS", # aggThrow @@ -130,10 +134,13 @@ def test_string_functions(start_cluster): functions = map(lambda x: x.strip(), functions) excludes = [ + # The argument of this function is not a seed, but an arbitrary expression needed for bypassing common subexpression elimination. "rand", "rand64", "randConstant", + "randCanonical", "generateUUIDv4", + "generateULID", # Syntax error otherwise "position", "substring", @@ -153,6 +160,18 @@ def test_string_functions(start_cluster): "tryBase64Decode", # Removed in 23.9 "meiliMatch", + # These functions require more than one argument. + "parseDateTimeInJodaSyntaxOrZero", + "parseDateTimeInJodaSyntaxOrNull", + "parseDateTimeOrNull", + "parseDateTimeOrZero", + "parseDateTime", + # The argument is effectively a disk name (and we don't have one with name foo) + "filesystemUnreserved", + "filesystemCapacity", + "filesystemAvailable", + # Exclude it for now. Looks like the result depends on the build type. + "farmHash64", ] functions = filter(lambda x: x not in excludes, functions) @@ -181,9 +200,7 @@ def test_string_functions(start_cluster): "Should start with ", # POINT/POLYGON/... "Cannot read input: expected a digit but got something else:", # ErrorCodes - "NUMBER_OF_ARGUMENTS_DOESNT_MATCH", "ILLEGAL_TYPE_OF_ARGUMENT", - "TOO_FEW_ARGUMENTS_FOR_FUNCTION", "DICTIONARIES_WAS_NOT_LOADED", "CANNOT_PARSE_UUID", "CANNOT_PARSE_DOMAIN_VALUE_FROM_STRING", @@ -203,8 +220,16 @@ def test_string_functions(start_cluster): "CANNOT_PARSE_TEXT", "CANNOT_PARSE_DATETIME", # Function X takes exactly one parameter: + "NUMBER_OF_ARGUMENTS_DOESNT_MATCH", + # Function X takes at least one argument + "TOO_FEW_ARGUMENTS_FOR_FUNCTION", + # Function X accepts at most 3 arguments, Y given + "TOO_MANY_ARGUMENTS_FOR_FUNCTION", # The function 'X' can only be used as a window function "BAD_ARGUMENTS", + # String foo is obviously not a valid IP address. + "CANNOT_PARSE_IPV4", + "CANNOT_PARSE_IPV6", ] if any(map(lambda x: x in error_message, allowed_errors)): logging.info("Skipping %s", function) diff --git a/tests/integration/test_broken_part_during_merge/test.py b/tests/integration/test_broken_part_during_merge/test.py index 19c22201fb0..0ba7beeb1fd 100644 --- a/tests/integration/test_broken_part_during_merge/test.py +++ b/tests/integration/test_broken_part_during_merge/test.py @@ -54,7 +54,7 @@ def test_merge_and_part_corruption(started_cluster): with Pool(1) as p: def optimize_with_delay(x): - node1.query("OPTIMIZE TABLE replicated_mt FINAL", timeout=30) + node1.query("OPTIMIZE TABLE replicated_mt FINAL", timeout=120) # corrupt part after merge already assigned, but not started res_opt = p.apply_async(optimize_with_delay, (1,)) @@ -70,7 +70,7 @@ def test_merge_and_part_corruption(started_cluster): node1.query( "ALTER TABLE replicated_mt UPDATE value = 7 WHERE 1", settings={"mutations_sync": 2}, - timeout=30, + timeout=120, ) assert node1.query("SELECT sum(value) FROM replicated_mt") == "2100000\n" diff --git a/tests/integration/test_broken_projections/test.py b/tests/integration/test_broken_projections/test.py index e198f98e4c5..a565f47449f 100644 --- a/tests/integration/test_broken_projections/test.py +++ b/tests/integration/test_broken_projections/test.py @@ -433,6 +433,7 @@ def test_broken_ignored_replicated(cluster): check(node, table_name, 1) create_table(node, table_name2, 2, table_name) + node.query(f"system sync replica {table_name}") check(node, table_name2, 1) break_projection(node, table_name, "proj1", "all_0_0_0", "data") diff --git a/tests/integration/test_config_substitutions/configs/000-server_overrides.xml b/tests/integration/test_config_substitutions/configs/000-server_overrides.xml new file mode 100644 index 00000000000..9335f663d68 --- /dev/null +++ b/tests/integration/test_config_substitutions/configs/000-server_overrides.xml @@ -0,0 +1,3 @@ + + 10000 + diff --git a/tests/integration/test_config_substitutions/configs/000-config_with_env_subst.xml b/tests/integration/test_config_substitutions/configs/000-users_with_env_subst.xml similarity index 100% rename from tests/integration/test_config_substitutions/configs/000-config_with_env_subst.xml rename to tests/integration/test_config_substitutions/configs/000-users_with_env_subst.xml diff --git a/tests/integration/test_config_substitutions/configs/010-server_with_env_subst.xml b/tests/integration/test_config_substitutions/configs/010-server_with_env_subst.xml new file mode 100644 index 00000000000..ea91f066a21 --- /dev/null +++ b/tests/integration/test_config_substitutions/configs/010-server_with_env_subst.xml @@ -0,0 +1,3 @@ + + + diff --git a/tests/integration/test_config_substitutions/test.py b/tests/integration/test_config_substitutions/test.py index faceab6fbcd..124dbcaedf7 100644 --- a/tests/integration/test_config_substitutions/test.py +++ b/tests/integration/test_config_substitutions/test.py @@ -39,9 +39,13 @@ node6 = cluster.add_instance( node7 = cluster.add_instance( "node7", user_configs=[ - "configs/000-config_with_env_subst.xml", + "configs/000-users_with_env_subst.xml", "configs/010-env_subst_override.xml", ], + main_configs=[ + "configs/000-server_overrides.xml", + "configs/010-server_with_env_subst.xml", + ], env_variables={ # overridden with 424242 "MAX_QUERY_SIZE": "121212", @@ -126,9 +130,9 @@ def test_config(start_cluster): ) -def test_config_invalid_overrides(start_cluster): +def test_config_from_env_overrides(start_cluster): node7.replace_config( - "/etc/clickhouse-server/users.d/000-config_with_env_subst.xml", + "/etc/clickhouse-server/users.d/000-users_with_env_subst.xml", """ @@ -156,7 +160,7 @@ def test_config_invalid_overrides(start_cluster): ): node7.query("SYSTEM RELOAD CONFIG") node7.replace_config( - "/etc/clickhouse-server/users.d/000-config_with_env_subst.xml", + "/etc/clickhouse-server/users.d/000-users_with_env_subst.xml", """ @@ -181,6 +185,24 @@ def test_config_invalid_overrides(start_cluster): node7.query("SYSTEM RELOAD CONFIG") +def test_config_merge_from_env_overrides(start_cluster): + assert ( + node7.query( + "SELECT value FROM system.server_settings WHERE name='max_thread_pool_size'" + ) + == "10000\n" + ) + node7.replace_config( + "/etc/clickhouse-server/config.d/010-server_with_env_subst.xml", + """ + + 9000 + +""", + ) + node7.query("SYSTEM RELOAD CONFIG") + + def test_include_config(start_cluster): # assert node4.query("select 1") diff --git a/tests/integration/test_crash_log/test.py b/tests/integration/test_crash_log/test.py index a5b82039a84..5a63e6ca6a7 100644 --- a/tests/integration/test_crash_log/test.py +++ b/tests/integration/test_crash_log/test.py @@ -60,6 +60,13 @@ def test_pkill(started_node): def test_pkill_query_log(started_node): + if ( + started_node.is_built_with_thread_sanitizer() + or started_node.is_built_with_address_sanitizer() + or started_node.is_built_with_memory_sanitizer() + ): + pytest.skip("doesn't fit in timeouts for stacktrace generation") + for signal in ["SEGV", "4"]: # force create query_log if it was not created started_node.query("SYSTEM FLUSH LOGS") diff --git a/tests/integration/test_dictionaries_replace/test.py b/tests/integration/test_dictionaries_replace/test.py index bf406f46cb1..efb65c15c49 100644 --- a/tests/integration/test_dictionaries_replace/test.py +++ b/tests/integration/test_dictionaries_replace/test.py @@ -134,3 +134,5 @@ def test_create_or_replace(database, instance_to_create_dictionary, instances_to expected_result = TSV([[0, 1], [5, 26], [7, 50], [11, 0]]) assert instance.query(select_query) == expected_result assert instance.query(select_query, user="dictget_user") == expected_result + + instance_to_create_dictionary.query(f"DROP DICTIONARY IF EXISTS {database}.dict") diff --git a/tests/integration/test_disk_configuration/test.py b/tests/integration/test_disk_configuration/test.py index c003ff85755..afc5303298c 100644 --- a/tests/integration/test_disk_configuration/test.py +++ b/tests/integration/test_disk_configuration/test.py @@ -208,13 +208,21 @@ def test_merge_tree_custom_disk_setting(start_cluster): secret_access_key='minio123'); """ ) - count = len(list(minio.list_objects(cluster.minio_bucket, "data/", recursive=True))) + + list1 = list(minio.list_objects(cluster.minio_bucket, "data/", recursive=True)) + count1 = len(list1) + node1.query(f"INSERT INTO {TABLE_NAME}_3 SELECT number FROM numbers(100)") assert int(node1.query(f"SELECT count() FROM {TABLE_NAME}_3")) == 100 - assert ( - len(list(minio.list_objects(cluster.minio_bucket, "data/", recursive=True))) - == count - ) + + list2 = list(minio.list_objects(cluster.minio_bucket, "data/", recursive=True)) + count2 = len(list2) + + if count1 != count2: + print("list1: ", list1) + print("list2: ", list2) + + assert count1 == count2 assert ( len(list(minio.list_objects(cluster.minio_bucket, "data2/", recursive=True))) > 0 diff --git a/tests/integration/test_disk_types/test.py b/tests/integration/test_disk_types/test.py index 1cc5048eb69..a5e2456ef4f 100644 --- a/tests/integration/test_disk_types/test.py +++ b/tests/integration/test_disk_types/test.py @@ -19,9 +19,9 @@ def cluster(): cluster = ClickHouseCluster(__file__) cluster.add_instance( "node", - main_configs=["configs/storage_arm.xml"] - if is_arm() - else ["configs/storage_amd.xml"], + main_configs=( + ["configs/storage_arm.xml"] if is_arm() else ["configs/storage_amd.xml"] + ), with_minio=True, with_hdfs=not is_arm(), ) diff --git a/tests/integration/test_disks_app_func/test.py b/tests/integration/test_disks_app_func/test.py index 97d5da787cd..56ea5c8846a 100644 --- a/tests/integration/test_disks_app_func/test.py +++ b/tests/integration/test_disks_app_func/test.py @@ -9,7 +9,9 @@ def started_cluster(): try: cluster = ClickHouseCluster(__file__) cluster.add_instance( - "disks_app_test", main_configs=["config.xml"], with_minio=True + "disks_app_test", + main_configs=["config.xml"], + with_minio=True, ) cluster.start() @@ -47,12 +49,18 @@ def test_disks_app_func_ld(started_cluster): source = cluster.instances["disks_app_test"] out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "list-disks"] + ["/usr/bin/clickhouse", "disks", "--save-logs", "--query", "list-disks"] ) - disks = out.split("\n") + disks = list( + sorted( + map( + lambda x: x.split(":")[0], filter(lambda x: len(x) > 1, out.split("\n")) + ) + ) + ) - assert disks[0] == "default" and disks[1] == "test1" and disks[2] == "test2" + assert disks[:4] == ["default", "local", "test1", "test2"] def test_disks_app_func_ls(started_cluster): @@ -61,7 +69,15 @@ def test_disks_app_func_ls(started_cluster): init_data(source) out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test1", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test1", + "--query", + "list .", + ] ) files = out.split("\n") @@ -75,9 +91,8 @@ def test_disks_app_func_ls(started_cluster): "--save-logs", "--disk", "test1", - "list", - ".", - "--recursive", + "--query", + "list . --recursive", ] ) @@ -102,8 +117,8 @@ def test_disks_app_func_cp(started_cluster): "--save-logs", "--disk", "test1", - "write", - "path1", + "--query", + "'write path1'", ] ), ] @@ -113,18 +128,21 @@ def test_disks_app_func_cp(started_cluster): [ "/usr/bin/clickhouse", "disks", - "copy", - "--disk-from", - "test1", - "--disk-to", - "test2", - ".", - ".", + "--query", + "copy --recursive --disk-from test1 --disk-to test2 . .", ] ) out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test2", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test2", + "--query", + "list .", + ] ) assert "path1" in out @@ -136,8 +154,8 @@ def test_disks_app_func_cp(started_cluster): "--save-logs", "--disk", "test2", - "remove", - "path1", + "--query", + "remove path1", ] ) @@ -148,21 +166,37 @@ def test_disks_app_func_cp(started_cluster): "--save-logs", "--disk", "test1", - "remove", - "path1", + "--query", + "remove path1", ] ) # alesapin: Why we need list one more time? # kssenii: it is an assertion that the file is indeed deleted out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test2", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test2", + "--query", + "list .", + ] ) assert "path1" not in out out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test1", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test1", + "--query", + "list .", + ] ) assert "path1" not in out @@ -177,14 +211,13 @@ def test_disks_app_func_ln(started_cluster): [ "/usr/bin/clickhouse", "disks", - "link", - "data/default/test_table", - "data/default/z_tester", + "--query", + "link data/default/test_table data/default/z_tester", ] ) out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "list", "data/default/"] + ["/usr/bin/clickhouse", "disks", "--save-logs", "--query", "list data/default/"] ) files = out.split("\n") @@ -209,15 +242,23 @@ def test_disks_app_func_rm(started_cluster): "--save-logs", "--disk", "test2", - "write", - "path3", + "--query", + "'write path3'", ] ), ] ) out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test2", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test2", + "--query", + "list .", + ] ) assert "path3" in out @@ -229,13 +270,21 @@ def test_disks_app_func_rm(started_cluster): "--save-logs", "--disk", "test2", - "remove", - "path3", + "--query", + "remove path3", ] ) out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test2", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test2", + "--query", + "list .", + ] ) assert "path3" not in out @@ -247,7 +296,15 @@ def test_disks_app_func_mv(started_cluster): init_data(source) out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test1", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test1", + "--query", + "list .", + ] ) files = out.split("\n") @@ -260,14 +317,21 @@ def test_disks_app_func_mv(started_cluster): "disks", "--disk", "test1", - "move", - "store", - "old_store", + "--query", + "move store old_store", ] ) out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test1", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test1", + "--query", + "list .", + ] ) files = out.split("\n") @@ -290,8 +354,8 @@ def test_disks_app_func_read_write(started_cluster): "--save-logs", "--disk", "test1", - "write", - "5.txt", + "--query", + "'write 5.txt'", ] ), ] @@ -304,8 +368,8 @@ def test_disks_app_func_read_write(started_cluster): "--save-logs", "--disk", "test1", - "read", - "5.txt", + "--query", + "read 5.txt", ] ) @@ -319,7 +383,15 @@ def test_remote_disk_list(started_cluster): init_data_s3(source) out = source.exec_in_container( - ["/usr/bin/clickhouse", "disks", "--save-logs", "--disk", "test3", "list", "."] + [ + "/usr/bin/clickhouse", + "disks", + "--save-logs", + "--disk", + "test3", + "--query", + "list .", + ] ) files = out.split("\n") @@ -333,9 +405,8 @@ def test_remote_disk_list(started_cluster): "--save-logs", "--disk", "test3", - "list", - ".", - "--recursive", + "--query", + "list . --recursive", ] ) diff --git a/tests/integration/test_host_resolver_fail_count/__init__.py b/tests/integration/test_disks_app_interactive/__init__.py similarity index 100% rename from tests/integration/test_host_resolver_fail_count/__init__.py rename to tests/integration/test_disks_app_interactive/__init__.py diff --git a/tests/integration/test_disks_app_interactive/configs/config.xml b/tests/integration/test_disks_app_interactive/configs/config.xml new file mode 100644 index 00000000000..bcbb107f0a2 --- /dev/null +++ b/tests/integration/test_disks_app_interactive/configs/config.xml @@ -0,0 +1,3 @@ + + /var/lib/clickhouse/ + \ No newline at end of file diff --git a/tests/integration/test_disks_app_interactive/test.py b/tests/integration/test_disks_app_interactive/test.py new file mode 100644 index 00000000000..ca4ba5d9065 --- /dev/null +++ b/tests/integration/test_disks_app_interactive/test.py @@ -0,0 +1,331 @@ +from helpers.cluster import ClickHouseCluster + +import pytest + +import pathlib + +import subprocess +import select +import io +from typing import List, Tuple, Dict, Union, Optional + +import os + + +class ClickHouseDisksException(Exception): + pass + + +@pytest.fixture(scope="module") +def started_cluster(): + global cluster + try: + cluster = ClickHouseCluster(__file__) + cluster.add_instance( + "disks_app_test", + main_configs=["server_configs/config.xml"], + with_minio=True, + ) + + cluster.start() + yield cluster + + finally: + cluster.shutdown() + + +class DisksClient(object): + SEPARATOR = b"\a\a\a\a\n" + local_client: Optional["DisksClient"] = None # static variable + default_disk_root_directory: str = "/var/lib/clickhouse" + + def __init__(self, bin_path: str, config_path: str, working_path: str): + self.bin_path = bin_path + self.working_path = working_path + + self.proc = subprocess.Popen( + [bin_path, "disks", "--test-mode", "--config", config_path], + stdin=subprocess.PIPE, + stdout=subprocess.PIPE, + stderr=subprocess.PIPE, + ) + + self.poller = select.epoll() + self.poller.register(self.proc.stdout) + self.poller.register(self.proc.stderr) + + self.stopped = False + + self._fd_nums = { + self.proc.stdout.fileno(): self.proc.stdout, + self.proc.stderr.fileno(): self.proc.stderr, + } + + def execute_query(self, query: str, timeout: float = 5.0) -> str: + output = io.BytesIO() + + self.proc.stdin.write(query.encode() + b"\n") + self.proc.stdin.flush() + + events = self.poller.poll(timeout) + if not events: + raise TimeoutError(f"Disks client returned no output") + + for fd_num, event in events: + if event & (select.EPOLLIN | select.EPOLLPRI): + file = self._fd_nums[fd_num] + + if file == self.proc.stdout: + while True: + chunk = file.readline() + if chunk.endswith(self.SEPARATOR): + break + + output.write(chunk) + + elif file == self.proc.stderr: + error_line = self.proc.stderr.readline() + print(error_line) + raise ClickHouseDisksException(error_line.strip().decode()) + + else: + raise ValueError(f"Failed to read from pipe. Flag {event}") + + data = output.getvalue().strip().decode() + return data + + def list_disks(self) -> List[Tuple[str, str]]: + output = self.execute_query("list-disks") + return list( + sorted( + map( + lambda x: (x.split(":")[0], ":".join(x.split(":")[1:])), + output.split("\n"), + ) + ) + ) + + def current_disk_with_path(self) -> Tuple[str, str]: + output = self.execute_query("current_disk_with_path") + disk_line = output.split("\n")[0] + path_line = output.split("\n")[1] + assert disk_line.startswith("Disk: ") + assert path_line.startswith("Path: ") + return disk_line[6:], path_line[6:] + + def ls( + self, path: str, recursive: bool = False, show_hidden: bool = False + ) -> Union[List[str], Dict[str, List[str]]]: + recursive_adding = "--recursive " if recursive else "" + show_hidden_adding = "--all " if show_hidden else "" + output = self.execute_query( + f"list {path} {recursive_adding} {show_hidden_adding}" + ) + if recursive: + answer: Dict[str, List[str]] = dict() + blocks = output.split("\n\n") + for block in blocks: + directory = block.split("\n")[0][:-1] + files = block.split("\n")[1:] + answer[directory] = files + return answer + else: + return output.split("\n") + + def switch_disk(self, disk: str, directory: Optional[str] = None): + directory_addition = f"--path {directory} " if directory is not None else "" + self.execute_query(f"switch-disk {disk} {directory_addition}") + + def cd(self, directory: str, disk: Optional[str] = None): + disk_addition = f"--disk {disk} " if disk is not None else "" + self.execute_query(f"cd {directory} {disk_addition}") + + def copy( + self, + path_from, + path_to, + disk_from: Optional[str] = None, + disk_to: Optional[str] = None, + recursive: bool = False, + ): + disk_from_option = f"--disk-from {disk_from} " if disk_from is not None else "" + disk_to_option = f"--disk-to {disk_to} " if disk_to is not None else "" + recursive_tag = "--recursive" if recursive else "" + + self.execute_query( + f"copy {recursive_tag} {path_from} {path_to} {disk_from_option} {disk_to_option}" + ) + + def move(self, path_from: str, path_to: str): + self.execute_query(f"move {path_from} {path_to}") + + def rm(self, path: str, recursive: bool = False): + recursive_tag = "--recursive" if recursive else "" + self.execute_query(f"rm {recursive_tag} {path}") + + def mkdir(self, path: str, recursive: bool = False): + recursive_adding = "--recursive " if recursive else "" + self.execute_query(f"mkdir {path} {recursive_adding}") + + def ln(self, path_from: str, path_to: str): + self.execute_query(f"link {path_from} {path_to}") + + def read(self, path_from: str, path_to: Optional[str] = None): + path_to_adding = f"--path-to {path_to} " if path_to is not None else "" + output = self.execute_query(f"read {path_from} {path_to_adding}") + return output + + def write( + self, path_from: str, path_to: str + ): # Writing from stdin is difficult to test (do not know how to do this in python) + path_from_adding = f"--path-from {path_from}" + self.execute_query(f"write {path_from_adding} {path_to}") + + @staticmethod + def getLocalDisksClient(refresh: bool): + if (DisksClient.local_client is None) or refresh: + binary_file = os.environ.get("CLICKHOUSE_TESTS_SERVER_BIN_PATH") + current_working_directory = str(pathlib.Path().resolve()) + config_file = f"{current_working_directory}/test_disks_app_interactive/configs/config.xml" + if not os.path.exists(DisksClient.default_disk_root_directory): + os.mkdir(DisksClient.default_disk_root_directory) + + DisksClient.local_client = DisksClient( + binary_file, config_file, current_working_directory + ) + return DisksClient.local_client + else: + return DisksClient.local_client + + +def test_disks_app_interactive_list_disks(): + client = DisksClient.getLocalDisksClient(True) + expected_disks_with_path = [ + ("default", "/"), + ("local", client.working_path), + ] + assert expected_disks_with_path == client.list_disks() + assert client.current_disk_with_path() == ("default", "/") + client.switch_disk("local") + assert client.current_disk_with_path() == ( + "local", + client.working_path, + ) + + +def test_disks_app_interactive_list_files_local(): + client = DisksClient.getLocalDisksClient(True) + client.switch_disk("local") + excepted_listed_files = sorted(os.listdir("test_disks_app_interactive/")) + listed_files = sorted(client.ls("test_disks_app_interactive/")) + assert excepted_listed_files == listed_files + + +def test_disks_app_interactive_list_directories_default(): + client = DisksClient.getLocalDisksClient(True) + traversed_dir = client.ls(".", recursive=True) + client.mkdir("dir1") + client.mkdir("dir2") + client.mkdir(".dir3") + client.cd("dir1") + client.mkdir("dir11") + client.mkdir(".dir12") + client.mkdir("dir13") + client.cd("../dir2") + client.mkdir("dir21") + client.mkdir("dir22") + client.mkdir(".dir23") + client.cd("../.dir3") + client.mkdir("dir31") + client.mkdir(".dir32") + client.cd("..") + traversed_dir = client.ls(".", recursive=True) + assert traversed_dir == { + ".": ["dir1", "dir2"], + "./dir1": ["dir11", "dir13"], + "./dir2": ["dir21", "dir22"], + "./dir1/dir11": [], + "./dir1/dir13": [], + "./dir2/dir21": [], + "./dir2/dir22": [], + } + traversed_dir = client.ls(".", recursive=True, show_hidden=True) + assert traversed_dir == { + ".": [".dir3", "dir1", "dir2"], + "./dir1": [".dir12", "dir11", "dir13"], + "./dir2": [".dir23", "dir21", "dir22"], + "./.dir3": [".dir32", "dir31"], + "./dir1/dir11": [], + "./dir1/.dir12": [], + "./dir1/dir13": [], + "./dir2/dir21": [], + "./dir2/dir22": [], + "./dir2/.dir23": [], + "./.dir3/dir31": [], + "./.dir3/.dir32": [], + } + client.rm("dir2", recursive=True) + traversed_dir = client.ls(".", recursive=True, show_hidden=True) + assert traversed_dir == { + ".": [".dir3", "dir1"], + "./dir1": [".dir12", "dir11", "dir13"], + "./.dir3": [".dir32", "dir31"], + "./dir1/dir11": [], + "./dir1/.dir12": [], + "./dir1/dir13": [], + "./.dir3/dir31": [], + "./.dir3/.dir32": [], + } + traversed_dir = client.ls(".", recursive=True, show_hidden=False) + assert traversed_dir == { + ".": ["dir1"], + "./dir1": ["dir11", "dir13"], + "./dir1/dir11": [], + "./dir1/dir13": [], + } + client.rm("dir1", recursive=True) + client.rm(".dir3", recursive=True) + assert client.ls(".", recursive=True, show_hidden=False) == {".": []} + + +def test_disks_app_interactive_cp_and_read(): + initial_text = "File content" + with open("a.txt", "w") as file: + file.write(initial_text) + client = DisksClient.getLocalDisksClient(True) + client.switch_disk("default") + client.copy("a.txt", "/a.txt", disk_from="local", disk_to="default") + read_text = client.read("a.txt") + assert initial_text == read_text + client.mkdir("dir1") + client.copy("a.txt", "/dir1/b.txt", disk_from="local", disk_to="default") + read_text = client.read("a.txt", path_to="dir1/b.txt") + assert "" == read_text + read_text = client.read("/dir1/b.txt") + assert read_text == initial_text + with open(f"{DisksClient.default_disk_root_directory}/dir1/b.txt", "r") as file: + read_text = file.read() + assert read_text == initial_text + os.remove("a.txt") + client.rm("a.txt") + client.rm("/dir1", recursive=True) + + +def test_disks_app_interactive_test_move_and_write(): + initial_text = "File content" + with open("a.txt", "w") as file: + file.write(initial_text) + client = DisksClient.getLocalDisksClient(True) + client.switch_disk("default") + client.copy("a.txt", "/a.txt", disk_from="local", disk_to="default") + files = client.ls(".") + assert files == ["a.txt"] + client.move("a.txt", "b.txt") + files = client.ls(".") + assert files == ["b.txt"] + read_text = client.read("/b.txt") + assert read_text == initial_text + client.write("b.txt", "c.txt") + read_text = client.read("c.txt") + assert read_text == initial_text + os.remove("a.txt") diff --git a/tests/integration/test_distributed_default_database/test.py b/tests/integration/test_distributed_default_database/test.py index ef69533416b..7da9a368997 100644 --- a/tests/integration/test_distributed_default_database/test.py +++ b/tests/integration/test_distributed_default_database/test.py @@ -5,6 +5,7 @@ in this test we write into per-node tables and read from the distributed table. The default database in the distributed table definition is left empty on purpose to test default database deduction. """ + import pytest from helpers.client import QueryRuntimeException diff --git a/tests/integration/test_distributed_inter_server_secret/test.py b/tests/integration/test_distributed_inter_server_secret/test.py index 50d7be4d11e..7ecb2cda257 100644 --- a/tests/integration/test_distributed_inter_server_secret/test.py +++ b/tests/integration/test_distributed_inter_server_secret/test.py @@ -7,7 +7,7 @@ import uuid import time from helpers.client import QueryRuntimeException -from helpers.cluster import ClickHouseCluster, CLICKHOUSE_CI_MIN_TESTED_VERSION +from helpers.cluster import ClickHouseCluster cluster = ClickHouseCluster(__file__) @@ -27,9 +27,6 @@ def make_instance(name, *args, **kwargs): ) -# DBMS_MIN_REVISION_WITH_INTERSERVER_SECRET_V2 added in 23.3, ensure that CLICKHOUSE_CI_MIN_TESTED_VERSION fits -assert CLICKHOUSE_CI_MIN_TESTED_VERSION < "23.3" - # _n1/_n2 contains cluster with different -- should fail # only n1 contains new_user n1 = make_instance( @@ -38,14 +35,6 @@ n1 = make_instance( user_configs=["configs/users.d/new_user.xml"], ) n2 = make_instance("n2", main_configs=["configs/remote_servers_n2.xml"]) -backward = make_instance( - "backward", - main_configs=["configs/remote_servers_backward.xml"], - image="clickhouse/clickhouse-server", - # version without DBMS_MIN_REVISION_WITH_INTERSERVER_SECRET_V2 - tag=CLICKHOUSE_CI_MIN_TESTED_VERSION, - with_installed_binary=True, -) users = pytest.mark.parametrize( "user,password", @@ -427,28 +416,6 @@ def test_per_user_protocol_settings_secure_cluster(user, password): ) -@users -def test_user_secure_cluster_with_backward(user, password): - id_ = "with-backward-query-dist_secure-" + user - n1.query( - f"SELECT *, '{id_}' FROM dist_secure_backward", user=user, password=password - ) - assert get_query_user_info(n1, id_) == [user, user] - assert get_query_user_info(backward, id_) == [user, user] - - -@users -def test_user_secure_cluster_from_backward(user, password): - id_ = "from-backward-query-dist_secure-" + user - backward.query(f"SELECT *, '{id_}' FROM dist_secure", user=user, password=password) - assert get_query_user_info(n1, id_) == [user, user] - assert get_query_user_info(backward, id_) == [user, user] - - assert n1.contains_in_log( - "Using deprecated interserver protocol because the client is too old. Consider upgrading all nodes in cluster." - ) - - def test_secure_cluster_distributed_over_distributed_different_users(): # This works because we will have initial_user='default' n1.query( diff --git a/tests/integration/test_force_deduplication/test.py b/tests/integration/test_force_deduplication/test.py index 87b2c45bbc5..14c11bc8500 100644 --- a/tests/integration/test_force_deduplication/test.py +++ b/tests/integration/test_force_deduplication/test.py @@ -29,6 +29,8 @@ def get_counts(): def test_basic(start_cluster): + old_src, old_a, old_b, old_c = 0, 0, 0, 0 + node.query( """ CREATE TABLE test (A Int64) ENGINE = ReplicatedMergeTree ('/clickhouse/test/tables/test','1') ORDER BY tuple(); @@ -39,6 +41,15 @@ def test_basic(start_cluster): INSERT INTO test values(999); """ ) + + src, a, b, c = get_counts() + assert src == old_src + 1 + assert a == old_a + 2 + assert b == old_b + 2 + assert c == old_c + 2 + old_src, old_a, old_b, old_c = src, a, b, c + + # that issert fails on test_mv_b due to partitions by A with pytest.raises(QueryRuntimeException): node.query( """ @@ -46,22 +57,23 @@ def test_basic(start_cluster): INSERT INTO test SELECT number FROM numbers(10); """ ) + src, a, b, c = get_counts() + assert src == old_src + 10 + assert a == old_a + 10 + assert b == old_b + assert c == old_c + 10 + old_src, old_a, old_b, old_c = src, a, b, c - old_src, old_a, old_b, old_c = get_counts() - # number of rows in test_mv_a and test_mv_c depends on order of inserts into views - assert old_src == 11 - assert old_a in (1, 11) - assert old_b == 1 - assert old_c in (1, 11) - + # deduplication only for src table node.query("INSERT INTO test SELECT number FROM numbers(10)") src, a, b, c = get_counts() - # no changes because of deduplication in source table assert src == old_src - assert a == old_a - assert b == old_b - assert c == old_c + assert a == old_a + 10 + assert b == old_b + 10 + assert c == old_c + 10 + old_src, old_a, old_b, old_c = src, a, b, c + # deduplication for MV tables does not work, because previous inserts have not written their deduplications tokens to the log due to `deduplicate_blocks_in_dependent_materialized_views = 0`. node.query( """ SET deduplicate_blocks_in_dependent_materialized_views = 1; @@ -69,11 +81,27 @@ def test_basic(start_cluster): """ ) src, a, b, c = get_counts() - assert src == 11 - assert a == old_a + 10 # first insert could be succesfull with disabled dedup - assert b == 11 + assert src == old_src + assert a == old_a + 10 + assert b == old_b + 10 assert c == old_c + 10 + old_src, old_a, old_b, old_c = src, a, b, c + # deduplication for all the tables + node.query( + """ + SET deduplicate_blocks_in_dependent_materialized_views = 1; + INSERT INTO test SELECT number FROM numbers(10); + """ + ) + src, a, b, c = get_counts() + assert src == old_src + assert a == old_a + assert b == old_b + assert c == old_c + old_src, old_a, old_b, old_c = src, a, b, c + + # that issert fails on test_mv_b due to partitions by A, it is an uniq data which is not deduplicated with pytest.raises(QueryRuntimeException): node.query( """ @@ -82,16 +110,23 @@ def test_basic(start_cluster): INSERT INTO test SELECT number FROM numbers(100,10); """ ) + src, a, b, c = get_counts() + assert src == old_src + 10 + assert a == old_a + 10 + assert b == old_b + assert c == old_c + 10 + old_src, old_a, old_b, old_c = src, a, b, c + # deduplication for all tables, except test_mv_b. For test_mv_b it is an uniq data which is not deduplicated due to exception at previous insert node.query( """ SET deduplicate_blocks_in_dependent_materialized_views = 1; INSERT INTO test SELECT number FROM numbers(100,10); """ ) - src, a, b, c = get_counts() - assert src == 21 - assert a == old_a + 20 - assert b == 21 - assert c == old_c + 20 + assert src == old_src + assert a == old_a + assert b == old_b + 10 + assert c == old_c + old_src, old_a, old_b, old_c = src, a, b, c diff --git a/tests/integration/test_grpc_protocol/test.py b/tests/integration/test_grpc_protocol/test.py index 851da99acf3..328ba3bc05c 100644 --- a/tests/integration/test_grpc_protocol/test.py +++ b/tests/integration/test_grpc_protocol/test.py @@ -39,6 +39,7 @@ node = cluster.add_instance( "TSAN_OPTIONS": "report_atomic_races=0 " + os.getenv("TSAN_OPTIONS", default="") }, ipv6_address=IPV6_ADDRESS, + stay_alive=True, ) main_channel = None @@ -369,47 +370,33 @@ def test_progress(): "SELECT number, sleep(0.31) FROM numbers(8) SETTINGS max_block_size=2, interactive_delay=100000", stream_output=True, ) - results = list(results) - for result in results: - result.time_zone = "" - result.query_id = "" - # print(results) - # Note: We can't convert those messages to string like `results = str(results)` and then compare it as a string - # because str() can serialize a protobuf message with any order of fields. - expected_results = [ - clickhouse_grpc_pb2.Result( - output_format="TabSeparated", - progress=clickhouse_grpc_pb2.Progress( - read_rows=2, read_bytes=16, total_rows_to_read=8 - ), - ), - clickhouse_grpc_pb2.Result(output=b"0\t0\n1\t0\n"), - clickhouse_grpc_pb2.Result( - progress=clickhouse_grpc_pb2.Progress(read_rows=2, read_bytes=16) - ), - clickhouse_grpc_pb2.Result(output=b"2\t0\n3\t0\n"), - clickhouse_grpc_pb2.Result( - progress=clickhouse_grpc_pb2.Progress(read_rows=2, read_bytes=16) - ), - clickhouse_grpc_pb2.Result(output=b"4\t0\n5\t0\n"), - clickhouse_grpc_pb2.Result( - progress=clickhouse_grpc_pb2.Progress(read_rows=2, read_bytes=16) - ), - clickhouse_grpc_pb2.Result(output=b"6\t0\n7\t0\n"), - clickhouse_grpc_pb2.Result( - stats=clickhouse_grpc_pb2.Stats( - rows=8, - blocks=4, - allocated_bytes=1092, - ) - ), + # Note: We can't compare results using a statement like `assert results == expected_results` + # because `results` can come in slightly different order. + # So we compare `outputs` and `progresses` separately and not `results` as a whole. + + outputs = [i.output for i in results if i.output] + progresses = [i.progress for i in results if i.HasField("progress")] + + # print(outputs) + # print(progresses) + + expected_outputs = [ + b"0\t0\n1\t0\n", + b"2\t0\n3\t0\n", + b"4\t0\n5\t0\n", + b"6\t0\n7\t0\n", ] - # Stats data can be returned, which broke the test - results = [i for i in results if not isinstance(i, clickhouse_grpc_pb2.Stats)] + expected_progresses = [ + clickhouse_grpc_pb2.Progress(read_rows=2, read_bytes=16, total_rows_to_read=8), + clickhouse_grpc_pb2.Progress(read_rows=2, read_bytes=16), + clickhouse_grpc_pb2.Progress(read_rows=2, read_bytes=16), + clickhouse_grpc_pb2.Progress(read_rows=2, read_bytes=16), + ] - assert results == expected_results + assert outputs == expected_outputs + assert progresses == expected_progresses def test_session_settings(): @@ -763,3 +750,9 @@ def test_opentelemetry_context_propagation(): ) == "SELECT 1\tsome custom state\n" ) + + +def test_restart(): + assert query("SELECT 1") == "1\n" + node.restart_clickhouse() + assert query("SELECT 2") == "2\n" diff --git a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/config.xml b/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/config.xml deleted file mode 100644 index 42a1f962705..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/config.xml +++ /dev/null @@ -1,4 +0,0 @@ - - 1 - 250 - diff --git a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/host_regexp.xml b/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/host_regexp.xml deleted file mode 100644 index 9329c8dbde2..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/host_regexp.xml +++ /dev/null @@ -1,11 +0,0 @@ - - - - - - test1\.example\.com$ - - default - - - diff --git a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/listen_host.xml b/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/listen_host.xml deleted file mode 100644 index 9c27c612f63..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/configs/listen_host.xml +++ /dev/null @@ -1,5 +0,0 @@ - - :: - 0.0.0.0 - 1 - diff --git a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/coredns_config/Corefile b/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/coredns_config/Corefile deleted file mode 100644 index 3edf37dafa5..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/coredns_config/Corefile +++ /dev/null @@ -1,8 +0,0 @@ -. { - hosts /example.com { - reload "20ms" - fallthrough - } - forward . 127.0.0.11 - log -} diff --git a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/coredns_config/example.com b/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/coredns_config/example.com deleted file mode 100644 index 9beb415c290..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/coredns_config/example.com +++ /dev/null @@ -1 +0,0 @@ -filled in runtime, but needs to exist in order to be volume mapped in docker \ No newline at end of file diff --git a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/scripts/stress_test.py b/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/scripts/stress_test.py deleted file mode 100644 index 70419f95dd3..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/scripts/stress_test.py +++ /dev/null @@ -1,62 +0,0 @@ -import pycurl -import threading -from io import BytesIO -import sys - -client_ip = sys.argv[1] -server_ip = sys.argv[2] - -mutex = threading.Lock() -success_counter = 0 -number_of_threads = 100 -number_of_iterations = 50 - - -def perform_request(): - buffer = BytesIO() - crl = pycurl.Curl() - crl.setopt(pycurl.INTERFACE, client_ip) - crl.setopt(crl.WRITEDATA, buffer) - crl.setopt(crl.URL, f"http://{server_ip}:8123/?query=select+1&user=test_dns") - - crl.perform() - - # End curl session - crl.close() - - str_response = buffer.getvalue().decode("iso-8859-1") - expected_response = "1\n" - - mutex.acquire() - - global success_counter - - if str_response == expected_response: - success_counter += 1 - - mutex.release() - - -def perform_multiple_requests(n): - for request_number in range(n): - perform_request() - - -threads = [] - - -for i in range(number_of_threads): - thread = threading.Thread( - target=perform_multiple_requests, args=(number_of_iterations,) - ) - thread.start() - threads.append(thread) - -for thread in threads: - thread.join() - - -if success_counter == number_of_threads * number_of_iterations: - exit(0) - -exit(1) diff --git a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/test.py b/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/test.py deleted file mode 100644 index d73e8813e79..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records_concurrent/test.py +++ /dev/null @@ -1,88 +0,0 @@ -import pytest -import socket -from helpers.cluster import ClickHouseCluster, get_docker_compose_path, run_and_check -from time import sleep -import os - -DOCKER_COMPOSE_PATH = get_docker_compose_path() -SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__)) - -cluster = ClickHouseCluster(__file__) - -ch_server = cluster.add_instance( - "clickhouse-server", - with_coredns=True, - main_configs=["configs/config.xml", "configs/listen_host.xml"], - user_configs=["configs/host_regexp.xml"], -) - -client = cluster.add_instance( - "clickhouse-client", -) - - -@pytest.fixture(scope="module") -def started_cluster(): - global cluster - try: - cluster.start() - yield cluster - - finally: - cluster.shutdown() - - -def check_ptr_record(ip, hostname): - try: - host, aliaslist, ipaddrlist = socket.gethostbyaddr(ip) - if hostname.lower() == host.lower(): - return True - except socket.herror: - pass - return False - - -def setup_dns_server(ip): - domains_string = "test3.example.com test2.example.com test1.example.com" - example_file_path = f'{ch_server.env_variables["COREDNS_CONFIG_DIR"]}/example.com' - run_and_check(f"echo '{ip} {domains_string}' > {example_file_path}", shell=True) - - # DNS server takes time to reload the configuration. - for try_num in range(10): - if all(check_ptr_record(ip, host) for host in domains_string.split()): - break - sleep(1) - - -def setup_ch_server(dns_server_ip): - ch_server.exec_in_container( - (["bash", "-c", f"echo 'nameserver {dns_server_ip}' > /etc/resolv.conf"]) - ) - ch_server.exec_in_container( - (["bash", "-c", "echo 'options ndots:0' >> /etc/resolv.conf"]) - ) - ch_server.query("SYSTEM DROP DNS CACHE") - - -def build_endpoint_v4(ip): - return f"'http://{ip}:8123/?query=SELECT+1&user=test_dns'" - - -def build_endpoint_v6(ip): - return build_endpoint_v4(f"[{ip}]") - - -def test_host_regexp_multiple_ptr_v4(started_cluster): - server_ip = cluster.get_instance_ip("clickhouse-server") - client_ip = cluster.get_instance_ip("clickhouse-client") - dns_server_ip = cluster.get_instance_ip(cluster.coredns_host) - - setup_dns_server(client_ip) - setup_ch_server(dns_server_ip) - - current_dir = os.path.dirname(__file__) - client.copy_file_to_container( - os.path.join(current_dir, "scripts", "stress_test.py"), "stress_test.py" - ) - - client.exec_in_container(["python3", f"stress_test.py", client_ip, server_ip]) diff --git a/tests/integration/test_host_resolver_fail_count/configs/config.d/cluster.xml b/tests/integration/test_host_resolver_fail_count/configs/config.d/cluster.xml deleted file mode 100644 index bde62b82719..00000000000 --- a/tests/integration/test_host_resolver_fail_count/configs/config.d/cluster.xml +++ /dev/null @@ -1,12 +0,0 @@ - - - - 5 - 5 - 5 - 5 - 5 - 5 - - - \ No newline at end of file diff --git a/tests/integration/test_host_resolver_fail_count/configs/config.d/s3.xml b/tests/integration/test_host_resolver_fail_count/configs/config.d/s3.xml deleted file mode 100644 index 94ac83b32ac..00000000000 --- a/tests/integration/test_host_resolver_fail_count/configs/config.d/s3.xml +++ /dev/null @@ -1,21 +0,0 @@ - - - - - s3 - http://minio1:9001/root/data/ - minio - minio123 - - - - - -
- s3 -
-
-
-
-
-
diff --git a/tests/integration/test_host_resolver_fail_count/test_case.py b/tests/integration/test_host_resolver_fail_count/test_case.py deleted file mode 100644 index d25681f0781..00000000000 --- a/tests/integration/test_host_resolver_fail_count/test_case.py +++ /dev/null @@ -1,126 +0,0 @@ -"""Test Interserver responses on configured IP.""" - -import pytest -import time -from helpers.cluster import ClickHouseCluster - -cluster = ClickHouseCluster(__file__) - -node = cluster.add_instance( - "node", - main_configs=["configs/config.d/cluster.xml", "configs/config.d/s3.xml"], - with_minio=True, -) - - -@pytest.fixture(scope="module") -def start_cluster(): - try: - cluster.start() - yield cluster - - finally: - cluster.shutdown() - - -# The same value as in ClickHouse, this can't be confugured via config now -DEFAULT_RESOLVE_TIME_HISTORY_SECONDS = 2 * 60 - - -def test_host_resolver(start_cluster): - minio_ip = cluster.get_instance_ip("minio1") - - # drop DNS cache - node.set_hosts( - [ - (minio_ip, "minio1"), - (node.ip_address, "minio1"), # no answer on 9001 port on this IP - ] - ) - - node.query("SYSTEM DROP DNS CACHE") - node.query("SYSTEM DROP CONNECTIONS CACHE") - - node.query( - """ - CREATE TABLE test (key UInt32, value UInt32) - Engine=MergeTree() - ORDER BY key PARTITION BY key - SETTINGS storage_policy='s3' - """ - ) - - initial_fails = "0\n" - k = 0 - limit = 100 - while initial_fails == "0\n": - node.query( - f""" - INSERT INTO test VALUES (0,{k}) - """ - ) - # HostResolver chooses IP randomly, so on single call can choose worked ID - initial_fails = node.query( - "SELECT value FROM system.events WHERE event LIKE 'AddressesMarkedAsFailed'" - ) - k += 1 - if k >= limit: - # Dead IP was not choosen for 100 iteration. - # This is not expected, but not an error actually. - # And test should be stopped. - return - - # initial_fails can be more than 1 if clickhouse does something in several parallel threads - - for j in range(10): - for i in range(10): - node.query( - f""" - INSERT INTO test VALUES ({i+1},{j+1}) - """ - ) - fails = node.query( - "SELECT value FROM system.events WHERE event LIKE 'AddressesMarkedAsFailed'" - ) - assert fails == initial_fails - - # Check that clickhouse tries to recheck IP after 2 minutes - time.sleep(DEFAULT_RESOLVE_TIME_HISTORY_SECONDS) - - intermediate_fails = initial_fails - limit = k + 100 - while intermediate_fails == initial_fails: - node.query( - f""" - INSERT INTO test VALUES (101,{k}) - """ - ) - intermediate_fails = node.query( - "SELECT value FROM system.events WHERE event LIKE 'AddressesMarkedAsFailed'" - ) - k += 1 - if k >= limit: - # Dead IP was not choosen for 100 iteration. - # This is not expected, but not an error actually. - # And test should be stopped. - return - - # After another 2 minutes shoudl not be new fails, next retry after 4 minutes - time.sleep(DEFAULT_RESOLVE_TIME_HISTORY_SECONDS) - - initial_fails = intermediate_fails - limit = k + 100 - while intermediate_fails == initial_fails: - node.query( - f""" - INSERT INTO test VALUES (102,{k}) - """ - ) - intermediate_fails = node.query( - "SELECT value FROM system.events WHERE event LIKE 'AddressesMarkedAsFailed'" - ) - k += 1 - if k >= limit: - break - - assert k == limit diff --git a/tests/integration/test_interserver_dns_retires/test.py b/tests/integration/test_interserver_dns_retires/test.py index f0c581e6450..7c81f278737 100644 --- a/tests/integration/test_interserver_dns_retires/test.py +++ b/tests/integration/test_interserver_dns_retires/test.py @@ -2,6 +2,7 @@ This test makes sure interserver cluster queries handle invalid DNS records for replicas. """ + from helpers.client import QueryRuntimeException from helpers.cluster import ClickHouseCluster, ClickHouseInstance diff --git a/tests/integration/test_keeper_four_word_command/test.py b/tests/integration/test_keeper_four_word_command/test.py index 44b2b50673a..83503122729 100644 --- a/tests/integration/test_keeper_four_word_command/test.py +++ b/tests/integration/test_keeper_four_word_command/test.py @@ -293,6 +293,16 @@ def test_cmd_conf(started_cluster): assert result["configuration_change_tries_count"] == "20" assert result["async_replication"] == "true" + + assert result["latest_logs_cache_size_threshold"] == "1073741824" + assert result["commit_logs_cache_size_threshold"] == "524288000" + + assert result["disk_move_retries_wait_ms"] == "1000" + assert result["disk_move_retries_during_init"] == "100" + + assert result["log_slow_total_threshold_ms"] == "5000" + assert result["log_slow_cpu_threshold_ms"] == "100" + assert result["log_slow_connection_operation_threshold_ms"] == "1000" finally: close_keeper_socket(client) diff --git a/tests/integration/test_keeper_map/test.py b/tests/integration/test_keeper_map/test.py index d7b4230d872..31316af7b1e 100644 --- a/tests/integration/test_keeper_map/test.py +++ b/tests/integration/test_keeper_map/test.py @@ -104,3 +104,24 @@ def test_keeper_map_without_zk(started_cluster): node.query("DETACH TABLE test_keeper_map_without_zk") client.stop() + + +def test_keeper_map_with_failed_drop(started_cluster): + run_query( + "CREATE TABLE test_keeper_map_with_failed_drop (key UInt64, value UInt64) ENGINE = KeeperMap('/test_keeper_map_with_failed_drop') PRIMARY KEY(key);" + ) + + run_query("INSERT INTO test_keeper_map_with_failed_drop VALUES (1, 11)") + run_query("SYSTEM ENABLE FAILPOINT keepermap_fail_drop_data") + node.query("DROP TABLE test_keeper_map_with_failed_drop SYNC") + + zk_client = get_genuine_zk() + assert ( + zk_client.get("/test_keeper_map/test_keeper_map_with_failed_drop/data") + is not None + ) + + run_query("SYSTEM DISABLE FAILPOINT keepermap_fail_drop_data") + run_query( + "CREATE TABLE test_keeper_map_with_failed_drop_another (key UInt64, value UInt64) ENGINE = KeeperMap('/test_keeper_map_with_failed_drop') PRIMARY KEY(key);" + ) diff --git a/tests/integration/test_lost_part/test.py b/tests/integration/test_lost_part/test.py index b8e67551d79..c2469d6c524 100644 --- a/tests/integration/test_lost_part/test.py +++ b/tests/integration/test_lost_part/test.py @@ -266,7 +266,7 @@ def test_lost_last_part(start_cluster): "ALTER TABLE mt3 UPDATE id = 777 WHERE 1", settings={"mutations_sync": "0"} ) - partition_id = node1.query("select partitionId('x')").strip() + partition_id = node1.query("select partitionID('x')").strip() remove_part_from_disk(node1, "mt3", f"{partition_id}_0_0_0") # other way to detect broken parts diff --git a/tests/integration/test_manipulate_statistics/test.py b/tests/integration/test_manipulate_statistics/test.py index a602cce63df..2541c9b946f 100644 --- a/tests/integration/test_manipulate_statistics/test.py +++ b/tests/integration/test_manipulate_statistics/test.py @@ -167,7 +167,8 @@ def test_replicated_table_ddl(started_cluster): check_stat_file_on_disk(node2, "test_stat", "all_0_0_0_1", "c", False) check_stat_file_on_disk(node2, "test_stat", "all_0_0_0_1", "d", True) node1.query( - "ALTER TABLE test_stat CLEAR STATISTICS d", settings={"alter_sync": "2"} + "ALTER TABLE test_stat CLEAR STATISTICS d", + settings={"alter_sync": "2", "mutations_sync": 2}, ) node1.query( "ALTER TABLE test_stat ADD STATISTICS b type tdigest", @@ -177,7 +178,8 @@ def test_replicated_table_ddl(started_cluster): check_stat_file_on_disk(node2, "test_stat", "all_0_0_0_2", "b", False) check_stat_file_on_disk(node2, "test_stat", "all_0_0_0_2", "d", False) node1.query( - "ALTER TABLE test_stat MATERIALIZE STATISTICS b", settings={"alter_sync": "2"} + "ALTER TABLE test_stat MATERIALIZE STATISTICS b", + settings={"alter_sync": "2", "mutations_sync": 2}, ) check_stat_file_on_disk(node2, "test_stat", "all_0_0_0_3", "a", True) check_stat_file_on_disk(node2, "test_stat", "all_0_0_0_3", "b", True) diff --git a/tests/integration/test_mask_sensitive_info/configs/users.xml b/tests/integration/test_mask_sensitive_info/configs/users.xml index f129a5bb3e3..f767216e907 100644 --- a/tests/integration/test_mask_sensitive_info/configs/users.xml +++ b/tests/integration/test_mask_sensitive_info/configs/users.xml @@ -2,6 +2,7 @@ 5 + 5 diff --git a/tests/integration/test_max_suspicious_broken_parts/test.py b/tests/integration/test_max_suspicious_broken_parts/test.py index c1f34adbb62..831531ed690 100644 --- a/tests/integration/test_max_suspicious_broken_parts/test.py +++ b/tests/integration/test_max_suspicious_broken_parts/test.py @@ -25,7 +25,7 @@ def break_part(table, part_name): [ "bash", "-c", - f"rm /var/lib/clickhouse/data/default/{table}/{part_name}/columns.txt", + f"rm /var/lib/clickhouse/data/default/{table}/{part_name}/primary.cidx", ] ) diff --git a/tests/integration/test_storage_kerberized_hdfs/__init__.py b/tests/integration/test_memory_limit/__init__.py similarity index 100% rename from tests/integration/test_storage_kerberized_hdfs/__init__.py rename to tests/integration/test_memory_limit/__init__.py diff --git a/tests/integration/test_memory_limit/configs/async_metrics_no.xml b/tests/integration/test_memory_limit/configs/async_metrics_no.xml new file mode 100644 index 00000000000..96cae3bf387 --- /dev/null +++ b/tests/integration/test_memory_limit/configs/async_metrics_no.xml @@ -0,0 +1,21 @@ + + + 86400 + + + + + + + + + + + + + + + + + + diff --git a/tests/integration/test_memory_limit/test.py b/tests/integration/test_memory_limit/test.py new file mode 100644 index 00000000000..6d6745711da --- /dev/null +++ b/tests/integration/test_memory_limit/test.py @@ -0,0 +1,54 @@ +#!/usr/bin/env python3 +import logging +import time +import pytest +from helpers.cluster import ClickHouseCluster +from multiprocessing.dummy import Pool + +cluster = ClickHouseCluster(__file__) + +node = cluster.add_instance( + "node", + main_configs=[ + "configs/async_metrics_no.xml", + ], + mem_limit="4g", + env_variables={"MALLOC_CONF": "dirty_decay_ms:0"}, +) + + +@pytest.fixture(scope="module", autouse=True) +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +def test_multiple_queries(): + if node.is_built_with_sanitizer(): + return + + p = Pool(15) + + def run_query(node): + try: + node.query("SELECT * FROM system.numbers GROUP BY number") + except Exception as ex: + print("Exception", ex) + raise ex + + tasks = [] + for i in range(30): + tasks.append(p.apply_async(run_query, (node,))) + time.sleep(i * 0.1) + + for task in tasks: + try: + task.get() + except Exception as ex: + print("Exception", ex) + + # test that we didn't kill the server + node.query("SELECT 1") diff --git a/tests/integration/test_memory_limit_observer/test.py b/tests/integration/test_memory_limit_observer/test.py index fe3acd9a0cf..0eda165b1d2 100644 --- a/tests/integration/test_memory_limit_observer/test.py +++ b/tests/integration/test_memory_limit_observer/test.py @@ -35,7 +35,7 @@ def get_latest_mem_limit(): ).strip() ) return mem_limit - except Exception as e: + except Exception: time.sleep(1) raise Exception("Cannot get memory limit") @@ -51,3 +51,29 @@ def test_observe_memory_limit(started_cluster): if new_max_mem > original_max_mem: return raise Exception("the memory limit does not increase as expected") + + +def test_memory_usage_doesnt_include_page_cache_size(started_cluster): + try: + # populate page cache with 4GB of data; it might be killed by OOM killer but it is fine + node1.exec_in_container( + ["dd", "if=/dev/zero", "of=outputfile", "bs=1M", "count=4K"] + ) + except Exception: + pass + + observer_refresh_period = int( + node1.query( + "select value from system.server_settings where name = 'cgroups_memory_usage_observer_wait_time'" + ).strip() + ) + time.sleep(observer_refresh_period + 1) + + max_mem_usage_from_cgroup = node1.query( + """ + SELECT max(toUInt64(replaceRegexpAll(message, 'Read current memory usage (\\d+) bytes.*', '\\1'))) AS max_mem + FROM system.text_log + WHERE logger_name = 'CgroupsMemoryUsageObserver' AND message LIKE 'Read current memory usage%bytes%' + """ + ).strip() + assert int(max_mem_usage_from_cgroup) < 2 * 2**30 diff --git a/tests/integration/test_merge_tree_s3/test.py b/tests/integration/test_merge_tree_s3/test.py index 0bf81e81383..e90fafa8461 100644 --- a/tests/integration/test_merge_tree_s3/test.py +++ b/tests/integration/test_merge_tree_s3/test.py @@ -962,7 +962,7 @@ def test_s3_engine_heavy_write_check_mem( "INSERT INTO s3_test SELECT number, toString(number) FROM numbers(50000000)" f" SETTINGS " f" max_memory_usage={2*memory}" - f", max_threads=1" # ParallelFormattingOutputFormat consumption depends on it + ", max_threads=1, optimize_trivial_insert_select=1" # ParallelFormattingOutputFormat consumption depends on it f", s3_max_inflight_parts_for_one_file={in_flight}", query_id=query_id, ) @@ -1010,9 +1010,10 @@ def test_s3_disk_heavy_write_check_mem(cluster, broken_s3, node_name): node.query( "INSERT INTO s3_test SELECT number, toString(number) FROM numbers(50000000)" f" SETTINGS max_memory_usage={2*memory}" - f", max_insert_block_size=50000000" - f", min_insert_block_size_rows=50000000" - f", min_insert_block_size_bytes=1000000000000", + ", max_insert_block_size=50000000" + ", min_insert_block_size_rows=50000000" + ", min_insert_block_size_bytes=1000000000000" + ", optimize_trivial_insert_select=1", query_id=query_id, ) diff --git a/tests/integration/test_named_collections/configs/config.d/named_collections_with_zookeeper.xml b/tests/integration/test_named_collections/configs/config.d/named_collections_with_zookeeper.xml index 2d7946d1587..43d80ee6f69 100644 --- a/tests/integration/test_named_collections/configs/config.d/named_collections_with_zookeeper.xml +++ b/tests/integration/test_named_collections/configs/config.d/named_collections_with_zookeeper.xml @@ -9,4 +9,21 @@ value1 + + + + + true + + node_with_keeper + 9000 + + + node_with_keeper_2 + 9000 + + + true + +
diff --git a/tests/integration/test_named_collections/configs/users.d/users.xml b/tests/integration/test_named_collections/configs/users.d/users.xml index 15da914f666..7d4f0543ff1 100644 --- a/tests/integration/test_named_collections/configs/users.d/users.xml +++ b/tests/integration/test_named_collections/configs/users.d/users.xml @@ -1,4 +1,9 @@ + + + 0 + + diff --git a/tests/integration/test_named_collections/test.py b/tests/integration/test_named_collections/test.py index dbc502236c0..32846c79d23 100644 --- a/tests/integration/test_named_collections/test.py +++ b/tests/integration/test_named_collections/test.py @@ -3,6 +3,8 @@ import pytest import os import time from helpers.cluster import ClickHouseCluster +from contextlib import nullcontext as does_not_raise +from helpers.client import QueryRuntimeException SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__)) NAMED_COLLECTIONS_CONFIG = os.path.join( @@ -761,3 +763,32 @@ def test_keeper_storage(cluster): check_dropped(node1) check_dropped(node2) + + +@pytest.mark.parametrize( + "ignore, expected_raise", + [(True, does_not_raise()), (False, pytest.raises(QueryRuntimeException))], +) +def test_keeper_storage_remove_on_cluster(cluster, ignore, expected_raise): + node = cluster.instances["node_with_keeper"] + + replace_in_users_config( + node, + "ignore_on_cluster_for_replicated_named_collections_queries>.", + f"ignore_on_cluster_for_replicated_named_collections_queries>{int(ignore)}", + ) + node.query("SYSTEM RELOAD CONFIG") + + with expected_raise: + node.query( + "DROP NAMED COLLECTION IF EXISTS test_nc ON CLUSTER `replicated_nc_nodes_cluster`" + ) + node.query( + f"CREATE NAMED COLLECTION test_nc ON CLUSTER `replicated_nc_nodes_cluster` AS key1=1, key2=2 OVERRIDABLE" + ) + node.query( + f"ALTER NAMED COLLECTION test_nc ON CLUSTER `replicated_nc_nodes_cluster` SET key2=3" + ) + node.query( + f"DROP NAMED COLLECTION test_nc ON CLUSTER `replicated_nc_nodes_cluster`" + ) diff --git a/tests/integration/test_odbc_interaction/test.py b/tests/integration/test_odbc_interaction/test.py index 06cbe70f7c6..0d0d7a0afb1 100644 --- a/tests/integration/test_odbc_interaction/test.py +++ b/tests/integration/test_odbc_interaction/test.py @@ -40,6 +40,16 @@ create_table_sql_template = """ PRIMARY KEY (`id`)) ENGINE=InnoDB; """ +create_table_sql_nullable_template = """ + CREATE TABLE `clickhouse`.`{}` ( + `id` integer not null, + `col1` integer, + `col2` decimal(15,10), + `col3` varchar(32), + `col4` datetime + ) + """ + def skip_test_msan(instance): if instance.is_built_with_memory_sanitizer(): @@ -77,6 +87,11 @@ def create_mysql_db(conn, name): cursor.execute("CREATE DATABASE {} DEFAULT CHARACTER SET 'utf8'".format(name)) +def create_mysql_nullable_table(conn, table_name): + with conn.cursor() as cursor: + cursor.execute(create_table_sql_nullable_template.format(table_name)) + + def create_mysql_table(conn, table_name): with conn.cursor() as cursor: cursor.execute(create_table_sql_template.format(table_name)) @@ -192,6 +207,46 @@ def started_cluster(): cluster.shutdown() +def test_mysql_odbc_select_nullable(started_cluster): + skip_test_msan(node1) + mysql_setup = node1.odbc_drivers["MySQL"] + + table_name = "test_insert_nullable_select" + conn = get_mysql_conn() + create_mysql_nullable_table(conn, table_name) + with conn.cursor() as cursor: + cursor.execute( + "INSERT INTO clickhouse.{} VALUES(1, 1, 1.23456, 'data1', '2010-01-01 00:00:00');".format( + table_name + ) + ) + cursor.execute( + "INSERT INTO clickhouse.{} VALUES(2, NULL, NULL, NULL, NULL);".format( + table_name + ) + ) + conn.commit() + + node1.query( + """ + CREATE TABLE {}(id UInt32, col1 Nullable(UInt32), col2 Nullable(Decimal(15, 10)), col3 Nullable(String), col4 Nullable(DateTime)) ENGINE = ODBC('DSN={}', 'clickhouse', '{}'); + """.format( + table_name, mysql_setup["DSN"], table_name + ) + ) + + assert ( + node1.query( + "SELECT id, col1, col2, col3, col4 from {} order by id asc".format( + table_name + ) + ) + == "1\t1\t1.23456\tdata1\t2010-01-01 00:00:00\n2\t\\N\t\\N\t\\N\t\\N\n" + ) + drop_mysql_table(conn, table_name) + conn.close() + + def test_mysql_simple_select_works(started_cluster): skip_test_msan(node1) diff --git a/tests/integration/test_parallel_replicas_custom_key/test.py b/tests/integration/test_parallel_replicas_custom_key/test.py index 07a9e2badff..375fe58d741 100644 --- a/tests/integration/test_parallel_replicas_custom_key/test.py +++ b/tests/integration/test_parallel_replicas_custom_key/test.py @@ -5,7 +5,10 @@ cluster = ClickHouseCluster(__file__) nodes = [ cluster.add_instance( - f"n{i}", main_configs=["configs/remote_servers.xml"], with_zookeeper=True + f"n{i}", + main_configs=["configs/remote_servers.xml"], + with_zookeeper=True, + macros={"replica": f"r{i}"}, ) for i in range(1, 5) ] @@ -20,34 +23,17 @@ def start_cluster(): cluster.shutdown() -def create_tables(cluster): - n1 = nodes[0] - n1.query("DROP TABLE IF EXISTS dist_table") - n1.query(f"DROP TABLE IF EXISTS test_table ON CLUSTER {cluster}") - - n1.query( - f"CREATE TABLE test_table ON CLUSTER {cluster} (key UInt32, value String) Engine=MergeTree ORDER BY (key, sipHash64(value))" - ) - n1.query( - f""" - CREATE TABLE dist_table AS test_table - Engine=Distributed( - {cluster}, - currentDatabase(), - test_table, - rand() - ) - """ +def insert_data(table_name, row_num, all_nodes=False): + query = ( + f"INSERT INTO {table_name} SELECT number % 4, number FROM numbers({row_num})" ) - -def insert_data(cluster, row_num): - create_tables(cluster) - n1 = nodes[0] - n1.query( - f"INSERT INTO dist_table SELECT number % 4, number FROM numbers({row_num})" - ) - n1.query("SYSTEM FLUSH DISTRIBUTED dist_table") + if all_nodes: + for n in nodes: + n.query(query) + else: + n1 = nodes[0] + n1.query(query) @pytest.mark.parametrize("custom_key", ["sipHash64(key)", "key"]) @@ -56,12 +42,36 @@ def insert_data(cluster, row_num): "cluster", ["test_multiple_shards_multiple_replicas", "test_single_shard_multiple_replicas"], ) -def test_parallel_replicas_custom_key(start_cluster, cluster, custom_key, filter_type): +def test_parallel_replicas_custom_key_distributed( + start_cluster, cluster, custom_key, filter_type +): for node in nodes: node.rotate_logs() row_num = 1000 - insert_data(cluster, row_num) + + n1 = nodes[0] + n1.query(f"DROP TABLE IF EXISTS dist_table ON CLUSTER {cluster} SYNC") + n1.query(f"DROP TABLE IF EXISTS test_table_for_dist ON CLUSTER {cluster} SYNC") + n1.query( + f"CREATE TABLE test_table_for_dist ON CLUSTER {cluster} (key UInt32, value String) Engine=MergeTree ORDER BY (key, sipHash64(value))" + ) + + n1.query( + f""" + CREATE TABLE dist_table AS test_table_for_dist + Engine=Distributed( + {cluster}, + currentDatabase(), + test_table_for_dist, + rand() + ) + """ + ) + + insert_data("dist_table", row_num) + + n1.query("SYSTEM FLUSH DISTRIBUTED dist_table") expected_result = "" for i in range(4): @@ -72,10 +82,10 @@ def test_parallel_replicas_custom_key(start_cluster, cluster, custom_key, filter n1.query( "SELECT key, count() FROM dist_table GROUP BY key ORDER BY key", settings={ - "prefer_localhost_replica": 0, "max_parallel_replicas": 4, "parallel_replicas_custom_key": custom_key, "parallel_replicas_custom_key_filter_type": filter_type, + "prefer_localhost_replica": 0, }, ) == expected_result @@ -87,3 +97,87 @@ def test_parallel_replicas_custom_key(start_cluster, cluster, custom_key, filter node.contains_in_log("Processing query on a replica using custom_key") for node in nodes ) + + +@pytest.mark.parametrize("custom_key", ["sipHash64(key)", "key"]) +@pytest.mark.parametrize("filter_type", ["default", "range"]) +@pytest.mark.parametrize( + "cluster", + ["test_single_shard_multiple_replicas"], +) +def test_parallel_replicas_custom_key_mergetree( + start_cluster, cluster, custom_key, filter_type +): + for node in nodes: + node.rotate_logs() + + row_num = 1000 + n1 = nodes[0] + n1.query(f"DROP TABLE IF EXISTS test_table_for_mt ON CLUSTER {cluster} SYNC") + n1.query( + f"CREATE TABLE test_table_for_mt ON CLUSTER {cluster} (key UInt32, value String) Engine=MergeTree ORDER BY (key, sipHash64(value))" + ) + + insert_data("test_table_for_mt", row_num, all_nodes=True) + + expected_result = "" + for i in range(4): + expected_result += f"{i}\t250\n" + + n1 = nodes[0] + assert ( + n1.query( + "SELECT key, count() FROM test_table_for_mt GROUP BY key ORDER BY key", + settings={ + "max_parallel_replicas": 4, + "parallel_replicas_custom_key": custom_key, + "parallel_replicas_custom_key_filter_type": filter_type, + "parallel_replicas_for_non_replicated_merge_tree": 1, + "cluster_for_parallel_replicas": cluster, + }, + ) + == expected_result + ) + + +@pytest.mark.parametrize("custom_key", ["sipHash64(key)", "key"]) +@pytest.mark.parametrize("filter_type", ["default", "range"]) +@pytest.mark.parametrize( + "cluster", + ["test_single_shard_multiple_replicas"], +) +def test_parallel_replicas_custom_key_replicatedmergetree( + start_cluster, cluster, custom_key, filter_type +): + for node in nodes: + node.rotate_logs() + + row_num = 1000 + n1 = nodes[0] + n1.query(f"DROP TABLE IF EXISTS test_table_for_rmt ON CLUSTER {cluster} SYNC") + n1.query( + f"CREATE TABLE test_table_for_rmt ON CLUSTER {cluster} (key UInt32, value String) Engine=ReplicatedMergeTree('/clickhouse/tables', '{{replica}}') ORDER BY (key, sipHash64(value))" + ) + + insert_data("test_table_for_rmt", row_num, all_nodes=False) + + for node in nodes: + node.query("SYSTEM SYNC REPLICA test_table_for_rmt LIGHTWEIGHT") + + expected_result = "" + for i in range(4): + expected_result += f"{i}\t250\n" + + n1 = nodes[0] + assert ( + n1.query( + "SELECT key, count() FROM test_table_for_rmt GROUP BY key ORDER BY key", + settings={ + "max_parallel_replicas": 4, + "parallel_replicas_custom_key": custom_key, + "parallel_replicas_custom_key_filter_type": filter_type, + "cluster_for_parallel_replicas": cluster, + }, + ) + == expected_result + ) diff --git a/tests/integration/test_parallel_replicas_custom_key_failover/test.py b/tests/integration/test_parallel_replicas_custom_key_failover/test.py index 3ba3ce092c3..f24a24f3238 100644 --- a/tests/integration/test_parallel_replicas_custom_key_failover/test.py +++ b/tests/integration/test_parallel_replicas_custom_key_failover/test.py @@ -76,11 +76,11 @@ def test_parallel_replicas_custom_key_failover( f"SELECT key, count() FROM cluster('{cluster_name}', currentDatabase(), test_table) GROUP BY key ORDER BY key", settings={ "log_comment": log_comment, - "prefer_localhost_replica": prefer_localhost_replica, "max_parallel_replicas": 4, "parallel_replicas_custom_key": custom_key, "parallel_replicas_custom_key_filter_type": filter_type, "use_hedged_requests": use_hedged_requests, + "prefer_localhost_replica": prefer_localhost_replica, # avoid considering replica delay on connection choice # otherwise connection can be not distributed evenly among available nodes # and so custom key secondary queries (we check it bellow) @@ -100,20 +100,19 @@ def test_parallel_replicas_custom_key_failover( assert query_id != "" query_id = query_id[:-1] - if prefer_localhost_replica == 0: + assert ( + node1.query( + f"SELECT 'subqueries', count() FROM clusterAllReplicas({cluster_name}, system.query_log) WHERE initial_query_id = '{query_id}' AND type ='QueryFinish' AND query_id != initial_query_id SETTINGS skip_unavailable_shards=1" + ) + == "subqueries\t4\n" + ) + + # With enabled hedged requests, we can't guarantee exact query distribution among nodes + # In case of a replica being slow in terms of responsiveness, hedged connection can change initial replicas choice + if use_hedged_requests == 0: assert ( node1.query( - f"SELECT 'subqueries', count() FROM clusterAllReplicas({cluster_name}, system.query_log) WHERE initial_query_id = '{query_id}' AND type ='QueryFinish' AND query_id != initial_query_id SETTINGS skip_unavailable_shards=1" + f"SELECT h, count() FROM clusterAllReplicas({cluster_name}, system.query_log) WHERE initial_query_id = '{query_id}' AND type ='QueryFinish' GROUP BY hostname() as h ORDER BY h SETTINGS skip_unavailable_shards=1" ) - == "subqueries\t4\n" + == "n1\t3\nn3\t2\n" ) - - # With enabled hedged requests, we can't guarantee exact query distribution among nodes - # In case of a replica being slow in terms of responsiveness, hedged connection can change initial replicas choice - if use_hedged_requests == 0: - assert ( - node1.query( - f"SELECT h, count() FROM clusterAllReplicas({cluster_name}, system.query_log) WHERE initial_query_id = '{query_id}' AND type ='QueryFinish' GROUP BY hostname() as h ORDER BY h SETTINGS skip_unavailable_shards=1" - ) - == "n1\t3\nn3\t2\n" - ) diff --git a/tests/integration/test_parallel_replicas_distributed_skip_shards/test.py b/tests/integration/test_parallel_replicas_distributed_skip_shards/test.py index af114ade2d7..0f54d053ce1 100644 --- a/tests/integration/test_parallel_replicas_distributed_skip_shards/test.py +++ b/tests/integration/test_parallel_replicas_distributed_skip_shards/test.py @@ -53,6 +53,9 @@ def create_tables(cluster, table_name): node1.query(f"INSERT INTO {table_name} SELECT number, number FROM numbers(1000)") node2.query(f"INSERT INTO {table_name} SELECT -number, -number FROM numbers(1000)") node1.query(f"INSERT INTO {table_name} SELECT number, number FROM numbers(3)") + # need to sync replicas to have consistent result + node1.query(f"SYSTEM SYNC REPLICA {table_name}") + node2.query(f"SYSTEM SYNC REPLICA {table_name}") @pytest.mark.parametrize( diff --git a/tests/integration/test_parallel_replicas_no_replicas/test.py b/tests/integration/test_parallel_replicas_no_replicas/test.py index 9f716459643..62d4b005d94 100644 --- a/tests/integration/test_parallel_replicas_no_replicas/test.py +++ b/tests/integration/test_parallel_replicas_no_replicas/test.py @@ -33,7 +33,10 @@ def create_tables(cluster, table_name): @pytest.mark.parametrize("skip_unavailable_shards", [1, 0]) -def test_skip_all_replicas(start_cluster, skip_unavailable_shards): +@pytest.mark.parametrize("max_parallel_replicas", [2, 3, 100]) +def test_skip_all_replicas( + start_cluster, skip_unavailable_shards, max_parallel_replicas +): cluster_name = "test_1_shard_3_unavaliable_replicas" table_name = "tt" create_tables(cluster_name, table_name) @@ -43,7 +46,7 @@ def test_skip_all_replicas(start_cluster, skip_unavailable_shards): f"SELECT key, count() FROM {table_name} GROUP BY key ORDER BY key", settings={ "allow_experimental_parallel_reading_from_replicas": 2, - "max_parallel_replicas": 3, + "max_parallel_replicas": max_parallel_replicas, "cluster_for_parallel_replicas": cluster_name, "skip_unavailable_shards": skip_unavailable_shards, }, diff --git a/tests/integration/test_profile_max_sessions_for_user/test.py b/tests/integration/test_profile_max_sessions_for_user/test.py index 133991fed7a..a2fa77e8dc9 100755 --- a/tests/integration/test_profile_max_sessions_for_user/test.py +++ b/tests/integration/test_profile_max_sessions_for_user/test.py @@ -7,7 +7,7 @@ import pytest import sys import threading -from helpers.cluster import ClickHouseCluster, run_and_check +from helpers.cluster import ClickHouseCluster from helpers.test_tools import assert_logs_contain_with_retry from helpers.uclient import client, prompt @@ -51,7 +51,7 @@ instance = cluster.add_instance( def get_query(name, id): - return f"SElECT '{name}', {id}, number from system.numbers" + return f"SELECT '{name}', {id}, COUNT(*) from system.numbers" def grpc_get_url(): @@ -90,7 +90,7 @@ def threaded_run_test(sessions): if len(sessions) > MAX_SESSIONS_FOR_USER: # High retry amount to avoid flakiness in ASAN (+Analyzer) tests assert_logs_contain_with_retry( - instance, "overflown session count", retry_count=60 + instance, "overflown session count", retry_count=120 ) instance.query(f"KILL QUERY WHERE user='{TEST_USER}' SYNC") diff --git a/tests/integration/test_replicated_database/test.py b/tests/integration/test_replicated_database/test.py index f23384b5c04..60a6e099b22 100644 --- a/tests/integration/test_replicated_database/test.py +++ b/tests/integration/test_replicated_database/test.py @@ -337,8 +337,12 @@ def test_alter_attach(started_cluster, attachable_part, engine): main_node.query(f"SELECT CounterID FROM {database}.alter_attach_test") == "123\n" ) + # On the other node, data is replicated only if using a Replicated table engine if engine == "ReplicatedMergeTree": + dummy_node.query( + f"SYSTEM SYNC REPLICA {database}.alter_attach_test LIGHTWEIGHT" + ) assert ( dummy_node.query(f"SELECT CounterID FROM {database}.alter_attach_test") == "123\n" diff --git a/tests/integration/test_runtime_configurable_cache_size/configs/smaller_query_cache.xml b/tests/integration/test_runtime_configurable_cache_size/configs/empty_query_cache.xml similarity index 64% rename from tests/integration/test_runtime_configurable_cache_size/configs/smaller_query_cache.xml rename to tests/integration/test_runtime_configurable_cache_size/configs/empty_query_cache.xml index 6f2de0fa8f5..c4872a0ce41 100644 --- a/tests/integration/test_runtime_configurable_cache_size/configs/smaller_query_cache.xml +++ b/tests/integration/test_runtime_configurable_cache_size/configs/empty_query_cache.xml @@ -1,7 +1,7 @@ - 1 + 0 diff --git a/tests/integration/test_runtime_configurable_cache_size/test.py b/tests/integration/test_runtime_configurable_cache_size/test.py index f761005f297..beaf83ea754 100644 --- a/tests/integration/test_runtime_configurable_cache_size/test.py +++ b/tests/integration/test_runtime_configurable_cache_size/test.py @@ -94,54 +94,61 @@ CONFIG_DIR = os.path.join(SCRIPT_DIR, "configs") def test_query_cache_size_is_runtime_configurable(start_cluster): - # the initial config specifies the maximum query cache size as 2, run 3 queries, expect 2 cache entries node.query("SYSTEM DROP QUERY CACHE") + + # The initial config allows at most two query cache entries but we don't mind node.query("SELECT 1 SETTINGS use_query_cache = 1, query_cache_ttl = 1") - node.query("SELECT 2 SETTINGS use_query_cache = 1, query_cache_ttl = 1") - node.query("SELECT 3 SETTINGS use_query_cache = 1, query_cache_ttl = 1") time.sleep(2) - node.query("SYSTEM RELOAD ASYNCHRONOUS METRICS") - res = node.query( - "SELECT value FROM system.asynchronous_metrics WHERE metric = 'QueryCacheEntries'", - ) - assert res == "2\n" + # At this point, the query cache contains one entry and it is stale - # switch to a config with a maximum query cache size of 1 + res = node.query( + "SELECT count(*) FROM system.query_cache", + ) + assert res == "1\n" + + # switch to a config with a maximum query cache size of _0_ node.copy_file_to_container( - os.path.join(CONFIG_DIR, "smaller_query_cache.xml"), + os.path.join(CONFIG_DIR, "empty_query_cache.xml"), "/etc/clickhouse-server/config.d/default.xml", ) node.query("SYSTEM RELOAD CONFIG") - # check that eviction worked as expected - time.sleep(2) - node.query("SYSTEM RELOAD ASYNCHRONOUS METRICS") res = node.query( - "SELECT value FROM system.asynchronous_metrics WHERE metric = 'QueryCacheEntries'", - ) - assert ( - res == "2\n" - ) # "Why not 1?", you think. Reason is that QC uses the TTLCachePolicy that evicts lazily only upon insert. - # Not a real issue, can be changed later, at least there's a test now. - - # Also, you may also wonder "why query_cache_ttl = 1"? Reason is that TTLCachePolicy only removes *stale* entries. With the default TTL - # (60 sec), no entries would be removed at all. Again: not a real issue, can be changed later and there's at least a test now. - - # check that the new query cache maximum size is respected when more queries run - node.query("SELECT 4 SETTINGS use_query_cache = 1, query_cache_ttl = 1") - node.query("SELECT 5 SETTINGS use_query_cache = 1, query_cache_ttl = 1") - - time.sleep(2) - node.query("SYSTEM RELOAD ASYNCHRONOUS METRICS") - res = node.query( - "SELECT value FROM system.asynchronous_metrics WHERE metric = 'QueryCacheEntries'", + "SELECT count(*) FROM system.query_cache", ) assert res == "1\n" + # "Why not 0?", I hear you say. Reason is that QC uses the TTLCachePolicy that evicts lazily only upon insert. + # Not a real issue, can be changed later, at least there's a test now. - # restore the original config + # The next SELECT will find a single stale entry which is one entry too much according to the new config. + # This triggers the eviction of all stale entries, in this case the 'SELECT 1' result. + # Then, it tries to insert the 'SELECT 2' result but it also cannot be added according to the config. + node.query("SELECT 2 SETTINGS use_query_cache = 1, query_cache_ttl = 1") + res = node.query( + "SELECT count(*) FROM system.query_cache", + ) + assert res == "0\n" + + # The new maximum cache size is respected when more queries run + node.query("SELECT 3 SETTINGS use_query_cache = 1, query_cache_ttl = 1") + res = node.query( + "SELECT count(*) FROM system.query_cache", + ) + assert res == "0\n" + + # Restore the original config node.copy_file_to_container( os.path.join(CONFIG_DIR, "default.xml"), "/etc/clickhouse-server/config.d/default.xml", ) + + node.query("SYSTEM RELOAD CONFIG") + + # It is possible to insert entries again + node.query("SELECT 4 SETTINGS use_query_cache = 1, query_cache_ttl = 1") + res = node.query( + "SELECT count(*) FROM system.query_cache", + ) + assert res == "1\n" diff --git a/tests/integration/test_shutdown_static_destructor_failure/__init__.py b/tests/integration/test_shutdown_static_destructor_failure/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_shutdown_static_destructor_failure/test.py b/tests/integration/test_shutdown_static_destructor_failure/test.py new file mode 100644 index 00000000000..b1d925cc432 --- /dev/null +++ b/tests/integration/test_shutdown_static_destructor_failure/test.py @@ -0,0 +1,21 @@ +#!/usr/bin/env python3 +import pytest +from helpers.cluster import ClickHouseCluster + +cluster = ClickHouseCluster(__file__) + +node = cluster.add_instance("node", main_configs=[], stay_alive=True) + + +@pytest.fixture(scope="module", autouse=True) +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +def test_shutdown(): + node.query("SYSTEM ENABLE FAILPOINT lazy_pipe_fds_fail_close") + node.stop_clickhouse() diff --git a/tests/integration/test_ssl_cert_authentication/certs/ca-cert.pem b/tests/integration/test_ssl_cert_authentication/certs/ca-cert.pem index a64cd623750..d1e4a3a88d9 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/ca-cert.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/ca-cert.pem @@ -1,32 +1,32 @@ -----BEGIN CERTIFICATE----- -MIIFhTCCA22gAwIBAgIUQOHzlr+pa+RiBlRROQnQgfkDRUMwDQYJKoZIhvcNAQEL +MIIFhTCCA22gAwIBAgIUZmPYBB6vdp8uxKlJcS8mI0SArqQwDQYJKoZIhvcNAQEL BQAwUjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUxITAfBgNVBAoM -GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDELMAkGA1UEAwwCY2EwHhcNMjIwODA4 -MTcwNTQwWhcNMzIwODA1MTcwNTQwWjBSMQswCQYDVQQGEwJSVTETMBEGA1UECAwK +GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDELMAkGA1UEAwwCY2EwHhcNMjQwNjI2 +MTAyNTAwWhcNMzQwNjI0MTAyNTAwWjBSMQswCQYDVQQGEwJSVTETMBEGA1UECAwK U29tZS1TdGF0ZTEhMB8GA1UECgwYSW50ZXJuZXQgV2lkZ2l0cyBQdHkgTHRkMQsw -CQYDVQQDDAJjYTCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBALgd5Tta -+M/XAGMhdi44sKEp7f81kTwqUib4mkHj9Nyp0MKMo9KodV+t60Fej/L0D9Fm1N3V -Q+CC4numWWyaBK6hFdfkU5wvbzZnBx8KozMoeiPLPes6QOl9VWkq3lpBDLAcEZM6 -LMsvNy1jsUdQ06lLy6j1lH9dgIe77qNf9UxZhtCJ4ZslogI8oY3q6D/UI1oCpx5L -J42OyOoCGbTFwzwtHhMw5WKI0EHXEOWwXubdM7P3ETxTjp2+vYPSXj5Irq5XVyfd -tkCJ7GWouB9x5vYa9Y6DC0J/cgOsdrLBbvLGMTubbFjO87musc1DCb9Svpu/IQDP -PawdFTn4ASPny3vt0TMRa/O8nBkrIW84O820ddXZhBb0tNT/q+ftffec2rOas7U4 -kE7YzbQthk9otBqvPX+VeCjFJ5Kx3KTOiPGc/eyWHoJbjvwNrnFuEWIiN/1TqhCc -Y3Qq8cud601yu3dBtCrNAEDCclfnRO0wdKNgkP3u2/lWY4xvMyJfxhNsb/R4R8Z6 -VVYQ4vJXoON0GGYs9D2KURlo+bMIz9hLtZLshK+voOetmRhUmYKa1gE3UxWLmJQM -/p8A7Zihr4OMv5ioH/hnXvVcSJj9VcsPMF3Z/RhllvOxN4TBLyZ0hW55oEz3B4Bn -IbA8LcRJUwfQTKtrVHyt07NLbQm0Kp7YYO8FAgMBAAGjUzBRMB0GA1UdDgQWBBRP -/aq+8kGTfMRAmPgrNaaEHkdKEzAfBgNVHSMEGDAWgBRP/aq+8kGTfMRAmPgrNaaE -HkdKEzAPBgNVHRMBAf8EBTADAQH/MA0GCSqGSIb3DQEBCwUAA4ICAQCycqlIfQMF -bxjczVV+ootUWeiD9Qo8w3VosQtR6PYxDkFZGgP2/aD3wIRNSBpmRfnHCdv1sURB -XkKpqG3E22NRDLHwjmXjI8y5BkUxX1xwL9WX5pe9yZTn3eZu6lPtXrhcPnuMvsxs -X+aP0Y30iMEpHTxLlvV4yNso7ZucXaunWfQTKV18FuXROPgwwnbcZscCQhTxOA+d -JrB+8WD3BwwzmJU8Whmojmt4pYhzS7q47OIfkxIGugtZbtSxAWXBMGrV0KJkipr+ -omYUV1qCbYXZgh+h2+JtNdBpIZ3K+i3esXT14N9e72oysmoYFQN4Qol8lMgZ3Jdr -cTH3m1zg1fOne7KT30XKyfTr4iYoph9WHrcv1XwxwYzPbMI+HdMJX2M30V3HEMRY -RLoafxUZNkFdpWcx29Dok1AI/aVU4vE4+32YdctSJNVPrT/V+Y3dX8skt8tgrnbg -JnrFCpEuVhkNiwgTS6ktH2ecdpY2VqesUENJtw+m85cCBjxd8XYhRNoFBPQp8SAv -hEeGc+hIjXYffy6AUo9p+45uOU+RBPKH4hSleESkrI7duajEXaPPl/wJeQYhqvWp -imbAJtqwI6aCf78NOlbzWiTWJt3g+4kla6oJTInGAdYHcOwwY1KwMWKtSO2ARHjM -wCCDUCIbtJgxTrUk1Kgty5ipLgP64m29Pw== +CQYDVQQDDAJjYTCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBALHzp+SR +T8EtrMBYNlgwIUGMZaXomQPwsjOpjt3RUsdE3LCQ15rLuyBwZ0SbLMDOazCDA7Xr ++AXHDYwg/PCJhe2N4NTzfgnWqbkIGwYLhys95Xxq+q+kV7Csk/27JSk1rsK3Nru/ +Iuj/NWgBkAQC8n10qfbJSXHcXm9wDVy/L8t7DXwxkX70LToGWsb6QW6VkiWe/QOW +QYTMVncqjrtLTCTnqGIk0s6ZIifgPLfaYJxtwK+LdIoJioSbAuVfHmVTe10vjonO +YqJ56KfqOn/G7dIf1l9gOLFBAe4jUf3RS1wkQMdk+oEGLbGnShgW+BQMfB48RkrQ +486h7slzwC29jBJLRARI2Oc9p8/zBDVph0pxkjVGka8dfIkZbmTD932h/1gfMgQl +F20G/H5FF1jk37bDcsczns0c24S1F2uJbzOlHjFLhqmH1IaVCWsYawfBs9khModW +VS6+WAv//cqWE3KmmJ2EdtAmzMCJzAQUEyrMZWmrFrBzpACyMq2zFEtyIXsCXpgq +eW4odxIIZlClibo1FGrflqN+hXnAhRyCj7WJBQ0ZPrEdRMRpBYhYdmygPJ+lWfsg +HOtNnshSuJTXGJTtsuJVr4Ioeq6tKfWGofRu4vvT6cILZjbp9VaxxfVLS7bTsjWR +c+5xHp+KbcjSfw2cJHQN85hlWTpMe9jPhgt/AgMBAAGjUzBRMB0GA1UdDgQWBBSJ +Kj1dSYN0jW+Mu0xFQrobZFmQpTAfBgNVHSMEGDAWgBSJKj1dSYN0jW+Mu0xFQrob +ZFmQpTAPBgNVHRMBAf8EBTADAQH/MA0GCSqGSIb3DQEBCwUAA4ICAQCEm/epjNWe +w0A+1M3i0y8adrXWdTeRXuP3Klgtjh3NGFKNm9WGnOAufsqUIpVwKLw1uYNIhiEA +fj5KuD7YIVU9CrWx4Z/+lILUZhyn/br5mACtRiRTF/75/QwLcDz5z7K9MyMzdL99 +DLQ9bd3JCuEgVj6zacPrALwWVhM8u9fNGJxdQANnIC8yTY5+ZE59/fn7UI787JuR +4njOGWSVnDavbTyJCPMPiUkgwqL+QSWBcNbGAPzMaAblvc1SL2Lj/ikFDAETAZs2 +T/3ZqBqHEOuVhFQYTAvMAdMQX3w8bYv/CGL8++W+qHazY+uqPypd9CLnICbnkZmr +P+id9WleGl2F//u1CQ+YA2Q3EazSFhwRLA7IKIVCrYVaBsbe/bpxxZb6+AQVfM/i ++7+fCbr7A5HDe9Fi4dClv6xPI0GZZkarhQPsoLPaDQeqM4OE+K6oPSHJnqfAB8v3 +NgTt1QuVnKhwveX5DDEP4t/Qt4j2n7AFpeoZDEA8aM33K0noXNrwqHz3991O1RWz +t/gd+cFG/Z1jRP8kYtfAV8go2nzt8QvqBhfIcNnMwD8cwuKJ5G7SdqLvDFj3XCCO +YqQAALl4QFs046eVDEWLajSESmj4fCaTmO05sHHLy7U5asoAo/MWGbmGmL+8ExUX +sPO9r12lPJ7IThJ13PSqbJIJnnloL/XCxA== -----END CERTIFICATE----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/ca-cert.srl b/tests/integration/test_ssl_cert_authentication/certs/ca-cert.srl index c02cd0a4526..cf47b0dc79c 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/ca-cert.srl +++ b/tests/integration/test_ssl_cert_authentication/certs/ca-cert.srl @@ -1 +1 @@ -05F10C67567FE30795D77AF2540F6AC8D4CF2461 +05F10C67567FE30795D77AF2540F6AC8D4CF2470 diff --git a/tests/integration/test_ssl_cert_authentication/certs/ca-key.pem b/tests/integration/test_ssl_cert_authentication/certs/ca-key.pem index 26616a084fb..2dea2ccd837 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/ca-key.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/ca-key.pem @@ -1,52 +1,52 @@ -----BEGIN PRIVATE KEY----- -MIIJQwIBADANBgkqhkiG9w0BAQEFAASCCS0wggkpAgEAAoICAQC4HeU7WvjP1wBj -IXYuOLChKe3/NZE8KlIm+JpB4/TcqdDCjKPSqHVfretBXo/y9A/RZtTd1UPgguJ7 -pllsmgSuoRXX5FOcL282ZwcfCqMzKHojyz3rOkDpfVVpKt5aQQywHBGTOizLLzct -Y7FHUNOpS8uo9ZR/XYCHu+6jX/VMWYbQieGbJaICPKGN6ug/1CNaAqceSyeNjsjq -Ahm0xcM8LR4TMOViiNBB1xDlsF7m3TOz9xE8U46dvr2D0l4+SK6uV1cn3bZAiexl -qLgfceb2GvWOgwtCf3IDrHaywW7yxjE7m2xYzvO5rrHNQwm/Ur6bvyEAzz2sHRU5 -+AEj58t77dEzEWvzvJwZKyFvODvNtHXV2YQW9LTU/6vn7X33nNqzmrO1OJBO2M20 -LYZPaLQarz1/lXgoxSeSsdykzojxnP3slh6CW478Da5xbhFiIjf9U6oQnGN0KvHL -netNcrt3QbQqzQBAwnJX50TtMHSjYJD97tv5VmOMbzMiX8YTbG/0eEfGelVWEOLy -V6DjdBhmLPQ9ilEZaPmzCM/YS7WS7ISvr6DnrZkYVJmCmtYBN1MVi5iUDP6fAO2Y -oa+DjL+YqB/4Z171XEiY/VXLDzBd2f0YZZbzsTeEwS8mdIVueaBM9weAZyGwPC3E -SVMH0Eyra1R8rdOzS20JtCqe2GDvBQIDAQABAoICACUNdyVz0GDbbuuEP2lb5Uf5 -zTiwChNmnPPU0g5ylsOVtTqB3TI2zrA7qv2TzzpgnZJRtIMUOT3EeVnTB2oQsOKH -7oJtI9ppvWLgZxF7LY7NFY+AhmyBY7DZlUe95dOaBbinXW4Ypg0Z6mvLBVFJ1d7b -LxhIm+Lp6UTnsEXWuOHtnnD3Xpy/fRQgKY8K+ERoJ9kkcYdqLoKReqOFWrEDn5/L -oKLSE0e6KtclNfRTgzXIzG/qR+CxHCFo0nDl9CtqqyTQ1Oxj9RqV5yoqQyuo24cp -s8NDYg+24DYXDig5P1/CPQQ26h4RQAFdQ9EALcUAHPXxB+5Gxq2NiMkKvQgBzPRC -92yD8IDarjwizeP5hKCOVCAyspHxtOFgMS8LpHQ4ul++sOPSYV8pL+v0dOhs0aky -96PZyZWsvnN/CVNXuuTAi0RvyUDTGhxt4mc9TUe8+XqFC9P1BHaCyg+jxbEjJCWR -YnJfBiL0auXSL6R/RZ+LSRQtS5RUMNKESYNHFYVik/mQHZoy/1W4pOoe25fnQ8f9 -/80IIzsGq7RDVHgsbPCKgwN7dpDMhTmWcjrC31+oOTWFmIkDkp+dS37ErO6f9kLF -e5syBKTvbV2yscKCTWXkoE9kgCk0TVlwysWKUQtRV3JKsy//YLH/PfjfH/PdWBvv -CXRmZJTH7+Ua5RIEGoT9AoIBAQDD+GMnfYkSOPZPNv6BA3AW7PvX3lwCE5eRorsD -G1hBfYbh582/yG7pRvSTxdxH400jwCU7Et2wlxj4ClZxOyBoeaOi+8wxZB4xFpGG -e7OhGaQJ2Vd4LWqjdb178Qk2a4q8oOsciHoQQzJhTZhd+iiJLpnTIDTR13OVJKeF -Xi4FFk3GPOq7TpugMdQXZ1XB7rh5t9lDjeYRVj0fDU5amBONPKcVYO8lIZElsZjQ -EzfxAeYEJxN+SaNkUUGew3QDZI/caF0nji4WcuEH2LracshoQaCl5MLaPJpMJIOv -aCa442AO1vpRWNyk9dZDjrm6+MjkR1gYPTSaaNll3lmqr2AvAoIBAQDwg/sJOFWa -Hq4lMD5MuX/u7We5HcT1QOzlVsXBEiGiUNcQUuQ+leIZRqJSosEpgnuVS33sbc4r -fDngZJLItwezQ7A/1LqzJVgHb0Qp4Sq0ak905ibjpRd19y9E3tx47KqNzziYcp0M -8t+gOBH5tDWHk8fjcujMuwRaQoHoJvDVl7jSoPhYVh8BIS1lKvYz5/hk/2yG43AF -MIjbUvC4b3yMiPXmvoUKjZNrAM/5f3E4A7SW1aCLEJ1Humepd3pLUrZsma6qq9x8 -lb2MQeW3UG/Yrfo1NquocX9KjHawGe3vTrPWoi0FQMYVHPKQB0oeLd/dkveYOl8B -DPEcUEBDrMMLAoIBAQC0fvCEjOFVAVYLu/FJTqtM0lVbjBheaUEvCB0LTYJKP52e -MGyW+br97TaTbKfpJngEN5OTB1tcrK54trzLadP9phuLoDOAQmB0gf7jSMg312/5 -Ck/ABsSzbVxq7aS4lPChr+0cyK70j4+1g0yVMjVntJ5FkOJVbAWvFqWc9xOX5+UW -c0EX59z6/SoFrE2WY5NRuH3J9MY0raN6GD1Uv8EyUuzpifB52KLhfuPqsejSeBct -N/iCVrzBRv4Thp4yCctfDFjQd+oHDyQon/vp+9KOA/Q9F6hIbY0IGJGZW2pe3D27 -LVI83oXLAgqpCqoShdYyO2vuV2E93thtNRCNXx5NAoIBAE3/Lvi1eSUqz+SleX6e -JXvRJ3Lj0YyLSoexFR5gh+HAf43+o6AcMR9rRZawyx1wAC3RNnvmvBZkCczYMLXA -jVG4IL8CeK1B7gMzNRKzv4qUc0IApRnr3ujViVG3SB46+bBVRBBEEuQxGw41QLcO -ltFpvkfnatGB0I6IxOIJRs6tjjVYGFFlVFakcLk/Lmp0zpvKLWKs/RXhwHAHvKLD -HMviWoRCwphCg00PDWLmzkuRAA5uJssSTz8Elztg0Jr+rsQXLoqQg0cvtDF46UsC -XdMR0HNTUGWmsNX3KUJAlmWlyzJOk3UBpXsRUSQeCQ4yaEfNsld+jnKjxMkeyUhp -DTMCggEBAJs4Ng8C7Pp3q/cP8/lRaei89NueSINNXKKxzuKNg6iy6w9kOqPZef6+ -MoAe14I0UX3051mUcz1RqJU34iOC1p2CjjfM7eT8Ue9TULHeMqc1wj6xseSqZYWM -3R41p96lum179pC+hRGRl2l418xx1/bo079sB8yS/qsUiDyBIBkASgiulXkYDBER -T7L99aqXHrBlEu1S1THcYt8GOqOUuYZ7+RK9i9+Irk58zN/Pkucj0QMHN4WVInYA -kepOlUn0KlUlyn8g9Ii7gZKHFKbv4F5QIdXE5tXX88sRtqHG7GYMFa7kZMGG6bIX -7MqACWG/2mFExJnaXtgA2HS8NYPKUcQ= +MIIJRAIBADANBgkqhkiG9w0BAQEFAASCCS4wggkqAgEAAoICAQCx86fkkU/BLazA +WDZYMCFBjGWl6JkD8LIzqY7d0VLHRNywkNeay7sgcGdEmyzAzmswgwO16/gFxw2M +IPzwiYXtjeDU834J1qm5CBsGC4crPeV8avqvpFewrJP9uyUpNa7Ctza7vyLo/zVo +AZAEAvJ9dKn2yUlx3F5vcA1cvy/Lew18MZF+9C06BlrG+kFulZIlnv0DlkGEzFZ3 +Ko67S0wk56hiJNLOmSIn4Dy32mCcbcCvi3SKCYqEmwLlXx5lU3tdL46JzmKieein +6jp/xu3SH9ZfYDixQQHuI1H90UtcJEDHZPqBBi2xp0oYFvgUDHwePEZK0OPOoe7J +c8AtvYwSS0QESNjnPafP8wQ1aYdKcZI1RpGvHXyJGW5kw/d9of9YHzIEJRdtBvx+ +RRdY5N+2w3LHM57NHNuEtRdriW8zpR4xS4aph9SGlQlrGGsHwbPZITKHVlUuvlgL +//3KlhNyppidhHbQJszAicwEFBMqzGVpqxawc6QAsjKtsxRLciF7Al6YKnluKHcS +CGZQpYm6NRRq35ajfoV5wIUcgo+1iQUNGT6xHUTEaQWIWHZsoDyfpVn7IBzrTZ7I +UriU1xiU7bLiVa+CKHqurSn1hqH0buL70+nCC2Y26fVWscX1S0u207I1kXPucR6f +im3I0n8NnCR0DfOYZVk6THvYz4YLfwIDAQABAoICABZRI14j5yen7cFVjsMyjgkl +bV4INKBs4DxgaF1jMglxlmfCUnfEUxx3XEwbVdp8SK8VzzJSfJFk7EsFnBMifBxV +rbunKchcFn7xCEqSyYnfwlb/J589cg3jJtAsVzW62MbsqT2Uc/FaiD0Z7RDDuduH +9QTRK5fO9jzthY97HqhbL07C/Kc6Qi3DvEC2A9y1f1WegcagYmkgIzvgp3PPtqXu +M4zTZ2inhcQQeCzqgzE7Bm49hAkHt0p4Ej3n1u0IMjF2lF6t9mq/9TCRzHJX5V1z +xrPBYnrAV1ihL1gwlk3g8grPnCbwOmzMORuaTdRd2HcGQh6B4f/5CPRUwuY5nkY7 +UMcX4rbRCcBDzG8s/NllTRaVC6yvEJPN4B/zOek6DI+tRy4tRj+BZO/1771bindD +nsYAklxbbpTiS7B073b+as3RFZW0xvmoqLyhzRQNW6GqSGj4C0KVigkWQ7Br69b7 +O2++oEurmLqI5Hm3RsJeJJ9obG4vKhPUPdtStF/W0W2TO2i0gpTOm+NeO5uYBRB1 +6OvhJH9dzMi+a0ekCpdQeD2xG4NLzwSe62/Ozz9Oi0rpAqHHhu+SvF+WEepLbkyO +2zx/OYpFK47idBRCAlHLC/9UyXpvw2yU9sleElVuHM41CzMe8Pwj3Qk0YdiHHDzS +Y19XEVHh/riXUufsAHVhAoIBAQDpUe+UJLgAIT6boz+MPLrPIGLeazCySEHEsdnp +jGuAx0M59PIDn5OeXQpMqCIRFDw6zhA/4gji0UegFRPIGA3Qduq+GsjVYRt6SHLC +N/rBRi2xg77yyKOMxv/+nwKFh/3TKIQbUc9EQj63WGBGCHu/EyMV7i9V0j8e7D2u +v/Z23nV/+XqawJXi4u2bHB3M/upOoclKfb3ewZBVouajzZc92kNC8XYfPn10Eofu +Pz7jxDX10IJNmzIYOozO9mlBsds7nFIBXS5wMz3iVg3GEUB05pPEKoEtZGrw474u +0M+gW9d7PV3qYdFgjSogiQf4JrSrOwXJQL/26nyfRX9QVplxAoIBAQDDP+fFT7Zl +eFLvouZu73lr++bV1+LogHgX+GHCSIujBELPyFEAyAeElFKFqnJ/NEOuPLG9X7tL +PDhh9NUijcBTPhVvwbH2/CRBdSX7Yf6RHh5fY+2Ik3hTF81L4bQcf0fgyX4roJY9 +YqpjQWFYGmSk4niCqWd+re/ZrYx/zpF+qgN21v37BAZNOCI+KidFScrq29p8kpLj +MdBWa0m7bhJcv4MPO46s2EJZVdczBU7iK86v5NVrGz7fPVm+tGxEDpwhyfYiu961 +U05XzT+trAaBa88KlSKdmUFq3qDdC4bFb6D+Ra4g+mhqNGtfBYXsavnENZgt0N99 +9M/dgaAEa/vvAoIBAQCm4YcFo9nDpgOc2H/Mc2d+XIC661oyOkJoSHk/dcMyodNw +scUkWFACdjO2ro9nPdzyho7S0n5elSew1UKH3MSMtXGjNY8uJ726kfUa+2UsNmDa +VgwOpPlt6KwTV3I7RhCDprgOvk4MWYF4LAr4LHsuKKbwuaM7tByXpotb4UuMrALI +3Q0XgOX0GTGvvsWF6VJ3mXpbAGL839+3kMN8p8EkaWewivpc0Jp0mgiFnNEDokSi +JFf+4CFNeRtbsJ2KcocHNQDmntpnQA9kQv6pC4/ZzU4lge1RJUDkOVC/NXU8ElSm +fjcdPIfAklduW/TKRgz1aEr0Lo7fMcqfNNsiAD7RAoIBAQCaeplZ13OsXMLhrrU6 +2GXtNeSxFJoG8n4SGQbfvJ4eYGSsGQVd5OVt1BxmfTERy7wwwvytpGx/XioN9rQb +HqQoOFqljU7M5zmYQKPIfQP4tSe6uUlaYbM1qwNXIkBqu5mXFFSrF+dGsiW1Wik2 +l8tBWZ2XY4jrBZtbUqBzDnC3ErSi9f8E92408lDFdnyTqYrOvxviq+VjtCnt9fzk +OnZ0w2FiT/DWeFQmcnBNgcmj0J07NYZVs7zOy6+R3xY50oVdhnkjihjuxfaaKV5U +fmK3SyEIcm5s2rCTaYlE2rXKyENMar0WgojSXp8FE02efBUZVH4O4c+xzFwaGVEN +rpIpAoIBAQDnAmhVy85n9oX6Bi+bXg9aZFa/teiCX33lCjNvNeJ5GljGqO0X6423 +6FVg91BYvnCbHQ0frqyKimVxNO/XYxCnid+Or48Cb9LwD31Wyip3UZABljg5sTb0 +fiNK0pxx6G8x1MsX0u97LogGwd5bMuff2lMi4xpinkz6ydB+fYsmM0UXGKsNkB/d +zR1izlqm87TMeQVi+pVZuOLmg/+hXVgISI2M7TlSoytODmpnSg5SZbaT7ut1IIwm +hJdWMTPHLv0X7NwsvhV4Knu27BJBw4470+okrOLfDuZ9+LZ6JHvYg22MywjCJx9s +MJuwAZJiZb+dQc0uwMkAEPMEfOQ1BI1+ -----END PRIVATE KEY----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client1-cert.pem b/tests/integration/test_ssl_cert_authentication/certs/client1-cert.pem index 3a953f1b941..068855d96ea 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client1-cert.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client1-cert.pem @@ -1,30 +1,30 @@ -----BEGIN CERTIFICATE----- -MIIFMDCCAxgCFC1il5+r7ZqfWcm+w5gJP/pffeI1MA0GCSqGSIb3DQEBCwUAMFIx +MIIFMDCCAxgCFAXxDGdWf+MHldd68lQPasjUzyRtMA0GCSqGSIb3DQEBCwUAMFIx CzAJBgNVBAYTAlJVMRMwEQYDVQQIDApTb21lLVN0YXRlMSEwHwYDVQQKDBhJbnRl -cm5ldCBXaWRnaXRzIFB0eSBMdGQxCzAJBgNVBAMMAmNhMB4XDTIyMDgwODE3MDU0 -OVoXDTMyMDgwNTE3MDU0OVowVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUt +cm5ldCBXaWRnaXRzIFB0eSBMdGQxCzAJBgNVBAMMAmNhMB4XDTI0MDYyNjEwMjUw +NFoXDTM0MDYyNDEwMjUwNFowVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUt U3RhdGUxITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEQMA4GA1UE -AwwHY2xpZW50MTCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAMGHbRBS -+W5wgB0Nv66ODxB8lU1xCj/4imLZPZcLzPdtL0fR1TS+G8KjGSrxznEph4NbnFR7 -cE6fKtuC/OSSUMcBeRiXZxjW1+uYjGp6HGJYuyaeVKMrhP2W7BD9GuMu1l0k6+6K -R8gGCW+09ROH8Uy6rvCI+aGl6pgwabKzYWIu04EgIsdCgj9aRpkyxoTwFdriDJng -bTutXrPCC+LaozYJBD4tnwBzSfpMlP+1rEPc/Mpt4beDyLA9vp15uVYQNaJKopvu -CoflqLE04QhcMdubsUjQSikutz5x/chElyWG8w/5kjAKZmdwJLa/yxg3NsCoPdmS -j4aVwQwK0seY5o6yUpdADc0W2BzQ8veuQoSX/rWsf/lWQe2VG7aEn84TSJoUWwgX -LIRQuFdyjE13w0VdprJo6Z07Yuuo3cJ9dnnmh/LOZL84tLC4o4qSsnb5UjjS7HWC -IDVtsvz61tKyApJZ1IvsrEshj602whrIDTCZ6jMhpBLpZIj8GRxSRKxpC+Nqu5Zp -sovCNWbpYAkBns7svwEZaRfKY1Sm6bbgcZk1VaMKUIPHqUjndVkKjZ7SeptXCf2K -v18xEaXDLerg99IhRzUlNvv/MKrG1Y9ukO2xb7UBvFPzkeiL09MHTR1bZcvowiSO -3IlBvCixjuHur5UtsGX4wszpSrhtaCYqZG37AgMBAAEwDQYJKoZIhvcNAQELBQAD -ggIBAFsyHyCiYGXGB1M/dSqnsXm4t7VtnaAKJxNYGtx4pv23jdgU56QgLpCXU6k/ -tOE0uNLTbIuDPRGPAPNlyazFG7Dk6OuEQZ6rTGpcgvgJMZisLynSUhSC1AO+10F+ -w/84EjzF11SrU/OuOh8UcKdNBQVJUu5MU3BVzoK6h+g5iG95KDAPpk/7yw46WMJJ -HI7PR2H2Xu8/7I9LyRl1kqMvzOEvLX0bLlQw5HZ8H/kEgxSOtUG8BqmQgxHL/EwI -kfYu7X7t/f9oqzi/AlFWtBBQg/SPDPpsf7uyxfcjduCiDtNxjAa7OId40WHrLoCD -5NqU8ssphuCKf9kxFp16SB8tkVjo7wJoWCgr8HHqArdOVpS+RNB3fjwvnDt+JXM6 -Xi6Ui6WBrTp7T/VQS2jMz04BbpVLnJcQXX1ri+zqQfM4KFsroZWTz/+WMdSD/tHS -6vnfDUKFEvN7GdN0hpV94r+YinbC9UTgRC7V6prrao24mU4EjaHjQJ+c6tymNzye -azPSoqJiYhDdFq/txxNp+OusBshz5sAl9yJye5vvvdCsClG/6USWVfcixBw9vhcp -m6LmgZd6Gc4cROHG5kGQNwPG8IHfr9hljGQGxnH2lvcRt8t/hEhP3NX+G4n4ihKx -g63Iv+ZMUHnHLs6qfQo7ll8150IrLTXubEKnH6M70/75j4wq +AwwHY2xpZW50MTCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBALaJgDgT +PPqKnBNHrxu/jVwMpfPa9ENvaHbtSi88jQrvu2O+nRV/EoOBb/ol3pQn7RveMyj0 +PV6Io3KzmkMQRcX2Z5H0BNPAM0WijXpzfxWSdS9yZ/BKbpyq3QfHPX9FTAcnnn0S +gBgKUDfSxnQt4tn32kNv09q8nk0H1JXzjxILwKAhllHweJlyVBt1a6AG/yNe9K+9 +atyhYgwuQHv7z0OdbF8NfBWhWODLPkZ4Lj4LktVakYSi5XS5sdTtYXwi8mEDkucZ +1CrlZswIzR44nlmkJVoZssMbGphLzxWw/XiDn2ZtO4QNU1cJYU/WbJiMY+SUjIyr +AsLf2GAbU5bUiluG+XZsYKUjvrzVV6gYHWncXPDVWvjz96PDRsc6M19rsrhMT9N1 +tMgydCPqpxJkweC8/IRt7GZhTlxQNL3ioFsjksPmvm112012XYHd+NiuVIZucY6P +c0dFRWi0VKZUjvLVRbtxWMlSawTk0S/C6sbL2r556GwxJTwkm+EIuK6nGDKg7Kmw +yLFlSyqtXkvUOnnAnIOzEH3VdjUyUniUbfFT4ODs6TLzIkFSSJDN7W4klP6p1Ot1 +ZUkB030FYpFt1r39AfWLPWLjwzKvMWenWaTSpZIRO3h8sXbh6gt7zVZKNMrf8AFJ +uyOnfYaQpUwrxvWvuJdWZETS7lFgoRrJxGDvAgMBAAEwDQYJKoZIhvcNAQELBQAD +ggIBAJ8XAILsGKVXyCBMazuq3k2rup8kkNU/hVg1RuYGmy4bNClNN6havzu/S1SO +/g00+tlXGBlSiPlRXq/p/oZgOI/efqIWSBfcmHuR9STdELfa7zugdFpscgaTOhTq +Ko5o1B81pZKw6wzVfn2JlGxnEy9e+lCC7ptMdFiBqc7SGrapkjCjLReszn1Jctk/ +9lsSvhWZ/7GhvRO/L93X3/ZM51K7VZxEwFnibULApObDZQBFipYdfKlrrVrqtdj2 +M7Plx2Hh+Ivt16Kj/DqRcRLcWVlaM8rp4QAtjn4bDYArFEGGi8ElWFRNjs5ztE12 +f0Iu+yqGmvDn0lHEocNf8fgxHIN1uJ2sYImS11Yn7xHp5FPb7efvYh8Ih6voCaTg +NojHi61q26YIU112A1ylStV4xMKgxt2rqRvmc6UTnWDtzNO9jp3NscQVHtUEJpv2 +Jd+JsDf1c/w42KTwTyOAz5j+D0acRmw1YRsv2BpO5tcly8nvdMX9k7JesdiQL9bx +ik863yOLG0AOMdWtZPVG1BAuiUUlbBS2RRUp3qsP4OuJ+eVKlXFieX+2NuzqyddV +CywaA+R05nutX5R34h3Cm2MmQOERAk9FUeHFX7cZMAXQRcmoBZKtUfKOGUKF0zOT +ZEs7xmHSqAOTx8ufDU26pnbyCxOBYwn1DVX9nYSskMGMSfGU -----END CERTIFICATE----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client1-key.pem b/tests/integration/test_ssl_cert_authentication/certs/client1-key.pem index 767afd6e2c9..8d9b887b033 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client1-key.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client1-key.pem @@ -1,52 +1,52 @@ -----BEGIN PRIVATE KEY----- -MIIJQgIBADANBgkqhkiG9w0BAQEFAASCCSwwggkoAgEAAoICAQDBh20QUvlucIAd -Db+ujg8QfJVNcQo/+Ipi2T2XC8z3bS9H0dU0vhvCoxkq8c5xKYeDW5xUe3BOnyrb -gvzkklDHAXkYl2cY1tfrmIxqehxiWLsmnlSjK4T9luwQ/RrjLtZdJOvuikfIBglv -tPUTh/FMuq7wiPmhpeqYMGmys2FiLtOBICLHQoI/WkaZMsaE8BXa4gyZ4G07rV6z -wgvi2qM2CQQ+LZ8Ac0n6TJT/taxD3PzKbeG3g8iwPb6deblWEDWiSqKb7gqH5aix -NOEIXDHbm7FI0EopLrc+cf3IRJclhvMP+ZIwCmZncCS2v8sYNzbAqD3Zko+GlcEM -CtLHmOaOslKXQA3NFtgc0PL3rkKEl/61rH/5VkHtlRu2hJ/OE0iaFFsIFyyEULhX -coxNd8NFXaayaOmdO2LrqN3CfXZ55ofyzmS/OLSwuKOKkrJ2+VI40ux1giA1bbL8 -+tbSsgKSWdSL7KxLIY+tNsIayA0wmeozIaQS6WSI/BkcUkSsaQvjaruWabKLwjVm -6WAJAZ7O7L8BGWkXymNUpum24HGZNVWjClCDx6lI53VZCo2e0nqbVwn9ir9fMRGl -wy3q4PfSIUc1JTb7/zCqxtWPbpDtsW+1AbxT85Hoi9PTB00dW2XL6MIkjtyJQbwo -sY7h7q+VLbBl+MLM6Uq4bWgmKmRt+wIDAQABAoICAF9pe5lFNyk+OMNaMGsIKaAo -s9kyxmlkiSGT1fweesZvo5RGfNVcdsY+b92OyodbAK7/3vJ9yxBQA6qmiTTNeBqs -/L1lg5Qo89n44x1pp43LDjXLcJHjllUJ78euaW+g+1l/pvv7W8lfRI64yez/pDT+ -gtoexefuq1qt36aVLolf6bGrHMpdXWaLhOvprxsCXNwFf03r6h62ak64d/C3dyjg -SCG5jz8DF/FZiUoKkOwCSvUoOFSoGazePBhaV0f5hN/G6SV0NefrD4CSP/HFtmT9 -fOmXm0Bzgi88tvBh9JyxTCtPkUWzEuqjTM52f1QogIt6Rsf6LF2ffkYIHgP+u4e5 -dDFw2RAZg1DFZrZOyIKW+fyDqftdSj5LEN/AKofc0os4qzCXktcl+x8KSxlGnhn5 -i9V3ZwvaV4ivMxWQ7EO1Jftqgy2OBwTiFHkeVQ8eC7676pTx18UmN0L2heybA0Rq -I9mrzYCDYBn0HGYdb0xplPpDuj/g7zLx+Edn6CEoaXGwU1kT7SoNwfVsix8/+k2x -jYee+f2EjbHmBKA7UeXkLcMa1QpjIipMX5gWBYa6ACV+xdDrvoMDjrRx7A2j4FF/ -T/tv2GjRWeqWB6bySVA7l92Lltm3RLsW82GBr7I9a8xScUYLa0AfNeeBmS5AIgCA -5FhQ7ivlZ56VWemfiTXpAoIBAQDNp6syo1svZJIBQiZIQRxh1H080wtYDi5Yan2R -loQbjbEitzXC1+rFNtspHuOc0frv5XBvbxOJwXmEsXcS8sWDQBKadr4RDdPGeMDa -SA0If8aWN2BjcILfY9VjJzbL4DY2Tfua9OmIrGvx0LMfPpl+6LXgZzCgPv/o+39i -Kv/ISHEsJITloVeuMq5Py0k6+KXepYWAvdrLkEfxg0gl7Ds2TZDOs58gZRb0lW9+ -u7VOtN9EFd6KZtaU1HfBXUBM57Jlt+aIBOGom8LB/XX7bBi6puIqZOK4VqGR+29j -wj1lnX7npfy8dZoCEplTvp3AY7pTO1BrkNluOcacmnSTGtfjAoIBAQDw59P0M0/z -82MwpGCDilXgqXY+SWk8JvRY5raAthqWrjiAkuA9hjh02PXmRLhExs9owsptuP5q -CckFOyoILZFA4IexlvPgyW2rPZFyrMpI98YYG6Lyw0aMSyDX/rBKJA9cServSjzc -1eLT9uPP43L3EkWLtgocHnD1lZ0YSkAe2ehTGEiNOnaW5H81csK07LPYdEvV8wup -aFNZdcUMJt/V8/C2nkAQpLsgGMLRKYgw3KX7JmtkQsldjK0j9rl7eVH65LdjBXuH -cRMS1qRJgHl5+Yemr7890aGomhD37RMYUKbd0NCQIn2ISxtWF0pWra8+zg0fhPiP -EGPbeNVVI30JAoIBAQC1B/Vhu/7Wtzb3OJ2uWPJ9A6nC0xxXRRRy7Drg35gnERFu -t8vxWlPliqZdij1enFDCwDu7PBH833zy260v1tka8lnt8rzZEkzrlvxcqbQfWSsc -rF6C6lWqA52hjLFlwla2cusauqCgmPbkhIxI4rgHyR3hDbT2Or7W/hxh8+v5CBvn -ebYq1V3zj9V1lENAUATi+t2MOJPTQYyzApeOGmb9JEZmTiOzRolwf+MHsoClaf5n -VsDxIBmgJW/NnSKvD/4wIDQkY/eojoRgc5dZ3QvfsmvAWdJh5pCPir/BFwko2/0M -OUdDNlp9nJWv7Em9Q4yPG9Vs+rMLnnxA+o3HuId1AoIBACSQWzg2TY1ORKDOYiO0 -7GHj9qFvjPxnQTD5G3wfp5t1J/hD0qsj4w/BGllv1rQBpNtWrVjH+j1n7M3RdAi3 -udMqAQ5wReW5TN7vwlKwbSd0C+n+z9We0+dZQ8vkyScHoBk20uSs1N4DzKC1WVBl -Sj671DhnUdOAv05W/fgA1QiZtExgZCqjU/qFBdW77Fd/kbBpvlTjxcJZpkTuvhCh -GdokY0WkcT7Vcd1mRLNwZU5dPwgGhcg65ss/HcxWl0JpYIr/CeKKo3wkKmvyjg7l -5AoiWHdxN0qPtcScVbT7k6leHGWQWwd1ZK46EBUaBdtwEygqKA2/peY4658VEPQS -JdkCggEAGLjOV5XjLMliu8tWrOYE/nBZT00KOd4er1nojcxrRd5q2MDbqMqSi5xV -hibl8Egasdx+Sc8+HLazGPFizNrxd6Ahs3tBdUpYClk4dqnV1X9XK9WV1Un6ULD7 -qzhYQ1gJhQLg/rPxg1BLR/Gl3LahlDdaLvDFTEYPz48oVTszC1yQ8A/Do466TuXf -qF5yPBkt/lxasqMa7ZLCUBMZDF/FGL6x0Z9HDcJc9nv1dLFc5vggoEzts8S8Rmue -WwNcwQsuXEuhpRv+Uf3pimoNm7qfPx9vKS4qxHswHCoC4yIxM2VaALUSCq4KicII -3UTpTl6z0FqLEeB20OY2dIyoPhdNoA== +MIIJRAIBADANBgkqhkiG9w0BAQEFAASCCS4wggkqAgEAAoICAQC2iYA4Ezz6ipwT +R68bv41cDKXz2vRDb2h27UovPI0K77tjvp0VfxKDgW/6Jd6UJ+0b3jMo9D1eiKNy +s5pDEEXF9meR9ATTwDNFoo16c38VknUvcmfwSm6cqt0Hxz1/RUwHJ559EoAYClA3 +0sZ0LeLZ99pDb9PavJ5NB9SV848SC8CgIZZR8HiZclQbdWugBv8jXvSvvWrcoWIM +LkB7+89DnWxfDXwVoVjgyz5GeC4+C5LVWpGEouV0ubHU7WF8IvJhA5LnGdQq5WbM +CM0eOJ5ZpCVaGbLDGxqYS88VsP14g59mbTuEDVNXCWFP1myYjGPklIyMqwLC39hg +G1OW1Ipbhvl2bGClI7681VeoGB1p3Fzw1Vr48/ejw0bHOjNfa7K4TE/TdbTIMnQj +6qcSZMHgvPyEbexmYU5cUDS94qBbI5LD5r5tddtNdl2B3fjYrlSGbnGOj3NHRUVo +tFSmVI7y1UW7cVjJUmsE5NEvwurGy9q+eehsMSU8JJvhCLiupxgyoOypsMixZUsq +rV5L1Dp5wJyDsxB91XY1MlJ4lG3xU+Dg7Oky8yJBUkiQze1uJJT+qdTrdWVJAdN9 +BWKRbda9/QH1iz1i48MyrzFnp1mk0qWSETt4fLF24eoLe81WSjTK3/ABSbsjp32G +kKVMK8b1r7iXVmRE0u5RYKEaycRg7wIDAQABAoICACgRktW8U1xj5NLOn3+l0q/s +DtmyrH/JCtNgTzKDRiqqaSYCB5VaaYP4e84bVfqLsR627eAFjRsdP1PEXQ5vmgFU +j3OYbx7UR+z3O7svcywXFCYwJOS4UgON9iro73Tqjz/a0I1/7CJa0TUPzYRfNjbG +k2DOQWD4mn8qQt4Pss4xSj1cYhTmhnKYiCHm6pMcNhFbnLafC8AWpOErnfgZVGvx +OIK9AQn2ev4NX0Q0yWHRRJAU63CEGX4/7OtimE2Zlj75e9vC7bHk3WXYYL5Li2b+ +Azz9+yGc53+a1IBcc6dqrSjcvX3FNxAZ/QR7eycZWiwo95lBSL/iRysBlJ29VglW +YScc2Il/xWNp97PORwsJEDpeWq5UYdinARFK6PAGjnxmADZNAeZHP+r1C5CQaw72 +y31aIrhL2s9wRPZ2DytIlkSmvffIoNpZJW2AyVdJn8L37Aot0Hwr0SsU8/zibvZ7 +4d+7/6rnPnE1jTZlpnDgyH5e5Mtn3YUYDlPAEQudfYyvh0QrNfSOMnetWSYTh/Oi +40iQM2vaKDiK91deTR50g90A88eSgxWMGG6WUzqNoE5CwwiNQxHPhrmFi4H1V6y2 +uaF3s0Gx6aF6j+ws1ImbgrkpAbvgTCENoDtmS8MYbZjXgTzwnG4UtIwqdc5bK2B5 +i9mdb5w1v2v6XLUxVvKhAoIBAQDhVgmw/ssCscde91dWLMylm5Wf+Q7Ry32eYSr0 +UXqYUbChHkYNK5HpVY5b6Br7C13FrhboDk0hsz3agOFexcrua1N2huyJ8gGjlAzz +i+286WuwbhsX9rYgqTvAZmQYpklAfWLZH8nlwtt3iafNhgSVaa//A2m4hhZagElT +pctVakSyG3OYaNDTXBDOnZi9xagc3eWmxkS8PWFaYC0DJCw9yf+9ynH6+FRZg75x +t7nnDd/eSxtW9QUALUCheOO+yIp/uJUiIyWR69cfojQ2vNx5t8FzpK6EqHFCujhq +e+kJB81BAc2P59O8oGqw9fvc9pzCQXyFbx7dtl/Xu/JyMEqnAoIBAQDPYH0afED6 +qyvtQ1le6bjLW3jGCaymflgGwc0sm/pm/3XY4WXhrmqeSIF3tbhP2HnANGinW0wP +nFd0vAi8RU9UxB7iIUAZ6wXRS8/YQmv5zIouPzSCpmvW0l0IcnqjDUS0IZPRo+UA +FTzS2KIQ/yOsHSZoVNQe/Tsdk7Z8XVAJlq+YZ7o7pGR25yGPleUUbVwbIhoEiBPq +EFA+4BErok4CFQB9J8jLRdzmTEQFjQ/w4w066ZkplcIy009a4+LXIvL+MCPG3qMD ++2K/HlTYfMd+CyozaF0ZGTECtakrK+PWbbTj+VV30SD9Sckk8ZqIFUq18Fb574gF +K2KSq5SkYSh5AoIBAQDdenJ2HEkvcctzJQsbsVbII58yKFsPi8IBjKHql7c2xXwl +MJtL0JpOzH/rB7yVKXvWk6ECHyRizkkqXeil/STTqHvVkRInF83SmO8N5mgaeRcW +x3Ir4JrsiUoodrtFmxN+pn8kx+DqytZprMxY7rPMo5+PuCwOaQTJmTP5Woj7gELb +CK5ajBNM2z3Nxwrc48yz6soRXOksV+w7JzK21rQBW2zZf4T+V1yYyyvBnALF/lYe +qJXLp3Jt1QykaSz4VSYEGUnDzuXbggHknspRTtopbJpg7ul1jBYeruhKiVXoQVnV +3k7MdeEgkk+rdWtDqMU1Daa1hB3Db8DOS3YmFB8bAoIBAQDPDD476F0UKTzlWf3r +9pzLZNuTlmsrnC+VJ4ALjvwWQ+7MiFapWfQXbrrc47FO/wqoLWtj1JJ/b5Ad+/MY +znajYmCXU61lczLOwcuV1tNph59bBz4NR82ZoVTDr1DkZMX4tyGYCPQF/i5JMYO2 +Rpa+LCiBuFhFTH3uTOHBD4Vu3WUaXE4jaEHqOWBXtMgQehOg/45MgfSoGHuWGy7p +itYp3AAt9T/UPD+OLA0qIaoNzxQRgtOqIlzPVA0B6U89jyZfRX8i+nx16FKyEL2T +nBmtrcYHp6Zz/aPiWa+6a8rB96zIhNOhmko+uaG7YgHw5pk+R+T/C/mZd7SmTetN +p7e5AoIBAQDXqOVl33+eRw3CxNCJZsfglrD/Jz8VuZv5MZwRolEB4IQwm/whXdnT +34y0UUpUQHnVepzAP3QjsLKANPUY2rKO+f8NAX4Uakzn43QLI+hZcxx6hVkV6OkJ +Hi9fwSEBZzx5DWEbxmYMIGlaRL1yVff8wevQci7WA4rrztb9D5B0c49ItCrMkLNs +X6+9Bh4zafL/FxJSkTahQLe+KGNXSGGGrYB9M31oLSKKM955ZTRnICPxuyA2hffx +8lmHZ/5hmP+eMKoAJ9khilX4LmnkdXJEZ2w5lQTPUTNP8ggaXvFWpijjUsaXEdkR +NMnXQHpKE2RaT22UJ6z3W+biQqNlhlVW -----END PRIVATE KEY----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client1-req.pem b/tests/integration/test_ssl_cert_authentication/certs/client1-req.pem index d60adf1f62d..d5cd522bc8f 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client1-req.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client1-req.pem @@ -1,27 +1,27 @@ -----BEGIN CERTIFICATE REQUEST----- MIIEnDCCAoQCAQAwVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUx ITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEQMA4GA1UEAwwHY2xp -ZW50MTCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAMGHbRBS+W5wgB0N -v66ODxB8lU1xCj/4imLZPZcLzPdtL0fR1TS+G8KjGSrxznEph4NbnFR7cE6fKtuC -/OSSUMcBeRiXZxjW1+uYjGp6HGJYuyaeVKMrhP2W7BD9GuMu1l0k6+6KR8gGCW+0 -9ROH8Uy6rvCI+aGl6pgwabKzYWIu04EgIsdCgj9aRpkyxoTwFdriDJngbTutXrPC -C+LaozYJBD4tnwBzSfpMlP+1rEPc/Mpt4beDyLA9vp15uVYQNaJKopvuCoflqLE0 -4QhcMdubsUjQSikutz5x/chElyWG8w/5kjAKZmdwJLa/yxg3NsCoPdmSj4aVwQwK -0seY5o6yUpdADc0W2BzQ8veuQoSX/rWsf/lWQe2VG7aEn84TSJoUWwgXLIRQuFdy -jE13w0VdprJo6Z07Yuuo3cJ9dnnmh/LOZL84tLC4o4qSsnb5UjjS7HWCIDVtsvz6 -1tKyApJZ1IvsrEshj602whrIDTCZ6jMhpBLpZIj8GRxSRKxpC+Nqu5ZpsovCNWbp -YAkBns7svwEZaRfKY1Sm6bbgcZk1VaMKUIPHqUjndVkKjZ7SeptXCf2Kv18xEaXD -Lerg99IhRzUlNvv/MKrG1Y9ukO2xb7UBvFPzkeiL09MHTR1bZcvowiSO3IlBvCix -juHur5UtsGX4wszpSrhtaCYqZG37AgMBAAGgADANBgkqhkiG9w0BAQsFAAOCAgEA -aCqXR9C9dZPY8ohqtRTPYcmYTURPAMIRnDvfjtUc9896GIJYR696gh3s7oxXo4c5 -85acI5E8LY0zrtq1yOHRSlsmoIp+Tm36a+K1C1H8IZGvwLhUGXogj7eWzGOTTFx2 -OK9m5QiaL3w4p0P2magOlDxA/PCJ1Uqnz5eWbdHT3TBFIgwpRosK7j+/n5eLfgP8 -gaAUhEW1cUxIcFy/nB1TUhrsj60IaDgfgDsKq98c3sDIf6pdzrRuk6m2ur3eVoHp -1gcDn/XhVcF57cvhi0kdBNA65pKfgvHVhFx2YUdb1nlPjkwDrkWTF/HyRxMxs7a6 -g97PvBHvTc8wojnZpRbXdQyaoNjAhJzpcXaJ8qPU8+P8FnFFEsX94nh+u0FTqtZF -DRm8opUwYGrPCznb/u70wlMMgeGFD8BSQ83TfwlEug6J85Kfh0Vp8Z9gD/GNN4sp -RLFChDgU58TmaG+gFAufhUJjDoSwZ2LepwhI585pdePvUNOL+q4hl6dL9pfGKVxu -gwdvM345CJGwbIOhnol6kfakjp3mSqejXGIjnxdzbTKJkGqhwLcL3A06Y37xykRJ -nkHN4ahhLnFEc/k9O1SwcvTTR1Ct06bYGRNbVrjy1RWCsjyCWokSArOdslh3K8K3 -rva3aKss6TWYg2Qjce10pMaluRbIoEkx+0iII9vujoc= +ZW50MTCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBALaJgDgTPPqKnBNH +rxu/jVwMpfPa9ENvaHbtSi88jQrvu2O+nRV/EoOBb/ol3pQn7RveMyj0PV6Io3Kz +mkMQRcX2Z5H0BNPAM0WijXpzfxWSdS9yZ/BKbpyq3QfHPX9FTAcnnn0SgBgKUDfS +xnQt4tn32kNv09q8nk0H1JXzjxILwKAhllHweJlyVBt1a6AG/yNe9K+9atyhYgwu +QHv7z0OdbF8NfBWhWODLPkZ4Lj4LktVakYSi5XS5sdTtYXwi8mEDkucZ1CrlZswI +zR44nlmkJVoZssMbGphLzxWw/XiDn2ZtO4QNU1cJYU/WbJiMY+SUjIyrAsLf2GAb +U5bUiluG+XZsYKUjvrzVV6gYHWncXPDVWvjz96PDRsc6M19rsrhMT9N1tMgydCPq +pxJkweC8/IRt7GZhTlxQNL3ioFsjksPmvm112012XYHd+NiuVIZucY6Pc0dFRWi0 +VKZUjvLVRbtxWMlSawTk0S/C6sbL2r556GwxJTwkm+EIuK6nGDKg7KmwyLFlSyqt +XkvUOnnAnIOzEH3VdjUyUniUbfFT4ODs6TLzIkFSSJDN7W4klP6p1Ot1ZUkB030F +YpFt1r39AfWLPWLjwzKvMWenWaTSpZIRO3h8sXbh6gt7zVZKNMrf8AFJuyOnfYaQ +pUwrxvWvuJdWZETS7lFgoRrJxGDvAgMBAAGgADANBgkqhkiG9w0BAQsFAAOCAgEA +L2KVSFlGtuYjmV6sTYF0GlA4V0RvUTbLM7qnd321957ReDR8iRqj9ZnqKDEHeH9g +jfuW+TV/BeQFjTTAXM5Gy+LRz23xOwqcvxKh0WYjGLhDXIjWx1zzzGBaAcyWMllq ++o6LAUadST3yhfncP0A79hPLxOsPgMGd3OXGajKskDNmU3MTTsbtQ065HsBwo07P +leMx2jDkapesUaNaTmXERg6MT86hsknGsbO/tU3dGNy4hBuOX5O6bEkijF6eESLd +U4Xc54yScVvxpWqTSEAz9xHjIOpOZNfW+enbLdpApxq6IZkeVM8z7yy8DNjTJ2SD +aS/xKexqrWjWFxNa/CtKezkaZgmLs9jGGan+hmlNBeuixvJEekPliv6Syj3wvLp/ +L3/PmLgBzZj6iRdw5fky0swCn1qwpgwYRjBSN+SL0E8yG6BGKFWByQfwWbdOu9DS +lN/CPBe73yi8kYY5gBvBmPsrt3VMVRbXBLNM16jO6lkyYzyC48jTdicqpLsHazZn +Z4I6GZoUQKc9WPzSdu6tEXjM6e/2lkT8kaPmrae3JOKnP+lzjZjfplV1NylICNQY +whPWBVGaRg0dy8dZSTGtzygTNMoHS3zYsBGE4MuGZtm/4+x/XLkz32n1k58wAKxJ +JKafNaOReYFxJKd+ML5XnYOVICuw3nxQY+CeVZlz1Bc= -----END CERTIFICATE REQUEST----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client2-cert.pem b/tests/integration/test_ssl_cert_authentication/certs/client2-cert.pem index 996bcf41d94..9d317f07963 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client2-cert.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client2-cert.pem @@ -1,30 +1,30 @@ -----BEGIN CERTIFICATE----- -MIIFMDCCAxgCFGtzLs/dg4kQgBYqaATBtAwv8dB3MA0GCSqGSIb3DQEBCwUAMFIx +MIIFMDCCAxgCFAXxDGdWf+MHldd68lQPasjUzyRuMA0GCSqGSIb3DQEBCwUAMFIx CzAJBgNVBAYTAlJVMRMwEQYDVQQIDApTb21lLVN0YXRlMSEwHwYDVQQKDBhJbnRl -cm5ldCBXaWRnaXRzIFB0eSBMdGQxCzAJBgNVBAMMAmNhMB4XDTIyMDgwODE3MDU0 -OVoXDTMyMDgwNTE3MDU0OVowVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUt +cm5ldCBXaWRnaXRzIFB0eSBMdGQxCzAJBgNVBAMMAmNhMB4XDTI0MDYyNjEwMjUw +NFoXDTM0MDYyNDEwMjUwNFowVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUt U3RhdGUxITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEQMA4GA1UE -AwwHY2xpZW50MjCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAPpDQjGP -dPUiGXybmTEEbneSxXDaz/+2H4Dl9QLq3NnqozavqqZksseYsvOwLbzdZ92sRLHf -5B1Wwo2h3SbOGtb6CD3NMqV5P/nvHghn3KD60l5Jy81W8aJ+pwra8tVf/O0oDcJ2 -qwHABhMYm7cf7vk3Llt4clQ2g9wj6o4UuCFjGXDqPLxO+xN2Qtx9YZdxUGTrUtww -FgaoJiUqoeuagMtJ2OlmTsRM8VNnddLqEnqTWNtV3hloH709fA4RzudCOhHDwx2p -+zKDwDTnOFgOVaX0I76CZ/eZ3qU4cwIZ5bJaJjowi0XCP0Pk3LYQ+wPrDsIuP1td -xVAlU0rXXePxi0yrwDLzTi8PPogwkMSkfSRf4xlgQRlnQjdya4A6h7VTvh92tpMD -i6EP1JA1nTNebOf5AjKwNHQ6B5XuTRP0PEUIDAfV9mNriR5vnn+oM44AM7FQ+tFw -Jbc8CQX6487M9KGsHmVsf60fWCBmgZicof0XSpVrnMDJdGARzgmxz4z/Eunrr9uu -p0SttZdWns1lLwWpbnrCk7I3F4SZ8On3Yf+RxFLZvBJjvHKfRopi81YkralkKfu+ -Se6TE0QpkPEEaXW5zqvVkt1gW9j79zeBZRFzjuT35F5m7fWi6e7V/W2crtw4lGxH -/LYX397ZP7i7cT6N+g4JYkeLgMy18S1jiZtHAgMBAAEwDQYJKoZIhvcNAQELBQAD -ggIBAJjXO5KD3JrWkcfdvYgdXVRLQDVecwnvUDAP1R2Cw7+iZRfmWccESKTly1HM -+71ThCt7wSFqSxkE/nl6/4cKgNGHG1Zw0Iy4RCMmA5vxiNzgSmUg/3jyl+smQZzf -8e1iDAezlsEJrohCcUTKocv5fl9qWKspZ7Kc6XKQ1q7YbUyh1ZZpzh3mHZ6XhSo1 -EGMXrlcKUst/hkKGiuONOP3qRjsb+lMRZ1IpIB1uIT/NddnQw08Ah25nVSRtc0z5 -b5edvvzspLZ25brsHBzKtHHg46FwMTi+UfMgQoEsV3DXNB/sWT5V+60AYrPA2mss -5MNhGSQlteinYD8f0LxO2Ocoxl5vArzlgDjx+BY1H2etI0xv3u5U09FXqvnbNQj2 -5kbjJXI+wBXxi+CSy029fBBaU4OpjT3TM8VTFFBL1MEe38ZkAX47HcWOenN+xHsq -dFDZ1so5ZYRbPgPhytEE0CWL8fNatjtPCTQTOrZVZW5uKJfUJog6gYbe/YxSKfqx -QtDf10xpSB8L9ooCyyBIx55YctpEtCuj48HXe0vMBBQvFTx+C9XikqTqGjc2kIdU -GNK2uImXLUas09UTXzm4rGTgf+hM2ixHd8/7K0TATh5eMlK0td/unXf+/yWmBEcz -FaZyYeygPcv4U1NXAFdcY//5qZuIF38H4HbQ2QISH5G6LPff +AwwHY2xpZW50MjCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBANUgBLgZ +EjiW3LmjGlCu17h2ERYc75YX+7esbw/iMWMrcrXfvNNutJ4H7hOsV81ZpMEQouHR +fog8InraKCuwmb28DLhJZJlpE49hZxeUJws4yN8VCCTgr/tLtTsvsSS+TGA58B29 +HHaglJwY0w2mOlOVcUxOkjne4VHokLOomhzzqcqTLCjwUslZqRn+SDgXyRw9P5re +J/m6E36dFyBeVglZvealVp4uK/TTqVVFJYBljD22M9wrOeo9AIvrru5VlgDNHu+r +wqgSeE/bGcwhX2J9J++lwOxsFDR7OILY33yoD7/6NN61SJEoRh7xZTQkr/Dc3Hpv +jYyj4YagdWKq2xDZzIxH+QfEMQssuFvulq5L76TJpXB3ceyCu2NEvb/563p1EvXQ +bp5Zjz7Ojo1elobs9epvfbCmiANmTxG8GLKteLXfjcph492gdIw0nsV9/bIzE5C7 +lnff4nEU9E/uEJz0FTw61VjcZPtpqEzLE/8abBU48pTj5HqKo8Nsjx9hPl6trO4h +81yMaqwbQDmza1KsU+CPIFiycyv8Hn4w6JEjwnUc08rOoQ3e7HjqLNpn8X6RirVQ +UrwSU7L8RTKeOCOBLg6AMXfH/frPRnNQjUG/Z7tBTjTJhm38qucxyHI3J5jwX6vn +/jBfdFHfT6510V0Q9fzgzp3H3fyHpnLW1qxDAgMBAAEwDQYJKoZIhvcNAQELBQAD +ggIBAF9fs1tF/yL+eBTf7F/RMdDrb1q9/YZCZ6btJH6CnxCuj4M3o4EkTW2PPSY5 +AeTX0zYaNXvlHT54vjdO+9H3ocyY0HfjVSzttw7qiTkvsssRLLW0PMpc8QMRBpz4 +CmD8vfjY63hKzE2cF5GyP1RveCuFVf7//wM2dfPwrQkIOtKrctejYjn1tOAfgJtX +It+RWvJ8T9t4e3KxYgKSa6eyYxyNMZV67X91C3jIJLgTTLwXXGQF5G8hH3KsclSl +RDE3CAYoyDTtaMlI6A3qDtmvfFKzeltKZc8w7uIbjgHvF49p+n4oh1WwDc/C8SUy +1QAx6DSSW1f470Egtfp0hJKT9yJh7C+/EdeAq8Oh1vMxYKBrtjswCsrFQ+bayEcl +2SzMLez2S/bIFSF0WaDqqIOZDzcjpXjbFlm/px01qoPDk5lkTPGA18Zq8mVc0y2N +R3vYzvfpigjkjXgMcOIfP1Jnlrx1x/4+txR723hUkHQd38nKENepsoEoLrcpmbIl +VAKYTALTle6jJKGf6oZf1TIs09Bc1Qs8Oo4IymubOXD+FlUSmggVwMiST15O5vQu +zdvidRHhAE581DKK04GLmWn0UE0Ko4uaNHAgl2gzZsuJQ5oZynOxmh/z6t+mgA7L +l2qS1WOq29Cq2qWrrfvqbl21LWLrf2X75UyTd3GAlQ19aqLV -----END CERTIFICATE----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client2-key.pem b/tests/integration/test_ssl_cert_authentication/certs/client2-key.pem index 76f56dd68f0..ed0d179712c 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client2-key.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client2-key.pem @@ -1,52 +1,52 @@ -----BEGIN PRIVATE KEY----- -MIIJQgIBADANBgkqhkiG9w0BAQEFAASCCSwwggkoAgEAAoICAQD6Q0Ixj3T1Ihl8 -m5kxBG53ksVw2s//th+A5fUC6tzZ6qM2r6qmZLLHmLLzsC283WfdrESx3+QdVsKN -od0mzhrW+gg9zTKleT/57x4IZ9yg+tJeScvNVvGifqcK2vLVX/ztKA3CdqsBwAYT -GJu3H+75Ny5beHJUNoPcI+qOFLghYxlw6jy8TvsTdkLcfWGXcVBk61LcMBYGqCYl -KqHrmoDLSdjpZk7ETPFTZ3XS6hJ6k1jbVd4ZaB+9PXwOEc7nQjoRw8Mdqfsyg8A0 -5zhYDlWl9CO+gmf3md6lOHMCGeWyWiY6MItFwj9D5Ny2EPsD6w7CLj9bXcVQJVNK -113j8YtMq8Ay804vDz6IMJDEpH0kX+MZYEEZZ0I3cmuAOoe1U74fdraTA4uhD9SQ -NZ0zXmzn+QIysDR0OgeV7k0T9DxFCAwH1fZja4keb55/qDOOADOxUPrRcCW3PAkF -+uPOzPShrB5lbH+tH1ggZoGYnKH9F0qVa5zAyXRgEc4Jsc+M/xLp66/brqdErbWX -Vp7NZS8FqW56wpOyNxeEmfDp92H/kcRS2bwSY7xyn0aKYvNWJK2pZCn7vknukxNE -KZDxBGl1uc6r1ZLdYFvY+/c3gWURc47k9+ReZu31ounu1f1tnK7cOJRsR/y2F9/e -2T+4u3E+jfoOCWJHi4DMtfEtY4mbRwIDAQABAoICAEnNkfTJqKUt9DQyMuAovWcX -6pAYh2SS0gGST0oX9x9wucdD0OCXK73/Ay8oUqSy6pGN3whRzZT1ZBSemnh6KaIi -RFHtdLUl578OTF4QOxliPq29t5OLw2C3Vw21eg2A7DcIK96gXlf6yA/TnwBHM5Nd -OZeSEq1RElvWX7Kc25xE/Fi3S0gBRrv7kUYy15fwu9O3Lk1vBN/bsLwfMXcorjjF -Q2m5WRs80aJWDYIws1hgocC1NvVpJosJWc9QYyMB/dwFTkIuQb0o64LueUhGme2B -7RSeooh7YRR6bDg/aOTK/yED9d6RpVShZpFuuofyRvzu5JJOOgUjrMeSccbqEbjq -VNjkdxoNl917TBezIR/KioXS/gauL4VF9coZostGohDnLIdJrVXwisAZFU9676pn -hFGaZ+vVl6oiWAFwwmbkSo423cIbc+dxMfHtYElbEfGm4LOLrybS3K3sg0gn8HZq -p/AISM2l/282peWjHfa1SXedvvDF1WFJRaB0GuzodkhsqEXXSyAQlc4veNiacNt4 -UQCYiOatcKzvBofgICyX/lTQN/TYdfo57tjCZxnOrDHwpcHH6xRcpuRnDphTV3KD -g2JgUKbvBcZgY03nS7JonzhuOMNd+AUT9Oof1VSS9KKRYLu/eJxl8/6dR7Mcmv5W -fCzEqNVvaigKkeVQEq9hAoIBAQD/73EFWGS+DfKuivyxGAdxAf37eyFRIQYL5Xz5 -xB1h+GjtfdxrigNM0hLHQ+B0209ZwZzl/O37OvW8x8V/Sj/YrY/BYQIGu5quD6lL -NbuEnhLMlU/yANukumNMDbDKo4Fgx+QY+y23zEIFNzraoKH61JH2VrWQ9qfrupaW -qVdUE/LtdZ6/Ar2pKnysLYsUpgZX9mashKg6fNfJWyLqfj3frVD7RM5NTNgwD5as -vnSAkvMLJV1997+57W68E3ALnGxf8UbTxAX8iQ6yig6Oo3bA5er6qt5wbH5m12ml -0UnPvhGNO1eRnX85rkH4okGJCG6ln+wPXbkc2igVH4Tlg/HxAoIBAQD6U3M4JkNE -kVRqke8di5D/HUpsIG2t21DTgihLrg6iNsjdaznvgzRGRAIQCLP14qRoibMsLN8k -zfr+I4QlE2dyIn++fJkaOq/xRvODSDNRRg5M1ttprvAK0QCvbgKgkYxxqpRzOuD9 -Aowpq5YGhHbe8g+aAoYGU9kYbx0XhzU2ZDCqbQP41e3f5OnFrNm8YCzazsJ1fNEU -H7ktT+vCP9g2DtZWF0zFdDeHo4f+h7DcLGaEkvAPpRWQ0qL5tZ7vPZL2GXO9gz6o -aeD5fiMZVUV/SOr51MWIcSIXa+oftxhkZxGqjJrm7Q0cJu3ic7yFXa+9nVaQd8r5 -oWRIvFxRCii3AoIBAQCgdf3VTRs0cUhDGI7fdWJM/uO/RO6Zj+vbj5Ewgz2sy/L1 -W2DT5560Dk2Js0vJ2CpC6vm01ERcrBWfu1xety2aaw7jPi/oCr0q/lS0+8ZKmlTN -AZwy7UjIWbeh2j+KfMLD6+9dnjWGs8B8xqjWo7mKqrWszaZecdbMG5sIcLl41F0N -dh4GQfCpXCL4TEGTu5fklG7BIRbcPKOJ8eLNREMEtwQ71WLG4jP0xgFA4tPmPLu2 -tEGOlcMWDf/MWR95mCP2by5p3M+oCkI1vArucRwmSsBtUq3NrKf/UHVDahkt70vP -0XHQsP8wKcng2dHe2XON0dtNswQ5S6mvvKg0wenxAoIBADmuEuWEQ4nAJwBrYfqf -1yhBmo1xp3QhDe2+lwRhNfQAxaXneDENPXVXZFZHexgUQifoWsW6DSzQ3Z3Dl0Zn -wzVUZ7T/xtN5ZGMnIyND9rcqek8QTvx6F8uWPx8tLSPMhd0HHi+zCHakKKHNbMNR -sIZMpnWpUTKSDXsI1149hHBlA6WxevHScX5eo9Mtsfoq+wrsC3jA6vhFGwkq2jsn -NO2324F42vLAha8Walam84S4ImM37GLeeiqlrnH1fIrJ2FwmnzmyzTRHOVSj0Pic -Ymgi6Cnq8h0vXFijQA5QxSkVaseuoF7HjnbHLZc5bd4ZKU56u6CMSdPdcVslM+xk -d98CggEAYleUQS65Nk2EWj6nLXD8HuVDWbKLXgP2Q4VpORDHpT4isboSTtR6mMwR -huncQX5lvUZqZNRvKIS/IfW/2CiAF+3IhUxx8r+AUyrc7OClp1keH26c52/RYoJM -sl4DfQE/DooWvieQTx4rILiEYrPWwQLYfuDs+dSgnaugr49s431PeVyxlH/849aI -dHnwInBd7sdbN11e6xErry4LU95imkRhg+y+QdiOE2N1Q6I3G2Jn+n4NhIWJfjlS -Dysosye3R96FsFQohVMosVLTL9mCT8J6LFIK8CjRcvhSQmaTmXjTuisIUntBA9zl -PK2zQzz0JkpsDD7mGM+S6apZHiLgHg== +MIIJRAIBADANBgkqhkiG9w0BAQEFAASCCS4wggkqAgEAAoICAQDVIAS4GRI4lty5 +oxpQrte4dhEWHO+WF/u3rG8P4jFjK3K137zTbrSeB+4TrFfNWaTBEKLh0X6IPCJ6 +2igrsJm9vAy4SWSZaROPYWcXlCcLOMjfFQgk4K/7S7U7L7EkvkxgOfAdvRx2oJSc +GNMNpjpTlXFMTpI53uFR6JCzqJoc86nKkywo8FLJWakZ/kg4F8kcPT+a3if5uhN+ +nRcgXlYJWb3mpVaeLiv006lVRSWAZYw9tjPcKznqPQCL667uVZYAzR7vq8KoEnhP +2xnMIV9ifSfvpcDsbBQ0eziC2N98qA+/+jTetUiRKEYe8WU0JK/w3Nx6b42Mo+GG +oHViqtsQ2cyMR/kHxDELLLhb7pauS++kyaVwd3HsgrtjRL2/+et6dRL10G6eWY8+ +zo6NXpaG7PXqb32wpogDZk8RvBiyrXi1343KYePdoHSMNJ7Fff2yMxOQu5Z33+Jx +FPRP7hCc9BU8OtVY3GT7aahMyxP/GmwVOPKU4+R6iqPDbI8fYT5erazuIfNcjGqs +G0A5s2tSrFPgjyBYsnMr/B5+MOiRI8J1HNPKzqEN3ux46izaZ/F+kYq1UFK8ElOy +/EUynjgjgS4OgDF3x/36z0ZzUI1Bv2e7QU40yYZt/KrnMchyNyeY8F+r5/4wX3RR +30+uddFdEPX84M6dx938h6Zy1tasQwIDAQABAoICAQDIuNYY+OvTRipt37IaCQF8 +Zh4jgG8ZIk9dJlaXVAYFi1cG+chiLSKIr5lHCArNiT8E4gE1wtNzxYcHw00QEMxL +CL/GFMFdRrw4TpkEePDovbtZdvprmP3FJAF00673lw5hlk+SApi7FPPBrBOiCEto +ixfgsSNAw6vcM7eMrR8wY0AnXMK7b9PYdMwxge5MfgJXyUuNNOvbY6eWmKa+Qnqv +ZcjXYCKa6YtWkr4pY+005u7U9DQViNSLypYoMXlYWFzlNkqLmW3EU1jihMzgFxI5 +tPwW1TpEsGm7H84SVeTuB26F9UUz9vJ4W8DmxZz2JhNaOvifi056BaKS466KlbWo +iZgt57ajj0VmYxB0ZL7QgQqb2xDZL12kU1AU09QAXJnpy/RqvV2HloKbqrOd5h4L +oME6j8vT6Q8o1vsh2zJuLXHAsMr30XK8x1HhLDDzln49gq/d3GrZkNrPDjcumiwI +o6PYR91Q4QI11kdqR/3005wV50g847uURFNF6J4ziDeDGsHqj2phmOIt6d8vWcFo +XBEovCZkXQUSx+4NgAAy1GRBjK6tLRRQnS9bGgkELS8+Jx84NlgKkH3m6+lKNJQ1 +o5SpUqmk1dYnpTv99U2+5qvA/o9SVy56wlfuo+u0GlbMjs3OmItXErg46UBPhd4d +HipFZrBItpw0DYAF+voLQQKCAQEA9ePjYyy53VGLq2+vRx02IZOMQOLaaBDfybtP +51ksHfCPg+XZ4jWsDH4EPn5DeUmjZ2nG8GYSuv8dAQ4v9kArToLDyy/qhXHzaING +uSd6KlTGrVrPK1Dyu2p69xYrnduD6Vm06sJ4olDq792rEj4/hdzVwrtgw+d1ZLXG +3ropWgrHQT8z7+B9CAIAOXhYlKrV7+UdbAod+j8OpCIHk5X3+NkT4Ht7biqzQvbo +pJJILFA1qHi230N9YR8ng3PHQYObYJ6NFBrxhpXIfXwbuPyrEApY3zaL3HbkYC52 +aAI3zy7WOqZSqRZ6aDzXdf2EMGusNSxj9/TAZhTAiJvwHdwBowKCAQEA3eNC/iMt +kmy4R3FQgti0Zq+CBUErMn46pQhBCcQreI/a5U4UT/iY5WGutKXp45d/BM2ztyQL +T/8p+85RkasVF/rJB2PwlzUZKAAq29nGXuV0I6N6EiMYa2LfFLzrrleNamPQ9Ubn +atp0kiyLiPZ6P0+Y5wZMirHlMyup+fzG6xsS7KVy0Z5Fy4YetP63r6xCVQ+Rdu3l +dvXqGb2Bdc9g4OxES1Zj7MKHg0b3ce2lYaL0cq0z3kJ52MAVbL9tQQOstJX41VYv +/QSVIjC5VACSa2qsqzquqxYOyT1U0l/8innHfD/uY/8907/q/JqoO1hU5LtvZ7OO +ZF/e/ycZCM2U4QKCAQAXUIJQ9v6wk3jQyoguEAD/8gOMa3YWA/OUJySOZRAfzp1s +/jBImJo1nQU9/67aIzdRKOBqDuObw3C2lufJS5BPo2p5K5PrD0DrGfdsuueEeAFW +kpOuIcDCMHh0US/ViejaCV10HPhfO5jrIXOFCU3wnV3PVwD30kx5PhsbJz+ggAEg +mKOODRUN21K2IEkV35TlaC3//n2VKsFyop9hSQj4GW0fDdZIPdg9czff0tbxDLHp +xXhhdv6+ZLvUZPfxqE7lPGNYEq3v+ufFrizav2pg3PpMP9nHD6bbz8v+VKeCB4jc +isSvr6fvlkU/tMgB51OuvwTDj/tmMnWG/nIoAqJNAoIBAQDWiLYsS8zzJwUhplDm +winiosz+0Zy3jE6dZBamH7K8NbK6RLzk+YKzPbgSV9yFPeQEu/KIH2SEqzxnh3tc +cWLKtaKK77keKaux/j9yI+RlukqJbrVHNgGVSppQTb096s8DT5Eopa54pNFSx5j+ +Cvn1nrtCm9eDvi7SQ+RrnVii1qF8hxc1z2bCOmIUM7dcNhxIa+4EZE2ZsHjw/EZg +puqPbkE16khhEVC+v+3djJ17gngBLK/atMFkrYvJgmhbFPd1/w8BDf0GENk0npGB +w6/OBez+/ZUGPCR9tDv/z+i35rjWzGVs78tSodvM8qe4AVbLdOJpDLWfHQbaAm51 +EXhhAoIBAQDmZVXAS4dTDAp/cGwPoXOyFxu+UNnGnAKO0S3aJW9qV6E5N7jsLqzI +4eD2Mk6chkBrO4Upmwgx4sLVnDMlHQGvoqpWUZES9k6oIgNZ6N7KXnvFm5GI7mlR +ySA2LftCeSb4BzQmwyX5wcVjOzfB6bkSgEkvuMFrRStSL6+79XZCoh54jBm+fW6g +up6oXa0+lJbyO4Qrx+oWoe2G9nrUJzsjV1Gj1njnxDECCMrmB+X5P4D02Ac2FgxP +rN+bxs0TpvO4fXsvBN4B+/dtF2Hjgo3rQm5FQ/NmpoO5lAs1VZPjVUiFCjhm3Fyk +Xe2nzT23gDTuPny4yivLAMHPZPfGLLg4 -----END PRIVATE KEY----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client2-req.pem b/tests/integration/test_ssl_cert_authentication/certs/client2-req.pem index 89cde777a45..f36eb94205b 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client2-req.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client2-req.pem @@ -1,27 +1,27 @@ -----BEGIN CERTIFICATE REQUEST----- MIIEnDCCAoQCAQAwVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUx ITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEQMA4GA1UEAwwHY2xp -ZW50MjCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAPpDQjGPdPUiGXyb -mTEEbneSxXDaz/+2H4Dl9QLq3NnqozavqqZksseYsvOwLbzdZ92sRLHf5B1Wwo2h -3SbOGtb6CD3NMqV5P/nvHghn3KD60l5Jy81W8aJ+pwra8tVf/O0oDcJ2qwHABhMY -m7cf7vk3Llt4clQ2g9wj6o4UuCFjGXDqPLxO+xN2Qtx9YZdxUGTrUtwwFgaoJiUq -oeuagMtJ2OlmTsRM8VNnddLqEnqTWNtV3hloH709fA4RzudCOhHDwx2p+zKDwDTn -OFgOVaX0I76CZ/eZ3qU4cwIZ5bJaJjowi0XCP0Pk3LYQ+wPrDsIuP1tdxVAlU0rX -XePxi0yrwDLzTi8PPogwkMSkfSRf4xlgQRlnQjdya4A6h7VTvh92tpMDi6EP1JA1 -nTNebOf5AjKwNHQ6B5XuTRP0PEUIDAfV9mNriR5vnn+oM44AM7FQ+tFwJbc8CQX6 -487M9KGsHmVsf60fWCBmgZicof0XSpVrnMDJdGARzgmxz4z/Eunrr9uup0SttZdW -ns1lLwWpbnrCk7I3F4SZ8On3Yf+RxFLZvBJjvHKfRopi81YkralkKfu+Se6TE0Qp -kPEEaXW5zqvVkt1gW9j79zeBZRFzjuT35F5m7fWi6e7V/W2crtw4lGxH/LYX397Z -P7i7cT6N+g4JYkeLgMy18S1jiZtHAgMBAAGgADANBgkqhkiG9w0BAQsFAAOCAgEA -h9/gIGo83p/F0NDDWGr4P9LxdrE+kkaZF2BxRv2rkty/OC1Qe2lcJPWaEOQY07an -witmxGPaZB6e764J9z/C2l8hoh9LFAIZ+gPBzMEqPBO2/5WYWjjPzzUlkTBczIdo -mCZywvYMrdPQ/F2LnZd8iQO0pmFUGC92vV9Hb/i3y7RthnJWRYv+2F/D5ZWE07sz -xA0gwmnY8d8/G+O2CwgGsHLl2tByfnePDqq+ogwRKXtsPeJwWloFvnPRHE0OmkHf -n2XHbbLMS/m8Wi5utN/LpMV+WITQHMxGPGXcX6XOWVFQuesfI7DKoqZzk/aB2IFO -tu2deFdfnrj3Md8PRAQBe0Ufig0gl6EvutdmiNCeiDloPwV4gLvH7SLQmLheOPP2 -CQZh1skRxhPmnJYD4rrsMGv86dlhEvtnxCO+cUrxnUAAr81BAX/Fo1Img0rPaanD -N4/FG7LnU/Rk8g9roN/8v8s62CnyxcQ4UAvSWAaRrudpErDk+L24Ib4UCtcYiGSB -Dj2tK5SMfcXn+bR2HTdVIKHWHIK1X6bS7Jn9ZXlDg/MCyCILeOmW523FoLfTvNVH -IY9MgNe5KDX7dpPCAJFWwmidekNz+sSbpu6Br8IgWd6SuTEx8Lmb9GB0V7P2CHlE -1ASW5YJMgVPJLQ9LQhRIim2+pL3Pz/SM+oijeSyEoSw= +ZW50MjCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBANUgBLgZEjiW3Lmj +GlCu17h2ERYc75YX+7esbw/iMWMrcrXfvNNutJ4H7hOsV81ZpMEQouHRfog8Inra +KCuwmb28DLhJZJlpE49hZxeUJws4yN8VCCTgr/tLtTsvsSS+TGA58B29HHaglJwY +0w2mOlOVcUxOkjne4VHokLOomhzzqcqTLCjwUslZqRn+SDgXyRw9P5reJ/m6E36d +FyBeVglZvealVp4uK/TTqVVFJYBljD22M9wrOeo9AIvrru5VlgDNHu+rwqgSeE/b +GcwhX2J9J++lwOxsFDR7OILY33yoD7/6NN61SJEoRh7xZTQkr/Dc3HpvjYyj4Yag +dWKq2xDZzIxH+QfEMQssuFvulq5L76TJpXB3ceyCu2NEvb/563p1EvXQbp5Zjz7O +jo1elobs9epvfbCmiANmTxG8GLKteLXfjcph492gdIw0nsV9/bIzE5C7lnff4nEU +9E/uEJz0FTw61VjcZPtpqEzLE/8abBU48pTj5HqKo8Nsjx9hPl6trO4h81yMaqwb +QDmza1KsU+CPIFiycyv8Hn4w6JEjwnUc08rOoQ3e7HjqLNpn8X6RirVQUrwSU7L8 +RTKeOCOBLg6AMXfH/frPRnNQjUG/Z7tBTjTJhm38qucxyHI3J5jwX6vn/jBfdFHf +T6510V0Q9fzgzp3H3fyHpnLW1qxDAgMBAAGgADANBgkqhkiG9w0BAQsFAAOCAgEA +dr0LKtpOa+Xu9PKwnlsM48/ltph4q9+tsu4CeC8XGoLFNbVIALuZZsKZDehTf+/d +bgEtjW8vVnBGAvVodo1MgCHnhPPensDLfyggAULT2X400cly+suGbKeu3kIOlKCs +TQsFdNKOPm17NcpuM1wTik2UT2EWLdzZ25Wy3Coid+ILrf5YZ75djqtxZlYbRiw4 +4IndIjN0bYsn8l6Z8Pt5HdJ1nQnbDZhQrx6FXWZ3eSSmpklfl4O07z0KlXi1Nmaf +OiVcOMvZUnM8pYmNvul8Jus/XmP8x3jSbYzJDNOJ3YV8+OD8DVG3pLM8U1FmjCZ7 +KiR5DNSxZFpHGXhUqDpTrhLgoqGK9chOqPdzU7Mp4taEO9FV8Goc7BCeOKB3Znxb +XDIszs0oBIHO/tsqUwEcWBI0vjyC2pBYQAYK++qwwmvbfWg5lrb7eH1ZO42DU9QD +AVR/5luxImAA11AmSsGf8i+FJ3F63PzSr0uUG7BnTLC03xna7dPdKXS/pGojNVBT +Q5A5J0rB3+4L2mZLE3mjst3t1xHfLW/0RVRqGwz0QUIloZkO6wPN6Jz6l5Q+TgCY +uEks1YN/qlwjHwI3ycT+Hr/sY5igT0OAySo7qa7lN13qTiO2z7eAMDgafNnq34kJ +4OQDCE28Bni0fFRIaqVCqTU31Kei5jbORif2wK81Zmw= -----END CERTIFICATE REQUEST----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client3-cert.pem b/tests/integration/test_ssl_cert_authentication/certs/client3-cert.pem index 848ecd9492d..376c85ab8f7 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client3-cert.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client3-cert.pem @@ -1,30 +1,30 @@ -----BEGIN CERTIFICATE----- -MIIFMDCCAxgCFDtXgZV+Jd7/OrySQd+e1dVblQe/MA0GCSqGSIb3DQEBCwUAMFIx +MIIFMDCCAxgCFAXxDGdWf+MHldd68lQPasjUzyRvMA0GCSqGSIb3DQEBCwUAMFIx CzAJBgNVBAYTAlJVMRMwEQYDVQQIDApTb21lLVN0YXRlMSEwHwYDVQQKDBhJbnRl -cm5ldCBXaWRnaXRzIFB0eSBMdGQxCzAJBgNVBAMMAmNhMB4XDTIyMDgwODE3MDU0 -OVoXDTMyMDgwNTE3MDU0OVowVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUt +cm5ldCBXaWRnaXRzIFB0eSBMdGQxCzAJBgNVBAMMAmNhMB4XDTI0MDYyNjEwMjUw +NFoXDTM0MDYyNDEwMjUwNFowVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUt U3RhdGUxITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEQMA4GA1UE -AwwHY2xpZW50MzCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAKHIdR/Q -waHkIn5z3cc+HNtMfHcKgVGzVDTobggUpWSwVUwa0DMq3OTcNrEnT6zZyUrrv1n1 -6aykGPjRzJ+SZX1ohu8X4EHssBOGaNXUH/AYyfOKMLMGN9AU7dQzNKjuJGkbBXsf -CtmQbQ+/ThMhE4X2bNxEULLudGEfKKQ09LZaqVjhhdVCbuOhx4SPMDJB58Ko10HQ -t7Mh1B3nUTJXFherTr5bcsazQhC6e5i5ySlBnJDnsa9+DMzopiUqMrqqb84WuMKs -zzqjlZXDxTRkAX7nGEU2wjGCx+moPaGLGZs2f1VjBOR7HoNGMAEhmHQhWsIgBCsZ -nDHgAc327Gz1xqsjVd/HrrywJyZfO7ZhxMdmF6MH7eRQjfZGe0+Oajc/7EtFWXg7 -fWdafJ38HGoO8hVjlthKfAFxM5NWjvS7I06IQRwMGGApP5bx3uFmUUixtl/FLa6t -jRKfzaXbX8b0p8HUMbeyvQAZemw+vA+nuKir3DtNIrpqfeXraCMUiEpI8fCRm29S -BvfEsDXCZxBje+nma8g27po8vCaHST+8sjwnNeiW4w6NpQbqqmnvzpf2ivm1U2su -2H1E0EA58zrUoKD13BQzFjccgwodlyutUfk0xYQLrRMOqggtMhsjFDIuNegnPgTH -t7DSyAAg9H0QBXlrd9Ic/OiFMLsb3bu6eeu/AgMBAAEwDQYJKoZIhvcNAQELBQAD -ggIBAHeimTo5afyFhpaH30D9j3EXXExt482nSCPZQbYm+taPVxEiJ4vAs9pa032S -LnA2CC4D74K2Ykd+B/mDGgT5lVpnWuP9VL3wpRErRy6TgkYAJwsEnRLGltNhbuT1 -lup3J4dFgR3tOgwxohjY9FlauZBA5Wu1neZDxXK9UTeAmP0HOb8iXh/goXEvmPLA -HAVHmCrSD0lgEpgB6mg72fb0AkPQq1wlzVBbVtaVgByQP561WmGW6eHO7sqwcO/a -/0Fhd299ChMdnzbHToRt6VFET+oEiCOwF+yEQBRWbjPjCjG+6nYHJh6FxE2ABtEr -Ebr3/7//Q6C8uD32swxXjZaCPEtBC0NNoDW5yi2D7xNHyc+4XHJnRo/v2rPry1RI -Bbwepp2aaCrs38uxut/qXka2xRTyDCimDezJFPxTigJoJ9CgxGTQeJe0R0d5uzlJ -FBtIdyJf6HDKzxNJqB0+wJTYiIiSl0VFPtBYJynMXA82SJuyvCMVgqj+uK4xBr51 -APqdWJR6nBoHaFURD105KiQRM9EVHrbnE38xn4DRN3STeKUlEP94zb3fo3UexJVE -+MWWqNJRdMtUE9j1LRX/P1So4c7BeFp0op0CxJrpXlRmRcWV5lBYhK+WtT8oiZHf -SVSJ8Chol77vm1gVVbJVHIrrH3cfWefv/2Y5fpwuQg6yk/u6 +AwwHY2xpZW50MzCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAPrdk2YZ +HRhd4RPfa89z2Ay/TOby545N5n9R+kpQvyhnCbr41afzIU5DrBX7y8cKcvo7y9Dk +Cdd17Xqn4oYliSvVNI8B2nwkBz87BUYT8pNVD+QUc3Jf2P4Wj6XQM4pd9Ntaw7rO +yIf2Oo3Tq58SSjbXYrqAbCYuX4cs/VWLWyq9PapBwxEPeQ756GZS3Xvq1IfCNIKv +NLYRgExctHzkUWPf6WAS0lKydBcCobYvSMfEzPkbBlIhuDRdMmzuji4FefODY4lI +zvnaD7+IIiRC4+IY9xNhdH6s0UobpSIqLCSxOJNYwqhUQt6gNAO1mdJhUirV/XIl +xG5nCGbQS77yeoBLIBEL1t7tpo3/AdEzkR+/wS11dSpcllSj+7BJaKBhTKVZrX1i +gMqUSbiTF1Et9PnAkS1jtUy1w3Ja7FyPWfM8nt/K6vfNRudg/xwY0iY1RFdCXuMw +kPZSr4W+QryGaPqm0RlpCpLHZBOxBDpf0ZcA37ullh5hjXsn5CRw/0ZYpanqrkrq +2cVZLnY98IrJI1QfQhHlDqUP7prR4Omk8C7edXBqQqE/0mqL7AEhppOyLedLFC7W +wUBepmc1bNH+Ho11CZeSdTZfIgwAcD3v6MiMA5kMTRcW6HAHNS309zNJeDf3Eesz +TBXOSCqNBBbk+oW8bxkTLRdHRgdlLT7N6qzLAgMBAAEwDQYJKoZIhvcNAQELBQAD +ggIBAADJZ+I3CJs6E9U2RjIzi1lMo2sYgdcKJS5+yWW8CNjB+DibKfkWjgWvq2K0 +i3hT0Uc6y+ter4OOeIkGtofOiUPekaZsQkPpi73sabwhDVnlki9QL9Ayrd1qDX82 +fMM5roL7w/a+YdKzTQE9hiwPoQhrpj/2mhu7LeYhidSqwzH1anU5YtTKHq3ZrdGN +imhnklcmbqfcNQU0K2l2bu5vuJXFs/v5FCp72ux2p6QDPWwMbwvr413wibt8o7ZT +bBGsQ1MtfJynRVwLGLosn+2t3NPJTfjd4dMEsZhkDY0EX4vbE1/X+K09EN7jPOHe +aJ2AOt3cO3A2EHCR3Dbmt055C6Lb/YR6s05dX4lBT8zY0knsWSL3R77kQoa3+7oR +hU46ydU6K/Kt67nO938WBvFgI81IatRVKVRsXfTIP2oEa0TkwzuvS7nzj3czNU8o +EOa9ixawVYRlEkcuE4KE7x3TcLEGa1gYJDGbsXAfJct1Hur1SJ/rTDwZvlc+qp3o +wWOLtN0mVHEH1OaGlWmeeTuRG16CuTcku2DYiqeuRNy5eZddSuMOag/DKnIN5ZqV +s1GNrpnxPCxd/KFKtdGl+l++Bc9dBmkd+r1dJ/kRGvhul77Zm2xEnGdyybIs64iQ +gvXq8d8ohbZOPxswiFo3p8fbBjWjv0qm3UnlU3P4B3RDMrrC -----END CERTIFICATE----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client3-key.pem b/tests/integration/test_ssl_cert_authentication/certs/client3-key.pem index 9807809578f..f88456215fe 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client3-key.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client3-key.pem @@ -1,52 +1,52 @@ -----BEGIN PRIVATE KEY----- -MIIJRAIBADANBgkqhkiG9w0BAQEFAASCCS4wggkqAgEAAoICAQChyHUf0MGh5CJ+ -c93HPhzbTHx3CoFRs1Q06G4IFKVksFVMGtAzKtzk3DaxJ0+s2clK679Z9emspBj4 -0cyfkmV9aIbvF+BB7LAThmjV1B/wGMnzijCzBjfQFO3UMzSo7iRpGwV7HwrZkG0P -v04TIROF9mzcRFCy7nRhHyikNPS2WqlY4YXVQm7joceEjzAyQefCqNdB0LezIdQd -51EyVxYXq06+W3LGs0IQunuYuckpQZyQ57GvfgzM6KYlKjK6qm/OFrjCrM86o5WV -w8U0ZAF+5xhFNsIxgsfpqD2hixmbNn9VYwTkex6DRjABIZh0IVrCIAQrGZwx4AHN -9uxs9carI1Xfx668sCcmXzu2YcTHZhejB+3kUI32RntPjmo3P+xLRVl4O31nWnyd -/BxqDvIVY5bYSnwBcTOTVo70uyNOiEEcDBhgKT+W8d7hZlFIsbZfxS2urY0Sn82l -21/G9KfB1DG3sr0AGXpsPrwPp7ioq9w7TSK6an3l62gjFIhKSPHwkZtvUgb3xLA1 -wmcQY3vp5mvINu6aPLwmh0k/vLI8JzXoluMOjaUG6qpp786X9or5tVNrLth9RNBA -OfM61KCg9dwUMxY3HIMKHZcrrVH5NMWEC60TDqoILTIbIxQyLjXoJz4Ex7ew0sgA -IPR9EAV5a3fSHPzohTC7G927unnrvwIDAQABAoICAB52pRtXD0cBMr+V6MJuPzfK -GLu/picwud/2jlWGMbiafY1FlUO1Q//yOXg1O0sEfWNWreLuK9Ce27bqpnm7stGg -/5uA5vDy9RrQGeh9t3/Z4xkxQcdMGfFcJ4ZoF/fhU8jY1jjfWQcrq2WmM7jlZww4 -ITp+wKVYwmBRYjE9aYD25GGAoddM2Z2SZoPfBIfKIi5k5ZeWiii4a30wA/mTLW1K -jBrLFHs24O9OdhyFi0vx913PywUNGrLw8ewYnJHKqx/m5mZ97M2RZQbc5MLfO+rd -8BOEYp+5DRTB4c0L6MfxxJta+F5fkpjMfBiSb2caqsDYD4DgIym/EQfsAYvo4b4m -R42rfg5SEiLV16K3ePA2lEHWGmgzsY7PZzmJUGtytHd+NrHMBWloGp0Jke3LV30g -+3WnvXX+/MZ/dtH2/hy1qpZelZR32h1fchSnKqGUnSBcfMjP1YQY6FrTqNxnAxbx -ufLve41IN0+rCScZ/bp9FEd63DfdAi+U+RKSYjF5promG9ao0u85Yw8iF0maTbEn -oOS7759oGi6Y8udWuAXqmbo1JRw7ZIVKiIU09cR8/IgnGBgOO6/xSOrY+nUCzFvz -gVYqOSWpqBilrZF70GUy8iRfPWUAen5Zm58UvyYmByn0xV/aUKyKKcxzfGeQ1JH/ -NCdITbwdLyVhqiKuw9l5AoIBAQDTiXYWOjyxkXQMlYczi/w8pRaCvYMNJGzM2qmP -hn9ChIoi2STv4DuqYUmbW6irT/mtcaTA8EPwC10gpj0FWXtM9YYVhnoXl7rfEZUy -oYT9DFRGx25cR6krqg+sxqV22He8Ld618jn6MISrQS7Zd3rLx9goQXeTkLfD7Zdz -fwduSTMKBFp6zQdBlWf6PhO0vu0Cjhhb0Mr9pTYOYMmZhuIKHeuGU2kZT+Myxn0f -+mlkEBQ5bGr0h9Su+ROvU0vXMmr7SklKbLk1TZDPkduqfub40N63aapto3aVMjUg -ff1urZh5/wiAgKzuLcucLapAWCkdhaoLa5mDyug/yRLs8vRpAoIBAQDDyczSH5CO -deWarOTcREWeFc0Up4AD6EbVZBiMDuIMEtT2hMO3gLiF+/XvEptBO46zMp7LK1hY -E7IP1f87OiH7rOtQTVtsVD5UWIYQaikvCG8OOtOM77zb/OyNYx7mp1Ne2ZuoPKgm -pAhIhdFw9XYiEsdBTnI541htujhpnx+x8Yb8H7CpQ7WR29KPd2sytb0KqFWcEN49 -fYyC4DhI1Y86VzuU3T6SLdBqtq0FtA9QpXYvoRvfqoK0wMb/k7m7ptld65fIQ0On -9pnJCEk63msWSVxPbJzwfQOgEeP9f+Blg8S65kn2kb7DvwD+SAfhtjItb5RLbJpC -MD0eND8RtDnnAoIBAQCyn6SjaFHP7277IVilYlOAMjcr9zMuac+lvA7qLzxOwvSS -MKJMHCDF4pjxIxjub/35Q7lHYps7m5zr8PQeDE1d0wWCL6fxPCKEMqi1tEZEF6Ei -k0zRh4GGaXgLAUK/dFLHCETDhuMGLOoaYlRZNdbvNLWGaKG8bbt/KqV01ZNEwXLj -xoFQqMizzKGcYcbqHT6tpadUAJ8oR9W5lmklxwwgVtuG9fANe7PyMEGAO0e7QwiQ -5Mf77KbfiVvh9IhaEyzbPQNeYSiTpgfd/uBqu/X1rQoj/on3Qszjdx39beYPC29x -tLVNLVrZVLpWCwl0g/1T/IZ2VkmvTCuJkRT2GMHhAoIBAQCXaOZG0TY/lZk8ptxO -I1YbTQzOHkL3wpeUytXY6mGRqLac9ktNC/SnWRT2D9OU/PP8TPdUc27cnk6jxICf -UvOY6D8KWOiMvBF/tP3oj9DNmJ4ZyRB0+6l4Dv740GDDSz9EKNEhp9b6Gvx06Vfx -HUKAUzlYncvkTJbENLEPrYkbWYdkTzWVkNFHvVH2tQlnq/hH16ptCPZ1YzRYugGN -AerD2VYwZ3DOJzP4ctEXigoV+f+OEe+2Zuyx0CuU1q9aGUwcP+efbbVSnXiMaSYI -qpzyHUWP/pTXvjYhgfRoxI3Ks75eM12bm1aFlp60BwxhVz8yuAlc0t3wtdFuHEVx -8YidAoIBAQCQYRslpvkKLynGY18m4oViMjaCvt27fepngAuuVFEZ9cJZeWY8GMcc -IJeq9qTtlMuiWHnq8oGVCL6BnFX6BIMT0W1knCSRqHcwjQByZ51X2Bo60TktdPOD -c6lILIdkYCFLs7fXv0xUZihyIIdYaxx3XpeNwgaqM+wFsbSclF3U4cdwm5U+ltOZ -L/3w1rFlyH/+ZWIItBC8N9pHD4bayiavHT99E+35Vgtol5jqIhKsiGVgoj3tdDW1 -+xBdyrg6JLXHFP/vobY5mvLLGdcwCXEd/b+jUK9Uhdbq2VC27XEllIwGpPK1SnNU -7tLJO1z1/eDbntbQC4cvewqNRYhwlnWe +MIIJRAIBADANBgkqhkiG9w0BAQEFAASCCS4wggkqAgEAAoICAQD63ZNmGR0YXeET +32vPc9gMv0zm8ueOTeZ/UfpKUL8oZwm6+NWn8yFOQ6wV+8vHCnL6O8vQ5AnXde16 +p+KGJYkr1TSPAdp8JAc/OwVGE/KTVQ/kFHNyX9j+Fo+l0DOKXfTbWsO6zsiH9jqN +06ufEko212K6gGwmLl+HLP1Vi1sqvT2qQcMRD3kO+ehmUt176tSHwjSCrzS2EYBM +XLR85FFj3+lgEtJSsnQXAqG2L0jHxMz5GwZSIbg0XTJs7o4uBXnzg2OJSM752g+/ +iCIkQuPiGPcTYXR+rNFKG6UiKiwksTiTWMKoVELeoDQDtZnSYVIq1f1yJcRuZwhm +0Eu+8nqASyARC9be7aaN/wHRM5Efv8EtdXUqXJZUo/uwSWigYUylWa19YoDKlEm4 +kxdRLfT5wJEtY7VMtcNyWuxcj1nzPJ7fyur3zUbnYP8cGNImNURXQl7jMJD2Uq+F +vkK8hmj6ptEZaQqSx2QTsQQ6X9GXAN+7pZYeYY17J+QkcP9GWKWp6q5K6tnFWS52 +PfCKySNUH0IR5Q6lD+6a0eDppPAu3nVwakKhP9Jqi+wBIaaTsi3nSxQu1sFAXqZn +NWzR/h6NdQmXknU2XyIMAHA97+jIjAOZDE0XFuhwBzUt9PczSXg39xHrM0wVzkgq +jQQW5PqFvG8ZEy0XR0YHZS0+zeqsywIDAQABAoICAQDAMTs48CqTPXEvyW6OS+EM +uw7OrO/r3RCnIIYRo1UgPfh9byA5AJLWpA/V88eF4SJ/RYp7qglEMcvTuYVZYq55 +j2kp2rCphOysa6o5qxSf/X4kLerYiEf1OhGpZh3mdt8doqbrmnqVd3YarD0CrH+B +DnhMDBFPGx4CsNwRSqd40ezJYIJyspj7eUisA/Y9doaGz6ltKY/HoRba6fc4667T +RntEKIdL5f38lv6PViB7M/IZMrQf/kdijrgQLp9s8LMiddmvFsHDN2XzRfdqMnjm +AlxgU7xtRDc/gHh9+TNClSeT81+GmK92YeQXp2yGehr6SGFYr0iTkIomQpSVYK2p +0haIIjQMHlc7E6WVkDELdpAxERgvV4uDN9iEkd4t9oNDPPRioPJQ4bhbMSxCO+CP +NdFHTxIbaDr39OdgqNNE14j7WJsFaCsYXH2NFF8jvwIkPQ3QVMQT/JPGStkyF+9P +5IjFfQ9aEF2i4mAVYiG0DE3NyD/OOI9/uF05POn15H9U+bA9hfBE0Rtm9nMqfVy+ +zgmajXkVb0jTHdL2t/UKv0YdgaglvDcWGFdEUskjJoB00NJwBGorSvcMZiSTxpLD +cGRqywRHOEqNIAbKv0Dt2AX5ZdBSQu7/z1/5Jcdmx8vp9lVhQKeMzYxsFKE4V7fr +ztDuPOlFGyffxpRenBIxUQKCAQEA/XVyoOW1cSFqeG46mjw+dbwjqRmLtEVhAMsG +TtW8pnMJHZ8u7lfM/UJyMN4NQEPJElrABns6I3dqPOwaKOy1a2leHMg5Vvd0/uqp +s5a2fduP6l9PXvhhWDN2sChbeKhl0jJDVnaTO7tiye8ZGMYOM/AlfQX/+PY4QNgd +O7UwcLKhoytxtPtHFZTOZp+cECdTvlmX9lZoNEzFp0nfzFaLVwDsy0B9e6KGt1xJ +fV3Drw7p7PeUyYBNKkyCRVee5S/pn5fT7pkIxMHvaL9BBnWVpwiH3Vi0hfTfFZk4 +8tLcVZgf3n0Y4dMVP2VQRF+kKBTL0coLne36HksQEJyk/4KZEwKCAQEA/WF4z6kc +YXwsU5847+ywq4ipq9efadkMDaGzI6Ez06TQjRYNsZGplCV9fiGxKX2YmZyFzjTf +4joqOmI6UANk+JZKW0Eyyak/TnxugrjMFq8WnK64cIz1TK054tAM/bHGkavaYb8K +bCfbKmaSkwkTbb/OasbQqsC7jbALdbM6Ae0PMrpPmI90YYIMYLRogIaBqCkB43vp +GEZN2VeNS7blhRMiq7YBDXn807aSMQ0+skNSQ7MA8F5i4BFvWyPb1nKZWux1RWLZ +O23IxGWmoGho1CAaEk55LXbqLygU5ZYlBSqkrP9N/elJykOp0LwpjoYBgjMPmanz +o6jy8XIUP78MaQKCAQEAi8+YjqaHosMTDyGG1AN9VMaWSTYdOTC4JI7ZiO0f5hU4 +pw1i/viRy/Y2NTyXxKZfqO9EU47v8BZ0FO0MNRz1qi1yS6Aq+Q0BjYh2Wek9+0j9 +JwSyLKoIUHX694sbggAqQnuVZ4F7EAz6nnd0uZSuyvmiREfl/jgbqbFM1t3IvbHb +tb1GONYPTRlLjZJnrQV0jWCwkaLyUj8zHGeEuxvWOwT4mdmWHnf1pfmTVEM/qTYp +1Zxwh4JtjnKrvYJq1PPMBEvlDQ1/p8FuxbISNXTxOzVadL/0vJvp3ukpX9Du14xV +sA4DhrZAVzsUvtKfI7jtAWlZZSGbwdAYKYGvBn7M3wKCAQAHZWX6YcxTSCWfF0G5 +NyZ9C1Mwke20UEKaz0KEYrs5jVENHTyvFzpk+actHFyogmMG8NuzBjYWy23aIG3l +UgQLgY+QFFogKtGPP/CV3kEO1HOLhUoa9vJeF5xd84a9jQfnzqVkPwhV2d/6392d +byFjDbs/wKfspA2VeDMNb3rc/Yd5CpkyMdXK1tn3pKx8O/Di8Ld+ZWqLa9nv4y9b +q24NsV5MttZXB12K7IRd7C4NVAu9sCbx3T9znO6sMWLEYrn5Pne528XNh0nZ+cGg +YwvUTU+VgzbkTdlOIRRjEzvnZ7RA3H7xT3L49XqqfiOUZnL60vS8nopfF5pn09Wl +erUpAoIBAQDWHJQT+Jvj+dXPC42oIRQKCiYtp1buM8YyL+dJNi7p73brF+2Oqx3k +XNT5eP9GthGqpVGJ732FWJDbPViuZB12zlx9tpGF3ghQTq3p/95KOhGEb2fG7mnl +bEcPqOoFEsAlc4DZYqsDDUvmsifimKm20ZWi4VjTqQJUHYCegJsjrA7D4obGbOxX +FujRMq7/idXRjEoWLloQTMPAQ0Uu4Omwnea25daRzrrrJ34uYrTO1sNgOKk9JAem +rGgrOzsRVG1aNwddcZT/t/icLKS25G7AszLnrNFxJB7DRAfgzpHkJBwNLpcPVtfR +KB6GTGRi7uqGHYScU6+wRMHjdVzdKGNM -----END PRIVATE KEY----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client3-req.pem b/tests/integration/test_ssl_cert_authentication/certs/client3-req.pem index a2b19bf835b..7c679b4b367 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/client3-req.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/client3-req.pem @@ -1,27 +1,27 @@ -----BEGIN CERTIFICATE REQUEST----- MIIEnDCCAoQCAQAwVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUx ITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEQMA4GA1UEAwwHY2xp -ZW50MzCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAKHIdR/QwaHkIn5z -3cc+HNtMfHcKgVGzVDTobggUpWSwVUwa0DMq3OTcNrEnT6zZyUrrv1n16aykGPjR -zJ+SZX1ohu8X4EHssBOGaNXUH/AYyfOKMLMGN9AU7dQzNKjuJGkbBXsfCtmQbQ+/ -ThMhE4X2bNxEULLudGEfKKQ09LZaqVjhhdVCbuOhx4SPMDJB58Ko10HQt7Mh1B3n -UTJXFherTr5bcsazQhC6e5i5ySlBnJDnsa9+DMzopiUqMrqqb84WuMKszzqjlZXD -xTRkAX7nGEU2wjGCx+moPaGLGZs2f1VjBOR7HoNGMAEhmHQhWsIgBCsZnDHgAc32 -7Gz1xqsjVd/HrrywJyZfO7ZhxMdmF6MH7eRQjfZGe0+Oajc/7EtFWXg7fWdafJ38 -HGoO8hVjlthKfAFxM5NWjvS7I06IQRwMGGApP5bx3uFmUUixtl/FLa6tjRKfzaXb -X8b0p8HUMbeyvQAZemw+vA+nuKir3DtNIrpqfeXraCMUiEpI8fCRm29SBvfEsDXC -ZxBje+nma8g27po8vCaHST+8sjwnNeiW4w6NpQbqqmnvzpf2ivm1U2su2H1E0EA5 -8zrUoKD13BQzFjccgwodlyutUfk0xYQLrRMOqggtMhsjFDIuNegnPgTHt7DSyAAg -9H0QBXlrd9Ic/OiFMLsb3bu6eeu/AgMBAAGgADANBgkqhkiG9w0BAQsFAAOCAgEA -f8vcJzjwqm2yUx1gYOt/BzfZ4+VNvP5CSIBxRAkT4judk4Wz07Pb1dQT351HcU8X -+pJ55HfIgUFyOSR1gKGJSV1HsREqYyaJV2KeBQM+klEeagYR+0Dt8R7NYTRtqUmV -lyoW7eHlUsbvUa0jCiwOK/t7WDr9qH4ZUKxVYSpJNa8FO058SoUcCRFue5TnTxF8 -tHH+J+kzcagcS0Rk5CCFWCtNE8+0FdfUs8IUYaV8cw8PEqdfrfJ2f/Zj0I37rh9P -pjuqe+GGPp7hv29YJ4bRd5TSe05vol2g+LYx2JNe1sr+NnGZVDVolsTg50cEwBo9 -gLW0ea/4Y+OoOAqFOdVM+RvfEbgpsT0LpHZAKXfiGi1PAMzZ0bJcOH8F77mV7OcR -qNcshdM1LkMSojGvoVQrRP/Bz0CVjSpwBcmkGiehESkaxNNsUyQBla84v0GDvuL6 -cA6NDfl8iPz5W3kk+2fypgO7sw0FXQVKjq63gz4XAQsGP8JzF1cC4fDnoRRsHO5E -UdWE98/AnVZ7mQ5bC11TAuDyzKGh1FNjrYFmsvTnMIWo3Ef5Tc5GXfYC+fVryfDf -BAbw71FprzMMFoAIxiSCPzK6y/am7BdGM5IZN09V4BBMg8QwZiXtzXWH5JX5PKm2 -f15IkScIvUliS0RepLfI0CXcFuzpJKi7eHLqca0cli8= +ZW50MzCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAPrdk2YZHRhd4RPf +a89z2Ay/TOby545N5n9R+kpQvyhnCbr41afzIU5DrBX7y8cKcvo7y9DkCdd17Xqn +4oYliSvVNI8B2nwkBz87BUYT8pNVD+QUc3Jf2P4Wj6XQM4pd9Ntaw7rOyIf2Oo3T +q58SSjbXYrqAbCYuX4cs/VWLWyq9PapBwxEPeQ756GZS3Xvq1IfCNIKvNLYRgExc +tHzkUWPf6WAS0lKydBcCobYvSMfEzPkbBlIhuDRdMmzuji4FefODY4lIzvnaD7+I +IiRC4+IY9xNhdH6s0UobpSIqLCSxOJNYwqhUQt6gNAO1mdJhUirV/XIlxG5nCGbQ +S77yeoBLIBEL1t7tpo3/AdEzkR+/wS11dSpcllSj+7BJaKBhTKVZrX1igMqUSbiT +F1Et9PnAkS1jtUy1w3Ja7FyPWfM8nt/K6vfNRudg/xwY0iY1RFdCXuMwkPZSr4W+ +QryGaPqm0RlpCpLHZBOxBDpf0ZcA37ullh5hjXsn5CRw/0ZYpanqrkrq2cVZLnY9 +8IrJI1QfQhHlDqUP7prR4Omk8C7edXBqQqE/0mqL7AEhppOyLedLFC7WwUBepmc1 +bNH+Ho11CZeSdTZfIgwAcD3v6MiMA5kMTRcW6HAHNS309zNJeDf3EeszTBXOSCqN +BBbk+oW8bxkTLRdHRgdlLT7N6qzLAgMBAAGgADANBgkqhkiG9w0BAQsFAAOCAgEA +WLhSuFZ6pnoDe8LQx6eMPXzRkQb1qsJpyjpegUxFe71o2e23V/1yMnTfFiO+DsBQ +PP8RkLWUKAvkAvqPyttJBx9U5ZYspsSsTVhPsCjUFZ4IG+fc/dVP1ZRid5HQJz2+ +bFf4KPgErZkJZR02Q2q6ZpKq9clRzbDkho56OZXLYI/o2Z4xADbhzpa0xt8sx533 +bm0rKvz85WxH3cimRjKaGKzuKg38ZaXmmUbsigV3dzImT00KDWmMmaW9SB8lIm2R +JToms0Qs+mOr9qD2NiRoiUd1wmgG2QpFDViIqAZKJjjeesmeV2CAcPfLztOZBim4 +6bRIOIXDhYYOyDgs52XuijXUr4BR8aQmqBrjnccCMcGE8Ol5ZH/IDg4pCRSduCWe +T7ThhH7BpAWYdgF3ITcp5oEcpXK8IdAMAst1/6vk7Z1JHIOejxksbLsGDYkaLM6w +yTn4X3Ak0X6bVmLAY+xAL/WjAJhVtDPqGYAmpx4iQ6QjYG/8gRdOiUI8H7MCK8+h +P0auhyyMmO+kdhNnzwuX/eeLXZfNvnyK4n2uHWYgwV5I+Kv282zw94UIQgwVQ2DN +/IbXD7K57s7+ff9Eff8L/B8rt1i1cmv01mEgQ4kMsLOClGaceGcz/ivfzDCosmsk +Xg/zVmdunUY0lswYL4SQM3BhWB3xJ4likHikfQHklM4= -----END CERTIFICATE REQUEST----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client4-cert.pem b/tests/integration/test_ssl_cert_authentication/certs/client4-cert.pem new file mode 100644 index 00000000000..5eae58da627 --- /dev/null +++ b/tests/integration/test_ssl_cert_authentication/certs/client4-cert.pem @@ -0,0 +1,31 @@ +-----BEGIN CERTIFICATE----- +MIIFWjCCA0KgAwIBAgIUBfEMZ1Z/4weV13ryVA9qyNTPJHAwDQYJKoZIhvcNAQEL +BQAwUjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUxITAfBgNVBAoM +GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDELMAkGA1UEAwwCY2EwHhcNMjQwNjI2 +MTAyNTA0WhcNMzQwNjI0MTAyNTA0WjBXMQswCQYDVQQGEwJSVTETMBEGA1UECAwK +U29tZS1TdGF0ZTEhMB8GA1UECgwYSW50ZXJuZXQgV2lkZ2l0cyBQdHkgTHRkMRAw +DgYDVQQDDAdjbGllbnQ0MIICIjANBgkqhkiG9w0BAQEFAAOCAg8AMIICCgKCAgEA +353z74lGXkwEc1n6r/M0FS1XTXhoVGMYIqK7HPCBOEGLeyyGwGfSQ7lRfT1xSkii +zBGG0Nod9cRT1CAewOSJ6BjVfkQcGEjlnVYm42nD6PMd9iFJj9Y5atPeFNvvr+wF +OFX+E8FRu8u9aEu7MIj+KCqoqBukFhFgJYX8sMbRROfLOPaCq0cSC+Vod4qR83+W +ITrQ5n8+/CC39uLY/oKgAKdVnmff595Uy76BVdYzuit1IRKwJxqIWMRrfNI+szmS +hdj0AHwgmwEGCaTNcOQyqvBLxW6qB5tc1FyV4LYv4iNftroqNQvlUbJ4UqVr55Fh +vZ38C1BQ4sWgo6FSS/B6u13clwpRzDh3H8tOMTTz1inUtg61Y49p2G8k3kNVH+QU +fRM4xvCkhFzIArgSiJ+/YUKboltSG5K28pegkk8RRMsaQK8g+NScKKu7/8ddRGE8 +454AqxPpzASij+djM0vxzgad6BB4e+iIVdj77NSjAxVAfg9GIjNHG1DZ87jLLgtk +SN2jaYsBRBRnmenslEGDtwO1SeWrzzicVfP9GRdiMLJwCkwv5P5hCzwrCB5eUPhm +tGHm4K8eXDAd+Ol9pKMySC79E5/W372wdbaO1fcAUKvpHhcRZnSusNAJkLGJYCkV +2gzTWlaeX4rGqjNVs4MSmNuMT+a0IafZeZivptxdgLkCAwEAAaMjMCEwHwYDVR0R +BBgwFoYUc3BpZmZlOi8vZm9vLmNvbS9iYXIwDQYJKoZIhvcNAQELBQADggIBAFox +3myVDr9yJkrF5+vB9gUlTv14JIPRd0OFCLcPOlHpvYKEwjRjTwT9oL3zU5PoRPX0 +AiD9sL5TOo0zraiFPUi1k5X6SoW/qU/kOJ/j5CgfChyyyit/V773LitM/cVXZGui +YX32V1zV9+RaCowC/16oHvfjMA8xNOYoYW83FgQ3GrKgRuqqVMT7JAHoDebVSqyb +w5W0G7RH3hHM1nCv51tnT1SZDn+qRBcX5faPUVARzdcRrZ/VSU2RoVIU/fPPiet8 +5TRioZFslZaFDWOLOuP0ZcOj5MsY3vQZtx2/NRgNc+iLF593YBUhRJYqfT5ePW3H +LwbZp/Rvd2kLucYd/W9WhKEzJKKvzm1V2hCDnh5dl32sZgdBzrdKzgyNB723cLR2 +cHFTIEj1Q/scay+iiSoV+VNfMSDQ71vkHqFHNhEqPFUpdF/SeooDFeQaDvYkomgr +Z9BJFtbp4kZRIEuPX+niTi0S/zwi7htiUn17wOIBcydcgG2GXBer5H3JyFnCXM1N +0jFQsuBFRj8xP71xzhN8YjA2Pe+MGYrMWiwaVMLTz8mdQ+Y2aEvOkfXFSaeNUqW3 +GYxAjEkhVCvzhOd6sD3QjLRX2qhwh8NJCJDbkTok66hno8QsHWASbaDiCMG9z7le +ci4dOHzu/buwqS4LVTmFWTn7mkd+FAlSY9Hj0WVI +-----END CERTIFICATE----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client4-ext.cnf b/tests/integration/test_ssl_cert_authentication/certs/client4-ext.cnf new file mode 100644 index 00000000000..f8425c275a1 --- /dev/null +++ b/tests/integration/test_ssl_cert_authentication/certs/client4-ext.cnf @@ -0,0 +1 @@ +subjectAltName=URI:spiffe://foo.com/bar diff --git a/tests/integration/test_ssl_cert_authentication/certs/client4-key.pem b/tests/integration/test_ssl_cert_authentication/certs/client4-key.pem new file mode 100644 index 00000000000..f1f17525a51 --- /dev/null +++ b/tests/integration/test_ssl_cert_authentication/certs/client4-key.pem @@ -0,0 +1,52 @@ +-----BEGIN PRIVATE KEY----- +MIIJRAIBADANBgkqhkiG9w0BAQEFAASCCS4wggkqAgEAAoICAQDfnfPviUZeTARz +Wfqv8zQVLVdNeGhUYxgiorsc8IE4QYt7LIbAZ9JDuVF9PXFKSKLMEYbQ2h31xFPU +IB7A5InoGNV+RBwYSOWdVibjacPo8x32IUmP1jlq094U2++v7AU4Vf4TwVG7y71o +S7swiP4oKqioG6QWEWAlhfywxtFE58s49oKrRxIL5Wh3ipHzf5YhOtDmfz78ILf2 +4tj+gqAAp1WeZ9/n3lTLvoFV1jO6K3UhErAnGohYxGt80j6zOZKF2PQAfCCbAQYJ +pM1w5DKq8EvFbqoHm1zUXJXgti/iI1+2uio1C+VRsnhSpWvnkWG9nfwLUFDixaCj +oVJL8Hq7XdyXClHMOHcfy04xNPPWKdS2DrVjj2nYbyTeQ1Uf5BR9EzjG8KSEXMgC +uBKIn79hQpuiW1Ibkrbyl6CSTxFEyxpAryD41Jwoq7v/x11EYTzjngCrE+nMBKKP +52MzS/HOBp3oEHh76IhV2Pvs1KMDFUB+D0YiM0cbUNnzuMsuC2RI3aNpiwFEFGeZ +6eyUQYO3A7VJ5avPOJxV8/0ZF2IwsnAKTC/k/mELPCsIHl5Q+Ga0Yebgrx5cMB34 +6X2kozJILv0Tn9bfvbB1to7V9wBQq+keFxFmdK6w0AmQsYlgKRXaDNNaVp5fisaq +M1WzgxKY24xP5rQhp9l5mK+m3F2AuQIDAQABAoICAAQfLTflF971F7/okK5dlUAu +rcVHyuSDTxaUWU6XQEqBKskCcRlq0H1fFRlx4Hy2Cgoo6ItA+fxlugXW8bosfD5C +9ux05O+tqE3WILFgabQJhyvaQTjdggFuFlHcG/bqKs53B0/l6FPF1Z/uhWzHmaez +4Zf3qnadq2AFsDqx73mNrDlIkfAGR1bgy6QocbhDSckjBGa7QbX0BHAQjl9imQBq +FTHuSDpF5to6kLe8UwfDdU0+wvB1lL3OIQ0T8wPqs8Cz1wuLPi6dPjc/SmoiSqzL +8RmaiJfLTVK8wiZ6NTe93y3HELAZoAh5ea5MTkjebSbJmrO6r0L+0Y8ykgnETP7O +Ug9PWeDDE15sNXIQCKtRe3QpHtJaoAJU1MGhNqwm9oKMcuSvBOV8XRuZORinTYRL +Q2ZD7czaT5VZXCQI4uHwE+KIlQF+658c9M9WETxClgUlhbzqig3ilUz3QUweaPvE +tqArjiYLsT2KtrgmsZ2DaDn2IlGSIRjXMZJ3kn6i49C0uhH2YkSZgU9/7kH2myse +3opxE1EbT4ARFWUbGgqXTOc/OSb9DAsxUK2u0eR/4yOMLQJkvxNJWQgwmi6N18iU +WdvTphNtMtmdsAhst9luwNaeJItzTDm7JeWx+MPs8f7PVOOkTz8HcBAvZnISH1Md +0i+0lBrBXbAcRK5X7tvhAoIBAQDwKPQA0uNk4Hemt5yke1jrg4B03OLimHeQ/1PY +I99hThh/RLncYaMxqsd5WkXXbjsWyGidKHYh3/cG9akmgU6D2Z16CFNMRhgBgRX2 ++LJkdS2QSuHPJlB9ERtOOiWFt7IDafB+tMKHE/VRQdxFRtvLe6pQMzP4veVXZsq8 +NNJGAQ8egUa6HDvkXzR2VDf2Kc61t4ZwT4JT6C12GnCfvXobaVkU1aWhcguoX8vI +o3UOkeracEKc/80ZRdFhA/nvPPXCobyjFjLi8WGp6PUySrVhN9eAdZaUBNeFHLdg +8urNvy5Q6mBAByEfHZeJNZbBeAEAw7S5YAxgL96blj2IPOerAoIBAQDuXazpCDXD +dG6XDZ9FS7MKBWriHUa54UgbX4JfQsOOFA8uxglIe5E4IKFSEbetlDORnKZjcmGa +SKTm0MLLi/kKrkhoDgi6HNbmbo9ZmKIhmEwws5L1TLeUdLrWj8GLgNphotOBKs1V +vQQkfh6rzovyFsMj44Xea8Kgx5ONVlB1L5pEepKdIyDRiQfxhwFox719HACSqCEa +06eFNGtUOLLqNMZhpur59bqgiVQtIZKis9juwzZID0svXBElpDrNvbWS1V5MCBOT +6AStW66YkmVWFCn7qQmNqMh4x19GveW8ajgrBSr/8GP/WXiACBDEsunWRORW57iS +KiPmC0uHlMUrAoIBAQCYTrCokRZLlJvtbIb4PY3gFw7xjmCJqn4xw+wNqHpzgI7C +r/hbjsRrrE5DZP/kJ3Fr+n92JAH/a8WDcWrsE5eSwQFBMmR5e/6ffZlLft/MHBBg +cU0SDc9/8chqbS/8xMotphMymDrCZeLvvKAQg2bDftM9d6ufNfdr3bH3eFxery9C +fmQ3hc5qAAMKhFDVWiBRWGn3ckVKJ3Ylb5E7jXQSTFaFgxU+9U/1YYOg5CFJszrJ +e+aTIRuWypOGPnpUwkluPRqgJ2TwTntMwYQ3d+/eDwcp3ek4SHXSYqrd3lERWQzr +niiakqrry92d1BGe8xdXv8Yuxn4yxkkcTUUK0O1vAoIBAQDLY0LW1BqL3B1A5m6w +QhdSxaydoz1l/cP5F1W20tDpulP6JSBmqIkQy0bbMCL6CSq3ZGLVGBQQAUwzZo3Q +AG9PncZKgy8PHux/UncejA5LfBgGtjL++6bpFXEXAzKyRhAQn065ODxcnBucx8CD ++ImQ17tKNClVz70SUzijsLKWSzfmlm/jhMXMBJCyle+t6EDXL72NZchZi5+1GTU7 +d+Wx0bY0PKji/7luob8hgzQLgEnp8MewVNxiXLyE0c0bIHR+BXGgjoOmAKN9CG3B +4ah1+l6YTXPJW+syo2u4gPA2BKxIiPBX0laA22bmV/t22vKL0dzECpSCo1JeR+T6 +mwZhAoIBAQDpLqRLxfZk2TK3wJ/bloXXsRg4TjSQ4m2Y3htVRiOQF83iERVRlAwg +9yKlyd99ux8tlYAK368Q+FFAwYTvUyghmfVTPARFTmeX0F5u+MX00WOa2FhPs3du ++ImYeQH3hg2O7qyVDMCwtqgIIuGLNwsVPqUUF5bx0He7wTwwzQmx3EVCOu6yZXG7 +Aw3qpOM2VhrtWgGP1mTONiUg5dh4sGbXX70gjG9cUpo/Owr69Q4Y8/OEyx9bzqSW +5BeVN0vONzQC+LHG5EvgNF6yOU7iCkuoDirZUrVchuAf+IDapK85TLIH8bm57LKN +Etg/x+MCoqlEQBVgnY7f3suMB89XerER +-----END PRIVATE KEY----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/client4-req.pem b/tests/integration/test_ssl_cert_authentication/certs/client4-req.pem new file mode 100644 index 00000000000..224484f6611 --- /dev/null +++ b/tests/integration/test_ssl_cert_authentication/certs/client4-req.pem @@ -0,0 +1,27 @@ +-----BEGIN CERTIFICATE REQUEST----- +MIIEnDCCAoQCAQAwVzELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUx +ITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEQMA4GA1UEAwwHY2xp +ZW50NDCCAiIwDQYJKoZIhvcNAQEBBQADggIPADCCAgoCggIBAN+d8++JRl5MBHNZ ++q/zNBUtV014aFRjGCKiuxzwgThBi3sshsBn0kO5UX09cUpIoswRhtDaHfXEU9Qg +HsDkiegY1X5EHBhI5Z1WJuNpw+jzHfYhSY/WOWrT3hTb76/sBThV/hPBUbvLvWhL +uzCI/igqqKgbpBYRYCWF/LDG0UTnyzj2gqtHEgvlaHeKkfN/liE60OZ/Pvwgt/bi +2P6CoACnVZ5n3+feVMu+gVXWM7ordSESsCcaiFjEa3zSPrM5koXY9AB8IJsBBgmk +zXDkMqrwS8VuqgebXNRcleC2L+IjX7a6KjUL5VGyeFKla+eRYb2d/AtQUOLFoKOh +Ukvwertd3JcKUcw4dx/LTjE089Yp1LYOtWOPadhvJN5DVR/kFH0TOMbwpIRcyAK4 +Eoifv2FCm6JbUhuStvKXoJJPEUTLGkCvIPjUnCiru//HXURhPOOeAKsT6cwEoo/n +YzNL8c4GnegQeHvoiFXY++zUowMVQH4PRiIzRxtQ2fO4yy4LZEjdo2mLAUQUZ5np +7JRBg7cDtUnlq884nFXz/RkXYjCycApML+T+YQs8KwgeXlD4ZrRh5uCvHlwwHfjp +faSjMkgu/ROf1t+9sHW2jtX3AFCr6R4XEWZ0rrDQCZCxiWApFdoM01pWnl+Kxqoz +VbODEpjbjE/mtCGn2XmYr6bcXYC5AgMBAAGgADANBgkqhkiG9w0BAQsFAAOCAgEA +2JVul/xcJ+YlepOHxJ9dczIcXEjjMOBxWuyK+G9/6wASgHg9e2SB+WS1VSeUARC6 +VkID3Jlwr1gEw4gR3lW2h5I21kdCaBfCHshUoOr9rV5uE76r9kxgyEsMUZMvSClC +eQd8VK4fUT9JEKigIJeCFT9IE9PyxrdH1xpp89jOLy40t3PkubDi8WR8dvPckg3N +juLU/6EtbrtgFMnCqB2TmH4mc6YSCeENUTvt+nSiBKZUblGDuIxu/edX3SscS0Yv +qPM5LPcNHEGeeMC5ZSfotaSzRP+x3OlV9VJNROG4brbaI+3kECtegBgFvKIiK+JY +m7dkt8oIpQc8CKZkM8Kk6e3JXHzKf4vAiWHf0Wyag3gqCukxdas/PMx/3ROi7iDm +XQN713lxhIjtqfXQZjZcRmYQwdnkaSY+H7hgAyhnavqkBmPLeMU5hffdBswrjH+0 +fD0FOIDOWNM9e2Q/qdtHxtglNUmox0ETvl/3gYRkN1I56zNan6FNzGMubilntt2z +xXQwxP4Jn+RoAwb5U9mIlaLJ73FDbl6KAvFSJHlZl34R/o1nKOOAiFSv4V+RcTMd +x49P7cyAcW+eSsIgDzqabhx1OcrFEFRtBy342w5m5Qdq62TpFmeALgRYhAarA9UZ +YY/XOglN88K/3iR+A5LO7Hdv0Q/wShokghcSdAE3JOo= +-----END CERTIFICATE REQUEST----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/generate_certs.sh b/tests/integration/test_ssl_cert_authentication/certs/generate_certs.sh index d6126d361f5..a09b7b2874e 100755 --- a/tests/integration/test_ssl_cert_authentication/certs/generate_certs.sh +++ b/tests/integration/test_ssl_cert_authentication/certs/generate_certs.sh @@ -13,11 +13,13 @@ openssl x509 -req -days 3650 -in server-req.pem -CA ca-cert.pem -CAkey ca-key.pe openssl req -newkey rsa:4096 -nodes -batch -keyout client1-key.pem -out client1-req.pem -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=client1" openssl req -newkey rsa:4096 -nodes -batch -keyout client2-key.pem -out client2-req.pem -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=client2" openssl req -newkey rsa:4096 -nodes -batch -keyout client3-key.pem -out client3-req.pem -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=client3" +openssl req -newkey rsa:4096 -nodes -batch -keyout client4-key.pem -out client4-req.pem -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=client4" # 5. Use CA's private key to sign client's CSR and get back the signed certificate openssl x509 -req -days 3650 -in client1-req.pem -CA ca-cert.pem -CAkey ca-key.pem -CAcreateserial -out client1-cert.pem openssl x509 -req -days 3650 -in client2-req.pem -CA ca-cert.pem -CAkey ca-key.pem -CAcreateserial -out client2-cert.pem openssl x509 -req -days 3650 -in client3-req.pem -CA ca-cert.pem -CAkey ca-key.pem -CAcreateserial -out client3-cert.pem +openssl x509 -req -days 3650 -in client4-req.pem -CA ca-cert.pem -CAkey ca-key.pem -CAcreateserial -extfile client4-ext.cnf -out client4-cert.pem # 6. Generate one more self-signed certificate and private key for using as wrong certificate (because it's not signed by CA) openssl req -newkey rsa:4096 -x509 -days 3650 -nodes -batch -keyout wrong-key.pem -out wrong-cert.pem -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=client" diff --git a/tests/integration/test_ssl_cert_authentication/certs/server-cert.pem b/tests/integration/test_ssl_cert_authentication/certs/server-cert.pem index 53ee3185b2a..073c6485bd2 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/server-cert.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/server-cert.pem @@ -1,33 +1,31 @@ -----BEGIN CERTIFICATE----- -MIIFpTCCA42gAwIBAgIUf7oSjl262zqxycxCt9R08BG75GYwDQYJKoZIhvcNAQEL +MIIFZTCCA02gAwIBAgIUBfEMZ1Z/4weV13ryVA9qyNTPJGwwDQYJKoZIhvcNAQEL BQAwUjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUxITAfBgNVBAoM -GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDELMAkGA1UEAwwCY2EwHhcNMjIwODA4 -MTcwNTQyWhcNMzIwODA1MTcwNTQyWjBWMQswCQYDVQQGEwJSVTETMBEGA1UECAwK +GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDELMAkGA1UEAwwCY2EwHhcNMjQwNjI2 +MTAyNTAxWhcNMzQwNjI0MTAyNTAxWjBWMQswCQYDVQQGEwJSVTETMBEGA1UECAwK U29tZS1TdGF0ZTEhMB8GA1UECgwYSW50ZXJuZXQgV2lkZ2l0cyBQdHkgTHRkMQ8w -DQYDVQQDDAZzZXJ2ZXIwggIiMA0GCSqGSIb3DQEBAQUAA4ICDwAwggIKAoICAQC+ -6f15rjbEl5DanEWi3YsMIPNwQ08waUrqFz3aCKeGcRujKb7uaX/I5LWdh8x9I++E -E2ccFlnJTd2dZKxjY9jd2pRXVHmVE7XLpl8qtlLtLjlJ889G3tbgwdLZqaClxJxH -0KQHH5wUwweqlfbteXeYUAIfhDRmoOL2qhUPLYi7E5/vpMeEL+tMn6fswuk7JIR5 -64NLUVC1/VXtjDli0YiIhE025iWL7FukUAifJKa/kYEAPen71bZcpT3uxYnALClf -rANg8uJ+DTDPeNjZbLjNTPRfqk4W7pwG8w5F6RhrMPwgqvBNrnne7OtCD2lzlFU/ -sQiBM8HA5gN2CJuDjl+F70KTY3nRkAxpmHvSl6RPTEx4egdcb69A85LmDjyV+S01 -8tJQgi8TEmXM+TadQo3Xz/6+MHBIr7MknRRs0l2wCCiNvkE5MxsT+Pv182wxGArF -aIvCU4dZI6bI2WZbYclR41Ud5O6XhQKYu2AhS7zY1+nhAMAvqyD5OmIIjKWi3GHP -vdSCljBQzs1tuHgTaKIeTTzFetnRDzyoRDbLbo/OhKEAybvIEqq5iXUmfz4RG+Am -ZUdxBZ6muueAf44mVQBKx/OB1BIzObaKy5s2gP+c486WR15lqVcu6FkPf8tO15c2 -lfpDXE5IvRBeSgi4vVdL4ceJtrfMavd9pa3YmGk+zwIDAQABo28wbTArBgNVHREE -JDAigiBpbnRlZ3JhdGlvbi10ZXN0cy5jbGlja2hvdXNlLmNvbTAdBgNVHQ4EFgQU -eebevYkmCqTDsOG16U3lTCIvqNcwHwYDVR0jBBgwFoAUT/2qvvJBk3zEQJj4KzWm -hB5HShMwDQYJKoZIhvcNAQELBQADggIBALVfgo+K+SHzNrerNVJI8U/50fi3WPsS -rsorNxYHaapTJB7/Ngn8Nv05YI1Er4Npl9X/9LjA0uwxYmW6zPGGoSVqGTXJD50o -2c2MVXrx3ZjkNLNw8OoIROU/JwFZRANYS9ECZVOYY2eHcci4S1D7izZP/7+8V+YV -l49Do+ht1nkpeOadWXsQDPZO3bVbUvkvuYYeaHGzULKWS4sHbXmFizrL3V4XeuyR -SAwAo1sRkYcJP5JUk8JCDW+5XFtk/X638RfZJ/9hGxW2gbX9T/Mgqsmi4TMajeNn -VWkq7+WmU4v9TKZARA0240CIiwmW95KVMYe5rWEB3i9yo0c4kei9H113q0Pp/3Kd -sNCZf14Wm8BhM8uUQTyOUyXQvDUx1JzJyZtXj3zGR86uqEGPPMJj4tWeP/FIcF3v -hpH3s0md010BIjpEVoov6q1qPe32WQL9eGappsiEbJKFpJ4YYXwtSgOnUy6vt0kp -TTuYQWBCUfdhKUmQHvJbFzA6OlIs4RRmWlUcFOFOcS79FzgcWS481LSyMNPNcQNQ -PGmLtxxcoRADs9++BltF+Q8V4MbH5o+ZQt8314Vg9n1AWV6L2poLNY5CxXMd6tVu -wDYHIe/VHyCBqsdh9u7XKpv27xfu4TdxDS1nNzyMm69FUlGvRC5IR2k+IEIKmQ2n -nyRbOKxUBKl6 +DQYDVQQDDAZzZXJ2ZXIwggIiMA0GCSqGSIb3DQEBAQUAA4ICDwAwggIKAoICAQDd +l4bW8vca9/iFDASgmGxyCQvEK5roWwPzE/DXdS+mMJM/fffyqB7Cldm1KW2ILDYU +6JGnlYdxvWLOEa7n4cIBrQ2za7vfDU4LzGJTBN/9C2TYGFP13FTfZi2gZOuipT+M +pkFCfJG1zMWR13yKNxiS6Ixf76c6ADUyt7GmRcuKSctKzM7hUARfeko5iBN8aHpD +nxYa6xSGlHXHDzHHEqEpsk+iQ/eK6o3EeElg7v74sZSzqleIkEfFLiRIPNvuAwEE +IRIQg+k/2S673kdANlyfszXXbUGCYU9lOOPYgAEFgb5ljSXZJBTW3uFlyIONK5I4 +EHgELy9Hs0qnqttXCx8d5I/eI/+Zx9GPF3x8gkMi5wQePiBpqde20kRYDvP2eGeZ +eFTnKYpqCshVAJ4+1Kg96f4S9+GZ1wfzsgS1lA7+9qr3qTvI+XnKhC2h3bqyWS+C +BQikgADbURgZT4EzhXvq7fSCFHFTA6xZzXNEVO7DkRrYbvkq06bIl1Ov9vcbCN4I +zOuJJMxlk2Dv3C3mUox3HmcOdO+vvtWnZt3HbFY7ZPCV68ObSf77YD3O5on5RFQu +hk+AinrsDIL6NiiVzALXBL+e8flkqZDzRk1mGphVXGcRP6nn4VtrfN+Bmbm8pu3m +6aYuqSX6vQXb7EHW1tAbvlfbxIlP/Hp5GoV8zqI/tQIDAQABoy8wLTArBgNVHREE +JDAigiBpbnRlZ3JhdGlvbi10ZXN0cy5jbGlja2hvdXNlLmNvbTANBgkqhkiG9w0B +AQsFAAOCAgEANvZ7QDkHIKRq/g4GPkuiU7DN44/TW4bOFe7rDC5S4z5sh/i/Tvur +JYW7m97vLui5PJf6Vbd7xyl5MFIiz2KzoLi26rlvYcI/BT8mIG8jMg7pJjp/wJGa +QdCxdO99a4SIwg7x8pvWChFAOij5e6RhrIvsEB0LN5kKRSnQ0sW5khfsdFbn+3Iy +VwyvvE+nsCqE+KK358EMHicn8FVD3Ze+YzckX0am9DbshL5+eVQ9nOhUV2B8PcbG +SGzqJF07wOBwCdcn3eY+V98SQqrpGC9tCXmv8qErfkq7pkUGWq15d+miF/gaUz+Y +yDPwgi1pephBJ34IhLUUk0IPZJ23uVv/zfB+SpZ9/5pjsmnapR3Zf725jWrhjeT8 +44i5kNeVCvZPzQO9cTOsLXJbWb0vqRzKsvuSvffDQZql4bMMvhPjMibqCiRuSHO/ +yPlWiJjhkZz52DPJX6+LOeP2pFfUe2TR6IqcFPfUs/bV6aD2L/s5UZfZXWS5i5FR +I8uvcKOWL7NBbdY+NVE5aT7DqfhaRurjp61Aym18FgXLHpDHYV9IpkU34+A1MBUi +bzHZRWhxZMRxYezC7jE4zsZ5CQtSq1miDPcDaeK5vMd/Vdys5MIekqCfKUh+Cd5Q +gfC2QgNgodcWRF2plNgA+3E0dUULR3+1s83gGWGC8/UFW+9dtYV4nv8= -----END CERTIFICATE----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/server-key.pem b/tests/integration/test_ssl_cert_authentication/certs/server-key.pem index 9d4aa59a125..067adf4e1fc 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/server-key.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/server-key.pem @@ -1,52 +1,52 @@ -----BEGIN PRIVATE KEY----- -MIIJQQIBADANBgkqhkiG9w0BAQEFAASCCSswggknAgEAAoICAQC+6f15rjbEl5Da -nEWi3YsMIPNwQ08waUrqFz3aCKeGcRujKb7uaX/I5LWdh8x9I++EE2ccFlnJTd2d -ZKxjY9jd2pRXVHmVE7XLpl8qtlLtLjlJ889G3tbgwdLZqaClxJxH0KQHH5wUwweq -lfbteXeYUAIfhDRmoOL2qhUPLYi7E5/vpMeEL+tMn6fswuk7JIR564NLUVC1/VXt -jDli0YiIhE025iWL7FukUAifJKa/kYEAPen71bZcpT3uxYnALClfrANg8uJ+DTDP -eNjZbLjNTPRfqk4W7pwG8w5F6RhrMPwgqvBNrnne7OtCD2lzlFU/sQiBM8HA5gN2 -CJuDjl+F70KTY3nRkAxpmHvSl6RPTEx4egdcb69A85LmDjyV+S018tJQgi8TEmXM -+TadQo3Xz/6+MHBIr7MknRRs0l2wCCiNvkE5MxsT+Pv182wxGArFaIvCU4dZI6bI -2WZbYclR41Ud5O6XhQKYu2AhS7zY1+nhAMAvqyD5OmIIjKWi3GHPvdSCljBQzs1t -uHgTaKIeTTzFetnRDzyoRDbLbo/OhKEAybvIEqq5iXUmfz4RG+AmZUdxBZ6muueA -f44mVQBKx/OB1BIzObaKy5s2gP+c486WR15lqVcu6FkPf8tO15c2lfpDXE5IvRBe -Sgi4vVdL4ceJtrfMavd9pa3YmGk+zwIDAQABAoICAFAqklEHmiDYIi/iAg2/KLWp -mn+QDA8gj6AXrVXCX5JBKaFMlMupCksR2r9w+CmuLtPPj/MH4aD5rO78CLVdhoG2 -WKcJJlq2oZaH4JtfXxgQWmcW1XmFdkZ/rSnQJFCnbBZt4orJN7GyKaR0f3E9mb4g -DpwsWBKmSVfZmKk8bhdcSMMI9uyncI9G1W1CdUxr66MEhafZV+JrpCrxQGGh6cql -f4TnhGmqkNrA7rXg3pI/p6Mx3HBuz7o8evKqCUtkX+U4Jl0N8JSMtmvQa4J1OG6g -+1a2fT786BC0/E/X7vSH579R+EEuXCeDZdBiB15MsbVigdc7JAd7roVgXOkTnkE1 -miQeGhP2J4b+OH8fTLS2KfZ9rW/uUFI3kO3duMv6+K14fIzXDZMGduW9f+Tf0gqf -bj5A8Me93fddU5UHgLcJPwqbKXwJjvpnc4c0Ntl5op/2NieakvSJ9l0SnqzGcLx/ -Ufgiz4djaRX/xd1qepOHYkXT0egVec5kJvY5uKgdkWMH5ZlL5viv+7dZNBANAzpl -3K6j5N/ay7ED5cifeUReuPeNw92w5Rvq7OVGWR6lEsexFf+J19fefPo43PQQFtLM -W4T2G/Y0NHkkUsY71CMFN/Oom9wNLq2EGq0apljphf3pAVf5aGNSNwJkLBlmwJDu -NBN2AZQDd99zJ5+LfkotAoIBAQDK2koIG46VTDmhLAOZFNm9wMBALjFExTwo7ds4 -J2GmAphAgjkaxq4KFePc9uzVBx3nkP/4QSWpq/IhJP1usFseSeAR4SCRx2ARvhvO -T+QQ2TTWIfH+LfVM+Vkg38eXJrAF6+UT2EFkvFg0gO6bW6q/SSCxZvltLKn2GX9m -ql3SeFq14r3q05D/sAetdJquAyY4mDZNUEh3rK9ueVNODXSw61GvCABEAJvrNVeD -4iTOmL+51hQaNubuEyxvmmMVv6ougo1MwQmC2MdnwlW5DmUDM4oHjiEZ/1bumEMK -HMUgAYHguwRug28YU4D7MMkKWnhCIa9tEeUKlCDi5eiqtoAzAoIBAQDw7vO61oAd -Aw1lS5UBKEovZfM3aDZKhFsqkhbRBB+FoYtWJyEm6OTZL7oJOVPS/BZFzI/Iy7XG -D+4aeTrK0iQsah1oim4VHp+A6jBOm8VuMJ1vJE/awXXqAkxPnoEaDPJI0DNyVJwx -ah0wH2qAMVi4eXvBHaYemx+KaVxYiXp6veupwWS6ZIQYxKsXbudbs0u8tKEblCGm -KFbeTN8lI2Dm0YwLsrYL4HU88AYYzW1xJAaeN7IV2okOdu8bxPZn33MrQT/Ag58V -FI08yWqxtxQ+2Xf+rpT9KV8GzrCkZ23aSxT3MKjhx4zg3IGEwZYeFmglDgxj4Xeu -WBanAMmddzr1AoIBABqRzPCS63ISsdmyciYy3PJFtOizJDDlxKN3xCbuwZOE83w/ -kks3isQ3ackfbpXYgMo768sQfWZj5ysANVGyN70X4Al3e1Sc0LCCPhIf1LeAO7Nw -bsnkKyUR4+KybOoXXybnZvHaeXZAVS1LVfVzZEH5yhZybmkmWHyrikxgNDnFGdyI -/mcrnupenCLUCw6PywnOpg4qXdOPAsttiMPIb2hc6i4K1j4R9lowem4DyN5mk3Y7 -3BYy6rx0NnXHuwK2QAwnfWp4Pk6tJjEo+yFqCUEKFyI2M4+8Kh7GQsGoUwmGKQTO -eh4fiQWeql48XZjFkMrYrqmSlnV0QXp2sqrpjJECggEANomwjEOP3oFZX+ubABcR -q+cFBi7F01pglKNbHNc1F3e45biXzYn8e5lNIFdkvSapGAW1KnvKWtoySLXWdCDb -ZV5j580mHAvBiVn5s2GZcFb63DS5CsiuG1mH0qILkU0K7yaJ7sBuVtUxZDpITlpd -Pezp8Y33k4gDvL4a1EPSgMRK+zM2zOaB7GVgYT6OinhslXvB2E9Qyp7pAwsdm/eF -MNqQO/mpMckOYMvoZWbi9jB0ew4ads3wJmEPwRZ1vI+dL1ZmyvpAYXI0gmUJjM7n -e06Y1gLI4QGjbBQPcjejaz2Bsm7GW81i+2eOvfFgPNFPo8upTFa2U7XG8ui/urBJ -nQKCAQBq2XWZJFgN9GLRWy7Yw+n0s6W5vyD/c8O9VTuTn1DafOVb5KxH+egcxDcj -0T39nyVm6XhchL96yu947bWgCVOiR2Yu5QQ6xw7em46VZdbTtwjKlW/xRYGJKW8v -utcd9C3kR1r92oLbtuWcoBFcCW2MMhamU8H+Wg6Mj5v9+6FhWcomqHaiS1E9ZA4V -qhPCheE6XZGlo6+ar9gIXJcGP2ktkStAWT+0O8stc8xvF5nS/y35ZJpIxrMOsNQS -9+kxj8ouH6D49LYm2ZFtgV5Ink32zynAzFUm4FcYe0fwo6ADzS/PH5eSL1fX3iZP -Hg4+vB3JZpIQ4jvvlZ+GdcG5eW61 +MIIJQwIBADANBgkqhkiG9w0BAQEFAASCCS0wggkpAgEAAoICAQDdl4bW8vca9/iF +DASgmGxyCQvEK5roWwPzE/DXdS+mMJM/fffyqB7Cldm1KW2ILDYU6JGnlYdxvWLO +Ea7n4cIBrQ2za7vfDU4LzGJTBN/9C2TYGFP13FTfZi2gZOuipT+MpkFCfJG1zMWR +13yKNxiS6Ixf76c6ADUyt7GmRcuKSctKzM7hUARfeko5iBN8aHpDnxYa6xSGlHXH +DzHHEqEpsk+iQ/eK6o3EeElg7v74sZSzqleIkEfFLiRIPNvuAwEEIRIQg+k/2S67 +3kdANlyfszXXbUGCYU9lOOPYgAEFgb5ljSXZJBTW3uFlyIONK5I4EHgELy9Hs0qn +qttXCx8d5I/eI/+Zx9GPF3x8gkMi5wQePiBpqde20kRYDvP2eGeZeFTnKYpqCshV +AJ4+1Kg96f4S9+GZ1wfzsgS1lA7+9qr3qTvI+XnKhC2h3bqyWS+CBQikgADbURgZ +T4EzhXvq7fSCFHFTA6xZzXNEVO7DkRrYbvkq06bIl1Ov9vcbCN4IzOuJJMxlk2Dv +3C3mUox3HmcOdO+vvtWnZt3HbFY7ZPCV68ObSf77YD3O5on5RFQuhk+AinrsDIL6 +NiiVzALXBL+e8flkqZDzRk1mGphVXGcRP6nn4VtrfN+Bmbm8pu3m6aYuqSX6vQXb +7EHW1tAbvlfbxIlP/Hp5GoV8zqI/tQIDAQABAoICAQDaRKlTDRwN+ndXRlFAhyM6 +6GIopvL9MLmhM+EluY5n2q0P+1rCMIusC8LYSahUW4gh7DucoRM7G9s5M/3e9mcN +E5LNSq9RtF9OC9JGCCVBsXlxyfTZ1l/bdWA3/3CDUtZYCmN5xA4az0tErsdDtaWE +/39V+E/2N8Iu5PYd293zp2CRm0+kbBcCnQiDxt+6yYa1GPzDIw+iyJWCsBrOBjGt +SrBaGyy4LvXZsspEquWHvhPFLWLvZ37qYNroNNpFhbv4f0K19dlJRPpdn0L7oxB1 +VicQvdOrQ4LbJ8B2vw9Ch1wt12ySiJHmXMAUa//4jBSJGN++72NY8uf0Y72N7ayF +Bck5QE0we4i7hhuN0WL+IftYD/O5NgOnprjMWifLOhQ8OECZfhOKgbRU+i3aJl8D ++raQBW7/GM0uL7xIoMcEZSwMs/sQR4loNCJ0UsIeWTdWNrhXrEuzDQGXoWcB8K/r +VVayDO5Qqx8R77HB82/pRdqEWNaNQd7DhPgnWiixISpIGm6zMvT3S0hzEkxu7FNb +uciq9i82BrBkKhg1kiF70FqG13VeMFqTJUuqtoRs1QEQgumvWB47n6FiVwHdDbbH +sUeKZYwbrY22Cn4rrfXH+0KKM9TDR0CiCv+CkSGmG57l5tW7aSUWun46qP8vh7sc +ztzb4LzyUt6XEBIWIqBIQQKCAQEA9+f4TyGo88qzTKaQoko5OAYhvAGr1UGAS6qh +sJpd7rHd9g6DnXyOInpQNglToiJ94mNdIK8f/oOjUXZh2E4CWuxeK291BNiaqCxe +s3LS3XjkdHQIHvqJUw/r4YJ+zfoGznthNbDwDkBob9x3h9rknTbGdLcgaTGi/0PZ +cFnyWDPNPskbcZ3Dxr41oDHiVsOx8n4d4HtspXzh+EPiQiJz5HVfwGNEyzhmFWIa +EzQaxnHL+WF1Pqe1wrzOwZia0Jss8lxbHcTnJupaV5WDvbxY0E4ynofShicv1U76 +B41xDKP/8hFWck9LiMDXk9vrbQIHvGAcsYr5N/jzIrDiEXxvGQKCAQEA5NOegb6m +Ak0mXg+XKhZnSdR1KaWT4/hbVWqYjwxAeAJfMaxjD4MXA8qeEAJo3l3ETkdFCHp/ +lr/BZzNYXnmHU6uvDn2Xq8gPO04ruSV0WWthikXb5gszXdkUH+6fryHn6L0kYJQH +NARQzOgdEcmTP7sy/5GDKzydWbT5qRNOnESUWgnJi9ePjGB9zWxn4jx9AfOYtozh +UmEgofSDGbFlamQic8HGnSJFgOxIZ0AfurPIRSR51gvXx2D5TcsPjLlDrY07IcF3 +DjqfJl0gC1XN5BXdpPvjvNrns+ZK/SRoGlgb8Q4tZLQevox9W110amvMeZj4yMTK +9mgGOSYCzZ6U/QKCAQEA1mBZ4Qwpj1DNRk6PqlfnLSRYTb1gO9UdvdE7a33CFuTn +HZ2lgS2xt+zvqhrcoMuU8o2cfeQTFcP+Gjb2G9gxvzDBqmwC1IL/Esjzx9hWssCV +RoMEds2OrS6Ke4OeZj59XldhU83DeX+HEJylHO1UXwN8EHg/5dfPrVCeGsMdh9qb +9VxxiAm2wAnCU9pvcTpfimQ3L+VrqZvZyRfi8+/ZKkm52KO/XMFTvdAM3mhjcxH7 +Ipd9jQX4bwNZBB8UWaqm7pqhDJg2j/d+0lhwCUZzwwasTV0E14/RlHNsUdWlWhoD +/e+yQr2BgyvIAIvgBW8JA4RVq86S/y0gC/LMO/TQGQKCAQBB2rlaY7DJJsTs+xWp +EiuFrvRNGQ734+j9KyFewcrn/t7An/keZL7B45UbzGW74UZ2tMIkT4TasLMLbVZ4 +UgdlSBqoU/LLiFcB3Vxt+16BwYqfzb0cdorA7pGBIx6nu11PuOd4OAHesYNDhWWg +Ud/jzo89x/X1AovSXmgfhaPxCzeatghgC5iPcNGjxhgbnwbnAeEoYGEUYUmP8pus +UEZ8mPblU5ZCcLOKB/ZKaMT46Xawl2/M7zmZcsos3kzKViMpFmU3MMN/v9U/qDtp +p7cKdlSEf82p82INfzCDq++d7U+VT1w3CDN06V/GZJ31ZrLBKAopVaGHyqZH2i2i +WYpNAoIBACmfr9BoJh1/mbLdd/WpOyORKTbnwstlMgoUcARRJn86iPZuyI4QoSSb +TePZqrWRVmO/K5M65hFjhUpqTWKJGJy5LYIZ4yuIbonJAPNUhjA0bkar9cULBFzy +rb0xmW6sRlBnqhv4aDlOkhHkkR9lB9rTIUW+ankuvVBiGWo4eE8DvZYo30frltku +2K/kqd3NppTl7dN4EnGTo8ROZvr3EMwSu6nE+wUr4G7YuCLdPxwb8gAB8dbmaUsn +AXocUh96kYqTwRxo8FO9SqgQYMf81/ovPUfv+7mwO40oygzy/YkGmB1shFIbQuzU +lJvRfdXyyC9DbllQkxWfdvaanLS3r1w= -----END PRIVATE KEY----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/server-req.pem b/tests/integration/test_ssl_cert_authentication/certs/server-req.pem index 714ef19ecf9..bd8e2e1fb7f 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/server-req.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/server-req.pem @@ -1,27 +1,27 @@ -----BEGIN CERTIFICATE REQUEST----- MIIEmzCCAoMCAQAwVjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUx ITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEPMA0GA1UEAwwGc2Vy -dmVyMIICIjANBgkqhkiG9w0BAQEFAAOCAg8AMIICCgKCAgEAvun9ea42xJeQ2pxF -ot2LDCDzcENPMGlK6hc92ginhnEboym+7ml/yOS1nYfMfSPvhBNnHBZZyU3dnWSs -Y2PY3dqUV1R5lRO1y6ZfKrZS7S45SfPPRt7W4MHS2amgpcScR9CkBx+cFMMHqpX2 -7Xl3mFACH4Q0ZqDi9qoVDy2IuxOf76THhC/rTJ+n7MLpOySEeeuDS1FQtf1V7Yw5 -YtGIiIRNNuYli+xbpFAInySmv5GBAD3p+9W2XKU97sWJwCwpX6wDYPLifg0wz3jY -2Wy4zUz0X6pOFu6cBvMORekYazD8IKrwTa553uzrQg9pc5RVP7EIgTPBwOYDdgib -g45fhe9Ck2N50ZAMaZh70pekT0xMeHoHXG+vQPOS5g48lfktNfLSUIIvExJlzPk2 -nUKN18/+vjBwSK+zJJ0UbNJdsAgojb5BOTMbE/j79fNsMRgKxWiLwlOHWSOmyNlm -W2HJUeNVHeTul4UCmLtgIUu82Nfp4QDAL6sg+TpiCIylotxhz73UgpYwUM7Nbbh4 -E2iiHk08xXrZ0Q88qEQ2y26PzoShAMm7yBKquYl1Jn8+ERvgJmVHcQWeprrngH+O -JlUASsfzgdQSMzm2isubNoD/nOPOlkdeZalXLuhZD3/LTteXNpX6Q1xOSL0QXkoI -uL1XS+HHiba3zGr3faWt2JhpPs8CAwEAAaAAMA0GCSqGSIb3DQEBCwUAA4ICAQAx -Sq/dJHjSVa1g4clFjtdKciFsSnCm/vgzGInyxGL4zTyTf4QXQ0PhfHkiFFCMkFSP -snxavti5HjSCJlUkhB/x4YpqFPQ+/9Uly8RCKKdlMTSiJ30IL/D4dWtmwA83UQAY -ZI6b6dvjdhBNMDb5M9Qzv4+PmF/KMB3KlFTQtDZoAqAnWrtahsVJzsaawK4PPc/e -4IINu2O/aAFnJt+ewwA1NDrkaSlD7Wgu+SAlQRPO+vAKS6Qbs69R/vDdVECJOTmB -FJ9uQlXuhwsR6u5Pl0Df3Jh4K+EXw0nY4LEko3915HnKAQt0F4BTrHjW3Sk2WnMN -AWtp+4D5epRvD5VpL+mwce0PLH6rUb4Ipe9zmApGQr2GAO3XjpfvusvUJPFcWe2b -EfnBxq/Asw1ALqLrT/LKpZHRvNN2YpBLl8ZrzOsNwqVPMDTPUYWf17wLS+FiuCHD -BTdMIoqZ0dmp1ZmENB8h5zM8W+XMlVQlg+LeTVqeEA5Jgr7zuMObQOar0K+MV00K -Jqi2ba/v/zFtN31rH+wULfV8BPdtrVTbJMTrCJKLpAwKjsO7wFoFn0Qk7WNEmPmD -+TA65ilk0xfok/04pkh1gd/Kqzh1LIOpG0kmh410U3AJ2jsF3Sop+apH+r+Blota -SsCHnBqnABNRs6gs5FA1pbD4t81pQl5xoXtCCuZbPA== +dmVyMIICIjANBgkqhkiG9w0BAQEFAAOCAg8AMIICCgKCAgEA3ZeG1vL3Gvf4hQwE +oJhscgkLxCua6FsD8xPw13UvpjCTP3338qgewpXZtSltiCw2FOiRp5WHcb1izhGu +5+HCAa0Ns2u73w1OC8xiUwTf/Qtk2BhT9dxU32YtoGTroqU/jKZBQnyRtczFkdd8 +ijcYkuiMX++nOgA1MrexpkXLiknLSszO4VAEX3pKOYgTfGh6Q58WGusUhpR1xw8x +xxKhKbJPokP3iuqNxHhJYO7++LGUs6pXiJBHxS4kSDzb7gMBBCESEIPpP9kuu95H +QDZcn7M1121BgmFPZTjj2IABBYG+ZY0l2SQU1t7hZciDjSuSOBB4BC8vR7NKp6rb +VwsfHeSP3iP/mcfRjxd8fIJDIucEHj4gaanXttJEWA7z9nhnmXhU5ymKagrIVQCe +PtSoPen+EvfhmdcH87IEtZQO/vaq96k7yPl5yoQtod26slkvggUIpIAA21EYGU+B +M4V76u30ghRxUwOsWc1zRFTuw5Ea2G75KtOmyJdTr/b3GwjeCMzriSTMZZNg79wt +5lKMdx5nDnTvr77Vp2bdx2xWO2TwlevDm0n++2A9zuaJ+URULoZPgIp67AyC+jYo +lcwC1wS/nvH5ZKmQ80ZNZhqYVVxnET+p5+Fba3zfgZm5vKbt5ummLqkl+r0F2+xB +1tbQG75X28SJT/x6eRqFfM6iP7UCAwEAAaAAMA0GCSqGSIb3DQEBCwUAA4ICAQBc +TPShRdB7ZIL4xQNAxWLGoEbshbY/UpJQZdjojxn27roVwEhwP6B1/KgiVKV2X6bE +a36LUnaWYllIMAh4oOHJkIm2gZ3xitdEK1anCf5NJga7TnkwGfD4jTZA91fWCynt +a/64s0KQggKsUVY12TTJVQvOH/l9RMrXOq+jgIh4OURwCBtHTCS6oOp3eF04pEu7 +w54dMAsxp/N9AQaLh14IZUZAQ2v5kdipL99EEQH/G6llU08XrJ4LfGgwDkUjJSsA +TzrX2uk9MLHJVwjpZ99ktjNBs8Gyr4fGOmstT5TXEOO6bVhqZDC6kEL7vrmFSnDZ +g/9lrd4wLUT/STt+E4Qukedi0n/419IpkIAE5C1HOXRnQaOUdcnrLixUB21mBHt/ +n7gkwdY7Cu77dYvBIShzeCnxiJED0+XrBPD5yPokxEjE3MmiIK6meHHCuIwqLjX8 +I78ysv4COeH0svwSjvJInveS9QRCAWBpdvskxPJR8dpoysAff+jiyp3ZvkwcIiUO +Vbsusorpp8pFJXZxvPDBXCy5TOlFnIG/9itjPj98pFRIl5bxzNwDf4wrkwHEpR3i +jpM6y+/RWZn69BpTeAG0vZHhGk3SuXdU56cRzatys4X3biCImgDqeJMUcAoxlrIZ +vgbJVTorwqQmPz5kt28b8ddnUVobbZEPlRITcqjwFg== -----END CERTIFICATE REQUEST----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/wrong-cert.pem b/tests/integration/test_ssl_cert_authentication/certs/wrong-cert.pem index 03ebf989764..b56a10f8a92 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/wrong-cert.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/wrong-cert.pem @@ -1,32 +1,32 @@ -----BEGIN CERTIFICATE----- -MIIFjTCCA3WgAwIBAgIUUghXstot43OmqoS1M2rmdSRRX54wDQYJKoZIhvcNAQEL +MIIFjTCCA3WgAwIBAgIUMcX2R8I2H8vNtASHi0EoufIgWEUwDQYJKoZIhvcNAQEL BQAwVjELMAkGA1UEBhMCUlUxEzARBgNVBAgMClNvbWUtU3RhdGUxITAfBgNVBAoM -GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEPMA0GA1UEAwwGY2xpZW50MB4XDTIy -MDgwODE3MDU1MloXDTMyMDgwNTE3MDU1MlowVjELMAkGA1UEBhMCUlUxEzARBgNV +GEludGVybmV0IFdpZGdpdHMgUHR5IEx0ZDEPMA0GA1UEAwwGY2xpZW50MB4XDTI0 +MDYyNjEwMjUwNVoXDTM0MDYyNDEwMjUwNVowVjELMAkGA1UEBhMCUlUxEzARBgNV BAgMClNvbWUtU3RhdGUxITAfBgNVBAoMGEludGVybmV0IFdpZGdpdHMgUHR5IEx0 ZDEPMA0GA1UEAwwGY2xpZW50MIICIjANBgkqhkiG9w0BAQEFAAOCAg8AMIICCgKC -AgEAjQjHTorL16xTiJFAaxeC0GDP0uCIbT1olT8NmWOePURqg7HpVzBtvffxiug8 -l8fF5kr3mfs9A2XuY9Of/A8uDTm+vdSEjdTwkAox4355oCHDQo3F0GsfJwy5f0vP -t1vKhMmvBHM7cN3aaPBnKJwM3Qd1XWxyEeCv2SDvaxhwMO9Mveal7WUGBKms35Rz -PYEVejM9ccy7nZ8+/wLAx2ixgs5I5uo/RkJkCMVGlKk6BPRLFpiG5tdvrUmFYNdp -yOV2iVEJ1McPKkViUfbVUII4UlvVrsgRNqsWn4ukD5jd7a7AzvfVQq6Mhe7SqDc2 -8c2aVLhoqxYpbKmcYYJGQAqlXgOObs6DEcyitXnK34RkltMgjrl1GyqqqS9hGkp4 -XBne+rw+mbH9jfBdQpi4Xp79l0NVQYahS2iX5HFYRBa4i8SCemMGpVpHrK4L+X8u -qINiLlVXIH7FBTBgz5EjvMsgihdBbEKlFLqUJsPJhRPyBmIewZMGZnsO8PR8av+P -jFp5iBFE1RtIcj4mg1QsjnYxA1QjUtPnqPeHqph7qxtIvjd/j+oXpcyal0xkGsh8 -G+sdYZXCktuocMDmm5ejJs5156znU2yHwN0/hdGJYGdKYWHjSVPT6sa9Q/blJ2IN -/CHSf7ao6FPq4XuMynN5N7K5RbQ22oYuCmqcGpQxtwWRcA0CAwEAAaNTMFEwHQYD -VR0OBBYEFFU4ee1zFXfU+/UiBHqDt7gbKn3aMB8GA1UdIwQYMBaAFFU4ee1zFXfU -+/UiBHqDt7gbKn3aMA8GA1UdEwEB/wQFMAMBAf8wDQYJKoZIhvcNAQELBQADggIB -AC3qz446kylKOhNkssTXnjmFrZJhZLD06ijE15rhIgHBP3FXk1RZYcWj0bvERZ2n -zz0IO+Hy68M1jBV0R5YE8SyNVwEec5QiTYhoFoGsBFvlf6zz8oBhSQKtrvpvefdO -8vu+BcvX5C/JcOPrG9y6feTEWa29O7007LM+GRJOe4i2x6xkHGyLWvCSAm9fvHRc -gs6BcRIJLULwqlRaIHHNt8R4pwVaanJ2qQ2Fctgbx9OYkQzQK6hOjo0UUMvykEhW -XW0DT3tYGb9keEG6lPHYcyOcbkans+zJmHxSqLiVPz7tm3pwDtd2OFCi869y60WS -zzQNBhXCYkh782dF+pnOxV6hmdPaLqx3tD/WrT72z1qNK4FEgVIzSy5ULCp3WcgZ -ARvMgCMV7CK+rDID2QHuwNmJIArTXX8JRrV/zkUgsdapGpF5zkQ8rNGRJcDKzBww -CiVPOOthSbf+F9jLQ5nzmlFa6rJ7RxHuj1PVwKhxc8u9bPHAgZSt3uGH4QaFh5bw -3aoGucu2yhnkv4aNK6w2nj04K9gYdLwRbJgTNqR1FRCEXzfaJF1szq0dhXQzlhf3 -kbP/Wh+RhDVqt2Eu90LbmbUl+DbIHiYdjc67XFNbppXNn8NpncZKSDH/xn4KvC7z -ORsznqTC1pyQp4SuvaKYFCOWBzMZ60G5T8CTehP1KmKR +AgEApABgxjCrAiDY529xEjboRE/PVwgVStD0dPE2vSOOqxfVYYOqP7VF7pG2PeA7 +Ek9Qqv2cqaDWbDWn5vT3AiN+aslfHtXJgGjPBrpOQ3Me6RSu2aUHPIguabUjz+kJ +VSkoj0pkbAqqIxKse1hZtUNHVwOmg/PthkpGPNxofNX1kCfBWoNJyuKFOoGeNgmY +6joY10zTaiHpq8hhA3b7WY35QdNGD7SwkYBvTjGGzBr/hu26fhX/DnceZ9kf9n6q +899gB2kZH5T1eTlShh12TI1sHa+BGz1YwR0HqM88zXDyAf7bWl7Hy5f8e9keZdx7 +Fx/ws93Bb3GusA5zwUm1iUe1OIwTLFlL+Kdkr0qzofDcQ0ZnNwrME4oAhCJFwSnJ +OWnrCmUou2XXj86Xl481uBana30zzJ9TniHM/hA54ttHsva/yB8tyoXcI4FASwk3 +GdihsOBbRS6KvmeKpEXQpsvBQ9GejSL/UUWuKg+O0ysHE9+QX/+OznFp66h/x7PP +Q7g6ipwAjgwuG5jm/Czz+dw4j3Qp5N5f7Dn3QhDzmXkKgzRzirKh9XVQqUFRwlLn +8VuzAhD5SjRN0JDE2jlt0Hin90zx/nkOV2b5hTYu9NVgmrfCye6uB/qsK7PQBh69 +9i4+8tBGXrcS2Nenm+Hm12fFhNum96A0ahj134H2ks4JcKcCAwEAAaNTMFEwHQYD +VR0OBBYEFIZYdI/00qzj+5JqEzEJfpj93AphMB8GA1UdIwQYMBaAFIZYdI/00qzj ++5JqEzEJfpj93AphMA8GA1UdEwEB/wQFMAMBAf8wDQYJKoZIhvcNAQELBQADggIB +AE+9y6Ea3pjIegOeffXzI725vXO0OSuv76vRoswUVQDKL7j7QcvqdcKCK/2hmeMW +MBNwKZneG5iumH2Yp5RQ14arihS9+SYEpgftvfNOZrzZ37RttJhSjBQ7GUorH7pD +uQ4s1unqKXv981JCtlKOQh5ZKthHAMsP/sfYuWg2ksVn1hvFTJZirewVUOgR8zPB +Djl6PdjZLDu1FlglQQ5YUNgsJbAqkBPcA0hEwGU0j5QEncvdspn4LiH2/mHhnzM9 +3QEvsXUxgJo989Am6+E77XNX0wcALj2zUmPPDWYgHgLKO/ZcSAOQ9JaauVhUre2D +7jPwnN47yLak5obVcyCpaDPKYk6sEUZkiWRaONvugoIbjYivmB/BJc0njfVA0kzT +FDwpUTtSddZgHzdTXe0p5C7IGmYkp/vgKlSpSYY6+aCiVApJSdJjL6FZKoOXqDnr +OgoQGSOJif4mDeipKOdrb2JtYwJkRl0c1S+tgOi8PU+ROvZxQGWI9/i20H58M7j0 +r/WhbudhcAqWglk5WOpCodbJhXffCrbUm5NjoFr7AKswxLJVz39WIe/duHPEGV7v +jLd/zj7eJRv5ycDyt91rbGxQ9NKzEx+by/5WIZTi+z+2PG75tdpQUwgEIh1c/XOt +6uXtS0sNnnjHVmXPBC+Myz+1NolYWjZMcBQ2xGIORvm8 -----END CERTIFICATE----- diff --git a/tests/integration/test_ssl_cert_authentication/certs/wrong-key.pem b/tests/integration/test_ssl_cert_authentication/certs/wrong-key.pem index 834f82fe80d..3924eac91c2 100644 --- a/tests/integration/test_ssl_cert_authentication/certs/wrong-key.pem +++ b/tests/integration/test_ssl_cert_authentication/certs/wrong-key.pem @@ -1,52 +1,52 @@ -----BEGIN PRIVATE KEY----- -MIIJQgIBADANBgkqhkiG9w0BAQEFAASCCSwwggkoAgEAAoICAQCNCMdOisvXrFOI -kUBrF4LQYM/S4IhtPWiVPw2ZY549RGqDselXMG299/GK6DyXx8XmSveZ+z0DZe5j -05/8Dy4NOb691ISN1PCQCjHjfnmgIcNCjcXQax8nDLl/S8+3W8qEya8Ecztw3dpo -8GconAzdB3VdbHIR4K/ZIO9rGHAw70y95qXtZQYEqazflHM9gRV6Mz1xzLudnz7/ -AsDHaLGCzkjm6j9GQmQIxUaUqToE9EsWmIbm12+tSYVg12nI5XaJUQnUxw8qRWJR -9tVQgjhSW9WuyBE2qxafi6QPmN3trsDO99VCroyF7tKoNzbxzZpUuGirFilsqZxh -gkZACqVeA45uzoMRzKK1ecrfhGSW0yCOuXUbKqqpL2EaSnhcGd76vD6Zsf2N8F1C -mLhenv2XQ1VBhqFLaJfkcVhEFriLxIJ6YwalWkesrgv5fy6og2IuVVcgfsUFMGDP -kSO8yyCKF0FsQqUUupQmw8mFE/IGYh7BkwZmew7w9Hxq/4+MWnmIEUTVG0hyPiaD -VCyOdjEDVCNS0+eo94eqmHurG0i+N3+P6helzJqXTGQayHwb6x1hlcKS26hwwOab -l6MmznXnrOdTbIfA3T+F0YlgZ0phYeNJU9Pqxr1D9uUnYg38IdJ/tqjoU+rhe4zK -c3k3srlFtDbahi4KapwalDG3BZFwDQIDAQABAoICABhnSHVittreyqV63LarHsuH -tO48Q98DHTGV83hzMFJRoEyhYErh6ltQp87tWG669pdeqXFPc8M+w/0qBSjMR5+q -OkMGntkb23RRD0jZ/ZmjvfGqLmxd3MoY/Y/1Qj/r9iL78vjGJ6bj/ILj9jWkquUk -AT7lIOfPBR6BAGx7kg4lmhXR7ywgquXrpioZuccOqvS2IQ+r6Vrgzzm49DqM9wo3 -P44VQlWZ701FNW1LyupIiyWhxtXyuTQ24IldPMSyT8x65USYHRunXdfo6ghAQBag -JV1U6QWYnRWGSCEaxmv2LQ6i+5fKNC6IoJLS2wMEHIxETHguIYTX5GNK83TpYquc -F2w0NcnO/xMPPFiNFwtX6N9/mexE7dDE+gDvUwWA0DodwHQFh9+IBphznr0OfrJ7 -G6I98tIJj5ecSsTnHyW6BzOxyoQ3KGXfwV8RnDoNtZXiK4boizGlrVrj3G5AzqAf -z4dnIox/zP3kPyhw6670a1PuesQ3aTvR63UqLFOwtm1moZCh5lJmxog32s4mikdn -cQiFFz83zd4hAMRsoJ7W5ABeZyM0V6DXd+lGO00bf3ex3cNFskLeSHLoArtiBt0n -xtWfXU7/h4URFCyD3rJPUruSPT8DPcvu+PpxB4MEnZhvpC+xOXp5LM+jHcSsIgXQ -Bp1wMpZtTN+tkJrtDGbRAoIBAQC/l9Ao+6I/NI1mO6AxMuwyFf4IVpSYH2j/wvrX -v7No+i8egiS5joi3boNFrwCFuNDYm4tu9DQmqOHOYFiJXCemRkeyglgn0tSnnfGH -jFQOlY4kX26lhI8A8EqCUiO3XYnEUsQOd2aJFDos1rr2UdIPFhmLIF7+6QNsFtsV -9r6JcNIl+tMlhAZ1LiPYaKsy3GWPptK0Ul3IykBocIf42CNBrbbhMjWeOHNVUzmQ -6EY5CEeGVEGOn8bkSq4mqtzDYVO9rFaX2ySsPtAwoelsxnxLRR+YFrVdIBS4zbJe -zUHMD7g0lDl+gLCTcwsi5hFqUIE0xldRjzkqZgyE5Y6sdaG3AoIBAQC8cfZtIxEr -4vVGl6ZqOsbqoXg1vNL99CQKK8KXBdi79ig4F0ItWgv/RfdZvLAQJ2fVDQcOQs7r -NYINYNJQ2+ASvwDpcjI//M/lCTFY/kNBbUdAYIIrMlNPVoAnfGXCSpxOmw6iTNeW -gvTHamvXR4hkI6XpxI60dmcBUxJExBsW9/Ua5oONZ37nYVL32/PZCba94nMpnJna -6sq+YYWewla0YOZBKgegL/9S4kSt23UaGEfP6tC55puVBEmJZ3ZF57eiNHRWjK5S -oLUSvoXghtNYD8kAE95DIMUfdZudnxcR7o0OZIXlXxdKuWELGj4rXLa52pTuXdUi -4SaoFBIBtqxbAoIBACHtKxImxOs2ThedVVeA8exIkqw9UKDDZx9rEmoj6wwayT84 -wnINOE5rBlnPjGMhf/HrUTx7h8SUwS9/Rg5EwEykQ0vr4tpibf+ka0GaasQOXZY5 -Q+mRWJVM95SuIN72dE/MbCU3pXN+nnKUL5UwlebP/gPx3oB7GH0Kaw8vmzvRO5U4 -Hi2VfGNBoV1P44jlGCLg4ozFzQ+iKNeVna5F77BQp4KnFnGHA31HeHkdeFnfCnNR -9i3gQyLnyoDUXrVGRcLfgpHvX0YU9V63PRNi7l2Rxu9yjfkIlkLHQd+hrxYIbpD4 -8llhRmNFyuIKJWlTCh4jRREPMIs+eNcLo0EMr28CggEAdbmER1325kHdFQK8VfTu -O7owDPIvpbMVqLz5zbs7UQFQ0unEz+xHPmyJQChenVjgxvpihAg8mDhPTOhFu3oB -G/TzLuPGfK8FeYIqr/v4LDGGp7Mi3rWL0i6p1QEj2SMfTwQTOlVjAYZCQSo8wLrR -fC3BAq9mr1qgsvUgnpEck1bIasvDN12GrCni2TExv08QOMP1SfOpMlgn3JO0FCA4 -Zmt6rQ8CwJH7od31Jd/J4kl043FAgEKOw6NwBUT7YGSIFe9LFp7AcKIxPMqFfGT+ -Ny91VHUVDz6zpxmj4+51TSCIQouOEmM7e9UfyP3WBOTgym7BR8Ca3DVoHuya6zMh -rQKCAQEAoy35YYbyHnb5CmVfCIBNk2ZrDERuT8NE6J7u2Y+1/yCmPJyLvA/JzYNm -LifFf9x8+WMKMVaANq5ZitBMr4gMhFQe2KflfSC9rEzfhAoWVqNGjCs79+xNEv3Q -eFv3ss/zU9AKbjepS6+wP6CuILqXVQ1lT/xMtJwc3+YQMyrIYHwUJ13IeGyCJRfY -1/81JcB5pm9sjrgvRccjNV739A8ACm78mlj5DRxasKfTXkIlx/oJUe/EXRGfuZwY -D2FxqP9LLa/gY9DzlUPYpWz3Y/KlbNO+nmakiAzHpPjwwca6FZug9xFW9Ua3SAwY -aH0VTefFJqNq7ADzlJxogNFALN+F0A== +MIIJQgIBADANBgkqhkiG9w0BAQEFAASCCSwwggkoAgEAAoICAQCkAGDGMKsCINjn +b3ESNuhET89XCBVK0PR08Ta9I46rF9Vhg6o/tUXukbY94DsST1Cq/ZypoNZsNafm +9PcCI35qyV8e1cmAaM8Guk5Dcx7pFK7ZpQc8iC5ptSPP6QlVKSiPSmRsCqojEqx7 +WFm1Q0dXA6aD8+2GSkY83Gh81fWQJ8Fag0nK4oU6gZ42CZjqOhjXTNNqIemryGED +dvtZjflB00YPtLCRgG9OMYbMGv+G7bp+Ff8Odx5n2R/2fqrz32AHaRkflPV5OVKG +HXZMjWwdr4EbPVjBHQeozzzNcPIB/ttaXsfLl/x72R5l3HsXH/Cz3cFvca6wDnPB +SbWJR7U4jBMsWUv4p2SvSrOh8NxDRmc3CswTigCEIkXBKck5aesKZSi7ZdePzpeX +jzW4FqdrfTPMn1OeIcz+EDni20ey9r/IHy3KhdwjgUBLCTcZ2KGw4FtFLoq+Z4qk +RdCmy8FD0Z6NIv9RRa4qD47TKwcT35Bf/47OcWnrqH/Hs89DuDqKnACODC4bmOb8 +LPP53DiPdCnk3l/sOfdCEPOZeQqDNHOKsqH1dVCpQVHCUufxW7MCEPlKNE3QkMTa +OW3QeKf3TPH+eQ5XZvmFNi701WCat8LJ7q4H+qwrs9AGHr32Lj7y0EZetxLY16eb +4ebXZ8WE26b3oDRqGPXfgfaSzglwpwIDAQABAoICACiTq1399qGtLN1+NjSyfx8/ +u+Ylqtb7AjDY6Zk8bfUpDXN2Fy5yFF5lkPiYPSVXmHbmDtftYoAdenBrVZ4i2All +z3IapSNvSyG4ANsxZYl3w5c3/KVecFVZKwYq+1MlvtJNLrGIpfXNjf1qq69seP8v +eQiW1sLuJ5ixU+znJz3GiFFzwFNBXoNORK3MDBiPzUufx4Mv5tfI2S/5RVEwDmbZ +9jC2nSUy6Nco69geKfDhas39dUDH+i7pir37MyLptqG+wCePPHkE1MU4Duf76a8i +mEf8ErSdESMUO0/9TPNvcihW4Qofjam624mKVq4vCegGyvBe6UDIIp3FNfREWLLC +ilit37ZVOHbq79qV+DobLADYXZGXrptzr6VqVZvvQUEwOjftD2B8GJzxbNxRl77F +BbeOhYA/IDr8YR7Qos0HjqDDekRavHKAa5kcf8yFVJaQycirHJZpjr3oNFktqg6H +9eb537SdPI5nHtgTSQSosxst+iBsjMCJ7rU7aowy9gKG75s9eME06hiQsukNcOI3 +hmBqQBeX+yLWh7Z6A2Y7MepHuYHWKereBGISR58bvTmNyI4mLWYwJZzjON2tot3a +MJM86gw83TuX1Qmp3+NjduQtdtMjDSXLN2yBbK4CufQYaTxK1xdHUoK/uvG9kIq3 +tP+/fiTHZHyuTSSgwOypAoIBAQDT2Vj2uahOypboLv33XtFr2kuDe1sogpLaAVu4 +Dv4vO835gto4fx79rK3u2fBxiVx0yv7kwitvpcwaoMdkUEoSBtOugYB9UezW+SX5 +91bpY0mBH0ToSmosPFVc6PEki6LRV+VGZ1gFXU7uZ4Wia9opENfT7d8cjBQ4NZ/M +sCyqHR2KYB82DHx6Lrtrs0eWn33N8BVgsy4xSbOi2YrgQCnJvfPWVYtcXjRbplj4 +jCVGnPlac0Z/bv3Kb7Q7EOS+d+RFi1ZpsFYPbW5KRhGshzOxGw5d/nCjkEXCV0ht +uK6KndjFOvCGfikZW7WVpw7bkCe0W2Ko/JSX99ccJBDyau1NAoIBAQDGLj/wVxss +dllwswBOAV3oSL9xbUPs+Xbr/FK4XKcL7wcY6cfdZBlsQLQCoAzyTZ8t+g756Hlt +a8qmW2/Wvdo+m63Z2ecnbI9pJsVyYeT8pVumx4biHuXbRBYO/0ZZPtB5kTT6Hzhv +ZHxoUj4jb7L/5kwEdEPFIZX4rVEXY20LJL5wtv2zEQylQk9kunjUgrP1L/GtcNh+ +QRzLXiJWAoC4HWcXmdxb/Hc0BU5geSwZL4bbl3YL3lwWvkd3aY17T90EjWy4uA6G +tsHCxbxauul1q8OkmCcLEfhYnDh95YoVddR97XhC33S0v4dYjX/iS46g8fJ0HhRo +9YGWsD+tRevDAoIBAFCp/5/iTV3C8fbyfa1FI0SH2Bz2SV2Bal0sCzpoKwzdHq6U +znaYoLpCl+/MeCsi/FtUN/3umQ9n9/FjqshdcfavNsbJdJ1DJoUsVPN65FL1hTVv +LJOuUgMJ7g70e21I5fQEHb7S9scEIlvQeye/HVBpo2SEvGFoTQKiGHid1EPp1ies +NfYkhvkW9jIqD2Yg0IwrkFhDoaEOySGG58Q/ainw8/l2lRvUmucSzenFoyPh/Wgd +YIiBQI1mPyAGbLLBf9+jEIIprHsvVcFeMLiaumoDPVM44LbG5mj7Rw7QNVV+iN2A +dbkgLJIFQ3z6IUQk/ZlE+qoRkprSuctzSCil4jkCggEAdiYWilNz6NL5yX193gNk +l9nfAGFS0JF8+31nV3AtSqkLAyhEtlE58ta0Oqhub3olPwTILucQlVJg80Kp700q +Mo8fWzRUYaWP7fFmXyXLnW97r3dei6o+ALWbrP81UnlnUkJmYgOA4q/2lz8Iupma +DoOeqD0kNf8q6KFzKc1lsfIK8ym1IC826cMZkAS3ioINhUw6+dq/xq1M3FVXhQ1i +7eDhmClrPQ/LhSDwtAUpbC5waLPodXTwU8LG2oL8DRr0ugUSXyGjz15fL54xB6pN +CpEHRzZKeIgTFci0ySGya87eiuCrBLsxWZyhtQJOznubIYp8sAtKwbQzuMGEhOmd +fwKCAQEAlZoi1SzHstg6PfpwrIHJV3imLa550k9hyAu610CKjMsg6IsFsgu9/J0b +9hkhlafeW+p9zhKSwjl3aCuWUMNE53R+zYmMIJJrBzejC+1H0SKW0Zix9+ghixOX +da1jRaUxUqApJYvvxUC8FbnATM/Eq0ofhGkG3o575SlO+54twJO+bXGAUf/C6xMY +AQUQh90pTbZ96Q3Wdm2Qmrhd/GUaC6k1vAHVHHU8WQHiLmo1fF4gL/TqRv5KEPUM +un6ld7h8BEWtMClhSIiL2h5nvSYGcB6Lai6rPO0UUbGkWBQFpGaeglUmoYi0ciC5 +lMRrRHGUiWHW9C4/siOKYrHBeH5oNQ== -----END PRIVATE KEY----- diff --git a/tests/integration/test_ssl_cert_authentication/configs/users_with_ssl_auth.xml b/tests/integration/test_ssl_cert_authentication/configs/users_with_ssl_auth.xml index c6ac737b632..4bd30163ea6 100644 --- a/tests/integration/test_ssl_cert_authentication/configs/users_with_ssl_auth.xml +++ b/tests/integration/test_ssl_cert_authentication/configs/users_with_ssl_auth.xml @@ -11,6 +11,12 @@ client3 + + + URI:spiffe://foo.com/bar + URI:spiffe://foo.com/baz + + diff --git a/tests/integration/test_ssl_cert_authentication/test.py b/tests/integration/test_ssl_cert_authentication/test.py index 22d41bb6e14..3af88759e82 100644 --- a/tests/integration/test_ssl_cert_authentication/test.py +++ b/tests/integration/test_ssl_cert_authentication/test.py @@ -43,15 +43,10 @@ def started_cluster(): config = """ - none - + strict {certificateFile} {privateKeyFile} {caConfig} - - - AcceptCertificateHandler - """ @@ -338,3 +333,39 @@ def test_create_user(): == 'emma\tssl_certificate\t{"common_names":["client2"]}\n' 'lucy\tssl_certificate\t{"common_names":["client2","client3"]}\n' ) + + +def test_x509_san_support(): + assert ( + execute_query_native( + instance, "SELECT currentUser()", user="jerome", cert_name="client4" + ) + == "jerome\n" + ) + assert ( + execute_query_https("SELECT currentUser()", user="jerome", cert_name="client4") + == "jerome\n" + ) + assert ( + instance.query( + "SELECT name, auth_type, auth_params FROM system.users WHERE name='jerome'" + ) + == 'jerome\tssl_certificate\t{"subject_alt_names":["URI:spiffe:\\\\/\\\\/foo.com\\\\/bar","URI:spiffe:\\\\/\\\\/foo.com\\\\/baz"]}\n' + ) + # user `jerome` is configured via xml config, but `show create` should work regardless. + assert ( + instance.query("SHOW CREATE USER jerome") + == "CREATE USER jerome IDENTIFIED WITH ssl_certificate SAN \\'URI:spiffe://foo.com/bar\\', \\'URI:spiffe://foo.com/baz\\'\n" + ) + + instance.query( + "CREATE USER jemma IDENTIFIED WITH ssl_certificate SAN 'URI:spiffe://foo.com/bar', 'URI:spiffe://foo.com/baz'" + ) + assert ( + execute_query_https("SELECT currentUser()", user="jemma", cert_name="client4") + == "jemma\n" + ) + assert ( + instance.query("SHOW CREATE USER jemma") + == "CREATE USER jemma IDENTIFIED WITH ssl_certificate SAN \\'URI:spiffe://foo.com/bar\\', \\'URI:spiffe://foo.com/baz\\'\n" + ) diff --git a/tests/integration/test_startup_scripts/__init__.py b/tests/integration/test_startup_scripts/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_startup_scripts/configs/config.d/query_log.xml b/tests/integration/test_startup_scripts/configs/config.d/query_log.xml new file mode 100644 index 00000000000..24d66fc674e --- /dev/null +++ b/tests/integration/test_startup_scripts/configs/config.d/query_log.xml @@ -0,0 +1,8 @@ + + + system + query_log
+ toYYYYMM(event_date) + 1000 +
+
diff --git a/tests/integration/test_startup_scripts/configs/config.d/startup_scripts.xml b/tests/integration/test_startup_scripts/configs/config.d/startup_scripts.xml new file mode 100644 index 00000000000..e8a711a926a --- /dev/null +++ b/tests/integration/test_startup_scripts/configs/config.d/startup_scripts.xml @@ -0,0 +1,17 @@ + + + + CREATE ROLE OR REPLACE testrole + + + GRANT CREATE USER, ALTER USER, DROP USER, SHOW USERS, SHOW CREATE USER ON *.* TO 'testrole' WITH GRANT OPTION; + + + CREATE TABLE TestTable (id UInt64) ENGINE=TinyLog + SELECT 1; + + + SELECT * FROM system.query_log LIMIT 1 + + + diff --git a/tests/integration/test_startup_scripts/configs/users.xml b/tests/integration/test_startup_scripts/configs/users.xml new file mode 100644 index 00000000000..f9917b034b2 --- /dev/null +++ b/tests/integration/test_startup_scripts/configs/users.xml @@ -0,0 +1,41 @@ + + + + + + + + 1 + + + + + + + + + + ::/0 + + + default + + default + + + + + + + + 3600 + + 0 + 0 + 0 + 0 + 0 + + + + diff --git a/tests/integration/test_startup_scripts/test.py b/tests/integration/test_startup_scripts/test.py new file mode 100644 index 00000000000..43a871a6fc5 --- /dev/null +++ b/tests/integration/test_startup_scripts/test.py @@ -0,0 +1,21 @@ +from helpers.cluster import ClickHouseCluster + + +def test_startup_scripts(): + cluster = ClickHouseCluster(__file__) + + node = cluster.add_instance( + "node", + main_configs=[ + "configs/config.d/query_log.xml", + "configs/config.d/startup_scripts.xml", + ], + with_zookeeper=False, + ) + + try: + cluster.start() + assert node.query("SHOW TABLES") == "TestTable\n" + + finally: + cluster.shutdown() diff --git a/tests/integration/test_storage_azure_blob_storage/test.py b/tests/integration/test_storage_azure_blob_storage/test.py index 20b004a7605..6fbe7634642 100644 --- a/tests/integration/test_storage_azure_blob_storage/test.py +++ b/tests/integration/test_storage_azure_blob_storage/test.py @@ -809,7 +809,7 @@ def test_read_subcolumn_time(cluster): def test_read_from_not_existing_container(cluster): node = cluster.instances["node"] query = ( - f"select * from azureBlobStorage('{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', 'cont_not_exists', 'test_table.csv', " + f"select * from azureBlobStorage('{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', 'cont-not-exists', 'test_table.csv', " f"'devstoreaccount1', 'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'CSV', 'auto')" ) expected_err_msg = "container does not exist" diff --git a/tests/integration/test_storage_delta/test.py b/tests/integration/test_storage_delta/test.py index 25f0b58e0f5..d3dd7cfe52a 100644 --- a/tests/integration/test_storage_delta/test.py +++ b/tests/integration/test_storage_delta/test.py @@ -153,7 +153,7 @@ def test_single_log_file(started_cluster): bucket = started_cluster.minio_bucket TABLE_NAME = "test_single_log_file" - inserted_data = "SELECT number, toString(number + 1) FROM numbers(100)" + inserted_data = "SELECT number as a, toString(number + 1) as b FROM numbers(100)" parquet_data_path = create_initial_data_file( started_cluster, instance, inserted_data, TABLE_NAME ) @@ -511,3 +511,201 @@ def test_restart_broken_table_function(started_cluster): upload_directory(minio_client, bucket, f"/{TABLE_NAME}", "") assert int(instance.query(f"SELECT count() FROM {TABLE_NAME}")) == 100 + + +def test_partition_columns(started_cluster): + instance = started_cluster.instances["node1"] + spark = started_cluster.spark_session + minio_client = started_cluster.minio_client + bucket = started_cluster.minio_bucket + TABLE_NAME = "test_partition_columns" + result_file = f"{TABLE_NAME}" + partition_columns = ["b", "c", "d", "e"] + + delta_table = ( + DeltaTable.create(spark) + .tableName(TABLE_NAME) + .location(f"/{result_file}") + .addColumn("a", "INT") + .addColumn("b", "STRING") + .addColumn("c", "DATE") + .addColumn("d", "INT") + .addColumn("e", "BOOLEAN") + .partitionedBy(partition_columns) + .execute() + ) + num_rows = 9 + + schema = StructType( + [ + StructField("a", IntegerType()), + StructField("b", StringType()), + StructField("c", DateType()), + StructField("d", IntegerType()), + StructField("e", BooleanType()), + ] + ) + + for i in range(1, num_rows + 1): + data = [ + ( + i, + "test" + str(i), + datetime.strptime(f"2000-01-0{i}", "%Y-%m-%d"), + i, + False, + ) + ] + df = spark.createDataFrame(data=data, schema=schema) + df.printSchema() + df.write.mode("append").format("delta").partitionBy(partition_columns).save( + f"/{TABLE_NAME}" + ) + + minio_client = started_cluster.minio_client + bucket = started_cluster.minio_bucket + + files = upload_directory(minio_client, bucket, f"/{TABLE_NAME}", "") + assert len(files) > 0 + print(f"Uploaded files: {files}") + + result = instance.query( + f"describe table deltaLake('http://{started_cluster.minio_ip}:{started_cluster.minio_port}/{bucket}/{result_file}/', 'minio', 'minio123')" + ).strip() + + assert ( + result + == "a\tNullable(Int32)\t\t\t\t\t\nb\tNullable(String)\t\t\t\t\t\nc\tNullable(Date32)\t\t\t\t\t\nd\tNullable(Int32)\t\t\t\t\t\ne\tNullable(Bool)" + ) + + result = int( + instance.query( + f"""SELECT count() + FROM deltaLake('http://{started_cluster.minio_ip}:{started_cluster.minio_port}/{bucket}/{result_file}/', 'minio', 'minio123') + """ + ) + ) + assert result == num_rows + result = int( + instance.query( + f"""SELECT count() + FROM deltaLake('http://{started_cluster.minio_ip}:{started_cluster.minio_port}/{bucket}/{result_file}/', 'minio', 'minio123') + WHERE c == toDateTime('2000/01/05') + """ + ) + ) + assert result == 1 + + instance.query( + f""" + DROP TABLE IF EXISTS {TABLE_NAME}; + CREATE TABLE {TABLE_NAME} (a Nullable(Int32), b Nullable(String), c Nullable(Date32), d Nullable(Int32), e Nullable(Bool)) + ENGINE=DeltaLake('http://{started_cluster.minio_ip}:{started_cluster.minio_port}/{bucket}/{result_file}/', 'minio', 'minio123')""" + ) + assert ( + """1 test1 2000-01-01 1 false +2 test2 2000-01-02 2 false +3 test3 2000-01-03 3 false +4 test4 2000-01-04 4 false +5 test5 2000-01-05 5 false +6 test6 2000-01-06 6 false +7 test7 2000-01-07 7 false +8 test8 2000-01-08 8 false +9 test9 2000-01-09 9 false""" + == instance.query(f"SELECT * FROM {TABLE_NAME} ORDER BY b").strip() + ) + + assert ( + int( + instance.query( + f"SELECT count() FROM {TABLE_NAME} WHERE c == toDateTime('2000/01/05')" + ) + ) + == 1 + ) + + # Subset of columns should work. + instance.query( + f""" + DROP TABLE IF EXISTS {TABLE_NAME}; + CREATE TABLE {TABLE_NAME} (b Nullable(String), c Nullable(Date32), d Nullable(Int32)) + ENGINE=DeltaLake('http://{started_cluster.minio_ip}:{started_cluster.minio_port}/{bucket}/{result_file}/', 'minio', 'minio123')""" + ) + assert ( + """test1 2000-01-01 1 +test2 2000-01-02 2 +test3 2000-01-03 3 +test4 2000-01-04 4 +test5 2000-01-05 5 +test6 2000-01-06 6 +test7 2000-01-07 7 +test8 2000-01-08 8 +test9 2000-01-09 9""" + == instance.query(f"SELECT * FROM {TABLE_NAME} ORDER BY b").strip() + ) + + for i in range(num_rows + 1, 2 * num_rows + 1): + data = [ + ( + i, + "test" + str(i), + datetime.strptime(f"2000-01-{i}", "%Y-%m-%d"), + i, + False, + ) + ] + df = spark.createDataFrame(data=data, schema=schema) + df.printSchema() + df.write.mode("append").format("delta").partitionBy(partition_columns).save( + f"/{TABLE_NAME}" + ) + + files = upload_directory(minio_client, bucket, f"/{TABLE_NAME}", "") + ok = False + for file in files: + if file.endswith("last_checkpoint"): + ok = True + assert ok + + result = int( + instance.query( + f"""SELECT count() + FROM deltaLake('http://{started_cluster.minio_ip}:{started_cluster.minio_port}/{bucket}/{result_file}/', 'minio', 'minio123') + """ + ) + ) + assert result == num_rows * 2 + + assert ( + """1 test1 2000-01-01 1 false +2 test2 2000-01-02 2 false +3 test3 2000-01-03 3 false +4 test4 2000-01-04 4 false +5 test5 2000-01-05 5 false +6 test6 2000-01-06 6 false +7 test7 2000-01-07 7 false +8 test8 2000-01-08 8 false +9 test9 2000-01-09 9 false +10 test10 2000-01-10 10 false +11 test11 2000-01-11 11 false +12 test12 2000-01-12 12 false +13 test13 2000-01-13 13 false +14 test14 2000-01-14 14 false +15 test15 2000-01-15 15 false +16 test16 2000-01-16 16 false +17 test17 2000-01-17 17 false +18 test18 2000-01-18 18 false""" + == instance.query( + f""" +SELECT * FROM deltaLake('http://{started_cluster.minio_ip}:{started_cluster.minio_port}/{bucket}/{result_file}/', 'minio', 'minio123') ORDER BY c + """ + ).strip() + ) + assert ( + int( + instance.query( + f"SELECT count() FROM {TABLE_NAME} WHERE c == toDateTime('2000/01/15')" + ) + ) + == 1 + ) diff --git a/tests/integration/test_storage_kerberized_hdfs/configs/hdfs.xml b/tests/integration/test_storage_kerberized_hdfs/configs/hdfs.xml deleted file mode 100644 index 2e9dc6cc06c..00000000000 --- a/tests/integration/test_storage_kerberized_hdfs/configs/hdfs.xml +++ /dev/null @@ -1,13 +0,0 @@ - - - /tmp/keytab/clickhouse.keytab - root@TEST.CLICKHOUSE.TECH - kerberos - - - specuser@TEST.CLICKHOUSE.TECH - - - /tmp/kerb_cache - - diff --git a/tests/integration/test_storage_kerberized_hdfs/hdfs_configs/bootstrap.sh b/tests/integration/test_storage_kerberized_hdfs/hdfs_configs/bootstrap.sh deleted file mode 100755 index db6921bc1c8..00000000000 --- a/tests/integration/test_storage_kerberized_hdfs/hdfs_configs/bootstrap.sh +++ /dev/null @@ -1,280 +0,0 @@ -#!/bin/bash - -: "${HADOOP_PREFIX:=/usr/local/hadoop}" - -cat >> $HADOOP_PREFIX/etc/hadoop/hadoop-env.sh < /usr/local/hadoop/etc/hadoop/core-site.xml - -cat >> /usr/local/hadoop/etc/hadoop/core-site.xml << EOF - - hadoop.security.authentication - kerberos - - - hadoop.security.authorization - true - - - fs.defaultFS - hdfs://kerberizedhdfs1:9010 - - - fs.default.name - hdfs://kerberizedhdfs1:9010 - - - -EOF - - -cat > /usr/local/hadoop/etc/hadoop/hdfs-site.xml << EOF - - - - dfs.replication - 1 - - - - dfs.block.access.token.enable - true - - - - dfs.namenode.keytab.file - /usr/local/hadoop/etc/hadoop/conf/hdfs.keytab - - - dfs.namenode.kerberos.principal - hdfs/_HOST@TEST.CLICKHOUSE.TECH - - - dfs.namenode.kerberos.internal.spnego.principal - HTTP/_HOST@TEST.CLICKHOUSE.TECH - - - - dfs.secondary.namenode.keytab.file - /usr/local/hadoop/etc/hadoop/conf/hdfs.keytab - - - dfs.secondary.namenode.kerberos.principal - hdfs/_HOST@TEST.CLICKHOUSE.TECH - - - dfs.secondary.namenode.kerberos.internal.spnego.principal - HTTP/_HOST@TEST.CLICKHOUSE.TECH - - - - dfs.datanode.data.dir.perm - 700 - - - dfs.datanode.address - 0.0.0.0:1004 - - - dfs.datanode.http.address - 0.0.0.0:1006 - - - - dfs.datanode.ipc.address - 0.0.0.0:0 - - - dfs.namenode.secondary.http-address - 0.0.0.0:0 - - - dfs.namenode.backup.address - 0.0.0.0:0 - - - dfs.namenode.backup.http-address - 0.0.0.0:0 - - - - - - dfs.datanode.keytab.file - /usr/local/hadoop/etc/hadoop/conf/hdfs.keytab - - - dfs.datanode.kerberos.principal - hdfs/_HOST@TEST.CLICKHOUSE.TECH - - - - - dfs.webhdfs.enabled - true - - -dfs.encrypt.data.transfer -false - - - dfs.web.authentication.kerberos.principal - HTTP/_HOST@TEST.CLICKHOUSE.TECH - - - dfs.web.authentication.kerberos.keytab - /usr/local/hadoop/etc/hadoop/conf/hdfs.keytab - - -EOF - - - -# cat > /usr/local/hadoop/etc/hadoop/ssl-server.xml << EOF -# -# -# ssl.server.truststore.location -# /usr/local/hadoop/etc/hadoop/conf/hdfs.jks -# -# -# ssl.server.truststore.password -# masterkey -# -# -# ssl.server.keystore.location -# /usr/local/hadoop/etc/hadoop/conf/hdfs.jks -# -# -# ssl.server.keystore.password -# masterkey -# -# -# ssl.server.keystore.keypassword -# masterkey -# -# -# EOF - -cat > /usr/local/hadoop/etc/hadoop/log4j.properties << EOF -# Set everything to be logged to the console -log4j.rootCategory=DEBUG, console -log4j.appender.console=org.apache.log4j.ConsoleAppender -log4j.appender.console.target=System.err -log4j.appender.console.layout=org.apache.log4j.PatternLayout -log4j.appender.console.layout.ConversionPattern=%d{yy/MM/dd HH:mm:ss} %p %c{1}: %m%n - -# Set the default spark-shell log level to WARN. When running the spark-shell, the -# log level for this class is used to overwrite the root logger's log level, so that -# the user can have different defaults for the shell and regular Spark apps. -log4j.logger.org.apache.spark.repl.Main=INFO - -# Settings to quiet third party logs that are too verbose -log4j.logger.org.spark_project.jetty=DEBUG -log4j.logger.org.spark_project.jetty.util.component.AbstractLifeCycle=ERROR -log4j.logger.org.apache.spark.repl.SparkIMain$exprTyper=INFO -log4j.logger.org.apache.spark.repl.SparkILoop$SparkILoopInterpreter=INFO -log4j.logger.org.apache.parquet=ERROR -log4j.logger.parquet=ERROR - -# SPARK-9183: Settings to avoid annoying messages when looking up nonexistent UDFs in SparkSQL with Hive support -log4j.logger.org.apache.hadoop.hive.metastore.RetryingHMSHandler=FATAL -log4j.logger.org.apache.hadoop.hive.ql.exec.FunctionRegistry=ERROR - -log4j.logger.org.apache.spark.deploy=DEBUG -log4j.logger.org.apache.spark.executor=DEBUG -log4j.logger.org.apache.spark.scheduler=DEBUG -EOF - -useradd -u 1098 hdfs - -# keytool -genkey -alias kerberized_hdfs1.test.clickhouse.com -keyalg rsa -keysize 1024 -dname "CN=kerberized_hdfs1.test.clickhouse.com" -keypass masterkey -keystore /usr/local/hadoop/etc/hadoop/conf/hdfs.jks -storepass masterkey -keytool -genkey -alias kerberizedhdfs1 -keyalg rsa -keysize 1024 -dname "CN=kerberizedhdfs1" -keypass masterkey -keystore /usr/local/hadoop/etc/hadoop/conf/hdfs.jks -storepass masterkey - -chmod g+r /usr/local/hadoop/etc/hadoop/conf/hdfs.jks - - -service sshd start - -# yum --quiet --assumeyes install krb5-workstation.x86_64 -# yum --quiet --assumeyes install tcpdump - -# cd /tmp -# curl http://archive.apache.org/dist/commons/daemon/source/commons-daemon-1.0.15-src.tar.gz -o commons-daemon-1.0.15-src.tar.gz -# tar xzf commons-daemon-1.0.15-src.tar.gz -# cd commons-daemon-1.0.15-src/src/native/unix -# ./configure && make -# cp ./jsvc /usr/local/hadoop/sbin - - -until kinit -kt /usr/local/hadoop/etc/hadoop/conf/hdfs.keytab hdfs/kerberizedhdfs1@TEST.CLICKHOUSE.TECH; do sleep 2; done -echo "KDC is up and ready to go... starting up" - -$HADOOP_PREFIX/sbin/start-dfs.sh -$HADOOP_PREFIX/sbin/start-yarn.sh -$HADOOP_PREFIX/sbin/mr-jobhistory-daemon.sh start historyserver - -chmod a+r /usr/local/hadoop/etc/hadoop/conf/hdfs.keytab # create dedicated keytab for hdfsuser - -$HADOOP_PREFIX/sbin/start-secure-dns.sh -sleep 3 - -/usr/local/hadoop/bin/hdfs dfsadmin -safemode leave - -/usr/local/hadoop/bin/hdfs dfs -mkdir /user/specuser -/usr/local/hadoop/bin/hdfs dfs -chown specuser /user/specuser -echo "chown_completed" | /usr/local/hadoop/bin/hdfs dfs -appendToFile - /preparations_done_marker - -kdestroy - - - -# adduser --groups hdfs hdfsuser - -# /usr/local/hadoop/sbin/hadoop-daemon.sh --config /usr/local/hadoop/etc/hadoop/ --script /usr/local/hadoop/sbin/hdfs start namenode -# /usr/local/hadoop/sbin/hadoop-daemon.sh --config /usr/local/hadoop/etc/hadoop/ --script /usr/local/hadoop/sbin/hdfs start datanode - - -if [[ $1 == "-d" ]]; then - while true; do sleep 1000; done -fi - -if [[ $1 == "-bash" ]]; then - /bin/bash -fi diff --git a/tests/integration/test_storage_kerberized_hdfs/kerberos_image_config.sh b/tests/integration/test_storage_kerberized_hdfs/kerberos_image_config.sh deleted file mode 100644 index 45fb93792e0..00000000000 --- a/tests/integration/test_storage_kerberized_hdfs/kerberos_image_config.sh +++ /dev/null @@ -1,140 +0,0 @@ -#!/bin/bash - - -set -x # trace - -: "${REALM:=TEST.CLICKHOUSE.TECH}" -: "${DOMAIN_REALM:=test.clickhouse.com}" -: "${KERB_MASTER_KEY:=masterkey}" -: "${KERB_ADMIN_USER:=admin}" -: "${KERB_ADMIN_PASS:=admin}" - -create_config() { - : "${KDC_ADDRESS:=$(hostname -f)}" - - cat>/etc/krb5.conf</var/kerberos/krb5kdc/kdc.conf< /var/kerberos/krb5kdc/kadm5.acl -} - -create_keytabs() { - rm /tmp/keytab/*.keytab - - - # kadmin.local -q "addprinc -randkey hdfs/kerberizedhdfs1.${DOMAIN_REALM}@${REALM}" - # kadmin.local -q "ktadd -norandkey -k /tmp/keytab/hdfs.keytab hdfs/kerberizedhdfs1.${DOMAIN_REALM}@${REALM}" - - # kadmin.local -q "addprinc -randkey HTTP/kerberizedhdfs1.${DOMAIN_REALM}@${REALM}" - # kadmin.local -q "ktadd -norandkey -k /tmp/keytab/hdfs.keytab HTTP/kerberizedhdfs1.${DOMAIN_REALM}@${REALM}" - kadmin.local -q "addprinc -randkey hdfs/kerberizedhdfs1@${REALM}" - kadmin.local -q "ktadd -norandkey -k /tmp/keytab/hdfs.keytab hdfs/kerberizedhdfs1@${REALM}" - - kadmin.local -q "addprinc -randkey HTTP/kerberizedhdfs1@${REALM}" - kadmin.local -q "ktadd -norandkey -k /tmp/keytab/hdfs.keytab HTTP/kerberizedhdfs1@${REALM}" - - kadmin.local -q "addprinc -randkey hdfsuser/node1@${REALM}" - kadmin.local -q "ktadd -norandkey -k /tmp/keytab/clickhouse.keytab hdfsuser/node1@${REALM}" - kadmin.local -q "addprinc -randkey hdfsuser@${REALM}" - kadmin.local -q "ktadd -norandkey -k /tmp/keytab/clickhouse.keytab hdfsuser@${REALM}" - kadmin.local -q "addprinc -randkey root@${REALM}" - kadmin.local -q "ktadd -norandkey -k /tmp/keytab/clickhouse.keytab root@${REALM}" - kadmin.local -q "addprinc -randkey specuser@${REALM}" - kadmin.local -q "ktadd -norandkey -k /tmp/keytab/clickhouse.keytab specuser@${REALM}" - - chmod g+r /tmp/keytab/clickhouse.keytab -} - -main() { - - if [ ! -f /kerberos_initialized ]; then - create_config - create_db - create_admin_user - start_kdc - - touch /kerberos_initialized - fi - - if [ ! -f /var/kerberos/krb5kdc/principal ]; then - while true; do sleep 1000; done - else - start_kdc - create_keytabs - tail -F /var/log/kerberos/krb5kdc.log - fi - -} - -[[ "$0" == "${BASH_SOURCE[0]}" ]] && main "$@" diff --git a/tests/integration/test_storage_kerberized_hdfs/secrets/krb.conf b/tests/integration/test_storage_kerberized_hdfs/secrets/krb.conf deleted file mode 100644 index dffdcaebe81..00000000000 --- a/tests/integration/test_storage_kerberized_hdfs/secrets/krb.conf +++ /dev/null @@ -1,25 +0,0 @@ -[logging] - default = FILE:/var/log/krb5libs.log - kdc = FILE:/var/log/krb5kdc.log - admin_server = FILE:/var/log/kadmind.log - -[libdefaults] - default_realm = TEST.CLICKHOUSE.TECH - dns_lookup_realm = false - dns_lookup_kdc = false - ticket_lifetime = 5s - forwardable = true - rdns = false - default_tgs_enctypes = des3-hmac-sha1 - default_tkt_enctypes = des3-hmac-sha1 - permitted_enctypes = des3-hmac-sha1 - -[realms] - TEST.CLICKHOUSE.TECH = { - kdc = hdfskerberos - admin_server = hdfskerberos - } - -[domain_realm] - .test.clickhouse.com = TEST.CLICKHOUSE.TECH - test.clickhouse.com = TEST.CLICKHOUSE.TECH diff --git a/tests/integration/test_storage_kerberized_hdfs/secrets/krb_long.conf b/tests/integration/test_storage_kerberized_hdfs/secrets/krb_long.conf deleted file mode 100644 index 43c009d2e98..00000000000 --- a/tests/integration/test_storage_kerberized_hdfs/secrets/krb_long.conf +++ /dev/null @@ -1,24 +0,0 @@ -[logging] - default = FILE:/var/log/krb5libs.log - kdc = FILE:/var/log/krb5kdc.log - admin_server = FILE:/var/log/kadmind.log - -[libdefaults] - default_realm = TEST.CLICKHOUSE.TECH - dns_lookup_realm = false - dns_lookup_kdc = false - ticket_lifetime = 15d - forwardable = true - default_tgs_enctypes = des3-hmac-sha1 - default_tkt_enctypes = des3-hmac-sha1 - permitted_enctypes = des3-hmac-sha1 - -[realms] - TEST.CLICKHOUSE.TECH = { - kdc = hdfskerberos - admin_server = hdfskerberos - } - -[domain_realm] - .test.clickhouse.com = TEST.CLICKHOUSE.TECH - test.clickhouse.com = TEST.CLICKHOUSE.TECH diff --git a/tests/integration/test_storage_kerberized_hdfs/test.py b/tests/integration/test_storage_kerberized_hdfs/test.py deleted file mode 100644 index ddfc1f6483d..00000000000 --- a/tests/integration/test_storage_kerberized_hdfs/test.py +++ /dev/null @@ -1,155 +0,0 @@ -import time -import pytest - -import os - -from helpers.cluster import ClickHouseCluster, is_arm -import subprocess - -if is_arm(): - pytestmark = pytest.mark.skip - -cluster = ClickHouseCluster(__file__) -node1 = cluster.add_instance( - "node1", - with_kerberized_hdfs=True, - user_configs=[], - main_configs=["configs/hdfs.xml"], -) - - -@pytest.fixture(scope="module") -def started_cluster(): - try: - cluster.start() - - yield cluster - - except Exception as ex: - print(ex) - raise ex - finally: - cluster.shutdown() - - -def test_read_table(started_cluster): - hdfs_api = started_cluster.hdfs_api - - data = "1\tSerialize\t555.222\n2\tData\t777.333\n" - hdfs_api.write_data("/simple_table_function", data) - - api_read = hdfs_api.read_data("/simple_table_function") - assert api_read == data - - select_read = node1.query( - "select * from hdfs('hdfs://kerberizedhdfs1:9010/simple_table_function', 'TSV', 'id UInt64, text String, number Float64')" - ) - assert select_read == data - - -def test_read_write_storage(started_cluster): - hdfs_api = started_cluster.hdfs_api - - node1.query( - "create table SimpleHDFSStorage2 (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://kerberizedhdfs1:9010/simple_storage1', 'TSV')" - ) - node1.query("insert into SimpleHDFSStorage2 values (1, 'Mark', 72.53)") - - api_read = hdfs_api.read_data("/simple_storage1") - assert api_read == "1\tMark\t72.53\n" - - select_read = node1.query("select * from SimpleHDFSStorage2") - assert select_read == "1\tMark\t72.53\n" - - -def test_write_storage_not_expired(started_cluster): - hdfs_api = started_cluster.hdfs_api - - node1.query( - "create table SimpleHDFSStorageNotExpired (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://kerberizedhdfs1:9010/simple_storage_not_expired', 'TSV')" - ) - - time.sleep(15) # wait for ticket expiration - node1.query("insert into SimpleHDFSStorageNotExpired values (1, 'Mark', 72.53)") - - api_read = hdfs_api.read_data("/simple_storage_not_expired") - assert api_read == "1\tMark\t72.53\n" - - select_read = node1.query("select * from SimpleHDFSStorageNotExpired") - assert select_read == "1\tMark\t72.53\n" - - -def test_two_users(started_cluster): - hdfs_api = started_cluster.hdfs_api - - node1.query( - "create table HDFSStorOne (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://kerberizedhdfs1:9010/storage_user_one', 'TSV')" - ) - node1.query("insert into HDFSStorOne values (1, 'Real', 86.00)") - - node1.query( - "create table HDFSStorTwo (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://suser@kerberizedhdfs1:9010/user/specuser/storage_user_two', 'TSV')" - ) - node1.query("insert into HDFSStorTwo values (1, 'Ideal', 74.00)") - - select_read_1 = node1.query( - "select * from hdfs('hdfs://kerberizedhdfs1:9010/user/specuser/storage_user_two', 'TSV', 'id UInt64, text String, number Float64')" - ) - - select_read_2 = node1.query( - "select * from hdfs('hdfs://suser@kerberizedhdfs1:9010/storage_user_one', 'TSV', 'id UInt64, text String, number Float64')" - ) - - -def test_read_table_expired(started_cluster): - hdfs_api = started_cluster.hdfs_api - - data = "1\tSerialize\t555.222\n2\tData\t777.333\n" - hdfs_api.write_data("/simple_table_function_relogin", data) - - started_cluster.pause_container("hdfskerberos") - time.sleep(15) - - try: - select_read = node1.query( - "select * from hdfs('hdfs://reloginuser&kerberizedhdfs1:9010/simple_table_function', 'TSV', 'id UInt64, text String, number Float64')" - ) - assert False, "Exception have to be thrown" - except Exception as ex: - assert "DB::Exception: KerberosInit failure:" in str(ex) - - started_cluster.unpause_container("hdfskerberos") - - -def test_prohibited(started_cluster): - node1.query( - "create table HDFSStorTwoProhibited (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://suser@kerberizedhdfs1:9010/storage_user_two_prohibited', 'TSV')" - ) - try: - node1.query("insert into HDFSStorTwoProhibited values (1, 'SomeOne', 74.00)") - assert False, "Exception have to be thrown" - except Exception as ex: - assert ( - "Unable to open HDFS file: /storage_user_two_prohibited (hdfs://suser@kerberizedhdfs1:9010/storage_user_two_prohibited) error: Permission denied: user=specuser, access=WRITE" - in str(ex) - ) - - -def test_cache_path(started_cluster): - node1.query( - "create table HDFSStorCachePath (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://dedicatedcachepath@kerberizedhdfs1:9010/storage_dedicated_cache_path', 'TSV')" - ) - try: - node1.query("insert into HDFSStorCachePath values (1, 'FatMark', 92.53)") - assert False, "Exception have to be thrown" - except Exception as ex: - assert ( - "DB::Exception: hadoop.security.kerberos.ticket.cache.path cannot be set per user" - in str(ex) - ) - - -if __name__ == "__main__": - cluster.start() - input("Cluster created, press any key to destroy...") - cluster.shutdown() diff --git a/tests/integration/test_storage_mysql/test.py b/tests/integration/test_storage_mysql/test.py index b131d8fe51c..5948954ff5f 100644 --- a/tests/integration/test_storage_mysql/test.py +++ b/tests/integration/test_storage_mysql/test.py @@ -908,6 +908,63 @@ def test_mysql_point(started_cluster): conn.close() +def test_joins(started_cluster): + conn = get_mysql_conn(started_cluster, cluster.mysql8_ip) + drop_mysql_table(conn, "test_joins_mysql_users") + with conn.cursor() as cursor: + cursor.execute( + "CREATE TABLE clickhouse.test_joins_mysql_users (id INT NOT NULL, name varchar(50) NOT NULL, created TIMESTAMP, PRIMARY KEY (`id`)) ENGINE=InnoDB;" + ) + cursor.execute( + f"INSERT INTO clickhouse.test_joins_mysql_users VALUES (469722, 'user@example.com', '2019-08-30 07:55:01')" + ) + + drop_mysql_table(conn, "test_joins_mysql_tickets") + with conn.cursor() as cursor: + cursor.execute( + "CREATE TABLE clickhouse.test_joins_mysql_tickets (id INT NOT NULL, subject varchar(50), created TIMESTAMP, creator INT NOT NULL, PRIMARY KEY (`id`)) ENGINE=InnoDB;" + ) + cursor.execute( + f"INSERT INTO clickhouse.test_joins_mysql_tickets VALUES (281607, 'Feedback', '2024-06-25 12:09:41', 469722)" + ) + + conn.commit() + + node1.query( + """ + CREATE TABLE test_joins_table_users + ( + `id` Int32, + `Name` String, + `Created` Nullable(DateTime) + ) + ENGINE = MySQL('mysql80:3306', 'clickhouse', 'test_joins_mysql_users', 'root', 'clickhouse'); + """ + ) + + node1.query( + """ + CREATE TABLE test_joins_table_tickets + ( + `id` Int32, + `Subject` Nullable(String), + `Created` Nullable(DateTime), + `Creator` Int32 + ) + ENGINE = MySQL('mysql80:3306', 'clickhouse', 'test_joins_mysql_tickets', 'root', 'clickhouse'); + """ + ) + + node1.query( + """ + SELECT test_joins_table_tickets.id, Subject, test_joins_table_tickets.Created, Name + FROM test_joins_table_tickets + LEFT JOIN test_joins_table_users ON test_joins_table_tickets.Creator = test_joins_table_users.id + WHERE test_joins_table_tickets.Created = '2024-06-25 12:09:41' + """ + ) == "281607\tFeedback\t2024-06-25 12:09:41\tuser@example.com\n" + + if __name__ == "__main__": with contextmanager(started_cluster)() as cluster: for name, instance in list(cluster.instances.items()): diff --git a/tests/integration/test_storage_postgresql/test.py b/tests/integration/test_storage_postgresql/test.py index d9f3a9917ab..12823f1f72d 100644 --- a/tests/integration/test_storage_postgresql/test.py +++ b/tests/integration/test_storage_postgresql/test.py @@ -834,6 +834,60 @@ def test_literal_escaping(started_cluster): cursor.execute(f"DROP TABLE escaping") +def test_filter_pushdown(started_cluster): + cursor = started_cluster.postgres_conn.cursor() + cursor.execute("CREATE SCHEMA test_filter_pushdown") + cursor.execute( + "CREATE TABLE test_filter_pushdown.test_table (id integer, value integer)" + ) + cursor.execute( + "INSERT INTO test_filter_pushdown.test_table VALUES (1, 10), (1, 110), (2, 0), (3, 33), (4, 0)" + ) + + node1.query( + """ + CREATE TABLE test_filter_pushdown_pg_table (id UInt32, value UInt32) + ENGINE PostgreSQL('postgres1:5432', 'postgres', 'test_table', 'postgres', 'mysecretpassword', 'test_filter_pushdown'); + """ + ) + + node1.query( + """ + CREATE TABLE test_filter_pushdown_local_table (id UInt32, value UInt32) ENGINE Memory AS SELECT * FROM test_filter_pushdown_pg_table + """ + ) + + node1.query( + "CREATE TABLE ch_table (id UInt32, pg_id UInt32) ENGINE MergeTree ORDER BY id" + ) + node1.query("INSERT INTO ch_table VALUES (1, 1), (2, 2), (3, 1), (4, 2), (5, 999)") + + def compare_results(query, **kwargs): + result1 = node1.query( + query.format(pg_table="test_filter_pushdown_pg_table", **kwargs) + ) + result2 = node1.query( + query.format(pg_table="test_filter_pushdown_local_table", **kwargs) + ) + assert result1 == result2 + + for kind in ["INNER", "LEFT", "RIGHT", "FULL", "ANY LEFT", "SEMI RIGHT"]: + for value in [0, 10]: + compare_results( + "SELECT * FROM ch_table {kind} JOIN {pg_table} as p ON ch_table.pg_id = p.id WHERE value = {value} ORDER BY ALL", + kind=kind, + value=value, + ) + + compare_results( + "SELECT * FROM {pg_table} as p {kind} JOIN ch_table ON ch_table.pg_id = p.id WHERE value = {value} ORDER BY ALL", + kind=kind, + value=value, + ) + + cursor.execute("DROP SCHEMA test_filter_pushdown CASCADE") + + if __name__ == "__main__": cluster.start() input("Cluster created, press any key to destroy...") diff --git a/tests/integration/test_storage_rabbitmq/test.py b/tests/integration/test_storage_rabbitmq/test.py index 3240039ee81..bc255b1964e 100644 --- a/tests/integration/test_storage_rabbitmq/test.py +++ b/tests/integration/test_storage_rabbitmq/test.py @@ -78,13 +78,13 @@ def wait_rabbitmq_to_start(rabbitmq_docker_id, cookie, timeout=180): def kill_rabbitmq(rabbitmq_id): p = subprocess.Popen(("docker", "stop", rabbitmq_id), stdout=subprocess.PIPE) - p.communicate() + p.wait(timeout=60) return p.returncode == 0 def revive_rabbitmq(rabbitmq_id, cookie): p = subprocess.Popen(("docker", "start", rabbitmq_id), stdout=subprocess.PIPE) - p.communicate() + p.wait(timeout=60) wait_rabbitmq_to_start(rabbitmq_id, cookie) @@ -2220,13 +2220,11 @@ def test_rabbitmq_commit_on_block_write(rabbitmq_cluster): def test_rabbitmq_no_connection_at_startup_1(rabbitmq_cluster): - # no connection when table is initialized - rabbitmq_cluster.pause_container("rabbitmq1") - instance.query_and_get_error( + error = instance.query_and_get_error( """ CREATE TABLE test.cs (key UInt64, value UInt64) ENGINE = RabbitMQ - SETTINGS rabbitmq_host_port = 'rabbitmq1:5672', + SETTINGS rabbitmq_host_port = 'no_connection_at_startup:5672', rabbitmq_exchange_name = 'cs', rabbitmq_format = 'JSONEachRow', rabbitmq_flush_interval_ms=1000, @@ -2234,7 +2232,7 @@ def test_rabbitmq_no_connection_at_startup_1(rabbitmq_cluster): rabbitmq_row_delimiter = '\\n'; """ ) - rabbitmq_cluster.unpause_container("rabbitmq1") + assert "CANNOT_CONNECT_RABBITMQ" in error def test_rabbitmq_no_connection_at_startup_2(rabbitmq_cluster): diff --git a/tests/integration/test_storage_s3/test.py b/tests/integration/test_storage_s3/test.py index 9a0cb352088..ab327afe90b 100644 --- a/tests/integration/test_storage_s3/test.py +++ b/tests/integration/test_storage_s3/test.py @@ -154,6 +154,7 @@ def test_put(started_cluster, maybe_auth, positive, compression): def test_partition_by(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] # type: ClickHouseInstance table_format = "column1 UInt32, column2 UInt32, column3 UInt32" @@ -161,26 +162,37 @@ def test_partition_by(started_cluster): values = "(1, 2, 3), (3, 2, 1), (78, 43, 45)" filename = "test_{_partition_id}.csv" put_query = f"""INSERT INTO TABLE FUNCTION - s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{filename}', 'CSV', '{table_format}') + s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{filename}', 'CSV', '{table_format}') PARTITION BY {partition_by} VALUES {values}""" run_query(instance, put_query) - assert "1,2,3\n" == get_s3_file_content(started_cluster, bucket, "test_3.csv") - assert "3,2,1\n" == get_s3_file_content(started_cluster, bucket, "test_1.csv") - assert "78,43,45\n" == get_s3_file_content(started_cluster, bucket, "test_45.csv") + assert "1,2,3\n" == get_s3_file_content(started_cluster, bucket, f"{id}/test_3.csv") + assert "3,2,1\n" == get_s3_file_content(started_cluster, bucket, f"{id}/test_1.csv") + assert "78,43,45\n" == get_s3_file_content( + started_cluster, bucket, f"{id}/test_45.csv" + ) filename = "test2_{_partition_id}.csv" instance.query( - f"create table p ({table_format}) engine=S3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{filename}', 'CSV') partition by column3" + f"create table p ({table_format}) engine=S3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{filename}', 'CSV') partition by column3" ) instance.query(f"insert into p values {values}") - assert "1,2,3\n" == get_s3_file_content(started_cluster, bucket, "test2_3.csv") - assert "3,2,1\n" == get_s3_file_content(started_cluster, bucket, "test2_1.csv") - assert "78,43,45\n" == get_s3_file_content(started_cluster, bucket, "test2_45.csv") + assert "1,2,3\n" == get_s3_file_content( + started_cluster, bucket, f"{id}/test2_3.csv" + ) + assert "3,2,1\n" == get_s3_file_content( + started_cluster, bucket, f"{id}/test2_1.csv" + ) + assert "78,43,45\n" == get_s3_file_content( + started_cluster, bucket, f"{id}/test2_45.csv" + ) + + instance.query("drop table p") def test_partition_by_string_column(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] # type: ClickHouseInstance table_format = "col_num UInt32, col_str String" @@ -188,19 +200,24 @@ def test_partition_by_string_column(started_cluster): values = "(1, 'foo/bar'), (3, 'йцук'), (78, '你好')" filename = "test_{_partition_id}.csv" put_query = f"""INSERT INTO TABLE FUNCTION - s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{filename}', 'CSV', '{table_format}') + s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{filename}', 'CSV', '{table_format}') PARTITION BY {partition_by} VALUES {values}""" run_query(instance, put_query) assert '1,"foo/bar"\n' == get_s3_file_content( - started_cluster, bucket, "test_foo/bar.csv" + started_cluster, bucket, f"{id}/test_foo/bar.csv" + ) + assert '3,"йцук"\n' == get_s3_file_content( + started_cluster, bucket, f"{id}/test_йцук.csv" + ) + assert '78,"你好"\n' == get_s3_file_content( + started_cluster, bucket, f"{id}/test_你好.csv" ) - assert '3,"йцук"\n' == get_s3_file_content(started_cluster, bucket, "test_йцук.csv") - assert '78,"你好"\n' == get_s3_file_content(started_cluster, bucket, "test_你好.csv") def test_partition_by_const_column(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] # type: ClickHouseInstance table_format = "column1 UInt32, column2 UInt32, column3 UInt32" @@ -209,12 +226,14 @@ def test_partition_by_const_column(started_cluster): values_csv = "1,2,3\n3,2,1\n78,43,45\n" filename = "test_{_partition_id}.csv" put_query = f"""INSERT INTO TABLE FUNCTION - s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{filename}', 'CSV', '{table_format}') + s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{filename}', 'CSV', '{table_format}') PARTITION BY {partition_by} VALUES {values}""" run_query(instance, put_query) - assert values_csv == get_s3_file_content(started_cluster, bucket, "test_88.csv") + assert values_csv == get_s3_file_content( + started_cluster, bucket, f"{id}/test_88.csv" + ) @pytest.mark.parametrize("special", ["space", "plus"]) @@ -274,46 +293,31 @@ def test_get_path_with_special(started_cluster, special): @pytest.mark.parametrize("auth", [pytest.param("'minio','minio123',", id="minio")]) def test_empty_put(started_cluster, auth): # type: (ClickHouseCluster, str) -> None - + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] # type: ClickHouseInstance table_format = "column1 UInt32, column2 UInt32, column3 UInt32" drop_empty_table_query = "DROP TABLE IF EXISTS empty_table" - create_empty_table_query = """ - CREATE TABLE empty_table ( - {} - ) ENGINE = Null() - """.format( - table_format + create_empty_table_query = ( + f"CREATE TABLE empty_table ({table_format}) ENGINE = Null()" ) run_query(instance, drop_empty_table_query) run_query(instance, create_empty_table_query) filename = "empty_put_test.csv" - put_query = "insert into table function s3('http://{}:{}/{}/{}', {}'CSV', '{}') select * from empty_table".format( - started_cluster.minio_ip, - MINIO_INTERNAL_PORT, - bucket, - filename, - auth, - table_format, - ) + put_query = f"""insert into table function + s3('http://{started_cluster.minio_ip}:{MINIO_INTERNAL_PORT}/{bucket}/{id}/{filename}', {auth} 'CSV', '{table_format}') + select * from empty_table""" run_query(instance, put_query) assert ( run_query( instance, - "select count(*) from s3('http://{}:{}/{}/{}', {}'CSV', '{}')".format( - started_cluster.minio_ip, - MINIO_INTERNAL_PORT, - bucket, - filename, - auth, - table_format, - ), + f"""select count(*) from + s3('http://{started_cluster.minio_ip}:{MINIO_INTERNAL_PORT}/{bucket}/{id}/{filename}', {auth} 'CSV', '{table_format}')""", ) == "0\n" ) @@ -497,6 +501,7 @@ def test_put_get_with_globs(started_cluster): def test_multipart(started_cluster, maybe_auth, positive): # type: (ClickHouseCluster, str, bool) -> None + id = uuid.uuid4() bucket = ( started_cluster.minio_bucket if not maybe_auth @@ -519,7 +524,7 @@ def test_multipart(started_cluster, maybe_auth, positive): assert len(csv_data) > min_part_size_bytes - filename = "test_multipart.csv" + filename = f"{id}/test_multipart.csv" put_query = "insert into table function s3('http://{}:{}/{}/{}', {}'CSV', '{}') format CSV".format( started_cluster.minio_redirect_host, started_cluster.minio_redirect_port, @@ -691,7 +696,7 @@ def test_s3_glob_many_objects_under_selection(started_cluster): def create_files(thread_num): for f_num in range(thread_num * 63, thread_num * 63 + 63): path = f"folder1/file{f_num}.csv" - query = "insert into table function s3('http://{}:{}/{}/{}', 'CSV', '{}') values {}".format( + query = "insert into table function s3('http://{}:{}/{}/{}', 'CSV', '{}') settings s3_truncate_on_insert=1 values {}".format( started_cluster.minio_ip, MINIO_INTERNAL_PORT, bucket, @@ -704,7 +709,7 @@ def test_s3_glob_many_objects_under_selection(started_cluster): jobs.append(threading.Thread(target=create_files, args=(thread_num,))) jobs[-1].start() - query = "insert into table function s3('http://{}:{}/{}/{}', 'CSV', '{}') values {}".format( + query = "insert into table function s3('http://{}:{}/{}/{}', 'CSV', '{}') settings s3_truncate_on_insert=1 values {}".format( started_cluster.minio_ip, MINIO_INTERNAL_PORT, bucket, @@ -879,7 +884,7 @@ def test_storage_s3_get_unstable(started_cluster): bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] table_format = "column1 Int64, column2 Int64, column3 Int64, column4 Int64" - get_query = f"SELECT count(), sum(column3), sum(column4) FROM s3('http://resolver:8081/{started_cluster.minio_bucket}/test.csv', 'CSV', '{table_format}') FORMAT CSV" + get_query = f"SELECT count(), sum(column3), sum(column4) FROM s3('http://resolver:8081/{started_cluster.minio_bucket}/test.csv', 'CSV', '{table_format}') SETTINGS s3_max_single_read_retries=30 FORMAT CSV" result = run_query(instance, get_query) assert result.splitlines() == ["500001,500000,0"] @@ -894,9 +899,10 @@ def test_storage_s3_get_slow(started_cluster): def test_storage_s3_put_uncompressed(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] - filename = "test_put_uncompressed.bin" + filename = f"{id}/test_put_uncompressed.bin" name = "test_put_uncompressed" data = [ "'Gloria Thompson',99", @@ -948,6 +954,7 @@ def test_storage_s3_put_uncompressed(started_cluster): r = result.strip().split("\t") assert int(r[0]) >= 1, blob_storage_log assert all(col == r[0] for col in r), blob_storage_log + run_query(instance, f"DROP TABLE {name}") @pytest.mark.parametrize( @@ -955,9 +962,10 @@ def test_storage_s3_put_uncompressed(started_cluster): [pytest.param("bin", "gzip", id="bin"), pytest.param("gz", "auto", id="gz")], ) def test_storage_s3_put_gzip(started_cluster, extension, method): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] - filename = f"test_put_gzip.{extension}" + filename = f"{id}/test_put_gzip.{extension}" name = f"test_put_gzip_{extension}" data = [ "'Joseph Tomlinson',5", @@ -994,6 +1002,7 @@ def test_storage_s3_put_gzip(started_cluster, extension, method): f = gzip.GzipFile(fileobj=buf, mode="rb") uncompressed_content = f.read().decode() assert sum([int(i.split(",")[1]) for i in uncompressed_content.splitlines()]) == 708 + run_query(instance, f"DROP TABLE {name}") def test_truncate_table(started_cluster): @@ -1019,14 +1028,24 @@ def test_truncate_table(started_cluster): len(list(minio.list_objects(started_cluster.minio_bucket, "truncate/"))) == 0 ): - return + break timeout -= 1 time.sleep(1) assert len(list(minio.list_objects(started_cluster.minio_bucket, "truncate/"))) == 0 - assert instance.query("SELECT * FROM {}".format(name)) == "" + # FIXME: there was a bug in test and it was never checked. + # Currently read from truncated table fails with + # DB::Exception: Failed to get object info: No response body.. + # HTTP response code: 404: while reading truncate: While executing S3Source + # assert instance.query("SELECT * FROM {}".format(name)) == "" + instance.query(f"DROP TABLE {name} SYNC") + assert ( + instance.query(f"SELECT count() FROM system.tables where name='{name}'") + == "0\n" + ) def test_predefined_connection_configuration(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances[ "dummy_without_named_collections" @@ -1054,7 +1073,9 @@ def test_predefined_connection_configuration(started_cluster): user="user", ) - instance.query(f"INSERT INTO {name} SELECT number FROM numbers(10)") + instance.query( + f"INSERT INTO {name} SELECT number FROM numbers(10) SETTINGS s3_truncate_on_insert=1" + ) result = instance.query(f"SELECT * FROM {name}") assert result == instance.query("SELECT number FROM numbers(10)") @@ -1068,9 +1089,11 @@ def test_predefined_connection_configuration(started_cluster): "To execute this query, it's necessary to have the grant NAMED COLLECTION ON no_collection" in error ) - instance = started_cluster.instances["dummy"] # has named collection access - error = instance.query_and_get_error("SELECT * FROM s3(no_collection)") + instance2 = started_cluster.instances["dummy"] # has named collection access + error = instance2.query_and_get_error("SELECT * FROM s3(no_collection)") assert "There is no named collection `no_collection`" in error + instance.query("DROP USER user") + instance.query(f"DROP TABLE {name}") result = "" @@ -1125,31 +1148,28 @@ def test_url_reconnect_in_the_middle(started_cluster): assert result == "1000000\t3914219105369203805\n" -def test_seekable_formats(started_cluster): - bucket = started_cluster.minio_bucket +# At the time of writing the actual read bytes are respectively 148 and 169, so -10% to not be flaky +@pytest.mark.parametrize( + "format_name,expected_bytes_read", [("Parquet", 133), ("ORC", 150)] +) +def test_seekable_formats(started_cluster, format_name, expected_bytes_read): + expected_lines = 1500000 instance = started_cluster.instances["dummy"] # type: ClickHouseInstance - table_function = f"s3(s3_parquet, structure='a Int32, b String', format='Parquet')" + table_function = f"s3(s3_{format_name.lower()}, structure='a Int32, b String', format='{format_name}')" exec_query_with_retry( instance, - f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(1000000) settings s3_truncate_on_insert=1", - timeout=100, + f"INSERT INTO TABLE FUNCTION {table_function} SELECT number, randomString(100) FROM numbers({expected_lines}) settings s3_truncate_on_insert=1", + timeout=300, ) result = instance.query(f"SELECT count() FROM {table_function}") - assert int(result) == 1000000 - - table_function = f"s3(s3_orc, structure='a Int32, b String', format='ORC')" - exec_query_with_retry( - instance, - f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(1500000) settings s3_truncate_on_insert=1", - timeout=100, - ) + assert int(result) == expected_lines result = instance.query( f"SELECT count() FROM {table_function} SETTINGS max_memory_usage='60M', max_download_threads=1" ) - assert int(result) == 1500000 + assert int(result) == expected_lines instance.query(f"SELECT * FROM {table_function} FORMAT Null") @@ -1160,35 +1180,31 @@ def test_seekable_formats(started_cluster): result = result.strip() assert result.endswith("MiB") result = result[: result.index(".")] - assert int(result) > 150 + assert int(result) > 140 -def test_seekable_formats_url(started_cluster): +@pytest.mark.parametrize("format_name", ["Parquet", "ORC"]) +def test_seekable_formats_url(started_cluster, format_name): bucket = started_cluster.minio_bucket + expected_lines = 1500000 instance = started_cluster.instances["dummy"] # type: ClickHouseInstance - table_function = f"s3(s3_parquet, structure='a Int32, b String', format='Parquet')" + format_name_lower = format_name.lower() + table_function = f"s3(s3_{format_name_lower}, structure='a Int32, b String', format='{format_name}')" exec_query_with_retry( instance, - f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(1500000) settings s3_truncate_on_insert=1", - timeout=100, + f"INSERT INTO TABLE FUNCTION {table_function} SELECT number, randomString(100) FROM numbers({expected_lines}) settings s3_truncate_on_insert=1", + timeout=300, ) result = instance.query(f"SELECT count() FROM {table_function}") - assert int(result) == 1500000 + assert int(result) == expected_lines - table_function = f"s3(s3_orc, structure='a Int32, b String', format='ORC')" - exec_query_with_retry( - instance, - f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(1500000) settings s3_truncate_on_insert=1", - timeout=100, - ) - - table_function = f"url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_parquet', 'Parquet', 'a Int32, b String')" + url_function = f"url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_{format_name_lower}', '{format_name}', 'a Int32, b String')" result = instance.query( - f"SELECT count() FROM {table_function} SETTINGS max_memory_usage='60M'" + f"SELECT count() FROM {url_function} SETTINGS max_memory_usage='60M'" ) - assert int(result) == 1500000 + assert int(result) == expected_lines def test_empty_file(started_cluster): @@ -1220,7 +1236,7 @@ def test_s3_schema_inference(started_cluster): instance = started_cluster.instances["dummy"] instance.query( - f"insert into table function s3(s3_native, structure='a Int32, b String', format='Native') select number, randomString(100) from numbers(5000000)" + f"insert into table function s3(s3_native, structure='a Int32, b String', format='Native') select number, randomString(100) from numbers(5000000) SETTINGS s3_truncate_on_insert=1" ) result = instance.query(f"desc s3(s3_native, format='Native')") assert result == "a\tInt32\t\t\t\t\t\nb\tString\t\t\t\t\t\n" @@ -1260,6 +1276,9 @@ def test_s3_schema_inference(started_cluster): result = instance.query(f"select count(*) from {table_function}") assert int(result) == 5000000 + instance.query("drop table schema_inference") + instance.query("drop table schema_inference_2") + def test_empty_file(started_cluster): bucket = started_cluster.minio_bucket @@ -1295,6 +1314,7 @@ def test_overwrite(started_cluster): result = instance.query(f"select count() from test_overwrite") assert int(result) == 200 + instance.query(f"drop table test_overwrite") def test_create_new_files_on_insert(started_cluster): @@ -1336,6 +1356,7 @@ def test_create_new_files_on_insert(started_cluster): result = instance.query(f"select count() from test_multiple_inserts") assert int(result) == 60 + instance.query("drop table test_multiple_inserts") def test_format_detection(started_cluster): @@ -1343,7 +1364,9 @@ def test_format_detection(started_cluster): instance = started_cluster.instances["dummy"] instance.query(f"create table arrow_table_s3 (x UInt64) engine=S3(s3_arrow)") - instance.query(f"insert into arrow_table_s3 select 1") + instance.query( + f"insert into arrow_table_s3 select 1 settings s3_truncate_on_insert=1" + ) result = instance.query(f"select * from s3(s3_arrow)") assert int(result) == 1 @@ -1358,7 +1381,9 @@ def test_format_detection(started_cluster): assert int(result) == 1 instance.query(f"create table parquet_table_s3 (x UInt64) engine=S3(s3_parquet2)") - instance.query(f"insert into parquet_table_s3 select 1") + instance.query( + f"insert into parquet_table_s3 select 1 settings s3_truncate_on_insert=1" + ) result = instance.query(f"select * from s3(s3_parquet2)") assert int(result) == 1 @@ -1371,64 +1396,67 @@ def test_format_detection(started_cluster): f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test.parquet')" ) assert int(result) == 1 + instance.query(f"drop table arrow_table_s3") + instance.query(f"drop table parquet_table_s3") def test_schema_inference_from_globs(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] instance.query( - f"insert into table function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test1.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select NULL" + f"insert into table function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test1.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select NULL" ) instance.query( - f"insert into table function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test2.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select 0" + f"insert into table function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test2.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select 0" ) url_filename = "test{1,2}.jsoncompacteachrow" result = instance.query( - f"desc url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{url_filename}') settings input_format_json_infer_incomplete_types_as_strings=0" + f"desc url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{url_filename}') settings input_format_json_infer_incomplete_types_as_strings=0" ) assert result.strip() == "c1\tNullable(Int64)" result = instance.query( - f"select * from url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{url_filename}') settings input_format_json_infer_incomplete_types_as_strings=0" + f"select * from url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{url_filename}') settings input_format_json_infer_incomplete_types_as_strings=0" ) assert sorted(result.split()) == ["0", "\\N"] result = instance.query( - f"desc s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test*.jsoncompacteachrow') settings input_format_json_infer_incomplete_types_as_strings=0" + f"desc s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test*.jsoncompacteachrow') settings input_format_json_infer_incomplete_types_as_strings=0" ) assert result.strip() == "c1\tNullable(Int64)" result = instance.query( - f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test*.jsoncompacteachrow') settings input_format_json_infer_incomplete_types_as_strings=0" + f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test*.jsoncompacteachrow') settings input_format_json_infer_incomplete_types_as_strings=0" ) assert sorted(result.split()) == ["0", "\\N"] instance.query( - f"insert into table function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test3.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select NULL" + f"insert into table function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test3.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select NULL" ) url_filename = "test{1,3}.jsoncompacteachrow" result = instance.query_and_get_error( - f"desc s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{url_filename}') settings schema_inference_use_cache_for_s3=0, input_format_json_infer_incomplete_types_as_strings=0" + f"desc s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{url_filename}') settings schema_inference_use_cache_for_s3=0, input_format_json_infer_incomplete_types_as_strings=0" ) assert "All attempts to extract table structure from files failed" in result result = instance.query_and_get_error( - f"desc url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{url_filename}') settings schema_inference_use_cache_for_url=0, input_format_json_infer_incomplete_types_as_strings=0" + f"desc url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{url_filename}') settings schema_inference_use_cache_for_url=0, input_format_json_infer_incomplete_types_as_strings=0" ) assert "All attempts to extract table structure from files failed" in result instance.query( - f"insert into table function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test0.jsoncompacteachrow', 'TSV', 'x String') select '[123;]'" + f"insert into table function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test0.jsoncompacteachrow', 'TSV', 'x String') select '[123;]'" ) result = instance.query_and_get_error( - f"desc s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test*.jsoncompacteachrow') settings schema_inference_use_cache_for_s3=0, input_format_json_infer_incomplete_types_as_strings=0" + f"desc s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test*.jsoncompacteachrow') settings schema_inference_use_cache_for_s3=0, input_format_json_infer_incomplete_types_as_strings=0" ) assert "CANNOT_EXTRACT_TABLE_STRUCTURE" in result @@ -1436,7 +1464,7 @@ def test_schema_inference_from_globs(started_cluster): url_filename = "test{0,1,2,3}.jsoncompacteachrow" result = instance.query_and_get_error( - f"desc url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{url_filename}') settings schema_inference_use_cache_for_url=0, input_format_json_infer_incomplete_types_as_strings=0" + f"desc url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/{url_filename}') settings schema_inference_use_cache_for_url=0, input_format_json_infer_incomplete_types_as_strings=0" ) assert "CANNOT_EXTRACT_TABLE_STRUCTURE" in result @@ -1496,9 +1524,12 @@ def test_signatures(started_cluster): ) assert "S3_ERROR" in error + instance.query(f"drop table test_signatures") + def test_select_columns(started_cluster): bucket = started_cluster.minio_bucket + id = uuid.uuid4() instance = started_cluster.instances["dummy"] name = "test_table2" structure = "id UInt32, value1 Int32, value2 Int32" @@ -1512,36 +1543,37 @@ def test_select_columns(started_cluster): instance.query( f"INSERT INTO {name} SELECT * FROM generateRandom('{structure}') LIMIT {limit} SETTINGS s3_truncate_on_insert=1" ) - instance.query(f"SELECT value2 FROM {name}") + instance.query(f"SELECT value2, '{id}' FROM {name}") instance.query("SYSTEM FLUSH LOGS") result1 = instance.query( - f"SELECT ProfileEvents['ReadBufferFromS3Bytes'] FROM system.query_log WHERE type='QueryFinish' and query LIKE 'SELECT value2 FROM {name}'" + f"SELECT ProfileEvents['ReadBufferFromS3Bytes'] FROM system.query_log WHERE type='QueryFinish' and query LIKE 'SELECT value2, ''{id}'' FROM {name}'" ) - instance.query(f"SELECT * FROM {name}") + instance.query(f"SELECT *, '{id}' FROM {name}") instance.query("SYSTEM FLUSH LOGS") result2 = instance.query( - f"SELECT ProfileEvents['ReadBufferFromS3Bytes'] FROM system.query_log WHERE type='QueryFinish' and query LIKE 'SELECT * FROM {name}'" + f"SELECT ProfileEvents['ReadBufferFromS3Bytes'] FROM system.query_log WHERE type='QueryFinish' and query LIKE 'SELECT *, ''{id}'' FROM {name}'" ) assert round(int(result2) / int(result1)) == 3 def test_insert_select_schema_inference(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_insert_select.native') select toUInt64(1) as x" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test_insert_select.native') select toUInt64(1) as x" ) result = instance.query( - f"desc s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_insert_select.native')" + f"desc s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test_insert_select.native')" ) assert result.strip() == "x\tUInt64" result = instance.query( - f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_insert_select.native')" + f"select * from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{id}/test_insert_select.native')" ) assert int(result) == 1 @@ -1551,7 +1583,7 @@ def test_parallel_reading_with_memory_limit(started_cluster): instance = started_cluster.instances["dummy"] instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_memory_limit.native') select * from numbers(1000000)" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_memory_limit.native') select * from numbers(1000000) SETTINGS s3_truncate_on_insert=1" ) result = instance.query_and_get_error( @@ -1572,7 +1604,7 @@ def test_wrong_format_usage(started_cluster): instance = started_cluster.instances["dummy"] instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_wrong_format.native') select * from numbers(10e6)" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_wrong_format.native') select * from numbers(10e6) SETTINGS s3_truncate_on_insert=1" ) # size(test_wrong_format.native) = 10e6*8+16(header) ~= 76MiB @@ -2095,11 +2127,11 @@ def test_read_subcolumns(started_cluster): instance = started_cluster.instances["dummy"] instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_subcolumns.tsv', auto, 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3)" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_subcolumns.tsv', auto, 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3) SETTINGS s3_truncate_on_insert=1" ) instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_subcolumns.jsonl', auto, 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3)" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_subcolumns.jsonl', auto, 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3) SETTINGS s3_truncate_on_insert=1" ) res = instance.query( @@ -2158,7 +2190,7 @@ def test_read_subcolumn_time(started_cluster): instance = started_cluster.instances["dummy"] instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_subcolumn_time.tsv', auto, 'a UInt32') select (42)" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_subcolumn_time.tsv', auto, 'a UInt32') select (42) SETTINGS s3_truncate_on_insert=1" ) res = instance.query( @@ -2169,29 +2201,30 @@ def test_read_subcolumn_time(started_cluster): def test_filtering_by_file_or_path(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["dummy"] instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_filter1.tsv', auto, 'x UInt64') select 1" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_filter1.tsv', auto, 'x UInt64') select 1 SETTINGS s3_truncate_on_insert=1" ) instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_filter2.tsv', auto, 'x UInt64') select 2" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_filter2.tsv', auto, 'x UInt64') select 2 SETTINGS s3_truncate_on_insert=1" ) instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_filter3.tsv', auto, 'x UInt64') select 3" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_filter3.tsv', auto, 'x UInt64') select 3 SETTINGS s3_truncate_on_insert=1" ) instance.query( - f"select count() from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_filter*.tsv') where _file = 'test_filter1.tsv'" + f"select count(), '{id}' from s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_filter*.tsv') where _file = 'test_filter1.tsv'" ) instance.query("SYSTEM FLUSH LOGS") result = instance.query( - f"SELECT ProfileEvents['EngineFileLikeReadFiles'] FROM system.query_log WHERE query like '%select%s3%test_filter%' AND type='QueryFinish'" + f"SELECT ProfileEvents['EngineFileLikeReadFiles'] FROM system.query_log WHERE query like '%{id}%' AND type='QueryFinish'" ) assert int(result) == 1 @@ -2204,54 +2237,56 @@ def test_filtering_by_file_or_path(started_cluster): def test_union_schema_inference_mode(started_cluster): + id = uuid.uuid4() bucket = started_cluster.minio_bucket instance = started_cluster.instances["s3_non_default"] + file_name_prefix = f"test_union_schema_inference_{id}_" instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference1.jsonl') select 1 as a" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}1.jsonl') select 1 as a SETTINGS s3_truncate_on_insert=1" ) instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference2.jsonl') select 2 as b" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}2.jsonl') select 2 as b SETTINGS s3_truncate_on_insert=1" ) instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference3.jsonl') select 2 as c" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}3.jsonl') select 2 as c SETTINGS s3_truncate_on_insert=1" ) instance.query( - f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference4.jsonl', TSV) select 'Error'" + f"insert into function s3('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}4.jsonl', TSV) select 'Error' SETTINGS s3_truncate_on_insert=1" ) for engine in ["s3", "url"]: instance.query("system drop schema cache for s3") result = instance.query( - f"desc {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference{{1,2,3}}.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"desc {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}{{1,2,3}}.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert result == "a\tNullable(Int64)\nb\tNullable(Int64)\nc\tNullable(Int64)\n" result = instance.query( - "select schema_inference_mode, splitByChar('/', source)[-1] as file, schema from system.schema_inference_cache where source like '%test_union_schema_inference%' order by file format TSV" + f"select schema_inference_mode, splitByChar('/', source)[-1] as file, schema from system.schema_inference_cache where source like '%{file_name_prefix}%' order by file format TSV" ) assert ( - result == "UNION\ttest_union_schema_inference1.jsonl\ta Nullable(Int64)\n" - "UNION\ttest_union_schema_inference2.jsonl\tb Nullable(Int64)\n" - "UNION\ttest_union_schema_inference3.jsonl\tc Nullable(Int64)\n" + result == f"UNION\t{file_name_prefix}1.jsonl\ta Nullable(Int64)\n" + f"UNION\t{file_name_prefix}2.jsonl\tb Nullable(Int64)\n" + f"UNION\t{file_name_prefix}3.jsonl\tc Nullable(Int64)\n" ) result = instance.query( - f"select * from {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference{{1,2,3}}.jsonl') order by tuple(*) settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"select * from {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}{{1,2,3}}.jsonl') order by tuple(*) settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert result == "1\t\\N\t\\N\n" "\\N\t2\t\\N\n" "\\N\t\\N\t2\n" instance.query(f"system drop schema cache for {engine}") result = instance.query( - f"desc {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference2.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"desc {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}2.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert result == "b\tNullable(Int64)\n" result = instance.query( - f"desc {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference{{1,2,3}}.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"desc {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}{{1,2,3}}.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert ( result == "a\tNullable(Int64)\n" @@ -2260,7 +2295,7 @@ def test_union_schema_inference_mode(started_cluster): ) error = instance.query_and_get_error( - f"desc {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_union_schema_inference{{1,2,3,4}}.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"desc {engine}('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{file_name_prefix}{{1,2,3,4}}.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert "CANNOT_EXTRACT_TABLE_STRUCTURE" in error diff --git a/tests/integration/test_storage_s3_queue/test.py b/tests/integration/test_storage_s3_queue/test.py index a59f3a2ce40..bf3c28c5429 100644 --- a/tests/integration/test_storage_s3_queue/test.py +++ b/tests/integration/test_storage_s3_queue/test.py @@ -12,6 +12,7 @@ import json AVAILABLE_MODES = ["unordered", "ordered"] DEFAULT_AUTH = ["'minio'", "'minio123'"] NO_AUTH = ["NOSIGN"] +AZURE_CONTAINER_NAME = "cont" def prepare_public_s3_bucket(started_cluster): @@ -84,6 +85,7 @@ def started_cluster(): "instance", user_configs=["configs/users.xml"], with_minio=True, + with_azurite=True, with_zookeeper=True, main_configs=[ "configs/zookeeper.xml", @@ -126,6 +128,11 @@ def started_cluster(): cluster.start() logging.info("Cluster started") + container_client = cluster.blob_service_client.get_container_client( + AZURE_CONTAINER_NAME + ) + container_client.create_container() + yield cluster finally: cluster.shutdown() @@ -145,6 +152,7 @@ def generate_random_files( started_cluster, files_path, count, + storage="s3", column_num=3, row_num=10, start_ind=0, @@ -166,7 +174,10 @@ def generate_random_files( values_csv = ( "\n".join((",".join(map(str, row)) for row in rand_values)) + "\n" ).encode() - put_s3_file_content(started_cluster, filename, values_csv, bucket) + if storage == "s3": + put_s3_file_content(started_cluster, filename, values_csv, bucket) + else: + put_azure_file_content(started_cluster, filename, values_csv, bucket) return total_values @@ -176,12 +187,21 @@ def put_s3_file_content(started_cluster, filename, data, bucket=None): started_cluster.minio_client.put_object(bucket, filename, buf, len(data)) +def put_azure_file_content(started_cluster, filename, data, bucket=None): + client = started_cluster.blob_service_client.get_blob_client( + AZURE_CONTAINER_NAME, filename + ) + buf = io.BytesIO(data) + client.upload_blob(buf, "BlockBlob", len(data)) + + def create_table( started_cluster, node, table_name, mode, files_path, + engine_name="S3Queue", format="column1 UInt32, column2 UInt32, column3 UInt32", additional_settings={}, file_format="CSV", @@ -200,11 +220,17 @@ def create_table( } settings.update(additional_settings) - url = f"http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{files_path}/" + engine_def = None + if engine_name == "S3Queue": + url = f"http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/{files_path}/" + engine_def = f"{engine_name}('{url}', {auth_params}, {file_format})" + else: + engine_def = f"{engine_name}('{started_cluster.env_variables['AZURITE_CONNECTION_STRING']}', 'cont', '{files_path}/', 'CSV')" + node.query(f"DROP TABLE IF EXISTS {table_name}") create_query = f""" CREATE TABLE {table_name} ({format}) - ENGINE = S3Queue('{url}', {auth_params}, {file_format}) + ENGINE = {engine_def} SETTINGS {",".join((k+"="+repr(v) for k, v in settings.items()))} """ @@ -235,17 +261,22 @@ def create_mv( ) -@pytest.mark.parametrize("mode", AVAILABLE_MODES) -def test_delete_after_processing(started_cluster, mode): +@pytest.mark.parametrize("mode", ["unordered", "ordered"]) +@pytest.mark.parametrize("engine_name", ["S3Queue", "AzureQueue"]) +def test_delete_after_processing(started_cluster, mode, engine_name): node = started_cluster.instances["instance"] - table_name = f"test.delete_after_processing_{mode}" + table_name = f"test.delete_after_processing_{mode}_{engine_name}" dst_table_name = f"{table_name}_dst" files_path = f"{table_name}_data" files_num = 5 row_num = 10 + if engine_name == "S3Queue": + storage = "s3" + else: + storage = "azure" total_values = generate_random_files( - started_cluster, files_path, files_num, row_num=row_num + started_cluster, files_path, files_num, row_num=row_num, storage=storage ) create_table( started_cluster, @@ -254,6 +285,7 @@ def test_delete_after_processing(started_cluster, mode): mode, files_path, additional_settings={"after_processing": "delete"}, + engine_name=engine_name, ) create_mv(node, table_name, dst_table_name) @@ -274,15 +306,24 @@ def test_delete_after_processing(started_cluster, mode): ).splitlines() ] == sorted(total_values, key=lambda x: (x[0], x[1], x[2])) - minio = started_cluster.minio_client - objects = list(minio.list_objects(started_cluster.minio_bucket, recursive=True)) - assert len(objects) == 0 + if engine_name == "S3Queue": + minio = started_cluster.minio_client + objects = list(minio.list_objects(started_cluster.minio_bucket, recursive=True)) + assert len(objects) == 0 + else: + client = started_cluster.blob_service_client.get_container_client( + AZURE_CONTAINER_NAME + ) + objects_iterator = client.list_blobs(files_path) + for objects in objects_iterator: + assert False -@pytest.mark.parametrize("mode", AVAILABLE_MODES) -def test_failed_retry(started_cluster, mode): +@pytest.mark.parametrize("mode", ["unordered", "ordered"]) +@pytest.mark.parametrize("engine_name", ["S3Queue", "AzureQueue"]) +def test_failed_retry(started_cluster, mode, engine_name): node = started_cluster.instances["instance"] - table_name = f"test.failed_retry_{mode}" + table_name = f"test.failed_retry_{mode}_{engine_name}" dst_table_name = f"{table_name}_dst" files_path = f"{table_name}_data" file_path = f"{files_path}/trash_test.csv" @@ -295,7 +336,10 @@ def test_failed_retry(started_cluster, mode): values_csv = ( "\n".join((",".join(map(str, row)) for row in values)) + "\n" ).encode() - put_s3_file_content(started_cluster, file_path, values_csv) + if engine_name == "S3Queue": + put_s3_file_content(started_cluster, file_path, values_csv) + else: + put_azure_file_content(started_cluster, file_path, values_csv) create_table( started_cluster, @@ -307,6 +351,7 @@ def test_failed_retry(started_cluster, mode): "s3queue_loading_retries": retries_num, "keeper_path": keeper_path, }, + engine_name=engine_name, ) create_mv(node, table_name, dst_table_name) @@ -792,12 +837,12 @@ def test_max_set_age(started_cluster): files_path, additional_settings={ "keeper_path": keeper_path, - "s3queue_tracked_file_ttl_sec": max_age, - "s3queue_cleanup_interval_min_ms": max_age / 3, - "s3queue_cleanup_interval_max_ms": max_age / 3, - "s3queue_loading_retries": 0, - "s3queue_processing_threads_num": 1, - "s3queue_loading_retries": 0, + "tracked_file_ttl_sec": max_age, + "cleanup_interval_min_ms": max_age / 3, + "cleanup_interval_max_ms": max_age / 3, + "loading_retries": 0, + "processing_threads_num": 1, + "loading_retries": 0, }, ) create_mv(node, table_name, dst_table_name) @@ -846,7 +891,7 @@ def test_max_set_age(started_cluster): failed_count = int( node.query( - "SELECT value FROM system.events WHERE name = 'S3QueueFailedFiles' SETTINGS system_events_show_zero_values=1" + "SELECT value FROM system.events WHERE name = 'ObjectStorageQueueFailedFiles' SETTINGS system_events_show_zero_values=1" ) ) @@ -861,7 +906,7 @@ def test_max_set_age(started_cluster): for _ in range(30): if failed_count + 1 == int( node.query( - "SELECT value FROM system.events WHERE name = 'S3QueueFailedFiles' SETTINGS system_events_show_zero_values=1" + "SELECT value FROM system.events WHERE name = 'ObjectStorageQueueFailedFiles' SETTINGS system_events_show_zero_values=1" ) ): break @@ -869,7 +914,7 @@ def test_max_set_age(started_cluster): assert failed_count + 1 == int( node.query( - "SELECT value FROM system.events WHERE name = 'S3QueueFailedFiles' SETTINGS system_events_show_zero_values=1" + "SELECT value FROM system.events WHERE name = 'ObjectStorageQueueFailedFiles' SETTINGS system_events_show_zero_values=1" ) ) @@ -891,7 +936,9 @@ def test_max_set_age(started_cluster): time.sleep(max_age + 1) assert failed_count + 2 == int( - node.query("SELECT value FROM system.events WHERE name = 'S3QueueFailedFiles'") + node.query( + "SELECT value FROM system.events WHERE name = 'ObjectStorageQueueFailedFiles'" + ) ) node.query("SYSTEM FLUSH LOGS") @@ -1435,7 +1482,7 @@ def test_settings_check(started_cluster): ) assert ( - "Existing table metadata in ZooKeeper differs in s3queue_buckets setting. Stored in ZooKeeper: 2, local: 3" + "Existing table metadata in ZooKeeper differs in buckets setting. Stored in ZooKeeper: 2, local: 3" in create_table( started_cluster, node_2, @@ -1733,11 +1780,12 @@ def test_commit_on_limit(started_cluster): if "test_999999.csv" in get_processed_files(): break time.sleep(1) + assert "test_999999.csv" in get_processed_files() assert 1 == int( node.query( - "SELECT value FROM system.events WHERE name = 'S3QueueFailedFiles' SETTINGS system_events_show_zero_values=1" + "SELECT value FROM system.events WHERE name = 'ObjectStorageQueueFailedFiles' SETTINGS system_events_show_zero_values=1" ) ) diff --git a/tests/integration/test_system_detached_tables/__init__.py b/tests/integration/test_system_detached_tables/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_system_detached_tables/configs/remote_servers.xml b/tests/integration/test_system_detached_tables/configs/remote_servers.xml new file mode 100644 index 00000000000..8b579724282 --- /dev/null +++ b/tests/integration/test_system_detached_tables/configs/remote_servers.xml @@ -0,0 +1,13 @@ + + + + + true + + replica1 + 9000 + + + + + \ No newline at end of file diff --git a/tests/integration/test_system_detached_tables/test.py b/tests/integration/test_system_detached_tables/test.py new file mode 100644 index 00000000000..2eb870efcbc --- /dev/null +++ b/tests/integration/test_system_detached_tables/test.py @@ -0,0 +1,112 @@ +import pytest +from helpers.cluster import ClickHouseCluster + +cluster = ClickHouseCluster(__file__) +node = cluster.add_instance( + "replica1", + with_zookeeper=True, + main_configs=["configs/remote_servers.xml"], + macros={"replica": "replica1"}, + stay_alive=True, +) + + +@pytest.fixture(scope="module", autouse=True) +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +@pytest.mark.parametrize( + "db_name,db_engine,table_engine,table_engine_params", + [ + pytest.param( + "test_db_atomic", + "Atomic", + "MergeTree", + "ORDER BY n", + id="Atomic db with MergeTree table", + ), + pytest.param( + "test_db_lazy", + "Lazy(60)", + "Log", + "", + id="Lazy db with Log table", + ), + pytest.param( + "test_db_repl", + "Replicated('/clickhouse/tables/test_table','shard1', 'replica1')", + "ReplicatedMergeTree", + "ORDER BY n", + id="Replicated db with ReplicatedMergeTree table", + ), + ], +) +def test_system_detached_tables( + start_cluster, db_name, db_engine, table_engine, table_engine_params +): + node.query(f"CREATE DATABASE IF NOT EXISTS {db_name} ENGINE={db_engine};") + + node.query( + f"CREATE TABLE {db_name}.test_table (n Int64) ENGINE={table_engine} {table_engine_params};" + ) + node.query( + f"CREATE TABLE {db_name}.test_table_perm (n Int64) ENGINE={table_engine} {table_engine_params};" + ) + + test_table_uuid = node.query( + "SELECT uuid FROM system.tables WHERE table='test_table'" + ).rstrip("\n") + test_table_metadata_path = node.query( + "SELECT metadata_path FROM system.tables WHERE table='test_table'" + ).rstrip("\n") + + test_table_perm_uuid = node.query( + "SELECT uuid FROM system.tables WHERE table='test_table_perm'" + ).rstrip("\n") + test_table_perm_metadata_path = node.query( + "SELECT metadata_path FROM system.tables WHERE table='test_table_perm'" + ).rstrip("\n") + + assert "" == node.query( + f"SELECT * FROM system.detached_tables WHERE database='{db_name}'" + ) + + node.query( + f"SET database_replicated_always_detach_permanently=1; DETACH TABLE {db_name}.test_table" + ) + node.query(f"DETACH TABLE {db_name}.test_table_perm PERMANENTLY") + + querry = f"SELECT database, table, is_permanently, uuid, metadata_path FROM system.detached_tables WHERE database='{db_name}' FORMAT Values" + result = node.query(querry) + + if db_engine.startswith("Repl"): + expected_before_restart = f"('{db_name}','test_table',1,'{test_table_uuid}','{test_table_metadata_path}'),('{db_name}','test_table_perm',1,'{test_table_perm_uuid}','{test_table_perm_metadata_path}')" + else: + expected_before_restart = f"('{db_name}','test_table',0,'{test_table_uuid}','{test_table_metadata_path}'),('{db_name}','test_table_perm',1,'{test_table_perm_uuid}','{test_table_perm_metadata_path}')" + + assert result == expected_before_restart + + if db_engine.startswith("Lazy"): + return + + node.restart_clickhouse() + + if db_engine.startswith("Repl"): + expected_after_restart = expected_before_restart + else: + expected_after_restart = f"('{db_name}','test_table_perm',1,'{test_table_perm_uuid}','{test_table_perm_metadata_path}')" + + result = node.query(querry) + assert result == expected_after_restart + + node.restart_clickhouse() + + result = node.query(querry) + assert result == expected_after_restart + + node.query(f"DROP DATABASE {db_name}") diff --git a/tests/integration/test_table_db_num_limit/test.py b/tests/integration/test_table_db_num_limit/test.py index aa8030b077c..b7f9d7c0b96 100644 --- a/tests/integration/test_table_db_num_limit/test.py +++ b/tests/integration/test_table_db_num_limit/test.py @@ -4,9 +4,7 @@ from helpers.cluster import ClickHouseCluster cluster = ClickHouseCluster(__file__) -node1 = cluster.add_instance( - "node1", main_configs=["config/config.xml"], with_zookeeper=True -) +node = cluster.add_instance("node", main_configs=["config/config.xml"]) @pytest.fixture(scope="module") @@ -21,23 +19,28 @@ def started_cluster(): def test_table_db_limit(started_cluster): - for i in range(10): - node1.query("create database db{}".format(i)) + # By the way, default database already exists. + for i in range(9): + node.query("create database db{}".format(i)) with pytest.raises(QueryRuntimeException) as exp_info: - node1.query("create database db_exp".format(i)) + node.query("create database db_exp".format(i)) assert "TOO_MANY_DATABASES" in str(exp_info) for i in range(10): - node1.query("create table t{} (a Int32) Engine = Log".format(i)) + node.query("create table t{} (a Int32) Engine = Log".format(i)) + + # This checks that system tables are not accounted in the number of tables. + node.query("system flush logs") - node1.query("system flush logs") for i in range(10): - node1.query("drop table t{}".format(i)) + node.query("drop table t{}".format(i)) + for i in range(10): - node1.query("create table t{} (a Int32) Engine = Log".format(i)) + node.query("create table t{} (a Int32) Engine = Log".format(i)) with pytest.raises(QueryRuntimeException) as exp_info: - node1.query("create table default.tx (a Int32) Engine = Log") + node.query("create table default.tx (a Int32) Engine = Log") + assert "TOO_MANY_TABLES" in str(exp_info) diff --git a/tests/integration/test_tcp_handler_http_responses/test_case.py b/tests/integration/test_tcp_handler_http_responses/test_case.py index 2fc53674ca4..98f4b74223e 100644 --- a/tests/integration/test_tcp_handler_http_responses/test_case.py +++ b/tests/integration/test_tcp_handler_http_responses/test_case.py @@ -1,4 +1,5 @@ """Test HTTP responses given by the TCP Handler.""" + from pathlib import Path import pytest from helpers.cluster import ClickHouseCluster diff --git a/tests/integration/test_tcp_handler_interserver_listen_host/test_case.py b/tests/integration/test_tcp_handler_interserver_listen_host/test_case.py index 62581996f3b..b20ab48795b 100644 --- a/tests/integration/test_tcp_handler_interserver_listen_host/test_case.py +++ b/tests/integration/test_tcp_handler_interserver_listen_host/test_case.py @@ -1,4 +1,5 @@ """Test Interserver responses on configured IP.""" + from pathlib import Path import pytest from helpers.cluster import ClickHouseCluster diff --git a/tests/integration/test_user_valid_until/test.py b/tests/integration/test_user_valid_until/test.py index d6d5bf8b18e..39ca5997067 100644 --- a/tests/integration/test_user_valid_until/test.py +++ b/tests/integration/test_user_valid_until/test.py @@ -5,7 +5,7 @@ from time import sleep from helpers.cluster import ClickHouseCluster cluster = ClickHouseCluster(__file__) -node = cluster.add_instance("node") +node = cluster.add_instance("node", stay_alive=True) @pytest.fixture(scope="module") @@ -84,3 +84,22 @@ def test_details(started_cluster): node.query("SHOW CREATE USER user_details_time_only") == f"CREATE USER user_details_time_only VALID UNTIL \\'{until_year}-01-01 22:03:40\\'\n" ) + + +def test_restart(started_cluster): + node.query("CREATE USER user_restart VALID UNTIL '06/11/2010 08:03:20 Z+3'") + + assert ( + node.query("SHOW CREATE USER user_restart") + == "CREATE USER user_restart VALID UNTIL \\'2010-11-06 05:03:20\\'\n" + ) + + node.restart_clickhouse() + + assert ( + node.query("SHOW CREATE USER user_restart") + == "CREATE USER user_restart VALID UNTIL \\'2010-11-06 05:03:20\\'\n" + ) + + error = "Authentication failed" + assert error in node.query_and_get_error("SELECT 1", user="user_restart") diff --git a/tests/integration/test_zookeeper_config_load_balancing/configs/zookeeper_load_balancing2.xml b/tests/integration/test_zookeeper_config_load_balancing/configs/zookeeper_load_balancing2.xml new file mode 100644 index 00000000000..fd416cad505 --- /dev/null +++ b/tests/integration/test_zookeeper_config_load_balancing/configs/zookeeper_load_balancing2.xml @@ -0,0 +1,35 @@ + + + + random + + 1 + + + 0 + 1 + + + + zoo1 + 2181 + az1 + + + zoo2 + 2181 + az2 + + + zoo3 + 2181 + az3 + + 3000 + + + + 0 + az2 + + diff --git a/tests/integration/test_zookeeper_config_load_balancing/test.py b/tests/integration/test_zookeeper_config_load_balancing/test.py index f17e0c3f03f..9cdf7db2b08 100644 --- a/tests/integration/test_zookeeper_config_load_balancing/test.py +++ b/tests/integration/test_zookeeper_config_load_balancing/test.py @@ -1,6 +1,8 @@ +import time import pytest from helpers.cluster import ClickHouseCluster from helpers.network import PartitionManager +from helpers.test_tools import assert_eq_with_retry cluster = ClickHouseCluster( __file__, zookeeper_config_path="configs/zookeeper_load_balancing.xml" @@ -17,6 +19,10 @@ node3 = cluster.add_instance( "nod3", with_zookeeper=True, main_configs=["configs/zookeeper_load_balancing.xml"] ) +node4 = cluster.add_instance( + "nod4", with_zookeeper=True, main_configs=["configs/zookeeper_load_balancing2.xml"] +) + def change_balancing(old, new, reload=True): line = "{}<" @@ -405,113 +411,57 @@ def test_hostname_levenshtein_distance(started_cluster): def test_round_robin(started_cluster): pm = PartitionManager() try: - pm._add_rule( - { - "source": node1.ip_address, - "destination": cluster.get_instance_ip("zoo1"), - "action": "REJECT --reject-with tcp-reset", - } - ) - pm._add_rule( - { - "source": node2.ip_address, - "destination": cluster.get_instance_ip("zoo1"), - "action": "REJECT --reject-with tcp-reset", - } - ) - pm._add_rule( - { - "source": node3.ip_address, - "destination": cluster.get_instance_ip("zoo1"), - "action": "REJECT --reject-with tcp-reset", - } - ) change_balancing("random", "round_robin") - - print( - str( - node1.exec_in_container( - [ - "bash", - "-c", - "lsof -a -i4 -i6 -itcp -w | grep ':2181' | grep ESTABLISHED", - ], - privileged=True, - user="root", - ) + for node in [node1, node2, node3]: + idx = int( + node.query("select index from system.zookeeper_connection").strip() ) - ) - assert ( - "1" - == str( - node1.exec_in_container( - [ - "bash", - "-c", - "lsof -a -i4 -i6 -itcp -w | grep 'testzookeeperconfigloadbalancing_zoo2_1.*testzookeeperconfigloadbalancing_default:2181' | grep ESTABLISHED | wc -l", - ], - privileged=True, - user="root", - ) - ).strip() - ) + new_idx = (idx + 1) % 3 - print( - str( - node2.exec_in_container( - [ - "bash", - "-c", - "lsof -a -i4 -i6 -itcp -w | grep ':2181' | grep ESTABLISHED", - ], - privileged=True, - user="root", - ) + pm._add_rule( + { + "source": node.ip_address, + "destination": cluster.get_instance_ip("zoo" + str(idx + 1)), + "action": "REJECT --reject-with tcp-reset", + } ) - ) - assert ( - "1" - == str( - node2.exec_in_container( - [ - "bash", - "-c", - "lsof -a -i4 -i6 -itcp -w | grep 'testzookeeperconfigloadbalancing_zoo2_1.*testzookeeperconfigloadbalancing_default:2181' | grep ESTABLISHED | wc -l", - ], - privileged=True, - user="root", - ) - ).strip() - ) - print( - str( - node3.exec_in_container( - [ - "bash", - "-c", - "lsof -a -i4 -i6 -itcp -w | grep ':2181' | grep ESTABLISHED", - ], - privileged=True, - user="root", - ) + assert_eq_with_retry( + node, + "select index from system.zookeeper_connection", + str(new_idx) + "\n", ) - ) - assert ( - "1" - == str( - node3.exec_in_container( - [ - "bash", - "-c", - "lsof -a -i4 -i6 -itcp -w | grep 'testzookeeperconfigloadbalancing_zoo2_1.*testzookeeperconfigloadbalancing_default:2181' | grep ESTABLISHED | wc -l", - ], - privileged=True, - user="root", - ) - ).strip() - ) - + pm.heal_all() finally: pm.heal_all() change_balancing("round_robin", "random", reload=False) + + +def test_az(started_cluster): + pm = PartitionManager() + try: + # make sure it disconnects from the optimal node + pm._add_rule( + { + "source": node4.ip_address, + "destination": cluster.get_instance_ip("zoo2"), + "action": "REJECT --reject-with tcp-reset", + } + ) + + node4.query_with_retry("select * from system.zookeeper where path='/'") + assert "az2\n" != node4.query( + "select availability_zone from system.zookeeper_connection" + ) + + # fallback_session_lifetime.max is 1 second, but it shouldn't drop current session until the node becomes available + + time.sleep(5) # this is fine + assert 5 <= int(node4.query("select zookeeperSessionUptime()").strip()) + + pm.heal_all() + assert_eq_with_retry( + node4, "select availability_zone from system.zookeeper_connection", "az2\n" + ) + finally: + pm.heal_all() diff --git a/tests/integration/test_zookeeper_fallback_session/test.py b/tests/integration/test_zookeeper_fallback_session/test.py index 9afabfa3da3..932bbe482d2 100644 --- a/tests/integration/test_zookeeper_fallback_session/test.py +++ b/tests/integration/test_zookeeper_fallback_session/test.py @@ -84,10 +84,28 @@ def test_fallback_session(started_cluster: ClickHouseCluster): ) # at this point network partitioning has been reverted. - # the nodes should switch to zoo1 automatically because of `in_order` load-balancing. + # the nodes should switch to zoo1 because of `in_order` load-balancing. # otherwise they would connect to a random replica + + # but there's no reason to reconnect because current session works + # and there's no "optimal" node with `in_order` load-balancing + # so we need to break the current session + for node in [node1, node2, node3]: - assert_uses_zk_node(node, "zoo1") + assert_uses_zk_node(node, "zoo3") + + with PartitionManager() as pm: + for node in started_cluster.instances.values(): + pm._add_rule( + { + "source": node.ip_address, + "destination": cluster.get_instance_ip("zoo3"), + "action": "REJECT --reject-with tcp-reset", + } + ) + + for node in [node1, node2, node3]: + assert_uses_zk_node(node, "zoo1") node1.query_with_retry("INSERT INTO simple VALUES ({0}, {0})".format(2)) for node in [node2, node3]: diff --git a/tests/performance/decimal_aggregates.xml b/tests/performance/decimal_aggregates.xml index ec88be0124f..724d0c5d0e6 100644 --- a/tests/performance/decimal_aggregates.xml +++ b/tests/performance/decimal_aggregates.xml @@ -4,8 +4,13 @@ CREATE TABLE t (x UInt64, d32 Decimal32(3), d64 Decimal64(4), d128 Decimal128(5)) ENGINE = Memory - - INSERT INTO t SELECT number AS x, x % 1000000 AS d32, x AS d64, x d128 FROM numbers_mt(500000000) SETTINGS max_threads = 8 + + INSERT INTO t SELECT number AS x, x % 1000000 AS d32, x AS d64, x d128 FROM numbers_mt(100000000) SETTINGS max_threads = 2 + INSERT INTO t SELECT number AS x, x % 1000000 AS d32, x AS d64, x d128 FROM numbers_mt(100000000, 100000000) SETTINGS max_threads = 2 + INSERT INTO t SELECT number AS x, x % 1000000 AS d32, x AS d64, x d128 FROM numbers_mt(200000000, 100000000) SETTINGS max_threads = 2 + INSERT INTO t SELECT number AS x, x % 1000000 AS d32, x AS d64, x d128 FROM numbers_mt(300000000, 100000000) SETTINGS max_threads = 2 + INSERT INTO t SELECT number AS x, x % 1000000 AS d32, x AS d64, x d128 FROM numbers_mt(400000000, 100000000) SETTINGS max_threads = 2 + DROP TABLE IF EXISTS t SELECT min(d32), max(d32), argMin(x, d32), argMax(x, d32) FROM t diff --git a/tests/performance/final_big_column.xml b/tests/performance/final_big_column.xml index 1fd586d2d90..5225b3d7ad4 100644 --- a/tests/performance/final_big_column.xml +++ b/tests/performance/final_big_column.xml @@ -10,8 +10,8 @@ PARTITION BY toYYYYMM(d) ORDER BY key - INSERT INTO optimized_select_final SELECT toDate('2000-01-01'), 2*number, randomPrintableASCII(1000) FROM numbers(5000000) - INSERT INTO optimized_select_final SELECT toDate('2020-01-01'), 2*number+1, randomPrintableASCII(1000) FROM numbers(5000000) + INSERT INTO optimized_select_final SELECT toDate('2000-01-01'), 2*number, randomPrintableASCII(1000) FROM numbers(2500000) + INSERT INTO optimized_select_final SELECT toDate('2020-01-01'), 2*number+1, randomPrintableASCII(1000) FROM numbers(2500000) SELECT * FROM optimized_select_final FINAL FORMAT Null SETTINGS max_threads = 8 SELECT * FROM optimized_select_final FINAL WHERE key % 10 = 0 FORMAT Null diff --git a/tests/performance/function_tokens.xml b/tests/performance/function_tokens.xml index 1ff56323d62..bc2bc71a933 100644 --- a/tests/performance/function_tokens.xml +++ b/tests/performance/function_tokens.xml @@ -1,5 +1,5 @@ with 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' as s select splitByChar(' ', materialize(s)) as w from numbers(1000000) - with 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' as s select splitByRegexp(' ', materialize(s)) as w from numbers(1000000) - with 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' as s select splitByRegexp('\s+', materialize(s)) as w from numbers(100000) + with 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' as s select splitByRegexp(' ', materialize(s)) as w from numbers(200000) + with 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' as s select splitByRegexp('\s+', materialize(s)) as w from numbers(20000) diff --git a/tests/performance/generate_table_function.xml b/tests/performance/generate_table_function.xml index c219d73b6cf..d56c585188d 100644 --- a/tests/performance/generate_table_function.xml +++ b/tests/performance/generate_table_function.xml @@ -7,7 +7,7 @@ SELECT sum(NOT ignore(*)) FROM (SELECT * FROM generateRandom('d Date, dt DateTime, dtm DateTime(\'Asia/Istanbul\')', 0, 10, 10) LIMIT 1000000000); SELECT sum(NOT ignore(*)) FROM (SELECT * FROM generateRandom('dt64 DateTime64, dts64 DateTime64(6), dtms64 DateTime64(6 ,\'Asia/Istanbul\')', 0, 10, 10) LIMIT 100000000); SELECT sum(NOT ignore(*)) FROM (SELECT * FROM generateRandom('f32 Float32, f64 Float64', 0, 10, 10) LIMIT 1000000000); - SELECT sum(NOT ignore(*)) FROM (SELECT * FROM generateRandom('d32 Decimal32(4), d64 Decimal64(8), d128 Decimal128(16)', 0, 10, 10) LIMIT 1000000000); + SELECT sum(NOT ignore(*)) FROM (SELECT * FROM generateRandom('d32 Decimal32(4), d64 Decimal64(8), d128 Decimal128(16)', 0, 10, 10) LIMIT 100000000); SELECT sum(NOT ignore(*)) FROM (SELECT * FROM generateRandom('i Tuple(Int32, Int64)', 0, 10, 10) LIMIT 1000000000); SELECT sum(NOT ignore(*)) FROM (SELECT * FROM generateRandom('i Array(Int8)', 0, 10, 10) LIMIT 100000000); SELECT sum(NOT ignore(*)) FROM (SELECT * FROM generateRandom('i Array(Nullable(Int32))', 0, 10, 10) LIMIT 100000000); diff --git a/tests/performance/polymorphic_parts_s.xml b/tests/performance/polymorphic_parts_s.xml index b4dd87a7ae3..5fe1ffffe1d 100644 --- a/tests/performance/polymorphic_parts_s.xml +++ b/tests/performance/polymorphic_parts_s.xml @@ -24,10 +24,10 @@ 1 - - INSERT INTO hits_wide(UserID) SELECT rand() FROM numbers(100) - INSERT INTO hits_compact(UserID) SELECT rand() FROM numbers(1000) - INSERT INTO hits_buffer(UserID) SELECT rand() FROM numbers(100) + + INSERT INTO hits_wide(UserID) SELECT rand() FROM numbers(50) + INSERT INTO hits_compact(UserID) SELECT rand() FROM numbers(500) + INSERT INTO hits_buffer(UserID) SELECT rand() FROM numbers(50) DROP TABLE IF EXISTS hits_wide DROP TABLE IF EXISTS hits_compact diff --git a/tests/performance/replaceRegexp_fallback.xml b/tests/performance/replaceRegexp_fallback.xml new file mode 100644 index 00000000000..509257efeb5 --- /dev/null +++ b/tests/performance/replaceRegexp_fallback.xml @@ -0,0 +1,12 @@ + + + + + WITH 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' AS s SELECT replaceRegexpAll(materialize(s), ' ', '\n') AS w FROM numbers_mt(5000000) FORMAT Null + WITH 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' AS s SELECT replaceRegexpOne(materialize(s), ' ', '\n') AS w FROM numbers_mt(5000000) FORMAT Null + + + + WITH 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' AS s SELECT replaceRegexpAll(materialize(s), '\s+', '\\0\n') AS w FROM numbers_mt(500000) FORMAT Null + WITH 'Many years later as he faced the firing squad, Colonel Aureliano Buendia was to remember that distant afternoon when his father took him to discover ice.' AS s SELECT replaceRegexpOne(materialize(s), '\s+', '\\0\n') AS w FROM numbers_mt(500000) FORMAT Null + diff --git a/tests/performance/scripts/report.py b/tests/performance/scripts/report.py index c2bc773bd54..e45d709ca2c 100755 --- a/tests/performance/scripts/report.py +++ b/tests/performance/scripts/report.py @@ -555,7 +555,7 @@ if args.report == "main": "Total client time for measured query runs, s", # 2 "Queries", # 3 "Longest query, total for measured runs, s", # 4 - "Wall clock time per query, s", # 5 + "Average query wall clock time, s", # 5 "Shortest query, total for measured runs, s", # 6 "", # Runs #7 ] diff --git a/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql b/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql index 0ca7df8ecd3..07c42d6d039 100644 --- a/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql +++ b/tests/queries/0_stateless/00002_log_and_exception_messages_formatting.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-fasttest +-- Tags: no-parallel, no-fasttest, no-ubsan, no-batch -- no-parallel because we want to run this test when most of the other tests already passed -- If this test fails, see the "Top patterns of log messages" diagnostics in the end of run.log diff --git a/tests/queries/0_stateless/00027_argMinMax.sql b/tests/queries/0_stateless/00027_argMinMax.sql index dbf7c9176d2..57f815add27 100644 --- a/tests/queries/0_stateless/00027_argMinMax.sql +++ b/tests/queries/0_stateless/00027_argMinMax.sql @@ -13,4 +13,4 @@ FROM SELECT arrayJoin([[10, 4, 3], [7, 5, 6], [8, 8, 2]]) AS num, arrayJoin([[1, 2, 4]]) AS id -) +); diff --git a/tests/queries/0_stateless/00133_long_shard_memory_tracker_and_exception_safety.sh b/tests/queries/0_stateless/00133_long_shard_memory_tracker_and_exception_safety.sh index a42fd58190a..d57efaa1f0e 100755 --- a/tests/queries/0_stateless/00133_long_shard_memory_tracker_and_exception_safety.sh +++ b/tests/queries/0_stateless/00133_long_shard_memory_tracker_and_exception_safety.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, shard, no-parallel +# Tags: long, shard CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/00186_very_long_arrays.sh b/tests/queries/0_stateless/00186_very_long_arrays.sh index 739b17ccc99..086303279fa 100755 --- a/tests/queries/0_stateless/00186_very_long_arrays.sh +++ b/tests/queries/0_stateless/00186_very_long_arrays.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long +# Tags: long, no-tsan CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/00307_format_xml.sql b/tests/queries/0_stateless/00307_format_xml.sql index 7fdca83b69f..29c733bb186 100644 --- a/tests/queries/0_stateless/00307_format_xml.sql +++ b/tests/queries/0_stateless/00307_format_xml.sql @@ -1,2 +1,2 @@ SET output_format_write_statistics = 0; -SELECT 'Hello & world' AS s, 'Hello\n', toDateTime('2001-02-03 04:05:06') AS time, arrayMap(x -> toString(x), range(10)) AS arr, (s, time) AS tpl SETTINGS extremes = 1 FORMAT XML; +SELECT 'Hello & world' AS s, 'Hello\n', toDateTime('2001-02-03 04:05:06') AS time, arrayMap(x -> toString(x), range(10)) AS arr, (s, time) AS tpl SETTINGS extremes = 1, enable_named_columns_in_function_tuple = 0 FORMAT XML; diff --git a/tests/queries/0_stateless/00309_formats.sql b/tests/queries/0_stateless/00309_formats.sql index 87a1ea454d0..b0939c00a10 100644 --- a/tests/queries/0_stateless/00309_formats.sql +++ b/tests/queries/0_stateless/00309_formats.sql @@ -1,4 +1,6 @@ SET output_format_write_statistics = 0; +SET enable_named_columns_in_function_tuple = 0; + SELECT number * 246 + 10 AS n, toDate('2000-01-01') + n AS d, range(n) AS arr, arrayStringConcat(arrayMap(x -> reinterpretAsString(x), arr)) AS s, (n, d) AS tuple FROM system.numbers LIMIT 2 FORMAT RowBinary; SELECT number * 246 + 10 AS n, toDate('2000-01-01') + n AS d, range(n) AS arr, arrayStringConcat(arrayMap(x -> reinterpretAsString(x), arr)) AS s, (n, d) AS tuple FROM system.numbers LIMIT 2 FORMAT RowBinaryWithNamesAndTypes; SELECT number * 246 + 10 AS n, toDate('2000-01-01') + n AS d, range(n) AS arr, arrayStringConcat(arrayMap(x -> reinterpretAsString(x), arr)) AS s, (n, d) AS tuple FROM system.numbers LIMIT 2 FORMAT TabSeparatedWithNamesAndTypes; diff --git a/tests/queries/0_stateless/00411_long_accurate_number_comparison.python b/tests/queries/0_stateless/00411_long_accurate_number_comparison.python index 183a2637d36..045de9ee7ee 100644 --- a/tests/queries/0_stateless/00411_long_accurate_number_comparison.python +++ b/tests/queries/0_stateless/00411_long_accurate_number_comparison.python @@ -50,7 +50,7 @@ TYPES = { "UInt32": {"bits": 32, "sign": False, "float": False}, "Int32": {"bits": 32, "sign": True, "float": False}, "UInt64": {"bits": 64, "sign": False, "float": False}, - "Int64": {"bits": 64, "sign": True, "float": False} + "Int64": {"bits": 64, "sign": True, "float": False}, # "Float32" : { "bits" : 32, "sign" : True, "float" : True }, # "Float64" : { "bits" : 64, "sign" : True, "float" : True } } diff --git a/tests/queries/0_stateless/00429_long_http_bufferization.sh b/tests/queries/0_stateless/00429_long_http_bufferization.sh index 98dd300e6ab..83a6a4e8043 100755 --- a/tests/queries/0_stateless/00429_long_http_bufferization.sh +++ b/tests/queries/0_stateless/00429_long_http_bufferization.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-parallel +# Tags: long set -e diff --git a/tests/queries/0_stateless/00504_mergetree_arrays_rw.sql b/tests/queries/0_stateless/00504_mergetree_arrays_rw.sql index 7c939d060ea..14929045356 100644 --- a/tests/queries/0_stateless/00504_mergetree_arrays_rw.sql +++ b/tests/queries/0_stateless/00504_mergetree_arrays_rw.sql @@ -1,5 +1,8 @@ set allow_deprecated_syntax_for_merge_tree=1; +set max_threads = 1; +set max_insert_threads = 1; + drop table if exists test_ins_arr; create table test_ins_arr (date Date, val Array(UInt64)) engine = MergeTree(date, (date), 8192); insert into test_ins_arr select toDate('2017-10-02'), [number, 42] from system.numbers limit 10000; diff --git a/tests/queries/0_stateless/00510_materizlized_view_and_deduplication_zookeeper.reference b/tests/queries/0_stateless/00510_materizlized_view_and_deduplication_zookeeper.reference index adf6abb7298..9c9281dc7e4 100644 --- a/tests/queries/0_stateless/00510_materizlized_view_and_deduplication_zookeeper.reference +++ b/tests/queries/0_stateless/00510_materizlized_view_and_deduplication_zookeeper.reference @@ -1,7 +1,7 @@ 2 3 -2 +3 3 1 diff --git a/tests/queries/0_stateless/00510_materizlized_view_and_deduplication_zookeeper.sql b/tests/queries/0_stateless/00510_materizlized_view_and_deduplication_zookeeper.sql index d3c4da86b41..51e6a513608 100644 --- a/tests/queries/0_stateless/00510_materizlized_view_and_deduplication_zookeeper.sql +++ b/tests/queries/0_stateless/00510_materizlized_view_and_deduplication_zookeeper.sql @@ -29,7 +29,7 @@ INSERT INTO without_deduplication VALUES (43); SELECT count() FROM with_deduplication; SELECT count() FROM without_deduplication; --- Implicit insert isn't deduplicated +-- Implicit insert isn't deduplicated, because deduplicate_blocks_in_dependent_materialized_views = 0 by default SELECT ''; SELECT countMerge(cnt) FROM with_deduplication_mv; SELECT countMerge(cnt) FROM without_deduplication_mv; diff --git a/tests/queries/0_stateless/00564_versioned_collapsing_merge_tree.sql b/tests/queries/0_stateless/00564_versioned_collapsing_merge_tree.sql index 22f6da71247..494d8243534 100644 --- a/tests/queries/0_stateless/00564_versioned_collapsing_merge_tree.sql +++ b/tests/queries/0_stateless/00564_versioned_collapsing_merge_tree.sql @@ -2,6 +2,7 @@ set allow_deprecated_syntax_for_merge_tree=1; set optimize_on_insert = 0; +set optimize_trivial_insert_select = 1; drop table if exists mult_tab; create table mult_tab (date Date, value String, version UInt64, sign Int8) engine = VersionedCollapsingMergeTree(date, (date), 8192, sign, version); diff --git a/tests/queries/0_stateless/00600_replace_running_query.sh b/tests/queries/0_stateless/00600_replace_running_query.sh index 6a682210489..7a71d17f19b 100755 --- a/tests/queries/0_stateless/00600_replace_running_query.sh +++ b/tests/queries/0_stateless/00600_replace_running_query.sh @@ -1,5 +1,4 @@ #!/usr/bin/env bash -# Tags: no-parallel CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL=none @@ -7,9 +6,10 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -q "drop user if exists u_00600" -${CLICKHOUSE_CLIENT} -q "create user u_00600 settings max_execution_time=60, readonly=1" -${CLICKHOUSE_CLIENT} -q "grant select on system.numbers to u_00600" +TEST_PREFIX=$RANDOM +${CLICKHOUSE_CLIENT} -q "drop user if exists u_00600${TEST_PREFIX}" +${CLICKHOUSE_CLIENT} -q "create user u_00600${TEST_PREFIX} settings max_execution_time=60, readonly=1" +${CLICKHOUSE_CLIENT} -q "grant select on system.numbers to u_00600${TEST_PREFIX}" function wait_for_query_to_start() { @@ -26,7 +26,7 @@ $CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL&query_id=hello&replace_running_query=1" -d # Wait for it to be replaced wait -${CLICKHOUSE_CLIENT_BINARY} --user=u_00600 --query_id=42 --query='SELECT 2, count() FROM system.numbers' 2>&1 | grep -cF 'was cancelled' & +${CLICKHOUSE_CLIENT_BINARY} --user=u_00600${TEST_PREFIX} --query_id=42 --query='SELECT 2, count() FROM system.numbers' 2>&1 | grep -cF 'was cancelled' & wait_for_query_to_start '42' # Trying to run another query with the same query_id @@ -43,4 +43,4 @@ wait_for_query_to_start '42' ${CLICKHOUSE_CLIENT} --query_id=42 --replace_running_query=1 --replace_running_query_max_wait_ms=500 --query='SELECT 43' 2>&1 | grep -F "can't be stopped" > /dev/null wait ${CLICKHOUSE_CLIENT} --query_id=42 --replace_running_query=1 --query='SELECT 44' -${CLICKHOUSE_CLIENT} -q "drop user u_00600" +${CLICKHOUSE_CLIENT} -q "drop user u_00600${TEST_PREFIX}" diff --git a/tests/queries/0_stateless/00623_truncate_all_tables.sql b/tests/queries/0_stateless/00623_truncate_all_tables.sql index 2d5e9d48f59..2626f7ed285 100644 --- a/tests/queries/0_stateless/00623_truncate_all_tables.sql +++ b/tests/queries/0_stateless/00623_truncate_all_tables.sql @@ -1,50 +1,43 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS truncate_test; - -CREATE DATABASE IF NOT EXISTS truncate_test; -CREATE TABLE IF NOT EXISTS truncate_test.truncate_test_set(id UInt64) ENGINE = Set; -CREATE TABLE IF NOT EXISTS truncate_test.truncate_test_log(id UInt64) ENGINE = Log; -CREATE TABLE IF NOT EXISTS truncate_test.truncate_test_memory(id UInt64) ENGINE = Memory; -CREATE TABLE IF NOT EXISTS truncate_test.truncate_test_tiny_log(id UInt64) ENGINE = TinyLog; -CREATE TABLE IF NOT EXISTS truncate_test.truncate_test_stripe_log(id UInt64) ENGINE = StripeLog; -CREATE TABLE IF NOT EXISTS truncate_test.truncate_test_merge_tree(p Date, k UInt64) ENGINE = MergeTree ORDER BY p; +CREATE TABLE IF NOT EXISTS truncate_test_set(id UInt64) ENGINE = Set; +CREATE TABLE IF NOT EXISTS truncate_test_log(id UInt64) ENGINE = Log; +CREATE TABLE IF NOT EXISTS truncate_test_memory(id UInt64) ENGINE = Memory; +CREATE TABLE IF NOT EXISTS truncate_test_tiny_log(id UInt64) ENGINE = TinyLog; +CREATE TABLE IF NOT EXISTS truncate_test_stripe_log(id UInt64) ENGINE = StripeLog; +CREATE TABLE IF NOT EXISTS truncate_test_merge_tree(p Date, k UInt64) ENGINE = MergeTree ORDER BY p; SELECT '======Before Truncate======'; -INSERT INTO truncate_test.truncate_test_set VALUES(0); -INSERT INTO truncate_test.truncate_test_log VALUES(1); -INSERT INTO truncate_test.truncate_test_memory VALUES(1); -INSERT INTO truncate_test.truncate_test_tiny_log VALUES(1); -INSERT INTO truncate_test.truncate_test_stripe_log VALUES(1); -INSERT INTO truncate_test.truncate_test_merge_tree VALUES('2000-01-01', 1); -SELECT * FROM system.numbers WHERE number NOT IN truncate_test.truncate_test_set LIMIT 1; -SELECT * FROM truncate_test.truncate_test_log; -SELECT * FROM truncate_test.truncate_test_memory; -SELECT * FROM truncate_test.truncate_test_tiny_log; -SELECT * FROM truncate_test.truncate_test_stripe_log; -SELECT * FROM truncate_test.truncate_test_merge_tree; +INSERT INTO truncate_test_set VALUES(0); +INSERT INTO truncate_test_log VALUES(1); +INSERT INTO truncate_test_memory VALUES(1); +INSERT INTO truncate_test_tiny_log VALUES(1); +INSERT INTO truncate_test_stripe_log VALUES(1); +INSERT INTO truncate_test_merge_tree VALUES('2000-01-01', 1); +SELECT * FROM system.numbers WHERE number NOT IN truncate_test_set LIMIT 1; +SELECT * FROM truncate_test_log; +SELECT * FROM truncate_test_memory; +SELECT * FROM truncate_test_tiny_log; +SELECT * FROM truncate_test_stripe_log; +SELECT * FROM truncate_test_merge_tree; SELECT '======After Truncate And Empty======'; -TRUNCATE ALL TABLES FROM IF EXISTS truncate_test; -SELECT * FROM system.numbers WHERE number NOT IN truncate_test.truncate_test_set LIMIT 1; -SELECT * FROM truncate_test.truncate_test_log; -SELECT * FROM truncate_test.truncate_test_memory; -SELECT * FROM truncate_test.truncate_test_tiny_log; -SELECT * FROM truncate_test.truncate_test_stripe_log; -SELECT * FROM truncate_test.truncate_test_merge_tree; +TRUNCATE ALL TABLES FROM IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; +SELECT * FROM system.numbers WHERE number NOT IN truncate_test_set LIMIT 1; +SELECT * FROM truncate_test_log; +SELECT * FROM truncate_test_memory; +SELECT * FROM truncate_test_tiny_log; +SELECT * FROM truncate_test_stripe_log; +SELECT * FROM truncate_test_merge_tree; SELECT '======After Truncate And Insert Data======'; -INSERT INTO truncate_test.truncate_test_set VALUES(0); -INSERT INTO truncate_test.truncate_test_log VALUES(1); -INSERT INTO truncate_test.truncate_test_memory VALUES(1); -INSERT INTO truncate_test.truncate_test_tiny_log VALUES(1); -INSERT INTO truncate_test.truncate_test_stripe_log VALUES(1); -INSERT INTO truncate_test.truncate_test_merge_tree VALUES('2000-01-01', 1); -SELECT * FROM system.numbers WHERE number NOT IN truncate_test.truncate_test_set LIMIT 1; -SELECT * FROM truncate_test.truncate_test_log; -SELECT * FROM truncate_test.truncate_test_memory; -SELECT * FROM truncate_test.truncate_test_tiny_log; -SELECT * FROM truncate_test.truncate_test_stripe_log; -SELECT * FROM truncate_test.truncate_test_merge_tree; - -DROP DATABASE IF EXISTS truncate_test; +INSERT INTO truncate_test_set VALUES(0); +INSERT INTO truncate_test_log VALUES(1); +INSERT INTO truncate_test_memory VALUES(1); +INSERT INTO truncate_test_tiny_log VALUES(1); +INSERT INTO truncate_test_stripe_log VALUES(1); +INSERT INTO truncate_test_merge_tree VALUES('2000-01-01', 1); +SELECT * FROM system.numbers WHERE number NOT IN truncate_test_set LIMIT 1; +SELECT * FROM truncate_test_log; +SELECT * FROM truncate_test_memory; +SELECT * FROM truncate_test_tiny_log; +SELECT * FROM truncate_test_stripe_log; +SELECT * FROM truncate_test_merge_tree; diff --git a/tests/queries/0_stateless/00623_truncate_table.sql b/tests/queries/0_stateless/00623_truncate_table.sql index 4a67e49acda..e35803db1d9 100644 --- a/tests/queries/0_stateless/00623_truncate_table.sql +++ b/tests/queries/0_stateless/00623_truncate_table.sql @@ -1,6 +1,5 @@ set allow_deprecated_syntax_for_merge_tree=1; -DROP DATABASE IF EXISTS truncate_test; DROP TABLE IF EXISTS truncate_test_log; DROP TABLE IF EXISTS truncate_test_memory; DROP TABLE IF EXISTS truncate_test_tiny_log; @@ -9,7 +8,6 @@ DROP TABLE IF EXISTS truncate_test_merge_tree; DROP TABLE IF EXISTS truncate_test_materialized_view; DROP TABLE IF EXISTS truncate_test_materialized_depend; -CREATE DATABASE truncate_test; CREATE TABLE truncate_test_set(id UInt64) ENGINE = Set; CREATE TABLE truncate_test_log(id UInt64) ENGINE = Log; CREATE TABLE truncate_test_memory(id UInt64) ENGINE = Memory; @@ -75,4 +73,3 @@ DROP TABLE IF EXISTS truncate_test_stripe_log; DROP TABLE IF EXISTS truncate_test_merge_tree; DROP TABLE IF EXISTS truncate_test_materialized_view; DROP TABLE IF EXISTS truncate_test_materialized_depend; -DROP DATABASE IF EXISTS truncate_test; diff --git a/tests/queries/0_stateless/00626_replace_partition_from_table_zookeeper.sh b/tests/queries/0_stateless/00626_replace_partition_from_table_zookeeper.sh index ffbf4df4ba7..13146f2eab0 100755 --- a/tests/queries/0_stateless/00626_replace_partition_from_table_zookeeper.sh +++ b/tests/queries/0_stateless/00626_replace_partition_from_table_zookeeper.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: zookeeper, no-s3-storage +# Tags: zookeeper, no-object-storage # Because REPLACE PARTITION does not forces immediate removal of replaced data parts from local filesystem # (it tries to do it as quick as possible, but it still performed in separate thread asynchronously) diff --git a/tests/queries/0_stateless/00632_get_sample_block_cache.sql b/tests/queries/0_stateless/00632_get_sample_block_cache.sql index c54ca0b084e..ae9b6bb7b2c 100644 --- a/tests/queries/0_stateless/00632_get_sample_block_cache.sql +++ b/tests/queries/0_stateless/00632_get_sample_block_cache.sql @@ -1,4 +1,4 @@ --- Tags: long, no-s3-storage, no-asan +-- Tags: long, no-object-storage, no-asan SET joined_subquery_requires_alias = 0; diff --git a/tests/queries/0_stateless/00633_materialized_view_and_too_many_parts_zookeeper.sh b/tests/queries/0_stateless/00633_materialized_view_and_too_many_parts_zookeeper.sh index 1fb219108da..8f7d19028b0 100755 --- a/tests/queries/0_stateless/00633_materialized_view_and_too_many_parts_zookeeper.sh +++ b/tests/queries/0_stateless/00633_materialized_view_and_too_many_parts_zookeeper.sh @@ -36,8 +36,8 @@ ${CLICKHOUSE_CLIENT} --query "DROP TABLE c" echo ${CLICKHOUSE_CLIENT} --query "CREATE TABLE root (d UInt64) ENGINE = Null" ${CLICKHOUSE_CLIENT} --query "CREATE MATERIALIZED VIEW d (d UInt64) ENGINE = ReplicatedMergeTree('/clickhouse/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/d', '1') ORDER BY d AS SELECT * FROM root" -${CLICKHOUSE_CLIENT} --query "INSERT INTO root VALUES (1)"; -${CLICKHOUSE_CLIENT} --query "INSERT INTO root VALUES (1)"; +${CLICKHOUSE_CLIENT} --query "INSERT INTO root SETTINGS deduplicate_blocks_in_dependent_materialized_views=1 VALUES (1)"; +${CLICKHOUSE_CLIENT} --query "INSERT INTO root SETTINGS deduplicate_blocks_in_dependent_materialized_views=1 VALUES (1)"; ${CLICKHOUSE_CLIENT} --query "SELECT * FROM d"; ${CLICKHOUSE_CLIENT} --query "DROP TABLE root" ${CLICKHOUSE_CLIENT} --query "DROP TABLE d" diff --git a/tests/queries/0_stateless/00673_subquery_prepared_set_performance.sql b/tests/queries/0_stateless/00673_subquery_prepared_set_performance.sql index 0591592344c..98c0802ffbc 100644 --- a/tests/queries/0_stateless/00673_subquery_prepared_set_performance.sql +++ b/tests/queries/0_stateless/00673_subquery_prepared_set_performance.sql @@ -3,10 +3,10 @@ DROP TABLE IF EXISTS mergetree_00673; CREATE TABLE mergetree_00673 (x UInt64) ENGINE = MergeTree ORDER BY x; INSERT INTO mergetree_00673 VALUES (1); -SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM mergetree_00673 WHERE x IN (SELECT * FROM numbers(10000000)))))))))))); +SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM mergetree_00673 WHERE x IN (SELECT * FROM numbers(1000000)))))))))))))))))))))); SET force_primary_key = 1; -SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM mergetree_00673 WHERE x IN (SELECT * FROM numbers(10000000)))))))))))); +SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM mergetree_00673 WHERE x IN (SELECT * FROM numbers(1000000)))))))))))))))))))))); DROP TABLE mergetree_00673; diff --git a/tests/queries/0_stateless/00688_low_cardinality_dictionary_deserialization.sql b/tests/queries/0_stateless/00688_low_cardinality_dictionary_deserialization.sql index c4613acf5f3..d359efd8d42 100644 --- a/tests/queries/0_stateless/00688_low_cardinality_dictionary_deserialization.sql +++ b/tests/queries/0_stateless/00688_low_cardinality_dictionary_deserialization.sql @@ -1,5 +1,5 @@ drop table if exists lc_dict_reading; -create table lc_dict_reading (val UInt64, str StringWithDictionary, pat String) engine = MergeTree order by val SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; +create table lc_dict_reading (val UInt64, str LowCardinality(String), pat String) engine = MergeTree order by val SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; insert into lc_dict_reading select number, if(number < 8192 * 4, number % 100, number) as s, s from system.numbers limit 1000000; select sum(toUInt64(str)), sum(toUInt64(pat)) from lc_dict_reading where val < 8129 or val > 8192 * 4; drop table if exists lc_dict_reading; diff --git a/tests/queries/0_stateless/00688_low_cardinality_in.sql b/tests/queries/0_stateless/00688_low_cardinality_in.sql index cb57fad51a4..c39fdb37160 100644 --- a/tests/queries/0_stateless/00688_low_cardinality_in.sql +++ b/tests/queries/0_stateless/00688_low_cardinality_in.sql @@ -1,6 +1,6 @@ set allow_suspicious_low_cardinality_types = 1; drop table if exists lc_00688; -create table lc_00688 (str StringWithDictionary, val UInt8WithDictionary) engine = MergeTree order by tuple(); +create table lc_00688 (str LowCardinality(String), val LowCardinality(UInt8)) engine = MergeTree order by tuple(); insert into lc_00688 values ('a', 1), ('b', 2); select str, str in ('a', 'd') from lc_00688; select val, val in (1, 3) from lc_00688; diff --git a/tests/queries/0_stateless/00688_low_cardinality_prewhere.sql b/tests/queries/0_stateless/00688_low_cardinality_prewhere.sql index a15b2540fe6..17c74b7ca05 100644 --- a/tests/queries/0_stateless/00688_low_cardinality_prewhere.sql +++ b/tests/queries/0_stateless/00688_low_cardinality_prewhere.sql @@ -1,5 +1,5 @@ drop table if exists lc_prewhere; -create table lc_prewhere (key UInt64, val UInt64, str StringWithDictionary, s String) engine = MergeTree order by key settings index_granularity = 8192; +create table lc_prewhere (key UInt64, val UInt64, str LowCardinality(String), s String) engine = MergeTree order by key settings index_granularity = 8192; insert into lc_prewhere select number, if(number < 10 or number > 8192 * 9, 1, 0), toString(number) as s, s from system.numbers limit 100000; select sum(toUInt64(str)), sum(toUInt64(s)) from lc_prewhere prewhere val == 1; drop table if exists lc_prewhere; diff --git a/tests/queries/0_stateless/00688_low_cardinality_serialization.sql b/tests/queries/0_stateless/00688_low_cardinality_serialization.sql index b4fe4b29200..1e4de3f3d3e 100644 --- a/tests/queries/0_stateless/00688_low_cardinality_serialization.sql +++ b/tests/queries/0_stateless/00688_low_cardinality_serialization.sql @@ -8,8 +8,8 @@ select 'MergeTree'; drop table if exists lc_small_dict; drop table if exists lc_big_dict; -create table lc_small_dict (str StringWithDictionary) engine = MergeTree order by str SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; -create table lc_big_dict (str StringWithDictionary) engine = MergeTree order by str SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; +create table lc_small_dict (str LowCardinality(String)) engine = MergeTree order by str SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; +create table lc_big_dict (str LowCardinality(String)) engine = MergeTree order by str SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; insert into lc_small_dict select toString(number % 1000) from system.numbers limit 1000000; insert into lc_big_dict select toString(number) from system.numbers limit 1000000; diff --git a/tests/queries/0_stateless/00688_low_cardinality_syntax.reference b/tests/queries/0_stateless/00688_low_cardinality_syntax.reference index ca27069a7df..b06beb0d6f0 100644 --- a/tests/queries/0_stateless/00688_low_cardinality_syntax.reference +++ b/tests/queries/0_stateless/00688_low_cardinality_syntax.reference @@ -1,13 +1,7 @@ a a -a -a 1 1 -1 -1 -ab -ab ab ab - diff --git a/tests/queries/0_stateless/00688_low_cardinality_syntax.sql b/tests/queries/0_stateless/00688_low_cardinality_syntax.sql index a11d9e2d9fe..dccdac1d95d 100644 --- a/tests/queries/0_stateless/00688_low_cardinality_syntax.sql +++ b/tests/queries/0_stateless/00688_low_cardinality_syntax.sql @@ -13,56 +13,32 @@ drop table if exists lc_null_fix_str_0; drop table if exists lc_null_fix_str_1; create table lc_str_0 (str LowCardinality(String)) engine = Memory; -create table lc_str_1 (str StringWithDictionary) engine = Memory; create table lc_null_str_0 (str LowCardinality(Nullable(String))) engine = Memory; -create table lc_null_str_1 (str NullableWithDictionary(String)) engine = Memory; create table lc_int8_0 (val LowCardinality(Int8)) engine = Memory; -create table lc_int8_1 (val Int8WithDictionary) engine = Memory; create table lc_null_int8_0 (val LowCardinality(Nullable(Int8))) engine = Memory; -create table lc_null_int8_1 (val NullableWithDictionary(Int8)) engine = Memory; create table lc_fix_str_0 (str LowCardinality(FixedString(2))) engine = Memory; -create table lc_fix_str_1 (str FixedStringWithDictionary(2)) engine = Memory; create table lc_null_fix_str_0 (str LowCardinality(Nullable(FixedString(2)))) engine = Memory; -create table lc_null_fix_str_1 (str NullableWithDictionary(FixedString(2))) engine = Memory; insert into lc_str_0 select 'a'; -insert into lc_str_1 select 'a'; insert into lc_null_str_0 select 'a'; -insert into lc_null_str_1 select 'a'; insert into lc_int8_0 select 1; -insert into lc_int8_1 select 1; insert into lc_null_int8_0 select 1; -insert into lc_null_int8_1 select 1; insert into lc_fix_str_0 select 'ab'; -insert into lc_fix_str_1 select 'ab'; insert into lc_null_fix_str_0 select 'ab'; -insert into lc_null_fix_str_1 select 'ab'; select str from lc_str_0; -select str from lc_str_1; select str from lc_null_str_0; -select str from lc_null_str_1; select val from lc_int8_0; -select val from lc_int8_1; select val from lc_null_int8_0; -select val from lc_null_int8_1; select str from lc_fix_str_0; -select str from lc_fix_str_1; select str from lc_null_fix_str_0; -select str from lc_null_fix_str_1; drop table if exists lc_str_0; -drop table if exists lc_str_1; drop table if exists lc_null_str_0; -drop table if exists lc_null_str_1; drop table if exists lc_int8_0; -drop table if exists lc_int8_1; drop table if exists lc_null_int8_0; -drop table if exists lc_null_int8_1; drop table if exists lc_fix_str_0; -drop table if exists lc_fix_str_1; drop table if exists lc_null_fix_str_0; -drop table if exists lc_null_fix_str_1; select '-'; SELECT toLowCardinality('a') AS s, toTypeName(s), toTypeName(length(s)) from system.one; @@ -73,7 +49,7 @@ select (toLowCardinality(z) as val) || 'b' from (select arrayJoin(['c', 'd']) a select '-'; drop table if exists lc_str_uuid; -create table lc_str_uuid(str1 String, str2 LowCardinality(String), str3 StringWithDictionary) ENGINE=Memory; +create table lc_str_uuid(str1 String, str2 LowCardinality(String), str3 LowCardinality(String)) ENGINE=Memory; select toUUID(str1), toUUID(str2), toUUID(str3) from lc_str_uuid; select toUUID(str1, '', NULL), toUUID(str2, '', NULL), toUUID(str3, '', NULL) from lc_str_uuid; insert into lc_str_uuid values ('61f0c404-5cb3-11e7-907b-a6006ad3dba0', '61f0c404-5cb3-11e7-907b-a6006ad3dba0', '61f0c404-5cb3-11e7-907b-a6006ad3dba0'); diff --git a/tests/queries/0_stateless/00705_drop_create_merge_tree.sh b/tests/queries/0_stateless/00705_drop_create_merge_tree.sh index d7754091290..ea8b9d02e49 100755 --- a/tests/queries/0_stateless/00705_drop_create_merge_tree.sh +++ b/tests/queries/0_stateless/00705_drop_create_merge_tree.sh @@ -5,8 +5,8 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -yes 'CREATE TABLE IF NOT EXISTS table (x UInt8) ENGINE = MergeTree ORDER BY tuple();' | head -n 1000 | $CLICKHOUSE_CLIENT --ignore-error -nm 2>/dev/null & -yes 'DROP TABLE table;' | head -n 1000 | $CLICKHOUSE_CLIENT --ignore-error -nm 2>/dev/null & +yes 'CREATE TABLE IF NOT EXISTS table (x UInt8) ENGINE = MergeTree ORDER BY tuple();' | head -n 1000 | $CLICKHOUSE_CLIENT --multiquery & +yes 'DROP TABLE IF EXISTS table;' | head -n 1000 | $CLICKHOUSE_CLIENT --multiquery & wait ${CLICKHOUSE_CLIENT} --query "DROP TABLE IF EXISTS table" diff --git a/tests/queries/0_stateless/00717_low_cardinaliry_group_by.sql b/tests/queries/0_stateless/00717_low_cardinaliry_group_by.sql index 02915d4e611..3115ab508fe 100644 --- a/tests/queries/0_stateless/00717_low_cardinaliry_group_by.sql +++ b/tests/queries/0_stateless/00717_low_cardinaliry_group_by.sql @@ -1,5 +1,5 @@ drop table if exists tab_00717; -create table tab_00717 (a String, b StringWithDictionary) engine = MergeTree order by a; +create table tab_00717 (a String, b LowCardinality(String)) engine = MergeTree order by a; insert into tab_00717 values ('a_1', 'b_1'), ('a_2', 'b_2'); select count() from tab_00717; select a from tab_00717 group by a order by a; diff --git a/tests/queries/0_stateless/00718_low_cardinaliry_alter.sql b/tests/queries/0_stateless/00718_low_cardinaliry_alter.sql index 591ff952132..524e396bcb1 100644 --- a/tests/queries/0_stateless/00718_low_cardinaliry_alter.sql +++ b/tests/queries/0_stateless/00718_low_cardinaliry_alter.sql @@ -7,7 +7,7 @@ alter table tab_00718 modify column b UInt32; select *, toTypeName(b) from tab_00718; alter table tab_00718 modify column b LowCardinality(UInt32); select *, toTypeName(b) from tab_00718; -alter table tab_00718 modify column b StringWithDictionary; +alter table tab_00718 modify column b LowCardinality(String); select *, toTypeName(b) from tab_00718; alter table tab_00718 modify column b LowCardinality(UInt32); select *, toTypeName(b) from tab_00718; diff --git a/tests/queries/0_stateless/00719_parallel_ddl_db.sh b/tests/queries/0_stateless/00719_parallel_ddl_db.sh index b7dea25c182..ac4e4eae4f2 100755 --- a/tests/queries/0_stateless/00719_parallel_ddl_db.sh +++ b/tests/queries/0_stateless/00719_parallel_ddl_db.sh @@ -1,13 +1,12 @@ #!/usr/bin/env bash -# Tags: no-parallel - set -e CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} --query "DROP DATABASE IF EXISTS parallel_ddl" +DB_SUFFIX=${RANDOM}${RANDOM}${RANDOM}${RANDOM} +${CLICKHOUSE_CLIENT} --query "DROP DATABASE IF EXISTS parallel_ddl_${DB_SUFFIX}" function query() { @@ -16,8 +15,8 @@ function query() while [ $SECONDS -lt "$TIMELIMIT" ] && [ $it -lt 50 ]; do it=$((it+1)) - ${CLICKHOUSE_CLIENT} --query "CREATE DATABASE IF NOT EXISTS parallel_ddl" - ${CLICKHOUSE_CLIENT} --query "DROP DATABASE IF EXISTS parallel_ddl" + ${CLICKHOUSE_CLIENT} --query "CREATE DATABASE IF NOT EXISTS parallel_ddl_${DB_SUFFIX}" + ${CLICKHOUSE_CLIENT} --query "DROP DATABASE IF EXISTS parallel_ddl_${DB_SUFFIX}" done } @@ -27,4 +26,4 @@ done wait -${CLICKHOUSE_CLIENT} --query "DROP DATABASE IF EXISTS parallel_ddl" +${CLICKHOUSE_CLIENT} --query "DROP DATABASE IF EXISTS parallel_ddl_${DB_SUFFIX}" diff --git a/tests/queries/0_stateless/00727_concat.reference b/tests/queries/0_stateless/00727_concat.reference index 6fb23c072d3..329ad36ad3c 100644 --- a/tests/queries/0_stateless/00727_concat.reference +++ b/tests/queries/0_stateless/00727_concat.reference @@ -72,3 +72,6 @@ foo \N \N Testing the alias +-- Empty argument tests + +String diff --git a/tests/queries/0_stateless/00727_concat.sql b/tests/queries/0_stateless/00727_concat.sql index 01792545b5a..76dae541261 100644 --- a/tests/queries/0_stateless/00727_concat.sql +++ b/tests/queries/0_stateless/00727_concat.sql @@ -93,4 +93,6 @@ SELECT concat(materialize(NULL :: Nullable(UInt64))); SELECT CONCAT('Testing the ', 'alias'); -SELECT concat(); -- { serverError TOO_FEW_ARGUMENTS_FOR_FUNCTION } +SELECT '-- Empty argument tests'; +SELECT concat(); +select toTypeName(concat()); diff --git a/tests/queries/0_stateless/00731_long_merge_tree_select_opened_files.sh b/tests/queries/0_stateless/00731_long_merge_tree_select_opened_files.sh index 1bb4dbd34de..5a4fd901f8d 100755 --- a/tests/queries/0_stateless/00731_long_merge_tree_select_opened_files.sh +++ b/tests/queries/0_stateless/00731_long_merge_tree_select_opened_files.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-s3-storage +# Tags: long, no-object-storage, no-tsan # no-s3 because read FileOpen metric set -e @@ -31,6 +31,6 @@ $CLICKHOUSE_CLIENT $settings -q "$touching_many_parts_query" &> /dev/null $CLICKHOUSE_CLIENT $settings -q "SYSTEM FLUSH LOGS" -$CLICKHOUSE_CLIENT $settings -q "SELECT ProfileEvents['FileOpen'] as opened_files FROM system.query_log WHERE query='$touching_many_parts_query' and current_database = currentDatabase() ORDER BY event_time DESC, opened_files DESC LIMIT 1;" +$CLICKHOUSE_CLIENT $settings -q "SELECT ProfileEvents['FileOpen'] as opened_files FROM system.query_log WHERE query = '$touching_many_parts_query' AND current_database = currentDatabase() AND event_date >= yesterday() ORDER BY event_time DESC, opened_files DESC LIMIT 1;" $CLICKHOUSE_CLIENT $settings -q "DROP TABLE IF EXISTS merge_tree_table;" diff --git a/tests/queries/0_stateless/00752_low_cardinality_mv_1.sql b/tests/queries/0_stateless/00752_low_cardinality_mv_1.sql index 60cc30ec2c3..4fdd0a2496e 100644 --- a/tests/queries/0_stateless/00752_low_cardinality_mv_1.sql +++ b/tests/queries/0_stateless/00752_low_cardinality_mv_1.sql @@ -1,7 +1,7 @@ drop table if exists lc_00752; drop table if exists lc_mv_00752; -create table lc_00752 (str StringWithDictionary) engine = MergeTree order by tuple(); +create table lc_00752 (str LowCardinality(String)) engine = MergeTree order by tuple(); insert into lc_00752 values ('a'), ('bbb'), ('ab'), ('accccc'), ('baasddas'), ('bcde'); @@ -12,4 +12,3 @@ select * from lc_mv_00752 order by letter; drop table if exists lc_00752; drop table if exists lc_mv_00752; - diff --git a/tests/queries/0_stateless/00753_system_columns_and_system_tables_long.sql b/tests/queries/0_stateless/00753_system_columns_and_system_tables_long.sql index 4613576cf4e..009fc0bbb9f 100644 --- a/tests/queries/0_stateless/00753_system_columns_and_system_tables_long.sql +++ b/tests/queries/0_stateless/00753_system_columns_and_system_tables_long.sql @@ -1,4 +1,4 @@ --- Tags: long, no-s3-storage, no-random-merge-tree-settings +-- Tags: long, no-object-storage, no-random-merge-tree-settings SET output_format_pretty_row_numbers = 0; DROP TABLE IF EXISTS check_system_tables; diff --git a/tests/queries/0_stateless/00763_lock_buffer_long.sh b/tests/queries/0_stateless/00763_lock_buffer_long.sh index 50680724149..444a66767aa 100755 --- a/tests/queries/0_stateless/00763_lock_buffer_long.sh +++ b/tests/queries/0_stateless/00763_lock_buffer_long.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: long +# Tags: long, no-object-storage, no-msan, no-asan, no-tsan, no-debug +# Some kind of stress test, it doesn't make sense to test in a non-release build set -e @@ -15,12 +16,12 @@ ${CLICKHOUSE_CLIENT} --query="CREATE TABLE buffer_00763_2 (s String) ENGINE = Bu function thread1() { - seq 1 500 | sed -r -e 's/.+/DROP TABLE IF EXISTS mt_00763_2; CREATE TABLE mt_00763_2 (s String) ENGINE = MergeTree ORDER BY s; INSERT INTO mt_00763_2 SELECT toString(number) FROM numbers(10);/' | ${CLICKHOUSE_CLIENT} --multiquery --ignore-error ||: + seq 1 500 | sed -r -e 's/.+/DROP TABLE IF EXISTS mt_00763_2; CREATE TABLE mt_00763_2 (s String) ENGINE = MergeTree ORDER BY s; INSERT INTO mt_00763_2 SELECT toString(number) FROM numbers(10);/' | ${CLICKHOUSE_CLIENT} --fsync-metadata 0 --multiquery --ignore-error ||: } function thread2() { - seq 1 1000 | sed -r -e 's/.+/SELECT count() FROM buffer_00763_2;/' | ${CLICKHOUSE_CLIENT} --multiquery --server_logs_file='/dev/null' --ignore-error 2>&1 | grep -vP '^0$|^10$|^Received exception|^Code: 60|^Code: 218|^Code: 473' | grep -v '(query: ' + seq 1 500 | sed -r -e 's/.+/SELECT count() FROM buffer_00763_2;/' | ${CLICKHOUSE_CLIENT} --multiquery --server_logs_file='/dev/null' --ignore-error 2>&1 | grep -vP '^0$|^10$|^Received exception|^Code: 60|^Code: 218|^Code: 473' | grep -v '(query: ' } thread1 & diff --git a/tests/queries/0_stateless/00816_long_concurrent_alter_column.sh b/tests/queries/0_stateless/00816_long_concurrent_alter_column.sh index 71acc11b971..0ed9593c689 100755 --- a/tests/queries/0_stateless/00816_long_concurrent_alter_column.sh +++ b/tests/queries/0_stateless/00816_long_concurrent_alter_column.sh @@ -50,7 +50,7 @@ export -f thread2; export -f thread3; export -f thread4; -TIMEOUT=30 +TIMEOUT=20 timeout $TIMEOUT bash -c thread1 2> /dev/null & timeout $TIMEOUT bash -c thread2 2> /dev/null & diff --git a/tests/queries/0_stateless/00830_join_overwrite.reference b/tests/queries/0_stateless/00830_join_overwrite.reference index 4792e70f333..e7d6081b647 100644 --- a/tests/queries/0_stateless/00830_join_overwrite.reference +++ b/tests/queries/0_stateless/00830_join_overwrite.reference @@ -1,2 +1,4 @@ 2 3 +2 +3 diff --git a/tests/queries/0_stateless/00830_join_overwrite.sql b/tests/queries/0_stateless/00830_join_overwrite.sql index cb7e277906b..bc3662528db 100644 --- a/tests/queries/0_stateless/00830_join_overwrite.sql +++ b/tests/queries/0_stateless/00830_join_overwrite.sql @@ -9,5 +9,14 @@ INSERT INTO kv_overwrite VALUES (1, 2); INSERT INTO kv_overwrite VALUES (1, 3); SELECT joinGet('kv_overwrite', 'v', toUInt32(1)); + +CREATE TABLE t2 (k UInt32, v UInt32) ENGINE = Memory; +INSERT INTO t2 VALUES (1, 2), (1, 3); + +SET allow_experimental_analyzer = 1; + +SELECT v FROM (SELECT 1 as k) t1 ANY INNER JOIN t2 USING (k) SETTINGS join_any_take_last_row = 0; +SELECT v FROM (SELECT 1 as k) t1 ANY INNER JOIN t2 USING (k) SETTINGS join_any_take_last_row = 1; + DROP TABLE kv; DROP TABLE kv_overwrite; diff --git a/tests/queries/0_stateless/00840_long_concurrent_select_and_drop_deadlock.sh b/tests/queries/0_stateless/00840_long_concurrent_select_and_drop_deadlock.sh index 238cdcea547..f0bc52ee356 100755 --- a/tests/queries/0_stateless/00840_long_concurrent_select_and_drop_deadlock.sh +++ b/tests/queries/0_stateless/00840_long_concurrent_select_and_drop_deadlock.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: deadlock, no-parallel, no-debug +# Tags: deadlock, no-debug, no-parallel # NOTE: database = $CLICKHOUSE_DATABASE is unwanted @@ -49,7 +49,7 @@ function thread_select() export -f thread_drop_create export -f thread_select -TIMEOUT=60 +TIMEOUT=30 thread_drop_create $TIMEOUT & thread_select $TIMEOUT & diff --git a/tests/queries/0_stateless/00910_buffer_prewhere.sql b/tests/queries/0_stateless/00910_buffer_prewhere.sql index deda0db85fb..e6b1cc424ad 100644 --- a/tests/queries/0_stateless/00910_buffer_prewhere.sql +++ b/tests/queries/0_stateless/00910_buffer_prewhere.sql @@ -1,9 +1,4 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS test_buffer; -CREATE DATABASE test_buffer; -CREATE TABLE test_buffer.mt (uid UInt64, ts DateTime, val Float64) ENGINE = MergeTree PARTITION BY toDate(ts) ORDER BY (uid, ts); -CREATE TABLE test_buffer.buf as test_buffer.mt ENGINE = Buffer(test_buffer, mt, 2, 10, 60, 10000, 100000, 1000000, 10000000); -INSERT INTO test_buffer.buf VALUES (1, '2019-03-01 10:00:00', 0.5), (2, '2019-03-02 10:00:00', 0.15), (1, '2019-03-03 10:00:00', 0.25); -SELECT count() from test_buffer.buf prewhere ts > toDateTime('2019-03-01 12:00:00') and ts < toDateTime('2019-03-02 12:00:00'); -DROP DATABASE test_buffer; +CREATE TABLE mt (uid UInt64, ts DateTime, val Float64) ENGINE = MergeTree PARTITION BY toDate(ts) ORDER BY (uid, ts); +CREATE TABLE buf as mt ENGINE = Buffer({CLICKHOUSE_DATABASE:Identifier}, mt, 2, 10, 60, 10000, 100000, 1000000, 10000000); +INSERT INTO buf VALUES (1, '2019-03-01 10:00:00', 0.5), (2, '2019-03-02 10:00:00', 0.15), (1, '2019-03-03 10:00:00', 0.25); +SELECT count() from buf prewhere ts > toDateTime('2019-03-01 12:00:00') and ts < toDateTime('2019-03-02 12:00:00'); diff --git a/tests/queries/0_stateless/00927_asof_join_correct_bt.reference b/tests/queries/0_stateless/00927_asof_join_correct_bt.reference index bb199d0159a..28c48d2e290 100644 --- a/tests/queries/0_stateless/00927_asof_join_correct_bt.reference +++ b/tests/queries/0_stateless/00927_asof_join_correct_bt.reference @@ -1,13 +1,36 @@ +-- { echoOn } +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B1 B USING(k,t) ORDER BY (A.k, A.t); 1 101 1 0 0 0 1 102 2 2 102 1 1 103 3 2 102 1 1 104 4 4 104 1 1 105 5 4 104 1 +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B2 B USING(k,t) ORDER BY (A.k, A.t); 1 101 1 0 0 0 1 102 2 2 102 1 1 103 3 2 102 1 1 104 4 4 104 1 1 105 5 4 104 1 +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B3 B USING(k,t) ORDER BY (A.k, A.t); +1 101 1 0 0 0 +1 102 2 2 102 1 +1 103 3 2 102 1 +1 104 4 4 104 1 +1 105 5 4 104 1 +SET join_algorithm = 'full_sorting_merge'; +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B1 B USING(k,t) ORDER BY (A.k, A.t); +1 101 1 0 0 0 +1 102 2 2 102 1 +1 103 3 2 102 1 +1 104 4 4 104 1 +1 105 5 4 104 1 +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B2 B USING(k,t) ORDER BY (A.k, A.t); +1 101 1 0 0 0 +1 102 2 2 102 1 +1 103 3 2 102 1 +1 104 4 4 104 1 +1 105 5 4 104 1 +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B3 B USING(k,t) ORDER BY (A.k, A.t); 1 101 1 0 0 0 1 102 2 2 102 1 1 103 3 2 102 1 diff --git a/tests/queries/0_stateless/00927_asof_join_correct_bt.sql b/tests/queries/0_stateless/00927_asof_join_correct_bt.sql index 281a81d51c0..d796b62d3b3 100644 --- a/tests/queries/0_stateless/00927_asof_join_correct_bt.sql +++ b/tests/queries/0_stateless/00927_asof_join_correct_bt.sql @@ -4,20 +4,29 @@ DROP TABLE IF EXISTS B; CREATE TABLE A(k UInt32, t UInt32, a UInt64) ENGINE = MergeTree() ORDER BY (k, t); INSERT INTO A(k,t,a) VALUES (1,101,1),(1,102,2),(1,103,3),(1,104,4),(1,105,5); -CREATE TABLE B(k UInt32, t UInt32, b UInt64) ENGINE = MergeTree() ORDER BY (k, t); -INSERT INTO B(k,t,b) VALUES (1,102,2), (1,104,4); -SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B USING(k,t) ORDER BY (A.k, A.t); -DROP TABLE B; +CREATE TABLE B1(k UInt32, t UInt32, b UInt64) ENGINE = MergeTree() ORDER BY (k, t); +INSERT INTO B1(k,t,b) VALUES (1,102,2), (1,104,4); +CREATE TABLE B2(t UInt32, k UInt32, b UInt64) ENGINE = MergeTree() ORDER BY (k, t); +INSERT INTO B2(k,t,b) VALUES (1,102,2), (1,104,4); -CREATE TABLE B(t UInt32, k UInt32, b UInt64) ENGINE = MergeTree() ORDER BY (k, t); -INSERT INTO B(k,t,b) VALUES (1,102,2), (1,104,4); -SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B USING(k,t) ORDER BY (A.k, A.t); -DROP TABLE B; +CREATE TABLE B3(k UInt32, b UInt64, t UInt32) ENGINE = MergeTree() ORDER BY (k, t); +INSERT INTO B3(k,t,b) VALUES (1,102,2), (1,104,4); -CREATE TABLE B(k UInt32, b UInt64, t UInt32) ENGINE = MergeTree() ORDER BY (k, t); -INSERT INTO B(k,t,b) VALUES (1,102,2), (1,104,4); -SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B USING(k,t) ORDER BY (A.k, A.t); -DROP TABLE B; +-- { echoOn } +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B1 B USING(k,t) ORDER BY (A.k, A.t); +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B2 B USING(k,t) ORDER BY (A.k, A.t); +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B3 B USING(k,t) ORDER BY (A.k, A.t); + +SET join_algorithm = 'full_sorting_merge'; +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B1 B USING(k,t) ORDER BY (A.k, A.t); +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B2 B USING(k,t) ORDER BY (A.k, A.t); +SELECT A.k, A.t, A.a, B.b, B.t, B.k FROM A ASOF LEFT JOIN B3 B USING(k,t) ORDER BY (A.k, A.t); + +-- { echoOff } + +DROP TABLE B1; +DROP TABLE B2; +DROP TABLE B3; DROP TABLE A; diff --git a/tests/queries/0_stateless/00927_asof_join_long.reference b/tests/queries/0_stateless/00927_asof_join_long.reference index d4f015c68e4..ec40d2bc463 100644 --- a/tests/queries/0_stateless/00927_asof_join_long.reference +++ b/tests/queries/0_stateless/00927_asof_join_long.reference @@ -1 +1,2 @@ 3000000 +3000000 diff --git a/tests/queries/0_stateless/00927_asof_join_long.sql b/tests/queries/0_stateless/00927_asof_join_long.sql index c03a06d48d4..7a73875e93e 100644 --- a/tests/queries/0_stateless/00927_asof_join_long.sql +++ b/tests/queries/0_stateless/00927_asof_join_long.sql @@ -2,15 +2,28 @@ DROP TABLE IF EXISTS tvs; +-- to use different algorithms for in subquery +SET allow_experimental_analyzer = 1; + CREATE TABLE tvs(k UInt32, t UInt32, tv UInt64) ENGINE = Memory; INSERT INTO tvs(k,t,tv) SELECT k, t, t FROM (SELECT toUInt32(number) AS k FROM numbers(1000)) keys -CROSS JOIN (SELECT toUInt32(number * 3) as t FROM numbers(10000)) tv_times; +CROSS JOIN (SELECT toUInt32(number * 3) as t FROM numbers(10000)) tv_times +SETTINGS join_algorithm = 'hash'; SELECT SUM(trades.price - tvs.tv) FROM (SELECT k, t, t as price FROM (SELECT toUInt32(number) AS k FROM numbers(1000)) keys - CROSS JOIN (SELECT toUInt32(number * 10) AS t FROM numbers(3000)) trade_times) trades + CROSS JOIN (SELECT toUInt32(number * 10) AS t FROM numbers(3000)) trade_times + SETTINGS join_algorithm = 'hash') trades ASOF LEFT JOIN tvs USING(k,t); +SELECT SUM(trades.price - tvs.tv) FROM +(SELECT k, t, t as price + FROM (SELECT toUInt32(number) AS k FROM numbers(1000)) keys + CROSS JOIN (SELECT toUInt32(number * 10) AS t FROM numbers(3000)) trade_times + SETTINGS join_algorithm = 'hash') trades +ASOF LEFT JOIN tvs USING(k,t) +SETTINGS join_algorithm = 'full_sorting_merge'; + DROP TABLE tvs; diff --git a/tests/queries/0_stateless/00927_asof_join_noninclusive.reference b/tests/queries/0_stateless/00927_asof_join_noninclusive.reference index fe2844a2a43..d856372fb4a 100644 --- a/tests/queries/0_stateless/00927_asof_join_noninclusive.reference +++ b/tests/queries/0_stateless/00927_asof_join_noninclusive.reference @@ -27,3 +27,32 @@ 2 1970-01-01 00:00:03 3 3 1970-01-01 00:00:03 2 2 1970-01-01 00:00:04 4 3 1970-01-01 00:00:03 2 2 1970-01-01 00:00:05 5 3 1970-01-01 00:00:03 2 +1 1970-01-01 00:00:01 1 0 1970-01-01 00:00:00 0 +1 1970-01-01 00:00:02 2 2 1970-01-01 00:00:02 1 +1 1970-01-01 00:00:03 3 2 1970-01-01 00:00:02 1 +1 1970-01-01 00:00:04 4 4 1970-01-01 00:00:04 1 +1 1970-01-01 00:00:05 5 4 1970-01-01 00:00:04 1 +2 1970-01-01 00:00:01 1 0 1970-01-01 00:00:00 0 +2 1970-01-01 00:00:02 2 0 1970-01-01 00:00:00 0 +2 1970-01-01 00:00:03 3 3 1970-01-01 00:00:03 2 +2 1970-01-01 00:00:04 4 3 1970-01-01 00:00:03 2 +2 1970-01-01 00:00:05 5 3 1970-01-01 00:00:03 2 +3 1970-01-01 00:00:01 1 0 1970-01-01 00:00:00 0 +3 1970-01-01 00:00:02 2 0 1970-01-01 00:00:00 0 +3 1970-01-01 00:00:03 3 0 1970-01-01 00:00:00 0 +3 1970-01-01 00:00:04 4 0 1970-01-01 00:00:00 0 +3 1970-01-01 00:00:05 5 0 1970-01-01 00:00:00 0 +1 1970-01-01 00:00:02 2 2 1970-01-01 00:00:02 1 +1 1970-01-01 00:00:03 3 2 1970-01-01 00:00:02 1 +1 1970-01-01 00:00:04 4 4 1970-01-01 00:00:04 1 +1 1970-01-01 00:00:05 5 4 1970-01-01 00:00:04 1 +2 1970-01-01 00:00:03 3 3 1970-01-01 00:00:03 2 +2 1970-01-01 00:00:04 4 3 1970-01-01 00:00:03 2 +2 1970-01-01 00:00:05 5 3 1970-01-01 00:00:03 2 +1 1970-01-01 00:00:02 2 2 1970-01-01 00:00:02 1 +1 1970-01-01 00:00:03 3 2 1970-01-01 00:00:02 1 +1 1970-01-01 00:00:04 4 4 1970-01-01 00:00:04 1 +1 1970-01-01 00:00:05 5 4 1970-01-01 00:00:04 1 +2 1970-01-01 00:00:03 3 3 1970-01-01 00:00:03 2 +2 1970-01-01 00:00:04 4 3 1970-01-01 00:00:03 2 +2 1970-01-01 00:00:05 5 3 1970-01-01 00:00:03 2 diff --git a/tests/queries/0_stateless/00927_asof_join_noninclusive.sql b/tests/queries/0_stateless/00927_asof_join_noninclusive.sql index 5f15f3b593d..3cc99df4462 100644 --- a/tests/queries/0_stateless/00927_asof_join_noninclusive.sql +++ b/tests/queries/0_stateless/00927_asof_join_noninclusive.sql @@ -11,9 +11,12 @@ INSERT INTO B(k,t,b) VALUES (1,2,2),(1,4,4); INSERT INTO B(k,t,b) VALUES (2,3,3); SELECT A.k, toString(A.t, 'UTC'), A.a, B.b, toString(B.t, 'UTC'), B.k FROM A ASOF LEFT JOIN B USING(k,t) ORDER BY (A.k, A.t); - SELECT A.k, toString(A.t, 'UTC'), A.a, B.b, toString(B.t, 'UTC'), B.k FROM A ASOF INNER JOIN B ON A.k == B.k AND A.t >= B.t ORDER BY (A.k, A.t); +SELECT A.k, toString(A.t, 'UTC'), A.a, B.b, toString(B.t, 'UTC'), B.k FROM A ASOF JOIN B USING(k,t) ORDER BY (A.k, A.t); +SET join_algorithm = 'full_sorting_merge'; +SELECT A.k, toString(A.t, 'UTC'), A.a, B.b, toString(B.t, 'UTC'), B.k FROM A ASOF LEFT JOIN B USING(k,t) ORDER BY (A.k, A.t); +SELECT A.k, toString(A.t, 'UTC'), A.a, B.b, toString(B.t, 'UTC'), B.k FROM A ASOF INNER JOIN B ON A.k == B.k AND A.t >= B.t ORDER BY (A.k, A.t); SELECT A.k, toString(A.t, 'UTC'), A.a, B.b, toString(B.t, 'UTC'), B.k FROM A ASOF JOIN B USING(k,t) ORDER BY (A.k, A.t); DROP TABLE A; diff --git a/tests/queries/0_stateless/00927_asof_join_other_types.reference b/tests/queries/0_stateless/00927_asof_join_other_types.reference index 80c85ec1ae3..ddbc24ff925 100644 --- a/tests/queries/0_stateless/00927_asof_join_other_types.reference +++ b/tests/queries/0_stateless/00927_asof_join_other_types.reference @@ -1,27 +1,72 @@ +- 2 1 1 0 2 3 3 3 2 5 5 3 +- 2 1 1 0 2 3 3 3 2 5 5 3 +- 2 1 1 0 2 3 3 3 2 5 5 3 +- 2 1 1 0 2 3 3 3 2 5 5 3 +- +2 1 1 0 +2 3 3 3 +2 5 5 3 +- +2 1 1 0 +2 3 3 3 +2 5 5 3 +- +2 1 1 0 +2 3 3 3 +2 5 5 3 +- +2 1 1 0 +2 3 3 3 +2 5 5 3 +- 2 1970-01-01 02:00:01 1 0 2 1970-01-01 02:00:03 3 3 2 1970-01-01 02:00:05 5 3 +- +2 1970-01-01 02:00:01 1 0 +2 1970-01-01 02:00:03 3 3 +2 1970-01-01 02:00:05 5 3 +- 2 1 1 0 2 3 3 3 2 5 5 3 +- 2 1 1 0 2 3 3 3 2 5 5 3 +- 2 1 1 0 2 3 3 3 2 5 5 3 +- +2 1 1 0 +2 3 3 3 +2 5 5 3 +- +2 1 1 0 +2 3 3 3 +2 5 5 3 +- +2 1 1 0 +2 3 3 3 +2 5 5 3 +- +2 1970-01-01 02:00:00.001 1 0 +2 1970-01-01 02:00:00.003 3 3 +2 1970-01-01 02:00:00.005 5 3 +- 2 1970-01-01 02:00:00.001 1 0 2 1970-01-01 02:00:00.003 3 3 2 1970-01-01 02:00:00.005 5 3 diff --git a/tests/queries/0_stateless/00927_asof_join_other_types.sh b/tests/queries/0_stateless/00927_asof_join_other_types.sh deleted file mode 100755 index 10173a3e43f..00000000000 --- a/tests/queries/0_stateless/00927_asof_join_other_types.sh +++ /dev/null @@ -1,27 +0,0 @@ -#!/usr/bin/env bash - -set -e - -CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# shellcheck source=../shell_config.sh -. "$CURDIR"/../shell_config.sh - -for typename in "UInt32" "UInt64" "Float64" "Float32" "DateTime('Asia/Istanbul')" "Decimal32(5)" "Decimal64(5)" "Decimal128(5)" "DateTime64(3, 'Asia/Istanbul')" -do - $CLICKHOUSE_CLIENT -mn <= B.t ORDER BY (A.a, A.t); SELECT count() FROM A ASOF LEFT JOIN B ON A.a == B.b AND B.t <= A.t; SELECT A.a, A.t, B.b, B.t FROM A ASOF INNER JOIN B ON B.t <= A.t AND A.a == B.b ORDER BY (A.a, A.t); @@ -28,5 +36,8 @@ ASOF INNER JOIN (SELECT * FROM B UNION ALL SELECT 1, 3) AS B ON B.t <= A.t AND A WHERE B.t != 3 ORDER BY (A.a, A.t) ; +{% endfor -%} +{% endfor -%} + DROP TABLE A; DROP TABLE B; diff --git a/tests/queries/0_stateless/00987_distributed_stack_overflow.sql b/tests/queries/0_stateless/00987_distributed_stack_overflow.sql index 5a22ac56413..ba58713fe0e 100644 --- a/tests/queries/0_stateless/00987_distributed_stack_overflow.sql +++ b/tests/queries/0_stateless/00987_distributed_stack_overflow.sql @@ -9,10 +9,6 @@ CREATE TABLE distr (x UInt8) ENGINE = Distributed(test_shard_localhost, currentD CREATE TABLE distr0 (x UInt8) ENGINE = Distributed(test_shard_localhost, '', distr0); -- { serverError INFINITE_LOOP } CREATE TABLE distr1 (x UInt8) ENGINE = Distributed(test_shard_localhost, currentDatabase(), distr2); -CREATE TABLE distr2 (x UInt8) ENGINE = Distributed(test_shard_localhost, currentDatabase(), distr1); - -SELECT * FROM distr1; -- { serverError TOO_LARGE_DISTRIBUTED_DEPTH } -SELECT * FROM distr2; -- { serverError TOO_LARGE_DISTRIBUTED_DEPTH } +CREATE TABLE distr2 (x UInt8) ENGINE = Distributed(test_shard_localhost, currentDatabase(), distr1); -- { serverError INFINITE_LOOP } DROP TABLE distr1; -DROP TABLE distr2; diff --git a/tests/queries/0_stateless/00989_parallel_parts_loading.sql b/tests/queries/0_stateless/00989_parallel_parts_loading.sql index a05515cf756..407e124f137 100644 --- a/tests/queries/0_stateless/00989_parallel_parts_loading.sql +++ b/tests/queries/0_stateless/00989_parallel_parts_loading.sql @@ -1,5 +1,3 @@ --- Tags: no-parallel - DROP TABLE IF EXISTS mt; CREATE TABLE mt (x UInt64) ENGINE = MergeTree ORDER BY x SETTINGS parts_to_delay_insert = 100000, parts_to_throw_insert = 100000; diff --git a/tests/queries/0_stateless/00992_system_parts_race_condition_zookeeper_long.sh b/tests/queries/0_stateless/00992_system_parts_race_condition_zookeeper_long.sh index d45cc3a6871..02a739ece4a 100755 --- a/tests/queries/0_stateless/00992_system_parts_race_condition_zookeeper_long.sh +++ b/tests/queries/0_stateless/00992_system_parts_race_condition_zookeeper_long.sh @@ -41,7 +41,7 @@ function thread3() function thread4() { - while true; do $CLICKHOUSE_CLIENT -q "OPTIMIZE TABLE alter_table0 FINAL"; done + while true; do $CLICKHOUSE_CLIENT --receive_timeout=1 -q "OPTIMIZE TABLE alter_table0 FINAL" | grep -Fv "Timeout exceeded while receiving data from server"; done } function thread5() diff --git a/tests/queries/0_stateless/00997_set_index_array.sql b/tests/queries/0_stateless/00997_set_index_array.sql index 34d0f0b71ec..ed972d1a545 100644 --- a/tests/queries/0_stateless/00997_set_index_array.sql +++ b/tests/queries/0_stateless/00997_set_index_array.sql @@ -12,10 +12,12 @@ ORDER BY (primary_key); INSERT INTO set_array select - toString(intDiv(number, 1000000)) as primary_key, + toString(intDiv(number, 100000)) as primary_key, array(number) as index_array from system.numbers -limit 10000000; +limit 1000000; + +OPTIMIZE TABLE set_array FINAL; SET max_rows_to_read = 8192; diff --git a/tests/queries/0_stateless/01004_rename_deadlock.sh b/tests/queries/0_stateless/01004_rename_deadlock.sh index f0adf136e94..60d67e1744a 100755 --- a/tests/queries/0_stateless/01004_rename_deadlock.sh +++ b/tests/queries/0_stateless/01004_rename_deadlock.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: deadlock +# Tags: deadlock, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01005_rwr_shard_deadlock.sh b/tests/queries/0_stateless/01005_rwr_shard_deadlock.sh index ef352606b69..51e8cdbc8e8 100755 --- a/tests/queries/0_stateless/01005_rwr_shard_deadlock.sh +++ b/tests/queries/0_stateless/01005_rwr_shard_deadlock.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: deadlock, shard +# Tags: deadlock, shard, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01006_simpod_empty_part_single_column_write.sh b/tests/queries/0_stateless/01006_simpod_empty_part_single_column_write.sh index 16ebf2e6e54..c3ad29d33a1 100755 --- a/tests/queries/0_stateless/01006_simpod_empty_part_single_column_write.sh +++ b/tests/queries/0_stateless/01006_simpod_empty_part_single_column_write.sh @@ -18,7 +18,7 @@ ${CLICKHOUSE_CLIENT} --query="CREATE TABLE table_with_empty_part ENGINE = MergeTree() ORDER BY id PARTITION BY id -SETTINGS vertical_merge_algorithm_min_rows_to_activate=0, vertical_merge_algorithm_min_columns_to_activate=0, remove_empty_parts = 0 +SETTINGS vertical_merge_algorithm_min_rows_to_activate=0, vertical_merge_algorithm_min_columns_to_activate=0, remove_empty_parts = 0, min_bytes_for_wide_part=0, min_bytes_for_full_part_storage = 0 " diff --git a/tests/queries/0_stateless/01007_r1r2_w_r2r1_deadlock.sh b/tests/queries/0_stateless/01007_r1r2_w_r2r1_deadlock.sh index 9f4b2241732..d602ff6df3a 100755 --- a/tests/queries/0_stateless/01007_r1r2_w_r2r1_deadlock.sh +++ b/tests/queries/0_stateless/01007_r1r2_w_r2r1_deadlock.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: deadlock +# Tags: deadlock, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01014_lazy_database_concurrent_recreate_reattach_and_show_tables.sh b/tests/queries/0_stateless/01014_lazy_database_concurrent_recreate_reattach_and_show_tables.sh index 0a6888a5c69..e4b3a31b13f 100755 --- a/tests/queries/0_stateless/01014_lazy_database_concurrent_recreate_reattach_and_show_tables.sh +++ b/tests/queries/0_stateless/01014_lazy_database_concurrent_recreate_reattach_and_show_tables.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -89,7 +89,7 @@ ${CLICKHOUSE_CLIENT} -n -q " " -TIMEOUT=30 +TIMEOUT=20 timeout $TIMEOUT bash -c recreate_lazy_func1 2> /dev/null & timeout $TIMEOUT bash -c recreate_lazy_func2 2> /dev/null & diff --git a/tests/queries/0_stateless/01035_concurrent_move_partition_from_table_zookeeper.sh b/tests/queries/0_stateless/01035_concurrent_move_partition_from_table_zookeeper.sh index 06a460f3600..59a1e3cb980 100755 --- a/tests/queries/0_stateless/01035_concurrent_move_partition_from_table_zookeeper.sh +++ b/tests/queries/0_stateless/01035_concurrent_move_partition_from_table_zookeeper.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: zookeeper, no-parallel, no-fasttest +# Tags: zookeeper, no-fasttest, long, no-parallel set -e @@ -61,7 +61,7 @@ export -f thread3; export -f thread4; export -f thread5; -TIMEOUT=30 +TIMEOUT=20 timeout $TIMEOUT bash -c thread1 2> /dev/null & timeout $TIMEOUT bash -c thread2 2> /dev/null & diff --git a/tests/queries/0_stateless/01036_no_superfluous_dict_reload_on_create_database.sql b/tests/queries/0_stateless/01036_no_superfluous_dict_reload_on_create_database.sql index 9040d7b3231..d0841124706 100644 --- a/tests/queries/0_stateless/01036_no_superfluous_dict_reload_on_create_database.sql +++ b/tests/queries/0_stateless/01036_no_superfluous_dict_reload_on_create_database.sql @@ -1,35 +1,25 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS dict_db_01036; -CREATE DATABASE dict_db_01036; - -CREATE TABLE dict_db_01036.dict_data (key UInt64, val UInt64) Engine=Memory(); -CREATE DICTIONARY dict_db_01036.dict +CREATE TABLE dict_data (key UInt64, val UInt64) Engine=Memory(); +CREATE DICTIONARY dict ( key UInt64 DEFAULT 0, val UInt64 DEFAULT 10 ) PRIMARY KEY key -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_data' PASSWORD '' DB 'dict_db_01036')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_data' PASSWORD '' DB currentDatabase())) LIFETIME(MIN 0 MAX 0) LAYOUT(FLAT()); -SELECT query_count FROM system.dictionaries WHERE database = 'dict_db_01036' AND name = 'dict'; -SELECT dictGetUInt64('dict_db_01036.dict', 'val', toUInt64(0)); -SELECT query_count FROM system.dictionaries WHERE database = 'dict_db_01036' AND name = 'dict'; +SELECT query_count FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; +SELECT dictGetUInt64('dict', 'val', toUInt64(0)); +SELECT query_count FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; SELECT 'SYSTEM RELOAD DICTIONARY'; -SYSTEM RELOAD DICTIONARY dict_db_01036.dict; -SELECT query_count FROM system.dictionaries WHERE database = 'dict_db_01036' AND name = 'dict'; -SELECT dictGetUInt64('dict_db_01036.dict', 'val', toUInt64(0)); -SELECT query_count FROM system.dictionaries WHERE database = 'dict_db_01036' AND name = 'dict'; +SYSTEM RELOAD DICTIONARY dict; +SELECT query_count FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; +SELECT dictGetUInt64('dict', 'val', toUInt64(0)); +SELECT query_count FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; SELECT 'CREATE DATABASE'; DROP DATABASE IF EXISTS empty_db_01036; -CREATE DATABASE empty_db_01036; -SELECT query_count FROM system.dictionaries WHERE database = 'dict_db_01036' AND name = 'dict'; - -DROP DICTIONARY dict_db_01036.dict; -DROP TABLE dict_db_01036.dict_data; -DROP DATABASE dict_db_01036; -DROP DATABASE empty_db_01036; +CREATE DATABASE IF NOT EXISTS empty_db_01036; +SELECT query_count FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; diff --git a/tests/queries/0_stateless/01037_polygon_dicts_correctness_all.ans b/tests/queries/0_stateless/01037_polygon_dicts_correctness_all.ans index 6e31edbdd40..0a3f4123eb8 100644 --- a/tests/queries/0_stateless/01037_polygon_dicts_correctness_all.ans +++ b/tests/queries/0_stateless/01037_polygon_dicts_correctness_all.ans @@ -1,1000 +1,1000 @@ -dictGet test_01037.dict_array (29.5699,2.50068) 101 -dictGet test_01037.dict_array (29.5796,1.55456) 101 -dictGet test_01037.dict_array (29.5796,2.36864) 101 -dictGet test_01037.dict_array (29.5844,1.59626) 101 -dictGet test_01037.dict_array (29.5886,4.03321) 101 -dictGet test_01037.dict_array (29.5914,3.02628) 101 -dictGet test_01037.dict_array (29.5926,-0.0965169) 101 -dictGet test_01037.dict_array (29.5968,2.37773) 101 -dictGet test_01037.dict_array (29.5984,0.755853) 101 -dictGet test_01037.dict_array (29.6066,3.47173) 101 -dictGet test_01037.dict_array (29.6085,-1.26007) 101 -dictGet test_01037.dict_array (29.6131,0.246565) 101 -dictGet test_01037.dict_array (29.6157,-0.266687) 101 -dictGet test_01037.dict_array (29.6164,2.94674) 101 -dictGet test_01037.dict_array (29.6195,-0.591941) 101 -dictGet test_01037.dict_array (29.6231,1.54818) 101 -dictGet test_01037.dict_array (29.6379,0.764114) 101 -dictGet test_01037.dict_array (29.6462,-0.772059) 934570 -dictGet test_01037.dict_array (29.6579,-1.07336) 101 -dictGet test_01037.dict_array (29.6618,-0.271842) 101 -dictGet test_01037.dict_array (29.6629,-0.303602) 101 -dictGet test_01037.dict_array (29.6659,-0.782823) 934570 -dictGet test_01037.dict_array (29.6736,-0.113832) 101 -dictGet test_01037.dict_array (29.6759,3.02905) 101 -dictGet test_01037.dict_array (29.6778,3.71898) 101 -dictGet test_01037.dict_array (29.6796,1.10433) 101 -dictGet test_01037.dict_array (29.6809,2.13677) 101 -dictGet test_01037.dict_array (29.6935,4.11894) 101 -dictGet test_01037.dict_array (29.6991,-1.4458199999999999) 101 -dictGet test_01037.dict_array (29.6997,3.17297) 101 -dictGet test_01037.dict_array (29.7043,3.6145899999999997) 101 -dictGet test_01037.dict_array (29.7065,3.24885) 101 -dictGet test_01037.dict_array (29.7126,0.28108) 101 -dictGet test_01037.dict_array (29.7192,0.174273) 101 -dictGet test_01037.dict_array (29.7217,-0.523481) 934570 -dictGet test_01037.dict_array (29.7271,1.67967) 101 -dictGet test_01037.dict_array (29.7311,4.12444) 101 -dictGet test_01037.dict_array (29.7347,1.88378) 101 -dictGet test_01037.dict_array (29.7358,0.67944) 101 -dictGet test_01037.dict_array (29.7366,-0.2973) 101 -dictGet test_01037.dict_array (29.7446,0.646536) 101 -dictGet test_01037.dict_array (29.7453,-0.567963) 101 -dictGet test_01037.dict_array (29.764,4.04217) 101 -dictGet test_01037.dict_array (29.7655,1.51372) 101 -dictGet test_01037.dict_array (29.7744,1.12435) 101 -dictGet test_01037.dict_array (29.7774,-0.0681196) 101 -dictGet test_01037.dict_array (29.7784,1.54864) 101 -dictGet test_01037.dict_array (29.7785,2.24139) 101 -dictGet test_01037.dict_array (29.7922,0.220808) 101 -dictGet test_01037.dict_array (29.7936,2.37709) 101 -dictGet test_01037.dict_array (29.8008,0.948536) 101 -dictGet test_01037.dict_array (29.8115,0.201227) 101 -dictGet test_01037.dict_array (29.814,0.149601) 101 -dictGet test_01037.dict_array (29.8193,-1.35858) 101 -dictGet test_01037.dict_array (29.8201,0.965518) 101 -dictGet test_01037.dict_array (29.8265,-0.727286) 101 -dictGet test_01037.dict_array (29.8277,-0.531746) 101 -dictGet test_01037.dict_array (29.8289,3.63009) 101 -dictGet test_01037.dict_array (29.8548,0.838047) 101 -dictGet test_01037.dict_array (29.8641,-0.845265) 101 -dictGet test_01037.dict_array (29.8649,0.0562212) 101 -dictGet test_01037.dict_array (29.8701,-1.02045) 101 -dictGet test_01037.dict_array (29.8733,2.76654) 101 -dictGet test_01037.dict_array (29.876,0.555475) 101 -dictGet test_01037.dict_array (29.8794,-0.800108) 101 -dictGet test_01037.dict_array (29.8813,2.7426399999999997) 101 -dictGet test_01037.dict_array (29.897100000000002,2.66193) 101 -dictGet test_01037.dict_array (29.908,4.01339) 101 -dictGet test_01037.dict_array (29.9165,-1.08246) 101 -dictGet test_01037.dict_array (29.9201,-0.420861) 101 -dictGet test_01037.dict_array (29.9217,3.03778) 101 -dictGet test_01037.dict_array (29.9355,0.773833) 101 -dictGet test_01037.dict_array (29.947,3.76517) 101 -dictGet test_01037.dict_array (29.9518,-0.60557) 101 -dictGet test_01037.dict_array (29.9564,-0.600163) 101 -dictGet test_01037.dict_array (29.959600000000002,4.16591) 101 -dictGet test_01037.dict_array (29.9615,-1.33708) 101 -dictGet test_01037.dict_array (29.9699,-0.392375) 101 -dictGet test_01037.dict_array (29.9776,1.04552) 101 -dictGet test_01037.dict_array (29.9784,4.02756) 101 -dictGet test_01037.dict_array (29.9819,4.00597) 101 -dictGet test_01037.dict_array (29.9826,1.2816100000000001) 101 -dictGet test_01037.dict_array (30.0026,2.76257) 101 -dictGet test_01037.dict_array (30.0126,3.68255) 101 -dictGet test_01037.dict_array (30.0131,0.796576) 101 -dictGet test_01037.dict_array (30.018,1.16523) 101 -dictGet test_01037.dict_array (30.0261,-0.210653) 101 -dictGet test_01037.dict_array (30.0472,-1.11007) 101 -dictGet test_01037.dict_array (30.0542,-0.479585) 101 -dictGet test_01037.dict_array (30.0613,1.6278000000000001) 101 -dictGet test_01037.dict_array (30.0617,-0.0551152) 101 -dictGet test_01037.dict_array (30.0637,2.62066) 101 -dictGet test_01037.dict_array (30.0721,1.6424400000000001) 101 -dictGet test_01037.dict_array (30.0769,-0.402636) 101 -dictGet test_01037.dict_array (30.0791,-0.277435) 101 -dictGet test_01037.dict_array (30.0931,0.0327512) 101 -dictGet test_01037.dict_array (30.1059,3.52623) 101 -dictGet test_01037.dict_array (30.1103,0.865466) 101 -dictGet test_01037.dict_array (30.1115,2.95243) 101 -dictGet test_01037.dict_array (30.1144,1.71029) 101 -dictGet test_01037.dict_array (30.1311,-0.864751) 101 -dictGet test_01037.dict_array (30.1336,-0.851386) 101 -dictGet test_01037.dict_array (30.1393,3.89901) 101 -dictGet test_01037.dict_array (30.1456,-0.531898) 101 -dictGet test_01037.dict_array (30.1492,2.07833) 101 -dictGet test_01037.dict_array (30.1575,2.43856) 101 -dictGet test_01037.dict_array (30.1682,1.19771) 101 -dictGet test_01037.dict_array (30.1716,3.9853300000000003) 101 -dictGet test_01037.dict_array (30.1849,2.78374) 101 -dictGet test_01037.dict_array (30.1866,0.65658) 101 -dictGet test_01037.dict_array (30.1885,1.56943) 101 -dictGet test_01037.dict_array (30.1959,-1.38202) 101 -dictGet test_01037.dict_array (30.1999,1.58413) 101 -dictGet test_01037.dict_array (30.2024,0.713081) 101 -dictGet test_01037.dict_array (30.2054,0.620143) 101 -dictGet test_01037.dict_array (30.2091,1.51641) 101 -dictGet test_01037.dict_array (30.2124,-0.331782) 101 -dictGet test_01037.dict_array (30.226,3.03527) 101 -dictGet test_01037.dict_array (30.2261,3.18486) 101 -dictGet test_01037.dict_array (30.2288,2.48407) 101 -dictGet test_01037.dict_array (30.2345,3.7462400000000002) 101 -dictGet test_01037.dict_array (30.2375,0.62046) 101 -dictGet test_01037.dict_array (30.2425,-0.472914) 101 -dictGet test_01037.dict_array (30.247,3.95863) 101 -dictGet test_01037.dict_array (30.2494,-0.305093) 101 -dictGet test_01037.dict_array (30.2499,2.54337) 101 -dictGet test_01037.dict_array (30.2606,2.16644) 101 -dictGet test_01037.dict_array (30.2672,3.94847) 101 -dictGet test_01037.dict_array (30.2709,-0.136264) 101 -dictGet test_01037.dict_array (30.2764,1.18654) 101 -dictGet test_01037.dict_array (30.2765,1.20383) 101 -dictGet test_01037.dict_array (30.2839,1.05762) 101 -dictGet test_01037.dict_array (30.286,0.469327) 101 -dictGet test_01037.dict_array (30.2927,3.1693) 101 -dictGet test_01037.dict_array (30.2935,3.49854) 101 -dictGet test_01037.dict_array (30.307,0.312338) 101 -dictGet test_01037.dict_array (30.3085,1.07791) 101 -dictGet test_01037.dict_array (30.3139,2.77248) 101 -dictGet test_01037.dict_array (30.314,0.822823) 101 -dictGet test_01037.dict_array (30.3227,-0.587351) 101 -dictGet test_01037.dict_array (30.332,1.00174) 101 -dictGet test_01037.dict_array (30.3388,0.844148) 101 -dictGet test_01037.dict_array (30.3485,0.561902) 101 -dictGet test_01037.dict_array (30.3497,0.180362) 101 -dictGet test_01037.dict_array (30.361,4.13016) 101 -dictGet test_01037.dict_array (30.3623,-0.0484027) 101 -dictGet test_01037.dict_array (30.3638,3.9845800000000002) 101 -dictGet test_01037.dict_array (30.3853,3.16051) 101 -dictGet test_01037.dict_array (30.3974,2.6617800000000003) 101 -dictGet test_01037.dict_array (30.4002,-1.15886) 101 -dictGet test_01037.dict_array (30.4008,-0.387015) 101 -dictGet test_01037.dict_array (30.4018,1.86493) 101 -dictGet test_01037.dict_array (30.4239,1.16818) 101 -dictGet test_01037.dict_array (30.4363,3.63938) 101 -dictGet test_01037.dict_array (30.4377,-0.81315) 101 -dictGet test_01037.dict_array (30.4391,3.54703) 101 -dictGet test_01037.dict_array (30.4424,-1.39435) 101 -dictGet test_01037.dict_array (30.4441,2.8463000000000003) 101 -dictGet test_01037.dict_array (30.4517,3.28117) 101 -dictGet test_01037.dict_array (30.4658,2.6928) 101 -dictGet test_01037.dict_array (30.4734,2.66161) 101 -dictGet test_01037.dict_array (30.4799,-1.07578) 101 -dictGet test_01037.dict_array (30.4837,-1.02486) 101 -dictGet test_01037.dict_array (30.485,1.06326) 101 -dictGet test_01037.dict_array (30.495,1.12306) 101 -dictGet test_01037.dict_array (30.501,2.27264) 101 -dictGet test_01037.dict_array (30.5027,1.99382) 101 -dictGet test_01037.dict_array (30.5194,-1.03943) 101 -dictGet test_01037.dict_array (30.5239,1.04328) 101 -dictGet test_01037.dict_array (30.528,3.82041) 101 -dictGet test_01037.dict_array (30.5299,-0.715248) 101 -dictGet test_01037.dict_array (30.5331,1.19603) 101 -dictGet test_01037.dict_array (30.535800000000002,2.71485) 101 -dictGet test_01037.dict_array (30.5405,0.804694) 101 -dictGet test_01037.dict_array (30.542,1.23739) 101 -dictGet test_01037.dict_array (30.5432,4.04189) 101 -dictGet test_01037.dict_array (30.5457,-0.956121) 101 -dictGet test_01037.dict_array (30.5506,3.07443) 101 -dictGet test_01037.dict_array (30.5539,3.87084) 101 -dictGet test_01037.dict_array (30.5578,3.78837) 101 -dictGet test_01037.dict_array (30.5588,0.966135) 101 -dictGet test_01037.dict_array (30.5637,2.5605) 101 -dictGet test_01037.dict_array (30.5647,-1.27328) 101 -dictGet test_01037.dict_array (30.5656,-0.0581332) 101 -dictGet test_01037.dict_array (30.5715,0.65755) 101 -dictGet test_01037.dict_array (30.5727,3.01604) 101 -dictGet test_01037.dict_array (30.5729,-0.976857) 101 -dictGet test_01037.dict_array (30.5751,0.60204) 101 -dictGet test_01037.dict_array (30.5854,3.02473) 101 -dictGet test_01037.dict_array (30.5866,0.174099) 101 -dictGet test_01037.dict_array (30.5947,0.875193) 101 -dictGet test_01037.dict_array (30.5992,-0.403901) 101 -dictGet test_01037.dict_array (30.6002,4.18891) 101 -dictGet test_01037.dict_array (30.6025,0.217712) 101 -dictGet test_01037.dict_array (30.6054,0.927203) 101 -dictGet test_01037.dict_array (30.6075,3.79359) 101 -dictGet test_01037.dict_array (30.6159,3.82773) 101 -dictGet test_01037.dict_array (30.627,3.84039) 101 -dictGet test_01037.dict_array (30.6308,0.77517) 101 -dictGet test_01037.dict_array (30.6338,0.179565) 101 -dictGet test_01037.dict_array (30.6461,1.3293599999999999) 101 -dictGet test_01037.dict_array (30.6674,-0.424547) 101 -dictGet test_01037.dict_array (30.669,1.76539) 101 -dictGet test_01037.dict_array (30.6788,4.01239) 101 -dictGet test_01037.dict_array (30.6864,3.59158) 101 -dictGet test_01037.dict_array (30.7049,-0.875413) 101 -dictGet test_01037.dict_array (30.705,1.3307) 101 -dictGet test_01037.dict_array (30.7063,-0.473192) 101 -dictGet test_01037.dict_array (30.7075,-1.1958199999999999) 101 -dictGet test_01037.dict_array (30.7101,-0.367562) 101 -dictGet test_01037.dict_array (30.7203,2.98725) 101 -dictGet test_01037.dict_array (30.7213,2.2745699999999998) 101 -dictGet test_01037.dict_array (30.7446,-0.334144) 101 -dictGet test_01037.dict_array (30.7468,3.82967) 101 -dictGet test_01037.dict_array (30.747,-0.384779) 101 -dictGet test_01037.dict_array (30.7681,0.904198) 101 -dictGet test_01037.dict_array (30.7757,1.78743) 101 -dictGet test_01037.dict_array (30.8021,-0.479212) 101 -dictGet test_01037.dict_array (30.8079,-1.40869) 101 -dictGet test_01037.dict_array (30.8206,-0.0608489) 101 -dictGet test_01037.dict_array (30.8218,0.43909) 101 -dictGet test_01037.dict_array (30.8239,0.10014) 101 -dictGet test_01037.dict_array (30.8282,4.15409) 101 -dictGet test_01037.dict_array (30.8288,-0.709528) 101 -dictGet test_01037.dict_array (30.8326,0.156011) 101 -dictGet test_01037.dict_array (30.8328,-1.03704) 101 -dictGet test_01037.dict_array (30.839,2.15528) 101 -dictGet test_01037.dict_array (30.8452,0.219377) 101 -dictGet test_01037.dict_array (30.8463,0.0515355) 101 -dictGet test_01037.dict_array (30.8526,2.06614) 101 -dictGet test_01037.dict_array (30.8566,0.517876) 101 -dictGet test_01037.dict_array (30.8588,-1.31738) 101 -dictGet test_01037.dict_array (30.8681,0.44207) 101 -dictGet test_01037.dict_array (30.8914,1.0072) 101 -dictGet test_01037.dict_array (30.897,0.483425) 101 -dictGet test_01037.dict_array (30.905,2.8731999999999998) 101 -dictGet test_01037.dict_array (30.9051,2.21956) 101 -dictGet test_01037.dict_array (30.9115,4.00663) 101 -dictGet test_01037.dict_array (30.9167,-0.834462) 101 -dictGet test_01037.dict_array (30.9252,-1.3289900000000001) 101 -dictGet test_01037.dict_array (30.9314,1.85384) 101 -dictGet test_01037.dict_array (30.9392,2.53236) 101 -dictGet test_01037.dict_array (30.9569,2.82038) 101 -dictGet test_01037.dict_array (30.9598,-0.641011) 101 -dictGet test_01037.dict_array (30.9601,-0.254928) 101 -dictGet test_01037.dict_array (30.9623,-1.3886) 101 -dictGet test_01037.dict_array (30.9707,0.888854) 101 -dictGet test_01037.dict_array (30.9766,2.81957) 101 -dictGet test_01037.dict_array (30.9775,2.69273) 101 -dictGet test_01037.dict_array (30.9821,0.587715) 101 -dictGet test_01037.dict_array (30.9887,4.0233) 101 -dictGet test_01037.dict_array (30.9914,0.259542) 101 -dictGet test_01037.dict_array (30.9986,-1.36832) 101 -dictGet test_01037.dict_array (31.008,0.628999) 101 -dictGet test_01037.dict_array (31.0168,-1.17462) 101 -dictGet test_01037.dict_array (31.0237,3.52547) 101 -dictGet test_01037.dict_array (31.0306,3.78522) 101 -dictGet test_01037.dict_array (31.0308,-0.72453) 101 -dictGet test_01037.dict_array (31.0463,2.41997) 101 -dictGet test_01037.dict_array (31.047,0.624184) 101 -dictGet test_01037.dict_array (31.0569,0.0706393) 5994232 -dictGet test_01037.dict_array (31.0583,1.3244099999999999) 101 -dictGet test_01037.dict_array (31.063,3.23861) 101 -dictGet test_01037.dict_array (31.068,0.695575) 101 -dictGet test_01037.dict_array (31.0687,1.85675) 101 -dictGet test_01037.dict_array (31.0692,0.254793) 101 -dictGet test_01037.dict_array (31.0766,0.828128) 101 -dictGet test_01037.dict_array (31.0833,0.0612782) 5994232 -dictGet test_01037.dict_array (31.0833,2.59748) 101 -dictGet test_01037.dict_array (31.0861,-1.3778299999999999) 101 -dictGet test_01037.dict_array (31.0874,3.07258) 101 -dictGet test_01037.dict_array (31.0882,1.4882) 101 -dictGet test_01037.dict_array (31.0924,3.42242) 101 -dictGet test_01037.dict_array (31.0927,2.67448) 101 -dictGet test_01037.dict_array (31.0936,1.12292) 101 -dictGet test_01037.dict_array (31.0952,-0.336928) 101 -dictGet test_01037.dict_array (31.0978,3.48482) 101 -dictGet test_01037.dict_array (31.1107,3.7513199999999998) 101 -dictGet test_01037.dict_array (31.1156,1.19171) 101 -dictGet test_01037.dict_array (31.1176,0.223509) 5994232 -dictGet test_01037.dict_array (31.1249,0.946838) 101 -dictGet test_01037.dict_array (31.1267,1.48983) 101 -dictGet test_01037.dict_array (31.138,-0.289981) 101 -dictGet test_01037.dict_array (31.1382,3.02904) 101 -dictGet test_01037.dict_array (31.1475,2.6178) 101 -dictGet test_01037.dict_array (31.1491,1.37873) 101 -dictGet test_01037.dict_array (31.1525,3.72105) 101 -dictGet test_01037.dict_array (31.1526,-1.4129800000000001) 101 -dictGet test_01037.dict_array (31.1526,-0.186457) 101 -dictGet test_01037.dict_array (31.1539,2.78789) 101 -dictGet test_01037.dict_array (31.1548,-1.08552) 101 -dictGet test_01037.dict_array (31.1567,-0.0768925) 101 -dictGet test_01037.dict_array (31.1613,1.49617) 101 -dictGet test_01037.dict_array (31.1653,1.03777) 101 -dictGet test_01037.dict_array (31.1662,3.4214700000000002) 101 -dictGet test_01037.dict_array (31.1672,-0.0813169) 101 -dictGet test_01037.dict_array (31.177,0.440843) 101 -dictGet test_01037.dict_array (31.1788,-0.737151) 101 -dictGet test_01037.dict_array (31.1856,-0.144396) 101 -dictGet test_01037.dict_array (31.1959,3.66813) 101 -dictGet test_01037.dict_array (31.1996,-0.353983) 101 -dictGet test_01037.dict_array (31.2019,2.86802) 101 -dictGet test_01037.dict_array (31.2087,2.31245) 101 -dictGet test_01037.dict_array (31.2125,3.2713200000000002) 101 -dictGet test_01037.dict_array (31.2137,-0.108129) 101 -dictGet test_01037.dict_array (31.216,3.9156) 101 -dictGet test_01037.dict_array (31.2201,-0.202141) 101 -dictGet test_01037.dict_array (31.2285,2.09058) 101 -dictGet test_01037.dict_array (31.2502,4.01526) 101 -dictGet test_01037.dict_array (31.2585,3.11524) 101 -dictGet test_01037.dict_array (31.2645,-0.620418) 101 -dictGet test_01037.dict_array (31.2684,2.74277) 101 -dictGet test_01037.dict_array (31.2821,-1.12772) 101 -dictGet test_01037.dict_array (31.2821,2.46769) 101 -dictGet test_01037.dict_array (31.2887,3.91396) 101 -dictGet test_01037.dict_array (31.295,1.49942) 101 -dictGet test_01037.dict_array (31.2997,3.46122) 101 -dictGet test_01037.dict_array (31.3017,3.3263) 101 -dictGet test_01037.dict_array (31.3022,3.16754) 101 -dictGet test_01037.dict_array (31.3048,0.364962) 101 -dictGet test_01037.dict_array (31.305,3.1967) 101 -dictGet test_01037.dict_array (31.3061,1.84303) 101 -dictGet test_01037.dict_array (31.3082,-0.173851) 101 -dictGet test_01037.dict_array (31.3315,3.90932) 101 -dictGet test_01037.dict_array (31.3351,2.80164) 101 -dictGet test_01037.dict_array (31.3388,0.168765) 5994233 -dictGet test_01037.dict_array (31.339,0.25535) 101 -dictGet test_01037.dict_array (31.3423,1.7036799999999999) 101 -dictGet test_01037.dict_array (31.349,0.386456) 101 -dictGet test_01037.dict_array (31.3558,-1.04336) 101 -dictGet test_01037.dict_array (31.3564,0.478876) 101 -dictGet test_01037.dict_array (31.3607,-0.0860507) 5994233 -dictGet test_01037.dict_array (31.3831,3.84469) 101 -dictGet test_01037.dict_array (31.3886,-0.731137) 101 -dictGet test_01037.dict_array (31.4043,-0.348907) 101 -dictGet test_01037.dict_array (31.4081,1.47391) 101 -dictGet test_01037.dict_array (31.4176,-0.583645) 101 -dictGet test_01037.dict_array (31.4177,1.36972) 101 -dictGet test_01037.dict_array (31.4182,0.958303) 101 -dictGet test_01037.dict_array (31.4199,3.1738) 101 -dictGet test_01037.dict_array (31.4221,2.74876) 101 -dictGet test_01037.dict_array (31.4301,-0.122643) 5994233 -dictGet test_01037.dict_array (31.4344,1.00661) 101 -dictGet test_01037.dict_array (31.4375,4.20304) 101 -dictGet test_01037.dict_array (31.4377,0.289608) 101 -dictGet test_01037.dict_array (31.4379,0.54744) 101 -dictGet test_01037.dict_array (31.4459,3.94945) 101 -dictGet test_01037.dict_array (31.4559,-0.345063) 101 -dictGet test_01037.dict_array (31.464,0.726129) 101 -dictGet test_01037.dict_array (31.4662,-0.299019) 5994233 -dictGet test_01037.dict_array (31.4671,1.9605299999999999) 101 -dictGet test_01037.dict_array (31.4673,-0.403676) 101 -dictGet test_01037.dict_array (31.4712,-0.237941) 5994233 -dictGet test_01037.dict_array (31.4816,0.120264) 5994233 -dictGet test_01037.dict_array (31.4875,0.323483) 101 -dictGet test_01037.dict_array (31.490099999999998,-0.338163) 101 -dictGet test_01037.dict_array (31.4932,0.517674) 101 -dictGet test_01037.dict_array (31.5112,1.9689299999999998) 101 -dictGet test_01037.dict_array (31.5122,2.92785) 101 -dictGet test_01037.dict_array (31.5151,0.166429) 101 -dictGet test_01037.dict_array (31.5174,2.94802) 101 -dictGet test_01037.dict_array (31.5182,4.18776) 101 -dictGet test_01037.dict_array (31.5238,1.18793) 101 -dictGet test_01037.dict_array (31.5271,3.07446) 101 -dictGet test_01037.dict_array (31.5393,1.58061) 101 -dictGet test_01037.dict_array (31.5421,3.13711) 101 -dictGet test_01037.dict_array (31.5479,2.39897) 101 -dictGet test_01037.dict_array (31.5519,0.99285) 101 -dictGet test_01037.dict_array (31.5685,3.47987) 101 -dictGet test_01037.dict_array (31.5959,0.437382) 101 -dictGet test_01037.dict_array (31.6003,0.194376) 101 -dictGet test_01037.dict_array (31.6026,2.15457) 101 -dictGet test_01037.dict_array (31.606,2.45365) 101 -dictGet test_01037.dict_array (31.6062,-0.453441) 101 -dictGet test_01037.dict_array (31.6107,1.35247) 101 -dictGet test_01037.dict_array (31.6155,3.85588) 101 -dictGet test_01037.dict_array (31.6222,2.03326) 101 -dictGet test_01037.dict_array (31.6231,-0.123059) 101 -dictGet test_01037.dict_array (31.6244,1.6885599999999998) 101 -dictGet test_01037.dict_array (31.6459,0.669716) 101 -dictGet test_01037.dict_array (31.6563,-0.0644741) 101 -dictGet test_01037.dict_array (31.6618,-0.551121) 101 -dictGet test_01037.dict_array (31.6725,-0.38922) 101 -dictGet test_01037.dict_array (31.6727,4.10336) 101 -dictGet test_01037.dict_array (31.6739,4.1391) 101 -dictGet test_01037.dict_array (31.6897,2.8694699999999997) 101 -dictGet test_01037.dict_array (31.6902,3.98792) 101 -dictGet test_01037.dict_array (31.6945,2.46687) 101 -dictGet test_01037.dict_array (31.6987,-1.3796) 101 -dictGet test_01037.dict_array (31.7012,2.34845) 101 -dictGet test_01037.dict_array (31.7036,0.0228348) 101 -dictGet test_01037.dict_array (31.7046,3.68111) 101 -dictGet test_01037.dict_array (31.7055,2.92556) 101 -dictGet test_01037.dict_array (31.7102,1.04532) 101 -dictGet test_01037.dict_array (31.7149,-0.443302) 101 -dictGet test_01037.dict_array (31.7195,2.99311) 101 -dictGet test_01037.dict_array (31.7274,0.166719) 101 -dictGet test_01037.dict_array (31.7565,-0.565382) 101 -dictGet test_01037.dict_array (31.7615,0.771626) 101 -dictGet test_01037.dict_array (31.7739,1.8970099999999999) 101 -dictGet test_01037.dict_array (31.7848,1.2623199999999999) 101 -dictGet test_01037.dict_array (31.7912,-0.788599) 101 -dictGet test_01037.dict_array (31.8011,2.65853) 101 -dictGet test_01037.dict_array (31.8032,-0.0590108) 101 -dictGet test_01037.dict_array (31.8038,1.9618799999999998) 101 -dictGet test_01037.dict_array (31.8098,-1.46851) 101 -dictGet test_01037.dict_array (31.8131,3.41982) 101 -dictGet test_01037.dict_array (31.8169,3.31059) 101 -dictGet test_01037.dict_array (31.8202,-0.193692) 101 -dictGet test_01037.dict_array (31.8306,1.57586) 101 -dictGet test_01037.dict_array (31.8382,-0.787948) 101 -dictGet test_01037.dict_array (31.8433,2.49692) 101 -dictGet test_01037.dict_array (31.8436,2.41851) 101 -dictGet test_01037.dict_array (31.8563,-1.10787) 101 -dictGet test_01037.dict_array (31.8683,0.996504) 101 -dictGet test_01037.dict_array (31.8693,-0.828142) 101 -dictGet test_01037.dict_array (31.8723,1.08929) 101 -dictGet test_01037.dict_array (31.8737,0.881127) 101 -dictGet test_01037.dict_array (31.8881,-0.58441) 101 -dictGet test_01037.dict_array (31.9011,0.121349) 101 -dictGet test_01037.dict_array (31.9066,2.13045) 101 -dictGet test_01037.dict_array (31.9142,1.03368) 101 -dictGet test_01037.dict_array (31.9155,3.38363) 101 -dictGet test_01037.dict_array (31.9168,1.3166) 101 -dictGet test_01037.dict_array (31.9185,-1.11879) 101 -dictGet test_01037.dict_array (31.9186,-0.647948) 101 -dictGet test_01037.dict_array (31.9311,3.96928) 101 -dictGet test_01037.dict_array (31.9335,1.47048) 101 -dictGet test_01037.dict_array (31.9443,-1.36175) 101 -dictGet test_01037.dict_array (31.9481,2.34231) 101 -dictGet test_01037.dict_array (31.9526,1.36565) 101 -dictGet test_01037.dict_array (31.9629,2.5208399999999997) 101 -dictGet test_01037.dict_array (31.9765,0.975783) 101 -dictGet test_01037.dict_array (31.9923,3.31773) 101 -dictGet test_01037.dict_array (31.9994,0.972816) 101 -dictGet test_01037.dict_array (32.001,3.47425) 101 -dictGet test_01037.dict_array (32.0127,2.13874) 101 -dictGet test_01037.dict_array (32.0244,3.2092) 101 -dictGet test_01037.dict_array (32.029,1.18039) 101 -dictGet test_01037.dict_array (32.0315,0.566073) 101 -dictGet test_01037.dict_array (32.0354,1.0766499999999999) 101 -dictGet test_01037.dict_array (32.0399,-1.11576) 101 -dictGet test_01037.dict_array (32.053,2.16849) 101 -dictGet test_01037.dict_array (32.0542,0.042328) 101 -dictGet test_01037.dict_array (32.0576,2.47001) 101 -dictGet test_01037.dict_array (32.061,3.7498899999999997) 101 -dictGet test_01037.dict_array (32.0623,1.25134) 101 -dictGet test_01037.dict_array (32.0626,1.9611399999999999) 101 -dictGet test_01037.dict_array (32.0666,-0.0904247) 101 -dictGet test_01037.dict_array (32.0681,2.28442) 101 -dictGet test_01037.dict_array (32.0692,1.50869) 101 -dictGet test_01037.dict_array (32.0724,4.03314) 101 -dictGet test_01037.dict_array (32.0729,-0.064324) 101 -dictGet test_01037.dict_array (32.079,0.293758) 101 -dictGet test_01037.dict_array (32.0847,-1.19814) 101 -dictGet test_01037.dict_array (32.0974,-0.91927) 101 -dictGet test_01037.dict_array (32.0979,-0.736979) 101 -dictGet test_01037.dict_array (32.106,-1.33063) 101 -dictGet test_01037.dict_array (32.1189,0.246715) 101 -dictGet test_01037.dict_array (32.1207,4.00883) 101 -dictGet test_01037.dict_array (32.1396,1.12402) 101 -dictGet test_01037.dict_array (32.1413,1.5668) 101 -dictGet test_01037.dict_array (32.143,1.35559) 101 -dictGet test_01037.dict_array (32.1538,1.32881) 101 -dictGet test_01037.dict_array (32.1549,4.06552) 101 -dictGet test_01037.dict_array (32.1555,-0.79275) 101 -dictGet test_01037.dict_array (32.163,1.17733) 101 -dictGet test_01037.dict_array (32.1634,2.94273) 101 -dictGet test_01037.dict_array (32.1644,1.85666) 101 -dictGet test_01037.dict_array (32.1745,0.435458) 101 -dictGet test_01037.dict_array (32.1765,1.65149) 101 -dictGet test_01037.dict_array (32.1893,2.08924) 101 -dictGet test_01037.dict_array (32.2024,0.222191) 101 -dictGet test_01037.dict_array (32.2107,1.34379) 101 -dictGet test_01037.dict_array (32.2109,3.9018699999999997) 101 -dictGet test_01037.dict_array (32.2123,1.85233) 101 -dictGet test_01037.dict_array (32.2144,3.72534) 101 -dictGet test_01037.dict_array (32.2218,2.5386699999999998) 101 -dictGet test_01037.dict_array (32.2279,2.84267) 101 -dictGet test_01037.dict_array (32.2345,3.33295) 101 -dictGet test_01037.dict_array (32.2435,3.85283) 101 -dictGet test_01037.dict_array (32.2527,-0.480608) 101 -dictGet test_01037.dict_array (32.2566,-0.837882) 101 -dictGet test_01037.dict_array (32.2627,2.57708) 101 -dictGet test_01037.dict_array (32.2733,0.244931) 101 -dictGet test_01037.dict_array (32.2761,4.05808) 101 -dictGet test_01037.dict_array (32.2764,3.78472) 101 -dictGet test_01037.dict_array (32.2814,-1.26011) 101 -dictGet test_01037.dict_array (32.2861,3.02427) 101 -dictGet test_01037.dict_array (32.2924,0.928609) 101 -dictGet test_01037.dict_array (32.2963,-0.78543) 101 -dictGet test_01037.dict_array (32.3039,3.21175) 101 -dictGet test_01037.dict_array (32.3107,0.698287) 101 -dictGet test_01037.dict_array (32.3138,0.0595677) 101 -dictGet test_01037.dict_array (32.3339,0.707056) 101 -dictGet test_01037.dict_array (32.3351,0.415474) 101 -dictGet test_01037.dict_array (32.342,-0.681023) 101 -dictGet test_01037.dict_array (32.3463,1.83196) 101 -dictGet test_01037.dict_array (32.3494,2.43799) 101 -dictGet test_01037.dict_array (32.3524,3.47049) 101 -dictGet test_01037.dict_array (32.3531,2.33115) 101 -dictGet test_01037.dict_array (32.3602,0.116106) 101 -dictGet test_01037.dict_array (32.3612,1.1598) 101 -dictGet test_01037.dict_array (32.3689,3.34847) 101 -dictGet test_01037.dict_array (32.3695,0.734055) 101 -dictGet test_01037.dict_array (32.3825,3.85017) 101 -dictGet test_01037.dict_array (32.3835,-1.25491) 101 -dictGet test_01037.dict_array (32.4018,-0.728568) 101 -dictGet test_01037.dict_array (32.4044,2.96727) 101 -dictGet test_01037.dict_array (32.4101,2.9988) 101 -dictGet test_01037.dict_array (32.417,-1.12908) 101 -dictGet test_01037.dict_array (32.4172,4.1952) 101 -dictGet test_01037.dict_array (32.4239,2.49512) 101 -dictGet test_01037.dict_array (32.4258,4.05137) 101 -dictGet test_01037.dict_array (32.4264,-0.427357) 101 -dictGet test_01037.dict_array (32.4274,3.59377) 101 -dictGet test_01037.dict_array (32.4286,-1.24757) 101 -dictGet test_01037.dict_array (32.4294,3.0665) 101 -dictGet test_01037.dict_array (32.4333,-0.353347) 101 -dictGet test_01037.dict_array (32.4391,3.64421) 101 -dictGet test_01037.dict_array (32.4401,3.70635) 101 -dictGet test_01037.dict_array (32.45,1.68918) 101 -dictGet test_01037.dict_array (32.4507,-0.133471) 101 -dictGet test_01037.dict_array (32.4592,0.976458) 101 -dictGet test_01037.dict_array (32.4595,1.89135) 101 -dictGet test_01037.dict_array (32.4604,0.280248) 101 -dictGet test_01037.dict_array (32.4835,0.472731) 101 -dictGet test_01037.dict_array (32.4855,2.01938) 101 -dictGet test_01037.dict_array (32.4872,2.01697) 101 -dictGet test_01037.dict_array (32.4911,0.613106) 101 -dictGet test_01037.dict_array (32.4918,2.17834) 101 -dictGet test_01037.dict_array (32.4947,2.34595) 101 -dictGet test_01037.dict_array (32.5035,2.92234) 101 -dictGet test_01037.dict_array (32.5132,-0.331206) 101 -dictGet test_01037.dict_array (32.5156,-0.412604) 7652581 -dictGet test_01037.dict_array (32.5158,2.9067499999999997) 101 -dictGet test_01037.dict_array (32.5249,2.44519) 101 -dictGet test_01037.dict_array (32.5293,-0.790952) 101 -dictGet test_01037.dict_array (32.5319,3.96854) 101 -dictGet test_01037.dict_array (32.5518,3.6093) 101 -dictGet test_01037.dict_array (32.5541,3.5225400000000002) 101 -dictGet test_01037.dict_array (32.5569,0.816123) 101 -dictGet test_01037.dict_array (32.5646,1.9775) 101 -dictGet test_01037.dict_array (32.5733,3.81271) 101 -dictGet test_01037.dict_array (32.5767,0.948327) 101 -dictGet test_01037.dict_array (32.5971,1.76179) 101 -dictGet test_01037.dict_array (32.6035,-0.716157) 101 -dictGet test_01037.dict_array (32.6087,4.21614) 101 -dictGet test_01037.dict_array (32.6171,0.024481) 101 -dictGet test_01037.dict_array (32.6189,-0.775391) 101 -dictGet test_01037.dict_array (32.6198,2.92081) 101 -dictGet test_01037.dict_array (32.621,-0.970784) 101 -dictGet test_01037.dict_array (32.6266,0.650009) 101 -dictGet test_01037.dict_array (32.6315,2.15144) 101 -dictGet test_01037.dict_array (32.6385,-0.436803) 101 -dictGet test_01037.dict_array (32.6449,-0.191292) 101 -dictGet test_01037.dict_array (32.6535,2.10385) 101 -dictGet test_01037.dict_array (32.6592,3.49973) 101 -dictGet test_01037.dict_array (32.6598,2.5980600000000003) 101 -dictGet test_01037.dict_array (32.6612,2.95681) 101 -dictGet test_01037.dict_array (32.6636,-0.57235) 101 -dictGet test_01037.dict_array (32.669,-0.382702) 101 -dictGet test_01037.dict_array (32.6752,1.30748) 101 -dictGet test_01037.dict_array (32.6811,2.9559800000000003) 101 -dictGet test_01037.dict_array (32.6821,0.57336) 101 -dictGet test_01037.dict_array (32.6828,3.91304) 101 -dictGet test_01037.dict_array (32.6979,3.96868) 101 -dictGet test_01037.dict_array (32.6983,3.15784) 101 -dictGet test_01037.dict_array (32.7122,0.794293) 101 -dictGet test_01037.dict_array (32.7131,-0.847256) 101 -dictGet test_01037.dict_array (32.7219,0.883461) 101 -dictGet test_01037.dict_array (32.7228,1.78808) 101 -dictGet test_01037.dict_array (32.7273,-0.206908) 101 -dictGet test_01037.dict_array (32.7292,0.259331) 101 -dictGet test_01037.dict_array (32.7304,-1.38317) 101 -dictGet test_01037.dict_array (32.7353,1.01601) 101 -dictGet test_01037.dict_array (32.7354,4.17574) 101 -dictGet test_01037.dict_array (32.7357,-0.190194) 101 -dictGet test_01037.dict_array (32.7465,-1.37598) 101 -dictGet test_01037.dict_array (32.7494,-0.275675) 101 -dictGet test_01037.dict_array (32.7514,0.128951) 101 -dictGet test_01037.dict_array (32.753,3.44207) 101 -dictGet test_01037.dict_array (32.7686,2.11713) 101 -dictGet test_01037.dict_array (32.7694,1.47159) 101 -dictGet test_01037.dict_array (32.7768,0.0401042) 101 -dictGet test_01037.dict_array (32.781,-1.34283) 101 -dictGet test_01037.dict_array (32.7814,1.73876) 101 -dictGet test_01037.dict_array (32.7856,-1.06363) 101 -dictGet test_01037.dict_array (32.792699999999996,-1.1255600000000001) 101 -dictGet test_01037.dict_array (32.7941,-0.645447) 101 -dictGet test_01037.dict_array (32.7946,1.48889) 101 -dictGet test_01037.dict_array (32.797,0.791753) 101 -dictGet test_01037.dict_array (32.7982,-0.537798) 101 -dictGet test_01037.dict_array (32.8091,2.3611) 101 -dictGet test_01037.dict_array (32.81,1.7130800000000002) 101 -dictGet test_01037.dict_array (32.8174,-0.288322) 101 -dictGet test_01037.dict_array (32.823,1.6546699999999999) 101 -dictGet test_01037.dict_array (32.8233,1.62108) 101 -dictGet test_01037.dict_array (32.8428,-0.400045) 101 -dictGet test_01037.dict_array (32.8479,2.13598) 101 -dictGet test_01037.dict_array (32.8524,0.199902) 101 -dictGet test_01037.dict_array (32.8543,3.23553) 101 -dictGet test_01037.dict_array (32.8562,1.31371) 101 -dictGet test_01037.dict_array (32.87,1.44256) 101 -dictGet test_01037.dict_array (32.8789,2.38192) 101 -dictGet test_01037.dict_array (32.8812,2.20734) 5999168 -dictGet test_01037.dict_array (32.8815,-0.54427) 101 -dictGet test_01037.dict_array (32.8853,2.4859) 5999168 -dictGet test_01037.dict_array (32.8909,0.513964) 101 -dictGet test_01037.dict_array (32.9035,2.38999) 101 -dictGet test_01037.dict_array (32.9097,2.48131) 5999168 -dictGet test_01037.dict_array (32.928,-0.943269) 101 -dictGet test_01037.dict_array (32.9322,1.13165) 101 -dictGet test_01037.dict_array (32.9348,1.22606) 101 -dictGet test_01037.dict_array (32.9417,3.77998) 101 -dictGet test_01037.dict_array (32.9428,3.11936) 101 -dictGet test_01037.dict_array (32.9482,1.18092) 101 -dictGet test_01037.dict_array (32.9506,0.0609364) 101 -dictGet test_01037.dict_array (32.953,-0.828308) 101 -dictGet test_01037.dict_array (32.9593,3.5209099999999998) 101 -dictGet test_01037.dict_array (32.9617,2.07711) 5999168 -dictGet test_01037.dict_array (32.966,0.693749) 101 -dictGet test_01037.dict_array (32.9668,-0.716432) 101 -dictGet test_01037.dict_array (32.9702,1.98555) 101 -dictGet test_01037.dict_array (32.9782,1.73819) 101 -dictGet test_01037.dict_array (32.9805,3.71151) 101 -dictGet test_01037.dict_array (32.9821,2.97225) 101 -dictGet test_01037.dict_array (32.995,-0.830301) 101 -dictGet test_01037.dict_array (33.0234,0.770848) 101 -dictGet test_01037.dict_array (33.0312,-0.340964) 101 -dictGet test_01037.dict_array (33.0366,-0.756795) 101 -dictGet test_01037.dict_array (33.0438,0.812871) 101 -dictGet test_01037.dict_array (33.0455,1.84843) 101 -dictGet test_01037.dict_array (33.0498,0.0913292) 101 -dictGet test_01037.dict_array (33.0506,1.53739) 101 -dictGet test_01037.dict_array (33.0554,2.4265) 101 -dictGet test_01037.dict_array (33.0741,3.61332) 101 -dictGet test_01037.dict_array (33.0765,-0.179985) 101 -dictGet test_01037.dict_array (33.087,1.46465) 101 -dictGet test_01037.dict_array (33.0906,-0.620383) 101 -dictGet test_01037.dict_array (33.1047,-1.28027) 101 -dictGet test_01037.dict_array (33.1072,1.96303) 101 -dictGet test_01037.dict_array (33.1081,-0.897874) 101 -dictGet test_01037.dict_array (33.1122,1.8950200000000001) 101 -dictGet test_01037.dict_array (33.1237,2.63993) 101 -dictGet test_01037.dict_array (33.1238,0.753963) 101 -dictGet test_01037.dict_array (33.1257,0.495668) 101 -dictGet test_01037.dict_array (33.1258,1.78341) 101 -dictGet test_01037.dict_array (33.127,2.59646) 101 -dictGet test_01037.dict_array (33.1324,-1.23742) 101 -dictGet test_01037.dict_array (33.1359,3.83491) 101 -dictGet test_01037.dict_array (33.1628,-0.379588) 101 -dictGet test_01037.dict_array (33.1679,1.25601) 101 -dictGet test_01037.dict_array (33.1688,-1.35553) 101 -dictGet test_01037.dict_array (33.181,2.10943) 101 -dictGet test_01037.dict_array (33.1871,2.81171) 101 -dictGet test_01037.dict_array (33.1877,0.771297) 101 -dictGet test_01037.dict_array (33.1883,-0.204797) 101 -dictGet test_01037.dict_array (33.1886,3.27998) 101 -dictGet test_01037.dict_array (33.1955,0.708907) 101 -dictGet test_01037.dict_array (33.2044,-0.769275) 101 -dictGet test_01037.dict_array (33.2182,3.36103) 101 -dictGet test_01037.dict_array (33.2192,3.43586) 101 -dictGet test_01037.dict_array (33.2322,-0.916753) 101 -dictGet test_01037.dict_array (33.2359,-0.81321) 101 -dictGet test_01037.dict_array (33.238,0.635072) 101 -dictGet test_01037.dict_array (33.2398,3.02588) 101 -dictGet test_01037.dict_array (33.2469,2.35698) 101 -dictGet test_01037.dict_array (33.247,2.3327) 101 -dictGet test_01037.dict_array (33.2579,2.8027100000000003) 101 -dictGet test_01037.dict_array (33.2607,0.321082) 101 -dictGet test_01037.dict_array (33.2653,0.243336) 101 -dictGet test_01037.dict_array (33.2758,0.831836) 101 -dictGet test_01037.dict_array (33.2771,0.886536) 101 -dictGet test_01037.dict_array (33.2914,1.16026) 101 -dictGet test_01037.dict_array (33.2914,1.38882) 101 -dictGet test_01037.dict_array (33.2982,-1.16604) 101 -dictGet test_01037.dict_array (33.2985,0.842556) 101 -dictGet test_01037.dict_array (33.3005,2.8338900000000002) 101 -dictGet test_01037.dict_array (33.305,0.0969475) 101 -dictGet test_01037.dict_array (33.3072,3.82163) 101 -dictGet test_01037.dict_array (33.312,3.41475) 101 -dictGet test_01037.dict_array (33.3129,2.46048) 101 -dictGet test_01037.dict_array (33.3134,3.46863) 101 -dictGet test_01037.dict_array (33.3203,2.33139) 101 -dictGet test_01037.dict_array (33.324,0.433701) 101 -dictGet test_01037.dict_array (33.3338,2.44705) 101 -dictGet test_01037.dict_array (33.337,4.06475) 101 -dictGet test_01037.dict_array (33.3469,1.08172) 101 -dictGet test_01037.dict_array (33.3538,0.717896) 101 -dictGet test_01037.dict_array (33.3618,1.37899) 101 -dictGet test_01037.dict_array (33.3698,0.547744) 101 -dictGet test_01037.dict_array (33.3705,0.957619) 101 -dictGet test_01037.dict_array (33.3821,3.07258) 101 -dictGet test_01037.dict_array (33.3881,3.0626) 101 -dictGet test_01037.dict_array (33.393,-0.816186) 101 -dictGet test_01037.dict_array (33.3945,0.869508) 101 -dictGet test_01037.dict_array (33.4001,1.24186) 101 -dictGet test_01037.dict_array (33.4008,2.34911) 101 -dictGet test_01037.dict_array (33.4166,-1.2808899999999999) 101 -dictGet test_01037.dict_array (33.4167,3.0655) 101 -dictGet test_01037.dict_array (33.4204,2.81887) 101 -dictGet test_01037.dict_array (33.4211,1.71128) 101 -dictGet test_01037.dict_array (33.4237,2.91761) 101 -dictGet test_01037.dict_array (33.4266,1.5955599999999999) 101 -dictGet test_01037.dict_array (33.4353,-0.391392) 101 -dictGet test_01037.dict_array (33.4362,-0.134658) 101 -dictGet test_01037.dict_array (33.4386,0.15396) 101 -dictGet test_01037.dict_array (33.4421,-0.50712) 101 -dictGet test_01037.dict_array (33.452,0.915829) 101 -dictGet test_01037.dict_array (33.463,-0.0882717) 101 -dictGet test_01037.dict_array (33.464,-1.00949) 101 -dictGet test_01037.dict_array (33.4692,0.954092) 101 -dictGet test_01037.dict_array (33.4716,1.9538799999999998) 101 -dictGet test_01037.dict_array (33.4756,1.85836) 101 -dictGet test_01037.dict_array (33.4859,4.0751) 101 -dictGet test_01037.dict_array (33.4899,3.54193) 101 -dictGet test_01037.dict_array (33.4935,3.49794) 101 -dictGet test_01037.dict_array (33.494,-0.983356) 101 -dictGet test_01037.dict_array (33.4955,-1.28128) 101 -dictGet test_01037.dict_array (33.4965,-0.278687) 101 -dictGet test_01037.dict_array (33.4991,0.647491) 101 -dictGet test_01037.dict_array (33.5076,2.2272) 101 -dictGet test_01037.dict_array (33.5079,-0.498199) 101 -dictGet test_01037.dict_array (33.5157,0.535034) 101 -dictGet test_01037.dict_array (33.5171,2.49677) 101 -dictGet test_01037.dict_array (33.5255,2.4447200000000002) 101 -dictGet test_01037.dict_array (33.526,4.01194) 101 -dictGet test_01037.dict_array (33.5288,0.789434) 101 -dictGet test_01037.dict_array (33.5356,-1.17671) 101 -dictGet test_01037.dict_array (33.5402,1.49152) 101 -dictGet test_01037.dict_array (33.5418,3.45757) 101 -dictGet test_01037.dict_array (33.5428,1.90712) 101 -dictGet test_01037.dict_array (33.5556,-0.55741) 101 -dictGet test_01037.dict_array (33.5564,0.876858) 101 -dictGet test_01037.dict_array (33.5567,-0.10208) 101 -dictGet test_01037.dict_array (33.5645,-0.124824) 101 -dictGet test_01037.dict_array (33.5663,3.4872) 101 -dictGet test_01037.dict_array (33.5716,-0.0107611) 101 -dictGet test_01037.dict_array (33.578,3.55714) 101 -dictGet test_01037.dict_array (33.5826,-0.49076) 101 -dictGet test_01037.dict_array (33.5909,0.773737) 101 -dictGet test_01037.dict_array (33.5958,2.9619999999999997) 5994231 -dictGet test_01037.dict_array (33.6193,-0.919755) 101 -dictGet test_01037.dict_array (33.6313,0.652132) 101 -dictGet test_01037.dict_array (33.632,0.823351) 101 -dictGet test_01037.dict_array (33.66,2.18998) 101 -dictGet test_01037.dict_array (33.6621,0.535395) 101 -dictGet test_01037.dict_array (33.6726,3.19367) 101 -dictGet test_01037.dict_array (33.6912,1.74522) 101 -dictGet test_01037.dict_array (33.705,0.706397) 101 -dictGet test_01037.dict_array (33.7076,0.7622) 101 -dictGet test_01037.dict_array (33.7112,1.70187) 101 -dictGet test_01037.dict_array (33.7246,-1.14837) 101 -dictGet test_01037.dict_array (33.7326,2.62413) 5994231 -dictGet test_01037.dict_array (33.7332,2.82137) 5994231 -dictGet test_01037.dict_array (33.7434,0.394672) 101 -dictGet test_01037.dict_array (33.7443,1.54557) 101 -dictGet test_01037.dict_array (33.7506,1.57317) 101 -dictGet test_01037.dict_array (33.7526,1.8578999999999999) 101 -dictGet test_01037.dict_array (33.766,4.15013) 101 -dictGet test_01037.dict_array (33.7834,2.41789) 101 -dictGet test_01037.dict_array (33.7864,0.230935) 101 -dictGet test_01037.dict_array (33.7965,3.05709) 101 -dictGet test_01037.dict_array (33.7998,3.32881) 101 -dictGet test_01037.dict_array (33.8003,2.97338) 5994231 -dictGet test_01037.dict_array (33.8007,-1.08962) 101 -dictGet test_01037.dict_array (33.8022,-0.139488) 101 -dictGet test_01037.dict_array (33.8065,2.70857) 5994231 -dictGet test_01037.dict_array (33.8169,-0.607788) 101 -dictGet test_01037.dict_array (33.8203,0.108512) 101 -dictGet test_01037.dict_array (33.8231,-1.03449) 101 -dictGet test_01037.dict_array (33.8312,3.49458) 101 -dictGet test_01037.dict_array (33.8342,0.297518) 101 -dictGet test_01037.dict_array (33.8352,0.165872) 101 -dictGet test_01037.dict_array (33.8354,1.87277) 101 -dictGet test_01037.dict_array (33.8371,1.60103) 101 -dictGet test_01037.dict_array (33.8387,1.9968) 101 -dictGet test_01037.dict_array (33.8403,3.5805) 101 -dictGet test_01037.dict_array (33.8414,-0.703067) 101 -dictGet test_01037.dict_array (33.844,-0.179472) 101 -dictGet test_01037.dict_array (33.8468,3.40137) 101 -dictGet test_01037.dict_array (33.8509,4.15334) 101 -dictGet test_01037.dict_array (33.8539,2.38339) 101 -dictGet test_01037.dict_array (33.858,-1.3122500000000001) 101 -dictGet test_01037.dict_array (33.859,3.72626) 101 -dictGet test_01037.dict_array (33.8616,2.24433) 101 -dictGet test_01037.dict_array (33.8621,3.01035) 101 -dictGet test_01037.dict_array (33.8623,1.17559) 101 -dictGet test_01037.dict_array (33.8682,2.706) 5994231 -dictGet test_01037.dict_array (33.8684,0.189231) 101 -dictGet test_01037.dict_array (33.872,1.93574) 101 -dictGet test_01037.dict_array (33.8844,3.80404) 101 -dictGet test_01037.dict_array (33.8888,0.594884) 101 -dictGet test_01037.dict_array (33.8946,2.74161) 101 -dictGet test_01037.dict_array (33.9023,0.6239) 101 -dictGet test_01037.dict_array (33.9057,0.873222) 101 -dictGet test_01037.dict_array (33.9157,-1.26607) 101 -dictGet test_01037.dict_array (33.92,2.06848) 101 -dictGet test_01037.dict_array (33.9298,-0.00526229) 101 -dictGet test_01037.dict_array (33.932,3.07063) 101 -dictGet test_01037.dict_array (33.9322,0.629385) 101 -dictGet test_01037.dict_array (33.9367,-1.41955) 101 -dictGet test_01037.dict_array (33.937,1.42532) 101 -dictGet test_01037.dict_array (33.9375,1.1467100000000001) 101 -dictGet test_01037.dict_array (33.9434,-1.05739) 101 -dictGet test_01037.dict_array (33.9477,3.34809) 101 -dictGet test_01037.dict_array (33.95,2.21715) 101 -dictGet test_01037.dict_array (33.955799999999996,0.305176) 101 -dictGet test_01037.dict_array (33.9686,-0.28273) 101 -dictGet test_01037.dict_array (33.9703,4.1255) 101 -dictGet test_01037.dict_array (33.9707,3.08199) 101 -dictGet test_01037.dict_array (33.9754,1.06203) 101 -dictGet test_01037.dict_array (33.9757,3.72468) 101 -dictGet test_01037.dict_array (33.9775,-0.0440599) 101 -dictGet test_01037.dict_array (33.9777,-0.251484) 101 -dictGet test_01037.dict_array (33.9789,-0.339374) 101 -dictGet test_01037.dict_array (33.9849,2.54515) 5994231 -dictGet test_01037.dict_array (33.9885,-0.318557) 101 -dictGet test_01037.dict_array (33.9977,1.07175) 101 -dictGet test_01037.dict_array (33.9984,-0.700517) 101 -dictGet test_01037.dict_array (34.0149,3.53338) 101 -dictGet test_01037.dict_array (34.0173,3.39155) 101 -dictGet test_01037.dict_array (34.0317,3.9579) 101 -dictGet test_01037.dict_array (34.0369,3.83612) 101 -dictGet test_01037.dict_array (34.043,-0.0887221) 101 -dictGet test_01037.dict_array (34.0487,1.14252) 101 -dictGet test_01037.dict_array (34.052,1.74832) 101 -dictGet test_01037.dict_array (34.0711,-0.898071) 101 -dictGet test_01037.dict_array (34.0747,1.55057) 101 -dictGet test_01037.dict_array (34.0803,3.16763) 101 -dictGet test_01037.dict_array (34.0872,3.75555) 101 -dictGet test_01037.dict_array (34.0965,1.62038) 101 -dictGet test_01037.dict_array (34.0977,-0.412691) 101 -dictGet test_01037.dict_array (34.0986,0.0294206) 101 -dictGet test_01037.dict_array (34.1072,3.15823) 101 -dictGet test_01037.dict_array (34.1092,3.09599) 101 -dictGet test_01037.dict_array (34.1206,1.04637) 5940222 -dictGet test_01037.dict_array (34.1209,3.13826) 101 -dictGet test_01037.dict_array (34.1265,3.95881) 101 -dictGet test_01037.dict_array (34.1286,-0.539319) 101 -dictGet test_01037.dict_array (34.1358,3.67451) 101 -dictGet test_01037.dict_array (34.1428,0.136115) 101 -dictGet test_01037.dict_array (34.157,1.73522) 101 -dictGet test_01037.dict_array (34.1581,1.48001) 101 -dictGet test_01037.dict_array (34.1682,3.42373) 101 -dictGet test_01037.dict_array (34.1683,-1.26511) 101 -dictGet test_01037.dict_array (34.1684,4.20007) 101 -dictGet test_01037.dict_array (34.1854,3.32089) 101 -dictGet test_01037.dict_array (34.2022,0.749536) 101 -dictGet test_01037.dict_array (34.2044,3.04865) 101 -dictGet test_01037.dict_array (34.22,-0.500055) 101 -dictGet test_01037.dict_array (34.2249,0.743775) 101 -dictGet test_01037.dict_array (34.2254,1.34702) 101 -dictGet test_01037.dict_array (34.2355,-0.898843) 101 -dictGet test_01037.dict_array (34.2394,2.0203699999999998) 101 -dictGet test_01037.dict_array (34.2466,1.83785) 101 -dictGet test_01037.dict_array (34.247,4.09563) 101 -dictGet test_01037.dict_array (34.2508,2.61312) 101 -dictGet test_01037.dict_array (34.2517,1.69642) 101 -dictGet test_01037.dict_array (34.2564,4.13033) 101 -dictGet test_01037.dict_array (34.2574,4.18928) 101 -dictGet test_01037.dict_array (34.2614,-0.478719) 101 -dictGet test_01037.dict_array (34.2625,2.38088) 101 -dictGet test_01037.dict_array (34.2666,3.1503) 101 -dictGet test_01037.dict_array (34.271,4.02223) 101 -dictGet test_01037.dict_array (34.2727,0.514755) 101 -dictGet test_01037.dict_array (34.278,1.98929) 101 -dictGet test_01037.dict_array (34.2798,-0.199208) 101 -dictGet test_01037.dict_array (34.2804,2.05184) 101 -dictGet test_01037.dict_array (34.2945,-1.11051) 101 -dictGet test_01037.dict_array (34.3168,-0.0829721) 101 -dictGet test_01037.dict_array (34.3345,3.4358) 101 -dictGet test_01037.dict_array (34.3377,1.13527) 5940222 -dictGet test_01037.dict_array (34.3383,1.27891) 5940222 -dictGet test_01037.dict_array (34.3391,1.47945) 5940222 -dictGet test_01037.dict_array (34.3441,0.627014) 101 -dictGet test_01037.dict_array (34.347,2.4853) 101 -dictGet test_01037.dict_array (34.3514,2.16247) 101 -dictGet test_01037.dict_array (34.3627,2.64533) 101 -dictGet test_01037.dict_array (34.3682,-0.227501) 101 -dictGet test_01037.dict_array (34.3756,4.21248) 101 -dictGet test_01037.dict_array (34.379,3.96604) 101 -dictGet test_01037.dict_array (34.3827,1.7518) 101 -dictGet test_01037.dict_array (34.3912,2.8834) 101 -dictGet test_01037.dict_array (34.3919,0.668829) 101 -dictGet test_01037.dict_array (34.3949,2.00338) 101 -dictGet test_01037.dict_array (34.3987,0.557268) 101 -dictGet test_01037.dict_array (34.4111,0.768558) 101 -dictGet test_01037.dict_array (34.4119,2.8742) 101 -dictGet test_01037.dict_array (34.416,3.50841) 101 -dictGet test_01037.dict_array (34.4212,1.24916) 5940222 -dictGet test_01037.dict_array (34.4251,0.457029) 101 -dictGet test_01037.dict_array (34.4274,-0.902559) 101 -dictGet test_01037.dict_array (34.4325,4.03159) 101 -dictGet test_01037.dict_array (34.438,1.63994) 101 -dictGet test_01037.dict_array (34.4403,-0.177594) 101 -dictGet test_01037.dict_array (34.4421,0.726712) 101 -dictGet test_01037.dict_array (34.4517,2.98611) 101 -dictGet test_01037.dict_array (34.4658,-1.312) 101 -dictGet test_01037.dict_array (34.4732,-0.0681338) 101 -dictGet test_01037.dict_array (34.4752,2.81646) 101 -dictGet test_01037.dict_array (34.4914,2.3858) 101 -dictGet test_01037.dict_array (34.4923,0.855231) 101 -dictGet test_01037.dict_array (34.5235,1.78468) 101 -dictGet test_01037.dict_array (34.5305,4.10608) 101 -dictGet test_01037.dict_array (34.5389,0.621937) 101 -dictGet test_01037.dict_array (34.5406,3.17145) 101 -dictGet test_01037.dict_array (34.5434,-0.56306) 101 -dictGet test_01037.dict_array (34.5449,3.13311) 101 -dictGet test_01037.dict_array (34.5491,2.31572) 101 -dictGet test_01037.dict_array (34.5539,2.94028) 101 -dictGet test_01037.dict_array (34.5546,-0.208825) 101 -dictGet test_01037.dict_array (34.5549,3.78486) 101 -dictGet test_01037.dict_array (34.5676,0.307148) 101 -dictGet test_01037.dict_array (34.5743,1.5217399999999999) 101 -dictGet test_01037.dict_array (34.5775,3.48046) 101 -dictGet test_01037.dict_array (34.5815,2.5243700000000002) 101 -dictGet test_01037.dict_array (34.5841,4.21191) 101 -dictGet test_01037.dict_array (34.5887,2.65083) 101 -dictGet test_01037.dict_array (34.5937,3.2143) 101 -dictGet test_01037.dict_array (34.6013,-1.0612) 101 -dictGet test_01037.dict_array (34.6089,1.36066) 101 -dictGet test_01037.dict_array (34.6103,3.40227) 101 -dictGet test_01037.dict_array (34.6128,1.92276) 101 -dictGet test_01037.dict_array (34.6175,2.43627) 101 -dictGet test_01037.dict_array (34.6209,3.43776) 101 -dictGet test_01037.dict_array (34.6234,2.60237) 101 -dictGet test_01037.dict_array (34.6275,3.52479) 101 -dictGet test_01037.dict_array (34.635,0.568558) 101 -dictGet test_01037.dict_array (34.6373,2.37692) 101 -dictGet test_01037.dict_array (34.6375,3.52234) 101 -dictGet test_01037.dict_array (34.6426,2.12397) 101 -dictGet test_01037.dict_array (34.6513,2.80915) 101 -dictGet test_01037.dict_array (34.6632,2.30039) 101 -dictGet test_01037.dict_array (34.6691,1.86582) 101 -dictGet test_01037.dict_array (34.6739,0.15342) 101 -dictGet test_01037.dict_array (34.6825,0.0499679) 101 -dictGet test_01037.dict_array (34.6893,0.454326) 101 -dictGet test_01037.dict_array (34.6957,-0.358598) 101 -dictGet test_01037.dict_array (34.6986,0.562679) 101 -dictGet test_01037.dict_array (34.712,1.12114) 101 -dictGet test_01037.dict_array (34.7126,-0.0057301) 101 -dictGet test_01037.dict_array (34.7137,0.0248501) 101 -dictGet test_01037.dict_array (34.7162,1.15623) 101 -dictGet test_01037.dict_array (34.7258,3.95142) 101 -dictGet test_01037.dict_array (34.7347,3.5232099999999997) 101 -dictGet test_01037.dict_array (34.7363,2.23374) 101 -dictGet test_01037.dict_array (34.7375,0.397841) 101 -dictGet test_01037.dict_array (34.7423,3.09198) 101 -dictGet test_01037.dict_array (34.7452,3.09029) 101 -dictGet test_01037.dict_array (34.7539,-1.06943) 101 -dictGet test_01037.dict_array (34.7733,-0.00912717) 101 -dictGet test_01037.dict_array (34.774,2.71088) 101 -dictGet test_01037.dict_array (34.7771,1.46009) 101 -dictGet test_01037.dict_array (34.7782,-1.28308) 101 -dictGet test_01037.dict_array (34.7924,3.63564) 101 -dictGet test_01037.dict_array (34.7939,-0.416676) 101 -dictGet test_01037.dict_array (34.7964,-0.401773) 101 -dictGet test_01037.dict_array (34.7974,0.0286873) 101 -dictGet test_01037.dict_array (34.7975,3.05965) 101 -dictGet test_01037.dict_array (34.8037,3.07263) 101 -dictGet test_01037.dict_array (34.8254,-0.390284) 101 -dictGet test_01037.dict_array (34.828,1.91869) 101 -dictGet test_01037.dict_array (34.8289,3.71058) 101 -dictGet test_01037.dict_array (34.8403,2.14606) 101 -dictGet test_01037.dict_array (34.8437,2.20617) 101 -dictGet test_01037.dict_array (34.8469,2.38435) 101 -dictGet test_01037.dict_array (34.86,1.45705) 101 -dictGet test_01037.dict_array (34.8612,0.914248) 101 -dictGet test_01037.dict_array (34.8663,3.4215400000000002) 101 -dictGet test_01037.dict_array (34.8724,-0.375144) 101 -dictGet test_01037.dict_array (34.8795,3.29317) 101 -dictGet test_01037.dict_array (34.8823,1.21988) 101 -dictGet test_01037.dict_array (34.8834,1.07657) 101 -dictGet test_01037.dict_array (34.8837,0.157648) 101 -dictGet test_01037.dict_array (34.8871,-0.9755) 101 -dictGet test_01037.dict_array (34.8871,1.8943699999999999) 101 -dictGet test_01037.dict_array (34.889,3.36756) 101 -dictGet test_01037.dict_array (34.8907,1.24874) 101 -dictGet test_01037.dict_array (34.8965,3.13508) 101 -dictGet test_01037.dict_array (34.9042,2.62092) 101 -dictGet test_01037.dict_array (34.9055,-0.0448967) 101 -dictGet test_01037.dict_array (34.9122,0.110576) 101 -dictGet test_01037.dict_array (34.9228,3.60183) 101 -dictGet test_01037.dict_array (34.9237,1.21715) 101 -dictGet test_01037.dict_array (34.9296,1.70459) 101 -dictGet test_01037.dict_array (34.941,-1.14663) 101 -dictGet test_01037.dict_array (34.9448,1.18923) 101 -dictGet test_01037.dict_array (34.9462,3.81678) 101 -dictGet test_01037.dict_array (34.9466,0.593463) 101 -dictGet test_01037.dict_array (34.9485,0.150307) 101 -dictGet test_01037.dict_array (34.9542,0.487238) 101 -dictGet test_01037.dict_array (34.9559,2.03473) 101 -dictGet test_01037.dict_array (34.9671,-0.960225) 101 -dictGet test_01037.dict_array (34.9711,2.63444) 101 -dictGet test_01037.dict_array (34.9892,0.354775) 101 -dictGet test_01037.dict_array (34.9907,1.40724) 101 -dictGet test_01037.dict_array (34.9916,-0.00173097) 101 -dictGet test_01037.dict_array (34.9919,2.06167) 101 +dictGet dict_array (29.5699,2.50068) 101 +dictGet dict_array (29.5796,1.55456) 101 +dictGet dict_array (29.5796,2.36864) 101 +dictGet dict_array (29.5844,1.59626) 101 +dictGet dict_array (29.5886,4.03321) 101 +dictGet dict_array (29.5914,3.02628) 101 +dictGet dict_array (29.5926,-0.0965169) 101 +dictGet dict_array (29.5968,2.37773) 101 +dictGet dict_array (29.5984,0.755853) 101 +dictGet dict_array (29.6066,3.47173) 101 +dictGet dict_array (29.6085,-1.26007) 101 +dictGet dict_array (29.6131,0.246565) 101 +dictGet dict_array (29.6157,-0.266687) 101 +dictGet dict_array (29.6164,2.94674) 101 +dictGet dict_array (29.6195,-0.591941) 101 +dictGet dict_array (29.6231,1.54818) 101 +dictGet dict_array (29.6379,0.764114) 101 +dictGet dict_array (29.6462,-0.772059) 934570 +dictGet dict_array (29.6579,-1.07336) 101 +dictGet dict_array (29.6618,-0.271842) 101 +dictGet dict_array (29.6629,-0.303602) 101 +dictGet dict_array (29.6659,-0.782823) 934570 +dictGet dict_array (29.6736,-0.113832) 101 +dictGet dict_array (29.6759,3.02905) 101 +dictGet dict_array (29.6778,3.71898) 101 +dictGet dict_array (29.6796,1.10433) 101 +dictGet dict_array (29.6809,2.13677) 101 +dictGet dict_array (29.6935,4.11894) 101 +dictGet dict_array (29.6991,-1.4458199999999999) 101 +dictGet dict_array (29.6997,3.17297) 101 +dictGet dict_array (29.7043,3.6145899999999997) 101 +dictGet dict_array (29.7065,3.24885) 101 +dictGet dict_array (29.7126,0.28108) 101 +dictGet dict_array (29.7192,0.174273) 101 +dictGet dict_array (29.7217,-0.523481) 934570 +dictGet dict_array (29.7271,1.67967) 101 +dictGet dict_array (29.7311,4.12444) 101 +dictGet dict_array (29.7347,1.88378) 101 +dictGet dict_array (29.7358,0.67944) 101 +dictGet dict_array (29.7366,-0.2973) 101 +dictGet dict_array (29.7446,0.646536) 101 +dictGet dict_array (29.7453,-0.567963) 101 +dictGet dict_array (29.764,4.04217) 101 +dictGet dict_array (29.7655,1.51372) 101 +dictGet dict_array (29.7744,1.12435) 101 +dictGet dict_array (29.7774,-0.0681196) 101 +dictGet dict_array (29.7784,1.54864) 101 +dictGet dict_array (29.7785,2.24139) 101 +dictGet dict_array (29.7922,0.220808) 101 +dictGet dict_array (29.7936,2.37709) 101 +dictGet dict_array (29.8008,0.948536) 101 +dictGet dict_array (29.8115,0.201227) 101 +dictGet dict_array (29.814,0.149601) 101 +dictGet dict_array (29.8193,-1.35858) 101 +dictGet dict_array (29.8201,0.965518) 101 +dictGet dict_array (29.8265,-0.727286) 101 +dictGet dict_array (29.8277,-0.531746) 101 +dictGet dict_array (29.8289,3.63009) 101 +dictGet dict_array (29.8548,0.838047) 101 +dictGet dict_array (29.8641,-0.845265) 101 +dictGet dict_array (29.8649,0.0562212) 101 +dictGet dict_array (29.8701,-1.02045) 101 +dictGet dict_array (29.8733,2.76654) 101 +dictGet dict_array (29.876,0.555475) 101 +dictGet dict_array (29.8794,-0.800108) 101 +dictGet dict_array (29.8813,2.7426399999999997) 101 +dictGet dict_array (29.897100000000002,2.66193) 101 +dictGet dict_array (29.908,4.01339) 101 +dictGet dict_array (29.9165,-1.08246) 101 +dictGet dict_array (29.9201,-0.420861) 101 +dictGet dict_array (29.9217,3.03778) 101 +dictGet dict_array (29.9355,0.773833) 101 +dictGet dict_array (29.947,3.76517) 101 +dictGet dict_array (29.9518,-0.60557) 101 +dictGet dict_array (29.9564,-0.600163) 101 +dictGet dict_array (29.959600000000002,4.16591) 101 +dictGet dict_array (29.9615,-1.33708) 101 +dictGet dict_array (29.9699,-0.392375) 101 +dictGet dict_array (29.9776,1.04552) 101 +dictGet dict_array (29.9784,4.02756) 101 +dictGet dict_array (29.9819,4.00597) 101 +dictGet dict_array (29.9826,1.2816100000000001) 101 +dictGet dict_array (30.0026,2.76257) 101 +dictGet dict_array (30.0126,3.68255) 101 +dictGet dict_array (30.0131,0.796576) 101 +dictGet dict_array (30.018,1.16523) 101 +dictGet dict_array (30.0261,-0.210653) 101 +dictGet dict_array (30.0472,-1.11007) 101 +dictGet dict_array (30.0542,-0.479585) 101 +dictGet dict_array (30.0613,1.6278000000000001) 101 +dictGet dict_array (30.0617,-0.0551152) 101 +dictGet dict_array (30.0637,2.62066) 101 +dictGet dict_array (30.0721,1.6424400000000001) 101 +dictGet dict_array (30.0769,-0.402636) 101 +dictGet dict_array (30.0791,-0.277435) 101 +dictGet dict_array (30.0931,0.0327512) 101 +dictGet dict_array (30.1059,3.52623) 101 +dictGet dict_array (30.1103,0.865466) 101 +dictGet dict_array (30.1115,2.95243) 101 +dictGet dict_array (30.1144,1.71029) 101 +dictGet dict_array (30.1311,-0.864751) 101 +dictGet dict_array (30.1336,-0.851386) 101 +dictGet dict_array (30.1393,3.89901) 101 +dictGet dict_array (30.1456,-0.531898) 101 +dictGet dict_array (30.1492,2.07833) 101 +dictGet dict_array (30.1575,2.43856) 101 +dictGet dict_array (30.1682,1.19771) 101 +dictGet dict_array (30.1716,3.9853300000000003) 101 +dictGet dict_array (30.1849,2.78374) 101 +dictGet dict_array (30.1866,0.65658) 101 +dictGet dict_array (30.1885,1.56943) 101 +dictGet dict_array (30.1959,-1.38202) 101 +dictGet dict_array (30.1999,1.58413) 101 +dictGet dict_array (30.2024,0.713081) 101 +dictGet dict_array (30.2054,0.620143) 101 +dictGet dict_array (30.2091,1.51641) 101 +dictGet dict_array (30.2124,-0.331782) 101 +dictGet dict_array (30.226,3.03527) 101 +dictGet dict_array (30.2261,3.18486) 101 +dictGet dict_array (30.2288,2.48407) 101 +dictGet dict_array (30.2345,3.7462400000000002) 101 +dictGet dict_array (30.2375,0.62046) 101 +dictGet dict_array (30.2425,-0.472914) 101 +dictGet dict_array (30.247,3.95863) 101 +dictGet dict_array (30.2494,-0.305093) 101 +dictGet dict_array (30.2499,2.54337) 101 +dictGet dict_array (30.2606,2.16644) 101 +dictGet dict_array (30.2672,3.94847) 101 +dictGet dict_array (30.2709,-0.136264) 101 +dictGet dict_array (30.2764,1.18654) 101 +dictGet dict_array (30.2765,1.20383) 101 +dictGet dict_array (30.2839,1.05762) 101 +dictGet dict_array (30.286,0.469327) 101 +dictGet dict_array (30.2927,3.1693) 101 +dictGet dict_array (30.2935,3.49854) 101 +dictGet dict_array (30.307,0.312338) 101 +dictGet dict_array (30.3085,1.07791) 101 +dictGet dict_array (30.3139,2.77248) 101 +dictGet dict_array (30.314,0.822823) 101 +dictGet dict_array (30.3227,-0.587351) 101 +dictGet dict_array (30.332,1.00174) 101 +dictGet dict_array (30.3388,0.844148) 101 +dictGet dict_array (30.3485,0.561902) 101 +dictGet dict_array (30.3497,0.180362) 101 +dictGet dict_array (30.361,4.13016) 101 +dictGet dict_array (30.3623,-0.0484027) 101 +dictGet dict_array (30.3638,3.9845800000000002) 101 +dictGet dict_array (30.3853,3.16051) 101 +dictGet dict_array (30.3974,2.6617800000000003) 101 +dictGet dict_array (30.4002,-1.15886) 101 +dictGet dict_array (30.4008,-0.387015) 101 +dictGet dict_array (30.4018,1.86493) 101 +dictGet dict_array (30.4239,1.16818) 101 +dictGet dict_array (30.4363,3.63938) 101 +dictGet dict_array (30.4377,-0.81315) 101 +dictGet dict_array (30.4391,3.54703) 101 +dictGet dict_array (30.4424,-1.39435) 101 +dictGet dict_array (30.4441,2.8463000000000003) 101 +dictGet dict_array (30.4517,3.28117) 101 +dictGet dict_array (30.4658,2.6928) 101 +dictGet dict_array (30.4734,2.66161) 101 +dictGet dict_array (30.4799,-1.07578) 101 +dictGet dict_array (30.4837,-1.02486) 101 +dictGet dict_array (30.485,1.06326) 101 +dictGet dict_array (30.495,1.12306) 101 +dictGet dict_array (30.501,2.27264) 101 +dictGet dict_array (30.5027,1.99382) 101 +dictGet dict_array (30.5194,-1.03943) 101 +dictGet dict_array (30.5239,1.04328) 101 +dictGet dict_array (30.528,3.82041) 101 +dictGet dict_array (30.5299,-0.715248) 101 +dictGet dict_array (30.5331,1.19603) 101 +dictGet dict_array (30.535800000000002,2.71485) 101 +dictGet dict_array (30.5405,0.804694) 101 +dictGet dict_array (30.542,1.23739) 101 +dictGet dict_array (30.5432,4.04189) 101 +dictGet dict_array (30.5457,-0.956121) 101 +dictGet dict_array (30.5506,3.07443) 101 +dictGet dict_array (30.5539,3.87084) 101 +dictGet dict_array (30.5578,3.78837) 101 +dictGet dict_array (30.5588,0.966135) 101 +dictGet dict_array (30.5637,2.5605) 101 +dictGet dict_array (30.5647,-1.27328) 101 +dictGet dict_array (30.5656,-0.0581332) 101 +dictGet dict_array (30.5715,0.65755) 101 +dictGet dict_array (30.5727,3.01604) 101 +dictGet dict_array (30.5729,-0.976857) 101 +dictGet dict_array (30.5751,0.60204) 101 +dictGet dict_array (30.5854,3.02473) 101 +dictGet dict_array (30.5866,0.174099) 101 +dictGet dict_array (30.5947,0.875193) 101 +dictGet dict_array (30.5992,-0.403901) 101 +dictGet dict_array (30.6002,4.18891) 101 +dictGet dict_array (30.6025,0.217712) 101 +dictGet dict_array (30.6054,0.927203) 101 +dictGet dict_array (30.6075,3.79359) 101 +dictGet dict_array (30.6159,3.82773) 101 +dictGet dict_array (30.627,3.84039) 101 +dictGet dict_array (30.6308,0.77517) 101 +dictGet dict_array (30.6338,0.179565) 101 +dictGet dict_array (30.6461,1.3293599999999999) 101 +dictGet dict_array (30.6674,-0.424547) 101 +dictGet dict_array (30.669,1.76539) 101 +dictGet dict_array (30.6788,4.01239) 101 +dictGet dict_array (30.6864,3.59158) 101 +dictGet dict_array (30.7049,-0.875413) 101 +dictGet dict_array (30.705,1.3307) 101 +dictGet dict_array (30.7063,-0.473192) 101 +dictGet dict_array (30.7075,-1.1958199999999999) 101 +dictGet dict_array (30.7101,-0.367562) 101 +dictGet dict_array (30.7203,2.98725) 101 +dictGet dict_array (30.7213,2.2745699999999998) 101 +dictGet dict_array (30.7446,-0.334144) 101 +dictGet dict_array (30.7468,3.82967) 101 +dictGet dict_array (30.747,-0.384779) 101 +dictGet dict_array (30.7681,0.904198) 101 +dictGet dict_array (30.7757,1.78743) 101 +dictGet dict_array (30.8021,-0.479212) 101 +dictGet dict_array (30.8079,-1.40869) 101 +dictGet dict_array (30.8206,-0.0608489) 101 +dictGet dict_array (30.8218,0.43909) 101 +dictGet dict_array (30.8239,0.10014) 101 +dictGet dict_array (30.8282,4.15409) 101 +dictGet dict_array (30.8288,-0.709528) 101 +dictGet dict_array (30.8326,0.156011) 101 +dictGet dict_array (30.8328,-1.03704) 101 +dictGet dict_array (30.839,2.15528) 101 +dictGet dict_array (30.8452,0.219377) 101 +dictGet dict_array (30.8463,0.0515355) 101 +dictGet dict_array (30.8526,2.06614) 101 +dictGet dict_array (30.8566,0.517876) 101 +dictGet dict_array (30.8588,-1.31738) 101 +dictGet dict_array (30.8681,0.44207) 101 +dictGet dict_array (30.8914,1.0072) 101 +dictGet dict_array (30.897,0.483425) 101 +dictGet dict_array (30.905,2.8731999999999998) 101 +dictGet dict_array (30.9051,2.21956) 101 +dictGet dict_array (30.9115,4.00663) 101 +dictGet dict_array (30.9167,-0.834462) 101 +dictGet dict_array (30.9252,-1.3289900000000001) 101 +dictGet dict_array (30.9314,1.85384) 101 +dictGet dict_array (30.9392,2.53236) 101 +dictGet dict_array (30.9569,2.82038) 101 +dictGet dict_array (30.9598,-0.641011) 101 +dictGet dict_array (30.9601,-0.254928) 101 +dictGet dict_array (30.9623,-1.3886) 101 +dictGet dict_array (30.9707,0.888854) 101 +dictGet dict_array (30.9766,2.81957) 101 +dictGet dict_array (30.9775,2.69273) 101 +dictGet dict_array (30.9821,0.587715) 101 +dictGet dict_array (30.9887,4.0233) 101 +dictGet dict_array (30.9914,0.259542) 101 +dictGet dict_array (30.9986,-1.36832) 101 +dictGet dict_array (31.008,0.628999) 101 +dictGet dict_array (31.0168,-1.17462) 101 +dictGet dict_array (31.0237,3.52547) 101 +dictGet dict_array (31.0306,3.78522) 101 +dictGet dict_array (31.0308,-0.72453) 101 +dictGet dict_array (31.0463,2.41997) 101 +dictGet dict_array (31.047,0.624184) 101 +dictGet dict_array (31.0569,0.0706393) 5994232 +dictGet dict_array (31.0583,1.3244099999999999) 101 +dictGet dict_array (31.063,3.23861) 101 +dictGet dict_array (31.068,0.695575) 101 +dictGet dict_array (31.0687,1.85675) 101 +dictGet dict_array (31.0692,0.254793) 101 +dictGet dict_array (31.0766,0.828128) 101 +dictGet dict_array (31.0833,0.0612782) 5994232 +dictGet dict_array (31.0833,2.59748) 101 +dictGet dict_array (31.0861,-1.3778299999999999) 101 +dictGet dict_array (31.0874,3.07258) 101 +dictGet dict_array (31.0882,1.4882) 101 +dictGet dict_array (31.0924,3.42242) 101 +dictGet dict_array (31.0927,2.67448) 101 +dictGet dict_array (31.0936,1.12292) 101 +dictGet dict_array (31.0952,-0.336928) 101 +dictGet dict_array (31.0978,3.48482) 101 +dictGet dict_array (31.1107,3.7513199999999998) 101 +dictGet dict_array (31.1156,1.19171) 101 +dictGet dict_array (31.1176,0.223509) 5994232 +dictGet dict_array (31.1249,0.946838) 101 +dictGet dict_array (31.1267,1.48983) 101 +dictGet dict_array (31.138,-0.289981) 101 +dictGet dict_array (31.1382,3.02904) 101 +dictGet dict_array (31.1475,2.6178) 101 +dictGet dict_array (31.1491,1.37873) 101 +dictGet dict_array (31.1525,3.72105) 101 +dictGet dict_array (31.1526,-1.4129800000000001) 101 +dictGet dict_array (31.1526,-0.186457) 101 +dictGet dict_array (31.1539,2.78789) 101 +dictGet dict_array (31.1548,-1.08552) 101 +dictGet dict_array (31.1567,-0.0768925) 101 +dictGet dict_array (31.1613,1.49617) 101 +dictGet dict_array (31.1653,1.03777) 101 +dictGet dict_array (31.1662,3.4214700000000002) 101 +dictGet dict_array (31.1672,-0.0813169) 101 +dictGet dict_array (31.177,0.440843) 101 +dictGet dict_array (31.1788,-0.737151) 101 +dictGet dict_array (31.1856,-0.144396) 101 +dictGet dict_array (31.1959,3.66813) 101 +dictGet dict_array (31.1996,-0.353983) 101 +dictGet dict_array (31.2019,2.86802) 101 +dictGet dict_array (31.2087,2.31245) 101 +dictGet dict_array (31.2125,3.2713200000000002) 101 +dictGet dict_array (31.2137,-0.108129) 101 +dictGet dict_array (31.216,3.9156) 101 +dictGet dict_array (31.2201,-0.202141) 101 +dictGet dict_array (31.2285,2.09058) 101 +dictGet dict_array (31.2502,4.01526) 101 +dictGet dict_array (31.2585,3.11524) 101 +dictGet dict_array (31.2645,-0.620418) 101 +dictGet dict_array (31.2684,2.74277) 101 +dictGet dict_array (31.2821,-1.12772) 101 +dictGet dict_array (31.2821,2.46769) 101 +dictGet dict_array (31.2887,3.91396) 101 +dictGet dict_array (31.295,1.49942) 101 +dictGet dict_array (31.2997,3.46122) 101 +dictGet dict_array (31.3017,3.3263) 101 +dictGet dict_array (31.3022,3.16754) 101 +dictGet dict_array (31.3048,0.364962) 101 +dictGet dict_array (31.305,3.1967) 101 +dictGet dict_array (31.3061,1.84303) 101 +dictGet dict_array (31.3082,-0.173851) 101 +dictGet dict_array (31.3315,3.90932) 101 +dictGet dict_array (31.3351,2.80164) 101 +dictGet dict_array (31.3388,0.168765) 5994233 +dictGet dict_array (31.339,0.25535) 101 +dictGet dict_array (31.3423,1.7036799999999999) 101 +dictGet dict_array (31.349,0.386456) 101 +dictGet dict_array (31.3558,-1.04336) 101 +dictGet dict_array (31.3564,0.478876) 101 +dictGet dict_array (31.3607,-0.0860507) 5994233 +dictGet dict_array (31.3831,3.84469) 101 +dictGet dict_array (31.3886,-0.731137) 101 +dictGet dict_array (31.4043,-0.348907) 101 +dictGet dict_array (31.4081,1.47391) 101 +dictGet dict_array (31.4176,-0.583645) 101 +dictGet dict_array (31.4177,1.36972) 101 +dictGet dict_array (31.4182,0.958303) 101 +dictGet dict_array (31.4199,3.1738) 101 +dictGet dict_array (31.4221,2.74876) 101 +dictGet dict_array (31.4301,-0.122643) 5994233 +dictGet dict_array (31.4344,1.00661) 101 +dictGet dict_array (31.4375,4.20304) 101 +dictGet dict_array (31.4377,0.289608) 101 +dictGet dict_array (31.4379,0.54744) 101 +dictGet dict_array (31.4459,3.94945) 101 +dictGet dict_array (31.4559,-0.345063) 101 +dictGet dict_array (31.464,0.726129) 101 +dictGet dict_array (31.4662,-0.299019) 5994233 +dictGet dict_array (31.4671,1.9605299999999999) 101 +dictGet dict_array (31.4673,-0.403676) 101 +dictGet dict_array (31.4712,-0.237941) 5994233 +dictGet dict_array (31.4816,0.120264) 5994233 +dictGet dict_array (31.4875,0.323483) 101 +dictGet dict_array (31.490099999999998,-0.338163) 101 +dictGet dict_array (31.4932,0.517674) 101 +dictGet dict_array (31.5112,1.9689299999999998) 101 +dictGet dict_array (31.5122,2.92785) 101 +dictGet dict_array (31.5151,0.166429) 101 +dictGet dict_array (31.5174,2.94802) 101 +dictGet dict_array (31.5182,4.18776) 101 +dictGet dict_array (31.5238,1.18793) 101 +dictGet dict_array (31.5271,3.07446) 101 +dictGet dict_array (31.5393,1.58061) 101 +dictGet dict_array (31.5421,3.13711) 101 +dictGet dict_array (31.5479,2.39897) 101 +dictGet dict_array (31.5519,0.99285) 101 +dictGet dict_array (31.5685,3.47987) 101 +dictGet dict_array (31.5959,0.437382) 101 +dictGet dict_array (31.6003,0.194376) 101 +dictGet dict_array (31.6026,2.15457) 101 +dictGet dict_array (31.606,2.45365) 101 +dictGet dict_array (31.6062,-0.453441) 101 +dictGet dict_array (31.6107,1.35247) 101 +dictGet dict_array (31.6155,3.85588) 101 +dictGet dict_array (31.6222,2.03326) 101 +dictGet dict_array (31.6231,-0.123059) 101 +dictGet dict_array (31.6244,1.6885599999999998) 101 +dictGet dict_array (31.6459,0.669716) 101 +dictGet dict_array (31.6563,-0.0644741) 101 +dictGet dict_array (31.6618,-0.551121) 101 +dictGet dict_array (31.6725,-0.38922) 101 +dictGet dict_array (31.6727,4.10336) 101 +dictGet dict_array (31.6739,4.1391) 101 +dictGet dict_array (31.6897,2.8694699999999997) 101 +dictGet dict_array (31.6902,3.98792) 101 +dictGet dict_array (31.6945,2.46687) 101 +dictGet dict_array (31.6987,-1.3796) 101 +dictGet dict_array (31.7012,2.34845) 101 +dictGet dict_array (31.7036,0.0228348) 101 +dictGet dict_array (31.7046,3.68111) 101 +dictGet dict_array (31.7055,2.92556) 101 +dictGet dict_array (31.7102,1.04532) 101 +dictGet dict_array (31.7149,-0.443302) 101 +dictGet dict_array (31.7195,2.99311) 101 +dictGet dict_array (31.7274,0.166719) 101 +dictGet dict_array (31.7565,-0.565382) 101 +dictGet dict_array (31.7615,0.771626) 101 +dictGet dict_array (31.7739,1.8970099999999999) 101 +dictGet dict_array (31.7848,1.2623199999999999) 101 +dictGet dict_array (31.7912,-0.788599) 101 +dictGet dict_array (31.8011,2.65853) 101 +dictGet dict_array (31.8032,-0.0590108) 101 +dictGet dict_array (31.8038,1.9618799999999998) 101 +dictGet dict_array (31.8098,-1.46851) 101 +dictGet dict_array (31.8131,3.41982) 101 +dictGet dict_array (31.8169,3.31059) 101 +dictGet dict_array (31.8202,-0.193692) 101 +dictGet dict_array (31.8306,1.57586) 101 +dictGet dict_array (31.8382,-0.787948) 101 +dictGet dict_array (31.8433,2.49692) 101 +dictGet dict_array (31.8436,2.41851) 101 +dictGet dict_array (31.8563,-1.10787) 101 +dictGet dict_array (31.8683,0.996504) 101 +dictGet dict_array (31.8693,-0.828142) 101 +dictGet dict_array (31.8723,1.08929) 101 +dictGet dict_array (31.8737,0.881127) 101 +dictGet dict_array (31.8881,-0.58441) 101 +dictGet dict_array (31.9011,0.121349) 101 +dictGet dict_array (31.9066,2.13045) 101 +dictGet dict_array (31.9142,1.03368) 101 +dictGet dict_array (31.9155,3.38363) 101 +dictGet dict_array (31.9168,1.3166) 101 +dictGet dict_array (31.9185,-1.11879) 101 +dictGet dict_array (31.9186,-0.647948) 101 +dictGet dict_array (31.9311,3.96928) 101 +dictGet dict_array (31.9335,1.47048) 101 +dictGet dict_array (31.9443,-1.36175) 101 +dictGet dict_array (31.9481,2.34231) 101 +dictGet dict_array (31.9526,1.36565) 101 +dictGet dict_array (31.9629,2.5208399999999997) 101 +dictGet dict_array (31.9765,0.975783) 101 +dictGet dict_array (31.9923,3.31773) 101 +dictGet dict_array (31.9994,0.972816) 101 +dictGet dict_array (32.001,3.47425) 101 +dictGet dict_array (32.0127,2.13874) 101 +dictGet dict_array (32.0244,3.2092) 101 +dictGet dict_array (32.029,1.18039) 101 +dictGet dict_array (32.0315,0.566073) 101 +dictGet dict_array (32.0354,1.0766499999999999) 101 +dictGet dict_array (32.0399,-1.11576) 101 +dictGet dict_array (32.053,2.16849) 101 +dictGet dict_array (32.0542,0.042328) 101 +dictGet dict_array (32.0576,2.47001) 101 +dictGet dict_array (32.061,3.7498899999999997) 101 +dictGet dict_array (32.0623,1.25134) 101 +dictGet dict_array (32.0626,1.9611399999999999) 101 +dictGet dict_array (32.0666,-0.0904247) 101 +dictGet dict_array (32.0681,2.28442) 101 +dictGet dict_array (32.0692,1.50869) 101 +dictGet dict_array (32.0724,4.03314) 101 +dictGet dict_array (32.0729,-0.064324) 101 +dictGet dict_array (32.079,0.293758) 101 +dictGet dict_array (32.0847,-1.19814) 101 +dictGet dict_array (32.0974,-0.91927) 101 +dictGet dict_array (32.0979,-0.736979) 101 +dictGet dict_array (32.106,-1.33063) 101 +dictGet dict_array (32.1189,0.246715) 101 +dictGet dict_array (32.1207,4.00883) 101 +dictGet dict_array (32.1396,1.12402) 101 +dictGet dict_array (32.1413,1.5668) 101 +dictGet dict_array (32.143,1.35559) 101 +dictGet dict_array (32.1538,1.32881) 101 +dictGet dict_array (32.1549,4.06552) 101 +dictGet dict_array (32.1555,-0.79275) 101 +dictGet dict_array (32.163,1.17733) 101 +dictGet dict_array (32.1634,2.94273) 101 +dictGet dict_array (32.1644,1.85666) 101 +dictGet dict_array (32.1745,0.435458) 101 +dictGet dict_array (32.1765,1.65149) 101 +dictGet dict_array (32.1893,2.08924) 101 +dictGet dict_array (32.2024,0.222191) 101 +dictGet dict_array (32.2107,1.34379) 101 +dictGet dict_array (32.2109,3.9018699999999997) 101 +dictGet dict_array (32.2123,1.85233) 101 +dictGet dict_array (32.2144,3.72534) 101 +dictGet dict_array (32.2218,2.5386699999999998) 101 +dictGet dict_array (32.2279,2.84267) 101 +dictGet dict_array (32.2345,3.33295) 101 +dictGet dict_array (32.2435,3.85283) 101 +dictGet dict_array (32.2527,-0.480608) 101 +dictGet dict_array (32.2566,-0.837882) 101 +dictGet dict_array (32.2627,2.57708) 101 +dictGet dict_array (32.2733,0.244931) 101 +dictGet dict_array (32.2761,4.05808) 101 +dictGet dict_array (32.2764,3.78472) 101 +dictGet dict_array (32.2814,-1.26011) 101 +dictGet dict_array (32.2861,3.02427) 101 +dictGet dict_array (32.2924,0.928609) 101 +dictGet dict_array (32.2963,-0.78543) 101 +dictGet dict_array (32.3039,3.21175) 101 +dictGet dict_array (32.3107,0.698287) 101 +dictGet dict_array (32.3138,0.0595677) 101 +dictGet dict_array (32.3339,0.707056) 101 +dictGet dict_array (32.3351,0.415474) 101 +dictGet dict_array (32.342,-0.681023) 101 +dictGet dict_array (32.3463,1.83196) 101 +dictGet dict_array (32.3494,2.43799) 101 +dictGet dict_array (32.3524,3.47049) 101 +dictGet dict_array (32.3531,2.33115) 101 +dictGet dict_array (32.3602,0.116106) 101 +dictGet dict_array (32.3612,1.1598) 101 +dictGet dict_array (32.3689,3.34847) 101 +dictGet dict_array (32.3695,0.734055) 101 +dictGet dict_array (32.3825,3.85017) 101 +dictGet dict_array (32.3835,-1.25491) 101 +dictGet dict_array (32.4018,-0.728568) 101 +dictGet dict_array (32.4044,2.96727) 101 +dictGet dict_array (32.4101,2.9988) 101 +dictGet dict_array (32.417,-1.12908) 101 +dictGet dict_array (32.4172,4.1952) 101 +dictGet dict_array (32.4239,2.49512) 101 +dictGet dict_array (32.4258,4.05137) 101 +dictGet dict_array (32.4264,-0.427357) 101 +dictGet dict_array (32.4274,3.59377) 101 +dictGet dict_array (32.4286,-1.24757) 101 +dictGet dict_array (32.4294,3.0665) 101 +dictGet dict_array (32.4333,-0.353347) 101 +dictGet dict_array (32.4391,3.64421) 101 +dictGet dict_array (32.4401,3.70635) 101 +dictGet dict_array (32.45,1.68918) 101 +dictGet dict_array (32.4507,-0.133471) 101 +dictGet dict_array (32.4592,0.976458) 101 +dictGet dict_array (32.4595,1.89135) 101 +dictGet dict_array (32.4604,0.280248) 101 +dictGet dict_array (32.4835,0.472731) 101 +dictGet dict_array (32.4855,2.01938) 101 +dictGet dict_array (32.4872,2.01697) 101 +dictGet dict_array (32.4911,0.613106) 101 +dictGet dict_array (32.4918,2.17834) 101 +dictGet dict_array (32.4947,2.34595) 101 +dictGet dict_array (32.5035,2.92234) 101 +dictGet dict_array (32.5132,-0.331206) 101 +dictGet dict_array (32.5156,-0.412604) 7652581 +dictGet dict_array (32.5158,2.9067499999999997) 101 +dictGet dict_array (32.5249,2.44519) 101 +dictGet dict_array (32.5293,-0.790952) 101 +dictGet dict_array (32.5319,3.96854) 101 +dictGet dict_array (32.5518,3.6093) 101 +dictGet dict_array (32.5541,3.5225400000000002) 101 +dictGet dict_array (32.5569,0.816123) 101 +dictGet dict_array (32.5646,1.9775) 101 +dictGet dict_array (32.5733,3.81271) 101 +dictGet dict_array (32.5767,0.948327) 101 +dictGet dict_array (32.5971,1.76179) 101 +dictGet dict_array (32.6035,-0.716157) 101 +dictGet dict_array (32.6087,4.21614) 101 +dictGet dict_array (32.6171,0.024481) 101 +dictGet dict_array (32.6189,-0.775391) 101 +dictGet dict_array (32.6198,2.92081) 101 +dictGet dict_array (32.621,-0.970784) 101 +dictGet dict_array (32.6266,0.650009) 101 +dictGet dict_array (32.6315,2.15144) 101 +dictGet dict_array (32.6385,-0.436803) 101 +dictGet dict_array (32.6449,-0.191292) 101 +dictGet dict_array (32.6535,2.10385) 101 +dictGet dict_array (32.6592,3.49973) 101 +dictGet dict_array (32.6598,2.5980600000000003) 101 +dictGet dict_array (32.6612,2.95681) 101 +dictGet dict_array (32.6636,-0.57235) 101 +dictGet dict_array (32.669,-0.382702) 101 +dictGet dict_array (32.6752,1.30748) 101 +dictGet dict_array (32.6811,2.9559800000000003) 101 +dictGet dict_array (32.6821,0.57336) 101 +dictGet dict_array (32.6828,3.91304) 101 +dictGet dict_array (32.6979,3.96868) 101 +dictGet dict_array (32.6983,3.15784) 101 +dictGet dict_array (32.7122,0.794293) 101 +dictGet dict_array (32.7131,-0.847256) 101 +dictGet dict_array (32.7219,0.883461) 101 +dictGet dict_array (32.7228,1.78808) 101 +dictGet dict_array (32.7273,-0.206908) 101 +dictGet dict_array (32.7292,0.259331) 101 +dictGet dict_array (32.7304,-1.38317) 101 +dictGet dict_array (32.7353,1.01601) 101 +dictGet dict_array (32.7354,4.17574) 101 +dictGet dict_array (32.7357,-0.190194) 101 +dictGet dict_array (32.7465,-1.37598) 101 +dictGet dict_array (32.7494,-0.275675) 101 +dictGet dict_array (32.7514,0.128951) 101 +dictGet dict_array (32.753,3.44207) 101 +dictGet dict_array (32.7686,2.11713) 101 +dictGet dict_array (32.7694,1.47159) 101 +dictGet dict_array (32.7768,0.0401042) 101 +dictGet dict_array (32.781,-1.34283) 101 +dictGet dict_array (32.7814,1.73876) 101 +dictGet dict_array (32.7856,-1.06363) 101 +dictGet dict_array (32.792699999999996,-1.1255600000000001) 101 +dictGet dict_array (32.7941,-0.645447) 101 +dictGet dict_array (32.7946,1.48889) 101 +dictGet dict_array (32.797,0.791753) 101 +dictGet dict_array (32.7982,-0.537798) 101 +dictGet dict_array (32.8091,2.3611) 101 +dictGet dict_array (32.81,1.7130800000000002) 101 +dictGet dict_array (32.8174,-0.288322) 101 +dictGet dict_array (32.823,1.6546699999999999) 101 +dictGet dict_array (32.8233,1.62108) 101 +dictGet dict_array (32.8428,-0.400045) 101 +dictGet dict_array (32.8479,2.13598) 101 +dictGet dict_array (32.8524,0.199902) 101 +dictGet dict_array (32.8543,3.23553) 101 +dictGet dict_array (32.8562,1.31371) 101 +dictGet dict_array (32.87,1.44256) 101 +dictGet dict_array (32.8789,2.38192) 101 +dictGet dict_array (32.8812,2.20734) 5999168 +dictGet dict_array (32.8815,-0.54427) 101 +dictGet dict_array (32.8853,2.4859) 5999168 +dictGet dict_array (32.8909,0.513964) 101 +dictGet dict_array (32.9035,2.38999) 101 +dictGet dict_array (32.9097,2.48131) 5999168 +dictGet dict_array (32.928,-0.943269) 101 +dictGet dict_array (32.9322,1.13165) 101 +dictGet dict_array (32.9348,1.22606) 101 +dictGet dict_array (32.9417,3.77998) 101 +dictGet dict_array (32.9428,3.11936) 101 +dictGet dict_array (32.9482,1.18092) 101 +dictGet dict_array (32.9506,0.0609364) 101 +dictGet dict_array (32.953,-0.828308) 101 +dictGet dict_array (32.9593,3.5209099999999998) 101 +dictGet dict_array (32.9617,2.07711) 5999168 +dictGet dict_array (32.966,0.693749) 101 +dictGet dict_array (32.9668,-0.716432) 101 +dictGet dict_array (32.9702,1.98555) 101 +dictGet dict_array (32.9782,1.73819) 101 +dictGet dict_array (32.9805,3.71151) 101 +dictGet dict_array (32.9821,2.97225) 101 +dictGet dict_array (32.995,-0.830301) 101 +dictGet dict_array (33.0234,0.770848) 101 +dictGet dict_array (33.0312,-0.340964) 101 +dictGet dict_array (33.0366,-0.756795) 101 +dictGet dict_array (33.0438,0.812871) 101 +dictGet dict_array (33.0455,1.84843) 101 +dictGet dict_array (33.0498,0.0913292) 101 +dictGet dict_array (33.0506,1.53739) 101 +dictGet dict_array (33.0554,2.4265) 101 +dictGet dict_array (33.0741,3.61332) 101 +dictGet dict_array (33.0765,-0.179985) 101 +dictGet dict_array (33.087,1.46465) 101 +dictGet dict_array (33.0906,-0.620383) 101 +dictGet dict_array (33.1047,-1.28027) 101 +dictGet dict_array (33.1072,1.96303) 101 +dictGet dict_array (33.1081,-0.897874) 101 +dictGet dict_array (33.1122,1.8950200000000001) 101 +dictGet dict_array (33.1237,2.63993) 101 +dictGet dict_array (33.1238,0.753963) 101 +dictGet dict_array (33.1257,0.495668) 101 +dictGet dict_array (33.1258,1.78341) 101 +dictGet dict_array (33.127,2.59646) 101 +dictGet dict_array (33.1324,-1.23742) 101 +dictGet dict_array (33.1359,3.83491) 101 +dictGet dict_array (33.1628,-0.379588) 101 +dictGet dict_array (33.1679,1.25601) 101 +dictGet dict_array (33.1688,-1.35553) 101 +dictGet dict_array (33.181,2.10943) 101 +dictGet dict_array (33.1871,2.81171) 101 +dictGet dict_array (33.1877,0.771297) 101 +dictGet dict_array (33.1883,-0.204797) 101 +dictGet dict_array (33.1886,3.27998) 101 +dictGet dict_array (33.1955,0.708907) 101 +dictGet dict_array (33.2044,-0.769275) 101 +dictGet dict_array (33.2182,3.36103) 101 +dictGet dict_array (33.2192,3.43586) 101 +dictGet dict_array (33.2322,-0.916753) 101 +dictGet dict_array (33.2359,-0.81321) 101 +dictGet dict_array (33.238,0.635072) 101 +dictGet dict_array (33.2398,3.02588) 101 +dictGet dict_array (33.2469,2.35698) 101 +dictGet dict_array (33.247,2.3327) 101 +dictGet dict_array (33.2579,2.8027100000000003) 101 +dictGet dict_array (33.2607,0.321082) 101 +dictGet dict_array (33.2653,0.243336) 101 +dictGet dict_array (33.2758,0.831836) 101 +dictGet dict_array (33.2771,0.886536) 101 +dictGet dict_array (33.2914,1.16026) 101 +dictGet dict_array (33.2914,1.38882) 101 +dictGet dict_array (33.2982,-1.16604) 101 +dictGet dict_array (33.2985,0.842556) 101 +dictGet dict_array (33.3005,2.8338900000000002) 101 +dictGet dict_array (33.305,0.0969475) 101 +dictGet dict_array (33.3072,3.82163) 101 +dictGet dict_array (33.312,3.41475) 101 +dictGet dict_array (33.3129,2.46048) 101 +dictGet dict_array (33.3134,3.46863) 101 +dictGet dict_array (33.3203,2.33139) 101 +dictGet dict_array (33.324,0.433701) 101 +dictGet dict_array (33.3338,2.44705) 101 +dictGet dict_array (33.337,4.06475) 101 +dictGet dict_array (33.3469,1.08172) 101 +dictGet dict_array (33.3538,0.717896) 101 +dictGet dict_array (33.3618,1.37899) 101 +dictGet dict_array (33.3698,0.547744) 101 +dictGet dict_array (33.3705,0.957619) 101 +dictGet dict_array (33.3821,3.07258) 101 +dictGet dict_array (33.3881,3.0626) 101 +dictGet dict_array (33.393,-0.816186) 101 +dictGet dict_array (33.3945,0.869508) 101 +dictGet dict_array (33.4001,1.24186) 101 +dictGet dict_array (33.4008,2.34911) 101 +dictGet dict_array (33.4166,-1.2808899999999999) 101 +dictGet dict_array (33.4167,3.0655) 101 +dictGet dict_array (33.4204,2.81887) 101 +dictGet dict_array (33.4211,1.71128) 101 +dictGet dict_array (33.4237,2.91761) 101 +dictGet dict_array (33.4266,1.5955599999999999) 101 +dictGet dict_array (33.4353,-0.391392) 101 +dictGet dict_array (33.4362,-0.134658) 101 +dictGet dict_array (33.4386,0.15396) 101 +dictGet dict_array (33.4421,-0.50712) 101 +dictGet dict_array (33.452,0.915829) 101 +dictGet dict_array (33.463,-0.0882717) 101 +dictGet dict_array (33.464,-1.00949) 101 +dictGet dict_array (33.4692,0.954092) 101 +dictGet dict_array (33.4716,1.9538799999999998) 101 +dictGet dict_array (33.4756,1.85836) 101 +dictGet dict_array (33.4859,4.0751) 101 +dictGet dict_array (33.4899,3.54193) 101 +dictGet dict_array (33.4935,3.49794) 101 +dictGet dict_array (33.494,-0.983356) 101 +dictGet dict_array (33.4955,-1.28128) 101 +dictGet dict_array (33.4965,-0.278687) 101 +dictGet dict_array (33.4991,0.647491) 101 +dictGet dict_array (33.5076,2.2272) 101 +dictGet dict_array (33.5079,-0.498199) 101 +dictGet dict_array (33.5157,0.535034) 101 +dictGet dict_array (33.5171,2.49677) 101 +dictGet dict_array (33.5255,2.4447200000000002) 101 +dictGet dict_array (33.526,4.01194) 101 +dictGet dict_array (33.5288,0.789434) 101 +dictGet dict_array (33.5356,-1.17671) 101 +dictGet dict_array (33.5402,1.49152) 101 +dictGet dict_array (33.5418,3.45757) 101 +dictGet dict_array (33.5428,1.90712) 101 +dictGet dict_array (33.5556,-0.55741) 101 +dictGet dict_array (33.5564,0.876858) 101 +dictGet dict_array (33.5567,-0.10208) 101 +dictGet dict_array (33.5645,-0.124824) 101 +dictGet dict_array (33.5663,3.4872) 101 +dictGet dict_array (33.5716,-0.0107611) 101 +dictGet dict_array (33.578,3.55714) 101 +dictGet dict_array (33.5826,-0.49076) 101 +dictGet dict_array (33.5909,0.773737) 101 +dictGet dict_array (33.5958,2.9619999999999997) 5994231 +dictGet dict_array (33.6193,-0.919755) 101 +dictGet dict_array (33.6313,0.652132) 101 +dictGet dict_array (33.632,0.823351) 101 +dictGet dict_array (33.66,2.18998) 101 +dictGet dict_array (33.6621,0.535395) 101 +dictGet dict_array (33.6726,3.19367) 101 +dictGet dict_array (33.6912,1.74522) 101 +dictGet dict_array (33.705,0.706397) 101 +dictGet dict_array (33.7076,0.7622) 101 +dictGet dict_array (33.7112,1.70187) 101 +dictGet dict_array (33.7246,-1.14837) 101 +dictGet dict_array (33.7326,2.62413) 5994231 +dictGet dict_array (33.7332,2.82137) 5994231 +dictGet dict_array (33.7434,0.394672) 101 +dictGet dict_array (33.7443,1.54557) 101 +dictGet dict_array (33.7506,1.57317) 101 +dictGet dict_array (33.7526,1.8578999999999999) 101 +dictGet dict_array (33.766,4.15013) 101 +dictGet dict_array (33.7834,2.41789) 101 +dictGet dict_array (33.7864,0.230935) 101 +dictGet dict_array (33.7965,3.05709) 101 +dictGet dict_array (33.7998,3.32881) 101 +dictGet dict_array (33.8003,2.97338) 5994231 +dictGet dict_array (33.8007,-1.08962) 101 +dictGet dict_array (33.8022,-0.139488) 101 +dictGet dict_array (33.8065,2.70857) 5994231 +dictGet dict_array (33.8169,-0.607788) 101 +dictGet dict_array (33.8203,0.108512) 101 +dictGet dict_array (33.8231,-1.03449) 101 +dictGet dict_array (33.8312,3.49458) 101 +dictGet dict_array (33.8342,0.297518) 101 +dictGet dict_array (33.8352,0.165872) 101 +dictGet dict_array (33.8354,1.87277) 101 +dictGet dict_array (33.8371,1.60103) 101 +dictGet dict_array (33.8387,1.9968) 101 +dictGet dict_array (33.8403,3.5805) 101 +dictGet dict_array (33.8414,-0.703067) 101 +dictGet dict_array (33.844,-0.179472) 101 +dictGet dict_array (33.8468,3.40137) 101 +dictGet dict_array (33.8509,4.15334) 101 +dictGet dict_array (33.8539,2.38339) 101 +dictGet dict_array (33.858,-1.3122500000000001) 101 +dictGet dict_array (33.859,3.72626) 101 +dictGet dict_array (33.8616,2.24433) 101 +dictGet dict_array (33.8621,3.01035) 101 +dictGet dict_array (33.8623,1.17559) 101 +dictGet dict_array (33.8682,2.706) 5994231 +dictGet dict_array (33.8684,0.189231) 101 +dictGet dict_array (33.872,1.93574) 101 +dictGet dict_array (33.8844,3.80404) 101 +dictGet dict_array (33.8888,0.594884) 101 +dictGet dict_array (33.8946,2.74161) 101 +dictGet dict_array (33.9023,0.6239) 101 +dictGet dict_array (33.9057,0.873222) 101 +dictGet dict_array (33.9157,-1.26607) 101 +dictGet dict_array (33.92,2.06848) 101 +dictGet dict_array (33.9298,-0.00526229) 101 +dictGet dict_array (33.932,3.07063) 101 +dictGet dict_array (33.9322,0.629385) 101 +dictGet dict_array (33.9367,-1.41955) 101 +dictGet dict_array (33.937,1.42532) 101 +dictGet dict_array (33.9375,1.1467100000000001) 101 +dictGet dict_array (33.9434,-1.05739) 101 +dictGet dict_array (33.9477,3.34809) 101 +dictGet dict_array (33.95,2.21715) 101 +dictGet dict_array (33.955799999999996,0.305176) 101 +dictGet dict_array (33.9686,-0.28273) 101 +dictGet dict_array (33.9703,4.1255) 101 +dictGet dict_array (33.9707,3.08199) 101 +dictGet dict_array (33.9754,1.06203) 101 +dictGet dict_array (33.9757,3.72468) 101 +dictGet dict_array (33.9775,-0.0440599) 101 +dictGet dict_array (33.9777,-0.251484) 101 +dictGet dict_array (33.9789,-0.339374) 101 +dictGet dict_array (33.9849,2.54515) 5994231 +dictGet dict_array (33.9885,-0.318557) 101 +dictGet dict_array (33.9977,1.07175) 101 +dictGet dict_array (33.9984,-0.700517) 101 +dictGet dict_array (34.0149,3.53338) 101 +dictGet dict_array (34.0173,3.39155) 101 +dictGet dict_array (34.0317,3.9579) 101 +dictGet dict_array (34.0369,3.83612) 101 +dictGet dict_array (34.043,-0.0887221) 101 +dictGet dict_array (34.0487,1.14252) 101 +dictGet dict_array (34.052,1.74832) 101 +dictGet dict_array (34.0711,-0.898071) 101 +dictGet dict_array (34.0747,1.55057) 101 +dictGet dict_array (34.0803,3.16763) 101 +dictGet dict_array (34.0872,3.75555) 101 +dictGet dict_array (34.0965,1.62038) 101 +dictGet dict_array (34.0977,-0.412691) 101 +dictGet dict_array (34.0986,0.0294206) 101 +dictGet dict_array (34.1072,3.15823) 101 +dictGet dict_array (34.1092,3.09599) 101 +dictGet dict_array (34.1206,1.04637) 5940222 +dictGet dict_array (34.1209,3.13826) 101 +dictGet dict_array (34.1265,3.95881) 101 +dictGet dict_array (34.1286,-0.539319) 101 +dictGet dict_array (34.1358,3.67451) 101 +dictGet dict_array (34.1428,0.136115) 101 +dictGet dict_array (34.157,1.73522) 101 +dictGet dict_array (34.1581,1.48001) 101 +dictGet dict_array (34.1682,3.42373) 101 +dictGet dict_array (34.1683,-1.26511) 101 +dictGet dict_array (34.1684,4.20007) 101 +dictGet dict_array (34.1854,3.32089) 101 +dictGet dict_array (34.2022,0.749536) 101 +dictGet dict_array (34.2044,3.04865) 101 +dictGet dict_array (34.22,-0.500055) 101 +dictGet dict_array (34.2249,0.743775) 101 +dictGet dict_array (34.2254,1.34702) 101 +dictGet dict_array (34.2355,-0.898843) 101 +dictGet dict_array (34.2394,2.0203699999999998) 101 +dictGet dict_array (34.2466,1.83785) 101 +dictGet dict_array (34.247,4.09563) 101 +dictGet dict_array (34.2508,2.61312) 101 +dictGet dict_array (34.2517,1.69642) 101 +dictGet dict_array (34.2564,4.13033) 101 +dictGet dict_array (34.2574,4.18928) 101 +dictGet dict_array (34.2614,-0.478719) 101 +dictGet dict_array (34.2625,2.38088) 101 +dictGet dict_array (34.2666,3.1503) 101 +dictGet dict_array (34.271,4.02223) 101 +dictGet dict_array (34.2727,0.514755) 101 +dictGet dict_array (34.278,1.98929) 101 +dictGet dict_array (34.2798,-0.199208) 101 +dictGet dict_array (34.2804,2.05184) 101 +dictGet dict_array (34.2945,-1.11051) 101 +dictGet dict_array (34.3168,-0.0829721) 101 +dictGet dict_array (34.3345,3.4358) 101 +dictGet dict_array (34.3377,1.13527) 5940222 +dictGet dict_array (34.3383,1.27891) 5940222 +dictGet dict_array (34.3391,1.47945) 5940222 +dictGet dict_array (34.3441,0.627014) 101 +dictGet dict_array (34.347,2.4853) 101 +dictGet dict_array (34.3514,2.16247) 101 +dictGet dict_array (34.3627,2.64533) 101 +dictGet dict_array (34.3682,-0.227501) 101 +dictGet dict_array (34.3756,4.21248) 101 +dictGet dict_array (34.379,3.96604) 101 +dictGet dict_array (34.3827,1.7518) 101 +dictGet dict_array (34.3912,2.8834) 101 +dictGet dict_array (34.3919,0.668829) 101 +dictGet dict_array (34.3949,2.00338) 101 +dictGet dict_array (34.3987,0.557268) 101 +dictGet dict_array (34.4111,0.768558) 101 +dictGet dict_array (34.4119,2.8742) 101 +dictGet dict_array (34.416,3.50841) 101 +dictGet dict_array (34.4212,1.24916) 5940222 +dictGet dict_array (34.4251,0.457029) 101 +dictGet dict_array (34.4274,-0.902559) 101 +dictGet dict_array (34.4325,4.03159) 101 +dictGet dict_array (34.438,1.63994) 101 +dictGet dict_array (34.4403,-0.177594) 101 +dictGet dict_array (34.4421,0.726712) 101 +dictGet dict_array (34.4517,2.98611) 101 +dictGet dict_array (34.4658,-1.312) 101 +dictGet dict_array (34.4732,-0.0681338) 101 +dictGet dict_array (34.4752,2.81646) 101 +dictGet dict_array (34.4914,2.3858) 101 +dictGet dict_array (34.4923,0.855231) 101 +dictGet dict_array (34.5235,1.78468) 101 +dictGet dict_array (34.5305,4.10608) 101 +dictGet dict_array (34.5389,0.621937) 101 +dictGet dict_array (34.5406,3.17145) 101 +dictGet dict_array (34.5434,-0.56306) 101 +dictGet dict_array (34.5449,3.13311) 101 +dictGet dict_array (34.5491,2.31572) 101 +dictGet dict_array (34.5539,2.94028) 101 +dictGet dict_array (34.5546,-0.208825) 101 +dictGet dict_array (34.5549,3.78486) 101 +dictGet dict_array (34.5676,0.307148) 101 +dictGet dict_array (34.5743,1.5217399999999999) 101 +dictGet dict_array (34.5775,3.48046) 101 +dictGet dict_array (34.5815,2.5243700000000002) 101 +dictGet dict_array (34.5841,4.21191) 101 +dictGet dict_array (34.5887,2.65083) 101 +dictGet dict_array (34.5937,3.2143) 101 +dictGet dict_array (34.6013,-1.0612) 101 +dictGet dict_array (34.6089,1.36066) 101 +dictGet dict_array (34.6103,3.40227) 101 +dictGet dict_array (34.6128,1.92276) 101 +dictGet dict_array (34.6175,2.43627) 101 +dictGet dict_array (34.6209,3.43776) 101 +dictGet dict_array (34.6234,2.60237) 101 +dictGet dict_array (34.6275,3.52479) 101 +dictGet dict_array (34.635,0.568558) 101 +dictGet dict_array (34.6373,2.37692) 101 +dictGet dict_array (34.6375,3.52234) 101 +dictGet dict_array (34.6426,2.12397) 101 +dictGet dict_array (34.6513,2.80915) 101 +dictGet dict_array (34.6632,2.30039) 101 +dictGet dict_array (34.6691,1.86582) 101 +dictGet dict_array (34.6739,0.15342) 101 +dictGet dict_array (34.6825,0.0499679) 101 +dictGet dict_array (34.6893,0.454326) 101 +dictGet dict_array (34.6957,-0.358598) 101 +dictGet dict_array (34.6986,0.562679) 101 +dictGet dict_array (34.712,1.12114) 101 +dictGet dict_array (34.7126,-0.0057301) 101 +dictGet dict_array (34.7137,0.0248501) 101 +dictGet dict_array (34.7162,1.15623) 101 +dictGet dict_array (34.7258,3.95142) 101 +dictGet dict_array (34.7347,3.5232099999999997) 101 +dictGet dict_array (34.7363,2.23374) 101 +dictGet dict_array (34.7375,0.397841) 101 +dictGet dict_array (34.7423,3.09198) 101 +dictGet dict_array (34.7452,3.09029) 101 +dictGet dict_array (34.7539,-1.06943) 101 +dictGet dict_array (34.7733,-0.00912717) 101 +dictGet dict_array (34.774,2.71088) 101 +dictGet dict_array (34.7771,1.46009) 101 +dictGet dict_array (34.7782,-1.28308) 101 +dictGet dict_array (34.7924,3.63564) 101 +dictGet dict_array (34.7939,-0.416676) 101 +dictGet dict_array (34.7964,-0.401773) 101 +dictGet dict_array (34.7974,0.0286873) 101 +dictGet dict_array (34.7975,3.05965) 101 +dictGet dict_array (34.8037,3.07263) 101 +dictGet dict_array (34.8254,-0.390284) 101 +dictGet dict_array (34.828,1.91869) 101 +dictGet dict_array (34.8289,3.71058) 101 +dictGet dict_array (34.8403,2.14606) 101 +dictGet dict_array (34.8437,2.20617) 101 +dictGet dict_array (34.8469,2.38435) 101 +dictGet dict_array (34.86,1.45705) 101 +dictGet dict_array (34.8612,0.914248) 101 +dictGet dict_array (34.8663,3.4215400000000002) 101 +dictGet dict_array (34.8724,-0.375144) 101 +dictGet dict_array (34.8795,3.29317) 101 +dictGet dict_array (34.8823,1.21988) 101 +dictGet dict_array (34.8834,1.07657) 101 +dictGet dict_array (34.8837,0.157648) 101 +dictGet dict_array (34.8871,-0.9755) 101 +dictGet dict_array (34.8871,1.8943699999999999) 101 +dictGet dict_array (34.889,3.36756) 101 +dictGet dict_array (34.8907,1.24874) 101 +dictGet dict_array (34.8965,3.13508) 101 +dictGet dict_array (34.9042,2.62092) 101 +dictGet dict_array (34.9055,-0.0448967) 101 +dictGet dict_array (34.9122,0.110576) 101 +dictGet dict_array (34.9228,3.60183) 101 +dictGet dict_array (34.9237,1.21715) 101 +dictGet dict_array (34.9296,1.70459) 101 +dictGet dict_array (34.941,-1.14663) 101 +dictGet dict_array (34.9448,1.18923) 101 +dictGet dict_array (34.9462,3.81678) 101 +dictGet dict_array (34.9466,0.593463) 101 +dictGet dict_array (34.9485,0.150307) 101 +dictGet dict_array (34.9542,0.487238) 101 +dictGet dict_array (34.9559,2.03473) 101 +dictGet dict_array (34.9671,-0.960225) 101 +dictGet dict_array (34.9711,2.63444) 101 +dictGet dict_array (34.9892,0.354775) 101 +dictGet dict_array (34.9907,1.40724) 101 +dictGet dict_array (34.9916,-0.00173097) 101 +dictGet dict_array (34.9919,2.06167) 101 diff --git a/tests/queries/0_stateless/01037_polygon_dicts_correctness_all.sh b/tests/queries/0_stateless/01037_polygon_dicts_correctness_all.sh index c2a35a3ef63..9a26f78a8ee 100755 --- a/tests/queries/0_stateless/01037_polygon_dicts_correctness_all.sh +++ b/tests/queries/0_stateless/01037_polygon_dicts_correctness_all.sh @@ -1,31 +1,32 @@ #!/usr/bin/env bash -# Tags: no-debug, no-parallel +# Tags: no-debug CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -TMP_DIR="/tmp" +TMP_DIR=${CLICKHOUSE_TMP}/tmp +DATA_DIR=${CLICKHOUSE_TMP}/data +mkdir -p $TMP_DIR +mkdir -p $DATA_DIR declare -a SearchTypes=("POLYGON" "POLYGON_SIMPLE" "POLYGON_INDEX_EACH" "POLYGON_INDEX_CELL") -tar -xf "${CURDIR}"/01037_test_data_search.tar.gz -C "${CURDIR}" +tar -xf "${CURDIR}"/01037_test_data_search.tar.gz -C "${DATA_DIR}" $CLICKHOUSE_CLIENT -n --query=" -DROP DATABASE IF EXISTS test_01037; -CREATE DATABASE test_01037; -DROP TABLE IF EXISTS test_01037.points; -CREATE TABLE test_01037.points (x Float64, y Float64) ENGINE = Memory; +DROP TABLE IF EXISTS points; +CREATE TABLE points (x Float64, y Float64) ENGINE = Memory; " -$CLICKHOUSE_CLIENT --query="INSERT INTO test_01037.points FORMAT TSV" --max_insert_block_size=100000 < "${CURDIR}/01037_point_data" +$CLICKHOUSE_CLIENT --query="INSERT INTO points FORMAT TSV" --max_insert_block_size=100000 < "${DATA_DIR}/01037_point_data" -rm "${CURDIR}"/01037_point_data +rm "${DATA_DIR}"/01037_point_data $CLICKHOUSE_CLIENT -n --query=" -DROP TABLE IF EXISTS test_01037.polygons_array; +DROP TABLE IF EXISTS polygons_array; -CREATE TABLE test_01037.polygons_array +CREATE TABLE polygons_array ( key Array(Array(Array(Array(Float64)))), name String, @@ -34,36 +35,31 @@ CREATE TABLE test_01037.polygons_array ENGINE = Memory; " -$CLICKHOUSE_CLIENT --query="INSERT INTO test_01037.polygons_array FORMAT JSONEachRow" --min_chunk_bytes_for_parallel_parsing=10485760 --max_insert_block_size=100000 < "${CURDIR}/01037_polygon_data" +$CLICKHOUSE_CLIENT --query="INSERT INTO polygons_array FORMAT JSONEachRow" --min_chunk_bytes_for_parallel_parsing=10485760 --max_insert_block_size=100000 < "${DATA_DIR}/01037_polygon_data" -rm "${CURDIR}"/01037_polygon_data +rm "${DATA_DIR}"/01037_polygon_data for type in "${SearchTypes[@]}"; do outputFile="${TMP_DIR}/results${type}.out" $CLICKHOUSE_CLIENT -n --query=" - DROP DICTIONARY IF EXISTS test_01037.dict_array; + DROP DICTIONARY IF EXISTS dict_array; - CREATE DICTIONARY test_01037.dict_array + CREATE DICTIONARY dict_array ( key Array(Array(Array(Array(Float64)))), name String DEFAULT 'qqq', value UInt64 DEFAULT 101 ) PRIMARY KEY key - SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'polygons_array' PASSWORD '' DB 'test_01037')) + SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'polygons_array' PASSWORD '' DB currentDatabase())) LIFETIME(0) LAYOUT($type()); - select 'dictGet', 'test_01037.dict_array' as dict_name, tuple(x, y) as key, - dictGet(dict_name, 'value', key) from test_01037.points order by x, y; + select 'dictGet', 'dict_array' as dict_name, tuple(x, y) as key, + dictGet(dict_name, 'value', key) from points order by x, y; " > "$outputFile" diff -q "${CURDIR}/01037_polygon_dicts_correctness_all.ans" "$outputFile" done - -$CLICKHOUSE_CLIENT -n --query=" -DROP TABLE test_01037.points; -DROP DATABASE test_01037; -" diff --git a/tests/queries/0_stateless/01037_polygon_dicts_correctness_fast.ans b/tests/queries/0_stateless/01037_polygon_dicts_correctness_fast.ans index 45fa7637421..297c8416096 100644 --- a/tests/queries/0_stateless/01037_polygon_dicts_correctness_fast.ans +++ b/tests/queries/0_stateless/01037_polygon_dicts_correctness_fast.ans @@ -1,1000 +1,1000 @@ -dictGet test_01037.dict_array (29.5699,2.50068) 101 -dictGet test_01037.dict_array (29.5796,1.55456) 101 -dictGet test_01037.dict_array (29.5796,2.36864) 101 -dictGet test_01037.dict_array (29.5844,1.59626) 101 -dictGet test_01037.dict_array (29.5886,4.03321) 101 -dictGet test_01037.dict_array (29.5914,3.02628) 101 -dictGet test_01037.dict_array (29.5926,-0.0965169) 101 -dictGet test_01037.dict_array (29.5968,2.37773) 101 -dictGet test_01037.dict_array (29.5984,0.755853) 101 -dictGet test_01037.dict_array (29.6066,3.47173) 101 -dictGet test_01037.dict_array (29.6085,-1.26007) 6489978 -dictGet test_01037.dict_array (29.6131,0.246565) 101 -dictGet test_01037.dict_array (29.6157,-0.266687) 101 -dictGet test_01037.dict_array (29.6164,2.94674) 101 -dictGet test_01037.dict_array (29.6195,-0.591941) 101 -dictGet test_01037.dict_array (29.6231,1.54818) 101 -dictGet test_01037.dict_array (29.6379,0.764114) 101 -dictGet test_01037.dict_array (29.6462,-0.772059) 934530 -dictGet test_01037.dict_array (29.6579,-1.07336) 6489978 -dictGet test_01037.dict_array (29.6618,-0.271842) 101 -dictGet test_01037.dict_array (29.6629,-0.303602) 101 -dictGet test_01037.dict_array (29.6659,-0.782823) 934530 -dictGet test_01037.dict_array (29.6736,-0.113832) 101 -dictGet test_01037.dict_array (29.6759,3.02905) 101 -dictGet test_01037.dict_array (29.6778,3.71898) 101 -dictGet test_01037.dict_array (29.6796,1.10433) 101 -dictGet test_01037.dict_array (29.6809,2.13677) 101 -dictGet test_01037.dict_array (29.6935,4.11894) 101 -dictGet test_01037.dict_array (29.6991,-1.4458199999999999) 101 -dictGet test_01037.dict_array (29.6997,3.17297) 101 -dictGet test_01037.dict_array (29.7043,3.6145899999999997) 101 -dictGet test_01037.dict_array (29.7065,3.24885) 101 -dictGet test_01037.dict_array (29.7126,0.28108) 101 -dictGet test_01037.dict_array (29.7192,0.174273) 101 -dictGet test_01037.dict_array (29.7217,-0.523481) 3501900 -dictGet test_01037.dict_array (29.7271,1.67967) 101 -dictGet test_01037.dict_array (29.7311,4.12444) 101 -dictGet test_01037.dict_array (29.7347,1.88378) 101 -dictGet test_01037.dict_array (29.7358,0.67944) 101 -dictGet test_01037.dict_array (29.7366,-0.2973) 101 -dictGet test_01037.dict_array (29.7446,0.646536) 101 -dictGet test_01037.dict_array (29.7453,-0.567963) 3501900 -dictGet test_01037.dict_array (29.764,4.04217) 101 -dictGet test_01037.dict_array (29.7655,1.51372) 101 -dictGet test_01037.dict_array (29.7744,1.12435) 101 -dictGet test_01037.dict_array (29.7774,-0.0681196) 3501895 -dictGet test_01037.dict_array (29.7784,1.54864) 101 -dictGet test_01037.dict_array (29.7785,2.24139) 101 -dictGet test_01037.dict_array (29.7922,0.220808) 101 -dictGet test_01037.dict_array (29.7936,2.37709) 101 -dictGet test_01037.dict_array (29.8008,0.948536) 101 -dictGet test_01037.dict_array (29.8115,0.201227) 101 -dictGet test_01037.dict_array (29.814,0.149601) 3501895 -dictGet test_01037.dict_array (29.8193,-1.35858) 101 -dictGet test_01037.dict_array (29.8201,0.965518) 101 -dictGet test_01037.dict_array (29.8265,-0.727286) 3501900 -dictGet test_01037.dict_array (29.8277,-0.531746) 3501900 -dictGet test_01037.dict_array (29.8289,3.63009) 101 -dictGet test_01037.dict_array (29.8548,0.838047) 101 -dictGet test_01037.dict_array (29.8641,-0.845265) 3501900 -dictGet test_01037.dict_array (29.8649,0.0562212) 3501895 -dictGet test_01037.dict_array (29.8701,-1.02045) 934530 -dictGet test_01037.dict_array (29.8733,2.76654) 101 -dictGet test_01037.dict_array (29.876,0.555475) 101 -dictGet test_01037.dict_array (29.8794,-0.800108) 3501900 -dictGet test_01037.dict_array (29.8813,2.7426399999999997) 101 -dictGet test_01037.dict_array (29.897100000000002,2.66193) 101 -dictGet test_01037.dict_array (29.908,4.01339) 101 -dictGet test_01037.dict_array (29.9165,-1.08246) 3501894 -dictGet test_01037.dict_array (29.9201,-0.420861) 3498054 -dictGet test_01037.dict_array (29.9217,3.03778) 101 -dictGet test_01037.dict_array (29.9355,0.773833) 101 -dictGet test_01037.dict_array (29.947,3.76517) 101 -dictGet test_01037.dict_array (29.9518,-0.60557) 3498056 -dictGet test_01037.dict_array (29.9564,-0.600163) 3498056 -dictGet test_01037.dict_array (29.959600000000002,4.16591) 101 -dictGet test_01037.dict_array (29.9615,-1.33708) 3501894 -dictGet test_01037.dict_array (29.9699,-0.392375) 3498054 -dictGet test_01037.dict_array (29.9776,1.04552) 101 -dictGet test_01037.dict_array (29.9784,4.02756) 101 -dictGet test_01037.dict_array (29.9819,4.00597) 101 -dictGet test_01037.dict_array (29.9826,1.2816100000000001) 101 -dictGet test_01037.dict_array (30.0026,2.76257) 101 -dictGet test_01037.dict_array (30.0126,3.68255) 101 -dictGet test_01037.dict_array (30.0131,0.796576) 3501892 -dictGet test_01037.dict_array (30.018,1.16523) 101 -dictGet test_01037.dict_array (30.0261,-0.210653) 3501896 -dictGet test_01037.dict_array (30.0472,-1.11007) 3501894 -dictGet test_01037.dict_array (30.0542,-0.479585) 3498054 -dictGet test_01037.dict_array (30.0613,1.6278000000000001) 101 -dictGet test_01037.dict_array (30.0617,-0.0551152) 3501895 -dictGet test_01037.dict_array (30.0637,2.62066) 101 -dictGet test_01037.dict_array (30.0721,1.6424400000000001) 101 -dictGet test_01037.dict_array (30.0769,-0.402636) 3498054 -dictGet test_01037.dict_array (30.0791,-0.277435) 3501896 -dictGet test_01037.dict_array (30.0931,0.0327512) 3501895 -dictGet test_01037.dict_array (30.1059,3.52623) 101 -dictGet test_01037.dict_array (30.1103,0.865466) 3501892 -dictGet test_01037.dict_array (30.1115,2.95243) 101 -dictGet test_01037.dict_array (30.1144,1.71029) 101 -dictGet test_01037.dict_array (30.1311,-0.864751) 3501899 -dictGet test_01037.dict_array (30.1336,-0.851386) 3501899 -dictGet test_01037.dict_array (30.1393,3.89901) 101 -dictGet test_01037.dict_array (30.1456,-0.531898) 3498054 -dictGet test_01037.dict_array (30.1492,2.07833) 101 -dictGet test_01037.dict_array (30.1575,2.43856) 101 -dictGet test_01037.dict_array (30.1682,1.19771) 101 -dictGet test_01037.dict_array (30.1716,3.9853300000000003) 101 -dictGet test_01037.dict_array (30.1849,2.78374) 101 -dictGet test_01037.dict_array (30.1866,0.65658) 3498021 -dictGet test_01037.dict_array (30.1885,1.56943) 101 -dictGet test_01037.dict_array (30.1959,-1.38202) 101 -dictGet test_01037.dict_array (30.1999,1.58413) 101 -dictGet test_01037.dict_array (30.2024,0.713081) 3498021 -dictGet test_01037.dict_array (30.2054,0.620143) 3498021 -dictGet test_01037.dict_array (30.2091,1.51641) 101 -dictGet test_01037.dict_array (30.2124,-0.331782) 3498031 -dictGet test_01037.dict_array (30.226,3.03527) 101 -dictGet test_01037.dict_array (30.2261,3.18486) 101 -dictGet test_01037.dict_array (30.2288,2.48407) 101 -dictGet test_01037.dict_array (30.2345,3.7462400000000002) 101 -dictGet test_01037.dict_array (30.2375,0.62046) 3498021 -dictGet test_01037.dict_array (30.2425,-0.472914) 3498054 -dictGet test_01037.dict_array (30.247,3.95863) 101 -dictGet test_01037.dict_array (30.2494,-0.305093) 3498031 -dictGet test_01037.dict_array (30.2499,2.54337) 101 -dictGet test_01037.dict_array (30.2606,2.16644) 101 -dictGet test_01037.dict_array (30.2672,3.94847) 101 -dictGet test_01037.dict_array (30.2709,-0.136264) 6088794 -dictGet test_01037.dict_array (30.2764,1.18654) 101 -dictGet test_01037.dict_array (30.2765,1.20383) 101 -dictGet test_01037.dict_array (30.2839,1.05762) 3498024 -dictGet test_01037.dict_array (30.286,0.469327) 3498021 -dictGet test_01037.dict_array (30.2927,3.1693) 101 -dictGet test_01037.dict_array (30.2935,3.49854) 101 -dictGet test_01037.dict_array (30.307,0.312338) 3498021 -dictGet test_01037.dict_array (30.3085,1.07791) 3498024 -dictGet test_01037.dict_array (30.3139,2.77248) 101 -dictGet test_01037.dict_array (30.314,0.822823) 3498024 -dictGet test_01037.dict_array (30.3227,-0.587351) 3498055 -dictGet test_01037.dict_array (30.332,1.00174) 3498024 -dictGet test_01037.dict_array (30.3388,0.844148) 3498024 -dictGet test_01037.dict_array (30.3485,0.561902) 3498021 -dictGet test_01037.dict_array (30.3497,0.180362) 6489998 -dictGet test_01037.dict_array (30.361,4.13016) 101 -dictGet test_01037.dict_array (30.3623,-0.0484027) 6489998 -dictGet test_01037.dict_array (30.3638,3.9845800000000002) 101 -dictGet test_01037.dict_array (30.3853,3.16051) 101 -dictGet test_01037.dict_array (30.3974,2.6617800000000003) 101 -dictGet test_01037.dict_array (30.4002,-1.15886) 101 -dictGet test_01037.dict_array (30.4008,-0.387015) 3498031 -dictGet test_01037.dict_array (30.4018,1.86493) 101 -dictGet test_01037.dict_array (30.4239,1.16818) 3498024 -dictGet test_01037.dict_array (30.4363,3.63938) 101 -dictGet test_01037.dict_array (30.4377,-0.81315) 3498063 -dictGet test_01037.dict_array (30.4391,3.54703) 101 -dictGet test_01037.dict_array (30.4424,-1.39435) 101 -dictGet test_01037.dict_array (30.4441,2.8463000000000003) 101 -dictGet test_01037.dict_array (30.4517,3.28117) 101 -dictGet test_01037.dict_array (30.4658,2.6928) 101 -dictGet test_01037.dict_array (30.4734,2.66161) 101 -dictGet test_01037.dict_array (30.4799,-1.07578) 101 -dictGet test_01037.dict_array (30.4837,-1.02486) 3501899 -dictGet test_01037.dict_array (30.485,1.06326) 3498024 -dictGet test_01037.dict_array (30.495,1.12306) 101 -dictGet test_01037.dict_array (30.501,2.27264) 101 -dictGet test_01037.dict_array (30.5027,1.99382) 101 -dictGet test_01037.dict_array (30.5194,-1.03943) 3501893 -dictGet test_01037.dict_array (30.5239,1.04328) 101 -dictGet test_01037.dict_array (30.528,3.82041) 101 -dictGet test_01037.dict_array (30.5299,-0.715248) 3498063 -dictGet test_01037.dict_array (30.5331,1.19603) 101 -dictGet test_01037.dict_array (30.535800000000002,2.71485) 101 -dictGet test_01037.dict_array (30.5405,0.804694) 3498023 -dictGet test_01037.dict_array (30.542,1.23739) 101 -dictGet test_01037.dict_array (30.5432,4.04189) 101 -dictGet test_01037.dict_array (30.5457,-0.956121) 3501893 -dictGet test_01037.dict_array (30.5506,3.07443) 101 -dictGet test_01037.dict_array (30.5539,3.87084) 101 -dictGet test_01037.dict_array (30.5578,3.78837) 101 -dictGet test_01037.dict_array (30.5588,0.966135) 3498022 -dictGet test_01037.dict_array (30.5637,2.5605) 101 -dictGet test_01037.dict_array (30.5647,-1.27328) 101 -dictGet test_01037.dict_array (30.5656,-0.0581332) 6088794 -dictGet test_01037.dict_array (30.5715,0.65755) 3498023 -dictGet test_01037.dict_array (30.5727,3.01604) 101 -dictGet test_01037.dict_array (30.5729,-0.976857) 3501893 -dictGet test_01037.dict_array (30.5751,0.60204) 3498023 -dictGet test_01037.dict_array (30.5854,3.02473) 101 -dictGet test_01037.dict_array (30.5866,0.174099) 6489998 -dictGet test_01037.dict_array (30.5947,0.875193) 3498023 -dictGet test_01037.dict_array (30.5992,-0.403901) 3498063 -dictGet test_01037.dict_array (30.6002,4.18891) 101 -dictGet test_01037.dict_array (30.6025,0.217712) 6489998 -dictGet test_01037.dict_array (30.6054,0.927203) 3498022 -dictGet test_01037.dict_array (30.6075,3.79359) 101 -dictGet test_01037.dict_array (30.6159,3.82773) 101 -dictGet test_01037.dict_array (30.627,3.84039) 101 -dictGet test_01037.dict_array (30.6308,0.77517) 3498023 -dictGet test_01037.dict_array (30.6338,0.179565) 6489998 -dictGet test_01037.dict_array (30.6461,1.3293599999999999) 101 -dictGet test_01037.dict_array (30.6674,-0.424547) 3498063 -dictGet test_01037.dict_array (30.669,1.76539) 101 -dictGet test_01037.dict_array (30.6788,4.01239) 101 -dictGet test_01037.dict_array (30.6864,3.59158) 101 -dictGet test_01037.dict_array (30.7049,-0.875413) 3501893 -dictGet test_01037.dict_array (30.705,1.3307) 101 -dictGet test_01037.dict_array (30.7063,-0.473192) 3498063 -dictGet test_01037.dict_array (30.7075,-1.1958199999999999) 101 -dictGet test_01037.dict_array (30.7101,-0.367562) 3498012 -dictGet test_01037.dict_array (30.7203,2.98725) 101 -dictGet test_01037.dict_array (30.7213,2.2745699999999998) 101 -dictGet test_01037.dict_array (30.7446,-0.334144) 3498012 -dictGet test_01037.dict_array (30.7468,3.82967) 101 -dictGet test_01037.dict_array (30.747,-0.384779) 3498012 -dictGet test_01037.dict_array (30.7681,0.904198) 3498022 -dictGet test_01037.dict_array (30.7757,1.78743) 101 -dictGet test_01037.dict_array (30.8021,-0.479212) 3498012 -dictGet test_01037.dict_array (30.8079,-1.40869) 101 -dictGet test_01037.dict_array (30.8206,-0.0608489) 3498012 -dictGet test_01037.dict_array (30.8218,0.43909) 3498023 -dictGet test_01037.dict_array (30.8239,0.10014) 3498012 -dictGet test_01037.dict_array (30.8282,4.15409) 101 -dictGet test_01037.dict_array (30.8288,-0.709528) 3501893 -dictGet test_01037.dict_array (30.8326,0.156011) 3498012 -dictGet test_01037.dict_array (30.8328,-1.03704) 101 -dictGet test_01037.dict_array (30.839,2.15528) 101 -dictGet test_01037.dict_array (30.8452,0.219377) 3498013 -dictGet test_01037.dict_array (30.8463,0.0515355) 3498012 -dictGet test_01037.dict_array (30.8526,2.06614) 101 -dictGet test_01037.dict_array (30.8566,0.517876) 3498023 -dictGet test_01037.dict_array (30.8588,-1.31738) 101 -dictGet test_01037.dict_array (30.8681,0.44207) 3498013 -dictGet test_01037.dict_array (30.8914,1.0072) 3498022 -dictGet test_01037.dict_array (30.897,0.483425) 3498013 -dictGet test_01037.dict_array (30.905,2.8731999999999998) 3501793 -dictGet test_01037.dict_array (30.9051,2.21956) 101 -dictGet test_01037.dict_array (30.9115,4.00663) 101 -dictGet test_01037.dict_array (30.9167,-0.834462) 3501893 -dictGet test_01037.dict_array (30.9252,-1.3289900000000001) 101 -dictGet test_01037.dict_array (30.9314,1.85384) 101 -dictGet test_01037.dict_array (30.9392,2.53236) 3501827 -dictGet test_01037.dict_array (30.9569,2.82038) 3501793 -dictGet test_01037.dict_array (30.9598,-0.641011) 3498012 -dictGet test_01037.dict_array (30.9601,-0.254928) 3498012 -dictGet test_01037.dict_array (30.9623,-1.3886) 101 -dictGet test_01037.dict_array (30.9707,0.888854) 3498022 -dictGet test_01037.dict_array (30.9766,2.81957) 3501793 -dictGet test_01037.dict_array (30.9775,2.69273) 3501793 -dictGet test_01037.dict_array (30.9821,0.587715) 3498013 -dictGet test_01037.dict_array (30.9887,4.0233) 101 -dictGet test_01037.dict_array (30.9914,0.259542) 3498013 -dictGet test_01037.dict_array (30.9986,-1.36832) 101 -dictGet test_01037.dict_array (31.008,0.628999) 3498013 -dictGet test_01037.dict_array (31.0168,-1.17462) 101 -dictGet test_01037.dict_array (31.0237,3.52547) 3501821 -dictGet test_01037.dict_array (31.0306,3.78522) 101 -dictGet test_01037.dict_array (31.0308,-0.72453) 3501893 -dictGet test_01037.dict_array (31.0463,2.41997) 3501825 -dictGet test_01037.dict_array (31.047,0.624184) 3498013 -dictGet test_01037.dict_array (31.0569,0.0706393) 3498015 -dictGet test_01037.dict_array (31.0583,1.3244099999999999) 3501926 -dictGet test_01037.dict_array (31.063,3.23861) 3501793 -dictGet test_01037.dict_array (31.068,0.695575) 3498022 -dictGet test_01037.dict_array (31.0687,1.85675) 101 -dictGet test_01037.dict_array (31.0692,0.254793) 3498014 -dictGet test_01037.dict_array (31.0766,0.828128) 3498022 -dictGet test_01037.dict_array (31.0833,0.0612782) 3498015 -dictGet test_01037.dict_array (31.0833,2.59748) 3501793 -dictGet test_01037.dict_array (31.0861,-1.3778299999999999) 101 -dictGet test_01037.dict_array (31.0874,3.07258) 3501793 -dictGet test_01037.dict_array (31.0882,1.4882) 3501926 -dictGet test_01037.dict_array (31.0924,3.42242) 3501821 -dictGet test_01037.dict_array (31.0927,2.67448) 3501793 -dictGet test_01037.dict_array (31.0936,1.12292) 3498022 -dictGet test_01037.dict_array (31.0952,-0.336928) 3498012 -dictGet test_01037.dict_array (31.0978,3.48482) 3501826 -dictGet test_01037.dict_array (31.1107,3.7513199999999998) 3501826 -dictGet test_01037.dict_array (31.1156,1.19171) 3501926 -dictGet test_01037.dict_array (31.1176,0.223509) 3498015 -dictGet test_01037.dict_array (31.1249,0.946838) 3498022 -dictGet test_01037.dict_array (31.1267,1.48983) 3501926 -dictGet test_01037.dict_array (31.138,-0.289981) 3501898 -dictGet test_01037.dict_array (31.1382,3.02904) 3501793 -dictGet test_01037.dict_array (31.1475,2.6178) 3501793 -dictGet test_01037.dict_array (31.1491,1.37873) 3501926 -dictGet test_01037.dict_array (31.1525,3.72105) 3501826 -dictGet test_01037.dict_array (31.1526,-1.4129800000000001) 101 -dictGet test_01037.dict_array (31.1526,-0.186457) 3501898 -dictGet test_01037.dict_array (31.1539,2.78789) 3501793 -dictGet test_01037.dict_array (31.1548,-1.08552) 101 -dictGet test_01037.dict_array (31.1567,-0.0768925) 3501898 -dictGet test_01037.dict_array (31.1613,1.49617) 3501926 -dictGet test_01037.dict_array (31.1653,1.03777) 3498022 -dictGet test_01037.dict_array (31.1662,3.4214700000000002) 3501826 -dictGet test_01037.dict_array (31.1672,-0.0813169) 3501898 -dictGet test_01037.dict_array (31.177,0.440843) 3498014 -dictGet test_01037.dict_array (31.1788,-0.737151) 3501893 -dictGet test_01037.dict_array (31.1856,-0.144396) 3501898 -dictGet test_01037.dict_array (31.1959,3.66813) 3501826 -dictGet test_01037.dict_array (31.1996,-0.353983) 3501898 -dictGet test_01037.dict_array (31.2019,2.86802) 3501793 -dictGet test_01037.dict_array (31.2087,2.31245) 3501825 -dictGet test_01037.dict_array (31.2125,3.2713200000000002) 3501793 -dictGet test_01037.dict_array (31.2137,-0.108129) 3501898 -dictGet test_01037.dict_array (31.216,3.9156) 101 -dictGet test_01037.dict_array (31.2201,-0.202141) 3501898 -dictGet test_01037.dict_array (31.2285,2.09058) 101 -dictGet test_01037.dict_array (31.2502,4.01526) 101 -dictGet test_01037.dict_array (31.2585,3.11524) 3501793 -dictGet test_01037.dict_array (31.2645,-0.620418) 3501890 -dictGet test_01037.dict_array (31.2684,2.74277) 3501793 -dictGet test_01037.dict_array (31.2821,-1.12772) 101 -dictGet test_01037.dict_array (31.2821,2.46769) 3501825 -dictGet test_01037.dict_array (31.2887,3.91396) 101 -dictGet test_01037.dict_array (31.295,1.49942) 3501926 -dictGet test_01037.dict_array (31.2997,3.46122) 3501826 -dictGet test_01037.dict_array (31.3017,3.3263) 3501826 -dictGet test_01037.dict_array (31.3022,3.16754) 3501793 -dictGet test_01037.dict_array (31.3048,0.364962) 3498014 -dictGet test_01037.dict_array (31.305,3.1967) 3501793 -dictGet test_01037.dict_array (31.3061,1.84303) 101 -dictGet test_01037.dict_array (31.3082,-0.173851) 3501898 -dictGet test_01037.dict_array (31.3315,3.90932) 101 -dictGet test_01037.dict_array (31.3351,2.80164) 3501793 -dictGet test_01037.dict_array (31.3388,0.168765) 3498015 -dictGet test_01037.dict_array (31.339,0.25535) 3498094 -dictGet test_01037.dict_array (31.3423,1.7036799999999999) 3501926 -dictGet test_01037.dict_array (31.349,0.386456) 3498014 -dictGet test_01037.dict_array (31.3558,-1.04336) 101 -dictGet test_01037.dict_array (31.3564,0.478876) 3498014 -dictGet test_01037.dict_array (31.3607,-0.0860507) 3498015 -dictGet test_01037.dict_array (31.3831,3.84469) 101 -dictGet test_01037.dict_array (31.3886,-0.731137) 3501890 -dictGet test_01037.dict_array (31.4043,-0.348907) 5457271 -dictGet test_01037.dict_array (31.4081,1.47391) 3501926 -dictGet test_01037.dict_array (31.4176,-0.583645) 5457271 -dictGet test_01037.dict_array (31.4177,1.36972) 3501926 -dictGet test_01037.dict_array (31.4182,0.958303) 3498022 -dictGet test_01037.dict_array (31.4199,3.1738) 3501793 -dictGet test_01037.dict_array (31.4221,2.74876) 3501825 -dictGet test_01037.dict_array (31.4301,-0.122643) 3498015 -dictGet test_01037.dict_array (31.4344,1.00661) 3498022 -dictGet test_01037.dict_array (31.4375,4.20304) 101 -dictGet test_01037.dict_array (31.4377,0.289608) 3498094 -dictGet test_01037.dict_array (31.4379,0.54744) 3498014 -dictGet test_01037.dict_array (31.4459,3.94945) 101 -dictGet test_01037.dict_array (31.4559,-0.345063) 5457271 -dictGet test_01037.dict_array (31.464,0.726129) 3498014 -dictGet test_01037.dict_array (31.4662,-0.299019) 3498015 -dictGet test_01037.dict_array (31.4671,1.9605299999999999) 3501794 -dictGet test_01037.dict_array (31.4673,-0.403676) 5457271 -dictGet test_01037.dict_array (31.4712,-0.237941) 3498015 -dictGet test_01037.dict_array (31.4816,0.120264) 3498015 -dictGet test_01037.dict_array (31.4875,0.323483) 3498014 -dictGet test_01037.dict_array (31.490099999999998,-0.338163) 5457271 -dictGet test_01037.dict_array (31.4932,0.517674) 3498014 -dictGet test_01037.dict_array (31.5112,1.9689299999999998) 3501794 -dictGet test_01037.dict_array (31.5122,2.92785) 3501791 -dictGet test_01037.dict_array (31.5151,0.166429) 3498094 -dictGet test_01037.dict_array (31.5174,2.94802) 3501791 -dictGet test_01037.dict_array (31.5182,4.18776) 101 -dictGet test_01037.dict_array (31.5238,1.18793) 3498003 -dictGet test_01037.dict_array (31.5271,3.07446) 3501791 -dictGet test_01037.dict_array (31.5393,1.58061) 3501794 -dictGet test_01037.dict_array (31.5421,3.13711) 3501791 -dictGet test_01037.dict_array (31.5479,2.39897) 3497970 -dictGet test_01037.dict_array (31.5519,0.99285) 3498003 -dictGet test_01037.dict_array (31.5685,3.47987) 3501824 -dictGet test_01037.dict_array (31.5959,0.437382) 3498014 -dictGet test_01037.dict_array (31.6003,0.194376) 3498094 -dictGet test_01037.dict_array (31.6026,2.15457) 3501794 -dictGet test_01037.dict_array (31.606,2.45365) 3497970 -dictGet test_01037.dict_array (31.6062,-0.453441) 3501890 -dictGet test_01037.dict_array (31.6107,1.35247) 3497974 -dictGet test_01037.dict_array (31.6155,3.85588) 101 -dictGet test_01037.dict_array (31.6222,2.03326) 3501794 -dictGet test_01037.dict_array (31.6231,-0.123059) 3498083 -dictGet test_01037.dict_array (31.6244,1.6885599999999998) 3497974 -dictGet test_01037.dict_array (31.6459,0.669716) 3498014 -dictGet test_01037.dict_array (31.6563,-0.0644741) 3498083 -dictGet test_01037.dict_array (31.6618,-0.551121) 3501890 -dictGet test_01037.dict_array (31.6725,-0.38922) 3498085 -dictGet test_01037.dict_array (31.6727,4.10336) 101 -dictGet test_01037.dict_array (31.6739,4.1391) 101 -dictGet test_01037.dict_array (31.6897,2.8694699999999997) 3501792 -dictGet test_01037.dict_array (31.6902,3.98792) 101 -dictGet test_01037.dict_array (31.6945,2.46687) 3497970 -dictGet test_01037.dict_array (31.6987,-1.3796) 101 -dictGet test_01037.dict_array (31.7012,2.34845) 3497970 -dictGet test_01037.dict_array (31.7036,0.0228348) 3501888 -dictGet test_01037.dict_array (31.7046,3.68111) 3501824 -dictGet test_01037.dict_array (31.7055,2.92556) 3501792 -dictGet test_01037.dict_array (31.7102,1.04532) 3498003 -dictGet test_01037.dict_array (31.7149,-0.443302) 3498085 -dictGet test_01037.dict_array (31.7195,2.99311) 3501791 -dictGet test_01037.dict_array (31.7274,0.166719) 3498094 -dictGet test_01037.dict_array (31.7565,-0.565382) 3498085 -dictGet test_01037.dict_array (31.7615,0.771626) 3498014 -dictGet test_01037.dict_array (31.7739,1.8970099999999999) 3497974 -dictGet test_01037.dict_array (31.7848,1.2623199999999999) 3498003 -dictGet test_01037.dict_array (31.7912,-0.788599) 101 -dictGet test_01037.dict_array (31.8011,2.65853) 3497970 -dictGet test_01037.dict_array (31.8032,-0.0590108) 3501888 -dictGet test_01037.dict_array (31.8038,1.9618799999999998) 3497974 -dictGet test_01037.dict_array (31.8098,-1.46851) 101 -dictGet test_01037.dict_array (31.8131,3.41982) 3501791 -dictGet test_01037.dict_array (31.8169,3.31059) 3501791 -dictGet test_01037.dict_array (31.8202,-0.193692) 3501888 -dictGet test_01037.dict_array (31.8306,1.57586) 3497974 -dictGet test_01037.dict_array (31.8382,-0.787948) 101 -dictGet test_01037.dict_array (31.8433,2.49692) 3497970 -dictGet test_01037.dict_array (31.8436,2.41851) 3497970 -dictGet test_01037.dict_array (31.8563,-1.10787) 101 -dictGet test_01037.dict_array (31.8683,0.996504) 3498002 -dictGet test_01037.dict_array (31.8693,-0.828142) 101 -dictGet test_01037.dict_array (31.8723,1.08929) 3498003 -dictGet test_01037.dict_array (31.8737,0.881127) 3498002 -dictGet test_01037.dict_array (31.8881,-0.58441) 101 -dictGet test_01037.dict_array (31.9011,0.121349) 3498094 -dictGet test_01037.dict_array (31.9066,2.13045) 3497965 -dictGet test_01037.dict_array (31.9142,1.03368) 3498002 -dictGet test_01037.dict_array (31.9155,3.38363) 3501791 -dictGet test_01037.dict_array (31.9168,1.3166) 3498004 -dictGet test_01037.dict_array (31.9185,-1.11879) 101 -dictGet test_01037.dict_array (31.9186,-0.647948) 101 -dictGet test_01037.dict_array (31.9311,3.96928) 101 -dictGet test_01037.dict_array (31.9335,1.47048) 3497974 -dictGet test_01037.dict_array (31.9443,-1.36175) 101 -dictGet test_01037.dict_array (31.9481,2.34231) 3497970 -dictGet test_01037.dict_array (31.9526,1.36565) 3498004 -dictGet test_01037.dict_array (31.9629,2.5208399999999997) 3497970 -dictGet test_01037.dict_array (31.9765,0.975783) 3498002 -dictGet test_01037.dict_array (31.9923,3.31773) 3501791 -dictGet test_01037.dict_array (31.9994,0.972816) 3498002 -dictGet test_01037.dict_array (32.001,3.47425) 3501791 -dictGet test_01037.dict_array (32.0127,2.13874) 3497965 -dictGet test_01037.dict_array (32.0244,3.2092) 3501792 -dictGet test_01037.dict_array (32.029,1.18039) 3498004 -dictGet test_01037.dict_array (32.0315,0.566073) 3498095 -dictGet test_01037.dict_array (32.0354,1.0766499999999999) 3498004 -dictGet test_01037.dict_array (32.0399,-1.11576) 101 -dictGet test_01037.dict_array (32.053,2.16849) 3497965 -dictGet test_01037.dict_array (32.0542,0.042328) 3498096 -dictGet test_01037.dict_array (32.0576,2.47001) 3497970 -dictGet test_01037.dict_array (32.061,3.7498899999999997) 101 -dictGet test_01037.dict_array (32.0623,1.25134) 3498004 -dictGet test_01037.dict_array (32.0626,1.9611399999999999) 3497965 -dictGet test_01037.dict_array (32.0666,-0.0904247) 3498096 -dictGet test_01037.dict_array (32.0681,2.28442) 3497970 -dictGet test_01037.dict_array (32.0692,1.50869) 3497981 -dictGet test_01037.dict_array (32.0724,4.03314) 101 -dictGet test_01037.dict_array (32.0729,-0.064324) 101 -dictGet test_01037.dict_array (32.079,0.293758) 3498094 -dictGet test_01037.dict_array (32.0847,-1.19814) 101 -dictGet test_01037.dict_array (32.0974,-0.91927) 101 -dictGet test_01037.dict_array (32.0979,-0.736979) 101 -dictGet test_01037.dict_array (32.106,-1.33063) 101 -dictGet test_01037.dict_array (32.1189,0.246715) 3498094 -dictGet test_01037.dict_array (32.1207,4.00883) 101 -dictGet test_01037.dict_array (32.1396,1.12402) 3498004 -dictGet test_01037.dict_array (32.1413,1.5668) 3497981 -dictGet test_01037.dict_array (32.143,1.35559) 3498004 -dictGet test_01037.dict_array (32.1538,1.32881) 3498004 -dictGet test_01037.dict_array (32.1549,4.06552) 101 -dictGet test_01037.dict_array (32.1555,-0.79275) 101 -dictGet test_01037.dict_array (32.163,1.17733) 3498004 -dictGet test_01037.dict_array (32.1634,2.94273) 3501792 -dictGet test_01037.dict_array (32.1644,1.85666) 3497965 -dictGet test_01037.dict_array (32.1745,0.435458) 3498095 -dictGet test_01037.dict_array (32.1765,1.65149) 3497981 -dictGet test_01037.dict_array (32.1893,2.08924) 3497965 -dictGet test_01037.dict_array (32.2024,0.222191) 3498093 -dictGet test_01037.dict_array (32.2107,1.34379) 3497981 -dictGet test_01037.dict_array (32.2109,3.9018699999999997) 101 -dictGet test_01037.dict_array (32.2123,1.85233) 3497965 -dictGet test_01037.dict_array (32.2144,3.72534) 101 -dictGet test_01037.dict_array (32.2218,2.5386699999999998) 3497970 -dictGet test_01037.dict_array (32.2279,2.84267) 3497245 -dictGet test_01037.dict_array (32.2345,3.33295) 3501792 -dictGet test_01037.dict_array (32.2435,3.85283) 101 -dictGet test_01037.dict_array (32.2527,-0.480608) 101 -dictGet test_01037.dict_array (32.2566,-0.837882) 101 -dictGet test_01037.dict_array (32.2627,2.57708) 3497970 -dictGet test_01037.dict_array (32.2733,0.244931) 3498096 -dictGet test_01037.dict_array (32.2761,4.05808) 101 -dictGet test_01037.dict_array (32.2764,3.78472) 101 -dictGet test_01037.dict_array (32.2814,-1.26011) 101 -dictGet test_01037.dict_array (32.2861,3.02427) 3497245 -dictGet test_01037.dict_array (32.2924,0.928609) 3498004 -dictGet test_01037.dict_array (32.2963,-0.78543) 101 -dictGet test_01037.dict_array (32.3039,3.21175) 3501792 -dictGet test_01037.dict_array (32.3107,0.698287) 3498004 -dictGet test_01037.dict_array (32.3138,0.0595677) 3498106 -dictGet test_01037.dict_array (32.3339,0.707056) 3498004 -dictGet test_01037.dict_array (32.3351,0.415474) 3498106 -dictGet test_01037.dict_array (32.342,-0.681023) 101 -dictGet test_01037.dict_array (32.3463,1.83196) 3497126 -dictGet test_01037.dict_array (32.3494,2.43799) 3497114 -dictGet test_01037.dict_array (32.3524,3.47049) 3501822 -dictGet test_01037.dict_array (32.3531,2.33115) 3497114 -dictGet test_01037.dict_array (32.3602,0.116106) 3498106 -dictGet test_01037.dict_array (32.3612,1.1598) 3498004 -dictGet test_01037.dict_array (32.3689,3.34847) 3501822 -dictGet test_01037.dict_array (32.3695,0.734055) 3498004 -dictGet test_01037.dict_array (32.3825,3.85017) 101 -dictGet test_01037.dict_array (32.3835,-1.25491) 101 -dictGet test_01037.dict_array (32.4018,-0.728568) 101 -dictGet test_01037.dict_array (32.4044,2.96727) 3497245 -dictGet test_01037.dict_array (32.4101,2.9988) 3497245 -dictGet test_01037.dict_array (32.417,-1.12908) 101 -dictGet test_01037.dict_array (32.4172,4.1952) 101 -dictGet test_01037.dict_array (32.4239,2.49512) 3497245 -dictGet test_01037.dict_array (32.4258,4.05137) 101 -dictGet test_01037.dict_array (32.4264,-0.427357) 101 -dictGet test_01037.dict_array (32.4274,3.59377) 3501822 -dictGet test_01037.dict_array (32.4286,-1.24757) 101 -dictGet test_01037.dict_array (32.4294,3.0665) 3497245 -dictGet test_01037.dict_array (32.4333,-0.353347) 101 -dictGet test_01037.dict_array (32.4391,3.64421) 3501822 -dictGet test_01037.dict_array (32.4401,3.70635) 3501822 -dictGet test_01037.dict_array (32.45,1.68918) 3497126 -dictGet test_01037.dict_array (32.4507,-0.133471) 101 -dictGet test_01037.dict_array (32.4592,0.976458) 3498105 -dictGet test_01037.dict_array (32.4595,1.89135) 3497126 -dictGet test_01037.dict_array (32.4604,0.280248) 3498106 -dictGet test_01037.dict_array (32.4835,0.472731) 3498106 -dictGet test_01037.dict_array (32.4855,2.01938) 3497126 -dictGet test_01037.dict_array (32.4872,2.01697) 3497126 -dictGet test_01037.dict_array (32.4911,0.613106) 3498105 -dictGet test_01037.dict_array (32.4918,2.17834) 3497114 -dictGet test_01037.dict_array (32.4947,2.34595) 3497114 -dictGet test_01037.dict_array (32.5035,2.92234) 3497245 -dictGet test_01037.dict_array (32.5132,-0.331206) 101 -dictGet test_01037.dict_array (32.5156,-0.412604) 3501887 -dictGet test_01037.dict_array (32.5158,2.9067499999999997) 3497245 -dictGet test_01037.dict_array (32.5249,2.44519) 3497114 -dictGet test_01037.dict_array (32.5293,-0.790952) 101 -dictGet test_01037.dict_array (32.5319,3.96854) 101 -dictGet test_01037.dict_array (32.5518,3.6093) 3501822 -dictGet test_01037.dict_array (32.5541,3.5225400000000002) 3501822 -dictGet test_01037.dict_array (32.5569,0.816123) 3498105 -dictGet test_01037.dict_array (32.5646,1.9775) 3497126 -dictGet test_01037.dict_array (32.5733,3.81271) 101 -dictGet test_01037.dict_array (32.5767,0.948327) 3498105 -dictGet test_01037.dict_array (32.5971,1.76179) 3497126 -dictGet test_01037.dict_array (32.6035,-0.716157) 101 -dictGet test_01037.dict_array (32.6087,4.21614) 101 -dictGet test_01037.dict_array (32.6171,0.024481) 101 -dictGet test_01037.dict_array (32.6189,-0.775391) 101 -dictGet test_01037.dict_array (32.6198,2.92081) 3497167 -dictGet test_01037.dict_array (32.621,-0.970784) 101 -dictGet test_01037.dict_array (32.6266,0.650009) 3498105 -dictGet test_01037.dict_array (32.6315,2.15144) 3497126 -dictGet test_01037.dict_array (32.6385,-0.436803) 101 -dictGet test_01037.dict_array (32.6449,-0.191292) 101 -dictGet test_01037.dict_array (32.6535,2.10385) 3497126 -dictGet test_01037.dict_array (32.6592,3.49973) 3501822 -dictGet test_01037.dict_array (32.6598,2.5980600000000003) 3497114 -dictGet test_01037.dict_array (32.6612,2.95681) 3497167 -dictGet test_01037.dict_array (32.6636,-0.57235) 101 -dictGet test_01037.dict_array (32.669,-0.382702) 101 -dictGet test_01037.dict_array (32.6752,1.30748) 3497981 -dictGet test_01037.dict_array (32.6811,2.9559800000000003) 3497167 -dictGet test_01037.dict_array (32.6821,0.57336) 3498105 -dictGet test_01037.dict_array (32.6828,3.91304) 101 -dictGet test_01037.dict_array (32.6979,3.96868) 101 -dictGet test_01037.dict_array (32.6983,3.15784) 3497167 -dictGet test_01037.dict_array (32.7122,0.794293) 3498105 -dictGet test_01037.dict_array (32.7131,-0.847256) 101 -dictGet test_01037.dict_array (32.7219,0.883461) 3498105 -dictGet test_01037.dict_array (32.7228,1.78808) 3497126 -dictGet test_01037.dict_array (32.7273,-0.206908) 101 -dictGet test_01037.dict_array (32.7292,0.259331) 3501889 -dictGet test_01037.dict_array (32.7304,-1.38317) 101 -dictGet test_01037.dict_array (32.7353,1.01601) 3498105 -dictGet test_01037.dict_array (32.7354,4.17574) 101 -dictGet test_01037.dict_array (32.7357,-0.190194) 101 -dictGet test_01037.dict_array (32.7465,-1.37598) 101 -dictGet test_01037.dict_array (32.7494,-0.275675) 101 -dictGet test_01037.dict_array (32.7514,0.128951) 3501889 -dictGet test_01037.dict_array (32.753,3.44207) 3501822 -dictGet test_01037.dict_array (32.7686,2.11713) 3497126 -dictGet test_01037.dict_array (32.7694,1.47159) 3497388 -dictGet test_01037.dict_array (32.7768,0.0401042) 101 -dictGet test_01037.dict_array (32.781,-1.34283) 101 -dictGet test_01037.dict_array (32.7814,1.73876) 3497388 -dictGet test_01037.dict_array (32.7856,-1.06363) 101 -dictGet test_01037.dict_array (32.792699999999996,-1.1255600000000001) 101 -dictGet test_01037.dict_array (32.7941,-0.645447) 101 -dictGet test_01037.dict_array (32.7946,1.48889) 3497388 -dictGet test_01037.dict_array (32.797,0.791753) 3501889 -dictGet test_01037.dict_array (32.7982,-0.537798) 101 -dictGet test_01037.dict_array (32.8091,2.3611) 3490438 -dictGet test_01037.dict_array (32.81,1.7130800000000002) 3497388 -dictGet test_01037.dict_array (32.8174,-0.288322) 101 -dictGet test_01037.dict_array (32.823,1.6546699999999999) 3497388 -dictGet test_01037.dict_array (32.8233,1.62108) 3497388 -dictGet test_01037.dict_array (32.8428,-0.400045) 101 -dictGet test_01037.dict_array (32.8479,2.13598) 3490438 -dictGet test_01037.dict_array (32.8524,0.199902) 3501889 -dictGet test_01037.dict_array (32.8543,3.23553) 3501820 -dictGet test_01037.dict_array (32.8562,1.31371) 3498117 -dictGet test_01037.dict_array (32.87,1.44256) 3498117 -dictGet test_01037.dict_array (32.8789,2.38192) 3490438 -dictGet test_01037.dict_array (32.8812,2.20734) 3497128 -dictGet test_01037.dict_array (32.8815,-0.54427) 101 -dictGet test_01037.dict_array (32.8853,2.4859) 3497128 -dictGet test_01037.dict_array (32.8909,0.513964) 3501889 -dictGet test_01037.dict_array (32.9035,2.38999) 3490438 -dictGet test_01037.dict_array (32.9097,2.48131) 3497128 -dictGet test_01037.dict_array (32.928,-0.943269) 101 -dictGet test_01037.dict_array (32.9322,1.13165) 3498104 -dictGet test_01037.dict_array (32.9348,1.22606) 3498117 -dictGet test_01037.dict_array (32.9417,3.77998) 3501822 -dictGet test_01037.dict_array (32.9428,3.11936) 3497167 -dictGet test_01037.dict_array (32.9482,1.18092) 3498118 -dictGet test_01037.dict_array (32.9506,0.0609364) 101 -dictGet test_01037.dict_array (32.953,-0.828308) 101 -dictGet test_01037.dict_array (32.9593,3.5209099999999998) 3501822 -dictGet test_01037.dict_array (32.9617,2.07711) 3497128 -dictGet test_01037.dict_array (32.966,0.693749) 3498104 -dictGet test_01037.dict_array (32.9668,-0.716432) 101 -dictGet test_01037.dict_array (32.9702,1.98555) 3497127 -dictGet test_01037.dict_array (32.9782,1.73819) 3497388 -dictGet test_01037.dict_array (32.9805,3.71151) 3501822 -dictGet test_01037.dict_array (32.9821,2.97225) 3497167 -dictGet test_01037.dict_array (32.995,-0.830301) 101 -dictGet test_01037.dict_array (33.0234,0.770848) 3498104 -dictGet test_01037.dict_array (33.0312,-0.340964) 101 -dictGet test_01037.dict_array (33.0366,-0.756795) 101 -dictGet test_01037.dict_array (33.0438,0.812871) 3498118 -dictGet test_01037.dict_array (33.0455,1.84843) 3497127 -dictGet test_01037.dict_array (33.0498,0.0913292) 101 -dictGet test_01037.dict_array (33.0506,1.53739) 3497364 -dictGet test_01037.dict_array (33.0554,2.4265) 3497363 -dictGet test_01037.dict_array (33.0741,3.61332) 3501822 -dictGet test_01037.dict_array (33.0765,-0.179985) 101 -dictGet test_01037.dict_array (33.087,1.46465) 3497399 -dictGet test_01037.dict_array (33.0906,-0.620383) 101 -dictGet test_01037.dict_array (33.1047,-1.28027) 101 -dictGet test_01037.dict_array (33.1072,1.96303) 3497127 -dictGet test_01037.dict_array (33.1081,-0.897874) 101 -dictGet test_01037.dict_array (33.1122,1.8950200000000001) 3497127 -dictGet test_01037.dict_array (33.1237,2.63993) 3497165 -dictGet test_01037.dict_array (33.1238,0.753963) 3498118 -dictGet test_01037.dict_array (33.1257,0.495668) 3498102 -dictGet test_01037.dict_array (33.1258,1.78341) 3497364 -dictGet test_01037.dict_array (33.127,2.59646) 3497166 -dictGet test_01037.dict_array (33.1324,-1.23742) 101 -dictGet test_01037.dict_array (33.1359,3.83491) 101 -dictGet test_01037.dict_array (33.1628,-0.379588) 101 -dictGet test_01037.dict_array (33.1679,1.25601) 3498117 -dictGet test_01037.dict_array (33.1688,-1.35553) 101 -dictGet test_01037.dict_array (33.181,2.10943) 3497363 -dictGet test_01037.dict_array (33.1871,2.81171) 3497165 -dictGet test_01037.dict_array (33.1877,0.771297) 3498118 -dictGet test_01037.dict_array (33.1883,-0.204797) 101 -dictGet test_01037.dict_array (33.1886,3.27998) 3501820 -dictGet test_01037.dict_array (33.1955,0.708907) 3498118 -dictGet test_01037.dict_array (33.2044,-0.769275) 101 -dictGet test_01037.dict_array (33.2182,3.36103) 3501820 -dictGet test_01037.dict_array (33.2192,3.43586) 3501822 -dictGet test_01037.dict_array (33.2322,-0.916753) 101 -dictGet test_01037.dict_array (33.2359,-0.81321) 101 -dictGet test_01037.dict_array (33.238,0.635072) 3498111 -dictGet test_01037.dict_array (33.2398,3.02588) 3497165 -dictGet test_01037.dict_array (33.2469,2.35698) 3497363 -dictGet test_01037.dict_array (33.247,2.3327) 3497363 -dictGet test_01037.dict_array (33.2579,2.8027100000000003) 3497165 -dictGet test_01037.dict_array (33.2607,0.321082) 101 -dictGet test_01037.dict_array (33.2653,0.243336) 101 -dictGet test_01037.dict_array (33.2758,0.831836) 3498118 -dictGet test_01037.dict_array (33.2771,0.886536) 3498118 -dictGet test_01037.dict_array (33.2914,1.16026) 3498117 -dictGet test_01037.dict_array (33.2914,1.38882) 3497399 -dictGet test_01037.dict_array (33.2982,-1.16604) 101 -dictGet test_01037.dict_array (33.2985,0.842556) 3498112 -dictGet test_01037.dict_array (33.3005,2.8338900000000002) 3497165 -dictGet test_01037.dict_array (33.305,0.0969475) 101 -dictGet test_01037.dict_array (33.3072,3.82163) 101 -dictGet test_01037.dict_array (33.312,3.41475) 3501820 -dictGet test_01037.dict_array (33.3129,2.46048) 3497166 -dictGet test_01037.dict_array (33.3134,3.46863) 3501820 -dictGet test_01037.dict_array (33.3203,2.33139) 3497166 -dictGet test_01037.dict_array (33.324,0.433701) 101 -dictGet test_01037.dict_array (33.3338,2.44705) 3497166 -dictGet test_01037.dict_array (33.337,4.06475) 101 -dictGet test_01037.dict_array (33.3469,1.08172) 3498126 -dictGet test_01037.dict_array (33.3538,0.717896) 3498112 -dictGet test_01037.dict_array (33.3618,1.37899) 3497399 -dictGet test_01037.dict_array (33.3698,0.547744) 3501862 -dictGet test_01037.dict_array (33.3705,0.957619) 3498112 -dictGet test_01037.dict_array (33.3821,3.07258) 3497165 -dictGet test_01037.dict_array (33.3881,3.0626) 3497165 -dictGet test_01037.dict_array (33.393,-0.816186) 101 -dictGet test_01037.dict_array (33.3945,0.869508) 3498110 -dictGet test_01037.dict_array (33.4001,1.24186) 3498117 -dictGet test_01037.dict_array (33.4008,2.34911) 3497166 -dictGet test_01037.dict_array (33.4166,-1.2808899999999999) 101 -dictGet test_01037.dict_array (33.4167,3.0655) 3497165 -dictGet test_01037.dict_array (33.4204,2.81887) 3497165 -dictGet test_01037.dict_array (33.4211,1.71128) 3497400 -dictGet test_01037.dict_array (33.4237,2.91761) 3497165 -dictGet test_01037.dict_array (33.4266,1.5955599999999999) 3497399 -dictGet test_01037.dict_array (33.4353,-0.391392) 101 -dictGet test_01037.dict_array (33.4362,-0.134658) 101 -dictGet test_01037.dict_array (33.4386,0.15396) 101 -dictGet test_01037.dict_array (33.4421,-0.50712) 101 -dictGet test_01037.dict_array (33.452,0.915829) 3498126 -dictGet test_01037.dict_array (33.463,-0.0882717) 101 -dictGet test_01037.dict_array (33.464,-1.00949) 101 -dictGet test_01037.dict_array (33.4692,0.954092) 3498126 -dictGet test_01037.dict_array (33.4716,1.9538799999999998) 3497400 -dictGet test_01037.dict_array (33.4756,1.85836) 3497400 -dictGet test_01037.dict_array (33.4859,4.0751) 101 -dictGet test_01037.dict_array (33.4899,3.54193) 3501820 -dictGet test_01037.dict_array (33.4935,3.49794) 3501820 -dictGet test_01037.dict_array (33.494,-0.983356) 101 -dictGet test_01037.dict_array (33.4955,-1.28128) 101 -dictGet test_01037.dict_array (33.4965,-0.278687) 101 -dictGet test_01037.dict_array (33.4991,0.647491) 3498110 -dictGet test_01037.dict_array (33.5076,2.2272) 3497424 -dictGet test_01037.dict_array (33.5079,-0.498199) 101 -dictGet test_01037.dict_array (33.5157,0.535034) 3501862 -dictGet test_01037.dict_array (33.5171,2.49677) 3497166 -dictGet test_01037.dict_array (33.5255,2.4447200000000002) 3497166 -dictGet test_01037.dict_array (33.526,4.01194) 101 -dictGet test_01037.dict_array (33.5288,0.789434) 3498110 -dictGet test_01037.dict_array (33.5356,-1.17671) 101 -dictGet test_01037.dict_array (33.5402,1.49152) 3497399 -dictGet test_01037.dict_array (33.5418,3.45757) 3501820 -dictGet test_01037.dict_array (33.5428,1.90712) 3497400 -dictGet test_01037.dict_array (33.5556,-0.55741) 101 -dictGet test_01037.dict_array (33.5564,0.876858) 3498128 -dictGet test_01037.dict_array (33.5567,-0.10208) 101 -dictGet test_01037.dict_array (33.5645,-0.124824) 101 -dictGet test_01037.dict_array (33.5663,3.4872) 3501820 -dictGet test_01037.dict_array (33.5716,-0.0107611) 101 -dictGet test_01037.dict_array (33.578,3.55714) 3501820 -dictGet test_01037.dict_array (33.5826,-0.49076) 101 -dictGet test_01037.dict_array (33.5909,0.773737) 3498110 -dictGet test_01037.dict_array (33.5958,2.9619999999999997) 3497425 -dictGet test_01037.dict_array (33.6193,-0.919755) 101 -dictGet test_01037.dict_array (33.6313,0.652132) 3498110 -dictGet test_01037.dict_array (33.632,0.823351) 3498128 -dictGet test_01037.dict_array (33.66,2.18998) 3497424 -dictGet test_01037.dict_array (33.6621,0.535395) 3498135 -dictGet test_01037.dict_array (33.6726,3.19367) 3497438 -dictGet test_01037.dict_array (33.6912,1.74522) 3497400 -dictGet test_01037.dict_array (33.705,0.706397) 3498135 -dictGet test_01037.dict_array (33.7076,0.7622) 3498128 -dictGet test_01037.dict_array (33.7112,1.70187) 3497400 -dictGet test_01037.dict_array (33.7246,-1.14837) 101 -dictGet test_01037.dict_array (33.7326,2.62413) 3497425 -dictGet test_01037.dict_array (33.7332,2.82137) 3497425 -dictGet test_01037.dict_array (33.7434,0.394672) 3498135 -dictGet test_01037.dict_array (33.7443,1.54557) 3497398 -dictGet test_01037.dict_array (33.7506,1.57317) 3497398 -dictGet test_01037.dict_array (33.7526,1.8578999999999999) 3497424 -dictGet test_01037.dict_array (33.766,4.15013) 101 -dictGet test_01037.dict_array (33.7834,2.41789) 3497439 -dictGet test_01037.dict_array (33.7864,0.230935) 101 -dictGet test_01037.dict_array (33.7965,3.05709) 3497438 -dictGet test_01037.dict_array (33.7998,3.32881) 3497438 -dictGet test_01037.dict_array (33.8003,2.97338) 3497425 -dictGet test_01037.dict_array (33.8007,-1.08962) 101 -dictGet test_01037.dict_array (33.8022,-0.139488) 101 -dictGet test_01037.dict_array (33.8065,2.70857) 3497425 -dictGet test_01037.dict_array (33.8169,-0.607788) 101 -dictGet test_01037.dict_array (33.8203,0.108512) 3501863 -dictGet test_01037.dict_array (33.8231,-1.03449) 101 -dictGet test_01037.dict_array (33.8312,3.49458) 3501829 -dictGet test_01037.dict_array (33.8342,0.297518) 3501863 -dictGet test_01037.dict_array (33.8352,0.165872) 101 -dictGet test_01037.dict_array (33.8354,1.87277) 3497424 -dictGet test_01037.dict_array (33.8371,1.60103) 3497398 -dictGet test_01037.dict_array (33.8387,1.9968) 3497424 -dictGet test_01037.dict_array (33.8403,3.5805) 3501829 -dictGet test_01037.dict_array (33.8414,-0.703067) 101 -dictGet test_01037.dict_array (33.844,-0.179472) 101 -dictGet test_01037.dict_array (33.8468,3.40137) 3501829 -dictGet test_01037.dict_array (33.8509,4.15334) 101 -dictGet test_01037.dict_array (33.8539,2.38339) 3497439 -dictGet test_01037.dict_array (33.858,-1.3122500000000001) 101 -dictGet test_01037.dict_array (33.859,3.72626) 3501829 -dictGet test_01037.dict_array (33.8616,2.24433) 3497424 -dictGet test_01037.dict_array (33.8621,3.01035) 3497438 -dictGet test_01037.dict_array (33.8623,1.17559) 3498129 -dictGet test_01037.dict_array (33.8682,2.706) 3497425 -dictGet test_01037.dict_array (33.8684,0.189231) 3501863 -dictGet test_01037.dict_array (33.872,1.93574) 3497424 -dictGet test_01037.dict_array (33.8844,3.80404) 3501829 -dictGet test_01037.dict_array (33.8888,0.594884) 3498135 -dictGet test_01037.dict_array (33.8946,2.74161) 3497438 -dictGet test_01037.dict_array (33.9023,0.6239) 3498135 -dictGet test_01037.dict_array (33.9057,0.873222) 3498136 -dictGet test_01037.dict_array (33.9157,-1.26607) 101 -dictGet test_01037.dict_array (33.92,2.06848) 3497397 -dictGet test_01037.dict_array (33.9298,-0.00526229) 101 -dictGet test_01037.dict_array (33.932,3.07063) 3497438 -dictGet test_01037.dict_array (33.9322,0.629385) 3501864 -dictGet test_01037.dict_array (33.9367,-1.41955) 101 -dictGet test_01037.dict_array (33.937,1.42532) 3498173 -dictGet test_01037.dict_array (33.9375,1.1467100000000001) 3498159 -dictGet test_01037.dict_array (33.9434,-1.05739) 101 -dictGet test_01037.dict_array (33.9477,3.34809) 3501829 -dictGet test_01037.dict_array (33.95,2.21715) 3497397 -dictGet test_01037.dict_array (33.955799999999996,0.305176) 3501859 -dictGet test_01037.dict_array (33.9686,-0.28273) 101 -dictGet test_01037.dict_array (33.9703,4.1255) 3501829 -dictGet test_01037.dict_array (33.9707,3.08199) 3497438 -dictGet test_01037.dict_array (33.9754,1.06203) 3498159 -dictGet test_01037.dict_array (33.9757,3.72468) 3501829 -dictGet test_01037.dict_array (33.9775,-0.0440599) 101 -dictGet test_01037.dict_array (33.9777,-0.251484) 101 -dictGet test_01037.dict_array (33.9789,-0.339374) 101 -dictGet test_01037.dict_array (33.9849,2.54515) 3497425 -dictGet test_01037.dict_array (33.9885,-0.318557) 101 -dictGet test_01037.dict_array (33.9977,1.07175) 3498159 -dictGet test_01037.dict_array (33.9984,-0.700517) 101 -dictGet test_01037.dict_array (34.0149,3.53338) 3501829 -dictGet test_01037.dict_array (34.0173,3.39155) 3501829 -dictGet test_01037.dict_array (34.0317,3.9579) 3501829 -dictGet test_01037.dict_array (34.0369,3.83612) 3501829 -dictGet test_01037.dict_array (34.043,-0.0887221) 101 -dictGet test_01037.dict_array (34.0487,1.14252) 3498159 -dictGet test_01037.dict_array (34.052,1.74832) 3497397 -dictGet test_01037.dict_array (34.0711,-0.898071) 101 -dictGet test_01037.dict_array (34.0747,1.55057) 3498173 -dictGet test_01037.dict_array (34.0803,3.16763) 3497438 -dictGet test_01037.dict_array (34.0872,3.75555) 3501829 -dictGet test_01037.dict_array (34.0965,1.62038) 3498173 -dictGet test_01037.dict_array (34.0977,-0.412691) 101 -dictGet test_01037.dict_array (34.0986,0.0294206) 101 -dictGet test_01037.dict_array (34.1072,3.15823) 3497438 -dictGet test_01037.dict_array (34.1092,3.09599) 3497438 -dictGet test_01037.dict_array (34.1206,1.04637) 3498160 -dictGet test_01037.dict_array (34.1209,3.13826) 3497438 -dictGet test_01037.dict_array (34.1265,3.95881) 3501829 -dictGet test_01037.dict_array (34.1286,-0.539319) 101 -dictGet test_01037.dict_array (34.1358,3.67451) 3501829 -dictGet test_01037.dict_array (34.1428,0.136115) 101 -dictGet test_01037.dict_array (34.157,1.73522) 3497397 -dictGet test_01037.dict_array (34.1581,1.48001) 3498172 -dictGet test_01037.dict_array (34.1682,3.42373) 3501829 -dictGet test_01037.dict_array (34.1683,-1.26511) 101 -dictGet test_01037.dict_array (34.1684,4.20007) 101 -dictGet test_01037.dict_array (34.1854,3.32089) 3501829 -dictGet test_01037.dict_array (34.2022,0.749536) 3501864 -dictGet test_01037.dict_array (34.2044,3.04865) 3497438 -dictGet test_01037.dict_array (34.22,-0.500055) 101 -dictGet test_01037.dict_array (34.2249,0.743775) 3501864 -dictGet test_01037.dict_array (34.2254,1.34702) 3498172 -dictGet test_01037.dict_array (34.2355,-0.898843) 101 -dictGet test_01037.dict_array (34.2394,2.0203699999999998) 3497439 -dictGet test_01037.dict_array (34.2466,1.83785) 3498251 -dictGet test_01037.dict_array (34.247,4.09563) 101 -dictGet test_01037.dict_array (34.2508,2.61312) 3497439 -dictGet test_01037.dict_array (34.2517,1.69642) 3498251 -dictGet test_01037.dict_array (34.2564,4.13033) 101 -dictGet test_01037.dict_array (34.2574,4.18928) 101 -dictGet test_01037.dict_array (34.2614,-0.478719) 101 -dictGet test_01037.dict_array (34.2625,2.38088) 3497439 -dictGet test_01037.dict_array (34.2666,3.1503) 3501829 -dictGet test_01037.dict_array (34.271,4.02223) 101 -dictGet test_01037.dict_array (34.2727,0.514755) 101 -dictGet test_01037.dict_array (34.278,1.98929) 3497439 -dictGet test_01037.dict_array (34.2798,-0.199208) 101 -dictGet test_01037.dict_array (34.2804,2.05184) 3497439 -dictGet test_01037.dict_array (34.2945,-1.11051) 101 -dictGet test_01037.dict_array (34.3168,-0.0829721) 101 -dictGet test_01037.dict_array (34.3345,3.4358) 3501829 -dictGet test_01037.dict_array (34.3377,1.13527) 3498162 -dictGet test_01037.dict_array (34.3383,1.27891) 3498161 -dictGet test_01037.dict_array (34.3391,1.47945) 3498161 -dictGet test_01037.dict_array (34.3441,0.627014) 101 -dictGet test_01037.dict_array (34.347,2.4853) 3497439 -dictGet test_01037.dict_array (34.3514,2.16247) 3497439 -dictGet test_01037.dict_array (34.3627,2.64533) 3497439 -dictGet test_01037.dict_array (34.3682,-0.227501) 101 -dictGet test_01037.dict_array (34.3756,4.21248) 101 -dictGet test_01037.dict_array (34.379,3.96604) 101 -dictGet test_01037.dict_array (34.3827,1.7518) 3498251 -dictGet test_01037.dict_array (34.3912,2.8834) 3501830 -dictGet test_01037.dict_array (34.3919,0.668829) 101 -dictGet test_01037.dict_array (34.3949,2.00338) 3497439 -dictGet test_01037.dict_array (34.3987,0.557268) 101 -dictGet test_01037.dict_array (34.4111,0.768558) 101 -dictGet test_01037.dict_array (34.4119,2.8742) 3501830 -dictGet test_01037.dict_array (34.416,3.50841) 3501829 -dictGet test_01037.dict_array (34.4212,1.24916) 3498161 -dictGet test_01037.dict_array (34.4251,0.457029) 101 -dictGet test_01037.dict_array (34.4274,-0.902559) 101 -dictGet test_01037.dict_array (34.4325,4.03159) 101 -dictGet test_01037.dict_array (34.438,1.63994) 3498251 -dictGet test_01037.dict_array (34.4403,-0.177594) 101 -dictGet test_01037.dict_array (34.4421,0.726712) 101 -dictGet test_01037.dict_array (34.4517,2.98611) 3501830 -dictGet test_01037.dict_array (34.4658,-1.312) 101 -dictGet test_01037.dict_array (34.4732,-0.0681338) 101 -dictGet test_01037.dict_array (34.4752,2.81646) 3501830 -dictGet test_01037.dict_array (34.4914,2.3858) 3497439 -dictGet test_01037.dict_array (34.4923,0.855231) 101 -dictGet test_01037.dict_array (34.5235,1.78468) 3498251 -dictGet test_01037.dict_array (34.5305,4.10608) 101 -dictGet test_01037.dict_array (34.5389,0.621937) 101 -dictGet test_01037.dict_array (34.5406,3.17145) 101 -dictGet test_01037.dict_array (34.5434,-0.56306) 101 -dictGet test_01037.dict_array (34.5449,3.13311) 3501829 -dictGet test_01037.dict_array (34.5491,2.31572) 3497439 -dictGet test_01037.dict_array (34.5539,2.94028) 3501830 -dictGet test_01037.dict_array (34.5546,-0.208825) 101 -dictGet test_01037.dict_array (34.5549,3.78486) 101 -dictGet test_01037.dict_array (34.5676,0.307148) 101 -dictGet test_01037.dict_array (34.5743,1.5217399999999999) 3501838 -dictGet test_01037.dict_array (34.5775,3.48046) 101 -dictGet test_01037.dict_array (34.5815,2.5243700000000002) 3501830 -dictGet test_01037.dict_array (34.5841,4.21191) 101 -dictGet test_01037.dict_array (34.5887,2.65083) 3501830 -dictGet test_01037.dict_array (34.5937,3.2143) 101 -dictGet test_01037.dict_array (34.6013,-1.0612) 101 -dictGet test_01037.dict_array (34.6089,1.36066) 3501838 -dictGet test_01037.dict_array (34.6103,3.40227) 101 -dictGet test_01037.dict_array (34.6128,1.92276) 3498251 -dictGet test_01037.dict_array (34.6175,2.43627) 3498251 -dictGet test_01037.dict_array (34.6209,3.43776) 101 -dictGet test_01037.dict_array (34.6234,2.60237) 3501830 -dictGet test_01037.dict_array (34.6275,3.52479) 101 -dictGet test_01037.dict_array (34.635,0.568558) 101 -dictGet test_01037.dict_array (34.6373,2.37692) 3498251 -dictGet test_01037.dict_array (34.6375,3.52234) 101 -dictGet test_01037.dict_array (34.6426,2.12397) 3498251 -dictGet test_01037.dict_array (34.6513,2.80915) 3501830 -dictGet test_01037.dict_array (34.6632,2.30039) 3498251 -dictGet test_01037.dict_array (34.6691,1.86582) 3498251 -dictGet test_01037.dict_array (34.6739,0.15342) 101 -dictGet test_01037.dict_array (34.6825,0.0499679) 101 -dictGet test_01037.dict_array (34.6893,0.454326) 101 -dictGet test_01037.dict_array (34.6957,-0.358598) 101 -dictGet test_01037.dict_array (34.6986,0.562679) 101 -dictGet test_01037.dict_array (34.712,1.12114) 101 -dictGet test_01037.dict_array (34.7126,-0.0057301) 101 -dictGet test_01037.dict_array (34.7137,0.0248501) 101 -dictGet test_01037.dict_array (34.7162,1.15623) 101 -dictGet test_01037.dict_array (34.7258,3.95142) 101 -dictGet test_01037.dict_array (34.7347,3.5232099999999997) 101 -dictGet test_01037.dict_array (34.7363,2.23374) 3501830 -dictGet test_01037.dict_array (34.7375,0.397841) 101 -dictGet test_01037.dict_array (34.7423,3.09198) 101 -dictGet test_01037.dict_array (34.7452,3.09029) 101 -dictGet test_01037.dict_array (34.7539,-1.06943) 101 -dictGet test_01037.dict_array (34.7733,-0.00912717) 101 -dictGet test_01037.dict_array (34.774,2.71088) 3501830 -dictGet test_01037.dict_array (34.7771,1.46009) 3501835 -dictGet test_01037.dict_array (34.7782,-1.28308) 101 -dictGet test_01037.dict_array (34.7924,3.63564) 101 -dictGet test_01037.dict_array (34.7939,-0.416676) 101 -dictGet test_01037.dict_array (34.7964,-0.401773) 101 -dictGet test_01037.dict_array (34.7974,0.0286873) 101 -dictGet test_01037.dict_array (34.7975,3.05965) 101 -dictGet test_01037.dict_array (34.8037,3.07263) 101 -dictGet test_01037.dict_array (34.8254,-0.390284) 101 -dictGet test_01037.dict_array (34.828,1.91869) 3498251 -dictGet test_01037.dict_array (34.8289,3.71058) 101 -dictGet test_01037.dict_array (34.8403,2.14606) 3501835 -dictGet test_01037.dict_array (34.8437,2.20617) 3501830 -dictGet test_01037.dict_array (34.8469,2.38435) 3501830 -dictGet test_01037.dict_array (34.86,1.45705) 101 -dictGet test_01037.dict_array (34.8612,0.914248) 101 -dictGet test_01037.dict_array (34.8663,3.4215400000000002) 101 -dictGet test_01037.dict_array (34.8724,-0.375144) 101 -dictGet test_01037.dict_array (34.8795,3.29317) 101 -dictGet test_01037.dict_array (34.8823,1.21988) 101 -dictGet test_01037.dict_array (34.8834,1.07657) 101 -dictGet test_01037.dict_array (34.8837,0.157648) 101 -dictGet test_01037.dict_array (34.8871,-0.9755) 101 -dictGet test_01037.dict_array (34.8871,1.8943699999999999) 3501835 -dictGet test_01037.dict_array (34.889,3.36756) 101 -dictGet test_01037.dict_array (34.8907,1.24874) 101 -dictGet test_01037.dict_array (34.8965,3.13508) 101 -dictGet test_01037.dict_array (34.9042,2.62092) 101 -dictGet test_01037.dict_array (34.9055,-0.0448967) 101 -dictGet test_01037.dict_array (34.9122,0.110576) 101 -dictGet test_01037.dict_array (34.9228,3.60183) 101 -dictGet test_01037.dict_array (34.9237,1.21715) 101 -dictGet test_01037.dict_array (34.9296,1.70459) 3501835 -dictGet test_01037.dict_array (34.941,-1.14663) 101 -dictGet test_01037.dict_array (34.9448,1.18923) 101 -dictGet test_01037.dict_array (34.9462,3.81678) 101 -dictGet test_01037.dict_array (34.9466,0.593463) 101 -dictGet test_01037.dict_array (34.9485,0.150307) 101 -dictGet test_01037.dict_array (34.9542,0.487238) 101 -dictGet test_01037.dict_array (34.9559,2.03473) 3501835 -dictGet test_01037.dict_array (34.9671,-0.960225) 101 -dictGet test_01037.dict_array (34.9711,2.63444) 101 -dictGet test_01037.dict_array (34.9892,0.354775) 101 -dictGet test_01037.dict_array (34.9907,1.40724) 101 -dictGet test_01037.dict_array (34.9916,-0.00173097) 101 -dictGet test_01037.dict_array (34.9919,2.06167) 101 +dictGet dict_array (29.5699,2.50068) 101 +dictGet dict_array (29.5796,1.55456) 101 +dictGet dict_array (29.5796,2.36864) 101 +dictGet dict_array (29.5844,1.59626) 101 +dictGet dict_array (29.5886,4.03321) 101 +dictGet dict_array (29.5914,3.02628) 101 +dictGet dict_array (29.5926,-0.0965169) 101 +dictGet dict_array (29.5968,2.37773) 101 +dictGet dict_array (29.5984,0.755853) 101 +dictGet dict_array (29.6066,3.47173) 101 +dictGet dict_array (29.6085,-1.26007) 6489978 +dictGet dict_array (29.6131,0.246565) 101 +dictGet dict_array (29.6157,-0.266687) 101 +dictGet dict_array (29.6164,2.94674) 101 +dictGet dict_array (29.6195,-0.591941) 101 +dictGet dict_array (29.6231,1.54818) 101 +dictGet dict_array (29.6379,0.764114) 101 +dictGet dict_array (29.6462,-0.772059) 934530 +dictGet dict_array (29.6579,-1.07336) 6489978 +dictGet dict_array (29.6618,-0.271842) 101 +dictGet dict_array (29.6629,-0.303602) 101 +dictGet dict_array (29.6659,-0.782823) 934530 +dictGet dict_array (29.6736,-0.113832) 101 +dictGet dict_array (29.6759,3.02905) 101 +dictGet dict_array (29.6778,3.71898) 101 +dictGet dict_array (29.6796,1.10433) 101 +dictGet dict_array (29.6809,2.13677) 101 +dictGet dict_array (29.6935,4.11894) 101 +dictGet dict_array (29.6991,-1.4458199999999999) 101 +dictGet dict_array (29.6997,3.17297) 101 +dictGet dict_array (29.7043,3.6145899999999997) 101 +dictGet dict_array (29.7065,3.24885) 101 +dictGet dict_array (29.7126,0.28108) 101 +dictGet dict_array (29.7192,0.174273) 101 +dictGet dict_array (29.7217,-0.523481) 3501900 +dictGet dict_array (29.7271,1.67967) 101 +dictGet dict_array (29.7311,4.12444) 101 +dictGet dict_array (29.7347,1.88378) 101 +dictGet dict_array (29.7358,0.67944) 101 +dictGet dict_array (29.7366,-0.2973) 101 +dictGet dict_array (29.7446,0.646536) 101 +dictGet dict_array (29.7453,-0.567963) 3501900 +dictGet dict_array (29.764,4.04217) 101 +dictGet dict_array (29.7655,1.51372) 101 +dictGet dict_array (29.7744,1.12435) 101 +dictGet dict_array (29.7774,-0.0681196) 3501895 +dictGet dict_array (29.7784,1.54864) 101 +dictGet dict_array (29.7785,2.24139) 101 +dictGet dict_array (29.7922,0.220808) 101 +dictGet dict_array (29.7936,2.37709) 101 +dictGet dict_array (29.8008,0.948536) 101 +dictGet dict_array (29.8115,0.201227) 101 +dictGet dict_array (29.814,0.149601) 3501895 +dictGet dict_array (29.8193,-1.35858) 101 +dictGet dict_array (29.8201,0.965518) 101 +dictGet dict_array (29.8265,-0.727286) 3501900 +dictGet dict_array (29.8277,-0.531746) 3501900 +dictGet dict_array (29.8289,3.63009) 101 +dictGet dict_array (29.8548,0.838047) 101 +dictGet dict_array (29.8641,-0.845265) 3501900 +dictGet dict_array (29.8649,0.0562212) 3501895 +dictGet dict_array (29.8701,-1.02045) 934530 +dictGet dict_array (29.8733,2.76654) 101 +dictGet dict_array (29.876,0.555475) 101 +dictGet dict_array (29.8794,-0.800108) 3501900 +dictGet dict_array (29.8813,2.7426399999999997) 101 +dictGet dict_array (29.897100000000002,2.66193) 101 +dictGet dict_array (29.908,4.01339) 101 +dictGet dict_array (29.9165,-1.08246) 3501894 +dictGet dict_array (29.9201,-0.420861) 3498054 +dictGet dict_array (29.9217,3.03778) 101 +dictGet dict_array (29.9355,0.773833) 101 +dictGet dict_array (29.947,3.76517) 101 +dictGet dict_array (29.9518,-0.60557) 3498056 +dictGet dict_array (29.9564,-0.600163) 3498056 +dictGet dict_array (29.959600000000002,4.16591) 101 +dictGet dict_array (29.9615,-1.33708) 3501894 +dictGet dict_array (29.9699,-0.392375) 3498054 +dictGet dict_array (29.9776,1.04552) 101 +dictGet dict_array (29.9784,4.02756) 101 +dictGet dict_array (29.9819,4.00597) 101 +dictGet dict_array (29.9826,1.2816100000000001) 101 +dictGet dict_array (30.0026,2.76257) 101 +dictGet dict_array (30.0126,3.68255) 101 +dictGet dict_array (30.0131,0.796576) 3501892 +dictGet dict_array (30.018,1.16523) 101 +dictGet dict_array (30.0261,-0.210653) 3501896 +dictGet dict_array (30.0472,-1.11007) 3501894 +dictGet dict_array (30.0542,-0.479585) 3498054 +dictGet dict_array (30.0613,1.6278000000000001) 101 +dictGet dict_array (30.0617,-0.0551152) 3501895 +dictGet dict_array (30.0637,2.62066) 101 +dictGet dict_array (30.0721,1.6424400000000001) 101 +dictGet dict_array (30.0769,-0.402636) 3498054 +dictGet dict_array (30.0791,-0.277435) 3501896 +dictGet dict_array (30.0931,0.0327512) 3501895 +dictGet dict_array (30.1059,3.52623) 101 +dictGet dict_array (30.1103,0.865466) 3501892 +dictGet dict_array (30.1115,2.95243) 101 +dictGet dict_array (30.1144,1.71029) 101 +dictGet dict_array (30.1311,-0.864751) 3501899 +dictGet dict_array (30.1336,-0.851386) 3501899 +dictGet dict_array (30.1393,3.89901) 101 +dictGet dict_array (30.1456,-0.531898) 3498054 +dictGet dict_array (30.1492,2.07833) 101 +dictGet dict_array (30.1575,2.43856) 101 +dictGet dict_array (30.1682,1.19771) 101 +dictGet dict_array (30.1716,3.9853300000000003) 101 +dictGet dict_array (30.1849,2.78374) 101 +dictGet dict_array (30.1866,0.65658) 3498021 +dictGet dict_array (30.1885,1.56943) 101 +dictGet dict_array (30.1959,-1.38202) 101 +dictGet dict_array (30.1999,1.58413) 101 +dictGet dict_array (30.2024,0.713081) 3498021 +dictGet dict_array (30.2054,0.620143) 3498021 +dictGet dict_array (30.2091,1.51641) 101 +dictGet dict_array (30.2124,-0.331782) 3498031 +dictGet dict_array (30.226,3.03527) 101 +dictGet dict_array (30.2261,3.18486) 101 +dictGet dict_array (30.2288,2.48407) 101 +dictGet dict_array (30.2345,3.7462400000000002) 101 +dictGet dict_array (30.2375,0.62046) 3498021 +dictGet dict_array (30.2425,-0.472914) 3498054 +dictGet dict_array (30.247,3.95863) 101 +dictGet dict_array (30.2494,-0.305093) 3498031 +dictGet dict_array (30.2499,2.54337) 101 +dictGet dict_array (30.2606,2.16644) 101 +dictGet dict_array (30.2672,3.94847) 101 +dictGet dict_array (30.2709,-0.136264) 6088794 +dictGet dict_array (30.2764,1.18654) 101 +dictGet dict_array (30.2765,1.20383) 101 +dictGet dict_array (30.2839,1.05762) 3498024 +dictGet dict_array (30.286,0.469327) 3498021 +dictGet dict_array (30.2927,3.1693) 101 +dictGet dict_array (30.2935,3.49854) 101 +dictGet dict_array (30.307,0.312338) 3498021 +dictGet dict_array (30.3085,1.07791) 3498024 +dictGet dict_array (30.3139,2.77248) 101 +dictGet dict_array (30.314,0.822823) 3498024 +dictGet dict_array (30.3227,-0.587351) 3498055 +dictGet dict_array (30.332,1.00174) 3498024 +dictGet dict_array (30.3388,0.844148) 3498024 +dictGet dict_array (30.3485,0.561902) 3498021 +dictGet dict_array (30.3497,0.180362) 6489998 +dictGet dict_array (30.361,4.13016) 101 +dictGet dict_array (30.3623,-0.0484027) 6489998 +dictGet dict_array (30.3638,3.9845800000000002) 101 +dictGet dict_array (30.3853,3.16051) 101 +dictGet dict_array (30.3974,2.6617800000000003) 101 +dictGet dict_array (30.4002,-1.15886) 101 +dictGet dict_array (30.4008,-0.387015) 3498031 +dictGet dict_array (30.4018,1.86493) 101 +dictGet dict_array (30.4239,1.16818) 3498024 +dictGet dict_array (30.4363,3.63938) 101 +dictGet dict_array (30.4377,-0.81315) 3498063 +dictGet dict_array (30.4391,3.54703) 101 +dictGet dict_array (30.4424,-1.39435) 101 +dictGet dict_array (30.4441,2.8463000000000003) 101 +dictGet dict_array (30.4517,3.28117) 101 +dictGet dict_array (30.4658,2.6928) 101 +dictGet dict_array (30.4734,2.66161) 101 +dictGet dict_array (30.4799,-1.07578) 101 +dictGet dict_array (30.4837,-1.02486) 3501899 +dictGet dict_array (30.485,1.06326) 3498024 +dictGet dict_array (30.495,1.12306) 101 +dictGet dict_array (30.501,2.27264) 101 +dictGet dict_array (30.5027,1.99382) 101 +dictGet dict_array (30.5194,-1.03943) 3501893 +dictGet dict_array (30.5239,1.04328) 101 +dictGet dict_array (30.528,3.82041) 101 +dictGet dict_array (30.5299,-0.715248) 3498063 +dictGet dict_array (30.5331,1.19603) 101 +dictGet dict_array (30.535800000000002,2.71485) 101 +dictGet dict_array (30.5405,0.804694) 3498023 +dictGet dict_array (30.542,1.23739) 101 +dictGet dict_array (30.5432,4.04189) 101 +dictGet dict_array (30.5457,-0.956121) 3501893 +dictGet dict_array (30.5506,3.07443) 101 +dictGet dict_array (30.5539,3.87084) 101 +dictGet dict_array (30.5578,3.78837) 101 +dictGet dict_array (30.5588,0.966135) 3498022 +dictGet dict_array (30.5637,2.5605) 101 +dictGet dict_array (30.5647,-1.27328) 101 +dictGet dict_array (30.5656,-0.0581332) 6088794 +dictGet dict_array (30.5715,0.65755) 3498023 +dictGet dict_array (30.5727,3.01604) 101 +dictGet dict_array (30.5729,-0.976857) 3501893 +dictGet dict_array (30.5751,0.60204) 3498023 +dictGet dict_array (30.5854,3.02473) 101 +dictGet dict_array (30.5866,0.174099) 6489998 +dictGet dict_array (30.5947,0.875193) 3498023 +dictGet dict_array (30.5992,-0.403901) 3498063 +dictGet dict_array (30.6002,4.18891) 101 +dictGet dict_array (30.6025,0.217712) 6489998 +dictGet dict_array (30.6054,0.927203) 3498022 +dictGet dict_array (30.6075,3.79359) 101 +dictGet dict_array (30.6159,3.82773) 101 +dictGet dict_array (30.627,3.84039) 101 +dictGet dict_array (30.6308,0.77517) 3498023 +dictGet dict_array (30.6338,0.179565) 6489998 +dictGet dict_array (30.6461,1.3293599999999999) 101 +dictGet dict_array (30.6674,-0.424547) 3498063 +dictGet dict_array (30.669,1.76539) 101 +dictGet dict_array (30.6788,4.01239) 101 +dictGet dict_array (30.6864,3.59158) 101 +dictGet dict_array (30.7049,-0.875413) 3501893 +dictGet dict_array (30.705,1.3307) 101 +dictGet dict_array (30.7063,-0.473192) 3498063 +dictGet dict_array (30.7075,-1.1958199999999999) 101 +dictGet dict_array (30.7101,-0.367562) 3498012 +dictGet dict_array (30.7203,2.98725) 101 +dictGet dict_array (30.7213,2.2745699999999998) 101 +dictGet dict_array (30.7446,-0.334144) 3498012 +dictGet dict_array (30.7468,3.82967) 101 +dictGet dict_array (30.747,-0.384779) 3498012 +dictGet dict_array (30.7681,0.904198) 3498022 +dictGet dict_array (30.7757,1.78743) 101 +dictGet dict_array (30.8021,-0.479212) 3498012 +dictGet dict_array (30.8079,-1.40869) 101 +dictGet dict_array (30.8206,-0.0608489) 3498012 +dictGet dict_array (30.8218,0.43909) 3498023 +dictGet dict_array (30.8239,0.10014) 3498012 +dictGet dict_array (30.8282,4.15409) 101 +dictGet dict_array (30.8288,-0.709528) 3501893 +dictGet dict_array (30.8326,0.156011) 3498012 +dictGet dict_array (30.8328,-1.03704) 101 +dictGet dict_array (30.839,2.15528) 101 +dictGet dict_array (30.8452,0.219377) 3498013 +dictGet dict_array (30.8463,0.0515355) 3498012 +dictGet dict_array (30.8526,2.06614) 101 +dictGet dict_array (30.8566,0.517876) 3498023 +dictGet dict_array (30.8588,-1.31738) 101 +dictGet dict_array (30.8681,0.44207) 3498013 +dictGet dict_array (30.8914,1.0072) 3498022 +dictGet dict_array (30.897,0.483425) 3498013 +dictGet dict_array (30.905,2.8731999999999998) 3501793 +dictGet dict_array (30.9051,2.21956) 101 +dictGet dict_array (30.9115,4.00663) 101 +dictGet dict_array (30.9167,-0.834462) 3501893 +dictGet dict_array (30.9252,-1.3289900000000001) 101 +dictGet dict_array (30.9314,1.85384) 101 +dictGet dict_array (30.9392,2.53236) 3501827 +dictGet dict_array (30.9569,2.82038) 3501793 +dictGet dict_array (30.9598,-0.641011) 3498012 +dictGet dict_array (30.9601,-0.254928) 3498012 +dictGet dict_array (30.9623,-1.3886) 101 +dictGet dict_array (30.9707,0.888854) 3498022 +dictGet dict_array (30.9766,2.81957) 3501793 +dictGet dict_array (30.9775,2.69273) 3501793 +dictGet dict_array (30.9821,0.587715) 3498013 +dictGet dict_array (30.9887,4.0233) 101 +dictGet dict_array (30.9914,0.259542) 3498013 +dictGet dict_array (30.9986,-1.36832) 101 +dictGet dict_array (31.008,0.628999) 3498013 +dictGet dict_array (31.0168,-1.17462) 101 +dictGet dict_array (31.0237,3.52547) 3501821 +dictGet dict_array (31.0306,3.78522) 101 +dictGet dict_array (31.0308,-0.72453) 3501893 +dictGet dict_array (31.0463,2.41997) 3501825 +dictGet dict_array (31.047,0.624184) 3498013 +dictGet dict_array (31.0569,0.0706393) 3498015 +dictGet dict_array (31.0583,1.3244099999999999) 3501926 +dictGet dict_array (31.063,3.23861) 3501793 +dictGet dict_array (31.068,0.695575) 3498022 +dictGet dict_array (31.0687,1.85675) 101 +dictGet dict_array (31.0692,0.254793) 3498014 +dictGet dict_array (31.0766,0.828128) 3498022 +dictGet dict_array (31.0833,0.0612782) 3498015 +dictGet dict_array (31.0833,2.59748) 3501793 +dictGet dict_array (31.0861,-1.3778299999999999) 101 +dictGet dict_array (31.0874,3.07258) 3501793 +dictGet dict_array (31.0882,1.4882) 3501926 +dictGet dict_array (31.0924,3.42242) 3501821 +dictGet dict_array (31.0927,2.67448) 3501793 +dictGet dict_array (31.0936,1.12292) 3498022 +dictGet dict_array (31.0952,-0.336928) 3498012 +dictGet dict_array (31.0978,3.48482) 3501826 +dictGet dict_array (31.1107,3.7513199999999998) 3501826 +dictGet dict_array (31.1156,1.19171) 3501926 +dictGet dict_array (31.1176,0.223509) 3498015 +dictGet dict_array (31.1249,0.946838) 3498022 +dictGet dict_array (31.1267,1.48983) 3501926 +dictGet dict_array (31.138,-0.289981) 3501898 +dictGet dict_array (31.1382,3.02904) 3501793 +dictGet dict_array (31.1475,2.6178) 3501793 +dictGet dict_array (31.1491,1.37873) 3501926 +dictGet dict_array (31.1525,3.72105) 3501826 +dictGet dict_array (31.1526,-1.4129800000000001) 101 +dictGet dict_array (31.1526,-0.186457) 3501898 +dictGet dict_array (31.1539,2.78789) 3501793 +dictGet dict_array (31.1548,-1.08552) 101 +dictGet dict_array (31.1567,-0.0768925) 3501898 +dictGet dict_array (31.1613,1.49617) 3501926 +dictGet dict_array (31.1653,1.03777) 3498022 +dictGet dict_array (31.1662,3.4214700000000002) 3501826 +dictGet dict_array (31.1672,-0.0813169) 3501898 +dictGet dict_array (31.177,0.440843) 3498014 +dictGet dict_array (31.1788,-0.737151) 3501893 +dictGet dict_array (31.1856,-0.144396) 3501898 +dictGet dict_array (31.1959,3.66813) 3501826 +dictGet dict_array (31.1996,-0.353983) 3501898 +dictGet dict_array (31.2019,2.86802) 3501793 +dictGet dict_array (31.2087,2.31245) 3501825 +dictGet dict_array (31.2125,3.2713200000000002) 3501793 +dictGet dict_array (31.2137,-0.108129) 3501898 +dictGet dict_array (31.216,3.9156) 101 +dictGet dict_array (31.2201,-0.202141) 3501898 +dictGet dict_array (31.2285,2.09058) 101 +dictGet dict_array (31.2502,4.01526) 101 +dictGet dict_array (31.2585,3.11524) 3501793 +dictGet dict_array (31.2645,-0.620418) 3501890 +dictGet dict_array (31.2684,2.74277) 3501793 +dictGet dict_array (31.2821,-1.12772) 101 +dictGet dict_array (31.2821,2.46769) 3501825 +dictGet dict_array (31.2887,3.91396) 101 +dictGet dict_array (31.295,1.49942) 3501926 +dictGet dict_array (31.2997,3.46122) 3501826 +dictGet dict_array (31.3017,3.3263) 3501826 +dictGet dict_array (31.3022,3.16754) 3501793 +dictGet dict_array (31.3048,0.364962) 3498014 +dictGet dict_array (31.305,3.1967) 3501793 +dictGet dict_array (31.3061,1.84303) 101 +dictGet dict_array (31.3082,-0.173851) 3501898 +dictGet dict_array (31.3315,3.90932) 101 +dictGet dict_array (31.3351,2.80164) 3501793 +dictGet dict_array (31.3388,0.168765) 3498015 +dictGet dict_array (31.339,0.25535) 3498094 +dictGet dict_array (31.3423,1.7036799999999999) 3501926 +dictGet dict_array (31.349,0.386456) 3498014 +dictGet dict_array (31.3558,-1.04336) 101 +dictGet dict_array (31.3564,0.478876) 3498014 +dictGet dict_array (31.3607,-0.0860507) 3498015 +dictGet dict_array (31.3831,3.84469) 101 +dictGet dict_array (31.3886,-0.731137) 3501890 +dictGet dict_array (31.4043,-0.348907) 5457271 +dictGet dict_array (31.4081,1.47391) 3501926 +dictGet dict_array (31.4176,-0.583645) 5457271 +dictGet dict_array (31.4177,1.36972) 3501926 +dictGet dict_array (31.4182,0.958303) 3498022 +dictGet dict_array (31.4199,3.1738) 3501793 +dictGet dict_array (31.4221,2.74876) 3501825 +dictGet dict_array (31.4301,-0.122643) 3498015 +dictGet dict_array (31.4344,1.00661) 3498022 +dictGet dict_array (31.4375,4.20304) 101 +dictGet dict_array (31.4377,0.289608) 3498094 +dictGet dict_array (31.4379,0.54744) 3498014 +dictGet dict_array (31.4459,3.94945) 101 +dictGet dict_array (31.4559,-0.345063) 5457271 +dictGet dict_array (31.464,0.726129) 3498014 +dictGet dict_array (31.4662,-0.299019) 3498015 +dictGet dict_array (31.4671,1.9605299999999999) 3501794 +dictGet dict_array (31.4673,-0.403676) 5457271 +dictGet dict_array (31.4712,-0.237941) 3498015 +dictGet dict_array (31.4816,0.120264) 3498015 +dictGet dict_array (31.4875,0.323483) 3498014 +dictGet dict_array (31.490099999999998,-0.338163) 5457271 +dictGet dict_array (31.4932,0.517674) 3498014 +dictGet dict_array (31.5112,1.9689299999999998) 3501794 +dictGet dict_array (31.5122,2.92785) 3501791 +dictGet dict_array (31.5151,0.166429) 3498094 +dictGet dict_array (31.5174,2.94802) 3501791 +dictGet dict_array (31.5182,4.18776) 101 +dictGet dict_array (31.5238,1.18793) 3498003 +dictGet dict_array (31.5271,3.07446) 3501791 +dictGet dict_array (31.5393,1.58061) 3501794 +dictGet dict_array (31.5421,3.13711) 3501791 +dictGet dict_array (31.5479,2.39897) 3497970 +dictGet dict_array (31.5519,0.99285) 3498003 +dictGet dict_array (31.5685,3.47987) 3501824 +dictGet dict_array (31.5959,0.437382) 3498014 +dictGet dict_array (31.6003,0.194376) 3498094 +dictGet dict_array (31.6026,2.15457) 3501794 +dictGet dict_array (31.606,2.45365) 3497970 +dictGet dict_array (31.6062,-0.453441) 3501890 +dictGet dict_array (31.6107,1.35247) 3497974 +dictGet dict_array (31.6155,3.85588) 101 +dictGet dict_array (31.6222,2.03326) 3501794 +dictGet dict_array (31.6231,-0.123059) 3498083 +dictGet dict_array (31.6244,1.6885599999999998) 3497974 +dictGet dict_array (31.6459,0.669716) 3498014 +dictGet dict_array (31.6563,-0.0644741) 3498083 +dictGet dict_array (31.6618,-0.551121) 3501890 +dictGet dict_array (31.6725,-0.38922) 3498085 +dictGet dict_array (31.6727,4.10336) 101 +dictGet dict_array (31.6739,4.1391) 101 +dictGet dict_array (31.6897,2.8694699999999997) 3501792 +dictGet dict_array (31.6902,3.98792) 101 +dictGet dict_array (31.6945,2.46687) 3497970 +dictGet dict_array (31.6987,-1.3796) 101 +dictGet dict_array (31.7012,2.34845) 3497970 +dictGet dict_array (31.7036,0.0228348) 3501888 +dictGet dict_array (31.7046,3.68111) 3501824 +dictGet dict_array (31.7055,2.92556) 3501792 +dictGet dict_array (31.7102,1.04532) 3498003 +dictGet dict_array (31.7149,-0.443302) 3498085 +dictGet dict_array (31.7195,2.99311) 3501791 +dictGet dict_array (31.7274,0.166719) 3498094 +dictGet dict_array (31.7565,-0.565382) 3498085 +dictGet dict_array (31.7615,0.771626) 3498014 +dictGet dict_array (31.7739,1.8970099999999999) 3497974 +dictGet dict_array (31.7848,1.2623199999999999) 3498003 +dictGet dict_array (31.7912,-0.788599) 101 +dictGet dict_array (31.8011,2.65853) 3497970 +dictGet dict_array (31.8032,-0.0590108) 3501888 +dictGet dict_array (31.8038,1.9618799999999998) 3497974 +dictGet dict_array (31.8098,-1.46851) 101 +dictGet dict_array (31.8131,3.41982) 3501791 +dictGet dict_array (31.8169,3.31059) 3501791 +dictGet dict_array (31.8202,-0.193692) 3501888 +dictGet dict_array (31.8306,1.57586) 3497974 +dictGet dict_array (31.8382,-0.787948) 101 +dictGet dict_array (31.8433,2.49692) 3497970 +dictGet dict_array (31.8436,2.41851) 3497970 +dictGet dict_array (31.8563,-1.10787) 101 +dictGet dict_array (31.8683,0.996504) 3498002 +dictGet dict_array (31.8693,-0.828142) 101 +dictGet dict_array (31.8723,1.08929) 3498003 +dictGet dict_array (31.8737,0.881127) 3498002 +dictGet dict_array (31.8881,-0.58441) 101 +dictGet dict_array (31.9011,0.121349) 3498094 +dictGet dict_array (31.9066,2.13045) 3497965 +dictGet dict_array (31.9142,1.03368) 3498002 +dictGet dict_array (31.9155,3.38363) 3501791 +dictGet dict_array (31.9168,1.3166) 3498004 +dictGet dict_array (31.9185,-1.11879) 101 +dictGet dict_array (31.9186,-0.647948) 101 +dictGet dict_array (31.9311,3.96928) 101 +dictGet dict_array (31.9335,1.47048) 3497974 +dictGet dict_array (31.9443,-1.36175) 101 +dictGet dict_array (31.9481,2.34231) 3497970 +dictGet dict_array (31.9526,1.36565) 3498004 +dictGet dict_array (31.9629,2.5208399999999997) 3497970 +dictGet dict_array (31.9765,0.975783) 3498002 +dictGet dict_array (31.9923,3.31773) 3501791 +dictGet dict_array (31.9994,0.972816) 3498002 +dictGet dict_array (32.001,3.47425) 3501791 +dictGet dict_array (32.0127,2.13874) 3497965 +dictGet dict_array (32.0244,3.2092) 3501792 +dictGet dict_array (32.029,1.18039) 3498004 +dictGet dict_array (32.0315,0.566073) 3498095 +dictGet dict_array (32.0354,1.0766499999999999) 3498004 +dictGet dict_array (32.0399,-1.11576) 101 +dictGet dict_array (32.053,2.16849) 3497965 +dictGet dict_array (32.0542,0.042328) 3498096 +dictGet dict_array (32.0576,2.47001) 3497970 +dictGet dict_array (32.061,3.7498899999999997) 101 +dictGet dict_array (32.0623,1.25134) 3498004 +dictGet dict_array (32.0626,1.9611399999999999) 3497965 +dictGet dict_array (32.0666,-0.0904247) 3498096 +dictGet dict_array (32.0681,2.28442) 3497970 +dictGet dict_array (32.0692,1.50869) 3497981 +dictGet dict_array (32.0724,4.03314) 101 +dictGet dict_array (32.0729,-0.064324) 101 +dictGet dict_array (32.079,0.293758) 3498094 +dictGet dict_array (32.0847,-1.19814) 101 +dictGet dict_array (32.0974,-0.91927) 101 +dictGet dict_array (32.0979,-0.736979) 101 +dictGet dict_array (32.106,-1.33063) 101 +dictGet dict_array (32.1189,0.246715) 3498094 +dictGet dict_array (32.1207,4.00883) 101 +dictGet dict_array (32.1396,1.12402) 3498004 +dictGet dict_array (32.1413,1.5668) 3497981 +dictGet dict_array (32.143,1.35559) 3498004 +dictGet dict_array (32.1538,1.32881) 3498004 +dictGet dict_array (32.1549,4.06552) 101 +dictGet dict_array (32.1555,-0.79275) 101 +dictGet dict_array (32.163,1.17733) 3498004 +dictGet dict_array (32.1634,2.94273) 3501792 +dictGet dict_array (32.1644,1.85666) 3497965 +dictGet dict_array (32.1745,0.435458) 3498095 +dictGet dict_array (32.1765,1.65149) 3497981 +dictGet dict_array (32.1893,2.08924) 3497965 +dictGet dict_array (32.2024,0.222191) 3498093 +dictGet dict_array (32.2107,1.34379) 3497981 +dictGet dict_array (32.2109,3.9018699999999997) 101 +dictGet dict_array (32.2123,1.85233) 3497965 +dictGet dict_array (32.2144,3.72534) 101 +dictGet dict_array (32.2218,2.5386699999999998) 3497970 +dictGet dict_array (32.2279,2.84267) 3497245 +dictGet dict_array (32.2345,3.33295) 3501792 +dictGet dict_array (32.2435,3.85283) 101 +dictGet dict_array (32.2527,-0.480608) 101 +dictGet dict_array (32.2566,-0.837882) 101 +dictGet dict_array (32.2627,2.57708) 3497970 +dictGet dict_array (32.2733,0.244931) 3498096 +dictGet dict_array (32.2761,4.05808) 101 +dictGet dict_array (32.2764,3.78472) 101 +dictGet dict_array (32.2814,-1.26011) 101 +dictGet dict_array (32.2861,3.02427) 3497245 +dictGet dict_array (32.2924,0.928609) 3498004 +dictGet dict_array (32.2963,-0.78543) 101 +dictGet dict_array (32.3039,3.21175) 3501792 +dictGet dict_array (32.3107,0.698287) 3498004 +dictGet dict_array (32.3138,0.0595677) 3498106 +dictGet dict_array (32.3339,0.707056) 3498004 +dictGet dict_array (32.3351,0.415474) 3498106 +dictGet dict_array (32.342,-0.681023) 101 +dictGet dict_array (32.3463,1.83196) 3497126 +dictGet dict_array (32.3494,2.43799) 3497114 +dictGet dict_array (32.3524,3.47049) 3501822 +dictGet dict_array (32.3531,2.33115) 3497114 +dictGet dict_array (32.3602,0.116106) 3498106 +dictGet dict_array (32.3612,1.1598) 3498004 +dictGet dict_array (32.3689,3.34847) 3501822 +dictGet dict_array (32.3695,0.734055) 3498004 +dictGet dict_array (32.3825,3.85017) 101 +dictGet dict_array (32.3835,-1.25491) 101 +dictGet dict_array (32.4018,-0.728568) 101 +dictGet dict_array (32.4044,2.96727) 3497245 +dictGet dict_array (32.4101,2.9988) 3497245 +dictGet dict_array (32.417,-1.12908) 101 +dictGet dict_array (32.4172,4.1952) 101 +dictGet dict_array (32.4239,2.49512) 3497245 +dictGet dict_array (32.4258,4.05137) 101 +dictGet dict_array (32.4264,-0.427357) 101 +dictGet dict_array (32.4274,3.59377) 3501822 +dictGet dict_array (32.4286,-1.24757) 101 +dictGet dict_array (32.4294,3.0665) 3497245 +dictGet dict_array (32.4333,-0.353347) 101 +dictGet dict_array (32.4391,3.64421) 3501822 +dictGet dict_array (32.4401,3.70635) 3501822 +dictGet dict_array (32.45,1.68918) 3497126 +dictGet dict_array (32.4507,-0.133471) 101 +dictGet dict_array (32.4592,0.976458) 3498105 +dictGet dict_array (32.4595,1.89135) 3497126 +dictGet dict_array (32.4604,0.280248) 3498106 +dictGet dict_array (32.4835,0.472731) 3498106 +dictGet dict_array (32.4855,2.01938) 3497126 +dictGet dict_array (32.4872,2.01697) 3497126 +dictGet dict_array (32.4911,0.613106) 3498105 +dictGet dict_array (32.4918,2.17834) 3497114 +dictGet dict_array (32.4947,2.34595) 3497114 +dictGet dict_array (32.5035,2.92234) 3497245 +dictGet dict_array (32.5132,-0.331206) 101 +dictGet dict_array (32.5156,-0.412604) 3501887 +dictGet dict_array (32.5158,2.9067499999999997) 3497245 +dictGet dict_array (32.5249,2.44519) 3497114 +dictGet dict_array (32.5293,-0.790952) 101 +dictGet dict_array (32.5319,3.96854) 101 +dictGet dict_array (32.5518,3.6093) 3501822 +dictGet dict_array (32.5541,3.5225400000000002) 3501822 +dictGet dict_array (32.5569,0.816123) 3498105 +dictGet dict_array (32.5646,1.9775) 3497126 +dictGet dict_array (32.5733,3.81271) 101 +dictGet dict_array (32.5767,0.948327) 3498105 +dictGet dict_array (32.5971,1.76179) 3497126 +dictGet dict_array (32.6035,-0.716157) 101 +dictGet dict_array (32.6087,4.21614) 101 +dictGet dict_array (32.6171,0.024481) 101 +dictGet dict_array (32.6189,-0.775391) 101 +dictGet dict_array (32.6198,2.92081) 3497167 +dictGet dict_array (32.621,-0.970784) 101 +dictGet dict_array (32.6266,0.650009) 3498105 +dictGet dict_array (32.6315,2.15144) 3497126 +dictGet dict_array (32.6385,-0.436803) 101 +dictGet dict_array (32.6449,-0.191292) 101 +dictGet dict_array (32.6535,2.10385) 3497126 +dictGet dict_array (32.6592,3.49973) 3501822 +dictGet dict_array (32.6598,2.5980600000000003) 3497114 +dictGet dict_array (32.6612,2.95681) 3497167 +dictGet dict_array (32.6636,-0.57235) 101 +dictGet dict_array (32.669,-0.382702) 101 +dictGet dict_array (32.6752,1.30748) 3497981 +dictGet dict_array (32.6811,2.9559800000000003) 3497167 +dictGet dict_array (32.6821,0.57336) 3498105 +dictGet dict_array (32.6828,3.91304) 101 +dictGet dict_array (32.6979,3.96868) 101 +dictGet dict_array (32.6983,3.15784) 3497167 +dictGet dict_array (32.7122,0.794293) 3498105 +dictGet dict_array (32.7131,-0.847256) 101 +dictGet dict_array (32.7219,0.883461) 3498105 +dictGet dict_array (32.7228,1.78808) 3497126 +dictGet dict_array (32.7273,-0.206908) 101 +dictGet dict_array (32.7292,0.259331) 3501889 +dictGet dict_array (32.7304,-1.38317) 101 +dictGet dict_array (32.7353,1.01601) 3498105 +dictGet dict_array (32.7354,4.17574) 101 +dictGet dict_array (32.7357,-0.190194) 101 +dictGet dict_array (32.7465,-1.37598) 101 +dictGet dict_array (32.7494,-0.275675) 101 +dictGet dict_array (32.7514,0.128951) 3501889 +dictGet dict_array (32.753,3.44207) 3501822 +dictGet dict_array (32.7686,2.11713) 3497126 +dictGet dict_array (32.7694,1.47159) 3497388 +dictGet dict_array (32.7768,0.0401042) 101 +dictGet dict_array (32.781,-1.34283) 101 +dictGet dict_array (32.7814,1.73876) 3497388 +dictGet dict_array (32.7856,-1.06363) 101 +dictGet dict_array (32.792699999999996,-1.1255600000000001) 101 +dictGet dict_array (32.7941,-0.645447) 101 +dictGet dict_array (32.7946,1.48889) 3497388 +dictGet dict_array (32.797,0.791753) 3501889 +dictGet dict_array (32.7982,-0.537798) 101 +dictGet dict_array (32.8091,2.3611) 3490438 +dictGet dict_array (32.81,1.7130800000000002) 3497388 +dictGet dict_array (32.8174,-0.288322) 101 +dictGet dict_array (32.823,1.6546699999999999) 3497388 +dictGet dict_array (32.8233,1.62108) 3497388 +dictGet dict_array (32.8428,-0.400045) 101 +dictGet dict_array (32.8479,2.13598) 3490438 +dictGet dict_array (32.8524,0.199902) 3501889 +dictGet dict_array (32.8543,3.23553) 3501820 +dictGet dict_array (32.8562,1.31371) 3498117 +dictGet dict_array (32.87,1.44256) 3498117 +dictGet dict_array (32.8789,2.38192) 3490438 +dictGet dict_array (32.8812,2.20734) 3497128 +dictGet dict_array (32.8815,-0.54427) 101 +dictGet dict_array (32.8853,2.4859) 3497128 +dictGet dict_array (32.8909,0.513964) 3501889 +dictGet dict_array (32.9035,2.38999) 3490438 +dictGet dict_array (32.9097,2.48131) 3497128 +dictGet dict_array (32.928,-0.943269) 101 +dictGet dict_array (32.9322,1.13165) 3498104 +dictGet dict_array (32.9348,1.22606) 3498117 +dictGet dict_array (32.9417,3.77998) 3501822 +dictGet dict_array (32.9428,3.11936) 3497167 +dictGet dict_array (32.9482,1.18092) 3498118 +dictGet dict_array (32.9506,0.0609364) 101 +dictGet dict_array (32.953,-0.828308) 101 +dictGet dict_array (32.9593,3.5209099999999998) 3501822 +dictGet dict_array (32.9617,2.07711) 3497128 +dictGet dict_array (32.966,0.693749) 3498104 +dictGet dict_array (32.9668,-0.716432) 101 +dictGet dict_array (32.9702,1.98555) 3497127 +dictGet dict_array (32.9782,1.73819) 3497388 +dictGet dict_array (32.9805,3.71151) 3501822 +dictGet dict_array (32.9821,2.97225) 3497167 +dictGet dict_array (32.995,-0.830301) 101 +dictGet dict_array (33.0234,0.770848) 3498104 +dictGet dict_array (33.0312,-0.340964) 101 +dictGet dict_array (33.0366,-0.756795) 101 +dictGet dict_array (33.0438,0.812871) 3498118 +dictGet dict_array (33.0455,1.84843) 3497127 +dictGet dict_array (33.0498,0.0913292) 101 +dictGet dict_array (33.0506,1.53739) 3497364 +dictGet dict_array (33.0554,2.4265) 3497363 +dictGet dict_array (33.0741,3.61332) 3501822 +dictGet dict_array (33.0765,-0.179985) 101 +dictGet dict_array (33.087,1.46465) 3497399 +dictGet dict_array (33.0906,-0.620383) 101 +dictGet dict_array (33.1047,-1.28027) 101 +dictGet dict_array (33.1072,1.96303) 3497127 +dictGet dict_array (33.1081,-0.897874) 101 +dictGet dict_array (33.1122,1.8950200000000001) 3497127 +dictGet dict_array (33.1237,2.63993) 3497165 +dictGet dict_array (33.1238,0.753963) 3498118 +dictGet dict_array (33.1257,0.495668) 3498102 +dictGet dict_array (33.1258,1.78341) 3497364 +dictGet dict_array (33.127,2.59646) 3497166 +dictGet dict_array (33.1324,-1.23742) 101 +dictGet dict_array (33.1359,3.83491) 101 +dictGet dict_array (33.1628,-0.379588) 101 +dictGet dict_array (33.1679,1.25601) 3498117 +dictGet dict_array (33.1688,-1.35553) 101 +dictGet dict_array (33.181,2.10943) 3497363 +dictGet dict_array (33.1871,2.81171) 3497165 +dictGet dict_array (33.1877,0.771297) 3498118 +dictGet dict_array (33.1883,-0.204797) 101 +dictGet dict_array (33.1886,3.27998) 3501820 +dictGet dict_array (33.1955,0.708907) 3498118 +dictGet dict_array (33.2044,-0.769275) 101 +dictGet dict_array (33.2182,3.36103) 3501820 +dictGet dict_array (33.2192,3.43586) 3501822 +dictGet dict_array (33.2322,-0.916753) 101 +dictGet dict_array (33.2359,-0.81321) 101 +dictGet dict_array (33.238,0.635072) 3498111 +dictGet dict_array (33.2398,3.02588) 3497165 +dictGet dict_array (33.2469,2.35698) 3497363 +dictGet dict_array (33.247,2.3327) 3497363 +dictGet dict_array (33.2579,2.8027100000000003) 3497165 +dictGet dict_array (33.2607,0.321082) 101 +dictGet dict_array (33.2653,0.243336) 101 +dictGet dict_array (33.2758,0.831836) 3498118 +dictGet dict_array (33.2771,0.886536) 3498118 +dictGet dict_array (33.2914,1.16026) 3498117 +dictGet dict_array (33.2914,1.38882) 3497399 +dictGet dict_array (33.2982,-1.16604) 101 +dictGet dict_array (33.2985,0.842556) 3498112 +dictGet dict_array (33.3005,2.8338900000000002) 3497165 +dictGet dict_array (33.305,0.0969475) 101 +dictGet dict_array (33.3072,3.82163) 101 +dictGet dict_array (33.312,3.41475) 3501820 +dictGet dict_array (33.3129,2.46048) 3497166 +dictGet dict_array (33.3134,3.46863) 3501820 +dictGet dict_array (33.3203,2.33139) 3497166 +dictGet dict_array (33.324,0.433701) 101 +dictGet dict_array (33.3338,2.44705) 3497166 +dictGet dict_array (33.337,4.06475) 101 +dictGet dict_array (33.3469,1.08172) 3498126 +dictGet dict_array (33.3538,0.717896) 3498112 +dictGet dict_array (33.3618,1.37899) 3497399 +dictGet dict_array (33.3698,0.547744) 3501862 +dictGet dict_array (33.3705,0.957619) 3498112 +dictGet dict_array (33.3821,3.07258) 3497165 +dictGet dict_array (33.3881,3.0626) 3497165 +dictGet dict_array (33.393,-0.816186) 101 +dictGet dict_array (33.3945,0.869508) 3498110 +dictGet dict_array (33.4001,1.24186) 3498117 +dictGet dict_array (33.4008,2.34911) 3497166 +dictGet dict_array (33.4166,-1.2808899999999999) 101 +dictGet dict_array (33.4167,3.0655) 3497165 +dictGet dict_array (33.4204,2.81887) 3497165 +dictGet dict_array (33.4211,1.71128) 3497400 +dictGet dict_array (33.4237,2.91761) 3497165 +dictGet dict_array (33.4266,1.5955599999999999) 3497399 +dictGet dict_array (33.4353,-0.391392) 101 +dictGet dict_array (33.4362,-0.134658) 101 +dictGet dict_array (33.4386,0.15396) 101 +dictGet dict_array (33.4421,-0.50712) 101 +dictGet dict_array (33.452,0.915829) 3498126 +dictGet dict_array (33.463,-0.0882717) 101 +dictGet dict_array (33.464,-1.00949) 101 +dictGet dict_array (33.4692,0.954092) 3498126 +dictGet dict_array (33.4716,1.9538799999999998) 3497400 +dictGet dict_array (33.4756,1.85836) 3497400 +dictGet dict_array (33.4859,4.0751) 101 +dictGet dict_array (33.4899,3.54193) 3501820 +dictGet dict_array (33.4935,3.49794) 3501820 +dictGet dict_array (33.494,-0.983356) 101 +dictGet dict_array (33.4955,-1.28128) 101 +dictGet dict_array (33.4965,-0.278687) 101 +dictGet dict_array (33.4991,0.647491) 3498110 +dictGet dict_array (33.5076,2.2272) 3497424 +dictGet dict_array (33.5079,-0.498199) 101 +dictGet dict_array (33.5157,0.535034) 3501862 +dictGet dict_array (33.5171,2.49677) 3497166 +dictGet dict_array (33.5255,2.4447200000000002) 3497166 +dictGet dict_array (33.526,4.01194) 101 +dictGet dict_array (33.5288,0.789434) 3498110 +dictGet dict_array (33.5356,-1.17671) 101 +dictGet dict_array (33.5402,1.49152) 3497399 +dictGet dict_array (33.5418,3.45757) 3501820 +dictGet dict_array (33.5428,1.90712) 3497400 +dictGet dict_array (33.5556,-0.55741) 101 +dictGet dict_array (33.5564,0.876858) 3498128 +dictGet dict_array (33.5567,-0.10208) 101 +dictGet dict_array (33.5645,-0.124824) 101 +dictGet dict_array (33.5663,3.4872) 3501820 +dictGet dict_array (33.5716,-0.0107611) 101 +dictGet dict_array (33.578,3.55714) 3501820 +dictGet dict_array (33.5826,-0.49076) 101 +dictGet dict_array (33.5909,0.773737) 3498110 +dictGet dict_array (33.5958,2.9619999999999997) 3497425 +dictGet dict_array (33.6193,-0.919755) 101 +dictGet dict_array (33.6313,0.652132) 3498110 +dictGet dict_array (33.632,0.823351) 3498128 +dictGet dict_array (33.66,2.18998) 3497424 +dictGet dict_array (33.6621,0.535395) 3498135 +dictGet dict_array (33.6726,3.19367) 3497438 +dictGet dict_array (33.6912,1.74522) 3497400 +dictGet dict_array (33.705,0.706397) 3498135 +dictGet dict_array (33.7076,0.7622) 3498128 +dictGet dict_array (33.7112,1.70187) 3497400 +dictGet dict_array (33.7246,-1.14837) 101 +dictGet dict_array (33.7326,2.62413) 3497425 +dictGet dict_array (33.7332,2.82137) 3497425 +dictGet dict_array (33.7434,0.394672) 3498135 +dictGet dict_array (33.7443,1.54557) 3497398 +dictGet dict_array (33.7506,1.57317) 3497398 +dictGet dict_array (33.7526,1.8578999999999999) 3497424 +dictGet dict_array (33.766,4.15013) 101 +dictGet dict_array (33.7834,2.41789) 3497439 +dictGet dict_array (33.7864,0.230935) 101 +dictGet dict_array (33.7965,3.05709) 3497438 +dictGet dict_array (33.7998,3.32881) 3497438 +dictGet dict_array (33.8003,2.97338) 3497425 +dictGet dict_array (33.8007,-1.08962) 101 +dictGet dict_array (33.8022,-0.139488) 101 +dictGet dict_array (33.8065,2.70857) 3497425 +dictGet dict_array (33.8169,-0.607788) 101 +dictGet dict_array (33.8203,0.108512) 3501863 +dictGet dict_array (33.8231,-1.03449) 101 +dictGet dict_array (33.8312,3.49458) 3501829 +dictGet dict_array (33.8342,0.297518) 3501863 +dictGet dict_array (33.8352,0.165872) 101 +dictGet dict_array (33.8354,1.87277) 3497424 +dictGet dict_array (33.8371,1.60103) 3497398 +dictGet dict_array (33.8387,1.9968) 3497424 +dictGet dict_array (33.8403,3.5805) 3501829 +dictGet dict_array (33.8414,-0.703067) 101 +dictGet dict_array (33.844,-0.179472) 101 +dictGet dict_array (33.8468,3.40137) 3501829 +dictGet dict_array (33.8509,4.15334) 101 +dictGet dict_array (33.8539,2.38339) 3497439 +dictGet dict_array (33.858,-1.3122500000000001) 101 +dictGet dict_array (33.859,3.72626) 3501829 +dictGet dict_array (33.8616,2.24433) 3497424 +dictGet dict_array (33.8621,3.01035) 3497438 +dictGet dict_array (33.8623,1.17559) 3498129 +dictGet dict_array (33.8682,2.706) 3497425 +dictGet dict_array (33.8684,0.189231) 3501863 +dictGet dict_array (33.872,1.93574) 3497424 +dictGet dict_array (33.8844,3.80404) 3501829 +dictGet dict_array (33.8888,0.594884) 3498135 +dictGet dict_array (33.8946,2.74161) 3497438 +dictGet dict_array (33.9023,0.6239) 3498135 +dictGet dict_array (33.9057,0.873222) 3498136 +dictGet dict_array (33.9157,-1.26607) 101 +dictGet dict_array (33.92,2.06848) 3497397 +dictGet dict_array (33.9298,-0.00526229) 101 +dictGet dict_array (33.932,3.07063) 3497438 +dictGet dict_array (33.9322,0.629385) 3501864 +dictGet dict_array (33.9367,-1.41955) 101 +dictGet dict_array (33.937,1.42532) 3498173 +dictGet dict_array (33.9375,1.1467100000000001) 3498159 +dictGet dict_array (33.9434,-1.05739) 101 +dictGet dict_array (33.9477,3.34809) 3501829 +dictGet dict_array (33.95,2.21715) 3497397 +dictGet dict_array (33.955799999999996,0.305176) 3501859 +dictGet dict_array (33.9686,-0.28273) 101 +dictGet dict_array (33.9703,4.1255) 3501829 +dictGet dict_array (33.9707,3.08199) 3497438 +dictGet dict_array (33.9754,1.06203) 3498159 +dictGet dict_array (33.9757,3.72468) 3501829 +dictGet dict_array (33.9775,-0.0440599) 101 +dictGet dict_array (33.9777,-0.251484) 101 +dictGet dict_array (33.9789,-0.339374) 101 +dictGet dict_array (33.9849,2.54515) 3497425 +dictGet dict_array (33.9885,-0.318557) 101 +dictGet dict_array (33.9977,1.07175) 3498159 +dictGet dict_array (33.9984,-0.700517) 101 +dictGet dict_array (34.0149,3.53338) 3501829 +dictGet dict_array (34.0173,3.39155) 3501829 +dictGet dict_array (34.0317,3.9579) 3501829 +dictGet dict_array (34.0369,3.83612) 3501829 +dictGet dict_array (34.043,-0.0887221) 101 +dictGet dict_array (34.0487,1.14252) 3498159 +dictGet dict_array (34.052,1.74832) 3497397 +dictGet dict_array (34.0711,-0.898071) 101 +dictGet dict_array (34.0747,1.55057) 3498173 +dictGet dict_array (34.0803,3.16763) 3497438 +dictGet dict_array (34.0872,3.75555) 3501829 +dictGet dict_array (34.0965,1.62038) 3498173 +dictGet dict_array (34.0977,-0.412691) 101 +dictGet dict_array (34.0986,0.0294206) 101 +dictGet dict_array (34.1072,3.15823) 3497438 +dictGet dict_array (34.1092,3.09599) 3497438 +dictGet dict_array (34.1206,1.04637) 3498160 +dictGet dict_array (34.1209,3.13826) 3497438 +dictGet dict_array (34.1265,3.95881) 3501829 +dictGet dict_array (34.1286,-0.539319) 101 +dictGet dict_array (34.1358,3.67451) 3501829 +dictGet dict_array (34.1428,0.136115) 101 +dictGet dict_array (34.157,1.73522) 3497397 +dictGet dict_array (34.1581,1.48001) 3498172 +dictGet dict_array (34.1682,3.42373) 3501829 +dictGet dict_array (34.1683,-1.26511) 101 +dictGet dict_array (34.1684,4.20007) 101 +dictGet dict_array (34.1854,3.32089) 3501829 +dictGet dict_array (34.2022,0.749536) 3501864 +dictGet dict_array (34.2044,3.04865) 3497438 +dictGet dict_array (34.22,-0.500055) 101 +dictGet dict_array (34.2249,0.743775) 3501864 +dictGet dict_array (34.2254,1.34702) 3498172 +dictGet dict_array (34.2355,-0.898843) 101 +dictGet dict_array (34.2394,2.0203699999999998) 3497439 +dictGet dict_array (34.2466,1.83785) 3498251 +dictGet dict_array (34.247,4.09563) 101 +dictGet dict_array (34.2508,2.61312) 3497439 +dictGet dict_array (34.2517,1.69642) 3498251 +dictGet dict_array (34.2564,4.13033) 101 +dictGet dict_array (34.2574,4.18928) 101 +dictGet dict_array (34.2614,-0.478719) 101 +dictGet dict_array (34.2625,2.38088) 3497439 +dictGet dict_array (34.2666,3.1503) 3501829 +dictGet dict_array (34.271,4.02223) 101 +dictGet dict_array (34.2727,0.514755) 101 +dictGet dict_array (34.278,1.98929) 3497439 +dictGet dict_array (34.2798,-0.199208) 101 +dictGet dict_array (34.2804,2.05184) 3497439 +dictGet dict_array (34.2945,-1.11051) 101 +dictGet dict_array (34.3168,-0.0829721) 101 +dictGet dict_array (34.3345,3.4358) 3501829 +dictGet dict_array (34.3377,1.13527) 3498162 +dictGet dict_array (34.3383,1.27891) 3498161 +dictGet dict_array (34.3391,1.47945) 3498161 +dictGet dict_array (34.3441,0.627014) 101 +dictGet dict_array (34.347,2.4853) 3497439 +dictGet dict_array (34.3514,2.16247) 3497439 +dictGet dict_array (34.3627,2.64533) 3497439 +dictGet dict_array (34.3682,-0.227501) 101 +dictGet dict_array (34.3756,4.21248) 101 +dictGet dict_array (34.379,3.96604) 101 +dictGet dict_array (34.3827,1.7518) 3498251 +dictGet dict_array (34.3912,2.8834) 3501830 +dictGet dict_array (34.3919,0.668829) 101 +dictGet dict_array (34.3949,2.00338) 3497439 +dictGet dict_array (34.3987,0.557268) 101 +dictGet dict_array (34.4111,0.768558) 101 +dictGet dict_array (34.4119,2.8742) 3501830 +dictGet dict_array (34.416,3.50841) 3501829 +dictGet dict_array (34.4212,1.24916) 3498161 +dictGet dict_array (34.4251,0.457029) 101 +dictGet dict_array (34.4274,-0.902559) 101 +dictGet dict_array (34.4325,4.03159) 101 +dictGet dict_array (34.438,1.63994) 3498251 +dictGet dict_array (34.4403,-0.177594) 101 +dictGet dict_array (34.4421,0.726712) 101 +dictGet dict_array (34.4517,2.98611) 3501830 +dictGet dict_array (34.4658,-1.312) 101 +dictGet dict_array (34.4732,-0.0681338) 101 +dictGet dict_array (34.4752,2.81646) 3501830 +dictGet dict_array (34.4914,2.3858) 3497439 +dictGet dict_array (34.4923,0.855231) 101 +dictGet dict_array (34.5235,1.78468) 3498251 +dictGet dict_array (34.5305,4.10608) 101 +dictGet dict_array (34.5389,0.621937) 101 +dictGet dict_array (34.5406,3.17145) 101 +dictGet dict_array (34.5434,-0.56306) 101 +dictGet dict_array (34.5449,3.13311) 3501829 +dictGet dict_array (34.5491,2.31572) 3497439 +dictGet dict_array (34.5539,2.94028) 3501830 +dictGet dict_array (34.5546,-0.208825) 101 +dictGet dict_array (34.5549,3.78486) 101 +dictGet dict_array (34.5676,0.307148) 101 +dictGet dict_array (34.5743,1.5217399999999999) 3501838 +dictGet dict_array (34.5775,3.48046) 101 +dictGet dict_array (34.5815,2.5243700000000002) 3501830 +dictGet dict_array (34.5841,4.21191) 101 +dictGet dict_array (34.5887,2.65083) 3501830 +dictGet dict_array (34.5937,3.2143) 101 +dictGet dict_array (34.6013,-1.0612) 101 +dictGet dict_array (34.6089,1.36066) 3501838 +dictGet dict_array (34.6103,3.40227) 101 +dictGet dict_array (34.6128,1.92276) 3498251 +dictGet dict_array (34.6175,2.43627) 3498251 +dictGet dict_array (34.6209,3.43776) 101 +dictGet dict_array (34.6234,2.60237) 3501830 +dictGet dict_array (34.6275,3.52479) 101 +dictGet dict_array (34.635,0.568558) 101 +dictGet dict_array (34.6373,2.37692) 3498251 +dictGet dict_array (34.6375,3.52234) 101 +dictGet dict_array (34.6426,2.12397) 3498251 +dictGet dict_array (34.6513,2.80915) 3501830 +dictGet dict_array (34.6632,2.30039) 3498251 +dictGet dict_array (34.6691,1.86582) 3498251 +dictGet dict_array (34.6739,0.15342) 101 +dictGet dict_array (34.6825,0.0499679) 101 +dictGet dict_array (34.6893,0.454326) 101 +dictGet dict_array (34.6957,-0.358598) 101 +dictGet dict_array (34.6986,0.562679) 101 +dictGet dict_array (34.712,1.12114) 101 +dictGet dict_array (34.7126,-0.0057301) 101 +dictGet dict_array (34.7137,0.0248501) 101 +dictGet dict_array (34.7162,1.15623) 101 +dictGet dict_array (34.7258,3.95142) 101 +dictGet dict_array (34.7347,3.5232099999999997) 101 +dictGet dict_array (34.7363,2.23374) 3501830 +dictGet dict_array (34.7375,0.397841) 101 +dictGet dict_array (34.7423,3.09198) 101 +dictGet dict_array (34.7452,3.09029) 101 +dictGet dict_array (34.7539,-1.06943) 101 +dictGet dict_array (34.7733,-0.00912717) 101 +dictGet dict_array (34.774,2.71088) 3501830 +dictGet dict_array (34.7771,1.46009) 3501835 +dictGet dict_array (34.7782,-1.28308) 101 +dictGet dict_array (34.7924,3.63564) 101 +dictGet dict_array (34.7939,-0.416676) 101 +dictGet dict_array (34.7964,-0.401773) 101 +dictGet dict_array (34.7974,0.0286873) 101 +dictGet dict_array (34.7975,3.05965) 101 +dictGet dict_array (34.8037,3.07263) 101 +dictGet dict_array (34.8254,-0.390284) 101 +dictGet dict_array (34.828,1.91869) 3498251 +dictGet dict_array (34.8289,3.71058) 101 +dictGet dict_array (34.8403,2.14606) 3501835 +dictGet dict_array (34.8437,2.20617) 3501830 +dictGet dict_array (34.8469,2.38435) 3501830 +dictGet dict_array (34.86,1.45705) 101 +dictGet dict_array (34.8612,0.914248) 101 +dictGet dict_array (34.8663,3.4215400000000002) 101 +dictGet dict_array (34.8724,-0.375144) 101 +dictGet dict_array (34.8795,3.29317) 101 +dictGet dict_array (34.8823,1.21988) 101 +dictGet dict_array (34.8834,1.07657) 101 +dictGet dict_array (34.8837,0.157648) 101 +dictGet dict_array (34.8871,-0.9755) 101 +dictGet dict_array (34.8871,1.8943699999999999) 3501835 +dictGet dict_array (34.889,3.36756) 101 +dictGet dict_array (34.8907,1.24874) 101 +dictGet dict_array (34.8965,3.13508) 101 +dictGet dict_array (34.9042,2.62092) 101 +dictGet dict_array (34.9055,-0.0448967) 101 +dictGet dict_array (34.9122,0.110576) 101 +dictGet dict_array (34.9228,3.60183) 101 +dictGet dict_array (34.9237,1.21715) 101 +dictGet dict_array (34.9296,1.70459) 3501835 +dictGet dict_array (34.941,-1.14663) 101 +dictGet dict_array (34.9448,1.18923) 101 +dictGet dict_array (34.9462,3.81678) 101 +dictGet dict_array (34.9466,0.593463) 101 +dictGet dict_array (34.9485,0.150307) 101 +dictGet dict_array (34.9542,0.487238) 101 +dictGet dict_array (34.9559,2.03473) 3501835 +dictGet dict_array (34.9671,-0.960225) 101 +dictGet dict_array (34.9711,2.63444) 101 +dictGet dict_array (34.9892,0.354775) 101 +dictGet dict_array (34.9907,1.40724) 101 +dictGet dict_array (34.9916,-0.00173097) 101 +dictGet dict_array (34.9919,2.06167) 101 diff --git a/tests/queries/0_stateless/01037_polygon_dicts_correctness_fast.sh b/tests/queries/0_stateless/01037_polygon_dicts_correctness_fast.sh index f6880ae5009..47f7a5c1c4f 100755 --- a/tests/queries/0_stateless/01037_polygon_dicts_correctness_fast.sh +++ b/tests/queries/0_stateless/01037_polygon_dicts_correctness_fast.sh @@ -1,31 +1,31 @@ #!/usr/bin/env bash -# Tags: no-debug, no-parallel +# Tags: no-debug CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -TMP_DIR="/tmp" +TMP_DIR=${CLICKHOUSE_TMP}/tmp +DATA_DIR=${CLICKHOUSE_TMP}/data +mkdir -p $TMP_DIR +mkdir -p $DATA_DIR declare -a SearchTypes=("POLYGON_INDEX_EACH" "POLYGON_INDEX_CELL") -tar -xf "${CURDIR}"/01037_test_data_perf.tar.gz -C "${CURDIR}" +tar -xf "${CURDIR}"/01037_test_data_perf.tar.gz -C "${DATA_DIR}" $CLICKHOUSE_CLIENT -n --query=" -DROP DATABASE IF EXISTS test_01037; -CREATE DATABASE test_01037; -DROP TABLE IF EXISTS test_01037.points; -CREATE TABLE test_01037.points (x Float64, y Float64) ENGINE = Memory; +CREATE TABLE points (x Float64, y Float64) ENGINE = Memory; " -$CLICKHOUSE_CLIENT --query="INSERT INTO test_01037.points FORMAT TSV" --min_chunk_bytes_for_parallel_parsing=10485760 --max_insert_block_size=100000 < "${CURDIR}/01037_point_data" +$CLICKHOUSE_CLIENT --query="INSERT INTO points FORMAT TSV" --min_chunk_bytes_for_parallel_parsing=10485760 --max_insert_block_size=100000 < "${DATA_DIR}/01037_point_data" -rm "${CURDIR}"/01037_point_data +rm "${DATA_DIR}"/01037_point_data $CLICKHOUSE_CLIENT -n --query=" -DROP TABLE IF EXISTS test_01037.polygons_array; +DROP TABLE IF EXISTS polygons_array; -CREATE TABLE test_01037.polygons_array +CREATE TABLE polygons_array ( key Array(Array(Array(Array(Float64)))), name String, @@ -34,36 +34,32 @@ CREATE TABLE test_01037.polygons_array ENGINE = Memory; " -$CLICKHOUSE_CLIENT --query="INSERT INTO test_01037.polygons_array FORMAT JSONEachRow" --min_chunk_bytes_for_parallel_parsing=10485760 --max_insert_block_size=100000 < "${CURDIR}/01037_polygon_data" +$CLICKHOUSE_CLIENT --query="INSERT INTO polygons_array FORMAT JSONEachRow" --min_chunk_bytes_for_parallel_parsing=10485760 --max_insert_block_size=100000 < "${DATA_DIR}/01037_polygon_data" -rm "${CURDIR}"/01037_polygon_data +rm "${DATA_DIR}"/01037_polygon_data for type in "${SearchTypes[@]}"; do outputFile="${TMP_DIR}/results${type}.out" $CLICKHOUSE_CLIENT -n --query=" - DROP DICTIONARY IF EXISTS test_01037.dict_array; + DROP DICTIONARY IF EXISTS dict_array; - CREATE DICTIONARY test_01037.dict_array + CREATE DICTIONARY dict_array ( key Array(Array(Array(Array(Float64)))), name String DEFAULT 'qqq', value UInt64 DEFAULT 101 ) PRIMARY KEY key - SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'polygons_array' PASSWORD '' DB 'test_01037')) + SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'polygons_array' PASSWORD '' DB currentDatabase())) LIFETIME(0) LAYOUT($type()); - select 'dictGet', 'test_01037.dict_array' as dict_name, tuple(x, y) as key, - dictGet(dict_name, 'value', key) from test_01037.points order by x, y; + select 'dictGet', 'dict_array' as dict_name, tuple(x, y) as key, + dictGet(dict_name, 'value', key) from points order by x, y; " > "$outputFile" diff -q "${CURDIR}/01037_polygon_dicts_correctness_fast.ans" "$outputFile" done -$CLICKHOUSE_CLIENT -n --query=" -DROP TABLE test_01037.points; -DROP DATABASE test_01037; -" diff --git a/tests/queries/0_stateless/01037_polygon_dicts_simple_functions.ans b/tests/queries/0_stateless/01037_polygon_dicts_simple_functions.ans index dfad14fb113..937539643ec 100644 --- a/tests/queries/0_stateless/01037_polygon_dicts_simple_functions.ans +++ b/tests/queries/0_stateless/01037_polygon_dicts_simple_functions.ans @@ -1,104 +1,104 @@ -dictGet test_01037.dict_array (-100,-42) qqq 101 -dictGet test_01037.dict_array (-1,0) Click South 423 -dictGet test_01037.dict_array (-0.1,0) Click South 423 -dictGet test_01037.dict_array (0,-2) Click West 424 -dictGet test_01037.dict_array (0,-1.1) Click West 424 -dictGet test_01037.dict_array (0,1.1) Click North 422 -dictGet test_01037.dict_array (0,2) Click North 422 -dictGet test_01037.dict_array (0.1,0) Click East 421 -dictGet test_01037.dict_array (0.99,2.99) Click North 422 -dictGet test_01037.dict_array (1,0) Click East 421 -dictGet test_01037.dict_array (3,3) House 314159 -dictGet test_01037.dict_array (5,6) Click 42 -dictGet test_01037.dict_array (7.01,7.01) qqq 101 -dictGetOrDefault test_01037.dict_array (-100,-42) www 1234 -dictGetOrDefault test_01037.dict_array (-1,0) Click South 423 -dictGetOrDefault test_01037.dict_array (-0.1,0) Click South 423 -dictGetOrDefault test_01037.dict_array (0,-2) Click West 424 -dictGetOrDefault test_01037.dict_array (0,-1.1) Click West 424 -dictGetOrDefault test_01037.dict_array (0,1.1) Click North 422 -dictGetOrDefault test_01037.dict_array (0,2) Click North 422 -dictGetOrDefault test_01037.dict_array (0.1,0) Click East 421 -dictGetOrDefault test_01037.dict_array (0.99,2.99) Click North 422 -dictGetOrDefault test_01037.dict_array (1,0) Click East 421 -dictGetOrDefault test_01037.dict_array (3,3) House 314159 -dictGetOrDefault test_01037.dict_array (5,6) Click 42 -dictGetOrDefault test_01037.dict_array (7.01,7.01) www 1234 -dictGetOrDefault test_01037.dict_array (-100,-42) dd 44 -dictGetOrDefault test_01037.dict_array (-1,0) Click South 423 -dictGetOrDefault test_01037.dict_array (-0.1,0) Click South 423 -dictGetOrDefault test_01037.dict_array (0,-2) Click West 424 -dictGetOrDefault test_01037.dict_array (0,-1.1) Click West 424 -dictGetOrDefault test_01037.dict_array (0,1.1) Click North 422 -dictGetOrDefault test_01037.dict_array (0,2) Click North 422 -dictGetOrDefault test_01037.dict_array (0.1,0) Click East 421 -dictGetOrDefault test_01037.dict_array (0.99,2.99) Click North 422 -dictGetOrDefault test_01037.dict_array (1,0) Click East 421 -dictGetOrDefault test_01037.dict_array (3,3) House 314159 -dictGetOrDefault test_01037.dict_array (5,6) Click 42 -dictGetOrDefault test_01037.dict_array (7.01,7.01) ee 55 -dictGet test_01037.dict_tuple (-100,-42) qqq 101 -dictGet test_01037.dict_tuple (-1,0) Click South 423 -dictGet test_01037.dict_tuple (-0.1,0) Click South 423 -dictGet test_01037.dict_tuple (0,-2) Click West 424 -dictGet test_01037.dict_tuple (0,-1.1) Click West 424 -dictGet test_01037.dict_tuple (0,1.1) Click North 422 -dictGet test_01037.dict_tuple (0,2) Click North 422 -dictGet test_01037.dict_tuple (0.1,0) Click East 421 -dictGet test_01037.dict_tuple (0.99,2.99) Click North 422 -dictGet test_01037.dict_tuple (1,0) Click East 421 -dictGet test_01037.dict_tuple (3,3) House 314159 -dictGet test_01037.dict_tuple (5,6) Click 42 -dictGet test_01037.dict_tuple (7.01,7.01) qqq 101 -dictGetOrDefault test_01037.dict_tuple (-100,-42) www 1234 -dictGetOrDefault test_01037.dict_tuple (-1,0) Click South 423 -dictGetOrDefault test_01037.dict_tuple (-0.1,0) Click South 423 -dictGetOrDefault test_01037.dict_tuple (0,-2) Click West 424 -dictGetOrDefault test_01037.dict_tuple (0,-1.1) Click West 424 -dictGetOrDefault test_01037.dict_tuple (0,1.1) Click North 422 -dictGetOrDefault test_01037.dict_tuple (0,2) Click North 422 -dictGetOrDefault test_01037.dict_tuple (0.1,0) Click East 421 -dictGetOrDefault test_01037.dict_tuple (0.99,2.99) Click North 422 -dictGetOrDefault test_01037.dict_tuple (1,0) Click East 421 -dictGetOrDefault test_01037.dict_tuple (3,3) House 314159 -dictGetOrDefault test_01037.dict_tuple (5,6) Click 42 -dictGetOrDefault test_01037.dict_tuple (7.01,7.01) www 1234 -dictGetOrDefault test_01037.dict_tuple (-100,-42) dd 44 -dictGetOrDefault test_01037.dict_tuple (-1,0) Click South 423 -dictGetOrDefault test_01037.dict_tuple (-0.1,0) Click South 423 -dictGetOrDefault test_01037.dict_tuple (0,-2) Click West 424 -dictGetOrDefault test_01037.dict_tuple (0,-1.1) Click West 424 -dictGetOrDefault test_01037.dict_tuple (0,1.1) Click North 422 -dictGetOrDefault test_01037.dict_tuple (0,2) Click North 422 -dictGetOrDefault test_01037.dict_tuple (0.1,0) Click East 421 -dictGetOrDefault test_01037.dict_tuple (0.99,2.99) Click North 422 -dictGetOrDefault test_01037.dict_tuple (1,0) Click East 421 -dictGetOrDefault test_01037.dict_tuple (3,3) House 314159 -dictGetOrDefault test_01037.dict_tuple (5,6) Click 42 -dictGetOrDefault test_01037.dict_tuple (7.01,7.01) ee 55 -dictHas test_01037.dict_array (-100,-42) 0 -dictHas test_01037.dict_array (-1,0) 1 -dictHas test_01037.dict_array (-0.1,0) 1 -dictHas test_01037.dict_array (0,-2) 1 -dictHas test_01037.dict_array (0,-1.1) 1 -dictHas test_01037.dict_array (0,1.1) 1 -dictHas test_01037.dict_array (0,2) 1 -dictHas test_01037.dict_array (0.1,0) 1 -dictHas test_01037.dict_array (0.99,2.99) 1 -dictHas test_01037.dict_array (1,0) 1 -dictHas test_01037.dict_array (3,3) 1 -dictHas test_01037.dict_array (5,6) 1 -dictHas test_01037.dict_array (7.01,7.01) 0 -dictHas test_01037.dict_tuple (-100,-42) 0 -dictHas test_01037.dict_tuple (-1,0) 1 -dictHas test_01037.dict_tuple (-0.1,0) 1 -dictHas test_01037.dict_tuple (0,-2) 1 -dictHas test_01037.dict_tuple (0,-1.1) 1 -dictHas test_01037.dict_tuple (0,1.1) 1 -dictHas test_01037.dict_tuple (0,2) 1 -dictHas test_01037.dict_tuple (0.1,0) 1 -dictHas test_01037.dict_tuple (0.99,2.99) 1 -dictHas test_01037.dict_tuple (1,0) 1 -dictHas test_01037.dict_tuple (3,3) 1 -dictHas test_01037.dict_tuple (5,6) 1 -dictHas test_01037.dict_tuple (7.01,7.01) 0 +dictGet dict_array (-100,-42) qqq 101 +dictGet dict_array (-1,0) Click South 423 +dictGet dict_array (-0.1,0) Click South 423 +dictGet dict_array (0,-2) Click West 424 +dictGet dict_array (0,-1.1) Click West 424 +dictGet dict_array (0,1.1) Click North 422 +dictGet dict_array (0,2) Click North 422 +dictGet dict_array (0.1,0) Click East 421 +dictGet dict_array (0.99,2.99) Click North 422 +dictGet dict_array (1,0) Click East 421 +dictGet dict_array (3,3) House 314159 +dictGet dict_array (5,6) Click 42 +dictGet dict_array (7.01,7.01) qqq 101 +dictGetOrDefault dict_array (-100,-42) www 1234 +dictGetOrDefault dict_array (-1,0) Click South 423 +dictGetOrDefault dict_array (-0.1,0) Click South 423 +dictGetOrDefault dict_array (0,-2) Click West 424 +dictGetOrDefault dict_array (0,-1.1) Click West 424 +dictGetOrDefault dict_array (0,1.1) Click North 422 +dictGetOrDefault dict_array (0,2) Click North 422 +dictGetOrDefault dict_array (0.1,0) Click East 421 +dictGetOrDefault dict_array (0.99,2.99) Click North 422 +dictGetOrDefault dict_array (1,0) Click East 421 +dictGetOrDefault dict_array (3,3) House 314159 +dictGetOrDefault dict_array (5,6) Click 42 +dictGetOrDefault dict_array (7.01,7.01) www 1234 +dictGetOrDefault dict_array (-100,-42) dd 44 +dictGetOrDefault dict_array (-1,0) Click South 423 +dictGetOrDefault dict_array (-0.1,0) Click South 423 +dictGetOrDefault dict_array (0,-2) Click West 424 +dictGetOrDefault dict_array (0,-1.1) Click West 424 +dictGetOrDefault dict_array (0,1.1) Click North 422 +dictGetOrDefault dict_array (0,2) Click North 422 +dictGetOrDefault dict_array (0.1,0) Click East 421 +dictGetOrDefault dict_array (0.99,2.99) Click North 422 +dictGetOrDefault dict_array (1,0) Click East 421 +dictGetOrDefault dict_array (3,3) House 314159 +dictGetOrDefault dict_array (5,6) Click 42 +dictGetOrDefault dict_array (7.01,7.01) ee 55 +dictGet dict_tuple (-100,-42) qqq 101 +dictGet dict_tuple (-1,0) Click South 423 +dictGet dict_tuple (-0.1,0) Click South 423 +dictGet dict_tuple (0,-2) Click West 424 +dictGet dict_tuple (0,-1.1) Click West 424 +dictGet dict_tuple (0,1.1) Click North 422 +dictGet dict_tuple (0,2) Click North 422 +dictGet dict_tuple (0.1,0) Click East 421 +dictGet dict_tuple (0.99,2.99) Click North 422 +dictGet dict_tuple (1,0) Click East 421 +dictGet dict_tuple (3,3) House 314159 +dictGet dict_tuple (5,6) Click 42 +dictGet dict_tuple (7.01,7.01) qqq 101 +dictGetOrDefault dict_tuple (-100,-42) www 1234 +dictGetOrDefault dict_tuple (-1,0) Click South 423 +dictGetOrDefault dict_tuple (-0.1,0) Click South 423 +dictGetOrDefault dict_tuple (0,-2) Click West 424 +dictGetOrDefault dict_tuple (0,-1.1) Click West 424 +dictGetOrDefault dict_tuple (0,1.1) Click North 422 +dictGetOrDefault dict_tuple (0,2) Click North 422 +dictGetOrDefault dict_tuple (0.1,0) Click East 421 +dictGetOrDefault dict_tuple (0.99,2.99) Click North 422 +dictGetOrDefault dict_tuple (1,0) Click East 421 +dictGetOrDefault dict_tuple (3,3) House 314159 +dictGetOrDefault dict_tuple (5,6) Click 42 +dictGetOrDefault dict_tuple (7.01,7.01) www 1234 +dictGetOrDefault dict_tuple (-100,-42) dd 44 +dictGetOrDefault dict_tuple (-1,0) Click South 423 +dictGetOrDefault dict_tuple (-0.1,0) Click South 423 +dictGetOrDefault dict_tuple (0,-2) Click West 424 +dictGetOrDefault dict_tuple (0,-1.1) Click West 424 +dictGetOrDefault dict_tuple (0,1.1) Click North 422 +dictGetOrDefault dict_tuple (0,2) Click North 422 +dictGetOrDefault dict_tuple (0.1,0) Click East 421 +dictGetOrDefault dict_tuple (0.99,2.99) Click North 422 +dictGetOrDefault dict_tuple (1,0) Click East 421 +dictGetOrDefault dict_tuple (3,3) House 314159 +dictGetOrDefault dict_tuple (5,6) Click 42 +dictGetOrDefault dict_tuple (7.01,7.01) ee 55 +dictHas dict_array (-100,-42) 0 +dictHas dict_array (-1,0) 1 +dictHas dict_array (-0.1,0) 1 +dictHas dict_array (0,-2) 1 +dictHas dict_array (0,-1.1) 1 +dictHas dict_array (0,1.1) 1 +dictHas dict_array (0,2) 1 +dictHas dict_array (0.1,0) 1 +dictHas dict_array (0.99,2.99) 1 +dictHas dict_array (1,0) 1 +dictHas dict_array (3,3) 1 +dictHas dict_array (5,6) 1 +dictHas dict_array (7.01,7.01) 0 +dictHas dict_tuple (-100,-42) 0 +dictHas dict_tuple (-1,0) 1 +dictHas dict_tuple (-0.1,0) 1 +dictHas dict_tuple (0,-2) 1 +dictHas dict_tuple (0,-1.1) 1 +dictHas dict_tuple (0,1.1) 1 +dictHas dict_tuple (0,2) 1 +dictHas dict_tuple (0.1,0) 1 +dictHas dict_tuple (0.99,2.99) 1 +dictHas dict_tuple (1,0) 1 +dictHas dict_tuple (3,3) 1 +dictHas dict_tuple (5,6) 1 +dictHas dict_tuple (7.01,7.01) 0 diff --git a/tests/queries/0_stateless/01037_polygon_dicts_simple_functions.sh b/tests/queries/0_stateless/01037_polygon_dicts_simple_functions.sh index be983ec1be4..d1ee3f283bc 100755 --- a/tests/queries/0_stateless/01037_polygon_dicts_simple_functions.sh +++ b/tests/queries/0_stateless/01037_polygon_dicts_simple_functions.sh @@ -1,56 +1,52 @@ #!/usr/bin/env bash -# Tags: no-debug, no-parallel +# Tags: no-debug CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -TMP_DIR="/tmp" +TMP_DIR=${CLICKHOUSE_TMP}/tmp +mkdir -p $TMP_DIR $CLICKHOUSE_CLIENT -n --query=" -DROP DATABASE IF EXISTS test_01037; +DROP TABLE IF EXISTS polygons_array; -CREATE DATABASE test_01037; +CREATE TABLE polygons_array (key Array(Array(Array(Array(Float64)))), name String, value UInt64) ENGINE = Memory; +INSERT INTO polygons_array VALUES ([[[[1, 3], [1, 1], [3, 1], [3, -1], [1, -1], [1, -3], [-1, -3], [-1, -1], [-3, -1], [-3, 1], [-1, 1], [-1, 3]]], [[[5, 5], [5, 1], [7, 1], [7, 7], [1, 7], [1, 5]]]], 'Click', 42); +INSERT INTO polygons_array VALUES ([[[[5, 5], [5, -5], [-5, -5], [-5, 5]], [[1, 3], [1, 1], [3, 1], [3, -1], [1, -1], [1, -3], [-1, -3], [-1, -1], [-3, -1], [-3, 1], [-1, 1], [-1, 3]]]], 'House', 314159); +INSERT INTO polygons_array VALUES ([[[[3, 1], [0, 1], [0, -1], [3, -1]]]], 'Click East', 421); +INSERT INTO polygons_array VALUES ([[[[-1, 1], [1, 1], [1, 3], [-1, 3]]]], 'Click North', 422); +INSERT INTO polygons_array VALUES ([[[[-3, 1], [-3, -1], [0, -1], [0, 1]]]], 'Click South', 423); +INSERT INTO polygons_array VALUES ([[[[-1, -1], [1, -1], [1, -3], [-1, -3]]]], 'Click West', 424); -DROP TABLE IF EXISTS test_01037.polygons_array; +DROP TABLE IF EXISTS polygons_tuple; -CREATE TABLE test_01037.polygons_array (key Array(Array(Array(Array(Float64)))), name String, value UInt64) ENGINE = Memory; -INSERT INTO test_01037.polygons_array VALUES ([[[[1, 3], [1, 1], [3, 1], [3, -1], [1, -1], [1, -3], [-1, -3], [-1, -1], [-3, -1], [-3, 1], [-1, 1], [-1, 3]]], [[[5, 5], [5, 1], [7, 1], [7, 7], [1, 7], [1, 5]]]], 'Click', 42); -INSERT INTO test_01037.polygons_array VALUES ([[[[5, 5], [5, -5], [-5, -5], [-5, 5]], [[1, 3], [1, 1], [3, 1], [3, -1], [1, -1], [1, -3], [-1, -3], [-1, -1], [-3, -1], [-3, 1], [-1, 1], [-1, 3]]]], 'House', 314159); -INSERT INTO test_01037.polygons_array VALUES ([[[[3, 1], [0, 1], [0, -1], [3, -1]]]], 'Click East', 421); -INSERT INTO test_01037.polygons_array VALUES ([[[[-1, 1], [1, 1], [1, 3], [-1, 3]]]], 'Click North', 422); -INSERT INTO test_01037.polygons_array VALUES ([[[[-3, 1], [-3, -1], [0, -1], [0, 1]]]], 'Click South', 423); -INSERT INTO test_01037.polygons_array VALUES ([[[[-1, -1], [1, -1], [1, -3], [-1, -3]]]], 'Click West', 424); +CREATE TABLE polygons_tuple (key Array(Array(Array(Tuple(Float64, Float64)))), name String, value UInt64) ENGINE = Memory; +INSERT INTO polygons_tuple VALUES ([[[(1, 3), (1, 1), (3, 1), (3, -1), (1, -1), (1, -3), (-1, -3), (-1, -1), (-3, -1), (-3, 1), (-1, 1), (-1, 3)]], [[(5, 5), (5, 1), (7, 1), (7, 7), (1, 7), (1, 5)]]], 'Click', 42); +INSERT INTO polygons_tuple VALUES ([[[(5, 5), (5, -5), (-5, -5), (-5, 5)], [(1, 3), (1, 1), (3, 1), (3, -1), (1, -1), (1, -3), (-1, -3), (-1, -1), (-3, -1), (-3, 1), (-1, 1), (-1, 3)]]], 'House', 314159); +INSERT INTO polygons_tuple VALUES ([[[(3, 1), (0, 1), (0, -1), (3, -1)]]], 'Click East', 421); +INSERT INTO polygons_tuple VALUES ([[[(-1, 1), (1, 1), (1, 3), (-1, 3)]]], 'Click North', 422); +INSERT INTO polygons_tuple VALUES ([[[(-3, 1), (-3, -1), (0, -1), (0, 1)]]], 'Click South', 423); +INSERT INTO polygons_tuple VALUES ([[[(-1, -1), (1, -1), (1, -3), (-1, -3)]]], 'Click West', 424); -DROP TABLE IF EXISTS test_01037.polygons_tuple; +DROP TABLE IF EXISTS points; -CREATE TABLE test_01037.polygons_tuple (key Array(Array(Array(Tuple(Float64, Float64)))), name String, value UInt64) ENGINE = Memory; -INSERT INTO test_01037.polygons_tuple VALUES ([[[(1, 3), (1, 1), (3, 1), (3, -1), (1, -1), (1, -3), (-1, -3), (-1, -1), (-3, -1), (-3, 1), (-1, 1), (-1, 3)]], [[(5, 5), (5, 1), (7, 1), (7, 7), (1, 7), (1, 5)]]], 'Click', 42); -INSERT INTO test_01037.polygons_tuple VALUES ([[[(5, 5), (5, -5), (-5, -5), (-5, 5)], [(1, 3), (1, 1), (3, 1), (3, -1), (1, -1), (1, -3), (-1, -3), (-1, -1), (-3, -1), (-3, 1), (-1, 1), (-1, 3)]]], 'House', 314159); -INSERT INTO test_01037.polygons_tuple VALUES ([[[(3, 1), (0, 1), (0, -1), (3, -1)]]], 'Click East', 421); -INSERT INTO test_01037.polygons_tuple VALUES ([[[(-1, 1), (1, 1), (1, 3), (-1, 3)]]], 'Click North', 422); -INSERT INTO test_01037.polygons_tuple VALUES ([[[(-3, 1), (-3, -1), (0, -1), (0, 1)]]], 'Click South', 423); -INSERT INTO test_01037.polygons_tuple VALUES ([[[(-1, -1), (1, -1), (1, -3), (-1, -3)]]], 'Click West', 424); - -DROP TABLE IF EXISTS test_01037.points; - -CREATE TABLE test_01037.points (x Float64, y Float64, def_i UInt64, def_s String) ENGINE = Memory; -INSERT INTO test_01037.points VALUES (0.1, 0.0, 112, 'aax'); -INSERT INTO test_01037.points VALUES (-0.1, 0.0, 113, 'aay'); -INSERT INTO test_01037.points VALUES (0.0, 1.1, 114, 'aaz'); -INSERT INTO test_01037.points VALUES (0.0, -1.1, 115, 'aat'); -INSERT INTO test_01037.points VALUES (3.0, 3.0, 22, 'bb'); -INSERT INTO test_01037.points VALUES (5.0, 6.0, 33, 'cc'); -INSERT INTO test_01037.points VALUES (-100.0, -42.0, 44, 'dd'); -INSERT INTO test_01037.points VALUES (7.01, 7.01, 55, 'ee'); -INSERT INTO test_01037.points VALUES (0.99, 2.99, 66, 'ee'); -INSERT INTO test_01037.points VALUES (1.0, 0.0, 771, 'ffa'); -INSERT INTO test_01037.points VALUES (-1.0, 0.0, 772, 'ffb'); -INSERT INTO test_01037.points VALUES (0.0, 2.0, 773, 'ffc'); -INSERT INTO test_01037.points VALUES (0.0, -2.0, 774, 'ffd'); +CREATE TABLE points (x Float64, y Float64, def_i UInt64, def_s String) ENGINE = Memory; +INSERT INTO points VALUES (0.1, 0.0, 112, 'aax'); +INSERT INTO points VALUES (-0.1, 0.0, 113, 'aay'); +INSERT INTO points VALUES (0.0, 1.1, 114, 'aaz'); +INSERT INTO points VALUES (0.0, -1.1, 115, 'aat'); +INSERT INTO points VALUES (3.0, 3.0, 22, 'bb'); +INSERT INTO points VALUES (5.0, 6.0, 33, 'cc'); +INSERT INTO points VALUES (-100.0, -42.0, 44, 'dd'); +INSERT INTO points VALUES (7.01, 7.01, 55, 'ee'); +INSERT INTO points VALUES (0.99, 2.99, 66, 'ee'); +INSERT INTO points VALUES (1.0, 0.0, 771, 'ffa'); +INSERT INTO points VALUES (-1.0, 0.0, 772, 'ffb'); +INSERT INTO points VALUES (0.0, 2.0, 773, 'ffc'); +INSERT INTO points VALUES (0.0, -2.0, 774, 'ffd'); " - declare -a SearchTypes=("POLYGON" "POLYGON_SIMPLE" "POLYGON_INDEX_EACH" "POLYGON_INDEX_CELL") for type in "${SearchTypes[@]}"; @@ -58,63 +54,62 @@ do outputFile="${TMP_DIR}/results${type}.out" $CLICKHOUSE_CLIENT -n --query=" - DROP DICTIONARY IF EXISTS test_01037.dict_array; - CREATE DICTIONARY test_01037.dict_array + DROP DICTIONARY IF EXISTS dict_array; + CREATE DICTIONARY dict_array ( key Array(Array(Array(Array(Float64)))), name String DEFAULT 'qqq', value UInt64 DEFAULT 101 ) PRIMARY KEY key - SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'polygons_array' PASSWORD '' DB 'test_01037')) + SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'polygons_array' PASSWORD '' DB currentDatabase())) LIFETIME(0) LAYOUT($type()); - DROP DICTIONARY IF EXISTS test_01037.dict_tuple; + DROP DICTIONARY IF EXISTS dict_tuple; - CREATE DICTIONARY test_01037.dict_tuple + CREATE DICTIONARY dict_tuple ( key Array(Array(Array(Tuple(Float64, Float64)))), name String DEFAULT 'qqq', value UInt64 DEFAULT 101 ) PRIMARY KEY key - SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'polygons_tuple' PASSWORD '' DB 'test_01037')) + SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'polygons_tuple' PASSWORD '' DB currentDatabase())) LIFETIME(0) LAYOUT($type()); - select 'dictGet', 'test_01037.dict_array' as dict_name, tuple(x, y) as key, + select 'dictGet', 'dict_array' as dict_name, tuple(x, y) as key, dictGet(dict_name, 'name', key), - dictGet(dict_name, 'value', key) from test_01037.points order by x, y; - select 'dictGetOrDefault', 'test_01037.dict_array' as dict_name, tuple(x, y) as key, + dictGet(dict_name, 'value', key) from points order by x, y; + select 'dictGetOrDefault', 'dict_array' as dict_name, tuple(x, y) as key, dictGetOrDefault(dict_name, 'name', key, 'www'), - dictGetOrDefault(dict_name, 'value', key, toUInt64(1234)) from test_01037.points order by x, y; - select 'dictGetOrDefault', 'test_01037.dict_array' as dict_name, tuple(x, y) as key, + dictGetOrDefault(dict_name, 'value', key, toUInt64(1234)) from points order by x, y; + select 'dictGetOrDefault', 'dict_array' as dict_name, tuple(x, y) as key, dictGetOrDefault(dict_name, 'name', key, def_s), - dictGetOrDefault(dict_name, 'value', key, def_i) from test_01037.points order by x, y; - select 'dictGet', 'test_01037.dict_tuple' as dict_name, tuple(x, y) as key, + dictGetOrDefault(dict_name, 'value', key, def_i) from points order by x, y; + select 'dictGet', 'dict_tuple' as dict_name, tuple(x, y) as key, dictGet(dict_name, 'name', key), - dictGet(dict_name, 'value', key) from test_01037.points order by x, y; - select 'dictGetOrDefault', 'test_01037.dict_tuple' as dict_name, tuple(x, y) as key, + dictGet(dict_name, 'value', key) from points order by x, y; + select 'dictGetOrDefault', 'dict_tuple' as dict_name, tuple(x, y) as key, dictGetOrDefault(dict_name, 'name', key, 'www'), - dictGetOrDefault(dict_name, 'value', key, toUInt64(1234)) from test_01037.points order by x, y; - select 'dictGetOrDefault', 'test_01037.dict_tuple' as dict_name, tuple(x, y) as key, + dictGetOrDefault(dict_name, 'value', key, toUInt64(1234)) from points order by x, y; + select 'dictGetOrDefault', 'dict_tuple' as dict_name, tuple(x, y) as key, dictGetOrDefault(dict_name, 'name', key, def_s), - dictGetOrDefault(dict_name, 'value', key, def_i) from test_01037.points order by x, y; - select 'dictHas', 'test_01037.dict_array' as dict_name, tuple(x, y) as key, - dictHas(dict_name, key) from test_01037.points order by x, y; - select 'dictHas', 'test_01037.dict_tuple' as dict_name, tuple(x, y) as key, - dictHas(dict_name, key) from test_01037.points order by x, y; + dictGetOrDefault(dict_name, 'value', key, def_i) from points order by x, y; + select 'dictHas', 'dict_array' as dict_name, tuple(x, y) as key, + dictHas(dict_name, key) from points order by x, y; + select 'dictHas', 'dict_tuple' as dict_name, tuple(x, y) as key, + dictHas(dict_name, key) from points order by x, y; " > "$outputFile" diff -q "${CURDIR}/01037_polygon_dicts_simple_functions.ans" "$outputFile" done $CLICKHOUSE_CLIENT -n --query=" -DROP DICTIONARY test_01037.dict_array; -DROP DICTIONARY test_01037.dict_tuple; -DROP TABLE test_01037.polygons_array; -DROP TABLE test_01037.polygons_tuple; -DROP TABLE test_01037.points; -DROP DATABASE test_01037; +DROP DICTIONARY dict_array; +DROP DICTIONARY dict_tuple; +DROP TABLE polygons_array; +DROP TABLE polygons_tuple; +DROP TABLE points; " diff --git a/tests/queries/0_stateless/01038_dictionary_lifetime_min_zero_sec.sh b/tests/queries/0_stateless/01038_dictionary_lifetime_min_zero_sec.sh index 1e754dce786..66732205f95 100755 --- a/tests/queries/0_stateless/01038_dictionary_lifetime_min_zero_sec.sh +++ b/tests/queries/0_stateless/01038_dictionary_lifetime_min_zero_sec.sh @@ -5,13 +5,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT --query "DROP DATABASE IF EXISTS db_01038" - -$CLICKHOUSE_CLIENT --query "CREATE DATABASE db_01038" - $CLICKHOUSE_CLIENT --query " -CREATE TABLE db_01038.table_for_dict +CREATE TABLE ${CLICKHOUSE_DATABASE}.table_for_dict ( key_column UInt64, value Float64 @@ -19,34 +15,34 @@ CREATE TABLE db_01038.table_for_dict ENGINE = MergeTree() ORDER BY key_column" -$CLICKHOUSE_CLIENT --query "INSERT INTO db_01038.table_for_dict VALUES (1, 1.1)" +$CLICKHOUSE_CLIENT --query "INSERT INTO ${CLICKHOUSE_DATABASE}.table_for_dict VALUES (1, 1.1)" $CLICKHOUSE_CLIENT --query " -CREATE DICTIONARY db_01038.dict_with_zero_min_lifetime +CREATE DICTIONARY ${CLICKHOUSE_DATABASE}.dict_with_zero_min_lifetime ( key_column UInt64, value Float64 DEFAULT 77.77 ) PRIMARY KEY key_column -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_for_dict' DB 'db_01038')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_for_dict' DB '${CLICKHOUSE_DATABASE}')) LIFETIME(1) LAYOUT(FLAT())" -$CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('db_01038.dict_with_zero_min_lifetime', 'value', toUInt64(1))" +$CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('${CLICKHOUSE_DATABASE}.dict_with_zero_min_lifetime', 'value', toUInt64(1))" -$CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('db_01038.dict_with_zero_min_lifetime', 'value', toUInt64(2))" +$CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('${CLICKHOUSE_DATABASE}.dict_with_zero_min_lifetime', 'value', toUInt64(2))" -$CLICKHOUSE_CLIENT --query "INSERT INTO db_01038.table_for_dict VALUES (2, 2.2)" +$CLICKHOUSE_CLIENT --query "INSERT INTO ${CLICKHOUSE_DATABASE}.table_for_dict VALUES (2, 2.2)" function check() { - query_result=$($CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('db_01038.dict_with_zero_min_lifetime', 'value', toUInt64(2))") + query_result=$($CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('${CLICKHOUSE_DATABASE}.dict_with_zero_min_lifetime', 'value', toUInt64(2))") while [ "$query_result" != "2.2" ] do - query_result=$($CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('db_01038.dict_with_zero_min_lifetime', 'value', toUInt64(2))") + query_result=$($CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('${CLICKHOUSE_DATABASE}.dict_with_zero_min_lifetime', 'value', toUInt64(2))") done } @@ -55,8 +51,6 @@ export -f check; timeout 10 bash -c check -$CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('db_01038.dict_with_zero_min_lifetime', 'value', toUInt64(1))" +$CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('${CLICKHOUSE_DATABASE}.dict_with_zero_min_lifetime', 'value', toUInt64(1))" -$CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('db_01038.dict_with_zero_min_lifetime', 'value', toUInt64(2))" - -$CLICKHOUSE_CLIENT --query "DROP DATABASE IF EXISTS db_01038" +$CLICKHOUSE_CLIENT --query "SELECT dictGetFloat64('${CLICKHOUSE_DATABASE}.dict_with_zero_min_lifetime', 'value', toUInt64(2))" diff --git a/tests/queries/0_stateless/01040_dictionary_invalidate_query_switchover_long.sh b/tests/queries/0_stateless/01040_dictionary_invalidate_query_switchover_long.sh index 6856f952a47..d558fbf465e 100755 --- a/tests/queries/0_stateless/01040_dictionary_invalidate_query_switchover_long.sh +++ b/tests/queries/0_stateless/01040_dictionary_invalidate_query_switchover_long.sh @@ -1,17 +1,12 @@ #!/usr/bin/env bash -# Tags: long, no-parallel +# Tags: long CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh - -$CLICKHOUSE_CLIENT --query "DROP DATABASE IF EXISTS dictdb_01041_01040" - -$CLICKHOUSE_CLIENT --query "CREATE DATABASE dictdb_01041_01040" - $CLICKHOUSE_CLIENT --query " -CREATE TABLE dictdb_01041_01040.dict_invalidate +CREATE TABLE dict_invalidate ENGINE = Memory AS SELECT 122 as dummy, @@ -20,31 +15,31 @@ FROM system.one" $CLICKHOUSE_CLIENT --query " -CREATE DICTIONARY dictdb_01041_01040.invalidate +CREATE DICTIONARY invalidate ( dummy UInt64, two UInt8 EXPRESSION dummy ) PRIMARY KEY dummy -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_invalidate' DB 'dictdb_01041_01040' INVALIDATE_QUERY 'select max(last_time) from dictdb_01041_01040.dict_invalidate')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_invalidate' DB currentDatabase() INVALIDATE_QUERY 'select max(last_time) from dict_invalidate')) LIFETIME(MIN 0 MAX 1) LAYOUT(FLAT())" -$CLICKHOUSE_CLIENT --query "SELECT dictGetUInt8('dictdb_01041_01040.invalidate', 'two', toUInt64(122))" +$CLICKHOUSE_CLIENT --query "SELECT dictGetUInt8('invalidate', 'two', toUInt64(122))" # No exception happened -$CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = 'dictdb_01041_01040' AND name = 'invalidate'" +$CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = currentDatabase() AND name = 'invalidate'" -$CLICKHOUSE_CLIENT --check_table_dependencies=0 --query "DROP TABLE dictdb_01041_01040.dict_invalidate" +$CLICKHOUSE_CLIENT --check_table_dependencies=0 --query "DROP TABLE dict_invalidate" function check_exception_detected() { - query_result=$($CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = 'dictdb_01041_01040' AND name = 'invalidate'" 2>&1) + query_result=$($CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = currentDatabase() AND name = 'invalidate'" 2>&1) while [ -z "$query_result" ] do - query_result=$($CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = 'dictdb_01041_01040' AND name = 'invalidate'" 2>&1) + query_result=$($CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = currentDatabase() AND name = 'invalidate'" 2>&1) sleep 0.1 done } @@ -53,10 +48,10 @@ function check_exception_detected() export -f check_exception_detected; timeout 30 bash -c check_exception_detected 2> /dev/null -$CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = 'dictdb_01041_01040' AND name = 'invalidate'" 2>&1 | grep -Eo "dictdb_01041_01040.dict_invalidate.*UNKNOWN_TABLE" | wc -l +$CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = currentDatabase() AND name = 'invalidate'" 2>&1 | grep -Eo "dict_invalidate.*UNKNOWN_TABLE" | wc -l $CLICKHOUSE_CLIENT --query " -CREATE TABLE dictdb_01041_01040.dict_invalidate +CREATE TABLE dict_invalidate ENGINE = Memory AS SELECT 133 as dummy, @@ -65,11 +60,11 @@ FROM system.one" function check_exception_fixed() { - query_result=$($CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = 'dictdb_01041_01040' AND name = 'invalidate'" 2>&1) + query_result=$($CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = currentDatabase() AND name = 'invalidate'" 2>&1) while [ "$query_result" ] do - query_result=$($CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = 'dictdb_01041_01040' AND name = 'invalidate'" 2>&1) + query_result=$($CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = currentDatabase() AND name = 'invalidate'" 2>&1) sleep 0.1 done } @@ -78,7 +73,5 @@ export -f check_exception_fixed; # it may take a while until dictionary reloads timeout 60 bash -c check_exception_fixed 2> /dev/null -$CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = 'dictdb_01041_01040' AND name = 'invalidate'" 2>&1 -$CLICKHOUSE_CLIENT --query "SELECT dictGetUInt8('dictdb_01041_01040.invalidate', 'two', toUInt64(133))" - -$CLICKHOUSE_CLIENT --query "DROP DATABASE IF EXISTS dictdb_01041_01040" +$CLICKHOUSE_CLIENT --query "SELECT last_exception FROM system.dictionaries WHERE database = currentDatabase() AND name = 'invalidate'" 2>&1 +$CLICKHOUSE_CLIENT --query "SELECT dictGetUInt8('invalidate', 'two', toUInt64(133))" diff --git a/tests/queries/0_stateless/01042_check_query_and_last_granule_size.sql b/tests/queries/0_stateless/01042_check_query_and_last_granule_size.sql index eccb2d25878..7b7d1706346 100644 --- a/tests/queries/0_stateless/01042_check_query_and_last_granule_size.sql +++ b/tests/queries/0_stateless/01042_check_query_and_last_granule_size.sql @@ -1,4 +1,6 @@ +SET optimize_trivial_insert_select = 1; SET check_query_single_value_result = 0; + DROP TABLE IF EXISTS check_query_test; CREATE TABLE check_query_test (SomeKey UInt64, SomeValue String) ENGINE = MergeTree() ORDER BY SomeKey SETTINGS min_bytes_for_wide_part = 0, min_rows_for_wide_part = 0; diff --git a/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.reference b/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.reference index f12dcd8258a..10bc7981d3e 100644 --- a/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.reference +++ b/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.reference @@ -1,6 +1,6 @@ 12 -> 102 13 -> 103 14 -> -1 -12(r) -> 102 -13(r) -> 103 -14(r) -> 104 +12 (after reloading) -> 102 +13 (after reloading) -> 103 +14 (after reloading) -> 104 diff --git a/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh b/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh index 9d34470c38d..42488ca946c 100755 --- a/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh +++ b/tests/queries/0_stateless/01042_system_reload_dictionary_reloads_completely.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: no-parallel +# Tags: no-random-settings +# Dictionaries are updated using the server time. CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -7,42 +8,42 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) set -e -o pipefail -# NOTE: dictionaries TTLs works with server timezone, so session_timeout cannot be used -$CLICKHOUSE_CLIENT --session_timezone '' --multiquery <<'EOF' -DROP DATABASE IF EXISTS dictdb_01042; -CREATE DATABASE dictdb_01042; -CREATE TABLE dictdb_01042.table(x Int64, y Int64, insert_time DateTime) ENGINE = MergeTree ORDER BY tuple(); -INSERT INTO dictdb_01042.table VALUES (12, 102, now()); +$CLICKHOUSE_CLIENT --multiquery < ', dictGetInt64('dictdb_01042.dict', 'y', toUInt64(12))" +$CLICKHOUSE_CLIENT --query "SELECT '12 -> ', dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(12))" -$CLICKHOUSE_CLIENT --query "INSERT INTO dictdb_01042.table VALUES (13, 103, now())" -$CLICKHOUSE_CLIENT --query "INSERT INTO dictdb_01042.table VALUES (14, 104, now() - INTERVAL 1 DAY)" +$CLICKHOUSE_CLIENT --query "INSERT INTO ${CLICKHOUSE_DATABASE}.table VALUES (13, 103, now())" +$CLICKHOUSE_CLIENT --query "INSERT INTO ${CLICKHOUSE_DATABASE}.table VALUES (14, 104, now() - INTERVAL 1 DAY)" -while [ "$(${CLICKHOUSE_CLIENT} --query "SELECT dictGetInt64('dictdb_01042.dict', 'y', toUInt64(13))")" = -1 ] - do - sleep 0.5 - done +# Wait when the dictionary will update the value for 13 on its own: +while [ "$(${CLICKHOUSE_CLIENT} --query "SELECT dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(13))")" = -1 ] +do + sleep 0.5 +done -$CLICKHOUSE_CLIENT --query "SELECT '13 -> ', dictGetInt64('dictdb_01042.dict', 'y', toUInt64(13))" -$CLICKHOUSE_CLIENT --query "SELECT '14 -> ', dictGetInt64('dictdb_01042.dict', 'y', toUInt64(14))" +$CLICKHOUSE_CLIENT --query "SELECT '13 -> ', dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(13))" -$CLICKHOUSE_CLIENT --query "SYSTEM RELOAD DICTIONARY 'dictdb_01042.dict'" +# By the way, the value for 14 is expected to not be updated at this moment, +# because the values were selected by the update field insert_time, and for 14 it was set as one day ago. +$CLICKHOUSE_CLIENT --query "SELECT '14 -> ', dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(14))" -$CLICKHOUSE_CLIENT --query "SELECT '12(r) -> ', dictGetInt64('dictdb_01042.dict', 'y', toUInt64(12))" -$CLICKHOUSE_CLIENT --query "SELECT '13(r) -> ', dictGetInt64('dictdb_01042.dict', 'y', toUInt64(13))" -$CLICKHOUSE_CLIENT --query "SELECT '14(r) -> ', dictGetInt64('dictdb_01042.dict', 'y', toUInt64(14))" +# SYSTEM RELOAD DICTIONARY reloads it completely, regardless of the update field, so we will see new values, even for key 14. +$CLICKHOUSE_CLIENT --query "SYSTEM RELOAD DICTIONARY '${CLICKHOUSE_DATABASE}.dict'" -$CLICKHOUSE_CLIENT --query "DROP DATABASE IF EXISTS dictdb_01042" +$CLICKHOUSE_CLIENT --query "SELECT '12 (after reloading) -> ', dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(12))" +$CLICKHOUSE_CLIENT --query "SELECT '13 (after reloading) -> ', dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(13))" +$CLICKHOUSE_CLIENT --query "SELECT '14 (after reloading) -> ', dictGetInt64('${CLICKHOUSE_DATABASE}.dict', 'y', toUInt64(14))" diff --git a/tests/queries/0_stateless/01045_order_by_pk_special_storages.sh b/tests/queries/0_stateless/01045_order_by_pk_special_storages.sh index 12421a2b308..0714b4c91ed 100755 --- a/tests/queries/0_stateless/01045_order_by_pk_special_storages.sh +++ b/tests/queries/0_stateless/01045_order_by_pk_special_storages.sh @@ -16,8 +16,8 @@ $CLICKHOUSE_CLIENT -q "CREATE TABLE s1 (a UInt32, s String) ENGINE = MergeTree O $CLICKHOUSE_CLIENT -q "CREATE TABLE s2 (a UInt32, s String) ENGINE = MergeTree ORDER BY a PARTITION BY a % 3 SETTINGS min_bytes_for_wide_part = 0, min_rows_for_wide_part = 0" $CLICKHOUSE_CLIENT -q "CREATE TABLE m (a UInt32, s String) engine = Merge('$CLICKHOUSE_DATABASE', 's[1,2]')" -$CLICKHOUSE_CLIENT -q "INSERT INTO s1 select (number % 20) * 2 as n, toString(number * number) from numbers(100000)" -$CLICKHOUSE_CLIENT -q "INSERT INTO s2 select (number % 20) * 2 + 1 as n, toString(number * number * number) from numbers(100000)" +$CLICKHOUSE_CLIENT --optimize_trivial_insert_select 1 -q "INSERT INTO s1 select (number % 20) * 2 as n, toString(number * number) from numbers(100000)" +$CLICKHOUSE_CLIENT --optimize_trivial_insert_select 1 -q "INSERT INTO s2 select (number % 20) * 2 + 1 as n, toString(number * number * number) from numbers(100000)" $CLICKHOUSE_CLIENT -q "SELECT '---StorageMerge---'" $CLICKHOUSE_CLIENT -q "SELECT a FROM m ORDER BY a LIMIT 5" diff --git a/tests/queries/0_stateless/01045_zookeeper_system_mutations_with_parts_names.sh b/tests/queries/0_stateless/01045_zookeeper_system_mutations_with_parts_names.sh index cd6501bbebf..1185498a5f7 100755 --- a/tests/queries/0_stateless/01045_zookeeper_system_mutations_with_parts_names.sh +++ b/tests/queries/0_stateless/01045_zookeeper_system_mutations_with_parts_names.sh @@ -25,7 +25,7 @@ ${CLICKHOUSE_CLIENT} --query="CREATE TABLE table_for_mutations(k UInt32, v1 UInt ${CLICKHOUSE_CLIENT} --query="SYSTEM STOP MERGES table_for_mutations" -${CLICKHOUSE_CLIENT} --query="INSERT INTO table_for_mutations select number, number from numbers(100000)" +${CLICKHOUSE_CLIENT} --optimize_trivial_insert_select 1 --query="INSERT INTO table_for_mutations select number, number from numbers(100000)" ${CLICKHOUSE_CLIENT} --query="SELECT sum(v1) FROM table_for_mutations" @@ -53,7 +53,7 @@ ${CLICKHOUSE_CLIENT} --query="CREATE TABLE replicated_table_for_mutations(k UInt ${CLICKHOUSE_CLIENT} --query="SYSTEM STOP MERGES replicated_table_for_mutations" # test relays on part ids, which are non-deterministic with keeper fault injections, so disable it -${CLICKHOUSE_CLIENT} --insert_keeper_fault_injection_probability=0 --query="INSERT INTO replicated_table_for_mutations select number, number from numbers(100000)" +${CLICKHOUSE_CLIENT} --optimize_trivial_insert_select 1 --insert_keeper_fault_injection_probability=0 --query="INSERT INTO replicated_table_for_mutations select number, number from numbers(100000)" ${CLICKHOUSE_CLIENT} --query="SELECT sum(v1) FROM replicated_table_for_mutations" diff --git a/tests/queries/0_stateless/01052_window_view_proc_tumble_to_now.sh b/tests/queries/0_stateless/01052_window_view_proc_tumble_to_now.sh index 4325ebeed24..5c70806ea7b 100755 --- a/tests/queries/0_stateless/01052_window_view_proc_tumble_to_now.sh +++ b/tests/queries/0_stateless/01052_window_view_proc_tumble_to_now.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-random-settings, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01053_ssd_dictionary.sh b/tests/queries/0_stateless/01053_ssd_dictionary.sh index b49144c9b1a..fdd40940ab5 100755 --- a/tests/queries/0_stateless/01053_ssd_dictionary.sh +++ b/tests/queries/0_stateless/01053_ssd_dictionary.sh @@ -1,17 +1,15 @@ #!/usr/bin/env bash # Tags: no-parallel, no-fasttest -# Tag no-fasttest: this test mistakenly requires acces to /var/lib/clickhouse -- can't run this locally, disabled +# Tag no-fasttest: this test mistakenly requires access to /var/lib/clickhouse -- can't run this locally, disabled CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT --allow_deprecated_database_ordinary=1 -n --query=" DROP DATABASE IF EXISTS 01053_db; - CREATE DATABASE 01053_db Engine = Ordinary; + CREATE DATABASE 01053_db; DROP TABLE IF EXISTS 01053_db.table_for_dict; diff --git a/tests/queries/0_stateless/01053_window_view_proc_hop_to_now.sh b/tests/queries/0_stateless/01053_window_view_proc_hop_to_now.sh index 8e28995980f..32c9c52ab09 100755 --- a/tests/queries/0_stateless/01053_window_view_proc_hop_to_now.sh +++ b/tests/queries/0_stateless/01053_window_view_proc_hop_to_now.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-random-settings, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01054_window_view_proc_tumble_to.sh b/tests/queries/0_stateless/01054_window_view_proc_tumble_to.sh index ee11b265ecd..ba566bb4ae6 100755 --- a/tests/queries/0_stateless/01054_window_view_proc_tumble_to.sh +++ b/tests/queries/0_stateless/01054_window_view_proc_tumble_to.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-random-settings, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01055_compact_parts_1.sql b/tests/queries/0_stateless/01055_compact_parts_1.sql index ff5ab722e0f..72048c59a41 100644 --- a/tests/queries/0_stateless/01055_compact_parts_1.sql +++ b/tests/queries/0_stateless/01055_compact_parts_1.sql @@ -1,8 +1,3 @@ --- Tags: no-parallel - -drop table if exists mt_compact; -drop table if exists mt_compact_2; - create table mt_compact (a Int, s String) engine = MergeTree order by a partition by a settings index_granularity_bytes = 0; alter table mt_compact modify setting min_rows_for_wide_part = 1000; -- { serverError NOT_IMPLEMENTED } @@ -25,5 +20,3 @@ alter table mt_compact modify setting parts_to_delay_insert = 300; alter table mt_compact modify setting min_rows_for_wide_part = 0; show create table mt_compact; - -drop table mt_compact diff --git a/tests/queries/0_stateless/01055_window_view_proc_hop_to.sh b/tests/queries/0_stateless/01055_window_view_proc_hop_to.sh index ea8ad372617..0db4173b3dc 100755 --- a/tests/queries/0_stateless/01055_window_view_proc_hop_to.sh +++ b/tests/queries/0_stateless/01055_window_view_proc_hop_to.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-random-settings, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01060_avro.sh b/tests/queries/0_stateless/01060_avro.sh index 3c70927db25..6ed26c8565f 100755 --- a/tests/queries/0_stateless/01060_avro.sh +++ b/tests/queries/0_stateless/01060_avro.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest set -e @@ -69,9 +69,6 @@ cat "$DATA_DIR"/simple.null.avro | ${CLICKHOUSE_LOCAL} --input-format Avro --out - - - # output echo '===' output diff --git a/tests/queries/0_stateless/01069_database_memory.sql b/tests/queries/0_stateless/01069_database_memory.sql index 5aab9175c58..5d2fa4ea11e 100644 --- a/tests/queries/0_stateless/01069_database_memory.sql +++ b/tests/queries/0_stateless/01069_database_memory.sql @@ -1,5 +1,3 @@ --- Tags: no-parallel - DROP DATABASE IF EXISTS memory_01069; CREATE DATABASE memory_01069 ENGINE = Memory; SHOW CREATE DATABASE memory_01069; diff --git a/tests/queries/0_stateless/01070_mutations_with_dependencies.sql b/tests/queries/0_stateless/01070_mutations_with_dependencies.sql index 813ebf3f5a7..4d1cd54306c 100644 --- a/tests/queries/0_stateless/01070_mutations_with_dependencies.sql +++ b/tests/queries/0_stateless/01070_mutations_with_dependencies.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-s3-storage +-- Tags: no-parallel, no-object-storage -- With s3 policy TTL TO DISK 'default' doesn't work (because we have no default, only 's3') drop table if exists ttl; diff --git a/tests/queries/0_stateless/01075_window_view_proc_tumble_to_now_populate.sh b/tests/queries/0_stateless/01075_window_view_proc_tumble_to_now_populate.sh index f7842af4dad..67c249a9d0e 100755 --- a/tests/queries/0_stateless/01075_window_view_proc_tumble_to_now_populate.sh +++ b/tests/queries/0_stateless/01075_window_view_proc_tumble_to_now_populate.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-random-settings, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01076_cache_dictionary_datarace_exception_ptr.sh b/tests/queries/0_stateless/01076_cache_dictionary_datarace_exception_ptr.sh index 17068dcbdf9..e003d2a26da 100755 --- a/tests/queries/0_stateless/01076_cache_dictionary_datarace_exception_ptr.sh +++ b/tests/queries/0_stateless/01076_cache_dictionary_datarace_exception_ptr.sh @@ -7,11 +7,8 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT --query="DROP DATABASE IF EXISTS dictdb_01076;" -$CLICKHOUSE_CLIENT --query="CREATE DATABASE dictdb_01076;" - $CLICKHOUSE_CLIENT --query=" -CREATE TABLE dictdb_01076.table_datarace +CREATE TABLE ${CLICKHOUSE_DATABASE}.table_datarace ( key_column UUID, value Float64 @@ -21,17 +18,17 @@ ORDER BY key_column; " $CLICKHOUSE_CLIENT --query=" -INSERT INTO dictdb_01076.table_datarace VALUES ('cd5db34f-0c25-4375-b10e-bfb3708ddc72', 1.1), ('cd5db34f-0c25-4375-b10e-bfb3708ddc72', 2.2), ('cd5db34f-0c25-4375-b10e-bfb3708ddc72', 3.3); +INSERT INTO ${CLICKHOUSE_DATABASE}.table_datarace VALUES ('cd5db34f-0c25-4375-b10e-bfb3708ddc72', 1.1), ('cd5db34f-0c25-4375-b10e-bfb3708ddc72', 2.2), ('cd5db34f-0c25-4375-b10e-bfb3708ddc72', 3.3); " $CLICKHOUSE_CLIENT --query=" -CREATE DICTIONARY IF NOT EXISTS dictdb_01076.dict_datarace +CREATE DICTIONARY IF NOT EXISTS ${CLICKHOUSE_DATABASE}.dict_datarace ( key_column UInt64, value Float64 DEFAULT 77.77 ) PRIMARY KEY key_column -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_datarace' DB 'dictdb_01076')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_datarace' DB '${CLICKHOUSE_DATABASE}')) LIFETIME(1) LAYOUT(CACHE(SIZE_IN_CELLS 10)); " @@ -41,7 +38,7 @@ function thread1() for _ in {1..50} do # This query will be ended with exception, because source dictionary has UUID as a key type. - $CLICKHOUSE_CLIENT --query="SELECT dictGetFloat64('dictdb_01076.dict_datarace', 'value', toUInt64(1));" + $CLICKHOUSE_CLIENT --query="SELECT dictGetFloat64('${CLICKHOUSE_DATABASE}.dict_datarace', 'value', toUInt64(1));" done } @@ -51,7 +48,7 @@ function thread2() for _ in {1..50} do # This query will be ended with exception, because source dictionary has UUID as a key type. - $CLICKHOUSE_CLIENT --query="SELECT dictGetFloat64('dictdb_01076.dict_datarace', 'value', toUInt64(2));" + $CLICKHOUSE_CLIENT --query="SELECT dictGetFloat64('${CLICKHOUSE_DATABASE}.dict_datarace', 'value', toUInt64(2));" done } @@ -67,6 +64,5 @@ wait echo OK -$CLICKHOUSE_CLIENT --query="DROP DICTIONARY dictdb_01076.dict_datarace;" -$CLICKHOUSE_CLIENT --query="DROP TABLE dictdb_01076.table_datarace;" -$CLICKHOUSE_CLIENT --query="DROP DATABASE dictdb_01076;" +$CLICKHOUSE_CLIENT --query="DROP DICTIONARY ${CLICKHOUSE_DATABASE}.dict_datarace;" +$CLICKHOUSE_CLIENT --query="DROP TABLE ${CLICKHOUSE_DATABASE}.table_datarace;" diff --git a/tests/queries/0_stateless/01078_merge_tree_read_one_thread.sql b/tests/queries/0_stateless/01078_merge_tree_read_one_thread.sql index 3a05e4507a2..166f44df2a7 100644 --- a/tests/queries/0_stateless/01078_merge_tree_read_one_thread.sql +++ b/tests/queries/0_stateless/01078_merge_tree_read_one_thread.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage -- Output slightly different plan drop table if exists t; diff --git a/tests/queries/0_stateless/01079_parallel_alter_add_drop_column_zookeeper.sh b/tests/queries/0_stateless/01079_parallel_alter_add_drop_column_zookeeper.sh index bfdea95fa9e..f05a0fed965 100755 --- a/tests/queries/0_stateless/01079_parallel_alter_add_drop_column_zookeeper.sh +++ b/tests/queries/0_stateless/01079_parallel_alter_add_drop_column_zookeeper.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: zookeeper, no-parallel, no-fasttest +# Tags: zookeeper, no-fasttest, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -63,7 +63,7 @@ export -f optimize_thread; export -f insert_thread; -TIMEOUT=30 +TIMEOUT=20 # Sometimes we detach and attach tables timeout $TIMEOUT bash -c alter_thread 2> /dev/null & diff --git a/tests/queries/0_stateless/01079_parallel_alter_modify_zookeeper_long.sh b/tests/queries/0_stateless/01079_parallel_alter_modify_zookeeper_long.sh index ba8d89aad3c..382e3279711 100755 --- a/tests/queries/0_stateless/01079_parallel_alter_modify_zookeeper_long.sh +++ b/tests/queries/0_stateless/01079_parallel_alter_modify_zookeeper_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, zookeeper, no-parallel, no-fasttest +# Tags: long, zookeeper, no-fasttest, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -76,7 +76,7 @@ export -f insert_thread; export -f select_thread; -TIMEOUT=30 +TIMEOUT=20 # Selects should run successfully diff --git a/tests/queries/0_stateless/01082_bit_test_out_of_bound.reference b/tests/queries/0_stateless/01082_bit_test_out_of_bound.reference index 708c5d9d994..26085389381 100644 --- a/tests/queries/0_stateless/01082_bit_test_out_of_bound.reference +++ b/tests/queries/0_stateless/01082_bit_test_out_of_bound.reference @@ -1,3 +1,22 @@ +-- bitTestAny +0 1 +1 0 +2 1 +3 0 +4 1 +5 0 +6 1 +7 0 +-- bitTestAll +0 1 +1 0 +2 1 +3 0 +4 1 +5 0 +6 1 +7 0 +-- bitTest 0 1 1 0 2 1 @@ -6,98 +25,6 @@ 5 0 6 1 7 0 -8 0 -9 0 -10 0 -11 0 -12 0 -13 0 -14 0 -15 0 -16 0 -17 0 -18 0 -19 0 -20 0 -21 0 -22 0 -23 0 -24 0 -25 0 -26 0 -27 0 -28 0 -29 0 -30 0 -31 0 -32 0 -33 0 -34 0 -35 0 -36 0 -37 0 -38 0 -39 0 -40 0 -41 0 -42 0 -43 0 -44 0 -45 0 -46 0 -47 0 -48 0 -49 0 -50 0 -51 0 -52 0 -53 0 -54 0 -55 0 -56 0 -57 0 -58 0 -59 0 -60 0 -61 0 -62 0 -63 0 -64 0 -65 0 -66 0 -67 0 -68 0 -69 0 -70 0 -71 0 -72 0 -73 0 -74 0 -75 0 -76 0 -77 0 -78 0 -79 0 -80 0 -81 0 -82 0 -83 0 -84 0 -85 0 -86 0 -87 0 -88 0 -89 0 -90 0 -91 0 -92 0 -93 0 -94 0 -95 0 -96 0 -97 0 -98 0 -99 0 0 1 1 0 2 1 @@ -107,94 +34,10 @@ 6 1 7 0 8 1 -9 1 +9 0 10 1 -11 1 +11 0 12 1 -13 1 +13 0 14 1 -15 1 -16 1 -17 1 -18 1 -19 1 -20 1 -21 1 -22 1 -23 1 -24 1 -25 1 -26 1 -27 1 -28 1 -29 1 -30 1 -31 1 -32 1 -33 1 -34 1 -35 1 -36 1 -37 1 -38 1 -39 1 -40 1 -41 1 -42 1 -43 1 -44 1 -45 1 -46 1 -47 1 -48 1 -49 1 -50 1 -51 1 -52 1 -53 1 -54 1 -55 1 -56 1 -57 1 -58 1 -59 1 -60 1 -61 1 -62 1 -63 1 -64 1 -65 1 -66 1 -67 1 -68 1 -69 1 -70 1 -71 1 -72 1 -73 1 -74 1 -75 1 -76 1 -77 1 -78 1 -79 1 -80 1 -81 1 -82 1 -83 1 -84 1 -85 1 -86 1 -87 1 -88 1 -89 1 -90 1 -91 1 -92 1 -93 1 -94 1 -95 1 -96 1 -97 1 -98 1 -99 1 +15 0 diff --git a/tests/queries/0_stateless/01082_bit_test_out_of_bound.sql b/tests/queries/0_stateless/01082_bit_test_out_of_bound.sql index 82e2c5a2380..e741cb249d0 100644 --- a/tests/queries/0_stateless/01082_bit_test_out_of_bound.sql +++ b/tests/queries/0_stateless/01082_bit_test_out_of_bound.sql @@ -1,2 +1,13 @@ -SELECT number, bitTestAny(toUInt8(1 + 4 + 16 + 64), number) FROM numbers(100); -SELECT number, bitTestAll(toUInt8(1 + 4 + 16 + 64), number) FROM numbers(100); +SELECT '-- bitTestAny'; +SELECT number, bitTestAny(toUInt8(1 + 4 + 16 + 64), number) FROM numbers(8); +SELECT number, bitTestAny(toUInt8(1 + 4 + 16 + 64), number) FROM numbers(8, 16); -- { serverError PARAMETER_OUT_OF_BOUND } + +SELECT '-- bitTestAll'; +SELECT number, bitTestAll(toUInt8(1 + 4 + 16 + 64), number) FROM numbers(8); +SELECT number, bitTestAll(toUInt8(1 + 4 + 16 + 64), number) FROM numbers(8, 16); -- { serverError PARAMETER_OUT_OF_BOUND } + +SELECT '-- bitTest'; +SELECT number, bitTest(toUInt8(1 + 4 + 16 + 64), number) FROM numbers(8); +SELECT number, bitTest(toUInt8(1 + 4 + 16 + 64), number) FROM numbers(8, 16); -- { serverError PARAMETER_OUT_OF_BOUND } +SELECT number, bitTest(toUInt16(1 + 4 + 16 + 64 + 256 + 1024 + 4096 + 16384 + 65536), number) FROM numbers(16); +SELECT -number, bitTest(toUInt16(1), -number) FROM numbers(8); -- { serverError PARAMETER_OUT_OF_BOUND } diff --git a/tests/queries/0_stateless/01083_expressions_in_engine_arguments.sql b/tests/queries/0_stateless/01083_expressions_in_engine_arguments.sql index 6268765aa27..697843be27f 100644 --- a/tests/queries/0_stateless/01083_expressions_in_engine_arguments.sql +++ b/tests/queries/0_stateless/01083_expressions_in_engine_arguments.sql @@ -28,7 +28,7 @@ CREATE TABLE url (n UInt64, col String) ENGINE=URL ( replace ( - 'https://localhost:8443/?query=' || 'select n, _table from ' || currentDatabase() || '.merge format CSV', ' ', '+' + 'https://localhost:' || getServerPort('https_port') || '/?query=' || 'select n, _table from ' || currentDatabase() || '.merge format CSV', ' ', '+' ), CSV ); @@ -39,7 +39,7 @@ CREATE VIEW view AS SELECT toInt64(n) as n FROM (SELECT toString(n) as n from me SELECT nonexistentsomething; -- { serverError UNKNOWN_IDENTIFIER } CREATE DICTIONARY dict (n UInt64, col String DEFAULT '42') PRIMARY KEY n -SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9440 SECURE 1 USER 'default' TABLE 'url')) LIFETIME(1) LAYOUT(CACHE(SIZE_IN_CELLS 1)); +SOURCE(CLICKHOUSE(HOST 'localhost' PORT getServerPort('tcp_port_secure') SECURE 1 USER 'default' TABLE 'url')) LIFETIME(1) LAYOUT(CACHE(SIZE_IN_CELLS 1)); -- dict --> url --> merge |-> distributed -> file (1) -- |-> distributed_tf -> buffer -> file (1) @@ -88,6 +88,7 @@ SELECT sum(n) from rich_syntax; SYSTEM DROP DNS CACHE; DROP TABLE file; +DROP DICTIONARY dict; DROP TABLE url; DROP TABLE view; DROP TABLE buffer; @@ -96,4 +97,3 @@ DROP TABLE merge_tf; DROP TABLE distributed; DROP TABLE distributed_tf; DROP TABLE rich_syntax; -DROP DICTIONARY dict; diff --git a/tests/queries/0_stateless/01098_msgpack_format.sh b/tests/queries/0_stateless/01098_msgpack_format.sh index e2ae026eb27..30956ac6c7f 100755 --- a/tests/queries/0_stateless/01098_msgpack_format.sh +++ b/tests/queries/0_stateless/01098_msgpack_format.sh @@ -1,14 +1,11 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel +# Tags: no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - - $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS msgpack"; $CLICKHOUSE_CLIENT --query="CREATE TABLE msgpack (uint8 UInt8, uint16 UInt16, uint32 UInt32, uint64 UInt64, int8 Int8, int16 Int16, int32 Int32, int64 Int64, float Float32, double Float64, string String, date Date, datetime DateTime('Asia/Istanbul'), datetime64 DateTime64(3, 'Asia/Istanbul'), array Array(UInt32)) ENGINE = Memory"; diff --git a/tests/queries/0_stateless/01103_check_cpu_instructions_at_startup.reference b/tests/queries/0_stateless/01103_check_cpu_instructions_at_startup.reference index 8984d35930a..03ed07cf1a4 100644 --- a/tests/queries/0_stateless/01103_check_cpu_instructions_at_startup.reference +++ b/tests/queries/0_stateless/01103_check_cpu_instructions_at_startup.reference @@ -2,6 +2,4 @@ Instruction check fail. The CPU does not support SSSE3 instruction set. Instruction check fail. The CPU does not support SSE4.1 instruction set. Instruction check fail. The CPU does not support SSE4.2 instruction set. Instruction check fail. The CPU does not support POPCNT instruction set. -: MADV_DONTNEED does not work (memset will be used instead) -: (This is the expected behaviour if you are running under QEMU) 1 diff --git a/tests/queries/0_stateless/01103_check_cpu_instructions_at_startup.sh b/tests/queries/0_stateless/01103_check_cpu_instructions_at_startup.sh index 01047aeb9ab..c37f1f95374 100755 --- a/tests/queries/0_stateless/01103_check_cpu_instructions_at_startup.sh +++ b/tests/queries/0_stateless/01103_check_cpu_instructions_at_startup.sh @@ -1,6 +1,6 @@ #!/usr/bin/env bash # Tags: no-tsan, no-asan, no-ubsan, no-msan, no-debug, no-fasttest, no-cpu-aarch64 -# Tag no-fasttest: avoid dependency on qemu -- invonvenient when running locally +# Tag no-fasttest: avoid dependency on qemu -- inconvenient when running locally CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01111_create_drop_replicated_db_stress.sh b/tests/queries/0_stateless/01111_create_drop_replicated_db_stress.sh index 91f7a276ea3..619b6e91d11 100755 --- a/tests/queries/0_stateless/01111_create_drop_replicated_db_stress.sh +++ b/tests/queries/0_stateless/01111_create_drop_replicated_db_stress.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: race, zookeeper +# Tags: race, zookeeper, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -87,7 +87,7 @@ function insert() -TIMEOUT=30 +TIMEOUT=20 create_db $TIMEOUT & sync_db $TIMEOUT & diff --git a/tests/queries/0_stateless/01113_local_dictionary_type_conversion.sql b/tests/queries/0_stateless/01113_local_dictionary_type_conversion.sql index 65a03993295..1dc727930ab 100644 --- a/tests/queries/0_stateless/01113_local_dictionary_type_conversion.sql +++ b/tests/queries/0_stateless/01113_local_dictionary_type_conversion.sql @@ -1,29 +1,21 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS database_for_dict; - -CREATE DATABASE database_for_dict; - -CREATE TABLE database_for_dict.table_for_dict ( +CREATE TABLE table_for_dict ( CompanyID String, OSType Enum('UNKNOWN' = 0, 'WINDOWS' = 1, 'LINUX' = 2, 'ANDROID' = 3, 'MAC' = 4), SomeID Int32 ) ENGINE = Memory(); -INSERT INTO database_for_dict.table_for_dict VALUES ('First', 'WINDOWS', 1), ('Second', 'LINUX', 2); +INSERT INTO table_for_dict VALUES ('First', 'WINDOWS', 1), ('Second', 'LINUX', 2); -CREATE DICTIONARY database_for_dict.dict_with_conversion +CREATE DICTIONARY dict_with_conversion ( CompanyID String DEFAULT '', OSType String DEFAULT '', SomeID Int32 DEFAULT 0 ) PRIMARY KEY CompanyID -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_for_dict' DB 'database_for_dict')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_for_dict' DB currentDatabase())) LIFETIME(MIN 1 MAX 20) LAYOUT(COMPLEX_KEY_HASHED()); -SELECT * FROM database_for_dict.dict_with_conversion ORDER BY CompanyID; - -DROP DATABASE IF EXISTS database_for_dict; +SELECT * FROM dict_with_conversion ORDER BY CompanyID; diff --git a/tests/queries/0_stateless/01114_database_atomic.sh b/tests/queries/0_stateless/01114_database_atomic.sh index 1b1f064ae0b..fed76727a27 100755 --- a/tests/queries/0_stateless/01114_database_atomic.sh +++ b/tests/queries/0_stateless/01114_database_atomic.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest # Tag no-fasttest: 45 seconds running # Creation of a database with Ordinary engine emits a warning. diff --git a/tests/queries/0_stateless/01114_mysql_database_engine_segfault.sql b/tests/queries/0_stateless/01114_mysql_database_engine_segfault.sql index 3379acf4d7b..783a728e336 100644 --- a/tests/queries/0_stateless/01114_mysql_database_engine_segfault.sql +++ b/tests/queries/0_stateless/01114_mysql_database_engine_segfault.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-fasttest +-- Tags: no-fasttest DROP DATABASE IF EXISTS conv_main; CREATE DATABASE conv_main ENGINE = MySQL('127.0.0.1:3456', conv_main, 'metrika', 'password'); -- { serverError CANNOT_CREATE_DATABASE } diff --git a/tests/queries/0_stateless/01116_asof_join_dolbyzerr.reference b/tests/queries/0_stateless/01116_asof_join_dolbyzerr.reference index 1055a67ea5b..0aa1a85f19d 100644 --- a/tests/queries/0_stateless/01116_asof_join_dolbyzerr.reference +++ b/tests/queries/0_stateless/01116_asof_join_dolbyzerr.reference @@ -1,3 +1,6 @@ v1 o1 ['s2','s1'] v1 o2 ['s4'] v2 o3 ['s5','s3'] +v1 o1 ['s2','s1'] +v1 o2 ['s4'] +v2 o3 ['s5','s3'] diff --git a/tests/queries/0_stateless/01116_asof_join_dolbyzerr.sql b/tests/queries/0_stateless/01116_asof_join_dolbyzerr.sql index 8a94b6ddd24..652cb35cf2a 100644 --- a/tests/queries/0_stateless/01116_asof_join_dolbyzerr.sql +++ b/tests/queries/0_stateless/01116_asof_join_dolbyzerr.sql @@ -16,3 +16,17 @@ GROUP BY ORDER BY visitorId ASC, orderId ASC; + +SELECT + visitorId, + orderId, + groupUniqArray(sessionId) +FROM sessions +ASOF INNER JOIN orders ON (sessions.visitorId = orders.visitorId) AND (sessions.date <= orders.date) +GROUP BY + visitorId, + orderId +ORDER BY + visitorId ASC, + orderId ASC +SETTINGS join_algorithm = 'full_sorting_merge'; diff --git a/tests/queries/0_stateless/01125_dict_ddl_cannot_add_column.sql b/tests/queries/0_stateless/01125_dict_ddl_cannot_add_column.sql index a324d278c12..6a818d94a58 100644 --- a/tests/queries/0_stateless/01125_dict_ddl_cannot_add_column.sql +++ b/tests/queries/0_stateless/01125_dict_ddl_cannot_add_column.sql @@ -1,11 +1,3 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS database_for_dict; - -CREATE DATABASE database_for_dict; - -use database_for_dict; - CREATE TABLE date_table ( id UInt32, @@ -24,7 +16,7 @@ CREATE DICTIONARY somedict end Date ) PRIMARY KEY id -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'date_table' DB 'database_for_dict')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'date_table' DB currentDatabase())) LAYOUT(RANGE_HASHED()) RANGE (MIN start MAX end) LIFETIME(MIN 300 MAX 360); @@ -35,5 +27,3 @@ SELECT * from somedict; SELECT 1 FROM somedict; SHOW TABLES; - -DROP DATABASE database_for_dict; diff --git a/tests/queries/0_stateless/01127_month_partitioning_consistency_select.sql b/tests/queries/0_stateless/01127_month_partitioning_consistency_select.sql index 2a1d04e6074..78632ab2463 100644 --- a/tests/queries/0_stateless/01127_month_partitioning_consistency_select.sql +++ b/tests/queries/0_stateless/01127_month_partitioning_consistency_select.sql @@ -1,6 +1,3 @@ --- Tags: no-parallel - -DROP TABLE IF EXISTS mt; set allow_deprecated_syntax_for_merge_tree=1; CREATE TABLE mt (d Date, x String) ENGINE = MergeTree(d, x, 8192); INSERT INTO mt VALUES ('2106-02-07', 'Hello'), ('1970-01-01', 'World'); diff --git a/tests/queries/0_stateless/01144_multiword_data_types.sql b/tests/queries/0_stateless/01144_multiword_data_types.sql index cc380f82d63..56def658ae0 100644 --- a/tests/queries/0_stateless/01144_multiword_data_types.sql +++ b/tests/queries/0_stateless/01144_multiword_data_types.sql @@ -23,7 +23,7 @@ CREATE TABLE multiword_types ( SHOW CREATE TABLE multiword_types; INSERT INTO multiword_types(a) VALUES (1); -SELECT toTypeName((*,)) FROM multiword_types; +SELECT toTypeName((*,)) FROM multiword_types SETTINGS enable_named_columns_in_function_tuple = 0; CREATE TABLE unsigned_types ( a TINYINT SIGNED, @@ -43,7 +43,7 @@ CREATE TABLE unsigned_types ( SHOW CREATE TABLE unsigned_types; INSERT INTO unsigned_types(a) VALUES (1); -SELECT toTypeName((*,)) FROM unsigned_types; +SELECT toTypeName((*,)) FROM unsigned_types SETTINGS enable_named_columns_in_function_tuple = 0; SELECT CAST('42' AS DOUBLE PRECISION), CAST(42, 'NATIONAL CHARACTER VARYING'), CAST(-1 AS tinyint UnSiGnEd), CAST(65535, ' sMaLlInT signed '); diff --git a/tests/queries/0_stateless/01154_move_partition_long.sh b/tests/queries/0_stateless/01154_move_partition_long.sh index 24ec58c9c17..bdffa028846 100755 --- a/tests/queries/0_stateless/01154_move_partition_long.sh +++ b/tests/queries/0_stateless/01154_move_partition_long.sh @@ -107,7 +107,7 @@ export -f drop_partition_thread; export -f optimize_thread; export -f drop_part_thread; -TIMEOUT=60 +TIMEOUT=40 #timeout $TIMEOUT bash -c "create_drop_thread ${engines[@]}" & timeout $TIMEOUT bash -c 'insert_thread src' & diff --git a/tests/queries/0_stateless/01158_zookeeper_log_long.sql b/tests/queries/0_stateless/01158_zookeeper_log_long.sql index 55d4162fc48..804cdf48fb6 100644 --- a/tests/queries/0_stateless/01158_zookeeper_log_long.sql +++ b/tests/queries/0_stateless/01158_zookeeper_log_long.sql @@ -29,14 +29,20 @@ select 'parts'; select type, has_watch, op_num, replace(path, toString(serverUUID()), ''), is_ephemeral, is_sequential, if(startsWith(path, '/clickhouse/sessions'), 1, version), requests_size, request_idx, error, watch_type, watch_state, path_created, stat_version, stat_cversion, stat_dataLength, stat_numChildren from system.zookeeper_log -where (session_id, xid) in (select session_id, xid from system.zookeeper_log where path='/test/01158/' || currentDatabase() || '/rmt/replicas/1/parts/all_0_0_0') +where (session_id, xid) in ( + select session_id, xid from system.zookeeper_log where path='/test/01158/' || currentDatabase() || '/rmt/replicas/1/parts/all_0_0_0' + and (query_id='' or query_id in (select query_id from system.query_log where current_database=currentDatabase() and event_date>=yesterday())) +) order by xid, type, request_idx; select 'blocks'; select type, has_watch, op_num, path, is_ephemeral, is_sequential, version, requests_size, request_idx, error, watch_type, watch_state, path_created, stat_version, stat_cversion, stat_dataLength, stat_numChildren from system.zookeeper_log -where (session_id, xid) in (select session_id, xid from system.zookeeper_log where path like '/test/01158/' || currentDatabase() || '/rmt/blocks/%' and op_num not in (1, 12, 500)) +where (session_id, xid) in ( + select session_id, xid from system.zookeeper_log where path like '/test/01158/' || currentDatabase() || '/rmt/blocks/%' and op_num not in (1, 12, 500) + and (query_id='' or query_id in (select query_id from system.query_log where current_database=currentDatabase() and event_date>=yesterday())) +) order by xid, type, request_idx; drop table rmt sync; diff --git a/tests/queries/0_stateless/01171_mv_select_insert_isolation_long.reference b/tests/queries/0_stateless/01171_mv_select_insert_isolation_long.reference index d8bb9e310e6..e69de29bb2d 100644 --- a/tests/queries/0_stateless/01171_mv_select_insert_isolation_long.reference +++ b/tests/queries/0_stateless/01171_mv_select_insert_isolation_long.reference @@ -1,4 +0,0 @@ -275 0 138 136 0 -275 0 -275 0 138 136 0 -275 0 diff --git a/tests/queries/0_stateless/01171_mv_select_insert_isolation_long.sh b/tests/queries/0_stateless/01171_mv_select_insert_isolation_long.sh index 8344bb6f426..3b9bb50517d 100755 --- a/tests/queries/0_stateless/01171_mv_select_insert_isolation_long.sh +++ b/tests/queries/0_stateless/01171_mv_select_insert_isolation_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-parallel, no-ordinary-database, no-debug +# Tags: long, no-parallel, no-ordinary-database # Test is too heavy, avoid parallel run in Flaky Check # shellcheck disable=SC2119 @@ -7,82 +7,126 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -set -e +set -ue $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS src"; $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS dst"; $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS mv"; $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS tmp"; -$CLICKHOUSE_CLIENT --query "CREATE TABLE src (n Int8, m Int8, CONSTRAINT c CHECK xxHash32(n+m) % 8 != 0) ENGINE=MergeTree ORDER BY n PARTITION BY 0 < n SETTINGS old_parts_lifetime=0"; -$CLICKHOUSE_CLIENT --query "CREATE TABLE dst (nm Int16, CONSTRAINT c CHECK xxHash32(nm) % 8 != 0) ENGINE=MergeTree ORDER BY nm SETTINGS old_parts_lifetime=0"; -$CLICKHOUSE_CLIENT --query "CREATE MATERIALIZED VIEW mv TO dst (nm Int16) AS SELECT n*m AS nm FROM src"; -$CLICKHOUSE_CLIENT --query "CREATE TABLE tmp (x UInt8, nm Int16) ENGINE=MergeTree ORDER BY (x, nm) SETTINGS old_parts_lifetime=0" +$CLICKHOUSE_CLIENT --query "CREATE TABLE src (n Int32, m Int32, CONSTRAINT c CHECK xxHash32(n+m) % 8 != 0) ENGINE=MergeTree ORDER BY n PARTITION BY 0 < n SETTINGS old_parts_lifetime=0"; +$CLICKHOUSE_CLIENT --query "CREATE TABLE dst (nm Int32, CONSTRAINT c CHECK xxHash32(nm) % 8 != 0) ENGINE=MergeTree ORDER BY nm SETTINGS old_parts_lifetime=0"; +$CLICKHOUSE_CLIENT --query "CREATE MATERIALIZED VIEW mv TO dst (nm Int32) AS SELECT n*m AS nm FROM src"; + +$CLICKHOUSE_CLIENT --query "CREATE TABLE tmp (x UInt32, nm Int32) ENGINE=MergeTree ORDER BY (x, nm) SETTINGS old_parts_lifetime=0" $CLICKHOUSE_CLIENT --query "INSERT INTO src VALUES (0, 0)" -# some transactions will fail due to constraint -function thread_insert_commit() +function get_now() { - set -e - for i in {1..100}; do - $CLICKHOUSE_CLIENT --multiquery --query " - BEGIN TRANSACTION; - INSERT INTO src VALUES /* ($i, $1) */ ($i, $1); - SELECT throwIf((SELECT sum(nm) FROM mv) != $(($i * $1))) FORMAT Null; - INSERT INTO src VALUES /* (-$i, $1) */ (-$i, $1); - COMMIT;" 2>&1| grep -Fv "is violated at row" | grep -Fv "Transaction is not in RUNNING state" | grep -F "Received from " ||: - done + date +%s } -function thread_insert_rollback() +is_pid_exist() +{ + local pid=$1 + ps -p $pid > /dev/null +} + +function run_until_deadline_and_at_least_times() { set -e - for _ in {1..100}; do - $CLICKHOUSE_CLIENT --multiquery --query " - BEGIN TRANSACTION; - INSERT INTO src VALUES /* (42, $1) */ (42, $1); - SELECT throwIf((SELECT count() FROM src WHERE n=42 AND m=$1) != 1) FORMAT Null; - ROLLBACK;" + + local deadline=$1; shift + local min_iterations=$1; shift + local function_to_run=$1; shift + + local started_time + started_time=$(get_now) + local i=0 + + while true + do + $function_to_run $i "$@" + + [[ $(get_now) -lt $deadline ]] || break + + i=$(($i + 1)) done + + [[ $i -gt $min_iterations ]] || echo "$i/$min_iterations : not enough iterations of $function_to_run has been made from $started_time until $deadline" >&2 +} + +function insert_commit_action() +{ + set -e + + local i=$1; shift + local tag=$1; shift + + # some transactions will fail due to constraint + $CLICKHOUSE_CLIENT --multiquery --query " + BEGIN TRANSACTION; + INSERT INTO src VALUES /* ($i, $tag) */ ($i, $tag); + SELECT throwIf((SELECT sum(nm) FROM mv) != $(($i * $tag))) /* ($i, $tag) */ FORMAT Null; + INSERT INTO src VALUES /* (-$i, $tag) */ (-$i, $tag); + COMMIT; + " 2>&1 \ + | grep -Fv "is violated at row" | grep -Fv "Transaction is not in RUNNING state" | grep -F "Received from " ||: +} + + +function insert_rollback_action() +{ + set -e + + local i=$1; shift + local tag=$1; shift + + $CLICKHOUSE_CLIENT --multiquery --query " + BEGIN TRANSACTION; + INSERT INTO src VALUES /* (42, $tag) */ (42, $tag); + SELECT throwIf((SELECT count() FROM src WHERE n=42 AND m=$tag) != 1) FORMAT Null; + ROLLBACK;" } # make merges more aggressive -function thread_optimize() +function optimize_action() { set -e - while true; do - optimize_query="OPTIMIZE TABLE src" - partition_id=$(( RANDOM % 2 )) - if (( RANDOM % 2 )); then - optimize_query="OPTIMIZE TABLE dst" - partition_id="all" - fi - if (( RANDOM % 2 )); then - optimize_query="$optimize_query PARTITION ID '$partition_id'" - fi - if (( RANDOM % 2 )); then - optimize_query="$optimize_query FINAL" - fi - action="COMMIT" - if (( RANDOM % 4 )); then - action="ROLLBACK" - fi - $CLICKHOUSE_CLIENT --multiquery --query " + optimize_query="OPTIMIZE TABLE src" + partition_id=$(( RANDOM % 2 )) + if (( RANDOM % 2 )); then + optimize_query="OPTIMIZE TABLE dst" + partition_id="all" + fi + if (( RANDOM % 2 )); then + optimize_query="$optimize_query PARTITION ID '$partition_id'" + fi + if (( RANDOM % 2 )); then + optimize_query="$optimize_query FINAL" + fi + action="COMMIT" + if (( RANDOM % 4 )); then + action="ROLLBACK" + fi + + $CLICKHOUSE_CLIENT --multiquery --query " BEGIN TRANSACTION; - $optimize_query; + $optimize_query; $action; - " 2>&1| grep -Fv "already exists, but it will be deleted soon" | grep -F "Received from " ||: - sleep 0.$RANDOM; - done + " 2>&1 \ + | grep -Fv "already exists, but it will be deleted soon" | grep -F "Received from " ||: + + sleep 0.$RANDOM; } -function thread_select() +function select_action() { set -e - while true; do - $CLICKHOUSE_CLIENT --multiquery --query " + + $CLICKHOUSE_CLIENT --multiquery --query " BEGIN TRANSACTION; SELECT throwIf((SELECT (sum(n), count() % 2) FROM src) != (0, 1)) FORMAT Null; SELECT throwIf((SELECT (sum(nm), count() % 2) FROM mv) != (0, 1)) FORMAT Null; @@ -90,14 +134,13 @@ function thread_select() SELECT throwIf((SELECT arraySort(groupArray(nm)) FROM mv) != (SELECT arraySort(groupArray(nm)) FROM dst)) FORMAT Null; SELECT throwIf((SELECT arraySort(groupArray(nm)) FROM mv) != (SELECT arraySort(groupArray(n*m)) FROM src)) FORMAT Null; COMMIT;" - done } -function thread_select_insert() +function select_insert_action() { set -e - while true; do - $CLICKHOUSE_CLIENT --multiquery --query " + + $CLICKHOUSE_CLIENT --multiquery --query " BEGIN TRANSACTION; SELECT throwIf((SELECT count() FROM tmp) != 0) FORMAT Null; INSERT INTO tmp SELECT 1, n*m FROM src; @@ -110,36 +153,69 @@ function thread_select_insert() SELECT throwIf(1 != (SELECT countDistinct(arr) FROM (SELECT x, arraySort(groupArray(nm)) AS arr FROM tmp WHERE x!=4 GROUP BY x))) FORMAT Null; SELECT throwIf((SELECT count(), sum(nm) FROM tmp WHERE x=4) != (SELECT count(), sum(nm) FROM tmp WHERE x!=4)) FORMAT Null; ROLLBACK;" - done } -thread_insert_commit 1 & PID_1=$! -thread_insert_commit 2 & PID_2=$! -thread_insert_rollback 3 & PID_3=$! +MAIN_TIME_PART=400 +SECOND_TIME_PART=30 +WAIT_FINISH=60 +LAST_TIME_GAP=10 -thread_optimize & PID_4=$! -thread_select & PID_5=$! -thread_select_insert & PID_6=$! -sleep 0.$RANDOM; -thread_select & PID_7=$! -thread_select_insert & PID_8=$! +if [[ $((MAIN_TIME_PART + SECOND_TIME_PART + WAIT_FINISH + LAST_TIME_GAP)) -ge 600 ]]; then + echo "time sttings are wrong" 2>&1 + exit 1 +fi -wait $PID_1 && wait $PID_2 && wait $PID_3 -kill -TERM $PID_4 -kill -TERM $PID_5 -kill -TERM $PID_6 -kill -TERM $PID_7 -kill -TERM $PID_8 -wait -wait_for_queries_to_finish 40 +START_TIME=$(get_now) +STOP_TIME=$((START_TIME + MAIN_TIME_PART)) +SECOND_STOP_TIME=$((STOP_TIME + SECOND_TIME_PART)) +MIN_ITERATIONS=20 + +run_until_deadline_and_at_least_times $STOP_TIME $MIN_ITERATIONS insert_commit_action 1 & PID_1=$! +run_until_deadline_and_at_least_times $STOP_TIME $MIN_ITERATIONS insert_commit_action 2 & PID_2=$! +run_until_deadline_and_at_least_times $STOP_TIME $MIN_ITERATIONS insert_rollback_action 3 & PID_3=$! + +run_until_deadline_and_at_least_times $SECOND_STOP_TIME $MIN_ITERATIONS optimize_action & PID_4=$! +run_until_deadline_and_at_least_times $SECOND_STOP_TIME $MIN_ITERATIONS select_action & PID_5=$! +run_until_deadline_and_at_least_times $SECOND_STOP_TIME $MIN_ITERATIONS select_insert_action & PID_6=$! +sleep 0.$RANDOM +run_until_deadline_and_at_least_times $SECOND_STOP_TIME $MIN_ITERATIONS select_action & PID_7=$! +run_until_deadline_and_at_least_times $SECOND_STOP_TIME $MIN_ITERATIONS select_insert_action & PID_8=$! + +wait $PID_1 || echo "insert_commit_action has failed with status $?" 2>&1 +wait $PID_2 || echo "second insert_commit_action has failed with status $?" 2>&1 +wait $PID_3 || echo "insert_rollback_action has failed with status $?" 2>&1 + +is_pid_exist $PID_4 || echo "optimize_action is not running" 2>&1 +is_pid_exist $PID_5 || echo "select_action is not running" 2>&1 +is_pid_exist $PID_6 || echo "select_insert_action is not running" 2>&1 +is_pid_exist $PID_7 || echo "second select_action is not running" 2>&1 +is_pid_exist $PID_8 || echo "second select_insert_action is not running" 2>&1 + +wait $PID_4 || echo "optimize_action has failed with status $?" 2>&1 +wait $PID_5 || echo "select_action has failed with status $?" 2>&1 +wait $PID_6 || echo "select_insert_action has failed with status $?" 2>&1 +wait $PID_7 || echo "second select_action has failed with status $?" 2>&1 +wait $PID_8 || echo "second select_insert_action has failed with status $?" 2>&1 + +wait_for_queries_to_finish $WAIT_FINISH $CLICKHOUSE_CLIENT --multiquery --query " -BEGIN TRANSACTION; -SELECT count(), sum(n), sum(m=1), sum(m=2), sum(m=3) FROM src; -SELECT count(), sum(nm) FROM mv"; + BEGIN TRANSACTION; + SELECT throwIf((SELECT (sum(n), count() % 2) FROM src) != (0, 1)) FORMAT Null; + SELECT throwIf((SELECT (sum(nm), count() % 2) FROM mv) != (0, 1)) FORMAT Null; + SELECT throwIf((SELECT (sum(nm), count() % 2) FROM dst) != (0, 1)) FORMAT Null; + SELECT throwIf((SELECT arraySort(groupArray(nm)) FROM mv) != (SELECT arraySort(groupArray(nm)) FROM dst)) FORMAT Null; + SELECT throwIf((SELECT arraySort(groupArray(nm)) FROM mv) != (SELECT arraySort(groupArray(n*m)) FROM src)) FORMAT Null; + COMMIT; +" -$CLICKHOUSE_CLIENT --query "SELECT count(), sum(n), sum(m=1), sum(m=2), sum(m=3) FROM src" -$CLICKHOUSE_CLIENT --query "SELECT count(), sum(nm) FROM mv" +$CLICKHOUSE_CLIENT --multiquery --query " + SELECT throwIf((SELECT (sum(n), count() % 2) FROM src) != (0, 1)) FORMAT Null; + SELECT throwIf((SELECT (sum(nm), count() % 2) FROM mv) != (0, 1)) FORMAT Null; + SELECT throwIf((SELECT (sum(nm), count() % 2) FROM dst) != (0, 1)) FORMAT Null; + SELECT throwIf((SELECT arraySort(groupArray(nm)) FROM mv) != (SELECT arraySort(groupArray(nm)) FROM dst)) FORMAT Null; + SELECT throwIf((SELECT arraySort(groupArray(nm)) FROM mv) != (SELECT arraySort(groupArray(n*m)) FROM src)) FORMAT Null; +" $CLICKHOUSE_CLIENT --query "DROP TABLE src"; $CLICKHOUSE_CLIENT --query "DROP TABLE dst"; diff --git a/tests/queries/0_stateless/01185_create_or_replace_table.sql b/tests/queries/0_stateless/01185_create_or_replace_table.sql index 11759d0bb0c..801a775e024 100644 --- a/tests/queries/0_stateless/01185_create_or_replace_table.sql +++ b/tests/queries/0_stateless/01185_create_or_replace_table.sql @@ -1,4 +1,4 @@ --- Tags: no-ordinary-database, no-parallel +-- Tags: no-ordinary-database drop table if exists t1; diff --git a/tests/queries/0_stateless/01188_attach_table_from_path.sql b/tests/queries/0_stateless/01188_attach_table_from_path.sql index d1b9493b6c2..026979a0132 100644 --- a/tests/queries/0_stateless/01188_attach_table_from_path.sql +++ b/tests/queries/0_stateless/01188_attach_table_from_path.sql @@ -1,4 +1,4 @@ --- Tags: no-replicated-database, no-parallel +-- Tags: no-replicated-database drop table if exists test; drop table if exists file; diff --git a/tests/queries/0_stateless/01191_rename_dictionary.sql b/tests/queries/0_stateless/01191_rename_dictionary.sql index 6666c3308ca..c5012dabc81 100644 --- a/tests/queries/0_stateless/01191_rename_dictionary.sql +++ b/tests/queries/0_stateless/01191_rename_dictionary.sql @@ -17,7 +17,7 @@ SELECT name, status FROM system.dictionaries WHERE database='test_01191'; SELECT name, engine FROM system.tables WHERE database='test_01191' ORDER BY name; RENAME DICTIONARY test_01191.table TO test_01191.table1; -- {serverError UNKNOWN_TABLE} -EXCHANGE DICTIONARIES test_01191._ AND test_01191.dict; -- {serverError INCORRECT_QUERY} +EXCHANGE DICTIONARIES test_01191._ AND test_01191.dict; -- {serverError INFINITE_LOOP} EXCHANGE TABLES test_01191.t AND test_01191.dict; SELECT name, status FROM system.dictionaries WHERE database='test_01191'; SELECT name, engine FROM system.tables WHERE database='test_01191' ORDER BY name; diff --git a/tests/queries/0_stateless/01194_http_query_id.sh b/tests/queries/0_stateless/01194_http_query_id.sh index 5aebdc10dfc..42321112185 100755 --- a/tests/queries/0_stateless/01194_http_query_id.sh +++ b/tests/queries/0_stateless/01194_http_query_id.sh @@ -4,14 +4,22 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -url="http://${CLICKHOUSE_HOST}:${CLICKHOUSE_PORT_HTTP}/?session_id=test_01194" -rnd=$RANDOM +rnd="$CLICKHOUSE_DATABASE" +url="${CLICKHOUSE_URL}&session_id=test_01194_${CLICKHOUSE_DATABASE}" ${CLICKHOUSE_CURL} -sS "$url&query=SELECT+'test_01194',$rnd,1" > /dev/null ${CLICKHOUSE_CURL} -sS "$url&query=SELECT+'test_01194',$rnd,2" > /dev/null ${CLICKHOUSE_CURL} -sS "$url" --data "SELECT 'test_01194',$rnd,3" > /dev/null ${CLICKHOUSE_CURL} -sS "$url" --data "SELECT 'test_01194',$rnd,4" > /dev/null -${CLICKHOUSE_CURL} -sS "$url" --data "SYSTEM FLUSH LOGS" +$CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS" -${CLICKHOUSE_CURL} -sS "$url&query=SELECT+count(DISTINCT+query_id)+FROM+system.query_log+WHERE+current_database+LIKE+currentDatabase()+AND+query+LIKE+'SELECT+''test_01194'',$rnd%25'" +$CLICKHOUSE_CLIENT -q " + SELECT + count(DISTINCT query_id) + FROM system.query_log + WHERE + current_database = currentDatabase() + AND event_date >= yesterday() + AND query LIKE 'SELECT ''test_01194'',$rnd%' + AND query_id != queryID()" diff --git a/tests/queries/0_stateless/01200_mutations_memory_consumption.sql b/tests/queries/0_stateless/01200_mutations_memory_consumption.sql index 2266da5fc8f..f2d071961ee 100644 --- a/tests/queries/0_stateless/01200_mutations_memory_consumption.sql +++ b/tests/queries/0_stateless/01200_mutations_memory_consumption.sql @@ -1,4 +1,5 @@ --- Tags: no-debug, no-parallel, long, no-s3-storage, no-random-settings, no-random-merge-tree-settings +-- Tags: no-debug, no-parallel, long, no-object-storage, no-random-settings, no-random-merge-tree-settings +SET optimize_trivial_insert_select = 1; DROP TABLE IF EXISTS table_with_single_pk; diff --git a/tests/queries/0_stateless/01202_array_auc_special.reference b/tests/queries/0_stateless/01202_array_auc_special.reference index 85c230fba58..8f3f0cf1efe 100644 --- a/tests/queries/0_stateless/01202_array_auc_special.reference +++ b/tests/queries/0_stateless/01202_array_auc_special.reference @@ -1,9 +1,9 @@ nan nan nan -0 -1 -0 0.5 1 -0.5 +0 +0.75 +1 +0.75 diff --git a/tests/queries/0_stateless/01221_system_settings.sql b/tests/queries/0_stateless/01221_system_settings.sql index fcffd6c45fe..da0204b37bd 100644 --- a/tests/queries/0_stateless/01221_system_settings.sql +++ b/tests/queries/0_stateless/01221_system_settings.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage select * from system.settings where name = 'send_timeout'; select * from system.merge_tree_settings order by length(description) limit 1; diff --git a/tests/queries/0_stateless/01225_drop_dictionary_as_table.sql b/tests/queries/0_stateless/01225_drop_dictionary_as_table.sql index 3e497f9e3a4..a0cacd8bc7a 100644 --- a/tests/queries/0_stateless/01225_drop_dictionary_as_table.sql +++ b/tests/queries/0_stateless/01225_drop_dictionary_as_table.sql @@ -1,22 +1,15 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS dict_db_01225; -CREATE DATABASE dict_db_01225; - -CREATE TABLE dict_db_01225.dict_data (key UInt64, val UInt64) Engine=Memory(); -CREATE DICTIONARY dict_db_01225.dict +CREATE TABLE dict_data (key UInt64, val UInt64) Engine=Memory(); +CREATE DICTIONARY dict ( key UInt64 DEFAULT 0, val UInt64 DEFAULT 10 ) PRIMARY KEY key -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_data' PASSWORD '' DB 'dict_db_01225')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_data' PASSWORD '' DB currentDatabase())) LIFETIME(MIN 0 MAX 0) LAYOUT(FLAT()); -SYSTEM RELOAD DICTIONARY dict_db_01225.dict; +SYSTEM RELOAD DICTIONARY dict; -DROP TABLE dict_db_01225.dict; -- { serverError CANNOT_DETACH_DICTIONARY_AS_TABLE } -DROP DICTIONARY dict_db_01225.dict; - -DROP DATABASE dict_db_01225; +DROP TABLE dict; -- { serverError CANNOT_DETACH_DICTIONARY_AS_TABLE } +DROP DICTIONARY dict; diff --git a/tests/queries/0_stateless/01232_untuple.reference b/tests/queries/0_stateless/01232_untuple.reference index 0358cde1354..3cd8eaa5611 100644 --- a/tests/queries/0_stateless/01232_untuple.reference +++ b/tests/queries/0_stateless/01232_untuple.reference @@ -2,7 +2,7 @@ hello 1 3 world 9 9 (0,1) -key tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'1\') tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'2\') tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'3\') tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'4\') tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'5\') +key tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'v1\') tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'v2\') tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'v3\') tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'v4\') tupleElement(argMax((v1, v2, v3, v4, v5), v1), \'v5\') 1 20 20 10 20 30 2 11 20 10 20 30 3 70 20 10 20 30 diff --git a/tests/queries/0_stateless/01232_untuple.sql b/tests/queries/0_stateless/01232_untuple.sql index ccefd13a772..391d08ab859 100644 --- a/tests/queries/0_stateless/01232_untuple.sql +++ b/tests/queries/0_stateless/01232_untuple.sql @@ -1,4 +1,5 @@ SET allow_experimental_analyzer = 1; +SET enable_named_columns_in_function_tuple = 1; select untuple((* except (b),)) from (select 1 a, 2 b, 3 c); select 'hello', untuple((* except (b),)), 'world' from (select 1 a, 2 b, 3 c); diff --git a/tests/queries/0_stateless/01246_buffer_flush.sh b/tests/queries/0_stateless/01246_buffer_flush.sh new file mode 100755 index 00000000000..27c3f01f216 --- /dev/null +++ b/tests/queries/0_stateless/01246_buffer_flush.sh @@ -0,0 +1,89 @@ +#!/usr/bin/env bash +# Tags: no-fasttest + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +set -e + +function wait_until() +{ + local expr=$1 && shift + while ! eval "$expr"; do + sleep 0.5 + done +} +function get_buffer_delay() +{ + local buffer_insert_id=$1 && shift + $CLICKHOUSE_CLIENT -nm -q " + SYSTEM FLUSH LOGS; + WITH + (SELECT event_time_microseconds FROM system.query_log WHERE current_database = currentDatabase() AND type = 'QueryStart' AND query_id = '$buffer_insert_id') AS begin_, + (SELECT max(event_time) FROM data_01256) AS end_ + SELECT dateDiff('seconds', begin_, end_)::UInt64; + " +} + +$CLICKHOUSE_CLIENT -nm -q " + drop table if exists data_01256; + drop table if exists buffer_01256; + + create table data_01256 (key UInt64, event_time DateTime(6) MATERIALIZED now64(6)) Engine=Memory(); +" + +echo "min" +$CLICKHOUSE_CLIENT -q " + create table buffer_01256 (key UInt64) Engine=Buffer(currentDatabase(), data_01256, 1, + 2, 100, /* time */ + 4, 100, /* rows */ + 1, 1e6 /* bytes */ + ) +" +min_query_id=$(random_str 10) +$CLICKHOUSE_CLIENT --query_id="$min_query_id" -q "insert into buffer_01256 select * from system.numbers limit 5" +$CLICKHOUSE_CLIENT -q "select count() from data_01256" +wait_until '[[ $($CLICKHOUSE_CLIENT -q "select count() from data_01256") -eq 5 ]]' +sec=$(get_buffer_delay "$min_query_id") +[[ $sec -ge 2 ]] || echo "Buffer flushed too early, min_time=2, flushed after $sec sec" +[[ $sec -lt 100 ]] || echo "Buffer flushed too late, max_time=100, flushed after $sec sec" +$CLICKHOUSE_CLIENT -q "select count() from data_01256" +$CLICKHOUSE_CLIENT -q "drop table buffer_01256" + +echo "max" +$CLICKHOUSE_CLIENT -q " + create table buffer_01256 (key UInt64) Engine=Buffer(currentDatabase(), data_01256, 1, + 100, 2, /* time */ + 0, 100, /* rows */ + 0, 1e6 /* bytes */ + ); +" +max_query_id=$(random_str 10) +$CLICKHOUSE_CLIENT --query_id="$max_query_id" -q "insert into buffer_01256 select * from system.numbers limit 5" +$CLICKHOUSE_CLIENT -q "select count() from data_01256" +wait_until '[[ $($CLICKHOUSE_CLIENT -q "select count() from data_01256") -eq 10 ]]' +sec=$(get_buffer_delay "$max_query_id") +[[ $sec -ge 2 ]] || echo "Buffer flushed too early, max_time=2, flushed after $sec sec" +$CLICKHOUSE_CLIENT -q "select count() from data_01256" +$CLICKHOUSE_CLIENT -q "drop table buffer_01256" + +echo "direct" +$CLICKHOUSE_CLIENT -nm -q " + create table buffer_01256 (key UInt64) Engine=Buffer(currentDatabase(), data_01256, 1, + 100, 100, /* time */ + 0, 9, /* rows */ + 0, 1e6 /* bytes */ + ); + insert into buffer_01256 select * from system.numbers limit 10; + select count() from data_01256; +" + +echo "drop" +$CLICKHOUSE_CLIENT -nm -q " + insert into buffer_01256 select * from system.numbers limit 10; + drop table if exists buffer_01256; + select count() from data_01256; +" + +$CLICKHOUSE_CLIENT -q "drop table data_01256" diff --git a/tests/queries/0_stateless/01246_buffer_flush.sql b/tests/queries/0_stateless/01246_buffer_flush.sql deleted file mode 100644 index 66f93371c29..00000000000 --- a/tests/queries/0_stateless/01246_buffer_flush.sql +++ /dev/null @@ -1,50 +0,0 @@ --- Tags: no-fasttest - -SET function_sleep_max_microseconds_per_block = 4000000; - -drop table if exists data_01256; -drop table if exists buffer_01256; - -create table data_01256 as system.numbers Engine=Memory(); - -select 'min'; -create table buffer_01256 as system.numbers Engine=Buffer(currentDatabase(), data_01256, 1, - 5, 100, /* time */ - 4, 100, /* rows */ - 1, 1e6 /* bytes */ -); -insert into buffer_01256 select * from system.numbers limit 5; -select count() from data_01256; --- It is enough to ensure that the buffer will be flushed earlier then 2*min_time (10 sec) -select sleepEachRow(9) FORMAT Null SETTINGS function_sleep_max_microseconds_per_block=10e6; -select count() from data_01256; -drop table buffer_01256; - -select 'max'; -create table buffer_01256 as system.numbers Engine=Buffer(currentDatabase(), data_01256, 1, - 100, 2, /* time */ - 0, 100, /* rows */ - 0, 1e6 /* bytes */ -); -insert into buffer_01256 select * from system.numbers limit 5; -select count() from data_01256; --- sleep 2 (min time) + 1 (round up) + bias (1) = 4 -select sleepEachRow(2) from numbers(2) FORMAT Null; -select count() from data_01256; -drop table buffer_01256; - -select 'direct'; -create table buffer_01256 as system.numbers Engine=Buffer(currentDatabase(), data_01256, 1, - 100, 100, /* time */ - 0, 9, /* rows */ - 0, 1e6 /* bytes */ -); -insert into buffer_01256 select * from system.numbers limit 10; -select count() from data_01256; - -select 'drop'; -insert into buffer_01256 select * from system.numbers limit 10; -drop table if exists buffer_01256; -select count() from data_01256; - -drop table data_01256; diff --git a/tests/queries/0_stateless/01254_dict_create_without_db.sql b/tests/queries/0_stateless/01254_dict_create_without_db.sql index 65a2ab52d23..2d4da5af9a9 100644 --- a/tests/queries/0_stateless/01254_dict_create_without_db.sql +++ b/tests/queries/0_stateless/01254_dict_create_without_db.sql @@ -1,9 +1,3 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS dict_db_01254; -CREATE DATABASE dict_db_01254; -USE dict_db_01254; - CREATE TABLE dict_data (key UInt64, val UInt64) Engine=Memory(); CREATE DICTIONARY dict ( @@ -11,15 +5,12 @@ CREATE DICTIONARY dict val UInt64 DEFAULT 10 ) PRIMARY KEY key -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_data' PASSWORD '' DB 'dict_db_01254')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_data' PASSWORD '' DB currentDatabase())) LIFETIME(MIN 0 MAX 0) LAYOUT(FLAT()); -SELECT query_count, status FROM system.dictionaries WHERE database = 'dict_db_01254' AND name = 'dict'; -SYSTEM RELOAD DICTIONARY dict_db_01254.dict; -SELECT query_count, status FROM system.dictionaries WHERE database = 'dict_db_01254' AND name = 'dict'; -SELECT dictGetUInt64('dict_db_01254.dict', 'val', toUInt64(0)); -SELECT query_count, status FROM system.dictionaries WHERE database = 'dict_db_01254' AND name = 'dict'; - -USE system; -DROP DATABASE dict_db_01254; +SELECT query_count, status FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; +SYSTEM RELOAD DICTIONARY dict; +SELECT query_count, status FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; +SELECT dictGetUInt64('dict', 'val', toUInt64(0)); +SELECT query_count, status FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; diff --git a/tests/queries/0_stateless/01254_dict_load_after_detach_attach.sql b/tests/queries/0_stateless/01254_dict_load_after_detach_attach.sql index 206ddeac612..11473c6ce32 100644 --- a/tests/queries/0_stateless/01254_dict_load_after_detach_attach.sql +++ b/tests/queries/0_stateless/01254_dict_load_after_detach_attach.sql @@ -1,26 +1,19 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS dict_db_01254; -CREATE DATABASE dict_db_01254; - -CREATE TABLE dict_db_01254.dict_data (key UInt64, val UInt64) Engine=Memory(); -CREATE DICTIONARY dict_db_01254.dict +CREATE TABLE dict_data (key UInt64, val UInt64) Engine=Memory(); +CREATE DICTIONARY dict ( key UInt64 DEFAULT 0, val UInt64 DEFAULT 10 ) PRIMARY KEY key -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_data' PASSWORD '' DB 'dict_db_01254')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'dict_data' PASSWORD '' DB currentDatabase())) LIFETIME(MIN 0 MAX 0) LAYOUT(FLAT()); -DETACH DATABASE dict_db_01254; -ATTACH DATABASE dict_db_01254; +DETACH DATABASE {CLICKHOUSE_DATABASE:Identifier}; +ATTACH DATABASE {CLICKHOUSE_DATABASE:Identifier}; -SELECT COALESCE((SELECT status FROM system.dictionaries WHERE database = 'dict_db_01254' AND name = 'dict')::Nullable(String), 'NOT_LOADED'); -SYSTEM RELOAD DICTIONARY dict_db_01254.dict; -SELECT query_count, status FROM system.dictionaries WHERE database = 'dict_db_01254' AND name = 'dict'; -SELECT dictGetUInt64('dict_db_01254.dict', 'val', toUInt64(0)); -SELECT query_count, status FROM system.dictionaries WHERE database = 'dict_db_01254' AND name = 'dict'; - -DROP DATABASE dict_db_01254; +SELECT COALESCE((SELECT status FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict')::Nullable(String), 'NOT_LOADED'); +SYSTEM RELOAD DICTIONARY dict; +SELECT query_count, status FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; +SELECT dictGetUInt64('dict', 'val', toUInt64(0)); +SELECT query_count, status FROM system.dictionaries WHERE database = currentDatabase() AND name = 'dict'; diff --git a/tests/queries/0_stateless/01259_dictionary_custom_settings_ddl.sql b/tests/queries/0_stateless/01259_dictionary_custom_settings_ddl.sql index 432256d33c2..be56806f8d6 100644 --- a/tests/queries/0_stateless/01259_dictionary_custom_settings_ddl.sql +++ b/tests/queries/0_stateless/01259_dictionary_custom_settings_ddl.sql @@ -1,12 +1,6 @@ --- Tags: no-parallel, no-fasttest +-- Tags: no-fasttest -DROP DATABASE IF EXISTS database_for_dict; - -CREATE DATABASE database_for_dict; - -DROP TABLE IF EXISTS database_for_dict.table_for_dict; - -CREATE TABLE database_for_dict.table_for_dict +CREATE TABLE table_for_dict ( key_column UInt64, second_column UInt64, @@ -15,7 +9,7 @@ CREATE TABLE database_for_dict.table_for_dict ENGINE = MergeTree() ORDER BY key_column; -INSERT INTO database_for_dict.table_for_dict VALUES (100500, 10000000, 'Hello world'); +INSERT INTO table_for_dict VALUES (100500, 10000000, 'Hello world'); DROP DATABASE IF EXISTS ordinary_db; @@ -30,7 +24,7 @@ CREATE DICTIONARY ordinary_db.dict1 third_column String DEFAULT 'qqq' ) PRIMARY KEY key_column -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_for_dict' PASSWORD '' DB 'database_for_dict')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_for_dict' PASSWORD '' DB currentDatabase())) LIFETIME(MIN 1 MAX 10) LAYOUT(FLAT()) SETTINGS(max_result_bytes=1); @@ -40,10 +34,6 @@ SELECT dictGetUInt64('ordinary_db.dict1', 'second_column', toUInt64(100500)); -- SELECT 'END'; -DROP DICTIONARY IF EXISTS ordinary_db.dict1; - DROP DATABASE IF EXISTS ordinary_db; -DROP TABLE IF EXISTS database_for_dict.table_for_dict; - -DROP DATABASE IF EXISTS database_for_dict; +DROP TABLE IF EXISTS table_for_dict; diff --git a/tests/queries/0_stateless/01268_procfs_metrics.sh b/tests/queries/0_stateless/01268_procfs_metrics.sh index 4f09d197596..7d6389bb86e 100755 --- a/tests/queries/0_stateless/01268_procfs_metrics.sh +++ b/tests/queries/0_stateless/01268_procfs_metrics.sh @@ -15,7 +15,7 @@ tmp_path=$(mktemp "$CURDIR/01268_procfs_metrics.XXXXXX") trap 'rm -f $tmp_path' EXIT truncate -s1025 "$tmp_path" -$CLICKHOUSE_LOCAL --profile-events-delay-ms=-1 --print-profile-events -q "SELECT * FROM file('$tmp_path', 'LineAsString') FORMAT Null" |& grep -m1 -F -o -e OSReadChars +$CLICKHOUSE_LOCAL --profile-events-delay-ms=-1 --print-profile-events --storage_file_read_method=pread -q "SELECT * FROM file('$tmp_path', 'LineAsString') FORMAT Null" |& grep -m1 -F -o -e OSReadChars # NOTE: that OSCPUVirtualTimeMicroseconds is in microseconds, so 1e6 is not enough. $CLICKHOUSE_LOCAL --profile-events-delay-ms=-1 --print-profile-events -q "SELECT * FROM numbers(1e8) FORMAT Null" |& grep -m1 -F -o -e OSCPUVirtualTimeMicroseconds exit 0 diff --git a/tests/queries/0_stateless/01269_alias_type_differs.sql b/tests/queries/0_stateless/01269_alias_type_differs.sql index 64abcf9e367..b78e46f62c8 100644 --- a/tests/queries/0_stateless/01269_alias_type_differs.sql +++ b/tests/queries/0_stateless/01269_alias_type_differs.sql @@ -1,5 +1,3 @@ --- Tags: no-parallel - DROP TABLE IF EXISTS data_01269; CREATE TABLE data_01269 ( diff --git a/tests/queries/0_stateless/01272_suspicious_codecs.sql b/tests/queries/0_stateless/01272_suspicious_codecs.sql index 082a8d08675..1c1d7b58dd0 100644 --- a/tests/queries/0_stateless/01272_suspicious_codecs.sql +++ b/tests/queries/0_stateless/01272_suspicious_codecs.sql @@ -1,9 +1,5 @@ --- Tags: no-parallel - DROP TABLE IF EXISTS codecs; --- test what should work - CREATE TABLE codecs ( a UInt8 CODEC(LZ4), diff --git a/tests/queries/0_stateless/01275_parallel_mv.reference b/tests/queries/0_stateless/01275_parallel_mv.reference index a9801e3b910..cf1f7054f6c 100644 --- a/tests/queries/0_stateless/01275_parallel_mv.reference +++ b/tests/queries/0_stateless/01275_parallel_mv.reference @@ -2,15 +2,20 @@ -select 'optimize_trivial_insert_select=0', 'max_insert_threads=0'; -optimize_trivial_insert_select=0 max_insert_threads=0 -insert into testX select number from numbers(10) settings + + + +select 'optimize_trivial_insert_select=0', 'max_insert_threads=0', 'iteration_num='; +optimize_trivial_insert_select=0 max_insert_threads=0 iteration_num= +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_1', log_queries=1, parallel_view_processing=0, optimize_trivial_insert_select=0, - max_insert_threads=0; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads=0; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and @@ -19,46 +24,46 @@ select arrayUniq(thread_ids) from system.query_log where Settings['max_insert_threads'] = '0'; 2 select count() from testX; -10 +200 select count() from testXA; -10 -select count() from testXB; -0 +200 select count() from testXC; -10 -select 'optimize_trivial_insert_select=0', 'max_insert_threads=16'; -optimize_trivial_insert_select=0 max_insert_threads=16 -insert into testX select number from numbers(10) settings +200 +select 'optimize_trivial_insert_select=0', 'max_insert_threads=5', 'iteration_num='; +optimize_trivial_insert_select=0 max_insert_threads=5 iteration_num= +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_2', log_queries=1, parallel_view_processing=0, optimize_trivial_insert_select=0, - max_insert_threads=16; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads=5; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and Settings['parallel_view_processing'] = '0' and Settings['optimize_trivial_insert_select'] = '0' and - Settings['max_insert_threads'] = '16'; + Settings['max_insert_threads'] = '5'; 2 select count() from testX; -20 +400 select count() from testXA; -20 -select count() from testXB; -0 +400 select count() from testXC; -20 -select 'optimize_trivial_insert_select=1', 'max_insert_threads=0'; -optimize_trivial_insert_select=1 max_insert_threads=0 -insert into testX select number from numbers(10) settings +400 +select 'optimize_trivial_insert_select=1', 'max_insert_threads=0', 'iteration_num='; +optimize_trivial_insert_select=1 max_insert_threads=0 iteration_num= +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_3', log_queries=1, parallel_view_processing=0, optimize_trivial_insert_select=1, - max_insert_threads=0; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads=0; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and @@ -67,46 +72,46 @@ select arrayUniq(thread_ids) from system.query_log where Settings['max_insert_threads'] = '0'; 2 select count() from testX; -30 +600 select count() from testXA; -30 -select count() from testXB; -0 +600 select count() from testXC; -30 -select 'optimize_trivial_insert_select=1', 'max_insert_threads=16'; -optimize_trivial_insert_select=1 max_insert_threads=16 -insert into testX select number from numbers(10) settings +600 +select 'optimize_trivial_insert_select=1', 'max_insert_threads=5', 'iteration_num='; +optimize_trivial_insert_select=1 max_insert_threads=5 iteration_num= +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_4', log_queries=1, parallel_view_processing=0, optimize_trivial_insert_select=1, - max_insert_threads=16; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads=5; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and Settings['parallel_view_processing'] = '0' and Settings['optimize_trivial_insert_select'] = '1' and - Settings['max_insert_threads'] = '16'; + Settings['max_insert_threads'] = '5'; 2 select count() from testX; -40 +800 select count() from testXA; -40 -select count() from testXB; -0 +800 select count() from testXC; -40 -select 'optimize_trivial_insert_select=0', 'max_insert_threads=0'; -optimize_trivial_insert_select=0 max_insert_threads=0 -insert into testX select number from numbers(10) settings +800 +select 'optimize_trivial_insert_select=0', 'max_insert_threads=0', 'iteration_num='; +optimize_trivial_insert_select=0 max_insert_threads=0 iteration_num= +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_5', log_queries=1, parallel_view_processing=1, optimize_trivial_insert_select=0, - max_insert_threads=0; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads=0; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and @@ -115,82 +120,80 @@ select arrayUniq(thread_ids) from system.query_log where Settings['max_insert_threads'] = '0'; 5 select count() from testX; -50 +1000 select count() from testXA; -50 -select count() from testXB; -0 +1000 select count() from testXC; -50 -select 'optimize_trivial_insert_select=0', 'max_insert_threads=16'; -optimize_trivial_insert_select=0 max_insert_threads=16 -insert into testX select number from numbers(10) settings +1000 +select 'optimize_trivial_insert_select=0', 'max_insert_threads=5', 'iteration_num='; +optimize_trivial_insert_select=0 max_insert_threads=5 iteration_num= +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_6', log_queries=1, parallel_view_processing=1, optimize_trivial_insert_select=0, - max_insert_threads=16; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads=5; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and Settings['parallel_view_processing'] = '1' and Settings['optimize_trivial_insert_select'] = '0' and - Settings['max_insert_threads'] = '16'; -5 + Settings['max_insert_threads'] = '5'; +12 select count() from testX; -60 +1200 select count() from testXA; -60 -select count() from testXB; -0 +1200 select count() from testXC; -60 -select 'optimize_trivial_insert_select=1', 'max_insert_threads=0'; -optimize_trivial_insert_select=1 max_insert_threads=0 -insert into testX select number from numbers(10) settings +1200 +select 'optimize_trivial_insert_select=1', 'max_insert_threads=0', 'iteration_num='; +optimize_trivial_insert_select=1 max_insert_threads=0 iteration_num= +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_7', log_queries=1, parallel_view_processing=1, optimize_trivial_insert_select=1, - max_insert_threads=0; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads=0; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and Settings['parallel_view_processing'] = '1' and Settings['optimize_trivial_insert_select'] = '1' and Settings['max_insert_threads'] = '0'; -5 +2 select count() from testX; -70 +1400 select count() from testXA; -70 -select count() from testXB; -0 +1400 select count() from testXC; -70 -select 'optimize_trivial_insert_select=1', 'max_insert_threads=16'; -optimize_trivial_insert_select=1 max_insert_threads=16 -insert into testX select number from numbers(10) settings +1400 +select 'optimize_trivial_insert_select=1', 'max_insert_threads=5', 'iteration_num='; +optimize_trivial_insert_select=1 max_insert_threads=5 iteration_num= +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_8', log_queries=1, parallel_view_processing=1, optimize_trivial_insert_select=1, - max_insert_threads=16; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads=5; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and Settings['parallel_view_processing'] = '1' and Settings['optimize_trivial_insert_select'] = '1' and - Settings['max_insert_threads'] = '16'; -5 + Settings['max_insert_threads'] = '5'; +7 select count() from testX; -80 +1600 select count() from testXA; -80 -select count() from testXB; -0 +1600 select count() from testXC; -80 +1600 diff --git a/tests/queries/0_stateless/01275_parallel_mv.sql.j2 b/tests/queries/0_stateless/01275_parallel_mv.sql.j2 index 047b1cc3ee7..6805b8902ec 100644 --- a/tests/queries/0_stateless/01275_parallel_mv.sql.j2 +++ b/tests/queries/0_stateless/01275_parallel_mv.sql.j2 @@ -1,9 +1,15 @@ --- Tags: no-s3-storage, no-parallel, no-fasttest --- no-s3-storage: s3 has 20 more threads +-- Tags: no-object-storage, no-parallel, no-fasttest +-- no-object-storage: s3 has 20 more threads -- no-parallel: it checks the number of threads, which can be lowered in presence of other queries -- avoid settings randomization by clickhouse-test -set max_threads = 0; +set max_threads = 10; + +-- more blocks to process +set max_block_size = 10; +set min_insert_block_size_rows = 10; + +set materialized_views_ignore_errors = 1; drop table if exists testX; drop table if exists testXA; @@ -16,19 +22,27 @@ create materialized view testXA engine=MergeTree order by tuple() as select slee create materialized view testXB engine=MergeTree order by tuple() as select sleep(0.2), throwIf(A=1) from testX; create materialized view testXC engine=MergeTree order by tuple() as select sleep(0.1) from testX; + +{% set count = namespace(value=0) %} + -- { echoOn } {% for parallel_view_processing in [0, 1] %} {% for optimize_trivial_insert_select in [0, 1] %} -{% for max_insert_threads in [0, 16] %} -select 'optimize_trivial_insert_select={{ optimize_trivial_insert_select }}', 'max_insert_threads={{ max_insert_threads }}'; +{% for max_insert_threads in [0, 5] %} -insert into testX select number from numbers(10) settings +{% set count.value = count.value + 1 %} + +select 'optimize_trivial_insert_select={{ optimize_trivial_insert_select }}', 'max_insert_threads={{ max_insert_threads }}', 'iteration_num={{ iteration_num }}'; + +insert into testX select number from numbers(200) settings + send_logs_level='fatal', + insert_deduplication_token='UT_{{ count.value }}', log_queries=1, parallel_view_processing={{ parallel_view_processing }}, optimize_trivial_insert_select={{ optimize_trivial_insert_select }}, - max_insert_threads={{ max_insert_threads }}; -- { serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + max_insert_threads={{ max_insert_threads }}; system flush logs; -select arrayUniq(thread_ids) from system.query_log where +select peak_threads_usage from system.query_log where current_database = currentDatabase() and type != 'QueryStart' and query like '%insert into testX %' and @@ -38,7 +52,6 @@ select arrayUniq(thread_ids) from system.query_log where select count() from testX; select count() from testXA; -select count() from testXB; select count() from testXC; {% endfor %} {% endfor %} diff --git a/tests/queries/0_stateless/01280_ssd_complex_key_dictionary.sh b/tests/queries/0_stateless/01280_ssd_complex_key_dictionary.sh index fb7bf5c6fc1..60a65b9a253 100755 --- a/tests/queries/0_stateless/01280_ssd_complex_key_dictionary.sh +++ b/tests/queries/0_stateless/01280_ssd_complex_key_dictionary.sh @@ -5,12 +5,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT -n --query=" - set allow_deprecated_database_ordinary=1; DROP DATABASE IF EXISTS 01280_db; - CREATE DATABASE 01280_db Engine = Ordinary; + CREATE DATABASE 01280_db; DROP TABLE IF EXISTS 01280_db.table_for_dict; CREATE TABLE 01280_db.table_for_dict ( diff --git a/tests/queries/0_stateless/01281_group_by_limit_memory_tracking.sh b/tests/queries/0_stateless/01281_group_by_limit_memory_tracking.sh index e83e49dffef..33b8f413fd5 100755 --- a/tests/queries/0_stateless/01281_group_by_limit_memory_tracking.sh +++ b/tests/queries/0_stateless/01281_group_by_limit_memory_tracking.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-replicated-database, no-parallel, no-fasttest, no-tsan, no-asan, no-random-settings, no-s3-storage, no-msan +# Tags: no-replicated-database, no-parallel, no-fasttest, no-tsan, no-asan, no-random-settings, no-object-storage, no-msan # Tag no-fasttest: max_memory_usage_for_user can interfere another queries running concurrently # Regression for MemoryTracker that had been incorrectly accounted diff --git a/tests/queries/0_stateless/01293_optimize_final_force.sh b/tests/queries/0_stateless/01293_optimize_final_force.sh index d3d3d3e1ac5..e838af8af9b 100755 --- a/tests/queries/0_stateless/01293_optimize_final_force.sh +++ b/tests/queries/0_stateless/01293_optimize_final_force.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, long, no-debug, no-s3-storage +# Tags: no-fasttest, long, no-debug, no-object-storage # This test is too slow with S3 storage and debug modes. CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) diff --git a/tests/queries/0_stateless/01294_lazy_database_concurrent_recreate_reattach_and_show_tables_long.sh b/tests/queries/0_stateless/01294_lazy_database_concurrent_recreate_reattach_and_show_tables_long.sh index 3c11dc5f772..21f46a34514 100755 --- a/tests/queries/0_stateless/01294_lazy_database_concurrent_recreate_reattach_and_show_tables_long.sh +++ b/tests/queries/0_stateless/01294_lazy_database_concurrent_recreate_reattach_and_show_tables_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-parallel, no-fasttest +# Tags: long, no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -91,7 +91,7 @@ ${CLICKHOUSE_CLIENT} -n -q " " -TIMEOUT=30 +TIMEOUT=20 timeout $TIMEOUT bash -c recreate_lazy_func1 2> /dev/null & timeout $TIMEOUT bash -c recreate_lazy_func2 2> /dev/null & diff --git a/tests/queries/0_stateless/01301_aggregate_state_exception_memory_leak.sh b/tests/queries/0_stateless/01301_aggregate_state_exception_memory_leak.sh index 2d18c45406c..d74092d828d 100755 --- a/tests/queries/0_stateless/01301_aggregate_state_exception_memory_leak.sh +++ b/tests/queries/0_stateless/01301_aggregate_state_exception_memory_leak.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01302_aggregate_state_exception_memory_leak.sh b/tests/queries/0_stateless/01302_aggregate_state_exception_memory_leak.sh index b9f1f81da1a..bbf2fd9177a 100755 --- a/tests/queries/0_stateless/01302_aggregate_state_exception_memory_leak.sh +++ b/tests/queries/0_stateless/01302_aggregate_state_exception_memory_leak.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01304_direct_io_long.sh b/tests/queries/0_stateless/01304_direct_io_long.sh index 97148dc268e..2e27c2f7728 100755 --- a/tests/queries/0_stateless/01304_direct_io_long.sh +++ b/tests/queries/0_stateless/01304_direct_io_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-s3-storage-with-slow-build +# Tags: long, no-object-storage-with-slow-build CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01312_comparison_with_constant_string_in_index_analysis.sql b/tests/queries/0_stateless/01312_comparison_with_constant_string_in_index_analysis.sql index b7778dfd780..9fca9b09e1f 100644 --- a/tests/queries/0_stateless/01312_comparison_with_constant_string_in_index_analysis.sql +++ b/tests/queries/0_stateless/01312_comparison_with_constant_string_in_index_analysis.sql @@ -1,3 +1,4 @@ +SET optimize_trivial_insert_select = 1; SET merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0; DROP TABLE IF EXISTS test; diff --git a/tests/queries/0_stateless/01323_too_many_threads_bug.sql b/tests/queries/0_stateless/01323_too_many_threads_bug.sql index c377e2c7570..5bf282808c3 100644 --- a/tests/queries/0_stateless/01323_too_many_threads_bug.sql +++ b/tests/queries/0_stateless/01323_too_many_threads_bug.sql @@ -14,11 +14,11 @@ set log_queries = 1; select x from table_01323_many_parts limit 10 format Null; system flush logs; -select arrayUniq(thread_ids) <= 4 from system.query_log where current_database = currentDatabase() AND event_date >= today() - 1 and query ilike '%select x from table_01323_many_parts%' and query not like '%system.query_log%' and type = 'QueryFinish' order by query_start_time desc limit 1; +select peak_threads_usage <= 4 from system.query_log where current_database = currentDatabase() AND event_date >= today() - 1 and query ilike '%select x from table_01323_many_parts%' and query not like '%system.query_log%' and type = 'QueryFinish' order by query_start_time desc limit 1; select x from table_01323_many_parts order by x limit 10 format Null; system flush logs; -select arrayUniq(thread_ids) <= 36 from system.query_log where current_database = currentDatabase() AND event_date >= today() - 1 and query ilike '%select x from table_01323_many_parts order by x%' and query not like '%system.query_log%' and type = 'QueryFinish' order by query_start_time desc limit 1; +select peak_threads_usage <= 36 from system.query_log where current_database = currentDatabase() AND event_date >= today() - 1 and query ilike '%select x from table_01323_many_parts order by x%' and query not like '%system.query_log%' and type = 'QueryFinish' order by query_start_time desc limit 1; drop table if exists table_01323_many_parts; diff --git a/tests/queries/0_stateless/01338_long_select_and_alter.reference b/tests/queries/0_stateless/01338_long_select_and_alter.reference index c2678e7052e..027109252e1 100644 --- a/tests/queries/0_stateless/01338_long_select_and_alter.reference +++ b/tests/queries/0_stateless/01338_long_select_and_alter.reference @@ -1,3 +1,3 @@ -10 5 -CREATE TABLE default.alter_mt\n(\n `key` UInt64,\n `value` UInt64\n)\nENGINE = MergeTree\nORDER BY key\nSETTINGS index_granularity = 8192 +5 +CREATE TABLE default.alter_mt\n(\n `key` Int64,\n `value` Int64\n)\nENGINE = MergeTree\nORDER BY key\nSETTINGS index_granularity = 8192 diff --git a/tests/queries/0_stateless/01338_long_select_and_alter.sh b/tests/queries/0_stateless/01338_long_select_and_alter.sh index fcdfa2dec82..5d2759ac884 100755 --- a/tests/queries/0_stateless/01338_long_select_and_alter.sh +++ b/tests/queries/0_stateless/01338_long_select_and_alter.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-parallel +# Tags: long CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -7,16 +7,16 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS alter_mt" -$CLICKHOUSE_CLIENT --query "CREATE TABLE alter_mt (key UInt64, value String) ENGINE=MergeTree() ORDER BY key" +$CLICKHOUSE_CLIENT --query "CREATE TABLE alter_mt (key Int64, value String) ENGINE=MergeTree() ORDER BY key" -$CLICKHOUSE_CLIENT --query "INSERT INTO alter_mt SELECT number, toString(number) FROM numbers(5)" +$CLICKHOUSE_CLIENT --query "INSERT INTO alter_mt SELECT number - 1 AS x, toString(x) FROM numbers(5)" $CLICKHOUSE_CLIENT --function_sleep_max_microseconds_per_block 10000000 --query "SELECT count(distinct concat(value, '_')) FROM alter_mt WHERE not sleepEachRow(2)" & -# to be sure that select took all required locks +# To be sure that select took all required locks for better test sensitivity, although it isn't guaranteed (then the test will also succeed). sleep 2 -$CLICKHOUSE_CLIENT --query "ALTER TABLE alter_mt MODIFY COLUMN value UInt64" +$CLICKHOUSE_CLIENT --query "ALTER TABLE alter_mt MODIFY COLUMN value Int64" $CLICKHOUSE_CLIENT --query "SELECT sum(value) FROM alter_mt" diff --git a/tests/queries/0_stateless/01338_long_select_and_alter_zookeeper.reference b/tests/queries/0_stateless/01338_long_select_and_alter_zookeeper.reference index b4ed8efab63..65e638bc3a4 100644 --- a/tests/queries/0_stateless/01338_long_select_and_alter_zookeeper.reference +++ b/tests/queries/0_stateless/01338_long_select_and_alter_zookeeper.reference @@ -1,3 +1,3 @@ -10 5 -CREATE TABLE default.alter_mt\n(\n `key` UInt64,\n `value` UInt64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/01338_long_select_and_alter_zookeeper_default/alter_mt\', \'1\')\nORDER BY key\nSETTINGS index_granularity = 8192 +5 +CREATE TABLE default.alter_mt\n(\n `key` Int64,\n `value` Int64\n)\nENGINE = ReplicatedMergeTree(\'/clickhouse/tables/01338_long_select_and_alter_zookeeper_default/alter_mt\', \'1\')\nORDER BY key\nSETTINGS index_granularity = 8192 diff --git a/tests/queries/0_stateless/01338_long_select_and_alter_zookeeper.sh b/tests/queries/0_stateless/01338_long_select_and_alter_zookeeper.sh index 50ade3fad45..593a96a7cc8 100755 --- a/tests/queries/0_stateless/01338_long_select_and_alter_zookeeper.sh +++ b/tests/queries/0_stateless/01338_long_select_and_alter_zookeeper.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, zookeeper, no-parallel +# Tags: long, zookeeper CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -7,16 +7,16 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS alter_mt" -$CLICKHOUSE_CLIENT --query "CREATE TABLE alter_mt (key UInt64, value String) ENGINE=ReplicatedMergeTree('/clickhouse/tables/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/alter_mt', '1') ORDER BY key" +$CLICKHOUSE_CLIENT --query "CREATE TABLE alter_mt (key Int64, value String) ENGINE=ReplicatedMergeTree('/clickhouse/tables/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/alter_mt', '1') ORDER BY key" -$CLICKHOUSE_CLIENT --query "INSERT INTO alter_mt SELECT number, toString(number) FROM numbers(5)" +$CLICKHOUSE_CLIENT --query "INSERT INTO alter_mt SELECT number - 1 AS x, toString(x) FROM numbers(5)" $CLICKHOUSE_CLIENT --function_sleep_max_microseconds_per_block 10000000 --query "SELECT count(distinct concat(value, '_')) FROM alter_mt WHERE not sleepEachRow(2)" & -# to be sure that select took all required locks +# To be sure that select took all required locks for better test sensitivity, although it isn't guaranteed (then the test will also succeed). sleep 2 -$CLICKHOUSE_CLIENT --query "ALTER TABLE alter_mt MODIFY COLUMN value UInt64" +$CLICKHOUSE_CLIENT --query "ALTER TABLE alter_mt MODIFY COLUMN value Int64" $CLICKHOUSE_CLIENT --query "SELECT sum(value) FROM alter_mt" diff --git a/tests/queries/0_stateless/01343_min_bytes_to_use_mmap_io.sql b/tests/queries/0_stateless/01343_min_bytes_to_use_mmap_io.sql index 614629351ef..15c9ec16700 100644 --- a/tests/queries/0_stateless/01343_min_bytes_to_use_mmap_io.sql +++ b/tests/queries/0_stateless/01343_min_bytes_to_use_mmap_io.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage DROP TABLE IF EXISTS test_01343; CREATE TABLE test_01343 (x String) ENGINE = MergeTree ORDER BY tuple() SETTINGS min_bytes_for_wide_part = 0; INSERT INTO test_01343 VALUES ('Hello, world'); diff --git a/tests/queries/0_stateless/01344_min_bytes_to_use_mmap_io_index.sql b/tests/queries/0_stateless/01344_min_bytes_to_use_mmap_io_index.sql index 2e5ec563641..76cb535dcb7 100644 --- a/tests/queries/0_stateless/01344_min_bytes_to_use_mmap_io_index.sql +++ b/tests/queries/0_stateless/01344_min_bytes_to_use_mmap_io_index.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage DROP TABLE IF EXISTS test_01344; CREATE TABLE test_01344 (x String, INDEX idx (x) TYPE set(10) GRANULARITY 1) ENGINE = MergeTree ORDER BY tuple() SETTINGS min_bytes_for_wide_part = 0; INSERT INTO test_01344 VALUES ('Hello, world'); diff --git a/tests/queries/0_stateless/01355_alter_column_with_order.sql b/tests/queries/0_stateless/01355_alter_column_with_order.sql index 0b1b4c42cce..405157fd891 100644 --- a/tests/queries/0_stateless/01355_alter_column_with_order.sql +++ b/tests/queries/0_stateless/01355_alter_column_with_order.sql @@ -1,28 +1,26 @@ --- Tags: no-parallel - -DROP TABLE IF EXISTS alter_test; +DROP TABLE IF EXISTS alter_01355; set allow_deprecated_syntax_for_merge_tree=1; -CREATE TABLE alter_test (CounterID UInt32, StartDate Date, UserID UInt32, VisitID UInt32, NestedColumn Nested(A UInt8, S String), ToDrop UInt32) ENGINE = MergeTree(StartDate, intHash32(UserID), (CounterID, StartDate, intHash32(UserID), VisitID), 8192); +CREATE TABLE alter_01355 (CounterID UInt32, StartDate Date, UserID UInt32, VisitID UInt32, NestedColumn Nested(A UInt8, S String), ToDrop UInt32) ENGINE = MergeTree(StartDate, intHash32(UserID), (CounterID, StartDate, intHash32(UserID), VisitID), 8192); -ALTER TABLE alter_test ADD COLUMN Added1 UInt32 FIRST; +ALTER TABLE alter_01355 ADD COLUMN Added1 UInt32 FIRST; -ALTER TABLE alter_test ADD COLUMN Added2 UInt32 AFTER NestedColumn; +ALTER TABLE alter_01355 ADD COLUMN Added2 UInt32 AFTER NestedColumn; -ALTER TABLE alter_test ADD COLUMN Added3 UInt32 AFTER ToDrop; +ALTER TABLE alter_01355 ADD COLUMN Added3 UInt32 AFTER ToDrop; -DESC alter_test; -DETACH TABLE alter_test; -ATTACH TABLE alter_test; -DESC alter_test; +DESC alter_01355; +DETACH TABLE alter_01355; +ATTACH TABLE alter_01355; +DESC alter_01355; -ALTER TABLE alter_test MODIFY COLUMN Added2 UInt32 FIRST; +ALTER TABLE alter_01355 MODIFY COLUMN Added2 UInt32 FIRST; -ALTER TABLE alter_test MODIFY COLUMN Added3 UInt32 AFTER CounterID; +ALTER TABLE alter_01355 MODIFY COLUMN Added3 UInt32 AFTER CounterID; -DESC alter_test; -DETACH TABLE alter_test; -ATTACH TABLE alter_test; -DESC alter_test; +DESC alter_01355; +DETACH TABLE alter_01355; +ATTACH TABLE alter_01355; +DESC alter_01355; -DROP TABLE IF EXISTS alter_test; +DROP TABLE IF EXISTS alter_01355; diff --git a/tests/queries/0_stateless/01355_ilike.sql b/tests/queries/0_stateless/01355_ilike.sql index 6bde62bf47e..1ceb878a5ef 100644 --- a/tests/queries/0_stateless/01355_ilike.sql +++ b/tests/queries/0_stateless/01355_ilike.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-fasttest +-- Tags: no-fasttest SELECT 'Hello' ILIKE ''; SELECT 'Hello' ILIKE '%'; @@ -53,11 +53,7 @@ SELECT 'ощщЁё' ILIKE '%щ%'; SELECT 'ощЩЁё' ILIKE '%ё%'; SHOW TABLES NOT ILIKE '%'; -DROP DATABASE IF EXISTS test_01355; -CREATE DATABASE test_01355; -USE test_01355; CREATE TABLE test1 (x UInt8) ENGINE = Memory; CREATE TABLE test2 (x UInt8) ENGINE = Memory; SHOW TABLES ILIKE 'tES%'; SHOW TABLES NOT ILIKE 'TeS%'; -DROP DATABASE test_01355; diff --git a/tests/queries/0_stateless/01388_clear_all_columns.sql b/tests/queries/0_stateless/01388_clear_all_columns.sql index cc395aa7fb4..07b4fb3de90 100644 --- a/tests/queries/0_stateless/01388_clear_all_columns.sql +++ b/tests/queries/0_stateless/01388_clear_all_columns.sql @@ -1,5 +1,3 @@ --- Tags: no-parallel - DROP TABLE IF EXISTS test; CREATE TABLE test (x UInt8) ENGINE = MergeTree ORDER BY tuple(); INSERT INTO test (x) VALUES (1), (2), (3); diff --git a/tests/queries/0_stateless/01391_join_on_dict_crash.sql b/tests/queries/0_stateless/01391_join_on_dict_crash.sql index 854da04b334..e056e147501 100644 --- a/tests/queries/0_stateless/01391_join_on_dict_crash.sql +++ b/tests/queries/0_stateless/01391_join_on_dict_crash.sql @@ -1,13 +1,3 @@ --- Tags: no-parallel - -DROP DATABASE IF EXISTS db_01391; -CREATE DATABASE db_01391; -USE db_01391; - -DROP TABLE IF EXISTS t; -DROP TABLE IF EXISTS d_src; -DROP DICTIONARY IF EXISTS d; - CREATE TABLE t (click_city_id UInt32, click_country_id UInt32) Engine = Memory; CREATE TABLE d_src (id UInt64, country_id UInt8, name String) Engine = Memory; @@ -16,14 +6,9 @@ INSERT INTO d_src VALUES (0, 0, 'n'); CREATE DICTIONARY d (id UInt32, country_id UInt8, name String) PRIMARY KEY id -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' DB 'db_01391' table 'd_src')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' DB currentDatabase() table 'd_src')) LIFETIME(MIN 1 MAX 1) LAYOUT(HASHED()); SELECT click_country_id FROM t AS cc LEFT JOIN d ON toUInt32(d.id) = cc.click_city_id; SELECT click_country_id FROM t AS cc LEFT JOIN d ON d.country_id < 99 AND d.id = cc.click_city_id; - -DROP DICTIONARY d; -DROP TABLE t; -DROP TABLE d_src; -DROP DATABASE IF EXISTS db_01391; diff --git a/tests/queries/0_stateless/01392_column_resolve.sql b/tests/queries/0_stateless/01392_column_resolve.sql index 72b6af4576a..90a7d9b169a 100644 --- a/tests/queries/0_stateless/01392_column_resolve.sql +++ b/tests/queries/0_stateless/01392_column_resolve.sql @@ -1,16 +1,11 @@ --- Tags: no-parallel +CREATE TABLE tableConversion (conversionId String, value Nullable(Double)) ENGINE = Log(); +CREATE TABLE tableClick (clickId String, conversionId String, value Nullable(Double)) ENGINE = Log(); +CREATE TABLE leftjoin (id String) ENGINE = Log(); -DROP DATABASE IF EXISTS test_01392; -CREATE DATABASE test_01392; - -CREATE TABLE test_01392.tableConversion (conversionId String, value Nullable(Double)) ENGINE = Log(); -CREATE TABLE test_01392.tableClick (clickId String, conversionId String, value Nullable(Double)) ENGINE = Log(); -CREATE TABLE test_01392.leftjoin (id String) ENGINE = Log(); - -INSERT INTO test_01392.tableConversion(conversionId, value) VALUES ('Conversion 1', 1); -INSERT INTO test_01392.tableClick(clickId, conversionId, value) VALUES ('Click 1', 'Conversion 1', 14); -INSERT INTO test_01392.tableClick(clickId, conversionId, value) VALUES ('Click 2', 'Conversion 1', 15); -INSERT INTO test_01392.tableClick(clickId, conversionId, value) VALUES ('Click 3', 'Conversion 1', 16); +INSERT INTO tableConversion(conversionId, value) VALUES ('Conversion 1', 1); +INSERT INTO tableClick(clickId, conversionId, value) VALUES ('Click 1', 'Conversion 1', 14); +INSERT INTO tableClick(clickId, conversionId, value) VALUES ('Click 2', 'Conversion 1', 15); +INSERT INTO tableClick(clickId, conversionId, value) VALUES ('Click 3', 'Conversion 1', 16); SELECT conversion.conversionId AS myConversionId, @@ -18,19 +13,13 @@ SELECT click.myValue AS myValue FROM ( SELECT conversionId, value as myValue - FROM test_01392.tableConversion + FROM tableConversion ) AS conversion INNER JOIN ( SELECT clickId, conversionId, value as myValue - FROM test_01392.tableClick + FROM tableClick ) AS click ON click.conversionId = conversion.conversionId LEFT JOIN ( - SELECT * FROM test_01392.leftjoin + SELECT * FROM leftjoin ) AS dummy ON (dummy.id = conversion.conversionId) ORDER BY myValue; - -DROP TABLE test_01392.tableConversion; -DROP TABLE test_01392.tableClick; -DROP TABLE test_01392.leftjoin; - -DROP DATABASE test_01392; diff --git a/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh b/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh index 1c1eb4489ee..bff85b3e29f 100755 --- a/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh +++ b/tests/queries/0_stateless/01396_inactive_replica_cleanup_nodes_zookeeper.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: replica, no-debug, no-parallel +# Tags: replica, no-debug CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -30,7 +30,7 @@ $CLICKHOUSE_CLIENT --max_block_size 1 --min_insert_block_size_rows 1 --min_inser for _ in {1..60}; do $CLICKHOUSE_CLIENT --query "SYSTEM FLUSH LOGS" - [[ $($CLICKHOUSE_CLIENT --query "SELECT sum(toUInt32(extract(message, 'Removed (\d+) old log entries'))) FROM system.text_log WHERE event_date >= yesterday() AND logger_name LIKE '%' || '$CLICKHOUSE_DATABASE' || '%r1%(ReplicatedMergeTreeCleanupThread)%' AND message LIKE '%Removed % old log entries%'") -gt $((SCALE - 100)) ]] && break; + [[ $($CLICKHOUSE_CLIENT --query "SELECT sum(toUInt32(extract(message, 'Removed (\d+) old log entries'))) FROM system.text_log WHERE event_date >= yesterday() AND logger_name LIKE '%' || '$CLICKHOUSE_DATABASE' || '%r1%(ReplicatedMergeTreeCleanupThread)%' AND message LIKE '%Removed % old log entries%'") -gt $((SCALE - 10)) ]] && break; sleep 1 done diff --git a/tests/queries/0_stateless/01412_cache_dictionary_race.sh b/tests/queries/0_stateless/01412_cache_dictionary_race.sh index 165a461193d..36295ca01ea 100755 --- a/tests/queries/0_stateless/01412_cache_dictionary_race.sh +++ b/tests/queries/0_stateless/01412_cache_dictionary_race.sh @@ -54,7 +54,7 @@ function drop_create_table_thread() export -f dict_get_thread; export -f drop_create_table_thread; -TIMEOUT=30 +TIMEOUT=20 timeout $TIMEOUT bash -c dict_get_thread 2> /dev/null & timeout $TIMEOUT bash -c dict_get_thread 2> /dev/null & diff --git a/tests/queries/0_stateless/01415_sticking_mutations.sh b/tests/queries/0_stateless/01415_sticking_mutations.sh index 821c83fe728..b7c8768a65d 100755 --- a/tests/queries/0_stateless/01415_sticking_mutations.sh +++ b/tests/queries/0_stateless/01415_sticking_mutations.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-replicated-database, no-parallel +# Tags: no-replicated-database set -e diff --git a/tests/queries/0_stateless/01442_merge_detach_attach_long.sh b/tests/queries/0_stateless/01442_merge_detach_attach_long.sh index 85fdf7ed764..9ba1fe93543 100755 --- a/tests/queries/0_stateless/01442_merge_detach_attach_long.sh +++ b/tests/queries/0_stateless/01442_merge_detach_attach_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-parallel, no-debug +# Tags: long, no-debug set -e @@ -27,7 +27,7 @@ function thread_ops() } export -f thread_ops -TIMEOUT=60 +TIMEOUT=30 thread_ops $TIMEOUT & wait diff --git a/tests/queries/0_stateless/01444_create_table_drop_database_race.sh b/tests/queries/0_stateless/01444_create_table_drop_database_race.sh index eb231e71525..ae74efa4e20 100755 --- a/tests/queries/0_stateless/01444_create_table_drop_database_race.sh +++ b/tests/queries/0_stateless/01444_create_table_drop_database_race.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: race, no-parallel +# Tags: race set -e @@ -8,20 +8,20 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh # This test reproduces "Directory not empty" error in DROP DATABASE query. +export DB=test_$RANDOM function thread1() { while true; do -# ${CLICKHOUSE_CLIENT} --query="SHOW TABLES FROM test_01444" - ${CLICKHOUSE_CLIENT} --query="DROP DATABASE IF EXISTS test_01444" 2>&1| grep -F "Code: " | grep -Fv "Code: 219" - ${CLICKHOUSE_CLIENT} --query="CREATE DATABASE IF NOT EXISTS test_01444" + ${CLICKHOUSE_CLIENT} --query="DROP DATABASE IF EXISTS $DB" 2>&1| grep -F "Code: " | grep -Fv "Code: 219" + ${CLICKHOUSE_CLIENT} --query="CREATE DATABASE IF NOT EXISTS $DB" done } function thread2() { while true; do - ${CLICKHOUSE_CLIENT} --query="CREATE TABLE IF NOT EXISTS test_01444.t$RANDOM (x UInt8) ENGINE = MergeTree ORDER BY tuple()" 2>/dev/null + ${CLICKHOUSE_CLIENT} --query="CREATE TABLE IF NOT EXISTS $DB.t$RANDOM (x UInt8) ENGINE = MergeTree ORDER BY tuple()" 2>/dev/null done } @@ -36,4 +36,4 @@ timeout $TIMEOUT bash -c thread2 & wait -${CLICKHOUSE_CLIENT} --query="DROP DATABASE IF EXISTS test_01444" 2>&1| grep -F "Code: " | grep -Fv "Code: 219" || exit 0 +${CLICKHOUSE_CLIENT} --query="DROP DATABASE IF EXISTS ${DB}" 2>&1| grep -F "Code: " | grep -Fv "Code: 219" || exit 0 diff --git a/tests/queries/0_stateless/01454_storagememory_data_race_challenge.sh b/tests/queries/0_stateless/01454_storagememory_data_race_challenge.sh index d83343b3cb3..ec9c5134059 100755 --- a/tests/queries/0_stateless/01454_storagememory_data_race_challenge.sh +++ b/tests/queries/0_stateless/01454_storagememory_data_race_challenge.sh @@ -11,12 +11,17 @@ $CLICKHOUSE_CLIENT -q "DROP TABLE IF EXISTS mem" $CLICKHOUSE_CLIENT -q "CREATE TABLE mem (x UInt64) engine = Memory" function f { + local TIMELIMIT=$((SECONDS+$1)) for _ in $(seq 1 300); do $CLICKHOUSE_CLIENT -q "SELECT count() FROM (SELECT * FROM mem SETTINGS max_threads=2) FORMAT Null;" + if [ $SECONDS -ge "$TIMELIMIT" ]; then + break + fi done } function g { + local TIMELIMIT=$((SECONDS+$1)) for _ in $(seq 1 100); do $CLICKHOUSE_CLIENT -n -q " INSERT INTO mem SELECT number FROM numbers(1000000); @@ -30,14 +35,18 @@ function g { INSERT INTO mem VALUES (1); TRUNCATE TABLE mem; " + if [ $SECONDS -ge "$TIMELIMIT" ]; then + break + fi done } export -f f; export -f g; -timeout 30 bash -c f > /dev/null & -timeout 30 bash -c g > /dev/null & +TIMEOUT=20 +f $TIMEOUT & +g $TIMEOUT & wait $CLICKHOUSE_CLIENT -q "DROP TABLE mem" diff --git a/tests/queries/0_stateless/01455_optimize_trivial_insert_select.sql b/tests/queries/0_stateless/01455_optimize_trivial_insert_select.sql index 09a93d94dc3..f3a664fa3e3 100644 --- a/tests/queries/0_stateless/01455_optimize_trivial_insert_select.sql +++ b/tests/queries/0_stateless/01455_optimize_trivial_insert_select.sql @@ -3,6 +3,7 @@ SET allow_deprecated_error_prone_window_functions = 1; DROP TABLE IF EXISTS t; CREATE TABLE t (x UInt64) ENGINE = StripeLog; -- For trivial INSERT SELECT, max_threads is lowered to max_insert_threads and max_block_size is changed to min_insert_block_size_rows. +SET optimize_trivial_insert_select = 1; INSERT INTO t SELECT * FROM numbers_mt(1000000); SET max_threads = 1; -- If data was inserted by more threads, we will probably see data out of order. diff --git a/tests/queries/0_stateless/01475_read_subcolumns.sql b/tests/queries/0_stateless/01475_read_subcolumns.sql index 8d4e3cb779b..d387230d9e5 100644 --- a/tests/queries/0_stateless/01475_read_subcolumns.sql +++ b/tests/queries/0_stateless/01475_read_subcolumns.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage, no-random-settings +-- Tags: no-object-storage, no-random-settings, no-parallel SET use_uncompressed_cache = 0; diff --git a/tests/queries/0_stateless/01475_read_subcolumns_storages.sh b/tests/queries/0_stateless/01475_read_subcolumns_storages.sh index 5a30f9e0f08..f74f6755e59 100755 --- a/tests/queries/0_stateless/01475_read_subcolumns_storages.sh +++ b/tests/queries/0_stateless/01475_read_subcolumns_storages.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-s3-storage +# Tags: no-object-storage CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01487_distributed_in_not_default_db.sql b/tests/queries/0_stateless/01487_distributed_in_not_default_db.sql deleted file mode 100644 index cd027530ac8..00000000000 --- a/tests/queries/0_stateless/01487_distributed_in_not_default_db.sql +++ /dev/null @@ -1,43 +0,0 @@ --- Tags: distributed, no-parallel - -CREATE DATABASE IF NOT EXISTS shard_0; -CREATE DATABASE IF NOT EXISTS shard_1; -CREATE DATABASE IF NOT EXISTS main_01487; -CREATE DATABASE IF NOT EXISTS test_01487; - -USE main_01487; - -DROP TABLE IF EXISTS shard_0.l; -DROP TABLE IF EXISTS shard_1.l; -DROP TABLE IF EXISTS d; -DROP TABLE IF EXISTS t; - -CREATE TABLE shard_0.l (value UInt8) ENGINE = MergeTree ORDER BY value; -CREATE TABLE shard_1.l (value UInt8) ENGINE = MergeTree ORDER BY value; -CREATE TABLE t (value UInt8) ENGINE = Memory; - -INSERT INTO shard_0.l VALUES (0); -INSERT INTO shard_1.l VALUES (1); -INSERT INTO t VALUES (0), (1), (2); - -CREATE TABLE d AS t ENGINE = Distributed(test_cluster_two_shards_different_databases, currentDatabase(), t); - -USE test_01487; -DROP DATABASE test_01487; - --- After the default database is dropped QueryAnalysisPass cannot process the following SELECT query. --- That query is invalid on the initiator node. -set allow_experimental_analyzer = 0; - -SELECT * FROM main_01487.d WHERE value IN (SELECT l.value FROM l) ORDER BY value; - -USE main_01487; - -DROP TABLE IF EXISTS shard_0.l; -DROP TABLE IF EXISTS shard_1.l; -DROP TABLE IF EXISTS d; -DROP TABLE IF EXISTS t; - -DROP DATABASE shard_0; -DROP DATABASE shard_1; -DROP DATABASE main_01487; diff --git a/tests/queries/0_stateless/01502_jemalloc_percpu_arena.reference b/tests/queries/0_stateless/01502_jemalloc_percpu_arena.reference index fe093e39a56..5accb577786 100644 --- a/tests/queries/0_stateless/01502_jemalloc_percpu_arena.reference +++ b/tests/queries/0_stateless/01502_jemalloc_percpu_arena.reference @@ -1,5 +1,3 @@ -: Number of CPUs detected is not deterministic. Per-CPU arena disabled. 1 -: Number of CPUs detected is not deterministic. Per-CPU arena disabled. 100000000 1 diff --git a/tests/queries/0_stateless/01502_jemalloc_percpu_arena.sh b/tests/queries/0_stateless/01502_jemalloc_percpu_arena.sh index b3ea6eca3f4..c1bd1e0e1fa 100755 --- a/tests/queries/0_stateless/01502_jemalloc_percpu_arena.sh +++ b/tests/queries/0_stateless/01502_jemalloc_percpu_arena.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-tsan, no-asan, no-msan, no-ubsan, no-fasttest +# Tags: no-tsan, no-asan, no-msan, no-ubsan, no-fasttest, no-debug # ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ # NOTE: jemalloc is disabled under sanitizers diff --git a/tests/queries/0_stateless/01502_long_log_tinylog_deadlock_race.sh b/tests/queries/0_stateless/01502_long_log_tinylog_deadlock_race.sh index b8efee89b4a..5a146968a3b 100755 --- a/tests/queries/0_stateless/01502_long_log_tinylog_deadlock_race.sh +++ b/tests/queries/0_stateless/01502_long_log_tinylog_deadlock_race.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: deadlock +# Tags: deadlock, no-parallel set -e diff --git a/tests/queries/0_stateless/01505_pipeline_executor_UAF.sh b/tests/queries/0_stateless/01505_pipeline_executor_UAF.sh index c2750ad31b2..35c2b796570 100755 --- a/tests/queries/0_stateless/01505_pipeline_executor_UAF.sh +++ b/tests/queries/0_stateless/01505_pipeline_executor_UAF.sh @@ -14,7 +14,7 @@ for _ in {1..10}; do ${CLICKHOUSE_LOCAL} -q 'select * from numbers_mt(100000000) settings max_threads=100 FORMAT Null' # Binding to specific CPU is not required, but this makes the test more reliable. taskset --cpu-list 0 ${CLICKHOUSE_LOCAL} -q 'select * from numbers_mt(100000000) settings max_threads=100 FORMAT Null' 2>&1 | { - # build with santiziers does not have jemalloc + # build with sanitiziers does not have jemalloc # and for jemalloc we have separate test # 01502_jemalloc_percpu_arena grep -v ': Number of CPUs detected is not deterministic. Per-CPU arena disabled.' diff --git a/tests/queries/0_stateless/01508_partition_pruning_long.sh b/tests/queries/0_stateless/01508_partition_pruning_long.sh deleted file mode 100755 index 7b56d8bbf03..00000000000 --- a/tests/queries/0_stateless/01508_partition_pruning_long.sh +++ /dev/null @@ -1,30 +0,0 @@ -#!/usr/bin/env bash -# Tags: long, no-polymorphic-parts, no-random-settings, no-random-merge-tree-settings, no-debug - -# Description of test result: -# Test the correctness of the partition pruning -# -# Script executes queries from a file 01508_partition_pruning_long.queries (1 line = 1 query) -# Queries are started with 'select' (but NOT with 'SELECT') are executed with log_level=debug - -CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# shellcheck source=../shell_config.sh -. "$CURDIR"/../shell_config.sh - - -queries="${CURDIR}/01508_partition_pruning_long.queries" -while IFS= read -r sql -do - [ -z "$sql" ] && continue - if [[ "$sql" == select* ]] ; - then - echo "$sql" - ${CLICKHOUSE_CLIENT} --query "$sql" - CLICKHOUSE_CLIENT=$(echo ${CLICKHOUSE_CLIENT} | sed 's/'"--send_logs_level=${CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL}"'/--send_logs_level=debug/g') - ${CLICKHOUSE_CLIENT} --query "$sql" 2>&1 | grep -oh "Selected .* parts by partition key, *. parts by primary key, .* marks by primary key, .* marks to read from .* ranges.*$" - CLICKHOUSE_CLIENT=$(echo ${CLICKHOUSE_CLIENT} | sed 's/--send_logs_level=debug/'"--send_logs_level=${CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL}"'/g') - echo "" - else - ${CLICKHOUSE_CLIENT} --query "$sql" - fi -done < "$queries" diff --git a/tests/queries/0_stateless/01508_partition_pruning_long.reference b/tests/queries/0_stateless/01508_partition_pruning_long_1.reference similarity index 50% rename from tests/queries/0_stateless/01508_partition_pruning_long.reference rename to tests/queries/0_stateless/01508_partition_pruning_long_1.reference index afdb4257505..3ea4cc4f6ee 100644 --- a/tests/queries/0_stateless/01508_partition_pruning_long.reference +++ b/tests/queries/0_stateless/01508_partition_pruning_long_1.reference @@ -123,122 +123,3 @@ select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 2 2 20000 Selected 2/3 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges ---------- tDD ---------------------------- -select uniqExact(_part), count() from tDD where toDate(d)=toDate('2020-09-24'); -1 10000 -Selected 1/4 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges - -select uniqExact(_part), count() FROM tDD WHERE toDate(d) = toDate('2020-09-24'); -1 10000 -Selected 1/4 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges - -select uniqExact(_part), count() FROM tDD WHERE toDate(d) = '2020-09-24'; -1 10000 -Selected 1/4 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges - -select uniqExact(_part), count() FROM tDD WHERE toDate(d) >= '2020-09-23' and toDate(d) <= '2020-09-26'; -3 40000 -Selected 3/4 parts by partition key, 3 parts by primary key, 4/4 marks by primary key, 4 marks to read from 3 ranges - -select uniqExact(_part), count() FROM tDD WHERE toYYYYMMDD(d) >= 20200923 and toDate(d) <= '2020-09-26'; -3 40000 -Selected 3/4 parts by partition key, 3 parts by primary key, 4/4 marks by primary key, 4 marks to read from 3 ranges - ---------- sDD ---------------------------- -select uniqExact(_part), count() from sDD; -6 30000 -Selected 6/6 parts by partition key, 6 parts by primary key, 6/6 marks by primary key, 6 marks to read from 6 ranges - -select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC')-1)+1 = 202010; -3 9999 -Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges - -select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC')-1) = 202010; -2 9999 -Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC')-1) = 202110; -0 0 -Selected 0/6 parts by partition key, 0 parts by primary key, 0/0 marks by primary key, 0 marks to read from 0 ranges - -select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC'))+1 > 202009 and toStartOfDay(toDateTime(intDiv(d,1000),'UTC')) < toDateTime('2020-10-02 00:00:00','UTC'); -3 11440 -Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges - -select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC'))+1 > 202009 and toDateTime(intDiv(d,1000),'UTC') < toDateTime('2020-10-01 00:00:00','UTC'); -2 10000 -Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from sDD where d >= 1598918400000; -4 20000 -Selected 4/6 parts by partition key, 4 parts by primary key, 4/4 marks by primary key, 4 marks to read from 4 ranges - -select uniqExact(_part), count() from sDD where d >= 1598918400000 and toYYYYMM(toDateTime(intDiv(d,1000),'UTC')-1) < 202010; -3 10001 -Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges - ---------- xMM ---------------------------- -select uniqExact(_part), count() from xMM where toStartOfDay(d) >= '2020-10-01 00:00:00'; -2 10000 -Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00'; -3 10001 -Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges - -select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-10-01 00:00:00'; -2 10000 -Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00' and a=1; -1 1 -Selected 1/6 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges - -select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00' and a<>3; -2 5001 -Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-10-01 00:00:00' and a<>3; -1 5000 -Selected 1/6 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges - -select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-11-01 00:00:00' and a = 1; -2 10000 -Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from xMM where a = 1; -3 15000 -Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges - -select uniqExact(_part), count() from xMM where a = 66; -0 0 -Selected 0/6 parts by partition key, 0 parts by primary key, 0/0 marks by primary key, 0 marks to read from 0 ranges - -select uniqExact(_part), count() from xMM where a <> 66; -6 30000 -Selected 6/6 parts by partition key, 6 parts by primary key, 6/6 marks by primary key, 6 marks to read from 6 ranges - -select uniqExact(_part), count() from xMM where a = 2; -2 10000 -Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from xMM where a = 1; -2 15000 -Selected 2/5 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from xMM where toStartOfDay(d) >= '2020-10-01 00:00:00'; -1 10000 -Selected 1/5 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges - -select uniqExact(_part), count() from xMM where a <> 66; -5 30000 -Selected 5/5 parts by partition key, 5 parts by primary key, 5/5 marks by primary key, 5 marks to read from 5 ranges - -select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00' and a<>3; -2 5001 -Selected 2/5 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges - -select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-10-01 00:00:00' and a<>3; -1 5000 -Selected 1/5 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges - diff --git a/tests/queries/0_stateless/01508_partition_pruning_long_1.sh b/tests/queries/0_stateless/01508_partition_pruning_long_1.sh new file mode 100755 index 00000000000..512cf8f5265 --- /dev/null +++ b/tests/queries/0_stateless/01508_partition_pruning_long_1.sh @@ -0,0 +1,79 @@ +#!/usr/bin/env bash +# Tags: long, no-polymorphic-parts, no-random-settings, no-random-merge-tree-settings, no-debug + +# Description of test result: +# Test the correctness of the partition pruning +# +# Script executes queries from a file 01508_partition_pruning_long.queries (1 line = 1 query) +# Queries are started with 'select' (but NOT with 'SELECT') are executed with log_level=debug + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +while IFS= read -r sql +do + [ -z "$sql" ] && continue + if [[ "$sql" == select* ]] ; + then + echo "$sql" + ${CLICKHOUSE_CLIENT} --query "$sql" + CLICKHOUSE_CLIENT=$(echo ${CLICKHOUSE_CLIENT} | sed 's/'"--send_logs_level=${CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL}"'/--send_logs_level=debug/g') + ${CLICKHOUSE_CLIENT} --query "$sql" 2>&1 | grep -oh "Selected .* parts by partition key, *. parts by primary key, .* marks by primary key, .* marks to read from .* ranges.*$" + CLICKHOUSE_CLIENT=$(echo ${CLICKHOUSE_CLIENT} | sed 's/--send_logs_level=debug/'"--send_logs_level=${CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL}"'/g') + echo "" + else + ${CLICKHOUSE_CLIENT} --query "$sql" + fi +done <<< " +DROP TABLE IF EXISTS tMM; + +CREATE TABLE tMM(d DateTime('Asia/Istanbul'), a Int64) ENGINE = MergeTree PARTITION BY toYYYYMM(d) ORDER BY tuple() SETTINGS index_granularity = 8192; +SYSTEM STOP MERGES tMM; +INSERT INTO tMM SELECT toDateTime('2020-08-16 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); +INSERT INTO tMM SELECT toDateTime('2020-08-16 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); +INSERT INTO tMM SELECT toDateTime('2020-09-01 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); +INSERT INTO tMM SELECT toDateTime('2020-09-01 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); +INSERT INTO tMM SELECT toDateTime('2020-10-01 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); +INSERT INTO tMM SELECT toDateTime('2020-10-15 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); + +SELECT '--------- tMM ----------------------------'; +select uniqExact(_part), count() from tMM where toDate(d)=toDate('2020-09-15'); +select uniqExact(_part), count() from tMM where toDate(d)=toDate('2020-09-01'); +select uniqExact(_part), count() from tMM where toDate(d)=toDate('2020-10-15'); +select uniqExact(_part), count() from tMM where toDate(d)='2020-09-15'; +select uniqExact(_part), count() from tMM where toYYYYMM(d)=202009; +select uniqExact(_part), count() from tMM where toYYYYMMDD(d)=20200816; +select uniqExact(_part), count() from tMM where toYYYYMMDD(d)=20201015; +select uniqExact(_part), count() from tMM where toDate(d)='2020-10-15'; +select uniqExact(_part), count() from tMM where d >= '2020-09-01 00:00:00' and d<'2020-10-15 00:00:00'; +select uniqExact(_part), count() from tMM where d >= '2020-01-16 00:00:00' and d < toDateTime('2021-08-17 00:00:00', 'Asia/Istanbul'); +select uniqExact(_part), count() from tMM where d >= '2020-09-16 00:00:00' and d < toDateTime('2020-10-01 00:00:00', 'Asia/Istanbul'); +select uniqExact(_part), count() from tMM where d >= '2020-09-12 00:00:00' and d < '2020-10-16 00:00:00'; +select uniqExact(_part), count() from tMM where toStartOfDay(d) >= '2020-09-12 00:00:00'; +select uniqExact(_part), count() from tMM where toStartOfDay(d) = '2020-09-01 00:00:00'; +select uniqExact(_part), count() from tMM where toStartOfDay(d) = '2020-10-01 00:00:00'; +select uniqExact(_part), count() from tMM where toStartOfDay(d) >= '2020-09-15 00:00:00' and d < '2020-10-16 00:00:00'; +select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 202010; +select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 202009; +select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 202010 and toStartOfDay(d) = '2020-10-01 00:00:00'; +select uniqExact(_part), count() from tMM where toYYYYMM(d) >= 202009 and toStartOfDay(d) < '2020-10-02 00:00:00'; +select uniqExact(_part), count() from tMM where toYYYYMM(d) > 202009 and toStartOfDay(d) < '2020-10-02 00:00:00'; +select uniqExact(_part), count() from tMM where toYYYYMM(d)+1 > 202009 and toStartOfDay(d) < '2020-10-02 00:00:00'; +select uniqExact(_part), count() from tMM where toYYYYMM(d)+1 > 202010 and toStartOfDay(d) < '2020-10-02 00:00:00'; +select uniqExact(_part), count() from tMM where toYYYYMM(d)+1 > 202010; +select uniqExact(_part), count() from tMM where toYYYYMM(d-1)+1 = 202010; +select uniqExact(_part), count() from tMM where toStartOfMonth(d) >= '2020-09-15'; +select uniqExact(_part), count() from tMM where toStartOfMonth(d) >= '2020-09-01'; +select uniqExact(_part), count() from tMM where toStartOfMonth(d) >= '2020-09-01' and toStartOfMonth(d) < '2020-10-01'; + +SYSTEM START MERGES tMM; +OPTIMIZE TABLE tMM FINAL; + +select uniqExact(_part), count() from tMM where toYYYYMM(d-1)+1 = 202010; +select uniqExact(_part), count() from tMM where toYYYYMM(d)+1 > 202010; +select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 202010; + +DROP TABLE tMM; +" diff --git a/tests/queries/0_stateless/01508_partition_pruning_long_2.reference b/tests/queries/0_stateless/01508_partition_pruning_long_2.reference new file mode 100644 index 00000000000..bc767f17031 --- /dev/null +++ b/tests/queries/0_stateless/01508_partition_pruning_long_2.reference @@ -0,0 +1,119 @@ +--------- tDD ---------------------------- +select uniqExact(_part), count() from tDD where toDate(d)=toDate('2020-09-24'); +1 10000 +Selected 1/4 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges + +select uniqExact(_part), count() FROM tDD WHERE toDate(d) = toDate('2020-09-24'); +1 10000 +Selected 1/4 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges + +select uniqExact(_part), count() FROM tDD WHERE toDate(d) = '2020-09-24'; +1 10000 +Selected 1/4 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges + +select uniqExact(_part), count() FROM tDD WHERE toDate(d) >= '2020-09-23' and toDate(d) <= '2020-09-26'; +3 40000 +Selected 3/4 parts by partition key, 3 parts by primary key, 4/4 marks by primary key, 4 marks to read from 3 ranges + +select uniqExact(_part), count() FROM tDD WHERE toYYYYMMDD(d) >= 20200923 and toDate(d) <= '2020-09-26'; +3 40000 +Selected 3/4 parts by partition key, 3 parts by primary key, 4/4 marks by primary key, 4 marks to read from 3 ranges + +--------- sDD ---------------------------- +select uniqExact(_part), count() from sDD; +6 30000 +Selected 6/6 parts by partition key, 6 parts by primary key, 6/6 marks by primary key, 6 marks to read from 6 ranges + +select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC')-1)+1 = 202010; +3 9999 +Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges + +select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC')-1) = 202010; +2 9999 +Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC')-1) = 202110; +0 0 +Selected 0/6 parts by partition key, 0 parts by primary key, 0/0 marks by primary key, 0 marks to read from 0 ranges + +select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC'))+1 > 202009 and toStartOfDay(toDateTime(intDiv(d,1000),'UTC')) < toDateTime('2020-10-02 00:00:00','UTC'); +3 11440 +Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges + +select uniqExact(_part), count() from sDD where toYYYYMM(toDateTime(intDiv(d,1000),'UTC'))+1 > 202009 and toDateTime(intDiv(d,1000),'UTC') < toDateTime('2020-10-01 00:00:00','UTC'); +2 10000 +Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from sDD where d >= 1598918400000; +4 20000 +Selected 4/6 parts by partition key, 4 parts by primary key, 4/4 marks by primary key, 4 marks to read from 4 ranges + +select uniqExact(_part), count() from sDD where d >= 1598918400000 and toYYYYMM(toDateTime(intDiv(d,1000),'UTC')-1) < 202010; +3 10001 +Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges + +--------- xMM ---------------------------- +select uniqExact(_part), count() from xMM where toStartOfDay(d) >= '2020-10-01 00:00:00'; +2 10000 +Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00'; +3 10001 +Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges + +select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-10-01 00:00:00'; +2 10000 +Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00' and a=1; +1 1 +Selected 1/6 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges + +select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00' and a<>3; +2 5001 +Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-10-01 00:00:00' and a<>3; +1 5000 +Selected 1/6 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges + +select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-11-01 00:00:00' and a = 1; +2 10000 +Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from xMM where a = 1; +3 15000 +Selected 3/6 parts by partition key, 3 parts by primary key, 3/3 marks by primary key, 3 marks to read from 3 ranges + +select uniqExact(_part), count() from xMM where a = 66; +0 0 +Selected 0/6 parts by partition key, 0 parts by primary key, 0/0 marks by primary key, 0 marks to read from 0 ranges + +select uniqExact(_part), count() from xMM where a <> 66; +6 30000 +Selected 6/6 parts by partition key, 6 parts by primary key, 6/6 marks by primary key, 6 marks to read from 6 ranges + +select uniqExact(_part), count() from xMM where a = 2; +2 10000 +Selected 2/6 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from xMM where a = 1; +2 15000 +Selected 2/5 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from xMM where toStartOfDay(d) >= '2020-10-01 00:00:00'; +1 10000 +Selected 1/5 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges + +select uniqExact(_part), count() from xMM where a <> 66; +5 30000 +Selected 5/5 parts by partition key, 5 parts by primary key, 5/5 marks by primary key, 5 marks to read from 5 ranges + +select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00' and a<>3; +2 5001 +Selected 2/5 parts by partition key, 2 parts by primary key, 2/2 marks by primary key, 2 marks to read from 2 ranges + +select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-10-01 00:00:00' and a<>3; +1 5000 +Selected 1/5 parts by partition key, 1 parts by primary key, 1/1 marks by primary key, 1 marks to read from 1 ranges + diff --git a/tests/queries/0_stateless/01508_partition_pruning_long.queries b/tests/queries/0_stateless/01508_partition_pruning_long_2.sh old mode 100644 new mode 100755 similarity index 58% rename from tests/queries/0_stateless/01508_partition_pruning_long.queries rename to tests/queries/0_stateless/01508_partition_pruning_long_2.sh index 0d64fc05f0f..3f8a89bdb20 --- a/tests/queries/0_stateless/01508_partition_pruning_long.queries +++ b/tests/queries/0_stateless/01508_partition_pruning_long_2.sh @@ -1,15 +1,35 @@ -DROP TABLE IF EXISTS tMM; +#!/usr/bin/env bash +# Tags: long, no-polymorphic-parts, no-random-settings, no-random-merge-tree-settings, no-debug + +# Description of test result: +# Test the correctness of the partition pruning +# +# Script executes queries from a file 01508_partition_pruning_long.queries (1 line = 1 query) +# Queries are started with 'select' (but NOT with 'SELECT') are executed with log_level=debug + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +while IFS= read -r sql +do + [ -z "$sql" ] && continue + if [[ "$sql" == select* ]] ; + then + echo "$sql" + ${CLICKHOUSE_CLIENT} --query "$sql" + CLICKHOUSE_CLIENT=$(echo ${CLICKHOUSE_CLIENT} | sed 's/'"--send_logs_level=${CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL}"'/--send_logs_level=debug/g') + ${CLICKHOUSE_CLIENT} --query "$sql" 2>&1 | grep -oh "Selected .* parts by partition key, *. parts by primary key, .* marks by primary key, .* marks to read from .* ranges.*$" + CLICKHOUSE_CLIENT=$(echo ${CLICKHOUSE_CLIENT} | sed 's/--send_logs_level=debug/'"--send_logs_level=${CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL}"'/g') + echo "" + else + ${CLICKHOUSE_CLIENT} --query "$sql" + fi +done <<< " DROP TABLE IF EXISTS tDD; DROP TABLE IF EXISTS sDD; DROP TABLE IF EXISTS xMM; -CREATE TABLE tMM(d DateTime('Asia/Istanbul'), a Int64) ENGINE = MergeTree PARTITION BY toYYYYMM(d) ORDER BY tuple() SETTINGS index_granularity = 8192; -SYSTEM STOP MERGES tMM; -INSERT INTO tMM SELECT toDateTime('2020-08-16 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); -INSERT INTO tMM SELECT toDateTime('2020-08-16 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); -INSERT INTO tMM SELECT toDateTime('2020-09-01 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); -INSERT INTO tMM SELECT toDateTime('2020-09-01 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); -INSERT INTO tMM SELECT toDateTime('2020-10-01 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); -INSERT INTO tMM SELECT toDateTime('2020-10-15 00:00:00', 'Asia/Istanbul') + number*60, number FROM numbers(5000); CREATE TABLE tDD(d DateTime('Asia/Istanbul'),a Int) ENGINE = MergeTree PARTITION BY toYYYYMMDD(d) ORDER BY tuple() SETTINGS index_granularity = 8192; SYSTEM STOP MERGES tDD; @@ -34,44 +54,6 @@ INSERT INTO xMM SELECT toDateTime('2020-10-01 00:00:00', 'Asia/Istanbul') + numb INSERT INTO xMM SELECT toDateTime('2020-10-15 00:00:00', 'Asia/Istanbul') + number*60, 1, number FROM numbers(5000); -SELECT '--------- tMM ----------------------------'; -select uniqExact(_part), count() from tMM where toDate(d)=toDate('2020-09-15'); -select uniqExact(_part), count() from tMM where toDate(d)=toDate('2020-09-01'); -select uniqExact(_part), count() from tMM where toDate(d)=toDate('2020-10-15'); -select uniqExact(_part), count() from tMM where toDate(d)='2020-09-15'; -select uniqExact(_part), count() from tMM where toYYYYMM(d)=202009; -select uniqExact(_part), count() from tMM where toYYYYMMDD(d)=20200816; -select uniqExact(_part), count() from tMM where toYYYYMMDD(d)=20201015; -select uniqExact(_part), count() from tMM where toDate(d)='2020-10-15'; -select uniqExact(_part), count() from tMM where d >= '2020-09-01 00:00:00' and d<'2020-10-15 00:00:00'; -select uniqExact(_part), count() from tMM where d >= '2020-01-16 00:00:00' and d < toDateTime('2021-08-17 00:00:00', 'Asia/Istanbul'); -select uniqExact(_part), count() from tMM where d >= '2020-09-16 00:00:00' and d < toDateTime('2020-10-01 00:00:00', 'Asia/Istanbul'); -select uniqExact(_part), count() from tMM where d >= '2020-09-12 00:00:00' and d < '2020-10-16 00:00:00'; -select uniqExact(_part), count() from tMM where toStartOfDay(d) >= '2020-09-12 00:00:00'; -select uniqExact(_part), count() from tMM where toStartOfDay(d) = '2020-09-01 00:00:00'; -select uniqExact(_part), count() from tMM where toStartOfDay(d) = '2020-10-01 00:00:00'; -select uniqExact(_part), count() from tMM where toStartOfDay(d) >= '2020-09-15 00:00:00' and d < '2020-10-16 00:00:00'; -select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 202010; -select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 202009; -select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 202010 and toStartOfDay(d) = '2020-10-01 00:00:00'; -select uniqExact(_part), count() from tMM where toYYYYMM(d) >= 202009 and toStartOfDay(d) < '2020-10-02 00:00:00'; -select uniqExact(_part), count() from tMM where toYYYYMM(d) > 202009 and toStartOfDay(d) < '2020-10-02 00:00:00'; -select uniqExact(_part), count() from tMM where toYYYYMM(d)+1 > 202009 and toStartOfDay(d) < '2020-10-02 00:00:00'; -select uniqExact(_part), count() from tMM where toYYYYMM(d)+1 > 202010 and toStartOfDay(d) < '2020-10-02 00:00:00'; -select uniqExact(_part), count() from tMM where toYYYYMM(d)+1 > 202010; -select uniqExact(_part), count() from tMM where toYYYYMM(d-1)+1 = 202010; -select uniqExact(_part), count() from tMM where toStartOfMonth(d) >= '2020-09-15'; -select uniqExact(_part), count() from tMM where toStartOfMonth(d) >= '2020-09-01'; -select uniqExact(_part), count() from tMM where toStartOfMonth(d) >= '2020-09-01' and toStartOfMonth(d) < '2020-10-01'; - -SYSTEM START MERGES tMM; -OPTIMIZE TABLE tMM FINAL; - -select uniqExact(_part), count() from tMM where toYYYYMM(d-1)+1 = 202010; -select uniqExact(_part), count() from tMM where toYYYYMM(d)+1 > 202010; -select uniqExact(_part), count() from tMM where toYYYYMM(d) between 202009 and 202010; - - SELECT '--------- tDD ----------------------------'; SYSTEM START MERGES tDD; OPTIMIZE TABLE tDD FINAL; @@ -116,9 +98,7 @@ select uniqExact(_part), count() from xMM where a <> 66; select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d <= '2020-10-01 00:00:00' and a<>3; select uniqExact(_part), count() from xMM where d >= '2020-09-01 00:00:00' and d < '2020-10-01 00:00:00' and a<>3; -DROP TABLE tMM; DROP TABLE tDD; DROP TABLE sDD; DROP TABLE xMM; - - +" diff --git a/tests/queries/0_stateless/01508_race_condition_rename_clear_zookeeper_long.sh b/tests/queries/0_stateless/01508_race_condition_rename_clear_zookeeper_long.sh index c3c87eeaf8b..6098c826e32 100755 --- a/tests/queries/0_stateless/01508_race_condition_rename_clear_zookeeper_long.sh +++ b/tests/queries/0_stateless/01508_race_condition_rename_clear_zookeeper_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: race, zookeeper, no-s3-storage +# Tags: race, zookeeper, no-object-storage CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01532_execute_merges_on_single_replica_long.sql b/tests/queries/0_stateless/01532_execute_merges_on_single_replica_long.sql index 49ef9d8b79f..e53f4476ec6 100644 --- a/tests/queries/0_stateless/01532_execute_merges_on_single_replica_long.sql +++ b/tests/queries/0_stateless/01532_execute_merges_on_single_replica_long.sql @@ -1,4 +1,4 @@ --- Tags: long, replica, no-replicated-database, no-parallel, no-s3-storage +-- Tags: long, replica, no-replicated-database, no-parallel, no-object-storage -- Tag no-replicated-database: Fails due to additional replicas or shards -- Tag no-parallel: static zk path diff --git a/tests/queries/0_stateless/01533_multiple_nested.sql b/tests/queries/0_stateless/01533_multiple_nested.sql index 1a6f0ec395e..4756bd4e919 100644 --- a/tests/queries/0_stateless/01533_multiple_nested.sql +++ b/tests/queries/0_stateless/01533_multiple_nested.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage, no-random-merge-tree-settings +-- Tags: no-object-storage, no-random-merge-tree-settings, no-parallel -- no-s3 because read FileOpen metric DROP TABLE IF EXISTS nested; diff --git a/tests/queries/0_stateless/01543_avro_deserialization_with_lc.sh b/tests/queries/0_stateless/01543_avro_deserialization_with_lc.sh index 697b32a77ae..a5697a62dc2 100755 --- a/tests/queries/0_stateless/01543_avro_deserialization_with_lc.sh +++ b/tests/queries/0_stateless/01543_avro_deserialization_with_lc.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT --multiquery --query " SET allow_suspicious_low_cardinality_types=1; CREATE TABLE IF NOT EXISTS test_01543 (value LowCardinality(String), value2 LowCardinality(UInt64)) ENGINE=Memory(); diff --git a/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.sql b/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.sql index c202ad349d6..b5ece08196e 100644 --- a/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.sql +++ b/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage, no-random-merge-tree-settings +-- Tags: no-object-storage, no-random-merge-tree-settings DROP TABLE IF EXISTS data_01551; @@ -10,6 +10,7 @@ PARTITION BY key%2 ORDER BY (key, key/2) SETTINGS index_granularity=10, index_granularity_bytes='10Mi'; +SET optimize_trivial_insert_select = 1; INSERT INTO data_01551 SELECT number FROM numbers(100000); SET max_threads=3; SET merge_tree_min_rows_for_concurrent_read=10000; diff --git a/tests/queries/0_stateless/01552_dict_fixedstring.sql b/tests/queries/0_stateless/01552_dict_fixedstring.sql index 01d55656e3c..0b19c9980a4 100644 --- a/tests/queries/0_stateless/01552_dict_fixedstring.sql +++ b/tests/queries/0_stateless/01552_dict_fixedstring.sql @@ -16,5 +16,5 @@ LIFETIME(MIN 10 MAX 10); SELECT dictGet(currentDatabase() || '.dict', 's', number) FROM numbers(2); -DROP TABLE src; DROP DICTIONARY dict; +DROP TABLE src; diff --git a/tests/queries/0_stateless/01563_distributed_query_finish.reference b/tests/queries/0_stateless/01563_distributed_query_finish.reference index c3688b553c4..b48979a492e 100644 --- a/tests/queries/0_stateless/01563_distributed_query_finish.reference +++ b/tests/queries/0_stateless/01563_distributed_query_finish.reference @@ -1,2 +1 @@ -1,0 NETWORK_ERROR=0 diff --git a/tests/queries/0_stateless/01563_distributed_query_finish.sh b/tests/queries/0_stateless/01563_distributed_query_finish.sh index 0019c714e40..e3c5928f108 100755 --- a/tests/queries/0_stateless/01563_distributed_query_finish.sh +++ b/tests/queries/0_stateless/01563_distributed_query_finish.sh @@ -19,20 +19,25 @@ create table dist_01247 as data_01247 engine=Distributed(test_cluster_two_shards select * from dist_01247 format Null; EOL -network_errors_before=$($CLICKHOUSE_CLIENT -q "SELECT value FROM system.errors WHERE name = 'NETWORK_ERROR'") +# NOTE: it is possible to got NETWORK_ERROR even with no-parallel, at least due to system.*_log_sender to the cloud +for ((i = 0; i < 100; ++i)); do + network_errors_before=$($CLICKHOUSE_CLIENT -q "SELECT value FROM system.errors WHERE name = 'NETWORK_ERROR'") -opts=( - "--max_distributed_connections=1" - "--optimize_skip_unused_shards=1" - "--optimize_distributed_group_by_sharding_key=1" - "--prefer_localhost_replica=0" -) -$CLICKHOUSE_CLIENT "${opts[@]}" --format CSV -nm < /dev/null & timeout $TIMEOUT bash -c kill_mutation_thread 2> /dev/null & diff --git a/tests/queries/0_stateless/01600_parts_states_metrics_long.reference b/tests/queries/0_stateless/01600_parts_states_metrics_long.reference new file mode 100644 index 00000000000..98fb6a68656 --- /dev/null +++ b/tests/queries/0_stateless/01600_parts_states_metrics_long.reference @@ -0,0 +1,4 @@ +1 +1 +1 +1 diff --git a/tests/queries/0_stateless/01600_parts_states_metrics_long.sh b/tests/queries/0_stateless/01600_parts_states_metrics_long.sh new file mode 100755 index 00000000000..2e47034e528 --- /dev/null +++ b/tests/queries/0_stateless/01600_parts_states_metrics_long.sh @@ -0,0 +1,40 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +# NOTE: database = $CLICKHOUSE_DATABASE is unwanted +verify_sql="SELECT + (SELECT sumIf(value, metric = 'PartsActive'), sumIf(value, metric = 'PartsOutdated') FROM system.metrics) + = (SELECT sum(active), sum(NOT active) FROM + (SELECT active FROM system.parts UNION ALL SELECT active FROM system.projection_parts UNION ALL SELECT 1 FROM system.dropped_tables_parts))" + +# The query is not atomic - it can compare states between system.parts and system.metrics from different points in time. +# So, there is inherent race condition. But it should get expected result eventually. +# In case of test failure, this code will do infinite loop and timeout. +verify() +{ + while true + do + result=$( $CLICKHOUSE_CLIENT -m --query="$verify_sql" ) + [ "$result" = "1" ] && break + sleep 0.1 + done + echo 1 +} + +$CLICKHOUSE_CLIENT --database_atomic_wait_for_drop_and_detach_synchronously=1 --query="DROP TABLE IF EXISTS test_table" +$CLICKHOUSE_CLIENT --query="CREATE TABLE test_table(data Date) ENGINE = MergeTree PARTITION BY toYear(data) ORDER BY data;" + +$CLICKHOUSE_CLIENT --query="INSERT INTO test_table VALUES ('1992-01-01')" +verify + +$CLICKHOUSE_CLIENT --query="INSERT INTO test_table VALUES ('1992-01-02')" +verify + +$CLICKHOUSE_CLIENT --query="OPTIMIZE TABLE test_table FINAL" +verify + +$CLICKHOUSE_CLIENT --database_atomic_wait_for_drop_and_detach_synchronously=1 --query="DROP TABLE test_table" +verify diff --git a/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql b/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql index 694e961bc4a..9b96ce3e586 100644 --- a/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql +++ b/tests/queries/0_stateless/01605_adaptive_granularity_block_borders.sql @@ -1,6 +1,6 @@ --- Tags: no-random-merge-tree-settings, no-tsan, no-debug, no-s3-storage +-- Tags: no-random-merge-tree-settings, no-tsan, no-debug, no-object-storage -- no-tsan: too slow --- no-s3-storage: for remote tables we use thread pool even when reading with one stream, so memory consumption is higher +-- no-object-storage: for remote tables we use thread pool even when reading with one stream, so memory consumption is higher SET use_uncompressed_cache = 0; SET allow_prefetched_read_pool_for_remote_filesystem=0; @@ -24,6 +24,7 @@ enable_vertical_merge_algorithm = 0; SET max_block_size=900; -- There are about 900 marks for our settings. +SET optimize_trivial_insert_select = 1; INSERT INTO adaptive_table SELECT number, if(number > 700, randomPrintableASCII(102400), randomPrintableASCII(1)) FROM numbers(10000); OPTIMIZE TABLE adaptive_table FINAL; diff --git a/tests/queries/0_stateless/01623_constraints_column_swap.sql b/tests/queries/0_stateless/01623_constraints_column_swap.sql index 3219ee3cda7..242be87938d 100644 --- a/tests/queries/0_stateless/01623_constraints_column_swap.sql +++ b/tests/queries/0_stateless/01623_constraints_column_swap.sql @@ -5,6 +5,7 @@ SET optimize_using_constraints = 1; SET optimize_move_to_prewhere = 1; SET optimize_substitute_columns = 1; SET optimize_append_index = 1; +SET optimize_trivial_insert_select = 1; DROP TABLE IF EXISTS column_swap_test_test; diff --git a/tests/queries/0_stateless/01643_merge_tree_fsync_smoke.sql b/tests/queries/0_stateless/01643_merge_tree_fsync_smoke.sql index dfc761e1764..f7622bcf98f 100644 --- a/tests/queries/0_stateless/01643_merge_tree_fsync_smoke.sql +++ b/tests/queries/0_stateless/01643_merge_tree_fsync_smoke.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage drop table if exists data_01643; diff --git a/tests/queries/0_stateless/01643_replicated_merge_tree_fsync_smoke.sql b/tests/queries/0_stateless/01643_replicated_merge_tree_fsync_smoke.sql index 54c30fa2b1a..992cc687c88 100644 --- a/tests/queries/0_stateless/01643_replicated_merge_tree_fsync_smoke.sql +++ b/tests/queries/0_stateless/01643_replicated_merge_tree_fsync_smoke.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-s3-storage +-- Tags: no-parallel, no-object-storage -- no-parallel -- for flaky check and to avoid "Removing leftovers from table" (for other tables) -- Temporarily skip warning 'table was created by another server at the same moment, will retry' diff --git a/tests/queries/0_stateless/01651_lc_insert_tiny_log.reference b/tests/queries/0_stateless/01651_lc_insert_tiny_log.reference index 3da44c57b27..5cc8909b6c8 100644 --- a/tests/queries/0_stateless/01651_lc_insert_tiny_log.reference +++ b/tests/queries/0_stateless/01651_lc_insert_tiny_log.reference @@ -1,12 +1,12 @@ 10000000 10000000 1274991808 -20000000 -20000000 2549983616 +30000000 +30000000 3824991808 10000000 10000000 1274991808 -20000000 -20000000 2549983616 +30000000 +30000000 3824991808 10000000 10000000 1274991808 -20000000 -20000000 2549983616 +30000000 +30000000 3824991808 diff --git a/tests/queries/0_stateless/01651_lc_insert_tiny_log.sql b/tests/queries/0_stateless/01651_lc_insert_tiny_log.sql index d405bb01fd9..d11c9120c61 100644 --- a/tests/queries/0_stateless/01651_lc_insert_tiny_log.sql +++ b/tests/queries/0_stateless/01651_lc_insert_tiny_log.sql @@ -8,7 +8,7 @@ INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000); select sum(length(arr)) from perf_lc_num; select sum(length(arr)), sum(num) from perf_lc_num; -INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000); +INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000, 20000000); select sum(length(arr)) from perf_lc_num; select sum(length(arr)), sum(num) from perf_lc_num; @@ -23,7 +23,7 @@ INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000); select sum(length(arr)) from perf_lc_num; select sum(length(arr)), sum(num) from perf_lc_num; -INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000); +INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000, 20000000); select sum(length(arr)) from perf_lc_num; select sum(length(arr)), sum(num) from perf_lc_num; @@ -38,7 +38,7 @@ INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000); select sum(length(arr)) from perf_lc_num; select sum(length(arr)), sum(num) from perf_lc_num; -INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000); +INSERT INTO perf_lc_num (num) SELECT toUInt8(number) FROM numbers(10000000, 20000000); select sum(length(arr)) from perf_lc_num; select sum(length(arr)), sum(num) from perf_lc_num; diff --git a/tests/queries/0_stateless/01655_plan_optimizations.reference b/tests/queries/0_stateless/01655_plan_optimizations.reference index a6af1f2170d..edf93b4b39f 100644 --- a/tests/queries/0_stateless/01655_plan_optimizations.reference +++ b/tests/queries/0_stateless/01655_plan_optimizations.reference @@ -163,6 +163,7 @@ Filter column: notEquals(__table1.y, 2_UInt8) > filter is pushed down before CreatingSets CreatingSets Filter +Filter 1 3 > one condition of filter is pushed down before LEFT JOIN diff --git a/tests/queries/0_stateless/01655_plan_optimizations_merge_filters.sql b/tests/queries/0_stateless/01655_plan_optimizations_merge_filters.sql index 1301135b4cb..2193fc7a8f4 100644 --- a/tests/queries/0_stateless/01655_plan_optimizations_merge_filters.sql +++ b/tests/queries/0_stateless/01655_plan_optimizations_merge_filters.sql @@ -1,3 +1,5 @@ +set query_plan_merge_filters=1; + set allow_experimental_analyzer=1; select explain from (explain actions = 1 select * from (select sum(number) as v, bitAnd(number, 15) as key from numbers(1e8) group by key having v != 0) where key = 7) where explain like '%Filter%' or explain like '%Aggregating%'; diff --git a/tests/queries/0_stateless/01658_read_file_to_stringcolumn.sh b/tests/queries/0_stateless/01658_read_file_to_stringcolumn.sh index 96327536f89..685fe69642a 100755 --- a/tests/queries/0_stateless/01658_read_file_to_stringcolumn.sh +++ b/tests/queries/0_stateless/01658_read_file_to_stringcolumn.sh @@ -7,18 +7,11 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. - -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -CLICKHOUSE_USER_FILES_PATH=$(clickhouse-client --query "select _path, _file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -mkdir -p ${CLICKHOUSE_USER_FILES_PATH}/ -echo -n aaaaaaaaa > ${CLICKHOUSE_USER_FILES_PATH}/a.txt -echo -n bbbbbbbbb > ${CLICKHOUSE_USER_FILES_PATH}/b.txt -echo -n ccccccccc > ${CLICKHOUSE_USER_FILES_PATH}/c.txt +echo -n aaaaaaaaa > ${USER_FILES_PATH}/a.txt +echo -n bbbbbbbbb > ${USER_FILES_PATH}/b.txt +echo -n ccccccccc > ${USER_FILES_PATH}/c.txt echo -n ccccccccc > /tmp/c.txt -mkdir -p ${CLICKHOUSE_USER_FILES_PATH}/dir +mkdir -p ${USER_FILES_PATH}/dir ### 1st TEST in CLIENT mode. @@ -85,15 +78,15 @@ echo "${CLICKHOUSE_LOCAL} --query "'"select file('"'dir'), file('b.txt')"'";echo # Test that the function is not injective -echo -n Hello > ${CLICKHOUSE_USER_FILES_PATH}/a -echo -n Hello > ${CLICKHOUSE_USER_FILES_PATH}/b -echo -n World > ${CLICKHOUSE_USER_FILES_PATH}/c +echo -n Hello > ${USER_FILES_PATH}/a +echo -n Hello > ${USER_FILES_PATH}/b +echo -n World > ${USER_FILES_PATH}/c ${CLICKHOUSE_CLIENT} --query "SELECT file(arrayJoin(['a', 'b', 'c'])) AS s, count() GROUP BY s ORDER BY s" ${CLICKHOUSE_CLIENT} --query "SELECT s, count() FROM file('?', TSV, 's String') GROUP BY s ORDER BY s" # Restore -rm ${CLICKHOUSE_USER_FILES_PATH}/{a,b,c}.txt -rm ${CLICKHOUSE_USER_FILES_PATH}/{a,b,c} +rm ${USER_FILES_PATH}/{a,b,c}.txt +rm ${USER_FILES_PATH}/{a,b,c} rm /tmp/c.txt -rm -rf ${CLICKHOUSE_USER_FILES_PATH}/dir +rm -rf ${USER_FILES_PATH}/dir diff --git a/tests/queries/0_stateless/01676_clickhouse_client_autocomplete.python b/tests/queries/0_stateless/01676_clickhouse_client_autocomplete.python index 9072dfeb09f..f363cb64018 100644 --- a/tests/queries/0_stateless/01676_clickhouse_client_autocomplete.python +++ b/tests/queries/0_stateless/01676_clickhouse_client_autocomplete.python @@ -10,18 +10,38 @@ DEBUG_LOG = os.path.join( os.path.basename(os.path.abspath(__file__)).strip(".python") + ".debuglog", ) +STATE_MAP = { + -1: "process did not start", + 0: "completion was found", + 1: "process started and said ':)'", + 2: "completion search was started", + 3: "completion is missing", +} + def run_with_timeout(func, args, timeout): - process = multiprocessing.Process(target=func, args=args) - process.start() - process.join(timeout) + for _ in range(5): + state = multiprocessing.Value("i", -1) + process = multiprocessing.Process( + target=func, args=args, kwargs={"state": state} + ) + process.start() + process.join(timeout) - if process.is_alive(): - process.terminate() - print("Timeout") + if state.value in (0, 3): + return + + if process.is_alive(): + process.terminate() + + if state.value == -1: + continue + + print(f"Timeout, state: {STATE_MAP[state.value]}") + return -def test_completion(program, argv, comp_word): +def test_completion(program, argv, comp_word, state=None): comp_begin = comp_word[:-3] shell_pid, master = pty.fork() @@ -41,6 +61,8 @@ def test_completion(program, argv, comp_word): debug_log_fd.write(repr(output_b) + "\n") debug_log_fd.flush() + state.value = 1 + os.write(master, b"SET " + bytes(comp_begin.encode())) output_b = os.read(master, 4096) output = output_b.decode() @@ -55,23 +77,28 @@ def test_completion(program, argv, comp_word): time.sleep(0.01) os.write(master, b"\t") + state.value = 2 + output_b = os.read(master, 4096) output = output_b.decode() debug_log_fd.write(repr(output_b) + "\n") debug_log_fd.flush() - # fail fast if there is a bell character in the output, - # meaning no concise completion is found - if "\x07" in output: - print(f"{comp_word}: FAIL") - return while not comp_word in output: + # fail fast if there is a bell character in the output, + # meaning no concise completion is found + if "\x07" in output: + print(f"{comp_word}: FAIL") + state.value = 3 + return + output_b = os.read(master, 4096) output += output_b.decode() debug_log_fd.write(repr(output_b) + "\n") debug_log_fd.flush() print(f"{comp_word}: OK") + state.value = 0 finally: os.close(master) debug_log_fd.close() diff --git a/tests/queries/0_stateless/01684_ssd_cache_dictionary_simple_key.sh b/tests/queries/0_stateless/01684_ssd_cache_dictionary_simple_key.sh index 9167a2d306f..0e5c2862066 100755 --- a/tests/queries/0_stateless/01684_ssd_cache_dictionary_simple_key.sh +++ b/tests/queries/0_stateless/01684_ssd_cache_dictionary_simple_key.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT -n --query=" DROP DATABASE IF EXISTS 01684_database_for_cache_dictionary; CREATE DATABASE 01684_database_for_cache_dictionary; diff --git a/tests/queries/0_stateless/01685_ssd_cache_dictionary_complex_key.sh b/tests/queries/0_stateless/01685_ssd_cache_dictionary_complex_key.sh index 9dd8a41ce5a..5583a9dd5e7 100755 --- a/tests/queries/0_stateless/01685_ssd_cache_dictionary_complex_key.sh +++ b/tests/queries/0_stateless/01685_ssd_cache_dictionary_complex_key.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT -n --query=" DROP DATABASE IF EXISTS 01685_database_for_cache_dictionary; CREATE DATABASE 01685_database_for_cache_dictionary; diff --git a/tests/queries/0_stateless/01705_normalize_create_alter_function_names.sql b/tests/queries/0_stateless/01705_normalize_create_alter_function_names.sql index be0f7e8b710..921d28e6399 100644 --- a/tests/queries/0_stateless/01705_normalize_create_alter_function_names.sql +++ b/tests/queries/0_stateless/01705_normalize_create_alter_function_names.sql @@ -1,4 +1,4 @@ --- Tags: zookeeper, no-replicated-database, no-parallel, no-s3-storage +-- Tags: zookeeper, no-replicated-database, no-parallel, no-object-storage drop table if exists x; diff --git a/tests/queries/0_stateless/01710_minmax_count_projection.sql b/tests/queries/0_stateless/01710_minmax_count_projection.sql index d0177da84d2..6c598bce440 100644 --- a/tests/queries/0_stateless/01710_minmax_count_projection.sql +++ b/tests/queries/0_stateless/01710_minmax_count_projection.sql @@ -16,7 +16,7 @@ select min(i), max(i), count() from d where _partition_value.1 = 10 group by _pa select min(i) from d where 1 = _partition_value.1; -- fuzz crash https://github.com/ClickHouse/ClickHouse/issues/37151 -SELECT min(i), max(i), count() FROM d WHERE (_partition_value.1) = 0 GROUP BY ignore(bitTest(ignore(NULL), 65535), NULL, (_partition_value.1) = 7, '10.25', bitTest(NULL, -9223372036854775808), NULL, ignore(ignore(-2147483647, NULL)), 1024), _partition_id ORDER BY _partition_id ASC NULLS FIRST; +SELECT min(i), max(i), count() FROM d WHERE (_partition_value.1) = 0 GROUP BY ignore(bitTest(ignore(NULL), 0), NULL, (_partition_value.1) = 7, '10.25', bitTest(NULL, 0), NULL, ignore(ignore(-2147483647, NULL)), 1024), _partition_id ORDER BY _partition_id ASC NULLS FIRST; drop table d; diff --git a/tests/queries/0_stateless/01710_projection_vertical_merges.sql b/tests/queries/0_stateless/01710_projection_vertical_merges.sql index 2c4378bb7a4..0f80d659e92 100644 --- a/tests/queries/0_stateless/01710_projection_vertical_merges.sql +++ b/tests/queries/0_stateless/01710_projection_vertical_merges.sql @@ -1,4 +1,4 @@ --- Tags: long, no-parallel +-- Tags: long drop table if exists t; diff --git a/tests/queries/0_stateless/01747_join_view_filter_dictionary.sql b/tests/queries/0_stateless/01747_join_view_filter_dictionary.sql index 050aa33464e..c7c525ba20e 100644 --- a/tests/queries/0_stateless/01747_join_view_filter_dictionary.sql +++ b/tests/queries/0_stateless/01747_join_view_filter_dictionary.sql @@ -25,7 +25,7 @@ create table dictst01747(some_name String, field1 String, field2 UInt8) Engine = as select 'name', 'test', 33; CREATE DICTIONARY default.dict01747 (some_name String, field1 String, field2 UInt8) -PRIMARY KEY some_name SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 +PRIMARY KEY some_name SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() TABLE dictst01747 DB currentDatabase() USER 'default')) LIFETIME(MIN 0 MAX 0) LAYOUT(COMPLEX_KEY_HASHED()); diff --git a/tests/queries/0_stateless/01747_system_session_log_long.sh b/tests/queries/0_stateless/01747_system_session_log_long.sh index ecddcb627b8..022bf488886 100755 --- a/tests/queries/0_stateless/01747_system_session_log_long.sh +++ b/tests/queries/0_stateless/01747_system_session_log_long.sh @@ -228,13 +228,13 @@ function testMySQL() echo "MySQL 'successful login' case is skipped for ${auth_type}." else executeQuery \ - <<< "SELECT 1 FROM mysql('127.0.0.1:9004', 'system', 'one', '${username}', '${password}') LIMIT 1 \ + <<< "SELECT 1 FROM mysql('127.0.0.1:${CLICKHOUSE_PORT_MYSQL}', 'system', 'one', '${username}', '${password}') LIMIT 1 \ FORMAT Null" fi echo 'Wrong username' executeQueryExpectError \ - <<< "SELECT 1 FROM mysql('127.0.0.1:9004', 'system', 'one', 'invalid_${username}', '${password}') LIMIT 1 \ + <<< "SELECT 1 FROM mysql('127.0.0.1:${CLICKHOUSE_PORT_MYSQL}', 'system', 'one', 'invalid_${username}', '${password}') LIMIT 1 \ FORMAT Null" \ | grep -Eq "Code: 279\. DB::Exception: .* invalid_${username}" @@ -246,7 +246,7 @@ function testMySQL() echo "MySQL 'wrong password' case is skipped for ${auth_type}." else executeQueryExpectError \ - <<< "SELECT 1 FROM mysql('127.0.0.1:9004', 'system', 'one', '${username}', 'invalid_${password}') LIMIT 1 \ + <<< "SELECT 1 FROM mysql('127.0.0.1:${CLICKHOUSE_PORT_MYSQL}', 'system', 'one', '${username}', 'invalid_${password}') LIMIT 1 \ FORMAT Null" | grep -Eq "Code: 279\. DB::Exception: .* ${username}" fi } @@ -267,11 +267,11 @@ function testMySQL() ## Loging\Logout ## CH is being able to log into itself via PostgreSQL protocol but query fails. #executeQueryExpectError \ - # <<< "SELECT 1 FROM postgresql('localhost:9005', 'system', 'one', '${username}', '${password}') LIMIT 1 FORMAT Null" \ + # <<< "SELECT 1 FROM postgresql('localhost:${CLICKHOUSE_PORT_POSTGRESQL', 'system', 'one', '${username}', '${password}') LIMIT 1 FORMAT Null" \ # Wrong username executeQueryExpectError \ - <<< "SELECT 1 FROM postgresql('localhost:9005', 'system', 'one', 'invalid_${username}', '${password}') LIMIT 1 FORMAT Null" \ + <<< "SELECT 1 FROM postgresql('localhost:${CLICKHOUSE_PORT_POSTGRESQL}', 'system', 'one', 'invalid_${username}', '${password}') LIMIT 1 FORMAT Null" \ | grep -Eq "Invalid user or password" if [[ "${auth_type}" == "no_password" ]] @@ -281,7 +281,7 @@ function testMySQL() else # Wrong password executeQueryExpectError \ - <<< "SELECT 1 FROM postgresql('localhost:9005', 'system', 'one', '${username}', 'invalid_${password}') LIMIT 1 FORMAT Null" \ + <<< "SELECT 1 FROM postgresql('localhost:${CLICKHOUSE_PORT_POSTGRESQL}', 'system', 'one', '${username}', 'invalid_${password}') LIMIT 1 FORMAT Null" \ | grep -Eq "Invalid user or password" fi } diff --git a/tests/queries/0_stateless/01748_dictionary_table_dot.sql b/tests/queries/0_stateless/01748_dictionary_table_dot.sql index a2364fdf823..993d2e1a635 100644 --- a/tests/queries/0_stateless/01748_dictionary_table_dot.sql +++ b/tests/queries/0_stateless/01748_dictionary_table_dot.sql @@ -22,7 +22,7 @@ CREATE DICTIONARY test_dict `value` String ) PRIMARY KEY key1, key2 -SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 USER 'default' TABLE `test.txt` PASSWORD '' DB currentDatabase())) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE `test.txt` PASSWORD '' DB currentDatabase())) LIFETIME(MIN 1 MAX 3600) LAYOUT(COMPLEX_KEY_HASHED()); diff --git a/tests/queries/0_stateless/01748_partition_id_pruning.sql b/tests/queries/0_stateless/01748_partition_id_pruning.sql index b637528bc6c..f492a4a8871 100644 --- a/tests/queries/0_stateless/01748_partition_id_pruning.sql +++ b/tests/queries/0_stateless/01748_partition_id_pruning.sql @@ -8,17 +8,17 @@ insert into x values (1, 1), (1, 2), (1, 3), (2, 4), (2, 5), (2, 6); set max_rows_to_read = 3; -select * from x where _partition_id = partitionId(1); +select * from x where _partition_id = partitionID(1); set max_rows_to_read = 5; -- one row for subquery + subquery -select * from x where _partition_id in (select partitionId(number + 1) from numbers(1)); +select * from x where _partition_id in (select partitionID(number + 1) from numbers(1)); -- trivial count optimization test set max_rows_to_read = 2; -- one row for subquery + subquery itself -- TODO: Relax the limits because we might build prepared set twice with _minmax_count_projection set max_rows_to_read = 3; -select count() from x where _partition_id in (select partitionId(number + 1) from numbers(1)); +select count() from x where _partition_id in (select partitionID(number + 1) from numbers(1)); drop table x; diff --git a/tests/queries/0_stateless/01753_direct_dictionary_simple_key.sql b/tests/queries/0_stateless/01753_direct_dictionary_simple_key.sql index 86af09f391d..93ed3f93c4e 100644 --- a/tests/queries/0_stateless/01753_direct_dictionary_simple_key.sql +++ b/tests/queries/0_stateless/01753_direct_dictionary_simple_key.sql @@ -41,7 +41,7 @@ SELECT dictGetOrDefault('01753_dictionary_db.direct_dictionary_simple_key_simple SELECT 'dictHas'; SELECT dictHas('01753_dictionary_db.direct_dictionary_simple_key_simple_attributes', number) FROM system.numbers LIMIT 4; SELECT 'select all values as input stream'; -SELECT * FROM 01753_dictionary_db.direct_dictionary_simple_key_simple_attributes; +SELECT * FROM 01753_dictionary_db.direct_dictionary_simple_key_simple_attributes ORDER BY ALL; DROP DICTIONARY 01753_dictionary_db.direct_dictionary_simple_key_simple_attributes; DROP TABLE 01753_dictionary_db.simple_key_simple_attributes_source_table; diff --git a/tests/queries/0_stateless/01760_ddl_dictionary_use_current_database_name.sql b/tests/queries/0_stateless/01760_ddl_dictionary_use_current_database_name.sql index 55c0d1e3678..a7f04921f1f 100644 --- a/tests/queries/0_stateless/01760_ddl_dictionary_use_current_database_name.sql +++ b/tests/queries/0_stateless/01760_ddl_dictionary_use_current_database_name.sql @@ -27,5 +27,5 @@ SELECT dictGet('ddl_dictionary_test', 'value', number) FROM system.numbers LIMIT SELECT 'dictHas'; SELECT dictHas('ddl_dictionary_test', number) FROM system.numbers LIMIT 3; -DROP TABLE ddl_dictonary_test_source; DROP DICTIONARY ddl_dictionary_test; +DROP TABLE ddl_dictonary_test_source; diff --git a/tests/queries/0_stateless/01760_system_dictionaries.sql b/tests/queries/0_stateless/01760_system_dictionaries.sql index a5609281e49..2e7d4184811 100644 --- a/tests/queries/0_stateless/01760_system_dictionaries.sql +++ b/tests/queries/0_stateless/01760_system_dictionaries.sql @@ -25,8 +25,8 @@ SELECT * FROM 01760_db.example_simple_key_dictionary; SELECT name, database, key.names, key.types, attribute.names, attribute.types, status FROM system.dictionaries WHERE database='01760_db'; -DROP TABLE 01760_db.example_simple_key_source; DROP DICTIONARY 01760_db.example_simple_key_dictionary; +DROP TABLE 01760_db.example_simple_key_source; SELECT name, database, key.names, key.types, attribute.names, attribute.types, status FROM system.dictionaries WHERE database='01760_db'; @@ -53,7 +53,7 @@ SELECT * FROM 01760_db.example_complex_key_dictionary; SELECT name, database, key.names, key.types, attribute.names, attribute.types, status FROM system.dictionaries WHERE database='01760_db'; -DROP TABLE 01760_db.example_complex_key_source; DROP DICTIONARY 01760_db.example_complex_key_dictionary; +DROP TABLE 01760_db.example_complex_key_source; DROP DATABASE 01760_db; diff --git a/tests/queries/0_stateless/01763_max_distributed_depth.sql b/tests/queries/0_stateless/01763_max_distributed_depth.sql index 08dc533876d..f722a88226d 100644 --- a/tests/queries/0_stateless/01763_max_distributed_depth.sql +++ b/tests/queries/0_stateless/01763_max_distributed_depth.sql @@ -17,19 +17,6 @@ ENGINE = Distributed('test_shard_localhost', '', 'tt7', rand()); DROP TABLE IF EXISTS tt7; -CREATE TABLE tt7 as tt6 ENGINE = Distributed('test_shard_localhost', '', 'tt6', rand()); - -INSERT INTO tt6 VALUES (1, 1, 1, 1, 'ok'); -- { serverError TOO_LARGE_DISTRIBUTED_DEPTH } - -SELECT * FROM tt6; -- { serverError TOO_LARGE_DISTRIBUTED_DEPTH } - -SET max_distributed_depth = 0; - --- stack overflow -INSERT INTO tt6 VALUES (1, 1, 1, 1, 'ok'); -- { serverError TOO_DEEP_RECURSION} - --- stack overflow -SELECT * FROM tt6; -- { serverError TOO_DEEP_RECURSION } +CREATE TABLE tt7 as tt6 ENGINE = Distributed('test_shard_localhost', '', 'tt6', rand()); -- {serverError INFINITE_LOOP} DROP TABLE tt6; -DROP TABLE tt7; diff --git a/tests/queries/0_stateless/01764_table_function_dictionary.sql b/tests/queries/0_stateless/01764_table_function_dictionary.sql index b642fdd741e..76e7213b367 100644 --- a/tests/queries/0_stateless/01764_table_function_dictionary.sql +++ b/tests/queries/0_stateless/01764_table_function_dictionary.sql @@ -23,5 +23,5 @@ LAYOUT(DIRECT()); SELECT * FROM dictionary('table_function_dictionary_test_dictionary'); -DROP TABLE table_function_dictionary_source_table; DROP DICTIONARY table_function_dictionary_test_dictionary; +DROP TABLE table_function_dictionary_source_table; diff --git a/tests/queries/0_stateless/01778_mmap_cache_infra.sql b/tests/queries/0_stateless/01778_mmap_cache_infra.sql index 29a84c5507b..50fdb6ffbba 100644 --- a/tests/queries/0_stateless/01778_mmap_cache_infra.sql +++ b/tests/queries/0_stateless/01778_mmap_cache_infra.sql @@ -1,3 +1,4 @@ +-- Tags: no-parallel -- We check the existence of queries and metrics and don't check the results (a smoke test). SYSTEM DROP MMAP CACHE; diff --git a/tests/queries/0_stateless/01780_clickhouse_dictionary_source_loop.sql b/tests/queries/0_stateless/01780_clickhouse_dictionary_source_loop.sql index 1eee4090112..3ebc85c47f7 100644 --- a/tests/queries/0_stateless/01780_clickhouse_dictionary_source_loop.sql +++ b/tests/queries/0_stateless/01780_clickhouse_dictionary_source_loop.sql @@ -10,7 +10,7 @@ CREATE DICTIONARY dict1 value String ) PRIMARY KEY id -SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 TABLE 'dict1')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() TABLE 'dict1')) LAYOUT(DIRECT()); SELECT * FROM dict1; --{serverError BAD_ARGUMENTS} @@ -24,7 +24,7 @@ CREATE DICTIONARY 01780_db.dict2 value String ) PRIMARY KEY id -SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 DATABASE '01780_db' TABLE 'dict2')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() DATABASE '01780_db' TABLE 'dict2')) LAYOUT(DIRECT()); SELECT * FROM 01780_db.dict2; --{serverError BAD_ARGUMENTS} @@ -45,7 +45,7 @@ CREATE DICTIONARY 01780_db.dict3 value String ) PRIMARY KEY id -SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 TABLE 'dict3_source' DATABASE '01780_db')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() TABLE 'dict3_source' DATABASE '01780_db')) LAYOUT(DIRECT()); SELECT * FROM 01780_db.dict3; diff --git a/tests/queries/0_stateless/01780_column_sparse_distinct.sql b/tests/queries/0_stateless/01780_column_sparse_distinct.sql index e98bada1aac..a0735e38f18 100644 --- a/tests/queries/0_stateless/01780_column_sparse_distinct.sql +++ b/tests/queries/0_stateless/01780_column_sparse_distinct.sql @@ -1,3 +1,5 @@ +SET optimize_trivial_insert_select = 1; + DROP TABLE IF EXISTS t_sparse_distinct; CREATE TABLE t_sparse_distinct (id UInt32, v UInt64) diff --git a/tests/queries/0_stateless/01780_column_sparse_filter.sql b/tests/queries/0_stateless/01780_column_sparse_filter.sql index f52beba50b0..245c7c121b7 100644 --- a/tests/queries/0_stateless/01780_column_sparse_filter.sql +++ b/tests/queries/0_stateless/01780_column_sparse_filter.sql @@ -1,3 +1,5 @@ +SET optimize_trivial_insert_select = 1; + DROP TABLE IF EXISTS t_sparse; CREATE TABLE t_sparse (id UInt64, u UInt64, s String) diff --git a/tests/queries/0_stateless/01804_dictionary_decimal256_type.sql b/tests/queries/0_stateless/01804_dictionary_decimal256_type.sql index 77e9abfb742..08a8d0feb27 100644 --- a/tests/queries/0_stateless/01804_dictionary_decimal256_type.sql +++ b/tests/queries/0_stateless/01804_dictionary_decimal256_type.sql @@ -25,6 +25,8 @@ LAYOUT(FLAT()); SELECT 'Flat dictionary'; SELECT dictGet('flat_dictionary', 'decimal_value', toUInt64(1)); +DROP DICTIONARY flat_dictionary; + DROP DICTIONARY IF EXISTS hashed_dictionary; CREATE DICTIONARY hashed_dictionary ( diff --git a/tests/queries/0_stateless/01810_max_part_removal_threads_long.sh b/tests/queries/0_stateless/01810_max_part_removal_threads_long.sh index 3782a7d3ad6..c38fc505fa8 100755 --- a/tests/queries/0_stateless/01810_max_part_removal_threads_long.sh +++ b/tests/queries/0_stateless/01810_max_part_removal_threads_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-s3-storage +# Tags: long, no-object-storage # Because parallel parts removal disabled for s3 storage # NOTE: this done as not .sql since we need to Ordinary database diff --git a/tests/queries/0_stateless/01821_table_comment.reference b/tests/queries/0_stateless/01821_table_comment.reference index 05acabae3d4..cdd87df43d2 100644 --- a/tests/queries/0_stateless/01821_table_comment.reference +++ b/tests/queries/0_stateless/01821_table_comment.reference @@ -1,4 +1,8 @@ -t1 this is a temtorary table +t1 this is a temporary table t2 this is a MergeTree table t3 this is a Log table -CREATE TABLE default.t1\n(\n `n` Int8\n)\nENGINE = Memory\nCOMMENT \'this is a temtorary table\' +t4 this is a Kafka table +t5 this is a EmbeddedRocksDB table +t6 this is a Executable table +t7 this is a WindowView table +CREATE TABLE default.t1\n(\n `n` Int8\n)\nENGINE = Memory\nCOMMENT \'this is a temporary table\' diff --git a/tests/queries/0_stateless/01821_table_comment.sql b/tests/queries/0_stateless/01821_table_comment.sql index 4bd71d3e278..4946e46d37a 100644 --- a/tests/queries/0_stateless/01821_table_comment.sql +++ b/tests/queries/0_stateless/01821_table_comment.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel +-- Tags: no-parallel, no-fasttest DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; @@ -9,7 +9,7 @@ CREATE TABLE t1 `n` Int8 ) ENGINE = Memory -COMMENT 'this is a temtorary table'; +COMMENT 'this is a temporary table'; CREATE TABLE t2 ( @@ -26,14 +26,57 @@ CREATE TABLE t3 ENGINE = Log COMMENT 'this is a Log table'; +CREATE TABLE t4 +( + `n` Int8 +) +ENGINE = Kafka +SETTINGS + kafka_broker_list = 'localhost:10000', + kafka_topic_list = 'test', + kafka_group_name = 'test', + kafka_format = 'JSONEachRow' +COMMENT 'this is a Kafka table'; + +CREATE TABLE t5 +( + `n` Int8 +) +ENGINE = EmbeddedRocksDB +PRIMARY KEY n +COMMENT 'this is a EmbeddedRocksDB table'; + +CREATE TABLE t6 +( + `n` Int8 +) +ENGINE = Executable('script.py', TabSeparated) +COMMENT 'this is a Executable table'; + +SET allow_experimental_window_view = 1; +-- New analyzer doesn't support WindowView tables +SET allow_experimental_analyzer = 0; + +CREATE WINDOW VIEW t7 +( + `n` Int8 +) +ENGINE MergeTree +ORDER BY n +AS SELECT 1 +GROUP BY tumble(now(), toIntervalDay('1')) +COMMENT 'this is a WindowView table'; + +SET allow_experimental_analyzer = 1; + SELECT name, comment FROM system.tables -WHERE name IN ('t1', 't2', 't3') AND database = currentDatabase() order by name; +WHERE name IN ('t1', 't2', 't3', 't4', 't5', 't6', 't7') + AND database = currentDatabase() order by name; SHOW CREATE TABLE t1; -DROP TABLE t1; -DROP TABLE t2; -DROP TABLE t3; +DROP TABLE t1, t2, t3, t4, t5, t6; +DROP VIEW t7; diff --git a/tests/queries/0_stateless/01825_type_json_btc.sh b/tests/queries/0_stateless/01825_type_json_btc.sh index 1e74166e7a7..ebc5482de7a 100755 --- a/tests/queries/0_stateless/01825_type_json_btc.sh +++ b/tests/queries/0_stateless/01825_type_json_btc.sh @@ -5,10 +5,9 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* -cp $CUR_DIR/data_json/btc_transactions.json ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +mkdir -p ${CLICKHOUSE_USER_FILES_UNIQUE}/ +rm -rf "${CLICKHOUSE_USER_FILES_UNIQUE:?}"/* +cp $CUR_DIR/data_json/btc_transactions.json ${CLICKHOUSE_USER_FILES_UNIQUE}/ ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS btc" @@ -27,4 +26,4 @@ ${CLICKHOUSE_CLIENT} -q "SELECT data.out.spending_outpoints AS outpoints FROM bt ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS btc" -rm ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/btc_transactions.json +rm ${CLICKHOUSE_USER_FILES_UNIQUE}/btc_transactions.json diff --git a/tests/queries/0_stateless/01825_type_json_from_map.sql b/tests/queries/0_stateless/01825_type_json_from_map.sql index 7cad50b363b..817d099c2d5 100644 --- a/tests/queries/0_stateless/01825_type_json_from_map.sql +++ b/tests/queries/0_stateless/01825_type_json_from_map.sql @@ -1,10 +1,11 @@ --- Tags: no-fasttest, no-random-merge-tree-settings +-- Tags: no-fasttest, no-random-settings, no-random-merge-tree-settings -- For example, it is 4 times slower with --merge_max_block_size=5967 --index_granularity=55 --min_bytes_for_wide_part=847510133 DROP TABLE IF EXISTS t_json; DROP TABLE IF EXISTS t_map; SET allow_experimental_object_type = 1; +SET optimize_trivial_insert_select = 1; CREATE TABLE t_json(id UInt64, obj JSON) ENGINE = MergeTree ORDER BY id; CREATE TABLE t_map(id UInt64, m Map(String, UInt64)) ENGINE = MergeTree ORDER BY id; diff --git a/tests/queries/0_stateless/01825_type_json_multiple_files.sh b/tests/queries/0_stateless/01825_type_json_multiple_files.sh index 089b7991784..453e7a3c78e 100755 --- a/tests/queries/0_stateless/01825_type_json_multiple_files.sh +++ b/tests/queries/0_stateless/01825_type_json_multiple_files.sh @@ -1,23 +1,22 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel +# Tags: no-fasttest CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -user_files_path=$($CLICKHOUSE_CLIENT --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep -E '^Code: 107.*FILE_DOESNT_EXIST' | head -1 | awk '{gsub("/nonexist.txt","",$9); print $9}') -for f in "$user_files_path"/01825_file_*.json; do +for f in "${USER_FILES_PATH:?}/${CLICKHOUSE_DATABASE}"_*.json; do [ -e $f ] && rm $f done for i in {0..5}; do - echo "{\"k$i\": 100}" > "$user_files_path"/01825_file_$i.json + echo "{\"k$i\": 100}" > "$USER_FILES_PATH/${CLICKHOUSE_DATABASE}_$i.json" done ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS t_json_files" ${CLICKHOUSE_CLIENT} -q "CREATE TABLE t_json_files (file String, data JSON) ENGINE = MergeTree ORDER BY tuple()" --allow_experimental_object_type 1 -${CLICKHOUSE_CLIENT} -q "INSERT INTO t_json_files SELECT _file, data FROM file('01825_file_*.json', 'JSONAsObject', 'data JSON')" --allow_experimental_object_type 1 +${CLICKHOUSE_CLIENT} -q "INSERT INTO t_json_files SELECT _file, data FROM file('${CLICKHOUSE_DATABASE}_*.json', 'JSONAsObject', 'data JSON')" --allow_experimental_object_type 1 ${CLICKHOUSE_CLIENT} -q "SELECT data FROM t_json_files ORDER BY file FORMAT JSONEachRow" --output_format_json_named_tuples_as_objects 1 ${CLICKHOUSE_CLIENT} -q "SELECT toTypeName(data) FROM t_json_files LIMIT 1" @@ -25,7 +24,7 @@ ${CLICKHOUSE_CLIENT} -q "SELECT toTypeName(data) FROM t_json_files LIMIT 1" ${CLICKHOUSE_CLIENT} -q "TRUNCATE TABLE IF EXISTS t_json_files" ${CLICKHOUSE_CLIENT} -q "INSERT INTO t_json_files \ - SELECT _file, data FROM file('01825_file_*.json', 'JSONAsObject', 'data JSON') \ + SELECT _file, data FROM file('${CLICKHOUSE_DATABASE}_*.json', 'JSONAsObject', 'data JSON') \ ORDER BY _file LIMIT 3" --max_threads 1 --min_insert_block_size_rows 1 --max_insert_block_size 1 --max_block_size 1 --allow_experimental_object_type 1 ${CLICKHOUSE_CLIENT} -q "SELECT data FROM t_json_files ORDER BY file, data FORMAT JSONEachRow" --output_format_json_named_tuples_as_objects 1 @@ -34,11 +33,11 @@ ${CLICKHOUSE_CLIENT} -q "SELECT toTypeName(data) FROM t_json_files LIMIT 1" ${CLICKHOUSE_CLIENT} -q "TRUNCATE TABLE IF EXISTS t_json_files" ${CLICKHOUSE_CLIENT} -q "INSERT INTO t_json_files \ - SELECT _file, data FROM file('01825_file_*.json', 'JSONAsObject', 'data JSON') \ - WHERE _file IN ('01825_file_1.json', '01825_file_3.json')" --allow_experimental_object_type 1 + SELECT _file, data FROM file('${CLICKHOUSE_DATABASE}_*.json', 'JSONAsObject', 'data JSON') \ + WHERE _file IN ('${CLICKHOUSE_DATABASE}_1.json', '${CLICKHOUSE_DATABASE}_3.json')" --allow_experimental_object_type 1 ${CLICKHOUSE_CLIENT} -q "SELECT data FROM t_json_files ORDER BY file FORMAT JSONEachRow" --output_format_json_named_tuples_as_objects 1 ${CLICKHOUSE_CLIENT} -q "SELECT toTypeName(data) FROM t_json_files LIMIT 1" ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS t_json_files" -rm "$user_files_path"/01825_file_*.json +rm "$USER_FILES_PATH"/${CLICKHOUSE_DATABASE}_*.json diff --git a/tests/queries/0_stateless/01825_type_json_schema_inference.sh b/tests/queries/0_stateless/01825_type_json_schema_inference.sh index 5fca608d8bb..e0c283b2230 100755 --- a/tests/queries/0_stateless/01825_type_json_schema_inference.sh +++ b/tests/queries/0_stateless/01825_type_json_schema_inference.sh @@ -10,11 +10,10 @@ ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS t_json_inference" ${CLICKHOUSE_CLIENT} -q "CREATE TABLE t_json_inference (id UInt64, obj Object(Nullable('json')), s String) \ ENGINE = MergeTree ORDER BY id" --allow_experimental_object_type 1 -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* -filename="${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/data.json" +filename="${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/data.json" echo '{"id": 1, "obj": {"k1": 1, "k2": {"k3": 2, "k4": [{"k5": 3}, {"k5": 4}]}}, "s": "foo"}' > $filename echo '{"id": 2, "obj": {"k2": {"k3": "str", "k4": [{"k6": 55}]}, "some": 42}, "s": "bar"}' >> $filename diff --git a/tests/queries/0_stateless/01825_type_json_sparse.sql b/tests/queries/0_stateless/01825_type_json_sparse.sql index cc7c66382a3..69ca1ff8406 100644 --- a/tests/queries/0_stateless/01825_type_json_sparse.sql +++ b/tests/queries/0_stateless/01825_type_json_sparse.sql @@ -3,6 +3,7 @@ DROP TABLE IF EXISTS t_json_sparse; SET allow_experimental_object_type = 1; +SET optimize_trivial_insert_select = 1; CREATE TABLE t_json_sparse (data JSON) ENGINE = MergeTree ORDER BY tuple() diff --git a/tests/queries/0_stateless/01852_dictionary_found_rate_long.sql b/tests/queries/0_stateless/01852_dictionary_found_rate_long.sql index d5108e98510..da364403893 100644 --- a/tests/queries/0_stateless/01852_dictionary_found_rate_long.sql +++ b/tests/queries/0_stateless/01852_dictionary_found_rate_long.sql @@ -310,6 +310,6 @@ SELECT name, found_rate FROM system.dictionaries WHERE database = currentDatabas SELECT tuple(x, y) as key, dictGet('polygon_dictionary_01862', 'name', key) FROM points_01862 FORMAT Null; SELECT name, found_rate FROM system.dictionaries WHERE database = currentDatabase() AND name = 'polygon_dictionary_01862'; +DROP DICTIONARY polygon_dictionary_01862; DROP TABLE polygons_01862; DROP TABLE points_01862; -DROP DICTIONARY polygon_dictionary_01862; diff --git a/tests/queries/0_stateless/01872_functions_to_subcolumns.reference b/tests/queries/0_stateless/01872_functions_to_subcolumns.reference deleted file mode 100644 index a1cd31e2dc9..00000000000 --- a/tests/queries/0_stateless/01872_functions_to_subcolumns.reference +++ /dev/null @@ -1,47 +0,0 @@ -0 0 1 -0 1 0 -SELECT - id IS NULL, - `n.null`, - NOT `n.null` -FROM t_func_to_subcolumns -3 0 1 0 -0 1 0 \N -SELECT - `arr.size0`, - `arr.size0` = 0, - `arr.size0` != 0, - empty(n) -FROM t_func_to_subcolumns -['foo','bar'] [1,2] -[] [] -SELECT - `m.keys`, - `m.values` -FROM t_func_to_subcolumns -1 -SELECT sum(NOT `n.null`) -FROM t_func_to_subcolumns -2 -SELECT count(id) -FROM t_func_to_subcolumns -1 0 0 -2 1 0 -3 0 0 -SELECT - id, - `n.null`, - right.n IS NULL -FROM t_func_to_subcolumns AS left -ALL FULL OUTER JOIN -( - SELECT - 1 AS id, - \'qqq\' AS n - UNION ALL - SELECT - 3 AS id, - \'www\' -) AS right USING (id) -0 10 -0 20 diff --git a/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.reference b/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.reference new file mode 100644 index 00000000000..32bacfba5ea --- /dev/null +++ b/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.reference @@ -0,0 +1,181 @@ +0 0 1 +0 1 0 +QUERY id: 0 + PROJECTION COLUMNS + isNull(id) UInt8 + isNull(n) UInt8 + isNotNull(n) UInt8 + PROJECTION + LIST id: 1, nodes: 3 + CONSTANT id: 2, constant_value: UInt64_0, constant_value_type: UInt8 + EXPRESSION + FUNCTION id: 3, function_name: isNull, function_type: ordinary, result_type: UInt8 + ARGUMENTS + LIST id: 4, nodes: 1 + COLUMN id: 5, column_name: id, result_type: UInt64, source_id: 6 + COLUMN id: 7, column_name: n.null, result_type: UInt8, source_id: 6 + FUNCTION id: 8, function_name: not, function_type: ordinary, result_type: UInt8 + ARGUMENTS + LIST id: 9, nodes: 1 + COLUMN id: 10, column_name: n.null, result_type: UInt8, source_id: 6 + JOIN TREE + TABLE id: 6, alias: __table1, table_name: default.t_func_to_subcolumns + +SELECT + _CAST(0, \'UInt8\') AS `isNull(id)`, + __table1.`n.null` AS `isNull(n)`, + NOT __table1.`n.null` AS `isNotNull(n)` +FROM default.t_func_to_subcolumns AS __table1 +3 0 1 0 +0 1 0 \N +QUERY id: 0 + PROJECTION COLUMNS + length(arr) UInt64 + empty(arr) UInt8 + notEmpty(arr) UInt8 + empty(n) Nullable(UInt8) + PROJECTION + LIST id: 1, nodes: 4 + COLUMN id: 2, column_name: arr.size0, result_type: UInt64, source_id: 3 + FUNCTION id: 4, function_name: equals, function_type: ordinary, result_type: UInt8 + ARGUMENTS + LIST id: 5, nodes: 2 + COLUMN id: 6, column_name: arr.size0, result_type: UInt64, source_id: 3 + CONSTANT id: 7, constant_value: UInt64_0, constant_value_type: UInt8 + FUNCTION id: 8, function_name: notEquals, function_type: ordinary, result_type: UInt8 + ARGUMENTS + LIST id: 9, nodes: 2 + COLUMN id: 10, column_name: arr.size0, result_type: UInt64, source_id: 3 + CONSTANT id: 11, constant_value: UInt64_0, constant_value_type: UInt8 + FUNCTION id: 12, function_name: empty, function_type: ordinary, result_type: Nullable(UInt8) + ARGUMENTS + LIST id: 13, nodes: 1 + COLUMN id: 14, column_name: n, result_type: Nullable(String), source_id: 3 + JOIN TREE + TABLE id: 3, alias: __table1, table_name: default.t_func_to_subcolumns + +SELECT + __table1.`arr.size0` AS `length(arr)`, + __table1.`arr.size0` = 0 AS `empty(arr)`, + __table1.`arr.size0` != 0 AS `notEmpty(arr)`, + empty(__table1.n) AS `empty(n)` +FROM default.t_func_to_subcolumns AS __table1 +['foo','bar'] [1,2] +[] [] +QUERY id: 0 + PROJECTION COLUMNS + mapKeys(m) Array(String) + mapValues(m) Array(UInt64) + PROJECTION + LIST id: 1, nodes: 2 + COLUMN id: 2, column_name: m.keys, result_type: Array(String), source_id: 3 + COLUMN id: 4, column_name: m.values, result_type: Array(UInt64), source_id: 3 + JOIN TREE + TABLE id: 3, alias: __table1, table_name: default.t_func_to_subcolumns + +SELECT + __table1.`m.keys` AS `mapKeys(m)`, + __table1.`m.values` AS `mapValues(m)` +FROM default.t_func_to_subcolumns AS __table1 +1 +QUERY id: 0 + PROJECTION COLUMNS + count(n) UInt64 + PROJECTION + LIST id: 1, nodes: 1 + FUNCTION id: 2, function_name: sum, function_type: aggregate, result_type: UInt64 + ARGUMENTS + LIST id: 3, nodes: 1 + FUNCTION id: 4, function_name: not, function_type: ordinary, result_type: UInt8 + ARGUMENTS + LIST id: 5, nodes: 1 + COLUMN id: 6, column_name: n.null, result_type: UInt8, source_id: 7 + JOIN TREE + TABLE id: 7, alias: __table1, table_name: default.t_func_to_subcolumns + +SELECT sum(NOT __table1.`n.null`) AS `count(n)` +FROM default.t_func_to_subcolumns AS __table1 +2 +QUERY id: 0 + PROJECTION COLUMNS + count(id) UInt64 + PROJECTION + LIST id: 1, nodes: 1 + FUNCTION id: 2, function_name: count, function_type: aggregate, result_type: UInt64 + ARGUMENTS + LIST id: 3, nodes: 1 + COLUMN id: 4, column_name: id, result_type: UInt64, source_id: 5 + JOIN TREE + TABLE id: 5, alias: __table1, table_name: default.t_func_to_subcolumns + +SELECT count(__table1.id) AS `count(id)` +FROM default.t_func_to_subcolumns AS __table1 +1 0 0 +2 1 0 +3 0 0 +QUERY id: 0 + PROJECTION COLUMNS + id UInt64 + isNull(n) UInt8 + isNull(right.n) UInt8 + PROJECTION + LIST id: 1, nodes: 3 + COLUMN id: 2, column_name: id, result_type: UInt64, source_id: 3 + COLUMN id: 4, column_name: n.null, result_type: UInt8, source_id: 3 + CONSTANT id: 5, constant_value: UInt64_0, constant_value_type: UInt8 + EXPRESSION + FUNCTION id: 6, function_name: isNull, function_type: ordinary, result_type: UInt8 + ARGUMENTS + LIST id: 7, nodes: 1 + COLUMN id: 8, column_name: n, result_type: String, source_id: 9 + JOIN TREE + JOIN id: 10, strictness: ALL, kind: FULL + LEFT TABLE EXPRESSION + TABLE id: 3, alias: __table1, table_name: default.t_func_to_subcolumns + RIGHT TABLE EXPRESSION + UNION id: 9, alias: __table2, is_subquery: 1, union_mode: UNION_ALL + QUERIES + LIST id: 11, nodes: 2 + QUERY id: 12, alias: __table3 + PROJECTION COLUMNS + id UInt8 + PROJECTION + LIST id: 13, nodes: 1 + CONSTANT id: 14, constant_value: UInt64_1, constant_value_type: UInt8 + JOIN TREE + TABLE id: 15, alias: __table4, table_name: system.one + QUERY id: 16, alias: __table5 + PROJECTION COLUMNS + id UInt8 + PROJECTION + LIST id: 17, nodes: 1 + CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 + JOIN TREE + TABLE id: 19, alias: __table6, table_name: system.one + JOIN EXPRESSION + LIST id: 20, nodes: 1 + COLUMN id: 21, column_name: id, result_type: UInt64, source_id: 10 + EXPRESSION + LIST id: 22, nodes: 2 + COLUMN id: 23, column_name: id, result_type: UInt64, source_id: 3 + COLUMN id: 24, column_name: id, result_type: UInt8, source_id: 9 + +SELECT + __table1.id AS id, + __table1.`n.null` AS `isNull(n)`, + _CAST(0, \'UInt8\') AS `isNull(right.n)` +FROM default.t_func_to_subcolumns AS __table1 +ALL FULL OUTER JOIN +( + ( + SELECT 1 AS id + FROM system.one AS __table4 + ) + UNION ALL + ( + SELECT 3 AS id + FROM system.one AS __table6 + ) +) AS __table2 USING (id) +0 10 +0 20 diff --git a/tests/queries/0_stateless/01872_functions_to_subcolumns.sql b/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.sql similarity index 62% rename from tests/queries/0_stateless/01872_functions_to_subcolumns.sql rename to tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.sql index 45f83bf20e5..b544f6829cf 100644 --- a/tests/queries/0_stateless/01872_functions_to_subcolumns.sql +++ b/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.sql @@ -1,5 +1,6 @@ DROP TABLE IF EXISTS t_func_to_subcolumns; +SET allow_experimental_analyzer = 1; SET optimize_functions_to_subcolumns = 1; CREATE TABLE t_func_to_subcolumns (id UInt64, arr Array(UInt64), n Nullable(String), m Map(String, UInt64)) @@ -8,24 +9,24 @@ ENGINE = MergeTree ORDER BY tuple(); INSERT INTO t_func_to_subcolumns VALUES (1, [1, 2, 3], 'abc', map('foo', 1, 'bar', 2)) (2, [], NULL, map()); SELECT id IS NULL, n IS NULL, n IS NOT NULL FROM t_func_to_subcolumns; -EXPLAIN SYNTAX SELECT id IS NULL, n IS NULL, n IS NOT NULL FROM t_func_to_subcolumns; +EXPLAIN QUERY TREE dump_tree = 1, dump_ast = 1 SELECT id IS NULL, n IS NULL, n IS NOT NULL FROM t_func_to_subcolumns; SELECT length(arr), empty(arr), notEmpty(arr), empty(n) FROM t_func_to_subcolumns; -EXPLAIN SYNTAX SELECT length(arr), empty(arr), notEmpty(arr), empty(n) FROM t_func_to_subcolumns; +EXPLAIN QUERY TREE dump_tree = 1, dump_ast = 1 SELECT length(arr), empty(arr), notEmpty(arr), empty(n) FROM t_func_to_subcolumns; SELECT mapKeys(m), mapValues(m) FROM t_func_to_subcolumns; -EXPLAIN SYNTAX SELECT mapKeys(m), mapValues(m) FROM t_func_to_subcolumns; +EXPLAIN QUERY TREE dump_tree = 1, dump_ast = 1 SELECT mapKeys(m), mapValues(m) FROM t_func_to_subcolumns; SELECT count(n) FROM t_func_to_subcolumns; -EXPLAIN SYNTAX SELECT count(n) FROM t_func_to_subcolumns; +EXPLAIN QUERY TREE dump_tree = 1, dump_ast = 1 SELECT count(n) FROM t_func_to_subcolumns; SELECT count(id) FROM t_func_to_subcolumns; -EXPLAIN SYNTAX SELECT count(id) FROM t_func_to_subcolumns; +EXPLAIN QUERY TREE dump_tree = 1, dump_ast = 1 SELECT count(id) FROM t_func_to_subcolumns; SELECT id, left.n IS NULL, right.n IS NULL FROM t_func_to_subcolumns AS left FULL JOIN (SELECT 1 AS id, 'qqq' AS n UNION ALL SELECT 3 AS id, 'www') AS right USING(id); -EXPLAIN SYNTAX SELECT id, left.n IS NULL, right.n IS NULL FROM t_func_to_subcolumns AS left +EXPLAIN QUERY TREE dump_tree = 1, dump_ast = 1 SELECT id, left.n IS NULL, right.n IS NULL FROM t_func_to_subcolumns AS left FULL JOIN (SELECT 1 AS id, 'qqq' AS n UNION ALL SELECT 3 AS id, 'www') AS right USING(id); DROP TABLE t_func_to_subcolumns; diff --git a/tests/queries/0_stateless/01875_ssd_cache_dictionary_decimal256_type.sh b/tests/queries/0_stateless/01875_ssd_cache_dictionary_decimal256_type.sh index 36a2165329b..8336229a643 100755 --- a/tests/queries/0_stateless/01875_ssd_cache_dictionary_decimal256_type.sh +++ b/tests/queries/0_stateless/01875_ssd_cache_dictionary_decimal256_type.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT -n --query=" SET allow_experimental_bigint_types = 1; diff --git a/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 b/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 index bd20d34b684..c2d85cefb18 100644 --- a/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 +++ b/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 @@ -72,7 +72,7 @@ SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t1.id + 2; -- { serverE SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t2.id + 2; -- { serverError 403 } SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t1.key; -- { serverError 43, 403 } SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t2.key; -- { serverError 43, 403 } -SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 AND (t1.id == t2.id OR isNull(t2.key2)); -- { serverError 403 } +SELECT * FROM t1 JOIN t2_nullable as t2 ON t2.key == t2.key2 AND (t1.id == t2.id OR isNull(t2.key2)); -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 OR t1.id == t2.id; -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON (t2.key == t2.key2 AND (t1.key == t1.key2 AND t1.key != 'XXX' OR t1.id == t2.id)) AND t1.id == t2.id; -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 AND t1.key == t1.key2 AND t1.key != 'XXX' AND t1.id == t2.id OR t2.key == t2.key2 AND t1.id == t2.id AND t1.id == t2.id; diff --git a/tests/queries/0_stateless/01881_join_on_conditions_merge.sql.j2 b/tests/queries/0_stateless/01881_join_on_conditions_merge.sql.j2 index e4b704247b2..13703771ac8 100644 --- a/tests/queries/0_stateless/01881_join_on_conditions_merge.sql.j2 +++ b/tests/queries/0_stateless/01881_join_on_conditions_merge.sql.j2 @@ -70,7 +70,7 @@ SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t1.id + 2; -- { serverE SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t2.id + 2; -- { serverError 403 } SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t1.key; -- { serverError 43, 403 } SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t2.key; -- { serverError 43, 403 } -SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 AND (t1.id == t2.id OR isNull(t2.key2)); -- { serverError 403 } +SELECT * FROM t1 JOIN t2_nullable as t2 ON t2.key == t2.key2 AND (t1.id == t2.id OR isNull(t2.key2)); -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 OR t1.id == t2.id; -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON (t2.key == t2.key2 AND (t1.key == t1.key2 AND t1.key != 'XXX' OR t1.id == t2.id)) AND t1.id == t2.id; -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 AND t1.key == t1.key2 AND t1.key != 'XXX' AND t1.id == t2.id OR t2.key == t2.key2 AND t1.id == t2.id AND t1.id == t2.id; -- { serverError 48 } diff --git a/tests/queries/0_stateless/01889_check_row_policy_defined_using_user_function.sh b/tests/queries/0_stateless/01889_check_row_policy_defined_using_user_function.sh index b5be39a91df..f79637a7635 100755 --- a/tests/queries/0_stateless/01889_check_row_policy_defined_using_user_function.sh +++ b/tests/queries/0_stateless/01889_check_row_policy_defined_using_user_function.sh @@ -1,5 +1,4 @@ #!/usr/bin/env bash -# Tags: no-parallel CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL=none @@ -7,23 +6,27 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -q "drop user if exists u_01889" -${CLICKHOUSE_CLIENT} -q "drop role if exists r_01889" -${CLICKHOUSE_CLIENT} -q "drop policy if exists t_01889_filter on t_01889" -${CLICKHOUSE_CLIENT} -q "create user u_01889 identified with plaintext_password by 'dfsdffdf5t123'" -${CLICKHOUSE_CLIENT} -q "revoke all on *.* from u_01889" -${CLICKHOUSE_CLIENT} -q "create role r_01889" +USER=u_01889$RANDOM +ROLE=r_01889$RANDOM +POLICY=t_01889_filter$RANDOM + +${CLICKHOUSE_CLIENT} -q "drop user if exists $USER" +${CLICKHOUSE_CLIENT} -q "drop role if exists ${ROLE}" +${CLICKHOUSE_CLIENT} -q "drop policy if exists ${POLICY} on t_01889" +${CLICKHOUSE_CLIENT} -q "create user $USER identified with plaintext_password by 'dfsdffdf5t123'" +${CLICKHOUSE_CLIENT} -q "revoke all on *.* from $USER" +${CLICKHOUSE_CLIENT} -q "create role ${ROLE}" ${CLICKHOUSE_CLIENT} -q "create table t_01889(a Int64, user_id String) Engine=MergeTree order by a" -${CLICKHOUSE_CLIENT} -q "insert into t_01889 select number, 'u_01889' from numbers(1000)" +${CLICKHOUSE_CLIENT} -q "insert into t_01889 select number, '$USER' from numbers(1000)" ${CLICKHOUSE_CLIENT} -q "insert into t_01889 select number, 'xxxxxxx' from numbers(1000)" -${CLICKHOUSE_CLIENT} -q "grant select on t_01889 to r_01889" -${CLICKHOUSE_CLIENT} -q "create row policy t_01889_filter ON t_01889 FOR SELECT USING user_id = user() TO r_01889" -${CLICKHOUSE_CLIENT} -q "grant r_01889 to u_01889" -${CLICKHOUSE_CLIENT} -q "alter user u_01889 default role r_01889 settings none" +${CLICKHOUSE_CLIENT} -q "grant select on t_01889 to ${ROLE}" +${CLICKHOUSE_CLIENT} -q "create row policy ${POLICY} ON t_01889 FOR SELECT USING user_id = user() TO ${ROLE}" +${CLICKHOUSE_CLIENT} -q "grant ${ROLE} to $USER" +${CLICKHOUSE_CLIENT} -q "alter user $USER default role ${ROLE} settings none" -${CLICKHOUSE_CLIENT_BINARY} --database=${CLICKHOUSE_DATABASE} --user=u_01889 --password=dfsdffdf5t123 --query="select count() from t_01889" +${CLICKHOUSE_CLIENT_BINARY} --database=${CLICKHOUSE_DATABASE} --user=$USER --password=dfsdffdf5t123 --query="select count() from t_01889" -${CLICKHOUSE_CLIENT} -q "drop user u_01889" -${CLICKHOUSE_CLIENT} -q "drop policy t_01889_filter on t_01889" -${CLICKHOUSE_CLIENT} -q "drop role r_01889" +${CLICKHOUSE_CLIENT} -q "drop user $USER" +${CLICKHOUSE_CLIENT} -q "drop policy ${POLICY} on t_01889" +${CLICKHOUSE_CLIENT} -q "drop role ${ROLE}" ${CLICKHOUSE_CLIENT} -q "drop table t_01889" diff --git a/tests/queries/0_stateless/01889_sqlite_read_write.sh b/tests/queries/0_stateless/01889_sqlite_read_write.sh index 30496af46f6..63ce5dc909c 100755 --- a/tests/queries/0_stateless/01889_sqlite_read_write.sh +++ b/tests/queries/0_stateless/01889_sqlite_read_write.sh @@ -6,15 +6,9 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -# See 01658_read_file_to_string_column.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -mkdir -p "${user_files_path}/" -chmod 777 "${user_files_path}" - export CURR_DATABASE="test_01889_sqllite_${CLICKHOUSE_DATABASE}" -DB_PATH=${user_files_path}/${CURR_DATABASE}_db1 +DB_PATH=${USER_FILES_PATH}/${CURR_DATABASE}_db1 DB_PATH2=$CUR_DIR/${CURR_DATABASE}_db2 function cleanup() diff --git a/tests/queries/0_stateless/01903_ssd_cache_dictionary_array_type.sh b/tests/queries/0_stateless/01903_ssd_cache_dictionary_array_type.sh index 3676f1429b2..853445daf3f 100755 --- a/tests/queries/0_stateless/01903_ssd_cache_dictionary_array_type.sh +++ b/tests/queries/0_stateless/01903_ssd_cache_dictionary_array_type.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT -n --query=" DROP TABLE IF EXISTS dictionary_array_source_table; CREATE TABLE dictionary_array_source_table diff --git a/tests/queries/0_stateless/01904_dictionary_default_nullable_type.sql b/tests/queries/0_stateless/01904_dictionary_default_nullable_type.sql index 4c623941a19..d28f9e5c4e6 100644 --- a/tests/queries/0_stateless/01904_dictionary_default_nullable_type.sql +++ b/tests/queries/0_stateless/01904_dictionary_default_nullable_type.sql @@ -111,6 +111,8 @@ LAYOUT(IP_TRIE()); SELECT 'IPTrie dictionary'; SELECT dictGet('ip_trie_dictionary', 'value', tuple(IPv4StringToNum('127.0.0.0'))); --{serverError UNSUPPORTED_METHOD} +DROP DICTIONARY ip_trie_dictionary; + DROP TABLE dictionary_nullable_source_table; DROP TABLE dictionary_nullable_default_source_table; diff --git a/tests/queries/0_stateless/01904_ssd_cache_dictionary_default_nullable_type.sh b/tests/queries/0_stateless/01904_ssd_cache_dictionary_default_nullable_type.sh index 6aecb20329a..0b555cf82c2 100755 --- a/tests/queries/0_stateless/01904_ssd_cache_dictionary_default_nullable_type.sh +++ b/tests/queries/0_stateless/01904_ssd_cache_dictionary_default_nullable_type.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT -n --query=" DROP TABLE IF EXISTS dictionary_nullable_source_table; CREATE TABLE dictionary_nullable_source_table diff --git a/tests/queries/0_stateless/01910_view_dictionary.sql b/tests/queries/0_stateless/01910_view_dictionary.sql index 1f9928735b4..51f46decadd 100644 --- a/tests/queries/0_stateless/01910_view_dictionary.sql +++ b/tests/queries/0_stateless/01910_view_dictionary.sql @@ -34,7 +34,7 @@ CREATE DICTIONARY flat_dictionary value_ru String ) PRIMARY KEY id -SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 USER 'default' PASSWORD '' TABLE 'dictionary_source_view')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' PASSWORD '' TABLE 'dictionary_source_view')) LIFETIME(MIN 1 MAX 1000) LAYOUT(FLAT()); @@ -45,5 +45,5 @@ FROM numbers(3); DROP TABLE dictionary_source_en; DROP TABLE dictionary_source_ru; -DROP TABLE dictionary_source_view; DROP DICTIONARY flat_dictionary; +DROP TABLE dictionary_source_view; diff --git a/tests/queries/0_stateless/01921_concurrent_ttl_and_normal_merges_zookeeper_long.sh b/tests/queries/0_stateless/01921_concurrent_ttl_and_normal_merges_zookeeper_long.sh index 5e1600a0673..0dfde4997a7 100755 --- a/tests/queries/0_stateless/01921_concurrent_ttl_and_normal_merges_zookeeper_long.sh +++ b/tests/queries/0_stateless/01921_concurrent_ttl_and_normal_merges_zookeeper_long.sh @@ -50,7 +50,7 @@ function insert_thread export -f insert_thread; export -f optimize_thread; -TIMEOUT=30 +TIMEOUT=20 timeout $TIMEOUT bash -c insert_thread 2> /dev/null & timeout $TIMEOUT bash -c insert_thread 2> /dev/null & diff --git a/tests/queries/0_stateless/01923_network_receive_time_metric_insert.sh b/tests/queries/0_stateless/01923_network_receive_time_metric_insert.sh index 4d7e79fae52..a6b83eba27d 100755 --- a/tests/queries/0_stateless/01923_network_receive_time_metric_insert.sh +++ b/tests/queries/0_stateless/01923_network_receive_time_metric_insert.sh @@ -9,7 +9,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) ${CLICKHOUSE_CLIENT} --multiquery --query "DROP TABLE IF EXISTS t; CREATE TABLE t (x UInt64) ENGINE = Memory;" # Rate limit is chosen for operation to spent more than one second. -seq 1 1000 | pv --quiet --rate-limit 500 | ${CLICKHOUSE_CLIENT} --query "INSERT INTO t FORMAT TSV" +seq 1 1000 | pv --quiet --rate-limit 400 | ${CLICKHOUSE_CLIENT} --query "INSERT INTO t FORMAT TSV" # We check that the value of NetworkReceiveElapsedMicroseconds correctly includes the time spent waiting data from the client. ${CLICKHOUSE_CLIENT} --multiquery --query "SYSTEM FLUSH LOGS; diff --git a/tests/queries/0_stateless/01927_query_views_log_current_database.sql b/tests/queries/0_stateless/01927_query_views_log_current_database.sql index ba42795333c..6287156daaf 100644 --- a/tests/queries/0_stateless/01927_query_views_log_current_database.sql +++ b/tests/queries/0_stateless/01927_query_views_log_current_database.sql @@ -16,6 +16,7 @@ CREATE MATERIALIZED VIEW matview_b_to_c TO table_c AS SELECT SUM(a + sleepEachRo CREATE MATERIALIZED VIEW matview_join_d_e TO table_f AS SELECT table_d.a as a, table_e.count + sleepEachRow(0.000003) as count FROM table_d LEFT JOIN table_e ON table_d.a = table_e.a; -- ENABLE LOGS +SET parallel_view_processing=0; SET log_query_views=1; SET log_queries_min_type='QUERY_FINISH'; SET log_queries=1; diff --git a/tests/queries/0_stateless/01946_test_zstd_decompression_with_escape_sequence_at_the_end_of_buffer.sh b/tests/queries/0_stateless/01946_test_zstd_decompression_with_escape_sequence_at_the_end_of_buffer.sh index 4b230e4f738..0aedef028a2 100755 --- a/tests/queries/0_stateless/01946_test_zstd_decompression_with_escape_sequence_at_the_end_of_buffer.sh +++ b/tests/queries/0_stateless/01946_test_zstd_decompression_with_escape_sequence_at_the_end_of_buffer.sh @@ -6,10 +6,8 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh -# See 01658_read_file_to_string_column.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -mkdir -p ${user_files_path}/ -cp $CUR_DIR/data_zstd/test_01946.zstd ${user_files_path}/ +mkdir -p ${USER_FILES_PATH}/ +cp $CUR_DIR/data_zstd/test_01946.zstd ${USER_FILES_PATH}/ ${CLICKHOUSE_CLIENT} --multiline --multiquery --query " set min_chunk_bytes_for_parallel_parsing=10485760; diff --git a/tests/queries/0_stateless/02003_compress_bz2.sh b/tests/queries/0_stateless/02003_compress_bz2.sh index b17effb20b6..edc433ad69b 100755 --- a/tests/queries/0_stateless/02003_compress_bz2.sh +++ b/tests/queries/0_stateless/02003_compress_bz2.sh @@ -6,7 +6,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') WORKING_FOLDER_02003="${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}" rm -rf "${WORKING_FOLDER_02003}" diff --git a/tests/queries/0_stateless/02008_complex_key_range_hashed_dictionary.sql b/tests/queries/0_stateless/02008_complex_key_range_hashed_dictionary.sql index 72cac481376..ea2dad5c732 100644 --- a/tests/queries/0_stateless/02008_complex_key_range_hashed_dictionary.sql +++ b/tests/queries/0_stateless/02008_complex_key_range_hashed_dictionary.sql @@ -53,8 +53,8 @@ SELECT CountryID, StartDate, Tax FROM range_dictionary ORDER BY CountryID, Start SELECT 'onlySpecificColumn'; SELECT Tax FROM range_dictionary ORDER BY CountryID, StartDate, EndDate; -DROP TABLE date_table; DROP DICTIONARY range_dictionary; +DROP TABLE date_table; CREATE TABLE date_table ( @@ -107,5 +107,5 @@ SELECT CountryID, StartDate, Tax FROM range_dictionary_nullable ORDER BY Country SELECT 'onlySpecificColumn'; SELECT Tax FROM range_dictionary_nullable ORDER BY CountryID, StartDate, EndDate; -DROP TABLE date_table; DROP DICTIONARY range_dictionary_nullable; +DROP TABLE date_table; diff --git a/tests/queries/0_stateless/02010_array_index_bad_cast.reference b/tests/queries/0_stateless/02010_array_index_bad_cast.reference index e69de29bb2d..e22493782f0 100644 --- a/tests/queries/0_stateless/02010_array_index_bad_cast.reference +++ b/tests/queries/0_stateless/02010_array_index_bad_cast.reference @@ -0,0 +1,3 @@ +1 +0 +0 diff --git a/tests/queries/0_stateless/02010_array_index_bad_cast.sql b/tests/queries/0_stateless/02010_array_index_bad_cast.sql index 14162e0d2e2..590e60eb42e 100644 --- a/tests/queries/0_stateless/02010_array_index_bad_cast.sql +++ b/tests/queries/0_stateless/02010_array_index_bad_cast.sql @@ -1,3 +1,4 @@ --- This query throws exception about uncomparable data types (but at least it does not introduce bad cast in code). SET allow_suspicious_low_cardinality_types=1; -SELECT has(materialize(CAST(['2021-07-14'] AS Array(LowCardinality(Nullable(DateTime))))), materialize('2021-07-14'::DateTime64(7))); -- { serverError ILLEGAL_COLUMN } +SELECT has(materialize(CAST(['2021-07-14'] AS Array(LowCardinality(Nullable(DateTime))))), materialize('2021-07-14'::DateTime64(7))); +SELECT has(materialize(CAST(['2021-07-14'] AS Array(LowCardinality(Nullable(DateTime))))), materialize('2021-07-14 00:00:01'::DateTime64(7))); +SELECT has(materialize(CAST(['2021-07-14'] AS Array(LowCardinality(Nullable(DateTime))))), materialize(NULL)); diff --git a/tests/queries/0_stateless/02012_compress_lz4.sh b/tests/queries/0_stateless/02012_compress_lz4.sh index aad437c8011..700bff613da 100755 --- a/tests/queries/0_stateless/02012_compress_lz4.sh +++ b/tests/queries/0_stateless/02012_compress_lz4.sh @@ -4,7 +4,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') WORKING_FOLDER_02012="${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}" rm -rf "${WORKING_FOLDER_02012}" diff --git a/tests/queries/0_stateless/02016_bit_shift_right_for_string_integer.reference b/tests/queries/0_stateless/02016_bit_shift_right_for_string_integer.reference index e6a2b2b6aaf..ab832478da0 100644 --- a/tests/queries/0_stateless/02016_bit_shift_right_for_string_integer.reference +++ b/tests/queries/0_stateless/02016_bit_shift_right_for_string_integer.reference @@ -41,8 +41,6 @@ String ConstConst 38 Hello 00000001 39 Hello 00000000 40 Hello -41 Hello -42 Hello FixedString ConstConst 1 0 Hello\0\0\0\0\0 01001000011001010110110001101100011011110000000000000000000000000000000000000000 @@ -92,10 +90,8 @@ FixedString ConstConst 78 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000001 79 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 80 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 -81 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 String VectorVector --1 Hello 0 Hello 0100100001100101011011000110110001101111 1 Hello 0010010000110010101101100011011000110111 7 Hello 0000000010010000110010101101100011011000 @@ -112,8 +108,6 @@ String VectorVector 33 Hello 00100100 39 Hello 00000000 40 Hello -41 Hello -42 Hello 7 Hel 000000001001000011001010 8 Hel 0100100001100101 9 Hel 0010010000110010 @@ -125,7 +119,6 @@ String VectorVector 9 Hel 0010010000110010 FixedString VectorVector --1 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 0 Hello\0\0\0\0\0 01001000011001010110110001101100011011110000000000000000000000000000000000000000 1 Hello\0\0\0\0\0 00100100001100101011011000110110001101111000000000000000000000000000000000000000 7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 @@ -142,8 +135,6 @@ FixedString VectorVector 33 Hello\0\0\0\0\0 00000000000000000000000000000000001001000011001010110110001101100011011110000000 39 Hello\0\0\0\0\0 00000000000000000000000000000000000000001001000011001010110110001101100011011110 40 Hello\0\0\0\0\0 00000000000000000000000000000000000000000100100001100101011011000110110001101111 -41 Hello\0\0\0\0\0 00000000000000000000000000000000000000000010010000110010101101100011011000110111 -42 Hello\0\0\0\0\0 00000000000000000000000000000000000000000001001000011001010110110001101100011011 7 Hel\0\0\0\0\0\0\0 00000000100100001100101011011000000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 00000000010010000110010101101100000000000000000000000000000000000000000000000000 9 Hel\0\0\0\0\0\0\0 00000000001001000011001010110110000000000000000000000000000000000000000000000000 @@ -171,9 +162,6 @@ String VectorConst 7 Hello 0000000010010000110010101101100011011000 7 Hello 0000000010010000110010101101100011011000 7 Hello 0000000010010000110010101101100011011000 -7 Hello 0000000010010000110010101101100011011000 -7 Hello 0000000010010000110010101101100011011000 -7 Hello 0000000010010000110010101101100011011000 7 Hel 000000001001000011001010 7 Hel 000000001001000011001010 7 Hel 000000001001000011001010 @@ -193,9 +181,6 @@ String VectorConst 8 Hello 01001000011001010110110001101100 8 Hello 01001000011001010110110001101100 8 Hello 01001000011001010110110001101100 -8 Hello 01001000011001010110110001101100 -8 Hello 01001000011001010110110001101100 -8 Hello 01001000011001010110110001101100 8 Hel 0100100001100101 8 Hel 0100100001100101 8 Hel 0100100001100101 @@ -217,9 +202,6 @@ FixedString VectorConst 7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 -7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 -7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 -7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00000000100100001100101011011000000000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00000000100100001100101011011000000000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00000000100100001100101011011000000000000000000000000000000000000000000000000000 @@ -239,15 +221,11 @@ FixedString VectorConst 8 Hello\0\0\0\0\0 00000000010010000110010101101100011011000110111100000000000000000000000000000000 8 Hello\0\0\0\0\0 00000000010010000110010101101100011011000110111100000000000000000000000000000000 8 Hello\0\0\0\0\0 00000000010010000110010101101100011011000110111100000000000000000000000000000000 -8 Hello\0\0\0\0\0 00000000010010000110010101101100011011000110111100000000000000000000000000000000 -8 Hello\0\0\0\0\0 00000000010010000110010101101100011011000110111100000000000000000000000000000000 -8 Hello\0\0\0\0\0 00000000010010000110010101101100011011000110111100000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 00000000010010000110010101101100000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 00000000010010000110010101101100000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 00000000010010000110010101101100000000000000000000000000000000000000000000000000 String ConstVector --1 Hello 0 Hello 0100100001100101011011000110110001101111 1 Hello 0010010000110010101101100011011000110111 7 Hello 0000000010010000110010101101100011011000 @@ -264,12 +242,9 @@ String ConstVector 33 Hello 00100100 39 Hello 00000000 40 Hello -41 Hello -42 Hello 7 Hello 0000000010010000110010101101100011011000 8 Hello 01001000011001010110110001101100 9 Hello 00100100001100101011011000110110 --1 Hel 0 Hel 010010000110010101101100 1 Hel 001001000011001010110110 7 Hel 000000001001000011001010 @@ -280,20 +255,11 @@ String ConstVector 17 Hel 00100100 23 Hel 00000000 24 Hel -25 Hel -31 Hel -32 Hel -33 Hel -39 Hel -40 Hel -41 Hel -42 Hel 7 Hel 000000001001000011001010 8 Hel 0100100001100101 9 Hel 0010010000110010 FixedString ConstVector --1 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 0 Hello\0\0\0\0\0 01001000011001010110110001101100011011110000000000000000000000000000000000000000 1 Hello\0\0\0\0\0 00100100001100101011011000110110001101111000000000000000000000000000000000000000 7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 @@ -310,12 +276,9 @@ FixedString ConstVector 33 Hello\0\0\0\0\0 00000000000000000000000000000000001001000011001010110110001101100011011110000000 39 Hello\0\0\0\0\0 00000000000000000000000000000000000000001001000011001010110110001101100011011110 40 Hello\0\0\0\0\0 00000000000000000000000000000000000000000100100001100101011011000110110001101111 -41 Hello\0\0\0\0\0 00000000000000000000000000000000000000000010010000110010101101100011011000110111 -42 Hello\0\0\0\0\0 00000000000000000000000000000000000000000001001000011001010110110001101100011011 7 Hello\0\0\0\0\0 00000000100100001100101011011000110110001101111000000000000000000000000000000000 8 Hello\0\0\0\0\0 00000000010010000110010101101100011011000110111100000000000000000000000000000000 9 Hello\0\0\0\0\0 00000000001001000011001010110110001101100011011110000000000000000000000000000000 --1 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 0 Hel\0\0\0\0\0\0\0 01001000011001010110110000000000000000000000000000000000000000000000000000000000 1 Hel\0\0\0\0\0\0\0 00100100001100101011011000000000000000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00000000100100001100101011011000000000000000000000000000000000000000000000000000 @@ -332,8 +295,6 @@ FixedString ConstVector 33 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000001001000011001010110110000000000000000000000000 39 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000001001000011001010110110000000000000000000 40 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000100100001100101011011000000000000000000 -41 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000010010000110010101101100000000000000000 -42 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000001001000011001010110110000000000000000 7 Hel\0\0\0\0\0\0\0 00000000100100001100101011011000000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 00000000010010000110010101101100000000000000000000000000000000000000000000000000 9 Hel\0\0\0\0\0\0\0 00000000001001000011001010110110000000000000000000000000000000000000000000000000 diff --git a/tests/queries/0_stateless/02016_bit_shift_right_for_string_integer.sql b/tests/queries/0_stateless/02016_bit_shift_right_for_string_integer.sql index 0ee04e408ba..40fccbc89e6 100644 --- a/tests/queries/0_stateless/02016_bit_shift_right_for_string_integer.sql +++ b/tests/queries/0_stateless/02016_bit_shift_right_for_string_integer.sql @@ -41,8 +41,6 @@ SELECT 37,'Hello',bin(bitShiftRight('Hello', 37)); SELECT 38,'Hello',bin(bitShiftRight('Hello', 38)); SELECT 39,'Hello',bin(bitShiftRight('Hello', 39)); SELECT 40,'Hello',bin(bitShiftRight('Hello', 40)); -SELECT 41,'Hello',bin(bitShiftRight('Hello', 41)); -SELECT 42,'Hello',bin(bitShiftRight('Hello', 42)); SELECT 'FixedString ConstConst'; SELECT bin(toFixedString('Hello', 10)) == bin(bitShiftRight(toFixedString('Hello', 10), 0)); @@ -93,40 +91,39 @@ SELECT 77,toFixedString('Hello', 10), bin(bitShiftRight(toFixedString('Hello', 1 SELECT 78,toFixedString('Hello', 10), bin(bitShiftRight(toFixedString('Hello', 10), 78)); SELECT 79,toFixedString('Hello', 10), bin(bitShiftRight(toFixedString('Hello', 10), 79)); SELECT 80,toFixedString('Hello', 10), bin(bitShiftRight(toFixedString('Hello', 10), 80)); -SELECT 81,toFixedString('Hello', 10), bin(bitShiftRight(toFixedString('Hello', 10), 81)); DROP TABLE IF EXISTS test_bit_shift_right_string_integer; CREATE TABLE test_bit_shift_right_string_integer (str String, fixedStr FixedString(10), id Int64) engine=Log; -INSERT INTO test_bit_shift_right_string_integer VALUES('Hello','Hello',-1)('Hello','Hello',0),('Hello','Hello',1),('Hello','Hello',7),('Hello','Hello',8),('Hello','Hello',9),('Hello','Hello',15),('Hello','Hello',16),('Hello','Hello',17),('Hello','Hello',23),('Hello','Hello',24),('Hello','Hello',25),('Hello','Hello',31),('Hello','Hello',32),('Hello','Hello',33),('Hello','Hello',39),('Hello','Hello',40),('Hello','Hello',41),('Hello','Hello',42),('Hel','Hel',7),('Hel','Hel',8),('Hel','Hel',9); +INSERT INTO test_bit_shift_right_string_integer VALUES('Hello','Hello',0),('Hello','Hello',1),('Hello','Hello',7),('Hello','Hello',8),('Hello','Hello',9),('Hello','Hello',15),('Hello','Hello',16),('Hello','Hello',17),('Hello','Hello',23),('Hello','Hello',24),('Hello','Hello',25),('Hello','Hello',31),('Hello','Hello',32),('Hello','Hello',33),('Hello','Hello',39),('Hello','Hello',40),('Hel','Hel',7),('Hel','Hel',8),('Hel','Hel',9); -SELECT bin(bitShiftRight('Hello', 42)); --A blank line +SELECT bin(bitShiftRight('Hello', 40)); --A blank line SELECT 'String VectorVector'; SELECT id as shift_right_bit,str as arg,bin(bitShiftRight(str, id)) as string_res FROM test_bit_shift_right_string_integer; SELECT id as shift_right_bit,str as arg,bin(bitShiftRight(str, id)) as string_res FROM test_bit_shift_right_string_integer WHERE (str='Hello' AND (id=23 OR id=24 OR id=25)) OR (str='Hel' AND (id=7 OR id=8 OR id=9)); -SELECT bin(bitShiftRight('Hello', 42)); +SELECT bin(bitShiftRight('Hello', 40)); SELECT 'FixedString VectorVector'; SELECT id as shift_right_bit,fixedStr as arg,bin(bitShiftRight(fixedStr, id)) as fixed_string_res FROM test_bit_shift_right_string_integer; SELECT id as shift_right_bit,fixedStr as arg,bin(bitShiftRight(fixedStr, id)) as fixed_string_res FROM test_bit_shift_right_string_integer WHERE (str='Hello' AND (id=23 OR id=24 OR id=25)) OR (str='Hel' AND (id=7 OR id=8 OR id=9)); -SELECT bin(bitShiftRight('Hello', 42)); --A blank line +SELECT bin(bitShiftRight('Hello', 40)); --A blank line SELECT 'String VectorConst'; SELECT 7 as shift_right_bit,str as arg,bin(bitShiftRight(str, 7)) as string_res FROM test_bit_shift_right_string_integer; SELECT 8 as shift_right_bit,str as arg,bin(bitShiftRight(str, 8)) as string_res FROM test_bit_shift_right_string_integer; -SELECT bin(bitShiftRight('Hello', 42)); --A blank line +SELECT bin(bitShiftRight('Hello', 40)); --A blank line SELECT 'FixedString VectorConst'; SELECT 7 as shift_right_bit,fixedStr as arg,bin(bitShiftRight(fixedStr, 7)) as fixed_string_res FROM test_bit_shift_right_string_integer; SELECT 8 as shift_right_bit,fixedStr as arg,bin(bitShiftRight(fixedStr, 8)) as fixed_string_res FROM test_bit_shift_right_string_integer; -SELECT bin(bitShiftRight('Hello', 42)); --A blank line +SELECT bin(bitShiftRight('Hello', 40)); --A blank line SELECT 'String ConstVector'; SELECT id as shift_right_bit,'Hello' as arg,bin(bitShiftRight('Hello', id)) as string_res FROM test_bit_shift_right_string_integer; -SELECT id as shift_right_bit,'Hel' as arg,bin(bitShiftRight('Hel', id)) as string_res FROM test_bit_shift_right_string_integer; +SELECT id as shift_right_bit,'Hel' as arg,bin(bitShiftRight('Hel', id)) as string_res FROM test_bit_shift_right_string_integer WHERE id <= 8 * 3; -SELECT bin(bitShiftRight('Hello', 42)); --A blank line +SELECT bin(bitShiftRight('Hello', 40)); --A blank line SELECT 'FixedString ConstVector'; SELECT id as shift_right_bit,toFixedString('Hello', 10) as arg,bin(bitShiftRight(toFixedString('Hello', 10), id)) as fixed_string_res FROM test_bit_shift_right_string_integer; SELECT id as shift_right_bit,toFixedString('Hel', 10) as arg,bin(bitShiftRight(toFixedString('Hel', 10), id)) as fixed_string_res FROM test_bit_shift_right_string_integer; diff --git a/tests/queries/0_stateless/02017_bit_shift_left_for_string_integer.reference b/tests/queries/0_stateless/02017_bit_shift_left_for_string_integer.reference index ff5a09c0d48..a20c44bbb9a 100644 --- a/tests/queries/0_stateless/02017_bit_shift_left_for_string_integer.reference +++ b/tests/queries/0_stateless/02017_bit_shift_left_for_string_integer.reference @@ -41,8 +41,6 @@ String ConstConst 38 Hello 00010010000110010101101100011011000110111100000000000000000000000000000000000000 39 Hello 00100100001100101011011000110110001101111000000000000000000000000000000000000000 40 Hello -41 Hello -42 Hello FixedString ConstConst 1 0 Hello\0\0\0\0\0 01001000011001010110110001101100011011110000000000000000000000000000000000000000 @@ -92,10 +90,8 @@ FixedString ConstConst 78 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 79 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 80 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 -81 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 String VectorVector --1 Hello 0 Hello 0100100001100101011011000110110001101111 1 Hello 000000001001000011001010110110001101100011011110 7 Hello 001001000011001010110110001101100011011110000000 @@ -112,8 +108,6 @@ String VectorVector 33 Hello 00000000100100001100101011011000110110001101111000000000000000000000000000000000 39 Hello 00100100001100101011011000110110001101111000000000000000000000000000000000000000 40 Hello -41 Hello -42 Hello 7 Hel 00100100001100101011011000000000 8 Hel 01001000011001010110110000000000 9 Hel 0000000010010000110010101101100000000000 @@ -125,7 +119,6 @@ String VectorVector 9 Hel 0000000010010000110010101101100000000000 FixedString VectorVector --1 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 0 Hello\0\0\0\0\0 01001000011001010110110001101100011011110000000000000000000000000000000000000000 1 Hello\0\0\0\0\0 10010000110010101101100011011000110111100000000000000000000000000000000000000000 7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 @@ -142,8 +135,6 @@ FixedString VectorVector 33 Hello\0\0\0\0\0 11011110000000000000000000000000000000000000000000000000000000000000000000000000 39 Hello\0\0\0\0\0 10000000000000000000000000000000000000000000000000000000000000000000000000000000 40 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 -41 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 -42 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00110010101101100000000000000000000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 01100101011011000000000000000000000000000000000000000000000000000000000000000000 9 Hel\0\0\0\0\0\0\0 11001010110110000000000000000000000000000000000000000000000000000000000000000000 @@ -171,9 +162,6 @@ String VectorConst 7 Hello 001001000011001010110110001101100011011110000000 7 Hello 001001000011001010110110001101100011011110000000 7 Hello 001001000011001010110110001101100011011110000000 -7 Hello 001001000011001010110110001101100011011110000000 -7 Hello 001001000011001010110110001101100011011110000000 -7 Hello 001001000011001010110110001101100011011110000000 7 Hel 00100100001100101011011000000000 7 Hel 00100100001100101011011000000000 7 Hel 00100100001100101011011000000000 @@ -193,9 +181,6 @@ String VectorConst 8 Hello 010010000110010101101100011011000110111100000000 8 Hello 010010000110010101101100011011000110111100000000 8 Hello 010010000110010101101100011011000110111100000000 -8 Hello 010010000110010101101100011011000110111100000000 -8 Hello 010010000110010101101100011011000110111100000000 -8 Hello 010010000110010101101100011011000110111100000000 8 Hel 01001000011001010110110000000000 8 Hel 01001000011001010110110000000000 8 Hel 01001000011001010110110000000000 @@ -217,9 +202,6 @@ FixedString VectorConst 7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 -7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 -7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 -7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00110010101101100000000000000000000000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00110010101101100000000000000000000000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00110010101101100000000000000000000000000000000000000000000000000000000000000000 @@ -239,15 +221,11 @@ FixedString VectorConst 8 Hello\0\0\0\0\0 01100101011011000110110001101111000000000000000000000000000000000000000000000000 8 Hello\0\0\0\0\0 01100101011011000110110001101111000000000000000000000000000000000000000000000000 8 Hello\0\0\0\0\0 01100101011011000110110001101111000000000000000000000000000000000000000000000000 -8 Hello\0\0\0\0\0 01100101011011000110110001101111000000000000000000000000000000000000000000000000 -8 Hello\0\0\0\0\0 01100101011011000110110001101111000000000000000000000000000000000000000000000000 -8 Hello\0\0\0\0\0 01100101011011000110110001101111000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 01100101011011000000000000000000000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 01100101011011000000000000000000000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 01100101011011000000000000000000000000000000000000000000000000000000000000000000 String ConstVector --1 Hello 0 Hello 0100100001100101011011000110110001101111 1 Hello 000000001001000011001010110110001101100011011110 7 Hello 001001000011001010110110001101100011011110000000 @@ -264,12 +242,9 @@ String ConstVector 33 Hello 00000000100100001100101011011000110110001101111000000000000000000000000000000000 39 Hello 00100100001100101011011000110110001101111000000000000000000000000000000000000000 40 Hello -41 Hello -42 Hello 7 Hello 001001000011001010110110001101100011011110000000 8 Hello 010010000110010101101100011011000110111100000000 9 Hello 00000000100100001100101011011000110110001101111000000000 --1 Hel 0 Hel 010010000110010101101100 1 Hel 00000000100100001100101011011000 7 Hel 00100100001100101011011000000000 @@ -280,20 +255,11 @@ String ConstVector 17 Hel 000000001001000011001010110110000000000000000000 23 Hel 001001000011001010110110000000000000000000000000 24 Hel -25 Hel -31 Hel -32 Hel -33 Hel -39 Hel -40 Hel -41 Hel -42 Hel 7 Hel 00100100001100101011011000000000 8 Hel 01001000011001010110110000000000 9 Hel 0000000010010000110010101101100000000000 FixedString ConstVector --1 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 0 Hello\0\0\0\0\0 01001000011001010110110001101100011011110000000000000000000000000000000000000000 1 Hello\0\0\0\0\0 10010000110010101101100011011000110111100000000000000000000000000000000000000000 7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 @@ -310,12 +276,9 @@ FixedString ConstVector 33 Hello\0\0\0\0\0 11011110000000000000000000000000000000000000000000000000000000000000000000000000 39 Hello\0\0\0\0\0 10000000000000000000000000000000000000000000000000000000000000000000000000000000 40 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 -41 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 -42 Hello\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 7 Hello\0\0\0\0\0 00110010101101100011011000110111100000000000000000000000000000000000000000000000 8 Hello\0\0\0\0\0 01100101011011000110110001101111000000000000000000000000000000000000000000000000 9 Hello\0\0\0\0\0 11001010110110001101100011011110000000000000000000000000000000000000000000000000 --1 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 0 Hel\0\0\0\0\0\0\0 01001000011001010110110000000000000000000000000000000000000000000000000000000000 1 Hel\0\0\0\0\0\0\0 10010000110010101101100000000000000000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00110010101101100000000000000000000000000000000000000000000000000000000000000000 @@ -332,8 +295,6 @@ FixedString ConstVector 33 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 39 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 40 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 -41 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 -42 Hel\0\0\0\0\0\0\0 00000000000000000000000000000000000000000000000000000000000000000000000000000000 7 Hel\0\0\0\0\0\0\0 00110010101101100000000000000000000000000000000000000000000000000000000000000000 8 Hel\0\0\0\0\0\0\0 01100101011011000000000000000000000000000000000000000000000000000000000000000000 9 Hel\0\0\0\0\0\0\0 11001010110110000000000000000000000000000000000000000000000000000000000000000000 diff --git a/tests/queries/0_stateless/02017_bit_shift_left_for_string_integer.sql b/tests/queries/0_stateless/02017_bit_shift_left_for_string_integer.sql index 5c7a9901dae..a8e66eda281 100644 --- a/tests/queries/0_stateless/02017_bit_shift_left_for_string_integer.sql +++ b/tests/queries/0_stateless/02017_bit_shift_left_for_string_integer.sql @@ -41,8 +41,6 @@ SELECT 37,'Hello',bin(bitShiftLeft('Hello', 37)); SELECT 38,'Hello',bin(bitShiftLeft('Hello', 38)); SELECT 39,'Hello',bin(bitShiftLeft('Hello', 39)); SELECT 40,'Hello',bin(bitShiftLeft('Hello', 40)); -SELECT 41,'Hello',bin(bitShiftLeft('Hello', 41)); -SELECT 42,'Hello',bin(bitShiftLeft('Hello', 42)); SELECT 'FixedString ConstConst'; SELECT bin(toFixedString('Hello', 10)) == bin(bitShiftLeft(toFixedString('Hello', 10), 0)); @@ -93,40 +91,39 @@ SELECT 77,toFixedString('Hello', 10), bin(bitShiftLeft(toFixedString('Hello', 10 SELECT 78,toFixedString('Hello', 10), bin(bitShiftLeft(toFixedString('Hello', 10), 78)); SELECT 79,toFixedString('Hello', 10), bin(bitShiftLeft(toFixedString('Hello', 10), 79)); SELECT 80,toFixedString('Hello', 10), bin(bitShiftLeft(toFixedString('Hello', 10), 80)); -SELECT 81,toFixedString('Hello', 10), bin(bitShiftLeft(toFixedString('Hello', 10), 81)); DROP TABLE IF EXISTS test_bit_shift_left_string_integer; CREATE TABLE test_bit_shift_left_string_integer (str String, fixedStr FixedString(10), id Int64) engine=Log; -INSERT INTO test_bit_shift_left_string_integer VALUES('Hello','Hello',-1)('Hello','Hello',0),('Hello','Hello',1),('Hello','Hello',7),('Hello','Hello',8),('Hello','Hello',9),('Hello','Hello',15),('Hello','Hello',16),('Hello','Hello',17),('Hello','Hello',23),('Hello','Hello',24),('Hello','Hello',25),('Hello','Hello',31),('Hello','Hello',32),('Hello','Hello',33),('Hello','Hello',39),('Hello','Hello',40),('Hello','Hello',41),('Hello','Hello',42),('Hel','Hel',7),('Hel','Hel',8),('Hel','Hel',9); +INSERT INTO test_bit_shift_left_string_integer VALUES('Hello','Hello',0),('Hello','Hello',1),('Hello','Hello',7),('Hello','Hello',8),('Hello','Hello',9),('Hello','Hello',15),('Hello','Hello',16),('Hello','Hello',17),('Hello','Hello',23),('Hello','Hello',24),('Hello','Hello',25),('Hello','Hello',31),('Hello','Hello',32),('Hello','Hello',33),('Hello','Hello',39),('Hello','Hello',40),('Hel','Hel',7),('Hel','Hel',8),('Hel','Hel',9); -SELECT bin(bitShiftLeft('Hello', 42)); --A blank line +SELECT bin(bitShiftLeft('Hello', 40)); --A blank line SELECT 'String VectorVector'; SELECT id as shift_right_bit,str as arg,bin(bitShiftLeft(str, id)) as string_res FROM test_bit_shift_left_string_integer; SELECT id as shift_right_bit,str as arg,bin(bitShiftLeft(str, id)) as string_res FROM test_bit_shift_left_string_integer WHERE (str='Hello' AND (id=23 OR id=24 OR id=25)) OR (str='Hel' AND (id=7 OR id=8 OR id=9)); -SELECT bin(bitShiftLeft('Hello', 42)); +SELECT bin(bitShiftLeft('Hello', 40)); SELECT 'FixedString VectorVector'; SELECT id as shift_right_bit,fixedStr as arg,bin(bitShiftLeft(fixedStr, id)) as fixed_string_res FROM test_bit_shift_left_string_integer; SELECT id as shift_right_bit,fixedStr as arg,bin(bitShiftLeft(fixedStr, id)) as fixed_string_res FROM test_bit_shift_left_string_integer WHERE (str='Hello' AND (id=23 OR id=24 OR id=25)) OR (str='Hel' AND (id=7 OR id=8 OR id=9)); -SELECT bin(bitShiftLeft('Hello', 42)); --A blank line +SELECT bin(bitShiftLeft('Hello', 40)); --A blank line SELECT 'String VectorConst'; SELECT 7 as shift_right_bit,str as arg,bin(bitShiftLeft(str, 7)) as string_res FROM test_bit_shift_left_string_integer; SELECT 8 as shift_right_bit,str as arg,bin(bitShiftLeft(str, 8)) as string_res FROM test_bit_shift_left_string_integer; -SELECT bin(bitShiftLeft('Hello', 42)); --A blank line +SELECT bin(bitShiftLeft('Hello', 40)); --A blank line SELECT 'FixedString VectorConst'; SELECT 7 as shift_right_bit,fixedStr as arg,bin(bitShiftLeft(fixedStr, 7)) as fixed_string_res FROM test_bit_shift_left_string_integer; SELECT 8 as shift_right_bit,fixedStr as arg,bin(bitShiftLeft(fixedStr, 8)) as fixed_string_res FROM test_bit_shift_left_string_integer; -SELECT bin(bitShiftLeft('Hello', 42)); --A blank line +SELECT bin(bitShiftLeft('Hello', 40)); --A blank line SELECT 'String ConstVector'; SELECT id as shift_right_bit,'Hello' as arg,bin(bitShiftLeft('Hello', id)) as string_res FROM test_bit_shift_left_string_integer; -SELECT id as shift_right_bit,'Hel' as arg,bin(bitShiftLeft('Hel', id)) as string_res FROM test_bit_shift_left_string_integer; +SELECT id as shift_right_bit,'Hel' as arg,bin(bitShiftLeft('Hel', id)) as string_res FROM test_bit_shift_left_string_integer WHERE id <= 8 * 3; -SELECT bin(bitShiftLeft('Hello', 42)); --A blank line +SELECT bin(bitShiftLeft('Hello', 40)); --A blank line SELECT 'FixedString ConstVector'; SELECT id as shift_right_bit,toFixedString('Hello', 10) as arg,bin(bitShiftLeft(toFixedString('Hello', 10), id)) as fixed_string_res FROM test_bit_shift_left_string_integer; SELECT id as shift_right_bit,toFixedString('Hel', 10) as arg,bin(bitShiftLeft(toFixedString('Hel', 10), id)) as fixed_string_res FROM test_bit_shift_left_string_integer; diff --git a/tests/queries/0_stateless/02022_storage_filelog_one_file.sh b/tests/queries/0_stateless/02022_storage_filelog_one_file.sh index ea703d69aa5..29bc28849c8 100755 --- a/tests/queries/0_stateless/02022_storage_filelog_one_file.sh +++ b/tests/queries/0_stateless/02022_storage_filelog_one_file.sh @@ -6,30 +6,26 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') for i in {1..20} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt done ${CLICKHOUSE_CLIENT} --query "drop table if exists file_log;" -${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt', 'CSV');" +${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt', 'CSV');" ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" for i in {100..120} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt done ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" # touch does not change file content, no event -touch ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt +touch ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" ${CLICKHOUSE_CLIENT} --query "detach table file_log;" @@ -40,4 +36,4 @@ ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_ ${CLICKHOUSE_CLIENT} --query "drop table file_log;" -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}.txt diff --git a/tests/queries/0_stateless/02023_storage_filelog.sh b/tests/queries/0_stateless/02023_storage_filelog.sh index 51c8dc8ab3e..798ffe66ed9 100755 --- a/tests/queries/0_stateless/02023_storage_filelog.sh +++ b/tests/queries/0_stateless/02023_storage_filelog.sh @@ -6,57 +6,52 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ - -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* for i in {1..20} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt done ${CLICKHOUSE_CLIENT} --query "drop table if exists file_log;" -${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" +${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.txt ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" for i in {100..120} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt done # touch does not change file content, no event -touch ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt +touch ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/c.txt -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/e.txt -mv ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/j.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/c.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/e.txt +mv ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/j.txt -rm ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt +rm ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" ${CLICKHOUSE_CLIENT} --query "detach table file_log;" -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/e.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/f.txt -mv ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/e.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/g.txt -mv ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/c.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/h.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/e.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/f.txt +mv ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/e.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/g.txt +mv ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/c.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/h.txt for i in {150..200} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/h.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/h.txt done for i in {200..250} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/i.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/i.txt done ${CLICKHOUSE_CLIENT} --query "attach table file_log;" @@ -68,11 +63,11 @@ ${CLICKHOUSE_CLIENT} --query "attach table file_log;" # should no records return ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" -truncate ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt --size 0 +truncate ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt --size 0 # exception happend ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" 2>&1 | grep -q "Code: 33" && echo 'OK' || echo 'FAIL' ${CLICKHOUSE_CLIENT} --query "drop table file_log;" -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} diff --git a/tests/queries/0_stateless/02024_storage_filelog_mv.sh b/tests/queries/0_stateless/02024_storage_filelog_mv.sh index 33c8693648c..a47ca6c7507 100755 --- a/tests/queries/0_stateless/02024_storage_filelog_mv.sh +++ b/tests/queries/0_stateless/02024_storage_filelog_mv.sh @@ -7,21 +7,16 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* for i in {1..20} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt done ${CLICKHOUSE_CLIENT} --query "drop table if exists file_log;" -${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" +${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" ${CLICKHOUSE_CLIENT} --query "drop table if exists mv;" ${CLICKHOUSE_CLIENT} --query "create Materialized View mv engine=MergeTree order by k as select * from file_log;" @@ -39,17 +34,17 @@ done ${CLICKHOUSE_CLIENT} --query "select * from mv order by k;" -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.txt # touch does not change file content, no event -touch ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt +touch ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/c.txt -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/c.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt for i in {100..120} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt done while true; do @@ -62,4 +57,4 @@ ${CLICKHOUSE_CLIENT} --query "select * from mv order by k;" ${CLICKHOUSE_CLIENT} --query "drop table mv;" ${CLICKHOUSE_CLIENT} --query "drop table file_log;" -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} diff --git a/tests/queries/0_stateless/02025_storage_filelog_virtual_col.sh b/tests/queries/0_stateless/02025_storage_filelog_virtual_col.sh index f027b61c3ef..7e414b8863e 100755 --- a/tests/queries/0_stateless/02025_storage_filelog_virtual_col.sh +++ b/tests/queries/0_stateless/02025_storage_filelog_virtual_col.sh @@ -6,42 +6,37 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ - -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* for i in {1..20} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt done ${CLICKHOUSE_CLIENT} --query "drop table if exists file_log;" -${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" +${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" ${CLICKHOUSE_CLIENT} --query "select *, _filename, _offset from file_log order by _filename, _offset settings stream_like_engine_allow_direct_select=1;" -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.txt ${CLICKHOUSE_CLIENT} --query "select *, _filename, _offset from file_log order by _filename, _offset settings stream_like_engine_allow_direct_select=1;" for i in {100..120} do - echo $i, $i >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt + echo $i, $i >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt done # touch does not change file content, no event -touch ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt +touch ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/c.txt -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt -cp ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/e.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/c.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt +cp ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/e.txt -rm ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt +rm ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/d.txt ${CLICKHOUSE_CLIENT} --query "select *, _filename, _offset from file_log order by _filename, _offset settings stream_like_engine_allow_direct_select=1;" @@ -51,11 +46,11 @@ ${CLICKHOUSE_CLIENT} --query "attach table file_log;" # should no records return ${CLICKHOUSE_CLIENT} --query "select *, _filename, _offset from file_log order by _filename, _offset settings stream_like_engine_allow_direct_select=1;" -truncate ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt --size 0 +truncate ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.txt --size 0 # exception happend ${CLICKHOUSE_CLIENT} --query "select * from file_log order by k settings stream_like_engine_allow_direct_select=1;" 2>&1 | grep -q "Code: 33" && echo 'OK' || echo 'FAIL' ${CLICKHOUSE_CLIENT} --query "drop table file_log;" -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} diff --git a/tests/queries/0_stateless/02026_storage_filelog_largefile.sh b/tests/queries/0_stateless/02026_storage_filelog_largefile.sh index b0a9a4357f3..b7377fb026a 100755 --- a/tests/queries/0_stateless/02026_storage_filelog_largefile.sh +++ b/tests/queries/0_stateless/02026_storage_filelog_largefile.sh @@ -7,34 +7,27 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ - -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* - -chmod 777 ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +chmod 777 ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ for i in {1..10} do - ${CLICKHOUSE_CLIENT} --query "insert into function file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/test$i.csv', 'CSV', 'k UInt32, v UInt32') select number, number from numbers(10000);" + ${CLICKHOUSE_CLIENT} --query "insert into function file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/test$i.csv', 'CSV', 'k UInt32, v UInt32') select number, number from numbers(10000);" done ${CLICKHOUSE_CLIENT} --query "drop table if exists file_log;" -${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt32, v UInt32) engine=FileLog('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" +${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt32, v UInt32) engine=FileLog('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" ${CLICKHOUSE_CLIENT} --query "select count() from file_log settings stream_like_engine_allow_direct_select=1;" for i in {11..20} do - ${CLICKHOUSE_CLIENT} --query "insert into function file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/test$i.csv', 'CSV', 'k UInt32, v UInt32') select number, number from numbers(10000);" + ${CLICKHOUSE_CLIENT} --query "insert into function file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/test$i.csv', 'CSV', 'k UInt32, v UInt32') select number, number from numbers(10000);" done ${CLICKHOUSE_CLIENT} --query "select count() from file_log settings stream_like_engine_allow_direct_select=1;" ${CLICKHOUSE_CLIENT} --query "drop table file_log;" -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} diff --git a/tests/queries/0_stateless/02030_capnp_format.sh b/tests/queries/0_stateless/02030_capnp_format.sh index 6fcfef23cc7..b34bdc9f670 100755 --- a/tests/queries/0_stateless/02030_capnp_format.sh +++ b/tests/queries/0_stateless/02030_capnp_format.sh @@ -1,17 +1,16 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel, no-replicated-database +# Tags: no-fasttest, no-replicated-database CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') CAPN_PROTO_FILE=$USER_FILES_PATH/data.capnp touch $CAPN_PROTO_FILE -SCHEMADIR=$($CLICKHOUSE_CLIENT_BINARY --query "select * from file('data.capnp', 'CapnProto', 'val1 char') settings format_schema='nonexist:Message'" 2>&1 | grep Exception | grep -oP "file \K.*(?=/nonexist.capnp)") +SCHEMADIR=${CLICKHOUSE_SCHEMA_FILES} CLIENT_SCHEMADIR=$CURDIR/format_schemas -SERVER_SCHEMADIR=test_02030 +SERVER_SCHEMADIR=${CLICKHOUSE_DATABASE} mkdir -p $SCHEMADIR/$SERVER_SCHEMADIR cp -r $CLIENT_SCHEMADIR/02030_* $SCHEMADIR/$SERVER_SCHEMADIR/ diff --git a/tests/queries/0_stateless/02044_url_glob_parallel_connection_refused.sh b/tests/queries/0_stateless/02044_url_glob_parallel_connection_refused.sh index 7e8579f7cbe..b4b0ee8a023 100755 --- a/tests/queries/0_stateless/02044_url_glob_parallel_connection_refused.sh +++ b/tests/queries/0_stateless/02044_url_glob_parallel_connection_refused.sh @@ -8,7 +8,10 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) i=0 retries=5 # Connecting to wrong address and checking for race condition +# http_max_tries is limited to 2 because with the default 10 retries the execution time might go as high as around 3 minutes (because of exponential back-off). +# because of that we might see wrong 'tests hung' reports depending on how close to the end of tests run this particular test was executed. +# proper fix should be implemented in https://github.com/ClickHouse/ClickHouse/issues/66656 while [[ $i -lt $retries ]]; do - timeout 5s ${CLICKHOUSE_CLIENT} --max_threads 10 --query "SELECT * FROM url('http://128.0.0.{1..10}:${CLICKHOUSE_PORT_HTTP}/?query=SELECT+sleep(1)', TSV, 'x UInt8')" --format Null 2>/dev/null + timeout 5s ${CLICKHOUSE_CLIENT} --max_threads 10 --http_max_tries 2 --query "SELECT * FROM url('http://128.0.0.{1..10}:${CLICKHOUSE_PORT_HTTP}/?query=SELECT+sleep(1)', TSV, 'x UInt8')" --format Null 2>/dev/null ((++i)) done diff --git a/tests/queries/0_stateless/02051_symlinks_to_user_files.sh b/tests/queries/0_stateless/02051_symlinks_to_user_files.sh index eab44e74d88..afdb14c4191 100755 --- a/tests/queries/0_stateless/02051_symlinks_to_user_files.sh +++ b/tests/queries/0_stateless/02051_symlinks_to_user_files.sh @@ -5,10 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -# See 01658_read_file_to_string_column.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -FILE_PATH="${user_files_path}/file" +FILE_PATH="${USER_FILES_PATH}/file" mkdir -p ${FILE_PATH} chmod 777 ${FILE_PATH} diff --git a/tests/queries/0_stateless/02103_tsv_csv_custom_null_representation.sh b/tests/queries/0_stateless/02103_tsv_csv_custom_null_representation.sh index 7ea6739e932..25ad4f37e45 100755 --- a/tests/queries/0_stateless/02103_tsv_csv_custom_null_representation.sh +++ b/tests/queries/0_stateless/02103_tsv_csv_custom_null_representation.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -DATA_FILE=$CLICKHOUSE_TMP/test_02103_null.data +DATA_FILE=${CLICKHOUSE_TMP}/${CLICKHOUSE_DATABASE}_null.data # Wrapper for clickhouse-client to always output in JSONEachRow format, that # way format settings will not affect output. diff --git a/tests/queries/0_stateless/02103_with_names_and_types_parallel_parsing.sh b/tests/queries/0_stateless/02103_with_names_and_types_parallel_parsing.sh index a6e704093a2..07d40539358 100755 --- a/tests/queries/0_stateless/02103_with_names_and_types_parallel_parsing.sh +++ b/tests/queries/0_stateless/02103_with_names_and_types_parallel_parsing.sh @@ -1,20 +1,17 @@ #!/usr/bin/env bash -# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$(clickhouse client --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -DATA_FILE=$USER_FILES_PATH/test_02103.data +DATA_FILE=$USER_FILES_PATH/${CLICKHOUSE_DATABASE}.data FORMATS=('TSVWithNames' 'TSVWithNamesAndTypes' 'TSVRawWithNames' 'TSVRawWithNamesAndTypes' 'CSVWithNames' 'CSVWithNamesAndTypes' 'JSONCompactEachRowWithNames' 'JSONCompactEachRowWithNamesAndTypes') for format in "${FORMATS[@]}" do $CLICKHOUSE_CLIENT -q "SELECT number, range(number + 10) AS array, toString(number) AS string FROM numbers(10) FORMAT $format" > $DATA_FILE - $CLICKHOUSE_CLIENT -q "SELECT * FROM file('test_02103.data', '$format', 'number UInt64, array Array(UInt64), string String') ORDER BY number SETTINGS input_format_parallel_parsing=1, min_chunk_bytes_for_parallel_parsing=40" + $CLICKHOUSE_CLIENT -q "SELECT * FROM file('${CLICKHOUSE_DATABASE}.data', '$format', 'number UInt64, array Array(UInt64), string String') ORDER BY number SETTINGS input_format_parallel_parsing=1, min_chunk_bytes_for_parallel_parsing=40" done rm $DATA_FILE diff --git a/tests/queries/0_stateless/02104_json_strings_nullable_string.sh b/tests/queries/0_stateless/02104_json_strings_nullable_string.sh index b3b156b5787..d46b0704b0e 100755 --- a/tests/queries/0_stateless/02104_json_strings_nullable_string.sh +++ b/tests/queries/0_stateless/02104_json_strings_nullable_string.sh @@ -5,7 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') DATA_FILE=$USER_FILES_PATH/test_02104_null.data echo -e '{"s" : "NULLSome string"}' > $DATA_FILE diff --git a/tests/queries/0_stateless/02105_table_function_file_partiotion_by.sh b/tests/queries/0_stateless/02105_table_function_file_partiotion_by.sh index c79b5d0eee5..38c68ca0005 100755 --- a/tests/queries/0_stateless/02105_table_function_file_partiotion_by.sh +++ b/tests/queries/0_stateless/02105_table_function_file_partiotion_by.sh @@ -5,13 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -# See 01658_read_file_to_string_column.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -mkdir -p "${user_files_path}/" -chmod 777 ${user_files_path} - -FILE_PATH="${user_files_path}/test_table_function_file" +FILE_PATH="${USER_FILES_PATH}/test_table_function_file" function cleanup() { diff --git a/tests/queries/0_stateless/02115_map_contains.reference b/tests/queries/0_stateless/02115_map_contains.reference deleted file mode 100644 index 975e9876237..00000000000 --- a/tests/queries/0_stateless/02115_map_contains.reference +++ /dev/null @@ -1,4 +0,0 @@ -SELECT has(`m.keys`, \'a\') -FROM t_map_contains -1 -0 diff --git a/tests/queries/0_stateless/02115_map_contains_analyzer.reference b/tests/queries/0_stateless/02115_map_contains_analyzer.reference new file mode 100644 index 00000000000..7da5243e727 --- /dev/null +++ b/tests/queries/0_stateless/02115_map_contains_analyzer.reference @@ -0,0 +1,4 @@ +SELECT has(__table1.`m.keys`, \'a\') AS `mapContains(m, \'a\')` +FROM default.t_map_contains AS __table1 +1 +0 diff --git a/tests/queries/0_stateless/02115_map_contains.sql b/tests/queries/0_stateless/02115_map_contains_analyzer.sql similarity index 70% rename from tests/queries/0_stateless/02115_map_contains.sql rename to tests/queries/0_stateless/02115_map_contains_analyzer.sql index 3c7f21cb4f1..46e02eca4f0 100644 --- a/tests/queries/0_stateless/02115_map_contains.sql +++ b/tests/queries/0_stateless/02115_map_contains_analyzer.sql @@ -5,8 +5,9 @@ CREATE TABLE t_map_contains (m Map(String, UInt32)) ENGINE = Memory; INSERT INTO t_map_contains VALUES (map('a', 1, 'b', 2)), (map('c', 3, 'd', 4)); SET optimize_functions_to_subcolumns = 1; +SET allow_experimental_analyzer = 1; -EXPLAIN SYNTAX SELECT mapContains(m, 'a') FROM t_map_contains; +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT mapContains(m, 'a') FROM t_map_contains; SELECT mapContains(m, 'a') FROM t_map_contains; DROP TABLE t_map_contains; diff --git a/tests/queries/0_stateless/02115_write_buffers_finalize.sh b/tests/queries/0_stateless/02115_write_buffers_finalize.sh index d8a3c29bbbd..fb3a77f6105 100755 --- a/tests/queries/0_stateless/02115_write_buffers_finalize.sh +++ b/tests/queries/0_stateless/02115_write_buffers_finalize.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel +# Tags: no-fasttest # Tag no-fasttest: depends on brotli and bzip2 CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) diff --git a/tests/queries/0_stateless/02116_tuple_element.reference b/tests/queries/0_stateless/02116_tuple_element.reference deleted file mode 100644 index 121b08d02f1..00000000000 --- a/tests/queries/0_stateless/02116_tuple_element.reference +++ /dev/null @@ -1,25 +0,0 @@ -1 -SELECT `t1.a` -FROM t_tuple_element -a -SELECT `t1.s` -FROM t_tuple_element -1 -SELECT `t1.a` -FROM t_tuple_element -2 -SELECT `t2.1` -FROM t_tuple_element -2 -SELECT `t2.1` -FROM t_tuple_element -1 2 -WITH (1, 2) AS t -SELECT - t.1, - t.2 -1 2 -WITH CAST(\'(1, 2)\', \'Tuple(a UInt32, b UInt32)\') AS t -SELECT - t.1, - tupleElement(t, \'b\') diff --git a/tests/queries/0_stateless/02116_tuple_element_analyzer.reference b/tests/queries/0_stateless/02116_tuple_element_analyzer.reference new file mode 100644 index 00000000000..22d48ffb2f3 --- /dev/null +++ b/tests/queries/0_stateless/02116_tuple_element_analyzer.reference @@ -0,0 +1,25 @@ +1 +SELECT __table1.`t1.a` AS `tupleElement(t1, 1)` +FROM default.t_tuple_element AS __table1 +a +SELECT __table1.`t1.s` AS `tupleElement(t1, 2)` +FROM default.t_tuple_element AS __table1 +1 +SELECT __table1.`t1.a` AS `tupleElement(t1, \'a\')` +FROM default.t_tuple_element AS __table1 +2 +SELECT __table1.`t2.1` AS `tupleElement(t2, 1)` +FROM default.t_tuple_element AS __table1 +2 +SELECT __table1.`t2.1` AS `tupleElement(t2, 1)` +FROM default.t_tuple_element AS __table1 +1 2 +SELECT + _CAST(1, \'UInt8\') AS `tupleElement(t, 1)`, + _CAST(2, \'UInt8\') AS `tupleElement(t, 2)` +FROM system.one AS __table1 +1 2 +SELECT + _CAST(1, \'UInt32\') AS `tupleElement(t, 1)`, + _CAST(2, \'UInt32\') AS `tupleElement(t, \'b\')` +FROM system.one AS __table1 diff --git a/tests/queries/0_stateless/02116_tuple_element.sql b/tests/queries/0_stateless/02116_tuple_element_analyzer.sql similarity index 60% rename from tests/queries/0_stateless/02116_tuple_element.sql rename to tests/queries/0_stateless/02116_tuple_element_analyzer.sql index 64d9b9db331..5aeb72c9ee4 100644 --- a/tests/queries/0_stateless/02116_tuple_element.sql +++ b/tests/queries/0_stateless/02116_tuple_element_analyzer.sql @@ -4,39 +4,40 @@ CREATE TABLE t_tuple_element(t1 Tuple(a UInt32, s String), t2 Tuple(UInt32, Stri INSERT INTO t_tuple_element VALUES ((1, 'a'), (2, 'b')); SET optimize_functions_to_subcolumns = 1; +SET allow_experimental_analyzer = 1; SELECT t1.1 FROM t_tuple_element; -EXPLAIN SYNTAX SELECT t1.1 FROM t_tuple_element; +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT t1.1 FROM t_tuple_element; SELECT tupleElement(t1, 2) FROM t_tuple_element; -EXPLAIN SYNTAX SELECT tupleElement(t1, 2) FROM t_tuple_element; +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT tupleElement(t1, 2) FROM t_tuple_element; SELECT tupleElement(t1, 'a') FROM t_tuple_element; -EXPLAIN SYNTAX SELECT tupleElement(t1, 'a') FROM t_tuple_element; +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT tupleElement(t1, 'a') FROM t_tuple_element; SELECT tupleElement(number, 1) FROM numbers(1); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } SELECT tupleElement(t1) FROM t_tuple_element; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } SELECT tupleElement(t1, 'b') FROM t_tuple_element; -- { serverError NOT_FOUND_COLUMN_IN_BLOCK, UNKNOWN_IDENTIFIER } -SELECT tupleElement(t1, 0) FROM t_tuple_element; -- { serverError ILLEGAL_INDEX, NOT_FOUND_COLUMN_IN_BLOCK } -SELECT tupleElement(t1, 3) FROM t_tuple_element; -- { serverError ILLEGAL_INDEX, NOT_FOUND_COLUMN_IN_BLOCK } -SELECT tupleElement(t1, materialize('a')) FROM t_tuple_element; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT, NOT_FOUND_COLUMN_IN_BLOCK } +SELECT tupleElement(t1, 0) FROM t_tuple_element; -- { serverError ARGUMENT_OUT_OF_BOUND, NOT_FOUND_COLUMN_IN_BLOCK } +SELECT tupleElement(t1, 3) FROM t_tuple_element; -- { serverError ARGUMENT_OUT_OF_BOUND, NOT_FOUND_COLUMN_IN_BLOCK } +SELECT tupleElement(t1, materialize('a')) FROM t_tuple_element; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } SELECT t2.1 FROM t_tuple_element; -EXPLAIN SYNTAX SELECT t2.1 FROM t_tuple_element; +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT t2.1 FROM t_tuple_element; SELECT tupleElement(t2, 1) FROM t_tuple_element; -EXPLAIN SYNTAX SELECT tupleElement(t2, 1) FROM t_tuple_element; +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT tupleElement(t2, 1) FROM t_tuple_element; SELECT tupleElement(t2) FROM t_tuple_element; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } SELECT tupleElement(t2, 'a') FROM t_tuple_element; -- { serverError NOT_FOUND_COLUMN_IN_BLOCK, UNKNOWN_IDENTIFIER } -SELECT tupleElement(t2, 0) FROM t_tuple_element; -- { serverError ILLEGAL_INDEX, NOT_FOUND_COLUMN_IN_BLOCK } -SELECT tupleElement(t2, 3) FROM t_tuple_element; -- { serverError ILLEGAL_INDEX, NOT_FOUND_COLUMN_IN_BLOCK } -SELECT tupleElement(t2, materialize(1)) FROM t_tuple_element; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT, NOT_FOUND_COLUMN_IN_BLOCK } +SELECT tupleElement(t2, 0) FROM t_tuple_element; -- { serverError ARGUMENT_OUT_OF_BOUND, NOT_FOUND_COLUMN_IN_BLOCK } +SELECT tupleElement(t2, 3) FROM t_tuple_element; -- { serverError ARGUMENT_OUT_OF_BOUND, NOT_FOUND_COLUMN_IN_BLOCK } +SELECT tupleElement(t2, materialize(1)) FROM t_tuple_element; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } DROP TABLE t_tuple_element; WITH (1, 2) AS t SELECT t.1, t.2; -EXPLAIN SYNTAX WITH (1, 2) AS t SELECT t.1, t.2; +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 WITH (1, 2) AS t SELECT t.1, t.2; WITH (1, 2)::Tuple(a UInt32, b UInt32) AS t SELECT t.1, tupleElement(t, 'b'); -EXPLAIN SYNTAX WITH (1, 2)::Tuple(a UInt32, b UInt32) AS t SELECT t.1, tupleElement(t, 'b'); +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 WITH (1, 2)::Tuple(a UInt32, b UInt32) AS t SELECT t.1, tupleElement(t, 'b'); diff --git a/tests/queries/0_stateless/02117_custom_separated_with_names_and_types.sh b/tests/queries/0_stateless/02117_custom_separated_with_names_and_types.sh index 400bf2a56fa..d925c962da4 100755 --- a/tests/queries/0_stateless/02117_custom_separated_with_names_and_types.sh +++ b/tests/queries/0_stateless/02117_custom_separated_with_names_and_types.sh @@ -1,5 +1,4 @@ #!/usr/bin/env bash -# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02117_show_create_table_system.reference b/tests/queries/0_stateless/02117_show_create_table_system.reference index 8f62eda9233..cfae4fee6c2 100644 --- a/tests/queries/0_stateless/02117_show_create_table_system.reference +++ b/tests/queries/0_stateless/02117_show_create_table_system.reference @@ -286,7 +286,6 @@ CREATE TABLE system.functions ( `name` String, `is_aggregate` UInt8, - `is_deterministic` Nullable(UInt8), `case_insensitive` UInt8, `alias_to` String, `create_query` String, diff --git a/tests/queries/0_stateless/02118_deserialize_whole_text.sh b/tests/queries/0_stateless/02118_deserialize_whole_text.sh index ccbfc5abe97..d544f1452a9 100755 --- a/tests/queries/0_stateless/02118_deserialize_whole_text.sh +++ b/tests/queries/0_stateless/02118_deserialize_whole_text.sh @@ -1,65 +1,65 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel, no-replicated-database +# Tags: no-fasttest, no-replicated-database CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -DATA_FILE=$USER_FILES_PATH/data_02118 +DATA_FILE=$USER_FILES_PATH/${CLICKHOUSE_DATABASE} +FILE=${CLICKHOUSE_DATABASE} echo "[\"[1,2,3]trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x Array(UInt32)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x Array(UInt32)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"1970-01-02trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x Date')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x Date')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"1970-01-02trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x Date32')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x Date32')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"1970-01-01 03:00:01trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x DateTime')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x DateTime')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"1970-01-01 03:00:01.0000trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x DateTime64(4)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x DateTime64(4)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"42trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x UInt32')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x UInt32')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"42.4242trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x Decimal32(4)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x Decimal32(4)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"255.255.255.255trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x IPv4')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x IPv4')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "255.255.255.255trash" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'TSV', 'x IPv4')" 2>&1 | grep -F -q "CANNOT_PARSE_INPUT_ASSERTION_FAILED" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'TSV', 'x IPv4')" 2>&1 | grep -F -q "CANNOT_PARSE_INPUT_ASSERTION_FAILED" && echo 'OK' || echo 'FAIL' echo "255.255.255.255trash" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'CSV', 'x IPv4')" 2>&1 | grep -F -q "INCORRECT_DATA" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'CSV', 'x IPv4')" 2>&1 | grep -F -q "INCORRECT_DATA" && echo 'OK' || echo 'FAIL' echo "[\"255.255.255.255trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactEachRow', 'x IPv4')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactEachRow', 'x IPv4')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"0000:0000:0000:0000:0000:ffff:192.168.100.228b1trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x IPv6')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x IPv6')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "0000:0000:0000:0000:0000:ffff:192.168.100.228b1trash" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'TSV', 'x IPv6')" 2>&1 | grep -F -q "CANNOT_PARSE_INPUT_ASSERTION_FAILED" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'TSV', 'x IPv6')" 2>&1 | grep -F -q "CANNOT_PARSE_INPUT_ASSERTION_FAILED" && echo 'OK' || echo 'FAIL' echo "0000:0000:0000:0000:0000:ffff:192.168.100.228b1trash" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'CSV', 'x IPv6')" 2>&1 | grep -F -q "INCORRECT_DATA" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'CSV', 'x IPv6')" 2>&1 | grep -F -q "INCORRECT_DATA" && echo 'OK' || echo 'FAIL' echo "[\"0000:0000:0000:0000:0000:ffff:192.168.100.228b1trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactEachRow', 'x IPv6')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactEachRow', 'x IPv6')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"{1:2, 2:3}trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x Map(UInt32, UInt32)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x Map(UInt32, UInt32)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"(1, 2)trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x Tuple(UInt32, UInt32)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x Tuple(UInt32, UInt32)')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' echo "[\"ed9fd45d-6287-47c1-ad9f-d45d628767c1trash\"]" > $DATA_FILE -$CLICKHOUSE_CLIENT -q "SELECT * FROM file('data_02118', 'JSONCompactStringsEachRow', 'x UUID')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_CLIENT -q "SELECT * FROM file('${FILE}', 'JSONCompactStringsEachRow', 'x UUID')" 2>&1 | grep -F -q "UNEXPECTED_DATA_AFTER_PARSED_VALUE" && echo 'OK' || echo 'FAIL' rm $DATA_FILE diff --git a/tests/queries/0_stateless/02124_insert_deduplication_token_materialized_views.reference b/tests/queries/0_stateless/02124_insert_deduplication_token_materialized_views.reference index e0cc8f0ce63..2d9f236ada9 100644 --- a/tests/queries/0_stateless/02124_insert_deduplication_token_materialized_views.reference +++ b/tests/queries/0_stateless/02124_insert_deduplication_token_materialized_views.reference @@ -1,8 +1,8 @@ -deduplicate_blocks_in_dependent_materialized_views=0, insert_deduplication_token = no, results inconsitent -18 18 9 18 -deduplicate_blocks_in_dependent_materialized_views=1, insert_deduplication_token = no, results inconsitent -18 9 9 9 -deduplicate_blocks_in_dependent_materialized_views=0, insert_deduplication_token = yes, results inconsitent -18 18 9 18 -deduplicate_blocks_in_dependent_materialized_views=1, insert_deduplication_token = yes, results consitent +deduplicate_blocks_in_dependent_materialized_views=0, insert_deduplication_token = no, results: test_mv_a and test_mv_c have all data, test_mv_b has data obly with max_partitions_per_insert_block=0 +18 36 27 36 +deduplicate_blocks_in_dependent_materialized_views=1, insert_deduplication_token = no, results: all tables have deduplicated data +18 18 18 18 +deduplicate_blocks_in_dependent_materialized_views=0, insert_deduplication_token = yes, results: test_mv_a and test_mv_c have all data, test_mv_b has data obly with max_partitions_per_insert_block=0 +18 36 27 36 +deduplicate_blocks_in_dependent_materialized_views=1, insert_deduplication_token = yes, results: all tables have deduplicated data 18 18 18 18 diff --git a/tests/queries/0_stateless/02124_insert_deduplication_token_materialized_views.sql b/tests/queries/0_stateless/02124_insert_deduplication_token_materialized_views.sql index fdd75b91b1f..465c8d6136c 100644 --- a/tests/queries/0_stateless/02124_insert_deduplication_token_materialized_views.sql +++ b/tests/queries/0_stateless/02124_insert_deduplication_token_materialized_views.sql @@ -1,6 +1,6 @@ -- Tags: long -select 'deduplicate_blocks_in_dependent_materialized_views=0, insert_deduplication_token = no, results inconsitent'; +select 'deduplicate_blocks_in_dependent_materialized_views=0, insert_deduplication_token = no, results: test_mv_a and test_mv_c have all data, test_mv_b has data obly with max_partitions_per_insert_block=0'; drop table if exists test sync; drop table if exists test_mv_a sync; @@ -35,7 +35,7 @@ select (select sum(c) from test_mv_c where test='case1'); -select 'deduplicate_blocks_in_dependent_materialized_views=1, insert_deduplication_token = no, results inconsitent'; +select 'deduplicate_blocks_in_dependent_materialized_views=1, insert_deduplication_token = no, results: all tables have deduplicated data'; set deduplicate_blocks_in_dependent_materialized_views=1; @@ -53,7 +53,7 @@ select (select sum(c) from test_mv_c where test='case2'); -select 'deduplicate_blocks_in_dependent_materialized_views=0, insert_deduplication_token = yes, results inconsitent'; +select 'deduplicate_blocks_in_dependent_materialized_views=0, insert_deduplication_token = yes, results: test_mv_a and test_mv_c have all data, test_mv_b has data obly with max_partitions_per_insert_block=0'; set deduplicate_blocks_in_dependent_materialized_views=0; @@ -70,7 +70,7 @@ select (select sum(c) from test_mv_b where test='case3'), (select sum(c) from test_mv_c where test='case3'); -select 'deduplicate_blocks_in_dependent_materialized_views=1, insert_deduplication_token = yes, results consitent'; +select 'deduplicate_blocks_in_dependent_materialized_views=1, insert_deduplication_token = yes, results: all tables have deduplicated data'; set deduplicate_blocks_in_dependent_materialized_views=1; diff --git a/tests/queries/0_stateless/02125_many_mutations.reference b/tests/queries/0_stateless/02125_many_mutations.reference index c98d8221c7f..a94b8b78480 100644 --- a/tests/queries/0_stateless/02125_many_mutations.reference +++ b/tests/queries/0_stateless/02125_many_mutations.reference @@ -2,5 +2,5 @@ 1 1 20000 0 -0 20000 -1 20001 +0 20001 +1 20002 diff --git a/tests/queries/0_stateless/02125_many_mutations.sh b/tests/queries/0_stateless/02125_many_mutations.sh index 474fc8ef199..239f30811d4 100755 --- a/tests/queries/0_stateless/02125_many_mutations.sh +++ b/tests/queries/0_stateless/02125_many_mutations.sh @@ -48,6 +48,7 @@ $CLICKHOUSE_CLIENT --multiquery -q " select count() from system.mutations where database = currentDatabase() and table = 'many_mutations' and not is_done; system start merges many_mutations; optimize table many_mutations final SETTINGS optimize_throw_if_noop = 1; +alter table many_mutations update y = y + 1 where 1 settings mutations_sync=2; select count() from system.mutations where database = currentDatabase() and table = 'many_mutations' and not is_done; select x, y from many_mutations order by x; truncate table many_mutations; diff --git a/tests/queries/0_stateless/02125_many_mutations_2.sh b/tests/queries/0_stateless/02125_many_mutations_2.sh index b7585991738..958f50287b2 100755 --- a/tests/queries/0_stateless/02125_many_mutations_2.sh +++ b/tests/queries/0_stateless/02125_many_mutations_2.sh @@ -52,6 +52,7 @@ $CLICKHOUSE_CLIENT --multiquery -q " select count() from system.mutations where database = currentDatabase() and table = 'many_mutations' and not is_done; system start merges many_mutations; optimize table many_mutations final SETTINGS optimize_throw_if_noop = 1; +alter table many_mutations update y = y + 1 where 1 settings mutations_sync=2; system flush logs; select count() from system.mutations where database = currentDatabase() and table = 'many_mutations' and not is_done; select count() from many_mutations; diff --git a/tests/queries/0_stateless/02125_query_views_log.sql b/tests/queries/0_stateless/02125_query_views_log.sql index d2d19b76a1f..ba50902ebea 100644 --- a/tests/queries/0_stateless/02125_query_views_log.sql +++ b/tests/queries/0_stateless/02125_query_views_log.sql @@ -8,7 +8,7 @@ create table dst (key Int) engine=Null(); create materialized view mv1 to dst as select * from src; create materialized view mv2 to dst as select * from src; -insert into src select * from numbers(1e6) settings log_queries=1, max_untracked_memory=0, parallel_view_processing=1; +insert into src select * from numbers(1e6) settings log_queries=1, max_untracked_memory=0, parallel_view_processing=0; system flush logs; -- { echo } diff --git a/tests/queries/0_stateless/02125_tskv_proper_names_reading.sh b/tests/queries/0_stateless/02125_tskv_proper_names_reading.sh index 0abf411d38f..4a169897520 100755 --- a/tests/queries/0_stateless/02125_tskv_proper_names_reading.sh +++ b/tests/queries/0_stateless/02125_tskv_proper_names_reading.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - DATA_FILE=$USER_FILES_PATH/test_02125.data echo "number=1" > $DATA_FILE diff --git a/tests/queries/0_stateless/02126_fix_filelog.sh b/tests/queries/0_stateless/02126_fix_filelog.sh index b266b582428..0e136a34c62 100755 --- a/tests/queries/0_stateless/02126_fix_filelog.sh +++ b/tests/queries/0_stateless/02126_fix_filelog.sh @@ -6,14 +6,8 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ - -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* ${CLICKHOUSE_CLIENT} --query "drop table if exists file_log;" @@ -21,8 +15,8 @@ ${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=Fil ${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('/tmp/aaa.csv', 'CSV');" 2>&1 | grep -q "Code: 36" && echo 'OK' || echo 'FAIL'; ${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('/tmp/aaa.csv', 'CSV');" 2>&1 | grep -q "Code: 36" && echo 'OK' || echo 'FAIL'; -${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" +${CLICKHOUSE_CLIENT} --query "create table file_log(k UInt8, v UInt8) engine=FileLog('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'CSV');" ${CLICKHOUSE_CLIENT} --query "drop table file_log;" -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} diff --git a/tests/queries/0_stateless/02129_skip_quoted_fields.sh b/tests/queries/0_stateless/02129_skip_quoted_fields.sh index ac702d3c750..701d7a30b68 100755 --- a/tests/queries/0_stateless/02129_skip_quoted_fields.sh +++ b/tests/queries/0_stateless/02129_skip_quoted_fields.sh @@ -1,5 +1,4 @@ #!/usr/bin/env bash -# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02130_parse_quoted_null.sh b/tests/queries/0_stateless/02130_parse_quoted_null.sh index 0c72d0e85a7..44e6ee93599 100755 --- a/tests/queries/0_stateless/02130_parse_quoted_null.sh +++ b/tests/queries/0_stateless/02130_parse_quoted_null.sh @@ -1,14 +1,12 @@ #!/usr/bin/env bash -# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -DATA_FILE=$USER_FILES_PATH/test_02130.data -SELECT_QUERY="select * from file('test_02130.data', 'CustomSeparated', 'x Nullable(Float64), y Nullable(UInt64)') settings input_format_parallel_parsing=0, format_custom_escaping_rule='Quoted'" +DATA_FILE=$USER_FILES_PATH/${CLICKHOUSE_DATABASE}.data +SELECT_QUERY="select * from file('${CLICKHOUSE_DATABASE}.data', 'CustomSeparated', 'x Nullable(Float64), y Nullable(UInt64)') settings input_format_parallel_parsing=0, format_custom_escaping_rule='Quoted'" $CLICKHOUSE_CLIENT -q "drop table if exists test_02130" diff --git a/tests/queries/0_stateless/02139_MV_with_scalar_subquery.sql b/tests/queries/0_stateless/02139_MV_with_scalar_subquery.sql index f0285bbec3d..63c894cfb85 100644 --- a/tests/queries/0_stateless/02139_MV_with_scalar_subquery.sql +++ b/tests/queries/0_stateless/02139_MV_with_scalar_subquery.sql @@ -16,7 +16,7 @@ SELECT FROM source_null GROUP BY count_subquery, min_subquery, max_subquery; - +SET optimize_trivial_insert_select = 1; INSERT INTO source SELECT number FROM numbers(2000) SETTINGS min_insert_block_size_rows=1500, max_insert_block_size=1500; SELECT count() FROM source; diff --git a/tests/queries/0_stateless/02149_external_schema_inference.sh b/tests/queries/0_stateless/02149_external_schema_inference.sh index 41f8bfee2bc..edb4e915701 100755 --- a/tests/queries/0_stateless/02149_external_schema_inference.sh +++ b/tests/queries/0_stateless/02149_external_schema_inference.sh @@ -1,20 +1,19 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -FILE_NAME=test_$CLICKHOUSE_TEST_UNIQUE_NAME.data +FILE_NAME=test_${CLICKHOUSE_TEST_UNIQUE_NAME}_${CLICKHOUSE_DATABASE}.data DATA_FILE=$USER_FILES_PATH/$FILE_NAME touch $DATA_FILE -SCHEMADIR=$($CLICKHOUSE_CLIENT_BINARY --query "select * from file('$FILE_NAME', 'CapnProto', 'val1 char') settings format_schema='nonexist:Message'" 2>&1 | grep Exception | grep -oP "file \K.*(?=/nonexist.capnp)") +SCHEMADIR=${CLICKHOUSE_SCHEMA_FILES} CLIENT_SCHEMADIR=$CURDIR/format_schemas -SERVER_SCHEMADIR=test_02149 +SERVER_SCHEMADIR=${CLICKHOUSE_DATABASE} mkdir -p $SCHEMADIR/$SERVER_SCHEMADIR cp -r $CLIENT_SCHEMADIR/* $SCHEMADIR/$SERVER_SCHEMADIR/ diff --git a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql index 0834b76d4ec..ae8c39b49bc 100644 --- a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql +++ b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql @@ -1,5 +1,6 @@ SET max_threads=0; SET optimize_read_in_order=1; +SET optimize_trivial_insert_select = 1; SET read_in_order_two_level_merge_threshold=100; DROP TABLE IF EXISTS t_read_in_order; diff --git a/tests/queries/0_stateless/02149_schema_inference.sh b/tests/queries/0_stateless/02149_schema_inference.sh index 856549f2215..fba1e6e9137 100755 --- a/tests/queries/0_stateless/02149_schema_inference.sh +++ b/tests/queries/0_stateless/02149_schema_inference.sh @@ -1,18 +1,16 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') FILE_NAME=test_$CLICKHOUSE_TEST_UNIQUE_NAME.data -DATA_FILE=${USER_FILES_PATH:?}/$FILE_NAME - +DATA_FILE=${CLICKHOUSE_USER_FILES:?}/$FILE_NAME touch $DATA_FILE -SCHEMADIR=$($CLICKHOUSE_CLIENT_BINARY --query "select * from file('$FILE_NAME', 'Template', 'val1 char') settings format_template_row='nonexist'" 2>&1 | grep Exception | grep -oP "file \K.*(?=/nonexist)") +SCHEMADIR=${CLICKHOUSE_SCHEMA_FILES} echo "TSV" diff --git a/tests/queries/0_stateless/02149_schema_inference_create_table_syntax.sh b/tests/queries/0_stateless/02149_schema_inference_create_table_syntax.sh index 8de2ab8c57a..bf247817323 100755 --- a/tests/queries/0_stateless/02149_schema_inference_create_table_syntax.sh +++ b/tests/queries/0_stateless/02149_schema_inference_create_table_syntax.sh @@ -1,14 +1,13 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -mkdir $USER_FILES_PATH/test_02149 -FILE_NAME=test_02149/data.Parquet +mkdir $USER_FILES_PATH/${CLICKHOUSE_DATABASE}/ +FILE_NAME=data.Parquet DATA_FILE=$USER_FILES_PATH/$FILE_NAME $CLICKHOUSE_CLIENT -q "select number as num, concat('Str: ', toString(number)) as str, [number, number + 1] as arr from numbers(10) format Parquet" > $DATA_FILE diff --git a/tests/queries/0_stateless/02151_hash_table_sizes_stats_joins.reference b/tests/queries/0_stateless/02151_hash_table_sizes_stats_joins.reference new file mode 100644 index 00000000000..d3d171221e8 --- /dev/null +++ b/tests/queries/0_stateless/02151_hash_table_sizes_stats_joins.reference @@ -0,0 +1,10 @@ +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 diff --git a/tests/queries/0_stateless/02151_hash_table_sizes_stats_joins.sh b/tests/queries/0_stateless/02151_hash_table_sizes_stats_joins.sh new file mode 100755 index 00000000000..6d715604d93 --- /dev/null +++ b/tests/queries/0_stateless/02151_hash_table_sizes_stats_joins.sh @@ -0,0 +1,74 @@ +#!/usr/bin/env bash +# Tags: long, distributed, no-debug, no-tsan, no-msan, no-ubsan, no-asan, no-random-settings, no-random-merge-tree-settings + +# shellcheck disable=SC2154 + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +opts=( + --join_algorithm='parallel_hash' +) + +$CLICKHOUSE_CLIENT -nq " + CREATE TABLE t1(a UInt32, b UInt32) ENGINE=MergeTree ORDER BY (); + INSERT INTO t1 SELECT number, number FROM numbers_mt(1e6); + + CREATE TABLE t2(a UInt32, b UInt32) ENGINE=MergeTree ORDER BY (); + INSERT INTO t2 SELECT number, number FROM numbers_mt(1e6); +" + +queries_without_preallocation=() +queries_with_preallocation=() + +run_new_query() { + query_id1="hash_table_sizes_stats_joins_$RANDOM$RANDOM" + # when we see a query for the first time we only collect it stats when execution ends. preallocation will happen only on the next run + queries_without_preallocation+=("$query_id1") + $CLICKHOUSE_CLIENT "${opts[@]}" --query_id="$query_id1" -q "$1" --format Null + + query_id2="hash_table_sizes_stats_joins_$RANDOM$RANDOM" + queries_with_preallocation+=("$query_id2") + $CLICKHOUSE_CLIENT "${opts[@]}" --query_id="$query_id2" -q "$1" --format Null +} + +run_new_query "SELECT * FROM t1 AS x INNER JOIN t2 AS y ON x.a = y.a" +# it only matters what columns from the right table are part of the join key, as soon as we change them - it is a new cache entry +run_new_query "SELECT * FROM t1 AS x INNER JOIN t2 AS y ON x.a = y.b" +run_new_query "SELECT * FROM t1 AS x INNER JOIN t2 AS y USING (a, b)" + +# we already had a join on t2.a, so cache should be populated +query_id="hash_table_sizes_stats_joins_$RANDOM$RANDOM" +queries_with_preallocation+=("$query_id") +$CLICKHOUSE_CLIENT "${opts[@]}" --query_id="$query_id" -q "SELECT * FROM t1 AS x INNER JOIN t2 AS y ON x.b = y.a" --format Null +# the same query with a different alias for the t2 +query_id="hash_table_sizes_stats_joins_$RANDOM$RANDOM" +queries_with_preallocation+=("$query_id") +$CLICKHOUSE_CLIENT "${opts[@]}" --query_id="$query_id" -q "SELECT * FROM t1 AS x INNER JOIN t2 AS z ON x.b = z.a" --format Null + +# now t1 is the right table +run_new_query "SELECT * FROM t2 AS x INNER JOIN t1 AS y ON x.a = y.a" + +################################## + +$CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS" + +for i in "${!queries_without_preallocation[@]}"; do + $CLICKHOUSE_CLIENT --param_query_id="${queries_without_preallocation[$i]}" -q " + -- the old analyzer is not supported + SELECT sum(if(getSetting('allow_experimental_analyzer'), ProfileEvents['HashJoinPreallocatedElementsInHashTables'] = 0, 1)) + FROM system.query_log + WHERE event_date >= yesterday() AND query_id = {query_id:String} AND current_database = currentDatabase() AND type = 'QueryFinish' + " +done + +for i in "${!queries_with_preallocation[@]}"; do + $CLICKHOUSE_CLIENT --param_query_id="${queries_with_preallocation[$i]}" -q " + -- the old analyzer is not supported + SELECT sum(if(getSetting('allow_experimental_analyzer'), ProfileEvents['HashJoinPreallocatedElementsInHashTables'] > 0, 1)) + FROM system.query_log + WHERE event_date >= yesterday() AND query_id = {query_id:String} AND current_database = currentDatabase() AND type = 'QueryFinish' + " +done diff --git a/tests/queries/0_stateless/02155_dictionary_comment.sql b/tests/queries/0_stateless/02155_dictionary_comment.sql index 30b85e16a7c..8ebc7b259fc 100644 --- a/tests/queries/0_stateless/02155_dictionary_comment.sql +++ b/tests/queries/0_stateless/02155_dictionary_comment.sql @@ -49,5 +49,5 @@ SELECT name, comment FROM system.tables WHERE name == '02155_test_dictionary_vie SELECT name, comment FROM system.tables WHERE name == '02155_test_dictionary_view' AND database == currentDatabase(); DROP TABLE 02155_test_dictionary_view; -DROP TABLE 02155_test_table; DROP DICTIONARY 02155_test_dictionary; +DROP TABLE 02155_test_table; diff --git a/tests/queries/0_stateless/02167_format_from_file_extension.sh b/tests/queries/0_stateless/02167_format_from_file_extension.sh index 14985233524..0a0efff3228 100755 --- a/tests/queries/0_stateless/02167_format_from_file_extension.sh +++ b/tests/queries/0_stateless/02167_format_from_file_extension.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -7,28 +7,26 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) function cleanup() { - # this command expects an error message like 'Code: 107. DB::Exception: Received <...> nonexist.txt doesn't exist. (FILE_DOESNT_EXIST)' - user_files_path=$($CLICKHOUSE_CLIENT --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep -E '^Code: 107.*FILE_DOESNT_EXIST' | head -1 | awk '{gsub("/nonexist.txt","",$9); print $9}') - rm $user_files_path/test_02167.* + rm ${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}_test_02167.* } trap cleanup EXIT for format in TSV TabSeparated TSVWithNames TSVWithNamesAndTypes CSV Parquet ORC Arrow JSONEachRow JSONCompactEachRow CustomSeparatedWithNamesAndTypes do - $CLICKHOUSE_CLIENT -q "insert into table function file('test_02167.$format', 'auto', 'x UInt64') select * from numbers(2)" - $CLICKHOUSE_CLIENT -q "select * from file('test_02167.$format')" - $CLICKHOUSE_CLIENT -q "select * from file('test_02167.$format', '$format')" + $CLICKHOUSE_CLIENT -q "insert into table function file('${CLICKHOUSE_DATABASE}_test_02167.$format', 'auto', 'x UInt64') select * from numbers(2)" + $CLICKHOUSE_CLIENT -q "select * from file('${CLICKHOUSE_DATABASE}_test_02167.$format')" + $CLICKHOUSE_CLIENT -q "select * from file('${CLICKHOUSE_DATABASE}_test_02167.$format', '$format')" done -$CLICKHOUSE_CLIENT -q "insert into table function file('test_02167.bin', 'auto', 'x UInt64') select * from numbers(2)" -$CLICKHOUSE_CLIENT -q "select * from file('test_02167.bin', 'auto', 'x UInt64')" -$CLICKHOUSE_CLIENT -q "select * from file('test_02167.bin', 'RowBinary', 'x UInt64')" +$CLICKHOUSE_CLIENT -q "insert into table function file('${CLICKHOUSE_DATABASE}_test_02167.bin', 'auto', 'x UInt64') select * from numbers(2)" +$CLICKHOUSE_CLIENT -q "select * from file('${CLICKHOUSE_DATABASE}_test_02167.bin', 'auto', 'x UInt64')" +$CLICKHOUSE_CLIENT -q "select * from file('${CLICKHOUSE_DATABASE}_test_02167.bin', 'RowBinary', 'x UInt64')" -$CLICKHOUSE_CLIENT -q "insert into table function file('test_02167.ndjson', 'auto', 'x UInt64') select * from numbers(2)" -$CLICKHOUSE_CLIENT -q "select * from file('test_02167.ndjson')" -$CLICKHOUSE_CLIENT -q "select * from file('test_02167.ndjson', 'JSONEachRow', 'x UInt64')" +$CLICKHOUSE_CLIENT -q "insert into table function file('${CLICKHOUSE_DATABASE}_test_02167.ndjson', 'auto', 'x UInt64') select * from numbers(2)" +$CLICKHOUSE_CLIENT -q "select * from file('${CLICKHOUSE_DATABASE}_test_02167.ndjson')" +$CLICKHOUSE_CLIENT -q "select * from file('${CLICKHOUSE_DATABASE}_test_02167.ndjson', 'JSONEachRow', 'x UInt64')" -$CLICKHOUSE_CLIENT -q "insert into table function file('test_02167.messagepack', 'auto', 'x UInt64') select * from numbers(2)" -$CLICKHOUSE_CLIENT -q "select * from file('test_02167.messagepack') settings input_format_msgpack_number_of_columns=1" -$CLICKHOUSE_CLIENT -q "select * from file('test_02167.messagepack', 'MsgPack', 'x UInt64')" +$CLICKHOUSE_CLIENT -q "insert into table function file('${CLICKHOUSE_DATABASE}_test_02167.messagepack', 'auto', 'x UInt64') select * from numbers(2)" +$CLICKHOUSE_CLIENT -q "select * from file('${CLICKHOUSE_DATABASE}_test_02167.messagepack') settings input_format_msgpack_number_of_columns=1" +$CLICKHOUSE_CLIENT -q "select * from file('${CLICKHOUSE_DATABASE}_test_02167.messagepack', 'MsgPack', 'x UInt64')" diff --git a/tests/queries/0_stateless/02183_dictionary_date_types.sql b/tests/queries/0_stateless/02183_dictionary_date_types.sql index e06863d5e53..5671f47cdab 100644 --- a/tests/queries/0_stateless/02183_dictionary_date_types.sql +++ b/tests/queries/0_stateless/02183_dictionary_date_types.sql @@ -170,8 +170,8 @@ LIFETIME(0); SELECT 'Polygon dictionary'; SELECT * FROM 02183_polygon_dictionary; -DROP TABLE 02183_polygon_dictionary_source_table; DROP DICTIONARY 02183_polygon_dictionary; +DROP TABLE 02183_polygon_dictionary_source_table; DROP TABLE IF EXISTS 02183_range_dictionary_source_table; CREATE TABLE 02183_range_dictionary_source_table diff --git a/tests/queries/0_stateless/02185_orc_corrupted_file.sh b/tests/queries/0_stateless/02185_orc_corrupted_file.sh index 12510ae3836..8cf4334845d 100755 --- a/tests/queries/0_stateless/02185_orc_corrupted_file.sh +++ b/tests/queries/0_stateless/02185_orc_corrupted_file.sh @@ -5,7 +5,6 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') cp $CUR_DIR/data_orc/corrupted.orc $USER_FILES_PATH/ ${CLICKHOUSE_CLIENT} --query="select * from file('corrupted.orc')" 2>&1 | grep -F -q 'CANNOT_EXTRACT_TABLE_STRUCTURE' && echo 'OK' || echo 'FAIL' diff --git a/tests/queries/0_stateless/02185_range_hashed_dictionary_open_ranges.sql b/tests/queries/0_stateless/02185_range_hashed_dictionary_open_ranges.sql index e6edee2ea18..a36c72de0ac 100644 --- a/tests/queries/0_stateless/02185_range_hashed_dictionary_open_ranges.sql +++ b/tests/queries/0_stateless/02185_range_hashed_dictionary_open_ranges.sql @@ -60,4 +60,5 @@ SELECT dictHas('02185_range_dictionary', 0, 0); SELECT dictHas('02185_range_dictionary', 0, 5001); SELECT dictHas('02185_range_dictionary', 0, 10001); +DROP DICTIONARY 02185_range_dictionary; DROP TABLE 02185_range_dictionary_source_table; diff --git a/tests/queries/0_stateless/02207_allow_plaintext_and_no_password.sh b/tests/queries/0_stateless/02207_allow_plaintext_and_no_password.sh index 0345a0e6394..dc3cb0de110 100755 --- a/tests/queries/0_stateless/02207_allow_plaintext_and_no_password.sh +++ b/tests/queries/0_stateless/02207_allow_plaintext_and_no_password.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -cp /etc/clickhouse-server/users.xml "$CURDIR"/users.xml +cp ${CLICKHOUSE_CONFIG_DIR}/users.xml "$CURDIR"/users.xml sed -i 's/<\/password>/c64c5e4e53ea1a9f1427d2713b3a22bbebe8940bc807adaf654744b1568c70ab<\/password_sha256_hex>/g' "$CURDIR"/users.xml sed -i 's//1<\/access_management>/g' "$CURDIR"/users.xml diff --git a/tests/queries/0_stateless/02210_processors_profile_log.sql b/tests/queries/0_stateless/02210_processors_profile_log.sql index 59edbb71457..75e5bcbb585 100644 --- a/tests/queries/0_stateless/02210_processors_profile_log.sql +++ b/tests/queries/0_stateless/02210_processors_profile_log.sql @@ -14,10 +14,12 @@ SELECT multiIf( -- ExpressionTransform executes sleep(), -- so IProcessor::work() will spend 1 sec. - name = 'ExpressionTransform', elapsed_us>=1e6 ? 1 : elapsed_us, + -- We use two different timers to measure time: CLOCK_MONOTONIC for sleep and CLOCK_MONOTONIC_COARSE for profiling + -- that's why we cannot compare directly with 1,000,000 microseconds - let's compare with 900,000 microseconds. + name = 'ExpressionTransform', elapsed_us >= 0.9e6 ? 1 : elapsed_us, -- SourceFromSingleChunk, that feed data to ExpressionTransform, -- will feed first block and then wait in PortFull. - name = 'SourceFromSingleChunk', output_wait_elapsed_us>=1e6 ? 1 : output_wait_elapsed_us, + name = 'SourceFromSingleChunk', output_wait_elapsed_us >= 0.9e6 ? 1 : output_wait_elapsed_us, -- NullSource/LazyOutputFormatLazyOutputFormat are the outputs -- so they cannot starts to execute before sleep(1) will be executed. input_wait_elapsed_us>=1e6 ? 1 : input_wait_elapsed_us) diff --git a/tests/queries/0_stateless/02222_create_table_without_columns_metadata.sh b/tests/queries/0_stateless/02222_create_table_without_columns_metadata.sh index d49c3610852..73d1c2b9b42 100755 --- a/tests/queries/0_stateless/02222_create_table_without_columns_metadata.sh +++ b/tests/queries/0_stateless/02222_create_table_without_columns_metadata.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT -q "insert into table function file(data.jsonl, 'JSONEachRow', 'x UInt32 default 42, y String') select number as x, 'String' as y from numbers(10)" $CLICKHOUSE_CLIENT -q "drop table if exists test" diff --git a/tests/queries/0_stateless/02226_filesystem_cache_profile_events.sh b/tests/queries/0_stateless/02226_filesystem_cache_profile_events.sh index 9d87542d84d..d0e61541b15 100755 --- a/tests/queries/0_stateless/02226_filesystem_cache_profile_events.sh +++ b/tests/queries/0_stateless/02226_filesystem_cache_profile_events.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel, no-s3-storage, no-random-settings, no-replicated-database +# Tags: no-fasttest, no-parallel, no-object-storage, no-random-settings, no-replicated-database # set -x diff --git a/tests/queries/0_stateless/02227_test_create_empty_sqlite_db.sh b/tests/queries/0_stateless/02227_test_create_empty_sqlite_db.sh index 37fdde95ea7..344452767cc 100755 --- a/tests/queries/0_stateless/02227_test_create_empty_sqlite_db.sh +++ b/tests/queries/0_stateless/02227_test_create_empty_sqlite_db.sh @@ -6,9 +6,6 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -# See 01658_read_file_to_string_column.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - function cleanup() { ${CLICKHOUSE_CLIENT} --query="DROP DATABASE IF EXISTS ${CURR_DATABASE}" @@ -18,7 +15,7 @@ trap cleanup EXIT export CURR_DATABASE="test_01889_sqllite_${CLICKHOUSE_DATABASE}" -DB_PATH=${user_files_path}/${CURR_DATABASE}_db1 +DB_PATH=${USER_FILES_PATH}/${CURR_DATABASE}_db1 ${CLICKHOUSE_CLIENT} --multiquery --multiline --query=""" DROP DATABASE IF EXISTS ${CURR_DATABASE}; diff --git a/tests/queries/0_stateless/02228_merge_tree_insert_memory_usage.sql b/tests/queries/0_stateless/02228_merge_tree_insert_memory_usage.sql index ca1ee2738c7..72b6cc06f26 100644 --- a/tests/queries/0_stateless/02228_merge_tree_insert_memory_usage.sql +++ b/tests/queries/0_stateless/02228_merge_tree_insert_memory_usage.sql @@ -1,5 +1,5 @@ --- Tags: long, no-parallel, no-s3-storage --- no-s3-storage: Avoid flakiness due to cache / buffer usage +-- Tags: long, no-object-storage +-- no-object-storage: Avoid flakiness due to cache / buffer usage SET insert_keeper_fault_injection_probability=0; -- to succeed this test can require too many retries due to 100 partitions, so disable fault injections -- regression for MEMORY_LIMIT_EXCEEDED error because of deferred final part flush diff --git a/tests/queries/0_stateless/02231_buffer_aggregate_states_leak.sql b/tests/queries/0_stateless/02231_buffer_aggregate_states_leak.sql index dbe18953957..3d6b69fe161 100644 --- a/tests/queries/0_stateless/02231_buffer_aggregate_states_leak.sql +++ b/tests/queries/0_stateless/02231_buffer_aggregate_states_leak.sql @@ -28,7 +28,8 @@ create materialized view mv_02231 to buffer_02231 as select from in_02231 group by key; -insert into in_02231 select * from numbers(10e6) settings max_memory_usage='310Mi', max_threads=1; +set optimize_trivial_insert_select = 1; +insert into in_02231 select * from numbers(10e6) settings max_memory_usage='400Mi', max_threads=1; drop table buffer_02231; drop table out_02231; diff --git a/tests/queries/0_stateless/02233_optimize_aggregation_in_order_prefix.sql b/tests/queries/0_stateless/02233_optimize_aggregation_in_order_prefix.sql index 8bc75040e5a..48af5ae0031 100644 --- a/tests/queries/0_stateless/02233_optimize_aggregation_in_order_prefix.sql +++ b/tests/queries/0_stateless/02233_optimize_aggregation_in_order_prefix.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage SET merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0; diff --git a/tests/queries/0_stateless/02235_add_part_offset_virtual_column.sql b/tests/queries/0_stateless/02235_add_part_offset_virtual_column.sql index 5af6565c03d..b23d2d25111 100644 --- a/tests/queries/0_stateless/02235_add_part_offset_virtual_column.sql +++ b/tests/queries/0_stateless/02235_add_part_offset_virtual_column.sql @@ -21,6 +21,7 @@ CREATE TABLE t_random_1 ) ENGINE = GenerateRandom(1, 5, 3); +SET optimize_trivial_insert_select = 1; INSERT INTO t_1 select rowNumberInAllBlocks(), *, '1984-01-01' from t_random_1 limit 1000000; OPTIMIZE TABLE t_1 FINAL; diff --git a/tests/queries/0_stateless/02235_remote_fs_cache_stress.sh b/tests/queries/0_stateless/02235_remote_fs_cache_stress.sh index 0b6b9f461b0..ffc38c0c1bd 100755 --- a/tests/queries/0_stateless/02235_remote_fs_cache_stress.sh +++ b/tests/queries/0_stateless/02235_remote_fs_cache_stress.sh @@ -28,7 +28,7 @@ ORDER BY tuple(); INSERT INTO t_01411_num (num) SELECT number % 1000 FROM numbers(100000); -create table lc_dict_reading (val UInt64, str StringWithDictionary, pat String) engine = MergeTree order by val; +create table lc_dict_reading (val UInt64, str LowCardinality(String), pat String) engine = MergeTree order by val; insert into lc_dict_reading select number, if(number < 8192 * 4, number % 100, number) as s, s from system.numbers limit 100000; """ diff --git a/tests/queries/0_stateless/02240_asof_join_biginteger.reference b/tests/queries/0_stateless/02240_asof_join_biginteger.reference index cac55eec430..f7eb4d74375 100644 --- a/tests/queries/0_stateless/02240_asof_join_biginteger.reference +++ b/tests/queries/0_stateless/02240_asof_join_biginteger.reference @@ -2,3 +2,7 @@ 0 340282366920938463463374607431768211457 0 18446744073709551617 0 340282366920938463463374607431768211457 +0 18446744073709551617 +0 340282366920938463463374607431768211457 +0 18446744073709551617 +0 340282366920938463463374607431768211457 diff --git a/tests/queries/0_stateless/02240_asof_join_biginteger.sql b/tests/queries/0_stateless/02240_asof_join_biginteger.sql index 6dc5b00f116..a5c1faae4ea 100644 --- a/tests/queries/0_stateless/02240_asof_join_biginteger.sql +++ b/tests/queries/0_stateless/02240_asof_join_biginteger.sql @@ -3,3 +3,11 @@ select * from (select 0 as k, toInt256('340282366920938463463374607431768211457' select * from (select 0 as k, toUInt128('18446744073709551617') as v) t1 asof join (select 0 as k, toUInt128('18446744073709551616') as v) t2 using(k, v); select * from (select 0 as k, toUInt256('340282366920938463463374607431768211457') as v) t1 asof join (select 0 as k, toUInt256('340282366920938463463374607431768211456') as v) t2 using(k, v); + +SET join_algorithm = 'full_sorting_merge'; + +select * from (select 0 as k, toInt128('18446744073709551617') as v) t1 asof join (select 0 as k, toInt128('18446744073709551616') as v) t2 using(k, v); +select * from (select 0 as k, toInt256('340282366920938463463374607431768211457') as v) t1 asof join (select 0 as k, toInt256('340282366920938463463374607431768211456') as v) t2 using(k, v); + +select * from (select 0 as k, toUInt128('18446744073709551617') as v) t1 asof join (select 0 as k, toUInt128('18446744073709551616') as v) t2 using(k, v); +select * from (select 0 as k, toUInt256('340282366920938463463374607431768211457') as v) t1 asof join (select 0 as k, toUInt256('340282366920938463463374607431768211456') as v) t2 using(k, v); diff --git a/tests/queries/0_stateless/02240_filesystem_cache_bypass_cache_threshold.sql b/tests/queries/0_stateless/02240_filesystem_cache_bypass_cache_threshold.sql index ee92931ec54..b791ee18e82 100644 --- a/tests/queries/0_stateless/02240_filesystem_cache_bypass_cache_threshold.sql +++ b/tests/queries/0_stateless/02240_filesystem_cache_bypass_cache_threshold.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-fasttest, no-s3-storage, no-random-settings +-- Tags: no-parallel, no-fasttest, no-object-storage, no-random-settings -- { echo } diff --git a/tests/queries/0_stateless/02240_filesystem_query_cache.sql b/tests/queries/0_stateless/02240_filesystem_query_cache.sql index a609702f22a..40c80e04697 100644 --- a/tests/queries/0_stateless/02240_filesystem_query_cache.sql +++ b/tests/queries/0_stateless/02240_filesystem_query_cache.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-fasttest, no-s3-storage, no-random-settings +-- Tags: no-parallel, no-fasttest, no-object-storage, no-random-settings -- { echo } diff --git a/tests/queries/0_stateless/02240_system_filesystem_cache_table.reference b/tests/queries/0_stateless/02240_system_filesystem_cache_table.reference index 93b6d4de94f..6b5dd182112 100644 --- a/tests/queries/0_stateless/02240_system_filesystem_cache_table.reference +++ b/tests/queries/0_stateless/02240_system_filesystem_cache_table.reference @@ -34,3 +34,21 @@ DOWNLOADED 0 79 80 DOWNLOADED 0 745 746 2 Expect no cache +Using storage policy: azure_cache +0 +Expect cache +DOWNLOADED 0 0 1 +DOWNLOADED 0 79 80 +DOWNLOADED 0 745 746 +3 +Expect cache +DOWNLOADED 0 0 1 +DOWNLOADED 0 79 80 +DOWNLOADED 0 745 746 +3 +Expect no cache +Expect cache +DOWNLOADED 0 79 80 +DOWNLOADED 0 745 746 +2 +Expect no cache diff --git a/tests/queries/0_stateless/02240_system_filesystem_cache_table.sh b/tests/queries/0_stateless/02240_system_filesystem_cache_table.sh index 9aa631c5d0a..8faf0a08f1f 100755 --- a/tests/queries/0_stateless/02240_system_filesystem_cache_table.sh +++ b/tests/queries/0_stateless/02240_system_filesystem_cache_table.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-fasttest, no-parallel, no-s3-storage, no-random-settings +# Tags: long, no-fasttest, no-parallel, no-object-storage, no-random-settings # set -x @@ -7,7 +7,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -for STORAGE_POLICY in 's3_cache' 'local_cache'; do +for STORAGE_POLICY in 's3_cache' 'local_cache' 'azure_cache'; do echo "Using storage policy: $STORAGE_POLICY" ${CLICKHOUSE_CLIENT} --query "SYSTEM DROP FILESYSTEM CACHE" ${CLICKHOUSE_CLIENT} --query "SYSTEM DROP MARK CACHE" diff --git a/tests/queries/0_stateless/02240_tskv_schema_inference_bug.sh b/tests/queries/0_stateless/02240_tskv_schema_inference_bug.sh index ce545a27317..d4a4e54acbd 100755 --- a/tests/queries/0_stateless/02240_tskv_schema_inference_bug.sh +++ b/tests/queries/0_stateless/02240_tskv_schema_inference_bug.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh - -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') FILE_NAME=test_02240.data DATA_FILE=${USER_FILES_PATH:?}/$FILE_NAME diff --git a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference index 186dcc1eeb2..f53f00992e7 100644 --- a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference +++ b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.reference @@ -218,3 +218,113 @@ SELECT count() FROM test_02241 5010500 SELECT count() FROM test_02241 WHERE value LIKE '%010%' 18816 +Using storage policy: azure_cache +DROP TABLE IF EXISTS test_02241 +CREATE TABLE test_02241 (key UInt32, value String) Engine=MergeTree() ORDER BY key SETTINGS storage_policy='azure_cache', min_bytes_for_wide_part = 10485760, compress_marks=false, compress_primary_key=false, ratio_of_defaults_for_sparse_serialization = 1 +SYSTEM STOP MERGES test_02241 +SYSTEM DROP FILESYSTEM CACHE +SELECT file_segment_range_begin, file_segment_range_end, size, state + FROM + ( + SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path + FROM + ( + SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path + FROM system.remote_data_paths + ) AS data_paths + INNER JOIN + system.filesystem_cache AS caches + ON data_paths.cache_path = caches.cache_path + ) + WHERE endsWith(local_path, 'data.bin') + FORMAT Vertical +SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path +0 +SELECT count(), sum(size) FROM system.filesystem_cache +0 0 +INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100) +SELECT file_segment_range_begin, file_segment_range_end, size, state + FROM + ( + SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path + FROM + ( + SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path + FROM system.remote_data_paths + ) AS data_paths + INNER JOIN + system.filesystem_cache AS caches + ON data_paths.cache_path = caches.cache_path + ) + WHERE endsWith(local_path, 'data.bin') + FORMAT Vertical +Row 1: +────── +file_segment_range_begin: 0 +file_segment_range_end: 745 +size: 746 +state: DOWNLOADED +SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path +8 +SELECT count(), sum(size) FROM system.filesystem_cache +8 1100 +SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0 +0 +SELECT * FROM test_02241 FORMAT Null +SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0 +2 +SELECT * FROM test_02241 FORMAT Null +SELECT count() FROM system.filesystem_cache WHERE cache_hits > 0 +2 +SELECT count(), sum(size) size FROM system.filesystem_cache +8 1100 +SYSTEM DROP FILESYSTEM CACHE +INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100, 200) +SELECT file_segment_range_begin, file_segment_range_end, size, state + FROM + ( + SELECT file_segment_range_begin, file_segment_range_end, size, state, local_path + FROM + ( + SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path + FROM system.remote_data_paths + ) AS data_paths + INNER JOIN + system.filesystem_cache AS caches + ON data_paths.cache_path = caches.cache_path + ) + WHERE endsWith(local_path, 'data.bin') + FORMAT Vertical; +Row 1: +────── +file_segment_range_begin: 0 +file_segment_range_end: 1659 +size: 1660 +state: DOWNLOADED +SELECT count() FROM (SELECT arrayJoin(cache_paths) AS cache_path, local_path, remote_path FROM system.remote_data_paths ) AS data_paths INNER JOIN system.filesystem_cache AS caches ON data_paths.cache_path = caches.cache_path +8 +SELECT count(), sum(size) FROM system.filesystem_cache +8 2014 +SELECT count(), sum(size) FROM system.filesystem_cache +8 2014 +INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100) SETTINGS enable_filesystem_cache_on_write_operations=0 +SELECT count(), sum(size) FROM system.filesystem_cache +8 2014 +INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(100) +INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(300, 10000) +SELECT count(), sum(size) FROM system.filesystem_cache +24 84045 +SYSTEM START MERGES test_02241 +OPTIMIZE TABLE test_02241 FINAL +SELECT count(), sum(size) FROM system.filesystem_cache +32 167243 +ALTER TABLE test_02241 UPDATE value = 'kek' WHERE key = 100 +SELECT count(), sum(size) FROM system.filesystem_cache +41 250541 +INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(5000000) +SYSTEM FLUSH LOGS +INSERT INTO test_02241 SELECT number, toString(number) FROM numbers(5000000) 0 +SELECT count() FROM test_02241 +5010500 +SELECT count() FROM test_02241 WHERE value LIKE '%010%' +18816 diff --git a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh index c1d930f54a7..f8e7b7e7e72 100755 --- a/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh +++ b/tests/queries/0_stateless/02241_filesystem_cache_on_write_operations.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-fasttest, no-parallel, no-s3-storage, no-random-settings +# Tags: long, no-fasttest, no-parallel, no-object-storage, no-random-settings # set -x @@ -7,7 +7,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -for STORAGE_POLICY in 's3_cache' 'local_cache'; do +for STORAGE_POLICY in 's3_cache' 'local_cache' 'azure_cache'; do echo "Using storage policy: $STORAGE_POLICY" $CLICKHOUSE_CLIENT --echo --query "DROP TABLE IF EXISTS test_02241" diff --git a/tests/queries/0_stateless/02241_join_rocksdb_bs.reference b/tests/queries/0_stateless/02241_join_rocksdb_bs.reference index 8416a2991c1..4dff9ef38ef 100644 --- a/tests/queries/0_stateless/02241_join_rocksdb_bs.reference +++ b/tests/queries/0_stateless/02241_join_rocksdb_bs.reference @@ -10,59 +10,3 @@ 1 1 1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 diff --git a/tests/queries/0_stateless/02241_join_rocksdb_bs.sql.j2 b/tests/queries/0_stateless/02241_join_rocksdb_bs.sql.j2 index 6121db6d6a2..e5703f99d62 100644 --- a/tests/queries/0_stateless/02241_join_rocksdb_bs.sql.j2 +++ b/tests/queries/0_stateless/02241_join_rocksdb_bs.sql.j2 @@ -1,4 +1,4 @@ --- Tags: use-rocksdb, long +-- Tags: use-rocksdb, long, no-s3-storage SET join_algorithm = 'direct'; @@ -13,27 +13,21 @@ INSERT INTO rdb_{{ table_size }} SELECT (sipHash64(number) % {{ table_size }}) as key, ('val' || toString(key)) AS value FROM numbers_mt({{ table_size }}); -{% for block_size in [10, 11, 128, 129, 65505, 65506, 70000] -%} - -{% if block_size * 5000 > table_size -%} - -SET max_block_size = {{ block_size }}; - {% for right_size in [table_size // 2, table_size + table_size // 4 + 1] -%} SELECT count() == (SELECT count() FROM rdb_{{ table_size }} WHERE key < {{ right_size }}) FROM (SELECT number as k FROM numbers_mt({{ right_size }})) as t1 INNER JOIN rdb_{{ table_size }} as rdb -ON rdb.key == t1.k; +ON rdb.key == t1.k +{% if table_size < 100 %}SETTINGS max_block_size = 1{% endif -%} +; SELECT count() == {{ right_size }} and countIf(value != '') == (SELECT count() FROM rdb_{{ table_size }} WHERE key < {{ right_size }}) FROM (SELECT number as k FROM numbers_mt({{ right_size }})) as t1 LEFT JOIN rdb_{{ table_size }} as rdb -ON rdb.key == t1.k; - -{% endfor -%} - -{% endif -%} +ON rdb.key == t1.k +{% if table_size < 100 %}SETTINGS max_block_size = 1{% endif -%} +; {% endfor -%} {% endfor -%} @@ -41,4 +35,3 @@ ON rdb.key == t1.k; {% for table_size in [10, 65555, 100000] -%} DROP TABLE IF EXISTS rdb_{{ table_size }}; {% endfor -%} - diff --git a/tests/queries/0_stateless/02242_arrow_orc_parquet_nullable_schema_inference.sh b/tests/queries/0_stateless/02242_arrow_orc_parquet_nullable_schema_inference.sh index e03c62cfc5f..0f37bff45d6 100755 --- a/tests/queries/0_stateless/02242_arrow_orc_parquet_nullable_schema_inference.sh +++ b/tests/queries/0_stateless/02242_arrow_orc_parquet_nullable_schema_inference.sh @@ -6,7 +6,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') FILE_NAME=test_02242.data DATA_FILE=$USER_FILES_PATH/$FILE_NAME diff --git a/tests/queries/0_stateless/02242_system_filesystem_cache_log_table.reference b/tests/queries/0_stateless/02242_system_filesystem_cache_log_table.reference index 99f31df7def..447e1a275fc 100644 --- a/tests/queries/0_stateless/02242_system_filesystem_cache_log_table.reference +++ b/tests/queries/0_stateless/02242_system_filesystem_cache_log_table.reference @@ -6,3 +6,7 @@ Using storage policy: local_cache (0,519) READ_FROM_FS_AND_DOWNLOADED_TO_CACHE (0,808110) READ_FROM_FS_AND_DOWNLOADED_TO_CACHE (0,808110) READ_FROM_CACHE +Using storage policy: azure_cache +(0,519) READ_FROM_FS_AND_DOWNLOADED_TO_CACHE +(0,808110) READ_FROM_FS_AND_DOWNLOADED_TO_CACHE +(0,808110) READ_FROM_CACHE diff --git a/tests/queries/0_stateless/02242_system_filesystem_cache_log_table.sh b/tests/queries/0_stateless/02242_system_filesystem_cache_log_table.sh index 4c92d1d2954..fe016f5a27f 100755 --- a/tests/queries/0_stateless/02242_system_filesystem_cache_log_table.sh +++ b/tests/queries/0_stateless/02242_system_filesystem_cache_log_table.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-fasttest, no-parallel, no-s3-storage, no-random-settings +# Tags: long, no-fasttest, no-parallel, no-object-storage, no-random-settings # set -x @@ -7,7 +7,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -for STORAGE_POLICY in 's3_cache' 'local_cache'; do +for STORAGE_POLICY in 's3_cache' 'local_cache' 'azure_cache'; do echo "Using storage policy: $STORAGE_POLICY" $CLICKHOUSE_CLIENT --query "SYSTEM DROP FILESYSTEM CACHE" diff --git a/tests/queries/0_stateless/02245_parquet_skip_unknown_type.sh b/tests/queries/0_stateless/02245_parquet_skip_unknown_type.sh index 8ff6e28b123..7bfeb747cc2 100755 --- a/tests/queries/0_stateless/02245_parquet_skip_unknown_type.sh +++ b/tests/queries/0_stateless/02245_parquet_skip_unknown_type.sh @@ -5,7 +5,6 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') FILE_NAME=test_02245.parquet DATA_FILE=$USER_FILES_PATH/$FILE_NAME diff --git a/tests/queries/0_stateless/02246_tsv_csv_best_effort_schema_inference.sh b/tests/queries/0_stateless/02246_tsv_csv_best_effort_schema_inference.sh index 233db7a534d..07c2a33c4d5 100755 --- a/tests/queries/0_stateless/02246_tsv_csv_best_effort_schema_inference.sh +++ b/tests/queries/0_stateless/02246_tsv_csv_best_effort_schema_inference.sh @@ -1,12 +1,11 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') FILE_NAME=test_02149.data DATA_FILE=${USER_FILES_PATH:?}/$FILE_NAME diff --git a/tests/queries/0_stateless/02247_names_order_in_json_and_tskv.sh b/tests/queries/0_stateless/02247_names_order_in_json_and_tskv.sh index e8e3bf88ac4..3385f62af38 100755 --- a/tests/queries/0_stateless/02247_names_order_in_json_and_tskv.sh +++ b/tests/queries/0_stateless/02247_names_order_in_json_and_tskv.sh @@ -6,7 +6,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') FILE_NAME=test_02247.data DATA_FILE=${USER_FILES_PATH:?}/$FILE_NAME diff --git a/tests/queries/0_stateless/02247_read_bools_as_numbers_json.sh b/tests/queries/0_stateless/02247_read_bools_as_numbers_json.sh index 523b5934543..76133df2b37 100755 --- a/tests/queries/0_stateless/02247_read_bools_as_numbers_json.sh +++ b/tests/queries/0_stateless/02247_read_bools_as_numbers_json.sh @@ -1,13 +1,12 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -FILE_NAME=test_02247.data +FILE_NAME=${CLICKHOUSE_DATABASE}.data DATA_FILE=${USER_FILES_PATH:?}/$FILE_NAME touch $DATA_FILE diff --git a/tests/queries/0_stateless/02263_lazy_mark_load.sh b/tests/queries/0_stateless/02263_lazy_mark_load.sh index 5f80d9d7f6d..f1602e47e01 100755 --- a/tests/queries/0_stateless/02263_lazy_mark_load.sh +++ b/tests/queries/0_stateless/02263_lazy_mark_load.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-s3-storage, no-random-settings, no-parallel +# Tags: no-object-storage, no-random-settings, no-parallel set -eo pipefail CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) diff --git a/tests/queries/0_stateless/02270_errors_in_files.sh b/tests/queries/0_stateless/02270_errors_in_files.sh index 517547c6ef8..ab8bb28787d 100755 --- a/tests/queries/0_stateless/02270_errors_in_files.sh +++ b/tests/queries/0_stateless/02270_errors_in_files.sh @@ -13,8 +13,6 @@ echo "Error" > "${CLICKHOUSE_TMP}"/test_02270_2.csv ${CLICKHOUSE_LOCAL} --query "SELECT * FROM file('${CLICKHOUSE_TMP}/test_02270*.csv', CSV, 'a String, b String')" 2>&1 | grep -o "test_02270_2.csv" ${CLICKHOUSE_LOCAL} --query "SELECT * FROM file('${CLICKHOUSE_TMP}/test_02270*.csv', CSV, 'a String, b String')" --input_format_parallel_parsing 0 2>&1 | grep -o "test_02270_2.csv" -user_files_path=$($CLICKHOUSE_CLIENT --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep -E '^Code: 107.*FILE_DOESNT_EXIST' | head -1 | awk '{gsub("/nonexist.txt","",$9); print $9}') - ${CLICKHOUSE_CLIENT} --query "INSERT INTO TABLE FUNCTION file('test_02270_1.csv') SELECT 'Hello', 'World'" ${CLICKHOUSE_CLIENT} --query "INSERT INTO TABLE FUNCTION file('test_02270_2.csv') SELECT 'Error'" @@ -27,9 +25,9 @@ ${CLICKHOUSE_CLIENT} --query "INSERT INTO TABLE FUNCTION file('test_02270_2.csv. ${CLICKHOUSE_CLIENT} --query "SELECT * FROM file('test_02270*.csv.gz', 'CSV', 'a String, b String')" 2>&1 | grep -o -m1 "test_02270_2.csv.gz" ${CLICKHOUSE_CLIENT} --query "SELECT * FROM file('test_02270*.csv.gz', 'CSV', 'a String, b String')" --input_format_parallel_parsing 0 2>&1 | grep -o -m1 "test_02270_2.csv.gz" -rm "${CLICKHOUSE_TMP}"/test_02270_1.csv -rm "${CLICKHOUSE_TMP}"/test_02270_2.csv -rm "${user_files_path}"/test_02270_1.csv -rm "${user_files_path}"/test_02270_2.csv -rm "${user_files_path}"/test_02270_1.csv.gz -rm "${user_files_path}"/test_02270_2.csv.gz +rm -f "${CLICKHOUSE_TMP}"/test_02270_1.csv +rm -f "${CLICKHOUSE_TMP}"/test_02270_2.csv +rm -f "${USER_FILES_PATH}"/test_02270_1.csv +rm -f "${USER_FILES_PATH}"/test_02270_2.csv +rm -f "${USER_FILES_PATH}"/test_02270_1.csv.gz +rm -f "${USER_FILES_PATH}"/test_02270_2.csv.gz diff --git a/tests/queries/0_stateless/02276_full_sort_join_unsupported.sql b/tests/queries/0_stateless/02276_full_sort_join_unsupported.sql index a4e60ff54dd..0b10101d8f2 100644 --- a/tests/queries/0_stateless/02276_full_sort_join_unsupported.sql +++ b/tests/queries/0_stateless/02276_full_sort_join_unsupported.sql @@ -19,8 +19,6 @@ SELECT * FROM t1 ANTI JOIN t2 ON t1.key = t2.key; -- { serverError NOT_IMPLEMENT SELECT * FROM t1 SEMI JOIN t2 ON t1.key = t2.key; -- { serverError NOT_IMPLEMENTED } -SELECT * FROM t1 ASOF JOIN t2 ON t1.key = t2.key AND t1.val > t2.val; -- { serverError NOT_IMPLEMENTED } - SELECT * FROM t1 ANY JOIN t2 ON t1.key = t2.key SETTINGS any_join_distinct_right_table_keys = 1; -- { serverError NOT_IMPLEMENTED } SELECT * FROM t1 JOIN t2 USING (key) SETTINGS join_use_nulls = 1; -- { serverError NOT_IMPLEMENTED } diff --git a/tests/queries/0_stateless/02286_drop_filesystem_cache.reference b/tests/queries/0_stateless/02286_drop_filesystem_cache.reference index b4e5b6715de..e3875dbabe1 100644 --- a/tests/queries/0_stateless/02286_drop_filesystem_cache.reference +++ b/tests/queries/0_stateless/02286_drop_filesystem_cache.reference @@ -16,3 +16,12 @@ Using storage policy: local_cache 1 1 0 +Using storage policy: azure_cache +0 +2 +0 +1 +1 +1 +1 +0 diff --git a/tests/queries/0_stateless/02286_drop_filesystem_cache.sh b/tests/queries/0_stateless/02286_drop_filesystem_cache.sh index 1e1841862e9..32c9e9cb060 100755 --- a/tests/queries/0_stateless/02286_drop_filesystem_cache.sh +++ b/tests/queries/0_stateless/02286_drop_filesystem_cache.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-fasttest, no-parallel, no-s3-storage, no-random-settings +# Tags: long, no-fasttest, no-parallel, no-object-storage, no-random-settings # set -x @@ -7,7 +7,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -for STORAGE_POLICY in 's3_cache' 'local_cache'; do +for STORAGE_POLICY in 's3_cache' 'local_cache' 'azure_cache'; do echo "Using storage policy: $STORAGE_POLICY" $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS test_02286" diff --git a/tests/queries/0_stateless/02286_mysql_dump_input_format.sh b/tests/queries/0_stateless/02286_mysql_dump_input_format.sh index 2f6167c3ddf..59528d97b93 100755 --- a/tests/queries/0_stateless/02286_mysql_dump_input_format.sh +++ b/tests/queries/0_stateless/02286_mysql_dump_input_format.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - cp $CURDIR/data_mysql_dump/dump*.sql $USER_FILES_PATH $CLICKHOUSE_CLIENT -q "select * from file(dump1.sql, MySQLDump, 'x Nullable(Int32), y Nullable(Int32)') order by x, y" diff --git a/tests/queries/0_stateless/02286_tuple_numeric_identifier.sql b/tests/queries/0_stateless/02286_tuple_numeric_identifier.sql index f723284ad61..151ff275f7b 100644 --- a/tests/queries/0_stateless/02286_tuple_numeric_identifier.sql +++ b/tests/queries/0_stateless/02286_tuple_numeric_identifier.sql @@ -12,9 +12,9 @@ SELECT * FROM t_tuple_numeric FORMAT JSONEachRow; SELECT `t`.`1`.`2`, `t`.`1`.`3`, `t`.`4` FROM t_tuple_numeric; SELECT t.1.1, t.1.2, t.2 FROM t_tuple_numeric; -SELECT t.1.3 FROM t_tuple_numeric; -- {serverError NOT_FOUND_COLUMN_IN_BLOCK} -SELECT t.4 FROM t_tuple_numeric; -- {serverError NOT_FOUND_COLUMN_IN_BLOCK} -SELECT `t`.`1`.`1`, `t`.`1`.`2`, `t`.`2` FROM t_tuple_numeric; -- {serverError UNKNOWN_IDENTIFIER} +SELECT t.1.3 FROM t_tuple_numeric; -- {serverError NOT_FOUND_COLUMN_IN_BLOCK, ARGUMENT_OUT_OF_BOUND} +SELECT t.4 FROM t_tuple_numeric; -- {serverError NOT_FOUND_COLUMN_IN_BLOCK, ARGUMENT_OUT_OF_BOUND} +SELECT `t`.`1`.`1`, `t`.`1`.`2`, `t`.`2` FROM t_tuple_numeric; -- {serverError UNKNOWN_IDENTIFIER, ARGUMENT_OUT_OF_BOUND} DROP TABLE t_tuple_numeric; diff --git a/tests/queries/0_stateless/02293_formats_json_columns.sh b/tests/queries/0_stateless/02293_formats_json_columns.sh index 4eae5a1abb4..bf605d6f591 100755 --- a/tests/queries/0_stateless/02293_formats_json_columns.sh +++ b/tests/queries/0_stateless/02293_formats_json_columns.sh @@ -5,8 +5,6 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -USER_FILES_PATH=$(clickhouse client --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - DATA_FILE=$USER_FILES_PATH/data_02293 $CLICKHOUSE_CLIENT -q "drop table if exists test_02293" diff --git a/tests/queries/0_stateless/02294_nothing_arguments_in_functions.sql b/tests/queries/0_stateless/02294_nothing_arguments_in_functions.sql index 4406a05df0c..ecf4f9cab93 100644 --- a/tests/queries/0_stateless/02294_nothing_arguments_in_functions.sql +++ b/tests/queries/0_stateless/02294_nothing_arguments_in_functions.sql @@ -1,3 +1,5 @@ +set enable_named_columns_in_function_tuple = 0; + select arrayMap(x -> 2 * x, []); select toTypeName(arrayMap(x -> 2 * x, [])); select arrayMap((x, y) -> x + y, [], []); diff --git a/tests/queries/0_stateless/02297_regex_parsing_file_names.sh b/tests/queries/0_stateless/02297_regex_parsing_file_names.sh index 5973e24844a..666cb4d87fc 100755 --- a/tests/queries/0_stateless/02297_regex_parsing_file_names.sh +++ b/tests/queries/0_stateless/02297_regex_parsing_file_names.sh @@ -5,29 +5,21 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. +rm -rf ${USER_FILES_PATH}/file_{0..10}.csv -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -CLICKHOUSE_USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path, _file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') +echo '0' > ${USER_FILES_PATH}/file_0.csv +echo '0' > ${USER_FILES_PATH}/file_1.csv +echo '0' > ${USER_FILES_PATH}/file_2.csv +echo '0' > ${USER_FILES_PATH}/file_3.csv +echo '0' > ${USER_FILES_PATH}/file_4.csv +echo '0' > ${USER_FILES_PATH}/file_5.csv +echo '0' > ${USER_FILES_PATH}/file_6.csv +echo '0' > ${USER_FILES_PATH}/file_7.csv +echo '0' > ${USER_FILES_PATH}/file_8.csv +echo '0' > ${USER_FILES_PATH}/file_9.csv +echo '0' > ${USER_FILES_PATH}/file_10.csv -mkdir -p ${CLICKHOUSE_USER_FILES_PATH}/ - -rm -rf ${CLICKHOUSE_USER_FILES_PATH}/file_{0..10}.csv - -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_0.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_1.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_2.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_3.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_4.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_5.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_6.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_7.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_8.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_9.csv -echo '0' > ${CLICKHOUSE_USER_FILES_PATH}/file_10.csv - -# echo '' > ${CLICKHOUSE_USER_FILES_PATH}/file_10.csv +# echo '' > ${USER_FILES_PATH}/file_10.csv ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS t_regex;" @@ -36,5 +28,5 @@ ${CLICKHOUSE_CLIENT} -q "CREATE TABLE t_regex (id UInt64) ENGINE = MergeTree() o ${CLICKHOUSE_CLIENT} -q "INSERT INTO t_regex SELECT * FROM file('file_{0..10}.csv','CSV');" ${CLICKHOUSE_CLIENT} -q "SELECT count() from t_regex;" -rm -rf ${CLICKHOUSE_USER_FILES_PATH}/file_{0..10}.csv; +rm -rf ${USER_FILES_PATH}/file_{0..10}.csv; ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS t_regex;" diff --git a/tests/queries/0_stateless/02310_clickhouse_local_INSERT_progress_profile_events.expect b/tests/queries/0_stateless/02310_clickhouse_local_INSERT_progress_profile_events.expect index bde00b306cf..d5b2a278180 100755 --- a/tests/queries/0_stateless/02310_clickhouse_local_INSERT_progress_profile_events.expect +++ b/tests/queries/0_stateless/02310_clickhouse_local_INSERT_progress_profile_events.expect @@ -1,4 +1,6 @@ #!/usr/bin/expect -f +# Tags: no-debug, no-tsan, no-msan, no-asan, no-ubsan, no-s3-storage +# ^ it can be slower than 60 seconds # This is the regression for the concurrent access in ProgressIndication, # so it is important to read enough rows here (10e6). diff --git a/tests/queries/0_stateless/02313_filesystem_cache_seeks.reference b/tests/queries/0_stateless/02313_filesystem_cache_seeks.reference index 062aac259a4..0a9e1c20b59 100644 --- a/tests/queries/0_stateless/02313_filesystem_cache_seeks.reference +++ b/tests/queries/0_stateless/02313_filesystem_cache_seeks.reference @@ -1,3 +1,4 @@ Using storage policy: s3_cache Using storage policy: local_cache Using storage policy: s3_cache_multi +Using storage policy: azure_cache diff --git a/tests/queries/0_stateless/02313_filesystem_cache_seeks.sh b/tests/queries/0_stateless/02313_filesystem_cache_seeks.sh index f5de4346fd6..b54e3d7f805 100755 --- a/tests/queries/0_stateless/02313_filesystem_cache_seeks.sh +++ b/tests/queries/0_stateless/02313_filesystem_cache_seeks.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-fasttest, no-parallel, no-s3-storage, no-random-settings +# Tags: long, no-fasttest, no-parallel, no-object-storage, no-random-settings # set -x @@ -8,7 +8,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh -for STORAGE_POLICY in 's3_cache' 'local_cache' 's3_cache_multi'; do +for STORAGE_POLICY in 's3_cache' 'local_cache' 's3_cache_multi' 'azure_cache'; do echo "Using storage policy: $STORAGE_POLICY" $CLICKHOUSE_CLIENT --query "SYSTEM DROP FILESYSTEM CACHE" diff --git a/tests/queries/0_stateless/02327_capnproto_protobuf_empty_messages.sh b/tests/queries/0_stateless/02327_capnproto_protobuf_empty_messages.sh index 650faf6985e..89e5c827a48 100755 --- a/tests/queries/0_stateless/02327_capnproto_protobuf_empty_messages.sh +++ b/tests/queries/0_stateless/02327_capnproto_protobuf_empty_messages.sh @@ -5,10 +5,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') touch $USER_FILES_PATH/data.capnp -SCHEMADIR=$($CLICKHOUSE_CLIENT_BINARY --query "select * from file('data.capnp', 'CapnProto', 'val1 char') settings format_schema='nonexist:Message'" 2>&1 | grep Exception | grep -oP "file \K.*(?=/nonexist.capnp)") +SCHEMADIR=${CLICKHOUSE_SCHEMA_FILES} CLIENT_SCHEMADIR=$CURDIR/format_schemas SERVER_SCHEMADIR=test_02327 mkdir -p $SCHEMADIR/$SERVER_SCHEMADIR diff --git a/tests/queries/0_stateless/02336_sparse_columns_s3.sql b/tests/queries/0_stateless/02336_sparse_columns_s3.sql index bf4622adedc..1dc1e980846 100644 --- a/tests/queries/0_stateless/02336_sparse_columns_s3.sql +++ b/tests/queries/0_stateless/02336_sparse_columns_s3.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-fasttest, no-s3-storage +-- Tags: no-parallel, no-fasttest, no-object-storage DROP TABLE IF EXISTS t_sparse_s3; diff --git a/tests/queries/0_stateless/02343_aggregation_pipeline.sql b/tests/queries/0_stateless/02343_aggregation_pipeline.sql index d73ac66763e..0f9dbd0247d 100644 --- a/tests/queries/0_stateless/02343_aggregation_pipeline.sql +++ b/tests/queries/0_stateless/02343_aggregation_pipeline.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage -- produces different pipeline if enabled set enable_memory_bound_merging_of_aggregation_results = 0; diff --git a/tests/queries/0_stateless/02346_non_negative_derivative.sql b/tests/queries/0_stateless/02346_non_negative_derivative.sql index 704241da16c..ab648f2ee73 100644 --- a/tests/queries/0_stateless/02346_non_negative_derivative.sql +++ b/tests/queries/0_stateless/02346_non_negative_derivative.sql @@ -58,8 +58,8 @@ SELECT ts, metric, nonNegativeDerivative(metric, ts, INTERVAL 13 YEAR) OVER (PAR -- test against wrong arguments/types SELECT ts, metric, nonNegativeDerivative(metric, 1, INTERVAL 3 NANOSECOND) OVER (PARTITION BY metric ORDER BY ts, metric ASC Rows BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS deriv FROM nnd; -- { serverError BAD_ARGUMENTS } SELECT ts, metric, nonNegativeDerivative('string not datetime', ts, INTERVAL 3 NANOSECOND) OVER (PARTITION BY metric ORDER BY ts, metric ASC Rows BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS deriv FROM nnd; -- { serverError BAD_ARGUMENTS } -SELECT ts, metric, nonNegativeDerivative(metric, ts, INTERVAL 3 NANOSECOND, id) OVER (PARTITION BY metric ORDER BY ts, metric ASC Rows BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS deriv FROM nnd; -- { serverError BAD_ARGUMENTS } -SELECT ts, metric, nonNegativeDerivative(metric) OVER (PARTITION BY metric ORDER BY ts, metric ASC Rows BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS deriv FROM nnd; -- { serverError BAD_ARGUMENTS } +SELECT ts, metric, nonNegativeDerivative(metric, ts, INTERVAL 3 NANOSECOND, id) OVER (PARTITION BY metric ORDER BY ts, metric ASC Rows BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS deriv FROM nnd; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT ts, metric, nonNegativeDerivative(metric) OVER (PARTITION BY metric ORDER BY ts, metric ASC Rows BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS deriv FROM nnd; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -- cleanup DROP TABLE IF EXISTS nnd; diff --git a/tests/queries/0_stateless/02350_views_max_insert_threads.sql b/tests/queries/0_stateless/02350_views_max_insert_threads.sql index 25e0fdeadba..a4f7e2546ed 100644 --- a/tests/queries/0_stateless/02350_views_max_insert_threads.sql +++ b/tests/queries/0_stateless/02350_views_max_insert_threads.sql @@ -10,7 +10,7 @@ create materialized view t_mv Engine = Null AS select now() as ts, max(a) from t insert into t select * from numbers_mt(10e6) settings max_threads = 16, max_insert_threads=16, max_block_size=100000; system flush logs; -select arrayUniq(thread_ids)>=16 from system.query_log where +select peak_threads_usage>=16 from system.query_log where event_date >= yesterday() and current_database = currentDatabase() and type = 'QueryFinish' and diff --git a/tests/queries/0_stateless/02353_compression_level.sh b/tests/queries/0_stateless/02353_compression_level.sh index 8d6a9c899ad..9e102d12fed 100755 --- a/tests/queries/0_stateless/02353_compression_level.sh +++ b/tests/queries/0_stateless/02353_compression_level.sh @@ -1,12 +1,11 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel +# Tags: no-fasttest # Tag no-fasttest: depends on brotli and bzip2 CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') WORKING_FOLDER_02353="${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}" rm -rf "${WORKING_FOLDER_02353}" diff --git a/tests/queries/0_stateless/02354_distributed_with_external_aggregation_memory_usage.sql b/tests/queries/0_stateless/02354_distributed_with_external_aggregation_memory_usage.sql index a5a3da82324..105fb500461 100644 --- a/tests/queries/0_stateless/02354_distributed_with_external_aggregation_memory_usage.sql +++ b/tests/queries/0_stateless/02354_distributed_with_external_aggregation_memory_usage.sql @@ -1,4 +1,4 @@ --- Tags: long, no-tsan, no-msan, no-asan, no-ubsan, no-debug, no-s3-storage +-- Tags: long, no-tsan, no-msan, no-asan, no-ubsan, no-debug, no-object-storage DROP TABLE IF EXISTS t_2354_dist_with_external_aggr; diff --git a/tests/queries/0_stateless/02358_file_default_value.sh b/tests/queries/0_stateless/02358_file_default_value.sh index a7c4c17c129..0fd97a09546 100755 --- a/tests/queries/0_stateless/02358_file_default_value.sh +++ b/tests/queries/0_stateless/02358_file_default_value.sh @@ -4,7 +4,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') WORKING_FOLDER_02357="${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}" rm -rf "${WORKING_FOLDER_02357}" diff --git a/tests/queries/0_stateless/02360_clickhouse_local_config-option.sh b/tests/queries/0_stateless/02360_clickhouse_local_config-option.sh index b58cfd7ec21..50e07ca8612 100755 --- a/tests/queries/0_stateless/02360_clickhouse_local_config-option.sh +++ b/tests/queries/0_stateless/02360_clickhouse_local_config-option.sh @@ -15,7 +15,7 @@ echo " true - 9000 + ${CLICKHOUSE_PORT_TCP} ${SAFE_DIR} diff --git a/tests/queries/0_stateless/02361_fsync_profile_events.sh b/tests/queries/0_stateless/02361_fsync_profile_events.sh index e150d70b896..98c9cf9b7b4 100755 --- a/tests/queries/0_stateless/02361_fsync_profile_events.sh +++ b/tests/queries/0_stateless/02361_fsync_profile_events.sh @@ -1,6 +1,6 @@ #!/usr/bin/env bash -# Tags: no-s3-storage, no-random-merge-tree-settings -# Tag no-s3-storage: s3 does not have fsync +# Tags: no-object-storage, no-random-merge-tree-settings +# Tag no-object-storage: s3 does not have fsync CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02366_kql_func_binary.reference b/tests/queries/0_stateless/02366_kql_func_binary.reference index 6276cd6d867..360c1aa9899 100644 --- a/tests/queries/0_stateless/02366_kql_func_binary.reference +++ b/tests/queries/0_stateless/02366_kql_func_binary.reference @@ -1,7 +1,4 @@ -- binary functions 4 7 -1 -1 -1 7 3 1 diff --git a/tests/queries/0_stateless/02366_kql_func_binary.sql b/tests/queries/0_stateless/02366_kql_func_binary.sql index 824022b564c..687f3afb5ee 100644 --- a/tests/queries/0_stateless/02366_kql_func_binary.sql +++ b/tests/queries/0_stateless/02366_kql_func_binary.sql @@ -1,8 +1,5 @@ set dialect='kusto'; print ' -- binary functions'; print binary_and(4,7), binary_or(4,7); -print binary_shift_left(1, 1) == binary_shift_left(1, 65); -print binary_shift_right(2, 1) == binary_shift_right(2, 65); -print binary_shift_right(binary_shift_left(1, 65), 65) == 1; print binary_xor(2, 5), bitset_count_ones(42); print bitset_count_ones(binary_shift_left(binary_and(4,7), 1)); diff --git a/tests/queries/0_stateless/02370_analyzer_in_function.reference b/tests/queries/0_stateless/02370_analyzer_in_function.reference index fda174c0b7c..49a080c2616 100644 --- a/tests/queries/0_stateless/02370_analyzer_in_function.reference +++ b/tests/queries/0_stateless/02370_analyzer_in_function.reference @@ -12,3 +12,6 @@ 1 1 0 +1 +1 +1 diff --git a/tests/queries/0_stateless/02370_analyzer_in_function.sql b/tests/queries/0_stateless/02370_analyzer_in_function.sql index 7287c94deda..a6e4400e101 100644 --- a/tests/queries/0_stateless/02370_analyzer_in_function.sql +++ b/tests/queries/0_stateless/02370_analyzer_in_function.sql @@ -21,3 +21,7 @@ SELECT (1, 2) IN 1; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } SELECT (1, 2) IN [1]; -- { serverError INCORRECT_ELEMENT_OF_SET } SELECT (1, 2) IN (((1, 2), (1, 2)), ((1, 2), (1, 2))); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } SELECT (1, 2) IN [((1, 2), (1, 2)), ((1, 2), (1, 2))]; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +select (select 1) in (1); +select in(untuple(((1), (1)))); +select in(untuple(((select 1), (1)))); diff --git a/tests/queries/0_stateless/02372_analyzer_join.reference b/tests/queries/0_stateless/02372_analyzer_join.reference index eefcb1e50dc..9204dded262 100644 --- a/tests/queries/0_stateless/02372_analyzer_join.reference +++ b/tests/queries/0_stateless/02372_analyzer_join.reference @@ -5,63 +5,63 @@ JOIN INNER SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value -FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; +FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 SELECT '--'; -- SELECT t1.value, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_1_Value_1 Join_2_Value_1 -SELECT id FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } -SELECT value FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT id FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT value FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0'; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 SELECT 'JOIN expression aliases'; JOIN expression aliases -SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id); +SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 SELECT '--'; -- -SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1_id = t2_id; +SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1_id = t2_id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 SELECT 'JOIN LEFT'; @@ -69,75 +69,75 @@ JOIN LEFT SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value -FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; +FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 SELECT '--'; -- SELECT t1.value, t2.value -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_1_Value_1 Join_2_Value_1 Join_1_Value_2 -SELECT id FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } -SELECT value FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT id FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT value FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0'; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 0 2 Join_1_Value_2 0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 0 2 Join_1_Value_2 0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 0 2 Join_1_Value_2 0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 SELECT 'JOIN expression aliases'; JOIN expression aliases -SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id); +SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 SELECT '--'; -- -SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1_id = t2_id; +SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1_id = t2_id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 @@ -146,182 +146,182 @@ JOIN RIGHT SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value -FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; +FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 -0 3 Join_2_Value_3 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 -0 3 Join_2_Value_3 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; +0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 -0 0 3 3 Join_2_Value_3 Join_2_Value_3 SELECT '--'; -- SELECT t1.value, t2.value -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_1_Value_1 Join_2_Value_1 - Join_2_Value_3 -SELECT id FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } -SELECT value FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT id FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT value FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0'; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 1 Join_2_Value_1 0 3 Join_2_Value_3 +0 Join_1_Value_0 0 Join_2_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 1 Join_2_Value_1 0 3 Join_2_Value_3 +0 Join_1_Value_0 0 Join_2_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; 0 1 Join_2_Value_1 0 3 Join_2_Value_3 +0 Join_1_Value_0 0 Join_2_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id; +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 -0 3 Join_2_Value_3 SELECT 'JOIN expression aliases'; JOIN expression aliases -SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id); +SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 -0 3 Join_2_Value_3 SELECT '--'; -- -SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1_id = t2_id; +SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1_id = t2_id ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 -0 3 Join_2_Value_3 SELECT 'JOIN FULL'; JOIN FULL SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value -FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; +FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 -0 3 Join_2_Value_3 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 -0 3 Join_2_Value_3 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; +0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 -0 0 3 3 Join_2_Value_3 Join_2_Value_3 SELECT '--'; -- SELECT t1.value, t2.value -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_1_Value_1 Join_2_Value_1 Join_1_Value_2 - Join_2_Value_3 -SELECT id FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } -SELECT value FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT id FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT value FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0'; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0' ORDER BY ALL; +0 1 Join_2_Value_1 +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 0 2 Join_1_Value_2 0 -0 1 Join_2_Value_1 -0 3 Join_2_Value_3 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; +0 1 Join_2_Value_1 +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 0 2 Join_1_Value_2 0 -0 1 Join_2_Value_1 -0 3 Join_2_Value_3 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; +0 1 Join_2_Value_1 +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 0 2 Join_1_Value_2 0 -0 1 Join_2_Value_1 -0 3 Join_2_Value_3 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 -0 3 Join_2_Value_3 SELECT 'JOIN expression aliases'; JOIN expression aliases -SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id); +SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 -0 3 Join_2_Value_3 SELECT '--'; -- -SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1_id = t2_id; +SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1_id = t2_id ORDER BY ALL; +0 3 Join_2_Value_3 0 Join_1_Value_0 0 Join_2_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 2 Join_1_Value_2 0 -0 3 Join_2_Value_3 SELECT 'First JOIN INNER second JOIN INNER'; First JOIN INNER second JOIN INNER SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; @@ -329,48 +329,48 @@ SELECT '--'; SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -INNER JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +INNER JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1_id = t2_id -INNER JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +INNER JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT 'First JOIN INNER second JOIN LEFT'; @@ -379,14 +379,14 @@ SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; @@ -394,48 +394,48 @@ SELECT '--'; SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -LEFT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +LEFT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1_id = t2_id -LEFT JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +LEFT JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT 'First JOIN INNER second JOIN RIGHT'; @@ -444,159 +444,159 @@ SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 -0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_3_Value_4 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_3_Value_4 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -RIGHT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +RIGHT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1_id = t2_id -RIGHT JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +RIGHT JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT 'First JOIN INNER second JOIN FULL'; First JOIN INNER second JOIN FULL SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 INNER JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 -0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_3_Value_4 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_3_Value_4 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -FULL JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +FULL JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1_id = t2_id -FULL JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +FULL JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT 'First JOIN LEFT second JOIN INNER'; First JOIN LEFT second JOIN INNER SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -604,7 +604,7 @@ SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -613,7 +613,7 @@ SELECT '--'; SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 0 0 Join_3_Value_0 Join_3_Value_0 @@ -621,7 +621,7 @@ SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 @@ -629,20 +629,20 @@ SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 @@ -650,7 +650,7 @@ SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -INNER JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +INNER JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -658,7 +658,7 @@ SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1_id = t2_id -INNER JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +INNER JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -668,7 +668,7 @@ SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -676,7 +676,7 @@ SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -685,7 +685,7 @@ SELECT '--'; SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 0 0 Join_3_Value_0 Join_3_Value_0 @@ -693,7 +693,7 @@ SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 @@ -701,7 +701,7 @@ SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 0 0 2 Join_1_Value_2 0 0 @@ -709,7 +709,7 @@ SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 0 0 2 Join_1_Value_2 0 0 @@ -718,7 +718,7 @@ JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 @@ -726,7 +726,7 @@ SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -LEFT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +LEFT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -734,7 +734,7 @@ SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1_id = t2_id -LEFT JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +LEFT JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -744,184 +744,184 @@ SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 0 0 Join_3_Value_0 Join_3_Value_0 -0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_3_Value_4 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_3_Value_4 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -RIGHT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +RIGHT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1_id = t2_id -RIGHT JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +RIGHT JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT 'First JOIN LEFT second JOIN FULL'; First JOIN LEFT second JOIN FULL SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 LEFT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 0 0 Join_3_Value_0 Join_3_Value_0 -0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_3_Value_4 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; +0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 0 0 2 Join_1_Value_2 0 0 -0 0 4 Join_3_Value_4 -0 0 1 Join_3_Value_1 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; +0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 0 0 2 Join_1_Value_2 0 0 -0 0 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_3_Value_4 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -FULL JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +FULL JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1_id = t2_id -FULL JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +FULL JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT 'First JOIN RIGHT second JOIN INNER'; First JOIN RIGHT second JOIN INNER SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; @@ -929,48 +929,48 @@ SELECT '--'; SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -INNER JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +INNER JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1_id = t2_id -INNER JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +INNER JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 SELECT 'First JOIN RIGHT second JOIN LEFT'; @@ -979,246 +979,246 @@ SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 3 Join_2_Value_3 0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 3 Join_2_Value_3 0 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 -0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_2_Value_3 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 1 Join_2_Value_1 0 +0 3 Join_2_Value_3 0 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 1 Join_2_Value_1 0 0 3 Join_2_Value_3 0 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_2_Value_3 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -LEFT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +LEFT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 3 Join_2_Value_3 0 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1_id = t2_id -LEFT JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +LEFT JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 3 Join_2_Value_3 0 SELECT 'First JOIN RIGHT second JOIN RIGHT'; First JOIN RIGHT second JOIN RIGHT SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 -0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_3_Value_4 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_3_Value_4 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -RIGHT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +RIGHT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1_id = t2_id -RIGHT JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +RIGHT JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT 'First JOIN RIGHT second JOIN FULL'; First JOIN RIGHT second JOIN FULL SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 RIGHT JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 3 Join_2_Value_3 0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 3 Join_2_Value_3 0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 +0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 -0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 -0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_3_Value_4 + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_2_Value_3 - Join_3_Value_4 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 -0 1 Join_2_Value_1 0 -0 0 4 Join_3_Value_4 +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 +0 1 Join_2_Value_1 0 +0 3 Join_2_Value_3 0 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 -0 1 Join_2_Value_1 0 -0 3 Join_2_Value_3 0 +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 0 0 4 Join_3_Value_4 +0 1 Join_2_Value_1 0 +0 3 Join_2_Value_3 0 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_3_Value_4 + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 - Join_2_Value_3 - Join_3_Value_4 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -FULL JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +FULL JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 0 4 Join_3_Value_4 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 3 Join_2_Value_3 0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1_id = t2_id -FULL JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +FULL JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 0 4 Join_3_Value_4 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 -0 3 Join_2_Value_3 0 -0 0 4 Join_3_Value_4 SELECT 'First JOIN FULL second JOIN INNER'; First JOIN FULL second JOIN INNER SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -1226,7 +1226,7 @@ SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -1235,7 +1235,7 @@ SELECT '--'; SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 0 0 Join_3_Value_0 Join_3_Value_0 @@ -1243,7 +1243,7 @@ SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +INNER JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 @@ -1251,20 +1251,20 @@ SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +INNER JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 @@ -1272,7 +1272,7 @@ SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -INNER JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +INNER JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -1280,7 +1280,7 @@ SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1_id = t2_id -INNER JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +INNER JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 @@ -1290,265 +1290,265 @@ SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 0 0 Join_3_Value_0 Join_3_Value_0 -0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +LEFT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_2_Value_3 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; +0 1 Join_2_Value_1 0 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 0 0 2 Join_1_Value_2 0 0 -0 3 Join_2_Value_3 0 -0 1 Join_2_Value_1 0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; +0 1 Join_2_Value_1 0 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 0 0 2 Join_1_Value_2 0 0 -0 1 Join_2_Value_1 0 -0 3 Join_2_Value_3 0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +LEFT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_2_Value_3 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -LEFT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +LEFT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1_id = t2_id -LEFT JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +LEFT JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 SELECT 'First JOIN FULL second JOIN RIGHT'; First JOIN FULL second JOIN RIGHT SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 0 0 Join_3_Value_0 Join_3_Value_0 -0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +RIGHT JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_3_Value_4 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; -0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; 0 0 1 Join_3_Value_1 0 0 4 Join_3_Value_4 +0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +RIGHT JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_3_Value_4 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_3_Value_4 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -RIGHT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +RIGHT JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1_id = t2_id -RIGHT JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +RIGHT JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 0 4 Join_3_Value_4 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 0 4 Join_3_Value_4 SELECT 'First JOIN FULL second JOIN FULL'; First JOIN FULL second JOIN FULL SELECT 'JOIN ON without conditions'; JOIN ON without conditions SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 FULL JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 4 Join_3_Value_4 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; +0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 +0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 0 0 Join_1_Value_0 Join_1_Value_0 0 0 Join_2_Value_0 Join_2_Value_0 0 0 Join_3_Value_0 Join_3_Value_0 1 1 Join_1_Value_1 Join_1_Value_1 1 1 Join_2_Value_1 Join_2_Value_1 1 1 Join_3_Value_1 Join_3_Value_1 2 2 Join_1_Value_2 Join_1_Value_2 0 0 0 0 Join_3_Value_0 Join_3_Value_0 -0 0 3 3 Join_2_Value_3 Join_2_Value_3 0 0 -0 0 0 0 4 4 Join_3_Value_4 Join_3_Value_4 SELECT '--'; -- SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +FULL JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; + Join_3_Value_4 + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_2_Value_3 - Join_3_Value_4 SELECT 'JOIN ON with conditions'; JOIN ON with conditions SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; +0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 +0 1 Join_2_Value_1 0 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 0 0 2 Join_1_Value_2 0 0 -0 3 Join_2_Value_3 0 -0 1 Join_2_Value_1 0 -0 0 4 Join_3_Value_4 -0 0 1 Join_3_Value_1 SELECT '--'; -- SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; +0 0 1 Join_3_Value_1 +0 0 4 Join_3_Value_4 +0 1 Join_2_Value_1 0 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 0 0 2 Join_1_Value_2 0 0 -0 1 Join_2_Value_1 0 -0 3 Join_2_Value_3 0 -0 0 1 Join_3_Value_1 -0 0 4 Join_3_Value_4 SELECT 'JOIN multiple clauses'; JOIN multiple clauses SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +FULL JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; + Join_3_Value_4 + Join_2_Value_3 Join_1_Value_0 Join_2_Value_0 Join_3_Value_0 Join_1_Value_1 Join_2_Value_1 Join_3_Value_1 Join_1_Value_2 Join_3_Value_0 - Join_2_Value_3 - Join_3_Value_4 SELECT 'JOIN expression aliases'; JOIN expression aliases SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -FULL JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +FULL JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; +0 0 4 Join_3_Value_4 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 -0 0 4 Join_3_Value_4 SELECT '--'; -- SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1_id = t2_id -FULL JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +FULL JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; +0 0 4 Join_3_Value_4 +0 3 Join_2_Value_3 0 0 Join_1_Value_0 0 Join_2_Value_0 0 Join_3_Value_0 1 Join_1_Value_1 1 Join_2_Value_1 1 Join_3_Value_1 2 Join_1_Value_2 0 0 Join_3_Value_0 -0 3 Join_2_Value_3 0 -0 0 4 Join_3_Value_4 diff --git a/tests/queries/0_stateless/02372_analyzer_join.sql.j2 b/tests/queries/0_stateless/02372_analyzer_join.sql.j2 index facf4dc018b..45ae63b9a49 100644 --- a/tests/queries/0_stateless/02372_analyzer_join.sql.j2 +++ b/tests/queries/0_stateless/02372_analyzer_join.sql.j2 @@ -45,59 +45,59 @@ SELECT 'JOIN {{ join_type }}'; SELECT 'JOIN ON without conditions'; SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value -FROM test_table_join_1 {{ join_type }} JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; +FROM test_table_join_1 {{ join_type }} JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; SELECT '--'; SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; SELECT '--'; SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value -FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; SELECT '--'; SELECT t1.value, t2.value -FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; -SELECT id FROM test_table_join_1 {{ join_type }} JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT id FROM test_table_join_1 {{ join_type }} JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } -SELECT value FROM test_table_join_1 {{ join_type }} JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id; -- { serverError AMBIGUOUS_IDENTIFIER } +SELECT value FROM test_table_join_1 {{ join_type }} JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id ORDER BY ALL; -- { serverError AMBIGUOUS_IDENTIFIER } SELECT 'JOIN ON with conditions'; SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0'; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' ORDER BY ALL; SELECT '--'; SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t2.value = 'Join_2_Value_0' ORDER BY ALL; SELECT '--'; SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; SELECT '--'; SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0'; +FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON toString(t1.id) = toString(t2.id) AND t1.value = 'Join_1_Value_0' AND t2.value = 'Join_2_Value_0' ORDER BY ALL; SELECT 'JOIN multiple clauses'; SELECT t1.id, t1.value, t2.id, t2.value -FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id; +FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id ORDER BY ALL; SELECT 'JOIN expression aliases'; -SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id); +SELECT t1_id, t1.value, t2_id, t2.value FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) ORDER BY ALL; SELECT '--'; -SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1_id = t2_id; +SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value FROM test_table_join_1 AS t1 {{ join_type }} JOIN test_table_join_2 AS t2 ON t1_id = t2_id ORDER BY ALL; {% endfor %} @@ -110,56 +110,56 @@ SELECT 'JOIN ON without conditions'; SELECT test_table_join_1.id, test_table_join_1.value, test_table_join_2.id, test_table_join_2.value, test_table_join_3.id, test_table_join_3.value FROM test_table_join_1 {{ first_join_type }} JOIN test_table_join_2 ON test_table_join_1.id = test_table_join_2.id -{{ second_join_type }} JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id; +{{ second_join_type }} JOIN test_table_join_3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; SELECT '--'; SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 {{ first_join_type }} JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -{{ second_join_type }} JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +{{ second_join_type }} JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; SELECT '--'; SELECT t1.id, test_table_join_1.id, t1.value, test_table_join_1.value, t2.id, test_table_join_2.id, t2.value, test_table_join_2.value, t3.id, test_table_join_3.id, t3.value, test_table_join_3.value FROM test_table_join_1 AS t1 {{ first_join_type }} JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -{{ second_join_type }} JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +{{ second_join_type }} JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; SELECT '--'; SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 {{ first_join_type }} JOIN test_table_join_2 AS t2 ON test_table_join_1.id = test_table_join_2.id -{{ second_join_type }} JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id; +{{ second_join_type }} JOIN test_table_join_3 AS t3 ON test_table_join_2.id = test_table_join_3.id ORDER BY ALL; SELECT 'JOIN ON with conditions'; SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 {{ first_join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' -{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0'; +{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' ORDER BY ALL; SELECT '--'; SELECT t1.id, t1.value, t2.id, t2.value, t3.id, t3.value FROM test_table_join_1 AS t1 {{ first_join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id AND t1.value == 'Join_1_Value_0' AND t2.value == 'Join_2_Value_0' -{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0'; +{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2.id = t3.id AND t2.value == 'Join_2_Value_0' AND t3.value == 'Join_3_Value_0' ORDER BY ALL; SELECT 'JOIN multiple clauses'; SELECT t1.value, t2.value, t3.value FROM test_table_join_1 AS t1 {{ first_join_type }} JOIN test_table_join_2 AS t2 ON t1.id = t2.id OR t1.id = t2.id -{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id; +{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2.id = t3.id OR t3.id = t2.id ORDER BY ALL; SELECT 'JOIN expression aliases'; SELECT t1_id, t1.value, t2_id, t2.value, t3_id, t3.value FROM test_table_join_1 AS t1 {{ first_join_type }} JOIN test_table_join_2 AS t2 ON (t1.id AS t1_id) = (t2.id AS t2_id) -{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id); +{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2_id = (t3.id AS t3_id) ORDER BY ALL; SELECT '--'; SELECT t1.id AS t1_id, t1.value, t2.id AS t2_id, t2.value, t3.id AS t3_id, t3.value FROM test_table_join_1 AS t1 {{ first_join_type }} JOIN test_table_join_2 AS t2 ON t1_id = t2_id -{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2_id = t3_id; +{{ second_join_type }} JOIN test_table_join_3 AS t3 ON t2_id = t3_id ORDER BY ALL; {% endfor %} {% endfor %} diff --git a/tests/queries/0_stateless/02373_analyzer_join_use_nulls.reference b/tests/queries/0_stateless/02373_analyzer_join_use_nulls.reference index 3722c23e4a0..2b3671e1ea6 100644 --- a/tests/queries/0_stateless/02373_analyzer_join_use_nulls.reference +++ b/tests/queries/0_stateless/02373_analyzer_join_use_nulls.reference @@ -1,27 +1,27 @@ -- { echoOn } SELECT t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; 0 UInt64 Join_1_Value_0 String 0 UInt64 Join_2_Value_0 String 1 UInt64 Join_1_Value_1 String 1 UInt64 Join_2_Value_1 String SELECT '--'; -- SELECT t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; 0 UInt64 Join_1_Value_0 String 0 Nullable(UInt64) Join_2_Value_0 Nullable(String) 1 UInt64 Join_1_Value_1 String 1 Nullable(UInt64) Join_2_Value_1 Nullable(String) 2 UInt64 Join_1_Value_2 String \N Nullable(UInt64) \N Nullable(String) SELECT '--'; -- SELECT t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; 0 Nullable(UInt64) Join_1_Value_0 Nullable(String) 0 UInt64 Join_2_Value_0 String 1 Nullable(UInt64) Join_1_Value_1 Nullable(String) 1 UInt64 Join_2_Value_1 String \N Nullable(UInt64) \N Nullable(String) 3 UInt64 Join_2_Value_3 String SELECT '--'; -- SELECT t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; 0 Nullable(UInt64) Join_1_Value_0 Nullable(String) 0 Nullable(UInt64) Join_2_Value_0 Nullable(String) 1 Nullable(UInt64) Join_1_Value_1 Nullable(String) 1 Nullable(UInt64) Join_2_Value_1 Nullable(String) 2 Nullable(UInt64) Join_1_Value_2 Nullable(String) \N Nullable(UInt64) \N Nullable(String) @@ -30,14 +30,14 @@ SELECT '--'; -- SELECT id AS using_id, toTypeName(using_id), t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 USING (id); +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 USING (id) ORDER BY ALL; 0 UInt64 0 UInt64 Join_1_Value_0 String 0 UInt64 Join_2_Value_0 String 1 UInt64 1 UInt64 Join_1_Value_1 String 1 UInt64 Join_2_Value_1 String SELECT '--'; -- SELECT id AS using_id, toTypeName(using_id), t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 USING (id); +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 USING (id) ORDER BY ALL; 0 UInt64 0 UInt64 Join_1_Value_0 String 0 Nullable(UInt64) Join_2_Value_0 Nullable(String) 1 UInt64 1 UInt64 Join_1_Value_1 String 1 Nullable(UInt64) Join_2_Value_1 Nullable(String) 2 UInt64 2 UInt64 Join_1_Value_2 String \N Nullable(UInt64) \N Nullable(String) @@ -45,7 +45,7 @@ SELECT '--'; -- SELECT id AS using_id, toTypeName(using_id), t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 USING (id); +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 USING (id) ORDER BY ALL; 0 UInt64 0 Nullable(UInt64) Join_1_Value_0 Nullable(String) 0 UInt64 Join_2_Value_0 String 1 UInt64 1 Nullable(UInt64) Join_1_Value_1 Nullable(String) 1 UInt64 Join_2_Value_1 String 3 UInt64 \N Nullable(UInt64) \N Nullable(String) 3 UInt64 Join_2_Value_3 String @@ -53,7 +53,7 @@ SELECT '--'; -- SELECT id AS using_id, toTypeName(using_id), t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 USING (id); +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 USING (id) ORDER BY ALL; 0 Nullable(UInt64) 0 Nullable(UInt64) Join_1_Value_0 Nullable(String) 0 Nullable(UInt64) Join_2_Value_0 Nullable(String) 1 Nullable(UInt64) 1 Nullable(UInt64) Join_1_Value_1 Nullable(String) 1 Nullable(UInt64) Join_2_Value_1 Nullable(String) 2 Nullable(UInt64) 2 Nullable(UInt64) Join_1_Value_2 Nullable(String) \N Nullable(UInt64) \N Nullable(String) diff --git a/tests/queries/0_stateless/02373_analyzer_join_use_nulls.sql b/tests/queries/0_stateless/02373_analyzer_join_use_nulls.sql index db7895084e8..bcec6d178a8 100644 --- a/tests/queries/0_stateless/02373_analyzer_join_use_nulls.sql +++ b/tests/queries/0_stateless/02373_analyzer_join_use_nulls.sql @@ -26,46 +26,46 @@ INSERT INTO test_table_join_2 VALUES (3, 'Join_2_Value_3'); -- { echoOn } SELECT t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; SELECT '--'; SELECT t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; SELECT '--'; SELECT t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; SELECT '--'; SELECT t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id; +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 ON t1.id = t2.id ORDER BY ALL; SELECT '--'; SELECT id AS using_id, toTypeName(using_id), t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 USING (id); +FROM test_table_join_1 AS t1 INNER JOIN test_table_join_2 AS t2 USING (id) ORDER BY ALL; SELECT '--'; SELECT id AS using_id, toTypeName(using_id), t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 USING (id); +FROM test_table_join_1 AS t1 LEFT JOIN test_table_join_2 AS t2 USING (id) ORDER BY ALL; SELECT '--'; SELECT id AS using_id, toTypeName(using_id), t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 USING (id); +FROM test_table_join_1 AS t1 RIGHT JOIN test_table_join_2 AS t2 USING (id) ORDER BY ALL; SELECT '--'; SELECT id AS using_id, toTypeName(using_id), t1.id AS t1_id, toTypeName(t1_id), t1.value AS t1_value, toTypeName(t1_value), t2.id AS t2_id, toTypeName(t2_id), t2.value AS t2_value, toTypeName(t2_value) -FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 USING (id); +FROM test_table_join_1 AS t1 FULL JOIN test_table_join_2 AS t2 USING (id) ORDER BY ALL; -- { echoOff } diff --git a/tests/queries/0_stateless/02373_heap_buffer_overflow_in_avro.sh b/tests/queries/0_stateless/02373_heap_buffer_overflow_in_avro.sh index 3461287d28a..e95bda2adfb 100755 --- a/tests/queries/0_stateless/02373_heap_buffer_overflow_in_avro.sh +++ b/tests/queries/0_stateless/02373_heap_buffer_overflow_in_avro.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - cp $CURDIR/data_avro/corrupted.avro $USER_FILES_PATH/ $CLICKHOUSE_CLIENT -q "select * from file(corrupted.avro)" 2>&1 | grep -F -q "Cannot read compressed data" && echo "OK" || echo "FAIL" diff --git a/tests/queries/0_stateless/02378_analyzer_projection_names.reference b/tests/queries/0_stateless/02378_analyzer_projection_names.reference index fd5bc7d4ae8..532414f117c 100644 --- a/tests/queries/0_stateless/02378_analyzer_projection_names.reference +++ b/tests/queries/0_stateless/02378_analyzer_projection_names.reference @@ -443,6 +443,18 @@ SELECT '--'; -- DESCRIBE (WITH test_table_in_cte AS (SELECT id FROM test_table) SELECT id IN test_table_in_cte FROM test_table); in(id, test_table_in_cte) UInt8 +SELECT '--'; +-- +DESCRIBE (WITH test_table_in_cte_1 AS (SELECT 1 AS c1), test_table_in_cte_2 AS (SELECT 1 AS c1) SELECT * +FROM test_table_in_cte_1 INNER JOIN test_table_in_cte_2 as test_table_in_cte_2 ON test_table_in_cte_1.c1 = test_table_in_cte_2.c1); +test_table_in_cte_1.c1 UInt8 +test_table_in_cte_2.c1 UInt8 +SELECT '--'; +-- +DESCRIBE (WITH test_table_in_cte_1 AS (SELECT 1 AS c1), test_table_in_cte_2 AS (SELECT 1 AS c1 UNION ALL SELECT 1 AS c1) SELECT * +FROM test_table_in_cte_1 INNER JOIN test_table_in_cte_2 as test_table_in_cte_2 ON test_table_in_cte_1.c1 = test_table_in_cte_2.c1); +test_table_in_cte_1.c1 UInt8 +test_table_in_cte_2.c1 UInt8 SELECT 'Joins'; Joins DESCRIBE (SELECT * FROM test_table_join_1, test_table_join_2); diff --git a/tests/queries/0_stateless/02378_analyzer_projection_names.sql b/tests/queries/0_stateless/02378_analyzer_projection_names.sql index f5ac5f7476f..f41afe6a950 100644 --- a/tests/queries/0_stateless/02378_analyzer_projection_names.sql +++ b/tests/queries/0_stateless/02378_analyzer_projection_names.sql @@ -408,6 +408,16 @@ SELECT '--'; DESCRIBE (WITH test_table_in_cte AS (SELECT id FROM test_table) SELECT id IN test_table_in_cte FROM test_table); +SELECT '--'; + +DESCRIBE (WITH test_table_in_cte_1 AS (SELECT 1 AS c1), test_table_in_cte_2 AS (SELECT 1 AS c1) SELECT * +FROM test_table_in_cte_1 INNER JOIN test_table_in_cte_2 as test_table_in_cte_2 ON test_table_in_cte_1.c1 = test_table_in_cte_2.c1); + +SELECT '--'; + +DESCRIBE (WITH test_table_in_cte_1 AS (SELECT 1 AS c1), test_table_in_cte_2 AS (SELECT 1 AS c1 UNION ALL SELECT 1 AS c1) SELECT * +FROM test_table_in_cte_1 INNER JOIN test_table_in_cte_2 as test_table_in_cte_2 ON test_table_in_cte_1.c1 = test_table_in_cte_2.c1); + SELECT 'Joins'; DESCRIBE (SELECT * FROM test_table_join_1, test_table_join_2); diff --git a/tests/queries/0_stateless/02381_client_prints_server_side_time.sh b/tests/queries/0_stateless/02381_client_prints_server_side_time.sh index e6cd63da95d..81376ee3791 100755 --- a/tests/queries/0_stateless/02381_client_prints_server_side_time.sh +++ b/tests/queries/0_stateless/02381_client_prints_server_side_time.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-tsan, no-asan, no-ubsan, no-msan, no-debug, no-s3-storage +# Tags: no-tsan, no-asan, no-ubsan, no-msan, no-debug, no-object-storage CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02381_compress_marks_and_primary_key.sql b/tests/queries/0_stateless/02381_compress_marks_and_primary_key.sql index 0c228c13f19..1a584b8b5b2 100644 --- a/tests/queries/0_stateless/02381_compress_marks_and_primary_key.sql +++ b/tests/queries/0_stateless/02381_compress_marks_and_primary_key.sql @@ -1,5 +1,7 @@ -- Tags: no-random-merge-tree-settings +SET optimize_trivial_insert_select = 1; + drop table if exists test_02381; create table test_02381(a UInt64, b UInt64) ENGINE = MergeTree order by (a, b) SETTINGS compress_marks = false, compress_primary_key = false, ratio_of_defaults_for_sparse_serialization = 1; insert into test_02381 select number, number * 10 from system.numbers limit 1000000; diff --git a/tests/queries/0_stateless/02383_arrow_dict_special_cases.sh b/tests/queries/0_stateless/02383_arrow_dict_special_cases.sh index 80743a97dd0..ae08941da63 100755 --- a/tests/queries/0_stateless/02383_arrow_dict_special_cases.sh +++ b/tests/queries/0_stateless/02383_arrow_dict_special_cases.sh @@ -5,7 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') UNIQ_DEST_PATH=$USER_FILES_PATH/test-02383-$RANDOM-$RANDOM mkdir -p $UNIQ_DEST_PATH diff --git a/tests/queries/0_stateless/02391_recursive_buffer.sql b/tests/queries/0_stateless/02391_recursive_buffer.sql index 1a630722b5a..60a2f0d1af1 100644 --- a/tests/queries/0_stateless/02391_recursive_buffer.sql +++ b/tests/queries/0_stateless/02391_recursive_buffer.sql @@ -10,9 +10,5 @@ DROP TABLE test; DROP TABLE IF EXISTS test1; DROP TABLE IF EXISTS test2; CREATE TABLE test1 (key UInt32) Engine = Buffer(currentDatabase(), test2, 16, 10, 100, 10000, 1000000, 10000000, 100000000); -CREATE TABLE test2 (key UInt32) Engine = Buffer(currentDatabase(), test1, 16, 10, 100, 10000, 1000000, 10000000, 100000000); -SELECT * FROM test1; -- { serverError TOO_DEEP_RECURSION } -SELECT * FROM test2; -- { serverError TOO_DEEP_RECURSION } -SELECT * FROM system.tables WHERE table IN ('test1', 'test2') AND database = currentDatabase(); -- { serverError TOO_DEEP_RECURSION } +CREATE TABLE test2 (key UInt32) Engine = Buffer(currentDatabase(), test1, 16, 10, 100, 10000, 1000000, 10000000, 100000000); -- { serverError INFINITE_LOOP } DROP TABLE test1; -DROP TABLE test2; diff --git a/tests/queries/0_stateless/02402_capnp_format_segments_overflow.sh b/tests/queries/0_stateless/02402_capnp_format_segments_overflow.sh index 8aad68ffe5c..3028451a5f5 100755 --- a/tests/queries/0_stateless/02402_capnp_format_segments_overflow.sh +++ b/tests/queries/0_stateless/02402_capnp_format_segments_overflow.sh @@ -5,11 +5,10 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') mkdir -p $USER_FILES_PATH/test_02402 cp $CURDIR/data_capnp/overflow.capnp $USER_FILES_PATH/test_02402/ -SCHEMADIR=$($CLICKHOUSE_CLIENT_BINARY --query "select * from file('test_02402/overflow.capnp', 'CapnProto', 'val1 char') settings format_schema='nonexist:Message'" 2>&1 | grep Exception | grep -oP "file \K.*(?=/nonexist.capnp)") +SCHEMADIR=${CLICKHOUSE_SCHEMA_FILES} CLIENT_SCHEMADIR=$CURDIR/format_schemas SERVER_SCHEMADIR=test_02402 diff --git a/tests/queries/0_stateless/02404_memory_bound_merging.sql b/tests/queries/0_stateless/02404_memory_bound_merging.sql index 5e017e79309..a2de19dff8a 100644 --- a/tests/queries/0_stateless/02404_memory_bound_merging.sql +++ b/tests/queries/0_stateless/02404_memory_bound_merging.sql @@ -7,6 +7,7 @@ drop table if exists t_different_dbs; drop table if exists dist_t; drop table if exists t; +set optimize_trivial_insert_select = 1; create table t(a UInt64, b UInt64) engine=MergeTree order by a; system stop merges t; diff --git a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference index a152066a460..8dd8910c858 100644 --- a/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference +++ b/tests/queries/0_stateless/02415_all_new_functions_must_be_documented.reference @@ -876,7 +876,6 @@ tryBase58Decode tumble tumbleEnd tumbleStart -tuple tupleConcat tupleDivide tupleDivideByNumber diff --git a/tests/queries/0_stateless/02417_load_marks_async.sh b/tests/queries/0_stateless/02417_load_marks_async.sh index 72b35a565df..950656e7ab6 100755 --- a/tests/queries/0_stateless/02417_load_marks_async.sh +++ b/tests/queries/0_stateless/02417_load_marks_async.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02421_record_errors_row_by_input_format.sh b/tests/queries/0_stateless/02421_record_errors_row_by_input_format.sh index df304eeeba5..72b55144348 100755 --- a/tests/queries/0_stateless/02421_record_errors_row_by_input_format.sh +++ b/tests/queries/0_stateless/02421_record_errors_row_by_input_format.sh @@ -9,10 +9,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # Data preparation. -CLICKHOUSE_USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path, _file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -mkdir -p ${CLICKHOUSE_USER_FILES_PATH}/ -echo -e "1,1\n2,a\nb,3\n4,4\n5,c\n6,6" > ${CLICKHOUSE_USER_FILES_PATH}/a.csv +echo -e "1,1\n2,a\nb,3\n4,4\n5,c\n6,6" > ${USER_FILES_PATH}/a.csv ${CLICKHOUSE_CLIENT} --query "drop table if exists data;" ${CLICKHOUSE_CLIENT} --query "create table data (A UInt8, B UInt8) engine=MergeTree() order by A;" @@ -23,12 +20,12 @@ sleep 2 ${CLICKHOUSE_CLIENT} --query "select * except (time) from file('errors_server', 'CSV', 'time DateTime, database Nullable(String), table Nullable(String), offset UInt32, reason String, raw_data String');" # Client side -${CLICKHOUSE_CLIENT} --input_format_allow_errors_num 4 --input_format_record_errors_file_path "${CLICKHOUSE_USER_FILES_PATH}/errors_client" --query "insert into data(A, B) format CSV" < ${CLICKHOUSE_USER_FILES_PATH}/a.csv +${CLICKHOUSE_CLIENT} --input_format_allow_errors_num 4 --input_format_record_errors_file_path "${USER_FILES_PATH}/errors_client" --query "insert into data(A, B) format CSV" < ${USER_FILES_PATH}/a.csv sleep 2 ${CLICKHOUSE_CLIENT} --query "select * except (time) from file('errors_client', 'CSV', 'time DateTime, database Nullable(String), table Nullable(String), offset UInt32, reason String, raw_data String');" # Restore ${CLICKHOUSE_CLIENT} --query "drop table if exists data;" -rm ${CLICKHOUSE_USER_FILES_PATH}/a.csv -rm ${CLICKHOUSE_USER_FILES_PATH}/errors_server -rm ${CLICKHOUSE_USER_FILES_PATH}/errors_client +rm ${USER_FILES_PATH}/a.csv +rm ${USER_FILES_PATH}/errors_server +rm ${USER_FILES_PATH}/errors_client diff --git a/tests/queries/0_stateless/02422_allow_implicit_no_password.sh b/tests/queries/0_stateless/02422_allow_implicit_no_password.sh index 013c367e079..3c433856be2 100755 --- a/tests/queries/0_stateless/02422_allow_implicit_no_password.sh +++ b/tests/queries/0_stateless/02422_allow_implicit_no_password.sh @@ -9,7 +9,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -cp /etc/clickhouse-server/users.xml "$CURDIR"/users.xml +cp ${CLICKHOUSE_CONFIG_DIR}/users.xml "$CURDIR"/users.xml sed -i 's/<\/password>/c64c5e4e53ea1a9f1427d2713b3a22bbebe8940bc807adaf654744b1568c70ab<\/password_sha256_hex>/g' "$CURDIR"/users.xml sed -i 's//1<\/access_management>/g' "$CURDIR"/users.xml diff --git a/tests/queries/0_stateless/02440_mutations_finalization.sql b/tests/queries/0_stateless/02440_mutations_finalization.sql index c522d8ab9df..92ed6a327ad 100644 --- a/tests/queries/0_stateless/02440_mutations_finalization.sql +++ b/tests/queries/0_stateless/02440_mutations_finalization.sql @@ -1,3 +1,4 @@ +-- Tags: no-parallel create table mut (n int) engine=ReplicatedMergeTree('/test/02440/{database}/mut', '1') order by tuple(); set insert_keeper_fault_injection_probability=0; diff --git a/tests/queries/0_stateless/02443_detach_attach_partition.sh b/tests/queries/0_stateless/02443_detach_attach_partition.sh index d72d771a150..6a47b7d8d61 100755 --- a/tests/queries/0_stateless/02443_detach_attach_partition.sh +++ b/tests/queries/0_stateless/02443_detach_attach_partition.sh @@ -73,7 +73,7 @@ kill -TERM $PID_1 && kill -TERM $PID_2 && kill -TERM $PID_3 && kill -TERM $PID_4 wait $CLICKHOUSE_CLIENT -q "SELECT '$CLICKHOUSE_DATABASE', 'threads finished'" -wait_for_queries_to_finish +wait_for_queries_to_finish 60 $CLICKHOUSE_CLIENT -q "SYSTEM SYNC REPLICA alter_table0" $CLICKHOUSE_CLIENT -q "SYSTEM SYNC REPLICA alter_table1" diff --git a/tests/queries/0_stateless/02446_parent_zero_copy_locks.sql b/tests/queries/0_stateless/02446_parent_zero_copy_locks.sql index 86eda526c72..a44322e02cf 100644 --- a/tests/queries/0_stateless/02446_parent_zero_copy_locks.sql +++ b/tests/queries/0_stateless/02446_parent_zero_copy_locks.sql @@ -7,7 +7,7 @@ create table rmt2 (n int, m int, k int) engine=ReplicatedMergeTree('/test/02446/ settings storage_policy='s3_cache', allow_remote_fs_zero_copy_replication=1, old_parts_lifetime=0, cleanup_delay_period=0, max_cleanup_delay_period=1, cleanup_delay_period_random_add=1, min_bytes_for_wide_part=0; -- FIXME zero-copy locks may remain in ZooKeeper forever if we failed to insert a part. --- Probably that's why we have to replace repsistent lock with ephemeral sometimes. +-- Probably that's why we have to replace persistent lock with ephemeral sometimes. -- See also "Replacing persistent lock with ephemeral for path {}. It can happen only in case of local part loss" -- in StorageReplicatedMergeTree::createZeroCopyLockNode set insert_keeper_fault_injection_probability=0; @@ -23,6 +23,10 @@ select sleepEachRow(0.5) as test_does_not_rely_on_this; insert into rmt1 values(5, 5, 5); alter table rmt2 update m = m * 10 where 1 settings mutations_sync=2; +-- wait for parts to be merged +select throwIf(name = 'all_0_5_1_6') from system.parts where database=currentDatabase() and table like 'rmt%' and active +format Null; -- { retry 30 until serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } + system sync replica rmt2; set optimize_throw_if_noop=1; optimize table rmt2 final; @@ -32,10 +36,10 @@ select 1, * from rmt1 order by n; system sync replica rmt1; select 2, * from rmt2 order by n; --- a funny way to wait for outdated parts to be removed -select sleep(1), sleepEachRow(0.1) from url('http://localhost:8123/?param_tries={1..10}&query=' || encodeURLComponent( - 'select *, _state from system.parts where database=''' || currentDatabase() || ''' and table like ''rmt%'' and active=0' - ), 'LineAsString', 's String') settings max_threads=1 format Null; +-- wait for outdated parts to be removed (do not ignore _state column, so it will count Deleting parts as well) +select throwIf(count() = 0), groupArray(_state) from ( +select *, _state from system.parts where database=currentDatabase() and table like 'rmt%' and active=0 +) format Null; -- { retry 30 until serverError FUNCTION_THROW_IF_VALUE_IS_NON_ZERO } select *, _state from system.parts where database=currentDatabase() and table like 'rmt%' and active=0; diff --git a/tests/queries/0_stateless/02447_drop_database_replica.reference b/tests/queries/0_stateless/02447_drop_database_replica.reference index bd3b689ca3c..d997b7ba830 100644 --- a/tests/queries/0_stateless/02447_drop_database_replica.reference +++ b/tests/queries/0_stateless/02447_drop_database_replica.reference @@ -12,11 +12,21 @@ t 2 rdb_default 1 1 s1 r1 1 2 -2 -2 +skip inactive s1 r1 OK 2 0 s1 r2 QUEUED 2 0 s2 r1 QUEUED 2 0 +s1 r1 OK 2 0 +s1 r2 QUEUED 2 0 +s2 r1 QUEUED 2 0 +timeout on active +2 +2 +s1 r1 OK 3 0 +s1 r2 QUEUED 3 0 +s2 r1 QUEUED 3 0 +s9 r9 QUEUED 3 0 +drop replica 2 rdb_default 1 1 s1 r1 1 rdb_default 1 2 s1 r2 0 @@ -24,6 +34,9 @@ rdb_default 1 2 s1 r2 0 2 t t2 +t22 t3 +t33 t4 +t44 rdb_default_4 1 1 s1 r1 1 diff --git a/tests/queries/0_stateless/02447_drop_database_replica.sh b/tests/queries/0_stateless/02447_drop_database_replica.sh index 1604d527f2b..93a5fcee8e2 100755 --- a/tests/queries/0_stateless/02447_drop_database_replica.sh +++ b/tests/queries/0_stateless/02447_drop_database_replica.sh @@ -33,10 +33,27 @@ $CLICKHOUSE_CLIENT -q "select cluster, shard_num, replica_num, database_shard_na $CLICKHOUSE_CLIENT -q "system drop database replica 's1|r1' from database $db2" 2>&1| grep -Fac "is active, cannot drop it" # Also check that it doesn't exceed distributed_ddl_task_timeout waiting for inactive replicas -timeout 60s $CLICKHOUSE_CLIENT --distributed_ddl_task_timeout=1000 --distributed_ddl_output_mode=none_only_active -q "create table $db.t2 (n int) engine=Log" 2>&1| grep -Fac "TIMEOUT_EXCEEDED" -timeout 60s $CLICKHOUSE_CLIENT --distributed_ddl_task_timeout=1000 --distributed_ddl_output_mode=throw_only_active -q "create table $db.t3 (n int) engine=Log" 2>&1| grep -Fac "TIMEOUT_EXCEEDED" +echo 'skip inactive' +timeout 60s $CLICKHOUSE_CLIENT --distributed_ddl_task_timeout=1000 --distributed_ddl_output_mode=none_only_active -q "create table $db.t2 (n int) engine=Log" +timeout 60s $CLICKHOUSE_CLIENT --distributed_ddl_task_timeout=1000 --distributed_ddl_output_mode=throw_only_active -q "create table $db.t3 (n int) engine=Log" | sort timeout 60s $CLICKHOUSE_CLIENT --distributed_ddl_task_timeout=1000 --distributed_ddl_output_mode=null_status_on_timeout_only_active -q "create table $db.t4 (n int) engine=Log" | sort +# And that it still throws TIMEOUT_EXCEEDED for active replicas +echo 'timeout on active' +db9="${db}_9" +$CLICKHOUSE_CLIENT -q "create database $db9 engine=Replicated('/test/$CLICKHOUSE_DATABASE/rdb', 's9', 'r9')" +$CLICKHOUSE_CLIENT -q "detach database $db9" +$CLICKHOUSE_CLIENT -q "insert into system.zookeeper(name, path, value) values ('active', '/test/$CLICKHOUSE_DATABASE/rdb/replicas/s9|r9', '$($CLICKHOUSE_CLIENT -q "select serverUUID()")')" + +$CLICKHOUSE_CLIENT --distributed_ddl_task_timeout=5 --distributed_ddl_output_mode=none_only_active -q "create table $db.t22 (n int) engine=Log" 2>&1| grep -Fac "TIMEOUT_EXCEEDED" +$CLICKHOUSE_CLIENT --distributed_ddl_task_timeout=5 --distributed_ddl_output_mode=throw_only_active -q "create table $db.t33 (n int) engine=Log" 2>&1| grep -Fac "TIMEOUT_EXCEEDED" +$CLICKHOUSE_CLIENT --distributed_ddl_task_timeout=5 --distributed_ddl_output_mode=null_status_on_timeout_only_active -q "create table $db.t44 (n int) engine=Log" | sort + +$CLICKHOUSE_CLIENT -q "attach database $db9" +$CLICKHOUSE_CLIENT -q "drop database $db9" + +echo 'drop replica' + $CLICKHOUSE_CLIENT -q "detach database $db3" $CLICKHOUSE_CLIENT -q "system drop database replica 'r1' from shard 's2' from database $db" $CLICKHOUSE_CLIENT -q "attach database $db3" 2>/dev/null diff --git a/tests/queries/0_stateless/02454_create_table_with_custom_disk.sql b/tests/queries/0_stateless/02454_create_table_with_custom_disk.sql index 6cb1c0774aa..a2d46cf6d1b 100644 --- a/tests/queries/0_stateless/02454_create_table_with_custom_disk.sql +++ b/tests/queries/0_stateless/02454_create_table_with_custom_disk.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage, no-replicated-database +-- Tags: no-object-storage, no-replicated-database DROP TABLE IF EXISTS test; diff --git a/tests/queries/0_stateless/02455_one_row_from_csv_memory_usage.sh b/tests/queries/0_stateless/02455_one_row_from_csv_memory_usage.sh index 05de3f05562..7906f2917c4 100755 --- a/tests/queries/0_stateless/02455_one_row_from_csv_memory_usage.sh +++ b/tests/queries/0_stateless/02455_one_row_from_csv_memory_usage.sh @@ -5,11 +5,9 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep -E '^Code: 107.*FILE_DOESNT_EXIST' | head -1 | awk '{gsub("/nonexist.txt","",$9); print $9}') +cp "$CUR_DIR"/data_csv/10m_rows.csv.xz $USER_FILES_PATH/${CLICKHOUSE_DATABASE}_10m_rows.csv.xz -cp "$CUR_DIR"/data_csv/10m_rows.csv.xz $USER_FILES_PATH/ +${CLICKHOUSE_CLIENT} --query="SELECT * FROM file('${CLICKHOUSE_DATABASE}_10m_rows.csv.xz' , 'CSVWithNames') order by identifier, number, name, surname, birthday LIMIT 1 settings input_format_parallel_parsing=1, max_threads=1, max_parsing_threads=16, min_chunk_bytes_for_parallel_parsing=10485760, max_memory_usage=1000000000" +${CLICKHOUSE_CLIENT} --query="SELECT * FROM file('${CLICKHOUSE_DATABASE}_10m_rows.csv.xz' , 'CSVWithNames') order by identifier, number, name, surname, birthday LIMIT 1 settings input_format_parallel_parsing=1, max_threads=1, max_parsing_threads=16, min_chunk_bytes_for_parallel_parsing=10485760, max_memory_usage=100000000" -${CLICKHOUSE_CLIENT} --query="SELECT * FROM file('10m_rows.csv.xz' , 'CSVWithNames') order by identifier, number, name, surname, birthday LIMIT 1 settings input_format_parallel_parsing=1, max_threads=1, max_parsing_threads=16, min_chunk_bytes_for_parallel_parsing=10485760, max_memory_usage=1000000000" -${CLICKHOUSE_CLIENT} --query="SELECT * FROM file('10m_rows.csv.xz' , 'CSVWithNames') order by identifier, number, name, surname, birthday LIMIT 1 settings input_format_parallel_parsing=1, max_threads=1, max_parsing_threads=16, min_chunk_bytes_for_parallel_parsing=10485760, max_memory_usage=100000000" - -rm $USER_FILES_PATH/10m_rows.csv.xz +rm $USER_FILES_PATH/${CLICKHOUSE_DATABASE}_10m_rows.csv.xz diff --git a/tests/queries/0_stateless/02457_bz2_concatenated.sh b/tests/queries/0_stateless/02457_bz2_concatenated.sh index 96e23cbfa2a..a9991cf44e7 100755 --- a/tests/queries/0_stateless/02457_bz2_concatenated.sh +++ b/tests/queries/0_stateless/02457_bz2_concatenated.sh @@ -6,7 +6,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') WORKING_FOLDER_02457="${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}" rm -rf "${WORKING_FOLDER_02457}" diff --git a/tests/queries/0_stateless/02459_glob_for_recursive_directory_traversal.sh b/tests/queries/0_stateless/02459_glob_for_recursive_directory_traversal.sh index b8430307ea3..b86385b72c4 100755 --- a/tests/queries/0_stateless/02459_glob_for_recursive_directory_traversal.sh +++ b/tests/queries/0_stateless/02459_glob_for_recursive_directory_traversal.sh @@ -5,28 +5,26 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -mkdir $user_files_path/d1 -touch $user_files_path/d1/text1.txt +mkdir $USER_FILES_PATH/d1 +touch $USER_FILES_PATH/d1/text1.txt for i in {1..2} do - echo $i$'\t'$i >> $user_files_path/d1/text1.txt + echo $i$'\t'$i >> $USER_FILES_PATH/d1/text1.txt done -mkdir $user_files_path/d1/d2 -touch $user_files_path/d1/d2/text2.txt +mkdir $USER_FILES_PATH/d1/d2 +touch $USER_FILES_PATH/d1/d2/text2.txt for i in {3..4} do - echo $i$'\t'$i >> $user_files_path/d1/d2/text2.txt + echo $i$'\t'$i >> $USER_FILES_PATH/d1/d2/text2.txt done -mkdir $user_files_path/d1/d2/d3 -touch $user_files_path/d1/d2/d3/text3.txt +mkdir $USER_FILES_PATH/d1/d2/d3 +touch $USER_FILES_PATH/d1/d2/d3/text3.txt for i in {5..6} do - echo $i$'\t'$i >> $user_files_path/d1/d2/d3/text3.txt + echo $i$'\t'$i >> $USER_FILES_PATH/d1/d2/d3/text3.txt done ${CLICKHOUSE_CLIENT} -q "SELECT * from file ('d1/*','TSV', 'Index UInt8, Number UInt8')" | sort --numeric-sort @@ -35,9 +33,9 @@ ${CLICKHOUSE_CLIENT} -q "SELECT * from file ('d1/*/tex*','TSV', 'Index UInt8, Nu ${CLICKHOUSE_CLIENT} -q "SELECT * from file ('d1/**/tex*','TSV', 'Index UInt8, Number UInt8')" | sort --numeric-sort -rm $user_files_path/d1/d2/d3/text3.txt -rmdir $user_files_path/d1/d2/d3 -rm $user_files_path/d1/d2/text2.txt -rmdir $user_files_path/d1/d2 -rm $user_files_path/d1/text1.txt -rmdir $user_files_path/d1 +rm $USER_FILES_PATH/d1/d2/d3/text3.txt +rmdir $USER_FILES_PATH/d1/d2/d3 +rm $USER_FILES_PATH/d1/d2/text2.txt +rmdir $USER_FILES_PATH/d1/d2 +rm $USER_FILES_PATH/d1/text1.txt +rmdir $USER_FILES_PATH/d1 diff --git a/tests/queries/0_stateless/02461_prewhere_row_level_policy_lightweight_delete.sql.j2 b/tests/queries/0_stateless/02461_prewhere_row_level_policy_lightweight_delete.sql.j2 index 0ec6b2ed144..d7cbf210506 100644 --- a/tests/queries/0_stateless/02461_prewhere_row_level_policy_lightweight_delete.sql.j2 +++ b/tests/queries/0_stateless/02461_prewhere_row_level_policy_lightweight_delete.sql.j2 @@ -1,5 +1,7 @@ {% for index_granularity in [999, 1000, 1001, 9999, 10000, 10001] %} +SET optimize_trivial_insert_select = 1; + DROP TABLE IF EXISTS url_na_log; CREATE TABLE url_na_log(SiteId UInt32, DateVisit Date, PRIMARY KEY (SiteId)) diff --git a/tests/queries/0_stateless/02475_bson_each_row_format.sh b/tests/queries/0_stateless/02475_bson_each_row_format.sh index 474a6cd0e47..2975b40c868 100755 --- a/tests/queries/0_stateless/02475_bson_each_row_format.sh +++ b/tests/queries/0_stateless/02475_bson_each_row_format.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-debug +# Tags: no-debug CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02477_age_datetime64.reference b/tests/queries/0_stateless/02477_age_datetime64.reference index 3b4459dd26d..fb085f461c9 100644 --- a/tests/queries/0_stateless/02477_age_datetime64.reference +++ b/tests/queries/0_stateless/02477_age_datetime64.reference @@ -111,3 +111,8 @@ SELECT age('day', materialize(toDateTime64('2015-08-18 00:00:00', 0, 'UTC')), ma 1 SELECT age('day', materialize(toDate('2015-08-18', 'UTC')), materialize(toDateTime64('2015-08-19 00:00:00', 3, 'UTC'))); 1 +-- UBsan bug #66638 +set session_timezone = 'UTC'; +SELECT age('second', toDateTime(1157339245694594829, 6, 'UTC'), toDate('2015-08-18')) + +-8973935999 diff --git a/tests/queries/0_stateless/02477_age_datetime64.sql b/tests/queries/0_stateless/02477_age_datetime64.sql index 1bed93991ca..b5fa4da8837 100644 --- a/tests/queries/0_stateless/02477_age_datetime64.sql +++ b/tests/queries/0_stateless/02477_age_datetime64.sql @@ -75,3 +75,7 @@ SELECT age('second', materialize(toDateTime64('2015-08-18 00:00:00', 0, 'UTC')), SELECT age('second', materialize(toDateTime('2015-08-18 00:00:00', 'UTC')), materialize(toDateTime64('2015-08-18 00:00:10', 3, 'UTC'))); SELECT age('day', materialize(toDateTime64('2015-08-18 00:00:00', 0, 'UTC')), materialize(toDate('2015-08-19', 'UTC'))); SELECT age('day', materialize(toDate('2015-08-18', 'UTC')), materialize(toDateTime64('2015-08-19 00:00:00', 3, 'UTC'))); + +-- UBsan bug #66638 +set session_timezone = 'UTC'; +SELECT age('second', toDateTime(1157339245694594829, 6, 'UTC'), toDate('2015-08-18')) diff --git a/tests/queries/0_stateless/02479_race_condition_between_insert_and_droppin_mv.sh b/tests/queries/0_stateless/02479_race_condition_between_insert_and_droppin_mv.sh index 6899b31d1d9..935ea03a947 100755 --- a/tests/queries/0_stateless/02479_race_condition_between_insert_and_droppin_mv.sh +++ b/tests/queries/0_stateless/02479_race_condition_between_insert_and_droppin_mv.sh @@ -38,7 +38,7 @@ ${CLICKHOUSE_CLIENT} -q "CREATE TABLE test_race_condition_landing (number Int64, export -f drop_mv; export -f insert; -TIMEOUT=55 +TIMEOUT=50 for i in {1..4} do diff --git a/tests/queries/0_stateless/02481_async_insert_race_long.sh b/tests/queries/0_stateless/02481_async_insert_race_long.sh index d8153967e9a..b0088017d32 100755 --- a/tests/queries/0_stateless/02481_async_insert_race_long.sh +++ b/tests/queries/0_stateless/02481_async_insert_race_long.sh @@ -13,7 +13,7 @@ function insert1() { local TIMELIMIT=$((SECONDS+$1)) while [ $SECONDS -lt "$TIMELIMIT" ]; do - ${MY_CLICKHOUSE_CLIENT} --wait_for_async_insert 0 -q 'INSERT INTO async_inserts_race FORMAT CSV 1,"a"' + ${MY_CLICKHOUSE_CLIENT} --insert_keeper_fault_injection_probability=0 --wait_for_async_insert 0 -q 'INSERT INTO async_inserts_race FORMAT CSV 1,"a"' done } @@ -21,7 +21,7 @@ function insert2() { local TIMELIMIT=$((SECONDS+$1)) while [ $SECONDS -lt "$TIMELIMIT" ]; do - ${MY_CLICKHOUSE_CLIENT} --wait_for_async_insert 0 -q 'INSERT INTO async_inserts_race FORMAT JSONEachRow {"id": 5, "s": "e"} {"id": 6, "s": "f"}' + ${MY_CLICKHOUSE_CLIENT} --insert_keeper_fault_injection_probability=0 --wait_for_async_insert 0 -q 'INSERT INTO async_inserts_race FORMAT JSONEachRow {"id": 5, "s": "e"} {"id": 6, "s": "f"}' done } @@ -29,7 +29,7 @@ function insert3() { local TIMELIMIT=$((SECONDS+$1)) while [ $SECONDS -lt "$TIMELIMIT" ]; do - ${MY_CLICKHOUSE_CLIENT} --wait_for_async_insert 1 -q "INSERT INTO async_inserts_race VALUES (7, 'g') (8, 'h')" & + ${MY_CLICKHOUSE_CLIENT} --insert_keeper_fault_injection_probability=0 --wait_for_async_insert 1 -q "INSERT INTO async_inserts_race VALUES (7, 'g') (8, 'h')" & sleep 0.05 done diff --git a/tests/queries/0_stateless/02482_capnp_list_of_structs.sh b/tests/queries/0_stateless/02482_capnp_list_of_structs.sh index 9d78b9893dd..a04c631c411 100755 --- a/tests/queries/0_stateless/02482_capnp_list_of_structs.sh +++ b/tests/queries/0_stateless/02482_capnp_list_of_structs.sh @@ -5,10 +5,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') touch $USER_FILES_PATH/data.capnp -SCHEMADIR=$($CLICKHOUSE_CLIENT_BINARY --query "select * from file('data.capnp', 'CapnProto', 'val1 char') settings format_schema='nonexist:Message'" 2>&1 | grep Exception | grep -oP "file \K.*(?=/nonexist.capnp)") +SCHEMADIR=${CLICKHOUSE_SCHEMA_FILES} CLIENT_SCHEMADIR=$CURDIR/format_schemas SERVER_SCHEMADIR=test_02482 mkdir -p $SCHEMADIR/$SERVER_SCHEMADIR diff --git a/tests/queries/0_stateless/02483_capnp_decimals.sh b/tests/queries/0_stateless/02483_capnp_decimals.sh index ef545a5539f..bc19b63fc8b 100755 --- a/tests/queries/0_stateless/02483_capnp_decimals.sh +++ b/tests/queries/0_stateless/02483_capnp_decimals.sh @@ -5,19 +5,17 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -touch $USER_FILES_PATH/data.capnp +touch $CLICKHOUSE_USER_FILES/data.capnp -SCHEMADIR=$($CLICKHOUSE_CLIENT_BINARY --query "select * from file('data.capnp', 'CapnProto', 'val1 char') settings format_schema='nonexist:Message'" 2>&1 | grep Exception | grep -oP "file \K.*(?=/nonexist.capnp)") CLIENT_SCHEMADIR=$CURDIR/format_schemas SERVER_SCHEMADIR=test_02483 -mkdir -p $SCHEMADIR/$SERVER_SCHEMADIR -cp -r $CLIENT_SCHEMADIR/02483_* $SCHEMADIR/$SERVER_SCHEMADIR/ +mkdir -p $CLICKHOUSE_SCHEMA_FILES/$SERVER_SCHEMADIR +cp -r $CLIENT_SCHEMADIR/02483_* $CLICKHOUSE_SCHEMA_FILES/$SERVER_SCHEMADIR/ $CLICKHOUSE_CLIENT -q "insert into function file(02483_data.capnp, auto, 'decimal32 Decimal32(3), decimal64 Decimal64(6)') select 42.42, 4242.424242 settings format_schema='$SERVER_SCHEMADIR/02483_decimals.capnp:Message', engine_file_truncate_on_insert=1" $CLICKHOUSE_CLIENT -q "select * from file(02483_data.capnp) settings format_schema='$SERVER_SCHEMADIR/02483_decimals.capnp:Message'" $CLICKHOUSE_CLIENT -q "select * from file(02483_data.capnp, auto, 'decimal64 Decimal64(6), decimal32 Decimal32(3)') settings format_schema='$SERVER_SCHEMADIR/02483_decimals.capnp:Message'" -rm $USER_FILES_PATH/data.capnp -rm $USER_FILES_PATH/02483_data.capnp +rm $CLICKHOUSE_USER_FILES/data.capnp +rm $CLICKHOUSE_USER_FILES/02483_data.capnp diff --git a/tests/queries/0_stateless/02494_query_cache_asynchronous_metrics.reference b/tests/queries/0_stateless/02494_query_cache_asynchronous_metrics.reference new file mode 100644 index 00000000000..6ed281c757a --- /dev/null +++ b/tests/queries/0_stateless/02494_query_cache_asynchronous_metrics.reference @@ -0,0 +1,2 @@ +1 +1 diff --git a/tests/queries/0_stateless/02494_query_cache_asynchronous_metrics.sql b/tests/queries/0_stateless/02494_query_cache_asynchronous_metrics.sql new file mode 100644 index 00000000000..d8de4facb38 --- /dev/null +++ b/tests/queries/0_stateless/02494_query_cache_asynchronous_metrics.sql @@ -0,0 +1,13 @@ +-- Tags: no-parallel +-- Tag no-parallel: Messes with internal cache + +SYSTEM DROP QUERY CACHE; + +-- Create an entry in the query cache +SELECT 1 SETTINGS use_query_cache = true; + +-- Asynchronous metrics must know about the entry +SYSTEM RELOAD ASYNCHRONOUS METRICS; +SELECT value FROM system.asynchronous_metrics WHERE metric = 'QueryCacheEntries'; + +SYSTEM DROP QUERY CACHE; diff --git a/tests/queries/0_stateless/02494_query_cache_empty_tuple.reference b/tests/queries/0_stateless/02494_query_cache_empty_tuple.reference new file mode 100644 index 00000000000..50e44edaecb --- /dev/null +++ b/tests/queries/0_stateless/02494_query_cache_empty_tuple.reference @@ -0,0 +1,2 @@ +() 0 +() 0 diff --git a/tests/queries/0_stateless/02494_query_cache_empty_tuple.sql b/tests/queries/0_stateless/02494_query_cache_empty_tuple.sql new file mode 100644 index 00000000000..8e133143ef8 --- /dev/null +++ b/tests/queries/0_stateless/02494_query_cache_empty_tuple.sql @@ -0,0 +1,2 @@ +SELECT tuple(), 0 FROM numbers(1) SETTINGS use_query_cache = true; +SELECT tuple(), 0 FROM numbers(1) SETTINGS use_query_cache = true; diff --git a/tests/queries/0_stateless/02496_remove_redundant_sorting.reference b/tests/queries/0_stateless/02496_remove_redundant_sorting.reference index 4a4e898c5bd..77ef213b36d 100644 --- a/tests/queries/0_stateless/02496_remove_redundant_sorting.reference +++ b/tests/queries/0_stateless/02496_remove_redundant_sorting.reference @@ -332,12 +332,13 @@ SETTINGS optimize_aggregators_of_group_by_keys=0 -- avoid removing any() as it d Expression (Projection) Sorting (Sorting for ORDER BY) Expression (Before ORDER BY) - Filter (((WHERE + (Projection + Before ORDER BY)) + HAVING)) - Aggregating - Expression ((Before GROUP BY + Projection)) - Sorting (Sorting for ORDER BY) - Expression ((Before ORDER BY + (Projection + Before ORDER BY))) - ReadFromSystemNumbers + Filter ((WHERE + (Projection + Before ORDER BY))) + Filter (HAVING) + Aggregating + Expression ((Before GROUP BY + Projection)) + Sorting (Sorting for ORDER BY) + Expression ((Before ORDER BY + (Projection + Before ORDER BY))) + ReadFromSystemNumbers -- execute 1 2 diff --git a/tests/queries/0_stateless/02497_trace_events_stress_long.sh b/tests/queries/0_stateless/02497_trace_events_stress_long.sh index c111ed40a29..dfd2f12b55b 100755 --- a/tests/queries/0_stateless/02497_trace_events_stress_long.sh +++ b/tests/queries/0_stateless/02497_trace_events_stress_long.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-parallel, no-tsan, no-asan, no-debug, no-s3-storage, no-fasttest, no-replicated-database +# Tags: long, no-parallel, no-tsan, no-asan, no-debug, no-object-storage, no-fasttest, no-replicated-database set -e diff --git a/tests/queries/0_stateless/02499_monotonicity_toUnixTimestamp64.sh b/tests/queries/0_stateless/02499_monotonicity_toUnixTimestamp64.sh index aa1f635f380..be87ea057ca 100755 --- a/tests/queries/0_stateless/02499_monotonicity_toUnixTimestamp64.sh +++ b/tests/queries/0_stateless/02499_monotonicity_toUnixTimestamp64.sh @@ -8,7 +8,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -q "create table t(ts DateTime64) engine=MergeTree order by ts as select * from numbers_mt(1e6);" +$CLICKHOUSE_CLIENT --optimize_trivial_insert_select 1 -q "create table t(ts DateTime64) engine=MergeTree order by ts as select * from numbers_mt(1e6);" max_block_size=8192 diff --git a/tests/queries/0_stateless/02503_cache_on_write_with_small_segment_size.sh b/tests/queries/0_stateless/02503_cache_on_write_with_small_segment_size.sh index 4f3fd0e54f6..5aeab4c746e 100755 --- a/tests/queries/0_stateless/02503_cache_on_write_with_small_segment_size.sh +++ b/tests/queries/0_stateless/02503_cache_on_write_with_small_segment_size.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest, no-s3-storage, no-random-settings +# Tags: no-parallel, no-fasttest, no-object-storage, no-random-settings CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL=none diff --git a/tests/queries/0_stateless/02504_regexp_dictionary_ua_parser.sh b/tests/queries/0_stateless/02504_regexp_dictionary_ua_parser.sh index d3a8743b880..e389cf410e8 100755 --- a/tests/queries/0_stateless/02504_regexp_dictionary_ua_parser.sh +++ b/tests/queries/0_stateless/02504_regexp_dictionary_ua_parser.sh @@ -1,18 +1,15 @@ #!/usr/bin/env bash - -# Tags: no-fasttest, no-parallel +# Tags: no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') +mkdir -p ${USER_FILES_PATH:?}/${CLICKHOUSE_DATABASE} -mkdir -p $user_files_path/test_02504 - -cp $CURDIR/data_ua_parser/os.yaml ${user_files_path}/test_02504/ -cp $CURDIR/data_ua_parser/browser.yaml ${user_files_path}/test_02504/ -cp $CURDIR/data_ua_parser/device.yaml ${user_files_path}/test_02504/ +cp $CURDIR/data_ua_parser/os.yaml ${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}/ +cp $CURDIR/data_ua_parser/browser.yaml ${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}/ +cp $CURDIR/data_ua_parser/device.yaml ${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}/ $CLICKHOUSE_CLIENT -n --query=" drop dictionary if exists regexp_os; @@ -29,7 +26,7 @@ create dictionary regexp_os os_v4_replacement String default '0' ) PRIMARY KEY(regex) -SOURCE(YAMLRegExpTree(PATH '${user_files_path}/test_02504/os.yaml')) +SOURCE(YAMLRegExpTree(PATH '${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}/os.yaml')) LIFETIME(0) LAYOUT(regexp_tree); @@ -41,7 +38,7 @@ create dictionary regexp_browser v2_replacement String default '0' ) PRIMARY KEY(regex) -SOURCE(YAMLRegExpTree(PATH '${user_files_path}/test_02504/browser.yaml')) +SOURCE(YAMLRegExpTree(PATH '${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}/browser.yaml')) LIFETIME(0) LAYOUT(regexp_tree); @@ -53,7 +50,7 @@ create dictionary regexp_device model_replacement String ) PRIMARY KEY(regex) -SOURCE(YAMLRegExpTree(PATH '${user_files_path}/test_02504/device.yaml')) +SOURCE(YAMLRegExpTree(PATH '${USER_FILES_PATH}/${CLICKHOUSE_DATABASE}/device.yaml')) LIFETIME(0) LAYOUT(regexp_tree); @@ -84,4 +81,4 @@ drop dictionary if exists regexp_device; drop table if exists user_agents; " -rm -rf "$user_files_path/test_02504" +rm -rf ${USER_FILES_PATH:?}/${CLICKHOUSE_DATABASE} diff --git a/tests/queries/0_stateless/02504_regexp_dictionary_yaml_source.sh b/tests/queries/0_stateless/02504_regexp_dictionary_yaml_source.sh index 7211372f2f7..68a87a14320 100755 --- a/tests/queries/0_stateless/02504_regexp_dictionary_yaml_source.sh +++ b/tests/queries/0_stateless/02504_regexp_dictionary_yaml_source.sh @@ -1,13 +1,10 @@ #!/usr/bin/env bash - # Tags: use-vectorscan, no-fasttest, no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - mkdir -p $USER_FILES_PATH/test_02504 yaml=$USER_FILES_PATH/test_02504/test.yaml diff --git a/tests/queries/0_stateless/02521_aggregation_by_partitions.sql b/tests/queries/0_stateless/02521_aggregation_by_partitions.sql index a90e56a9822..b4d31e234d8 100644 --- a/tests/queries/0_stateless/02521_aggregation_by_partitions.sql +++ b/tests/queries/0_stateless/02521_aggregation_by_partitions.sql @@ -1,4 +1,4 @@ --- Tags: long, no-s3-storage +-- Tags: long, no-object-storage SET merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0; @@ -6,6 +6,7 @@ set max_threads = 16; set allow_aggregate_partitions_independently = 1; set force_aggregate_partitions_independently = 1; set optimize_use_projections = 0; +set optimize_trivial_insert_select = 1; set allow_prefetched_read_pool_for_remote_filesystem = 0; set allow_prefetched_read_pool_for_local_filesystem = 0; diff --git a/tests/queries/0_stateless/02530_dictionaries_update_field.reference b/tests/queries/0_stateless/02530_dictionaries_update_field.reference index 4d5a7447a49..ada13443a10 100644 --- a/tests/queries/0_stateless/02530_dictionaries_update_field.reference +++ b/tests/queries/0_stateless/02530_dictionaries_update_field.reference @@ -1,102 +1,42 @@ flat --- { echoOn } - INSERT INTO table_for_update_field_dictionary VALUES (1, 'First', now()); -SELECT key, value FROM dict_flat ORDER BY key ASC; 1 First -INSERT INTO table_for_update_field_dictionary VALUES (2, 'Second', now()); -SELECT sleepEachRow(1) FROM numbers(10) SETTINGS function_sleep_max_microseconds_per_block = 10000000 FORMAT Null; -SELECT key, value FROM dict_flat ORDER BY key ASC; 1 First 2 Second -INSERT INTO table_for_update_field_dictionary VALUES (2, 'SecondUpdated', now()); -INSERT INTO table_for_update_field_dictionary VALUES (3, 'Third', now()); -SELECT sleepEachRow(1) FROM numbers(20) SETTINGS function_sleep_max_microseconds_per_block = 20000000 FORMAT Null; -SELECT key, value FROM dict_flat ORDER BY key ASC; 1 First 2 SecondUpdated 3 Third flat/custom --- { echoOn } - INSERT INTO table_for_update_field_dictionary VALUES (1, 'First', now()); -SELECT key, value FROM dict_flat_custom ORDER BY key ASC; 1 First -INSERT INTO table_for_update_field_dictionary VALUES (2, 'Second', now()); -SELECT sleepEachRow(1) FROM numbers(10) SETTINGS function_sleep_max_microseconds_per_block = 10000000 FORMAT Null; -SELECT key, value FROM dict_flat_custom ORDER BY key ASC; 1 First 2 Second -INSERT INTO table_for_update_field_dictionary VALUES (2, 'SecondUpdated', now()); -INSERT INTO table_for_update_field_dictionary VALUES (3, 'Third', now()); -SELECT sleepEachRow(1) FROM numbers(20) SETTINGS function_sleep_max_microseconds_per_block = 20000000 FORMAT Null; -SELECT key, value FROM dict_flat_custom ORDER BY key ASC; 1 First 2 SecondUpdated 3 Third hashed --- { echoOn } - INSERT INTO table_for_update_field_dictionary VALUES (1, 'First', now()); -SELECT key, value FROM dict_hashed ORDER BY key ASC; 1 First -INSERT INTO table_for_update_field_dictionary VALUES (2, 'Second', now()); -SELECT sleepEachRow(1) FROM numbers(10) SETTINGS function_sleep_max_microseconds_per_block = 10000000 FORMAT Null; -SELECT key, value FROM dict_hashed ORDER BY key ASC; 1 First 2 Second -INSERT INTO table_for_update_field_dictionary VALUES (2, 'SecondUpdated', now()); -INSERT INTO table_for_update_field_dictionary VALUES (3, 'Third', now()); -SELECT sleepEachRow(1) FROM numbers(20) SETTINGS function_sleep_max_microseconds_per_block = 20000000 FORMAT Null; -SELECT key, value FROM dict_hashed ORDER BY key ASC; 1 First 2 SecondUpdated 3 Third hashed/custom --- { echoOn } - INSERT INTO table_for_update_field_dictionary VALUES (1, 'First', now()); -SELECT key, value FROM dict_hashed_custom ORDER BY key ASC; 1 First -INSERT INTO table_for_update_field_dictionary VALUES (2, 'Second', now()); -SELECT sleepEachRow(1) FROM numbers(10) SETTINGS function_sleep_max_microseconds_per_block = 10000000 FORMAT Null; -SELECT key, value FROM dict_hashed_custom ORDER BY key ASC; 1 First 2 Second -INSERT INTO table_for_update_field_dictionary VALUES (2, 'SecondUpdated', now()); -INSERT INTO table_for_update_field_dictionary VALUES (3, 'Third', now()); -SELECT sleepEachRow(1) FROM numbers(20) SETTINGS function_sleep_max_microseconds_per_block = 20000000 FORMAT Null; -SELECT key, value FROM dict_hashed_custom ORDER BY key ASC; 1 First 2 SecondUpdated 3 Third complex_key_hashed --- { echoOn } - INSERT INTO table_for_update_field_dictionary VALUES (1, 'First', now()); -SELECT key, value FROM dict_complex_key_hashed ORDER BY key ASC; 1 First -INSERT INTO table_for_update_field_dictionary VALUES (2, 'Second', now()); -SELECT sleepEachRow(1) FROM numbers(10) SETTINGS function_sleep_max_microseconds_per_block = 10000000 FORMAT Null; -SELECT key, value FROM dict_complex_key_hashed ORDER BY key ASC; 1 First 2 Second -INSERT INTO table_for_update_field_dictionary VALUES (2, 'SecondUpdated', now()); -INSERT INTO table_for_update_field_dictionary VALUES (3, 'Third', now()); -SELECT sleepEachRow(1) FROM numbers(20) SETTINGS function_sleep_max_microseconds_per_block = 20000000 FORMAT Null; -SELECT key, value FROM dict_complex_key_hashed ORDER BY key ASC; 1 First 2 SecondUpdated 3 Third complex_key_hashed/custom --- { echoOn } - INSERT INTO table_for_update_field_dictionary VALUES (1, 'First', now()); -SELECT key, value FROM dict_complex_key_hashed_custom ORDER BY key ASC; 1 First -INSERT INTO table_for_update_field_dictionary VALUES (2, 'Second', now()); -SELECT sleepEachRow(1) FROM numbers(10) SETTINGS function_sleep_max_microseconds_per_block = 10000000 FORMAT Null; -SELECT key, value FROM dict_complex_key_hashed_custom ORDER BY key ASC; 1 First 2 Second -INSERT INTO table_for_update_field_dictionary VALUES (2, 'SecondUpdated', now()); -INSERT INTO table_for_update_field_dictionary VALUES (3, 'Third', now()); -SELECT sleepEachRow(1) FROM numbers(20) SETTINGS function_sleep_max_microseconds_per_block = 20000000 FORMAT Null; -SELECT key, value FROM dict_complex_key_hashed_custom ORDER BY key ASC; 1 First 2 SecondUpdated 3 Third diff --git a/tests/queries/0_stateless/02530_dictionaries_update_field.sh b/tests/queries/0_stateless/02530_dictionaries_update_field.sh index 9ced78a1196..a8e5804fba4 100755 --- a/tests/queries/0_stateless/02530_dictionaries_update_field.sh +++ b/tests/queries/0_stateless/02530_dictionaries_update_field.sh @@ -35,7 +35,7 @@ for layout in "${layouts[@]}"; do echo "$layout" fi - $CLICKHOUSE_CLIENT -nm -q " + $CLICKHOUSE_CLIENT --multiquery " TRUNCATE TABLE table_for_update_field_dictionary; CREATE DICTIONARY $dictionary_name @@ -49,24 +49,31 @@ for layout in "${layouts[@]}"; do LAYOUT($layout()) LIFETIME(1); - -- { echoOn } - INSERT INTO table_for_update_field_dictionary VALUES (1, 'First', now()); - SELECT key, value FROM $dictionary_name ORDER BY key ASC; + INSERT INTO table_for_update_field_dictionary VALUES (1, 'First', now());" - INSERT INTO table_for_update_field_dictionary VALUES (2, 'Second', now()); - SELECT sleepEachRow(1) FROM numbers(10) SETTINGS function_sleep_max_microseconds_per_block = 10000000 FORMAT Null; + while true + do + $CLICKHOUSE_CLIENT --query "SELECT key, value FROM $dictionary_name ORDER BY key ASC" | grep -A10 -B10 'First' && break; + sleep .1; + done - SELECT key, value FROM $dictionary_name ORDER BY key ASC; + $CLICKHOUSE_CLIENT --query "INSERT INTO table_for_update_field_dictionary VALUES (2, 'Second', now());" - INSERT INTO table_for_update_field_dictionary VALUES (2, 'SecondUpdated', now()); - INSERT INTO table_for_update_field_dictionary VALUES (3, 'Third', now()); - SELECT sleepEachRow(1) FROM numbers(20) SETTINGS function_sleep_max_microseconds_per_block = 20000000 FORMAT Null; + while true + do + $CLICKHOUSE_CLIENT --query "SELECT key, value FROM $dictionary_name ORDER BY key ASC" | grep -A10 -B10 'Second' && break; + sleep .1; + done - SELECT key, value FROM $dictionary_name ORDER BY key ASC; - -- { echoOff } + $CLICKHOUSE_CLIENT --query "INSERT INTO table_for_update_field_dictionary VALUES (2, 'SecondUpdated', now()), (3, 'Third', now())" - DROP DICTIONARY $dictionary_name; - " + while true + do + $CLICKHOUSE_CLIENT --query "SELECT key, value FROM $dictionary_name ORDER BY key ASC" | grep -A10 -B10 'SecondUpdated' && break; + sleep .1; + done + + $CLICKHOUSE_CLIENT --query "DROP DICTIONARY $dictionary_name" done done diff --git a/tests/queries/0_stateless/02532_send_logs_level_test.reference b/tests/queries/0_stateless/02532_send_logs_level_test.reference index 7e51b888d9c..72f4ea06184 100644 --- a/tests/queries/0_stateless/02532_send_logs_level_test.reference +++ b/tests/queries/0_stateless/02532_send_logs_level_test.reference @@ -1,3 +1,4 @@ + MergeTreeReadPoolBase: Will use min_marks_per_task=24 MergeTreeMarksLoader: Loading marks from path data.cmrk3 MergeTreeRangeReader: First reader returned: num_rows: 1, columns: 1, total_rows_per_granule: 1, no filter, column[0]: Int32(size = 1), requested columns: key MergeTreeRangeReader: read() returned num_rows: 1, columns: 1, total_rows_per_granule: 1, no filter, column[0]: Int32(size = 1), sample block key diff --git a/tests/queries/0_stateless/02532_send_logs_level_test.sh b/tests/queries/0_stateless/02532_send_logs_level_test.sh index 4afc6d4496b..506ac2331f2 100755 --- a/tests/queries/0_stateless/02532_send_logs_level_test.sh +++ b/tests/queries/0_stateless/02532_send_logs_level_test.sh @@ -1,7 +1,8 @@ #!/usr/bin/env bash -# Tags: no-s3-storage, no-debug -# - no-s3-storage - S3 has additional logging +# Tags: no-object-storage, no-debug, no-random-merge-tree-settings +# - no-object-storage - S3 has additional logging # - no-debug - debug builds also has additional logging +# - no-random-merge-tree-settings - changes content of log messages CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02535_analyzer_group_by_use_nulls.reference b/tests/queries/0_stateless/02535_analyzer_group_by_use_nulls.reference index 63610604ddd..858fbe98838 100644 --- a/tests/queries/0_stateless/02535_analyzer_group_by_use_nulls.reference +++ b/tests/queries/0_stateless/02535_analyzer_group_by_use_nulls.reference @@ -264,3 +264,44 @@ SETTINGS group_by_use_nulls = 1, max_bytes_before_external_sort=10; 9 \N 9 \N 0 20 \N 1 25 +CREATE TABLE test +ENGINE = ReplacingMergeTree +PRIMARY KEY id +AS SELECT number AS id FROM numbers(100); +SELECT id +FROM test +GROUP BY id + WITH CUBE +HAVING id IN ( + SELECT id + FROM test +) +FORMAT `NUll` +SETTINGS allow_experimental_analyzer = 1, group_by_use_nulls = true; +SELECT id +FROM test +FINAL +GROUP BY id + WITH CUBE +HAVING id IN ( + SELECT DISTINCT id + FROM test + FINAL +) +FORMAT `NUll` +SETTINGS allow_experimental_analyzer = 1, group_by_use_nulls = true; +SELECT id +FROM test +FINAL +GROUP BY + GROUPING SETS ((id)) +ORDER BY + id IN ( + SELECT DISTINCT id + FROM test + FINAL + LIMIT 4 + ) ASC +LIMIT 256 BY id +FORMAT `NUll` +SETTINGS allow_experimental_analyzer = 1, group_by_use_nulls=true; diff --git a/tests/queries/0_stateless/02535_analyzer_group_by_use_nulls.sql b/tests/queries/0_stateless/02535_analyzer_group_by_use_nulls.sql index a4d4f2f8bc9..4ae5df9629a 100644 --- a/tests/queries/0_stateless/02535_analyzer_group_by_use_nulls.sql +++ b/tests/queries/0_stateless/02535_analyzer_group_by_use_nulls.sql @@ -83,3 +83,48 @@ GROUP BY ) ORDER BY 1, tuple(val) SETTINGS group_by_use_nulls = 1, max_bytes_before_external_sort=10; + +CREATE TABLE test +ENGINE = ReplacingMergeTree +PRIMARY KEY id +AS SELECT number AS id FROM numbers(100); + +SELECT id +FROM test +GROUP BY id + WITH CUBE +HAVING id IN ( + SELECT id + FROM test +) +FORMAT `NUll` +SETTINGS allow_experimental_analyzer = 1, group_by_use_nulls = true; + +SELECT id +FROM test +FINAL +GROUP BY id + WITH CUBE +HAVING id IN ( + SELECT DISTINCT id + FROM test + FINAL +) +FORMAT `NUll` +SETTINGS allow_experimental_analyzer = 1, group_by_use_nulls = true; + +SELECT id +FROM test +FINAL +GROUP BY + GROUPING SETS ((id)) +ORDER BY + id IN ( + SELECT DISTINCT id + FROM test + FINAL + LIMIT 4 + ) ASC +LIMIT 256 BY id +FORMAT `NUll` +SETTINGS allow_experimental_analyzer = 1, group_by_use_nulls=true; diff --git a/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key.reference b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key.reference deleted file mode 100644 index 8d0f56ba185..00000000000 --- a/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key.reference +++ /dev/null @@ -1,173 +0,0 @@ -query='SELECT * FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key)' with custom_key='sipHash64(x)' -filter_type='default' max_replicas=1 prefer_localhost_replica=0 -Hello -filter_type='default' max_replicas=2 prefer_localhost_replica=0 -Hello -filter_type='default' max_replicas=3 prefer_localhost_replica=0 -Hello -filter_type='range' max_replicas=1 prefer_localhost_replica=0 -Hello -filter_type='range' max_replicas=2 prefer_localhost_replica=0 -Hello -filter_type='range' max_replicas=3 prefer_localhost_replica=0 -Hello -filter_type='default' max_replicas=1 prefer_localhost_replica=1 -Hello -filter_type='default' max_replicas=2 prefer_localhost_replica=1 -Hello -filter_type='default' max_replicas=3 prefer_localhost_replica=1 -Hello -filter_type='range' max_replicas=1 prefer_localhost_replica=1 -Hello -filter_type='range' max_replicas=2 prefer_localhost_replica=1 -Hello -filter_type='range' max_replicas=3 prefer_localhost_replica=1 -Hello -query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key) GROUP BY y ORDER BY y' with custom_key='y' -filter_type='default' max_replicas=1 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=2 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=3 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=1 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=2 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=3 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=1 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=2 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=3 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=1 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=2 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=3 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key) GROUP BY y ORDER BY y' with custom_key='cityHash64(y)' -filter_type='default' max_replicas=1 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=2 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=3 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=1 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=2 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=3 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=1 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=2 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=3 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=1 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=2 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=3 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key) GROUP BY y ORDER BY y' with custom_key='cityHash64(y) + 1' -filter_type='default' max_replicas=1 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=2 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=3 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=1 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=2 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=3 prefer_localhost_replica=0 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=1 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=2 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='default' max_replicas=3 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=1 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=2 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -filter_type='range' max_replicas=3 prefer_localhost_replica=1 -0 334 -1 333 -2 333 -1 diff --git a/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key.sh b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key.sh deleted file mode 100755 index dccb680be42..00000000000 --- a/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key.sh +++ /dev/null @@ -1,46 +0,0 @@ -#!/usr/bin/env bash -# Tags: no-parallel, long - -CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# shellcheck source=../shell_config.sh -. "$CURDIR"/../shell_config.sh - -function run_with_custom_key { - echo "query='$1' with custom_key='$2'" - for prefer_localhost_replica in 0 1; do - for filter_type in 'default' 'range'; do - for max_replicas in {1..3}; do - echo "filter_type='$filter_type' max_replicas=$max_replicas prefer_localhost_replica=$prefer_localhost_replica" - query="$1 SETTINGS max_parallel_replicas=$max_replicas\ - , parallel_replicas_custom_key='$2'\ - , parallel_replicas_custom_key_filter_type='$filter_type'\ - , prefer_localhost_replica=$prefer_localhost_replica" - $CLICKHOUSE_CLIENT --query="$query" - done - done - done -} - -$CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS 02535_custom_key"; - -$CLICKHOUSE_CLIENT --query="CREATE TABLE 02535_custom_key (x String) ENGINE = MergeTree ORDER BY x"; -$CLICKHOUSE_CLIENT --query="INSERT INTO 02535_custom_key VALUES ('Hello')"; - -run_with_custom_key "SELECT * FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key)" "sipHash64(x)" - -$CLICKHOUSE_CLIENT --query="DROP TABLE 02535_custom_key" - -$CLICKHOUSE_CLIENT --query="CREATE TABLE 02535_custom_key (x String, y UInt32) ENGINE = MergeTree ORDER BY cityHash64(x)" -$CLICKHOUSE_CLIENT --query="INSERT INTO 02535_custom_key SELECT toString(number), number % 3 FROM numbers(1000)" - -function run_count_with_custom_key { - run_with_custom_key "SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key) GROUP BY y ORDER BY y" "$1" -} - -run_count_with_custom_key "y" -run_count_with_custom_key "cityHash64(y)" -run_count_with_custom_key "cityHash64(y) + 1" - -$CLICKHOUSE_CLIENT --query="SELECT count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key) as t1 JOIN 02535_custom_key USING y" --allow_repeated_settings --parallel_replicas_custom_key="y" --send_logs_level="trace" 2>&1 | grep -Fac "JOINs are not supported with" - -$CLICKHOUSE_CLIENT --query="DROP TABLE 02535_custom_key" diff --git a/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_mt.reference b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_mt.reference new file mode 100644 index 00000000000..1bb07f0d916 --- /dev/null +++ b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_mt.reference @@ -0,0 +1,177 @@ +query='SELECT * FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_mt)' with custom_key='sipHash64(x)' +filter_type='default' max_replicas=1 +Hello +filter_type='default' max_replicas=2 +Hello +filter_type='default' max_replicas=3 +Hello +filter_type='range' max_replicas=1 +Hello +filter_type='range' max_replicas=2 +Hello +filter_type='range' max_replicas=3 +Hello +query='SELECT * FROM 02535_custom_key_mt' with custom_key='sipHash64(x)' +filter_type='default' max_replicas=1 +Hello +filter_type='default' max_replicas=2 +Hello +filter_type='default' max_replicas=3 +Hello +filter_type='range' max_replicas=1 +Hello +filter_type='range' max_replicas=2 +Hello +filter_type='range' max_replicas=3 +Hello +query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_mt) GROUP BY y ORDER BY y' with custom_key='y' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_mt) GROUP BY y ORDER BY y' with custom_key='cityHash64(y)' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_mt) GROUP BY y ORDER BY y' with custom_key='cityHash64(y) + 1' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM 02535_custom_key_mt GROUP BY y ORDER BY y' with custom_key='y' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM 02535_custom_key_mt GROUP BY y ORDER BY y' with custom_key='cityHash64(y)' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM 02535_custom_key_mt GROUP BY y ORDER BY y' with custom_key='cityHash64(y) + 1' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +1 diff --git a/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_mt.sh b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_mt.sh new file mode 100755 index 00000000000..fad43ea9070 --- /dev/null +++ b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_mt.sh @@ -0,0 +1,54 @@ +#!/usr/bin/env bash +# Tags: no-parallel, long + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +function run_with_custom_key { + echo "query='$1' with custom_key='$2'" + for filter_type in 'default' 'range'; do + for max_replicas in {1..3}; do + echo "filter_type='$filter_type' max_replicas=$max_replicas" + query="$1 SETTINGS max_parallel_replicas=$max_replicas\ +, parallel_replicas_custom_key='$2'\ +, parallel_replicas_custom_key_filter_type='$filter_type'\ +, parallel_replicas_for_non_replicated_merge_tree=1 \ +, cluster_for_parallel_replicas='test_cluster_one_shard_three_replicas_localhost'" + $CLICKHOUSE_CLIENT --query="$query" + done + done +} + +$CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS 02535_custom_key_mt"; + +$CLICKHOUSE_CLIENT --query="CREATE TABLE 02535_custom_key_mt (x String) ENGINE = MergeTree ORDER BY x"; +$CLICKHOUSE_CLIENT --query="INSERT INTO 02535_custom_key_mt VALUES ('Hello')"; + +run_with_custom_key "SELECT * FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_mt)" "sipHash64(x)" +run_with_custom_key "SELECT * FROM 02535_custom_key_mt" "sipHash64(x)" + +$CLICKHOUSE_CLIENT --query="DROP TABLE 02535_custom_key_mt" + +$CLICKHOUSE_CLIENT --query="CREATE TABLE 02535_custom_key_mt (x String, y UInt32) ENGINE = MergeTree ORDER BY cityHash64(x)" +$CLICKHOUSE_CLIENT --query="INSERT INTO 02535_custom_key_mt SELECT toString(number), number % 3 FROM numbers(1000)" + +function run_count_with_custom_key_distributed { + run_with_custom_key "SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_mt) GROUP BY y ORDER BY y" "$1" +} + +run_count_with_custom_key_distributed "y" +run_count_with_custom_key_distributed "cityHash64(y)" +run_count_with_custom_key_distributed "cityHash64(y) + 1" + +function run_count_with_custom_key_merge_tree { + run_with_custom_key "SELECT y, count() FROM 02535_custom_key_mt GROUP BY y ORDER BY y" "$1" +} + +run_count_with_custom_key_merge_tree "y" +run_count_with_custom_key_merge_tree "cityHash64(y)" +run_count_with_custom_key_merge_tree "cityHash64(y) + 1" + +$CLICKHOUSE_CLIENT --query="SELECT count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_mt) as t1 JOIN 02535_custom_key_mt USING y" --allow_repeated_settings --parallel_replicas_custom_key="y" --send_logs_level="trace" 2>&1 | grep -Fac "JOINs are not supported with" + +$CLICKHOUSE_CLIENT --query="DROP TABLE 02535_custom_key_mt" diff --git a/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_rmt.reference b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_rmt.reference new file mode 100644 index 00000000000..c6526b506d3 --- /dev/null +++ b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_rmt.reference @@ -0,0 +1,177 @@ +query='SELECT * FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_rmt)' with custom_key='sipHash64(x)' +filter_type='default' max_replicas=1 +Hello +filter_type='default' max_replicas=2 +Hello +filter_type='default' max_replicas=3 +Hello +filter_type='range' max_replicas=1 +Hello +filter_type='range' max_replicas=2 +Hello +filter_type='range' max_replicas=3 +Hello +query='SELECT * FROM 02535_custom_key_rmt' with custom_key='sipHash64(x)' +filter_type='default' max_replicas=1 +Hello +filter_type='default' max_replicas=2 +Hello +filter_type='default' max_replicas=3 +Hello +filter_type='range' max_replicas=1 +Hello +filter_type='range' max_replicas=2 +Hello +filter_type='range' max_replicas=3 +Hello +query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_rmt_hash) GROUP BY y ORDER BY y' with custom_key='y' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_rmt_hash) GROUP BY y ORDER BY y' with custom_key='cityHash64(y)' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_rmt_hash) GROUP BY y ORDER BY y' with custom_key='cityHash64(y) + 1' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM 02535_custom_key_rmt_hash GROUP BY y ORDER BY y' with custom_key='y' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM 02535_custom_key_rmt_hash GROUP BY y ORDER BY y' with custom_key='cityHash64(y)' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +query='SELECT y, count() FROM 02535_custom_key_rmt_hash GROUP BY y ORDER BY y' with custom_key='cityHash64(y) + 1' +filter_type='default' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='default' max_replicas=3 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=1 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=2 +0 334 +1 333 +2 333 +filter_type='range' max_replicas=3 +0 334 +1 333 +2 333 +1 diff --git a/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_rmt.sh b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_rmt.sh new file mode 100755 index 00000000000..6350f5027f9 --- /dev/null +++ b/tests/queries/0_stateless/02535_max_parallel_replicas_custom_key_rmt.sh @@ -0,0 +1,54 @@ +#!/usr/bin/env bash +# Tags: no-parallel, long + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +function run_with_custom_key { + echo "query='$1' with custom_key='$2'" + for filter_type in 'default' 'range'; do + for max_replicas in {1..3}; do + echo "filter_type='$filter_type' max_replicas=$max_replicas" + query="$1 SETTINGS max_parallel_replicas=$max_replicas\ +, parallel_replicas_custom_key='$2'\ +, parallel_replicas_custom_key_filter_type='$filter_type'\ +, parallel_replicas_for_non_replicated_merge_tree=1 \ +, cluster_for_parallel_replicas='test_cluster_one_shard_three_replicas_localhost'" + $CLICKHOUSE_CLIENT --query="$query" + done + done +} + +$CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS 02535_custom_key_rmt"; + +$CLICKHOUSE_CLIENT --query="CREATE TABLE 02535_custom_key_rmt (x String) ENGINE = ReplicatedMergeTree('/clickhouse/tables/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/test_02535', 'r1') ORDER BY x"; +$CLICKHOUSE_CLIENT --query="INSERT INTO 02535_custom_key_rmt VALUES ('Hello')"; + +run_with_custom_key "SELECT * FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_rmt)" "sipHash64(x)" +run_with_custom_key "SELECT * FROM 02535_custom_key_rmt" "sipHash64(x)" + +$CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS 02535_custom_key_rmt_hash"; + +$CLICKHOUSE_CLIENT --query="CREATE TABLE 02535_custom_key_rmt_hash (x String, y UInt32) ENGINE = ReplicatedMergeTree('/clickhouse/tables/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/test_02535_hash', 'r1') ORDER BY cityHash64(x)" +$CLICKHOUSE_CLIENT --query="INSERT INTO 02535_custom_key_rmt_hash SELECT toString(number), number % 3 FROM numbers(1000)" + +function run_count_with_custom_key { + run_with_custom_key "SELECT y, count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_rmt_hash) GROUP BY y ORDER BY y" "$1" +} + +run_count_with_custom_key "y" +run_count_with_custom_key "cityHash64(y)" +run_count_with_custom_key "cityHash64(y) + 1" + +function run_count_with_custom_key_merge_tree { + run_with_custom_key "SELECT y, count() FROM 02535_custom_key_rmt_hash GROUP BY y ORDER BY y" "$1" +} + +run_count_with_custom_key_merge_tree "y" +run_count_with_custom_key_merge_tree "cityHash64(y)" +run_count_with_custom_key_merge_tree "cityHash64(y) + 1" + +$CLICKHOUSE_CLIENT --query="SELECT count() FROM cluster(test_cluster_one_shard_three_replicas_localhost, currentDatabase(), 02535_custom_key_rmt_hash) as t1 JOIN 02535_custom_key_rmt_hash USING y" --allow_repeated_settings --parallel_replicas_custom_key="y" --send_logs_level="trace" 2>&1 | grep -Fac "JOINs are not supported with" + +$CLICKHOUSE_CLIENT --query="DROP TABLE 02535_custom_key_rmt_hash" diff --git a/tests/queries/0_stateless/02541_tuple_element_with_null.sql b/tests/queries/0_stateless/02541_tuple_element_with_null.sql index d2062b60d49..e1581ce3755 100644 --- a/tests/queries/0_stateless/02541_tuple_element_with_null.sql +++ b/tests/queries/0_stateless/02541_tuple_element_with_null.sql @@ -9,7 +9,7 @@ SETTINGS index_granularity = 8192; INSERT INTO test_tuple_element VALUES (tuple(1,2)), (tuple(NULL, 3)); -SELECT +SELECT tupleElement(tuple, 'k1', 0) fine_k1_with_0, tupleElement(tuple, 'k1', NULL) k1_with_null, tupleElement(tuple, 'k2', 0) k2_with_0, diff --git a/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.reference b/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.reference index 70bcd7f255b..9bb0c022752 100644 --- a/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.reference +++ b/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.reference @@ -29,16 +29,20 @@ WHERE type_1 = \'all\' ExpressionTransform × 2 (Filter) FilterTransform × 2 - (Aggregating) - ExpressionTransform × 2 - AggregatingTransform × 2 - Copy 1 → 2 - (Expression) - ExpressionTransform - (Expression) - ExpressionTransform - (ReadFromMergeTree) - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + (Filter) + FilterTransform × 2 + (Filter) + FilterTransform × 2 + (Aggregating) + ExpressionTransform × 2 + AggregatingTransform × 2 + Copy 1 → 2 + (Expression) + ExpressionTransform + (Expression) + ExpressionTransform + (ReadFromMergeTree) + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 (Expression) ExpressionTransform × 2 (Filter) @@ -64,10 +68,14 @@ ExpressionTransform × 2 ExpressionTransform × 2 AggregatingTransform × 2 Copy 1 → 2 - (Expression) - ExpressionTransform - (ReadFromMergeTree) - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + (Filter) + FilterTransform + (Filter) + FilterTransform + (Expression) + ExpressionTransform + (ReadFromMergeTree) + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 (Expression) ExpressionTransform × 2 (Aggregating) diff --git a/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.sql b/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.sql index 0891f1aa8a2..f926b9037d2 100644 --- a/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.sql +++ b/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage DROP TABLE IF EXISTS test_grouping_sets_predicate; diff --git a/tests/queries/0_stateless/02560_vertical_merge_memory_usage.sql b/tests/queries/0_stateless/02560_vertical_merge_memory_usage.sql index 785fb10f70b..361305bac6d 100644 --- a/tests/queries/0_stateless/02560_vertical_merge_memory_usage.sql +++ b/tests/queries/0_stateless/02560_vertical_merge_memory_usage.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage drop table if exists tvm; create table tvm (c0 UInt64, c1 UInt64, c2 UInt64, c3 UInt64, c4 UInt64, c5 UInt64, c6 UInt64, c7 UInt64, c8 UInt64, c9 UInt64, c10 UInt64, c11 UInt64, c12 UInt64, c13 UInt64, c14 UInt64, c15 UInt64, c16 UInt64, c17 UInt64, c18 UInt64, c19 UInt64, c20 UInt64, c21 UInt64, c22 UInt64, c23 UInt64, c24 UInt64, c25 UInt64, c26 UInt64, c27 UInt64, c28 UInt64, c29 UInt64, c30 UInt64, c31 UInt64, c32 UInt64, c33 UInt64, c34 UInt64, c35 UInt64, c36 UInt64, c37 UInt64, c38 UInt64, c39 UInt64, c40 UInt64, c41 UInt64, c42 UInt64, c43 UInt64, c44 UInt64, c45 UInt64, c46 UInt64, c47 UInt64, c48 UInt64, c49 UInt64, c50 UInt64, c51 UInt64, c52 UInt64, c53 UInt64, c54 UInt64, c55 UInt64, c56 UInt64, c57 UInt64, c58 UInt64, c59 UInt64, c60 UInt64, c61 UInt64, c62 UInt64, c63 UInt64, c64 UInt64, c65 UInt64, c66 UInt64, c67 UInt64, c68 UInt64, c69 UInt64, c70 UInt64, c71 UInt64, c72 UInt64, c73 UInt64, c74 UInt64, c75 UInt64, c76 UInt64, c77 UInt64, c78 UInt64, c79 UInt64, c80 UInt64, c81 UInt64, c82 UInt64, c83 UInt64, c84 UInt64, c85 UInt64, c86 UInt64, c87 UInt64, c88 UInt64, c89 UInt64, c90 UInt64, c91 UInt64, c92 UInt64, c93 UInt64, c94 UInt64, c95 UInt64, c96 UInt64, c97 UInt64, c98 UInt64, c99 UInt64, c100 UInt64, c101 UInt64, c102 UInt64, c103 UInt64, c104 UInt64, c105 UInt64, c106 UInt64, c107 UInt64, c108 UInt64, c109 UInt64, c110 UInt64, c111 UInt64, c112 UInt64, c113 UInt64, c114 UInt64, c115 UInt64, c116 UInt64, c117 UInt64, c118 UInt64, c119 UInt64, c120 UInt64, c121 UInt64, c122 UInt64, c123 UInt64, c124 UInt64, c125 UInt64, c126 UInt64, c127 UInt64, c128 UInt64, c129 UInt64, c130 UInt64, c131 UInt64, c132 UInt64, c133 UInt64, c134 UInt64, c135 UInt64, c136 UInt64, c137 UInt64, c138 UInt64, c139 UInt64, c140 UInt64, c141 UInt64, c142 UInt64, c143 UInt64, c144 UInt64, c145 UInt64, c146 UInt64, c147 UInt64, c148 UInt64, c149 UInt64, c150 UInt64, c151 UInt64, c152 UInt64, c153 UInt64, c154 UInt64, c155 UInt64, c156 UInt64, c157 UInt64, c158 UInt64, c159 UInt64, c160 UInt64, c161 UInt64, c162 UInt64, c163 UInt64, c164 UInt64, c165 UInt64, c166 UInt64, c167 UInt64, c168 UInt64, c169 UInt64, c170 UInt64, c171 UInt64, c172 UInt64, c173 UInt64, c174 UInt64, c175 UInt64, c176 UInt64, c177 UInt64, c178 UInt64, c179 UInt64, c180 UInt64, c181 UInt64, c182 UInt64, c183 UInt64, c184 UInt64, c185 UInt64, c186 UInt64, c187 UInt64, c188 UInt64, c189 UInt64, c190 UInt64, c191 UInt64, c192 UInt64, c193 UInt64, c194 UInt64, c195 UInt64, c196 UInt64, c197 UInt64, c198 UInt64, c199 UInt64, c200 UInt64, c201 UInt64, c202 UInt64, c203 UInt64, c204 UInt64, c205 UInt64, c206 UInt64, c207 UInt64, c208 UInt64, c209 UInt64, c210 UInt64, c211 UInt64, c212 UInt64, c213 UInt64, c214 UInt64, c215 UInt64, c216 UInt64, c217 UInt64, c218 UInt64, c219 UInt64, c220 UInt64, c221 UInt64, c222 UInt64, c223 UInt64, c224 UInt64, c225 UInt64, c226 UInt64, c227 UInt64, c228 UInt64, c229 UInt64, c230 UInt64, c231 UInt64, c232 UInt64, c233 UInt64, c234 UInt64, c235 UInt64, c236 UInt64, c237 UInt64, c238 UInt64, c239 UInt64, c240 UInt64, c241 UInt64, c242 UInt64, c243 UInt64, c244 UInt64, c245 UInt64, c246 UInt64, c247 UInt64, c248 UInt64, c249 UInt64, c250 UInt64, c251 UInt64, c252 UInt64, c253 UInt64, c254 UInt64, c255 UInt64, c256 UInt64, c257 UInt64, c258 UInt64, c259 UInt64, c260 UInt64, c261 UInt64, c262 UInt64, c263 UInt64, c264 UInt64, c265 UInt64, c266 UInt64, c267 UInt64, c268 UInt64, c269 UInt64, c270 UInt64, c271 UInt64, c272 UInt64, c273 UInt64, c274 UInt64, c275 UInt64, c276 UInt64, c277 UInt64, c278 UInt64, c279 UInt64, c280 UInt64, c281 UInt64, c282 UInt64, c283 UInt64, c284 UInt64, c285 UInt64, c286 UInt64, c287 UInt64, c288 UInt64, c289 UInt64, c290 UInt64, c291 UInt64, c292 UInt64, c293 UInt64, c294 UInt64, c295 UInt64, c296 UInt64, c297 UInt64, c298 UInt64, c299 UInt64) engine = MergeTree order by tuple() settings min_rows_for_wide_part = 10, min_bytes_for_wide_part=0, vertical_merge_algorithm_min_rows_to_activate=1; diff --git a/tests/queries/0_stateless/02560_window_ntile.reference b/tests/queries/0_stateless/02560_window_ntile.reference index d877b2034cb..5f01832d075 100644 --- a/tests/queries/0_stateless/02560_window_ntile.reference +++ b/tests/queries/0_stateless/02560_window_ntile.reference @@ -213,6 +213,8 @@ select a, b, ntile('2') over (partition by a order by b) from(select intDiv(numb select a, b, ntile(0) over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } select a, b, ntile(-2) over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } select a, b, ntile(b + 1) over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } +select a, b, ntile() over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +select a, b, ntile(3, 2) over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -- Bad window type select a, b, ntile(2) over (partition by a) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } select a, b, ntile(2) over (partition by a order by b rows between 4 preceding and unbounded following) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } diff --git a/tests/queries/0_stateless/02560_window_ntile.sql b/tests/queries/0_stateless/02560_window_ntile.sql index 44e9b865052..d0e4d557e58 100644 --- a/tests/queries/0_stateless/02560_window_ntile.sql +++ b/tests/queries/0_stateless/02560_window_ntile.sql @@ -16,6 +16,8 @@ select a, b, ntile('2') over (partition by a order by b) from(select intDiv(numb select a, b, ntile(0) over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } select a, b, ntile(-2) over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } select a, b, ntile(b + 1) over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } +select a, b, ntile() over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +select a, b, ntile(3, 2) over (partition by a order by b) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -- Bad window type select a, b, ntile(2) over (partition by a) from(select intDiv(number,10) as a, number%10 as b from numbers(20)); -- { serverError BAD_ARGUMENTS } diff --git a/tests/queries/0_stateless/02581_share_big_sets_between_multiple_mutations_tasks_long.reference b/tests/queries/0_stateless/02581_share_big_sets_between_multiple_mutations_tasks_long.reference index 3a92fcf283d..d3e7e113ffe 100644 --- a/tests/queries/0_stateless/02581_share_big_sets_between_multiple_mutations_tasks_long.reference +++ b/tests/queries/0_stateless/02581_share_big_sets_between_multiple_mutations_tasks_long.reference @@ -3,7 +3,11 @@ all_1_1_0 all_2_2_0 all_3_3_0 all_4_4_0 +40000 5000 all_1_1_0_9 5000 all_2_2_0_9 5000 all_3_3_0_9 5000 all_4_4_0_9 +mutation_version has_parts_for_which_set_was_built has_parts_that_shared_set +8 1 1 +9 1 1 diff --git a/tests/queries/0_stateless/02581_share_big_sets_between_multiple_mutations_tasks_long.sql b/tests/queries/0_stateless/02581_share_big_sets_between_multiple_mutations_tasks_long.sql index 741d0177971..6b0677a80ae 100644 --- a/tests/queries/0_stateless/02581_share_big_sets_between_multiple_mutations_tasks_long.sql +++ b/tests/queries/0_stateless/02581_share_big_sets_between_multiple_mutations_tasks_long.sql @@ -18,12 +18,35 @@ SELECT name FROM system.parts WHERE database=currentDatabase() AND table = '0258 -- Start multiple mutations simultaneously SYSTEM STOP MERGES 02581_trips; -ALTER TABLE 02581_trips UPDATE description='5' WHERE id IN (SELECT (number*10 + 5)::UInt32 FROM numbers(200000000)) SETTINGS mutations_sync=0; -ALTER TABLE 02581_trips UPDATE description='6' WHERE id IN (SELECT (number*10 + 6)::UInt32 FROM numbers(200000000)) SETTINGS mutations_sync=0; -ALTER TABLE 02581_trips DELETE WHERE id IN (SELECT (number*10 + 7)::UInt32 FROM numbers(200000000)) SETTINGS mutations_sync=0; -ALTER TABLE 02581_trips UPDATE description='8' WHERE id IN (SELECT (number*10 + 8)::UInt32 FROM numbers(200000000)) SETTINGS mutations_sync=0; +ALTER TABLE 02581_trips UPDATE description='5' WHERE id IN (SELECT (number*10 + 5)::UInt32 FROM numbers(10000000)) SETTINGS mutations_sync=0; +ALTER TABLE 02581_trips UPDATE description='6' WHERE id IN (SELECT (number*10 + 6)::UInt32 FROM numbers(10000000)) SETTINGS mutations_sync=0; +ALTER TABLE 02581_trips DELETE WHERE id IN (SELECT (number*10 + 7)::UInt32 FROM numbers(10000000)) SETTINGS mutations_sync=0; +ALTER TABLE 02581_trips UPDATE description='8' WHERE id IN (SELECT (number*10 + 8)::UInt32 FROM numbers(10000000)) SETTINGS mutations_sync=0; SYSTEM START MERGES 02581_trips; -DELETE FROM 02581_trips WHERE id IN (SELECT (number*10 + 9)::UInt32 FROM numbers(200000000)); -SELECT count(), _part from 02581_trips WHERE description = '' GROUP BY _part ORDER BY _part; + +-- Wait for mutations to finish +SELECT count() FROM 02581_trips SETTINGS select_sequential_consistency = 1; + +DELETE FROM 02581_trips WHERE id IN (SELECT (number*10 + 9)::UInt32 FROM numbers(10000000)) SETTINGS lightweight_deletes_sync = 2; +SELECT count(), _part from 02581_trips WHERE description = '' GROUP BY _part ORDER BY _part SETTINGS select_sequential_consistency=1; + +SYSTEM FLUSH LOGS; +-- Check that in every mutation there were parts that built sets (log messages like 'Created Set with 10000000 entries from 10000000 rows in 0.388989187 sec.' ) +-- and parts that shared sets (log messages like 'Got set from cache in 0.388930505 sec.' ) +WITH ( + SELECT uuid + FROM system.tables + WHERE (database = currentDatabase()) AND (name = '02581_trips') + ) AS table_uuid +SELECT + CAST(splitByChar('_', query_id)[5], 'UInt64') AS mutation_version, -- '5521485f-8a40-4aba-87a2-00342c369563::all_3_3_0_6' + sum(message LIKE 'Created Set with % entries%') >= 1 AS has_parts_for_which_set_was_built, + sum(message LIKE 'Got set from cache%') >= 1 AS has_parts_that_shared_set +FROM system.text_log +WHERE + query_id LIKE concat(CAST(table_uuid, 'String'), '::all\\_%') + AND (event_date >= yesterday()) + AND (message LIKE 'Created Set with % entries%' OR message LIKE 'Got set from cache%') +GROUP BY mutation_version ORDER BY mutation_version FORMAT TSVWithNames; DROP TABLE 02581_trips; diff --git a/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks_long.reference b/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks_long.reference index 3a7410d925f..eecd768eb5e 100644 --- a/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks_long.reference +++ b/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks_long.reference @@ -10,3 +10,11 @@ all_4_4_0 20000 16000 12000 +mutation_version has_parts_for_which_set_was_built has_parts_that_shared_set +5 1 1 +6 1 1 +7 1 1 +8 1 1 +9 1 1 +10 1 1 +11 1 1 diff --git a/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks_long.sql b/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks_long.sql index b7314c8fa47..091a9c8171d 100644 --- a/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks_long.sql +++ b/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks_long.sql @@ -18,42 +18,63 @@ SELECT count() from 02581_trips WHERE description = ''; SELECT name FROM system.parts WHERE database=currentDatabase() AND table = '02581_trips' AND active ORDER BY name; -- Run mutation with `id` a 'IN big subquery' -ALTER TABLE 02581_trips UPDATE description='a' WHERE id IN (SELECT (number*10)::UInt32 FROM numbers(200000000)) SETTINGS mutations_sync=2; +ALTER TABLE 02581_trips UPDATE description='a' WHERE id IN (SELECT (number*10)::UInt32 FROM numbers(10000000)) SETTINGS mutations_sync=2; SELECT count() from 02581_trips WHERE description = ''; -ALTER TABLE 02581_trips UPDATE description='a' WHERE id IN (SELECT (number*10 + 1)::UInt32 FROM numbers(200000000)) SETTINGS mutations_sync=2, max_rows_in_set=1000; +ALTER TABLE 02581_trips UPDATE description='a' WHERE id IN (SELECT (number*10 + 1)::UInt32 FROM numbers(10000000)) SETTINGS mutations_sync=2, max_rows_in_set=1000; SELECT count() from 02581_trips WHERE description = ''; -- Run mutation with func(`id`) IN big subquery -ALTER TABLE 02581_trips UPDATE description='b' WHERE id::UInt64 IN (SELECT (number*10 + 2)::UInt32 FROM numbers(200000000)) SETTINGS mutations_sync=2; +ALTER TABLE 02581_trips UPDATE description='b' WHERE id::UInt64 IN (SELECT (number*10 + 2)::UInt32 FROM numbers(10000000)) SETTINGS mutations_sync=2; SELECT count() from 02581_trips WHERE description = ''; -- Run mutation with non-PK `id2` IN big subquery -ALTER TABLE 02581_trips UPDATE description='c' WHERE id2 IN (SELECT (number*10 + 3)::UInt32 FROM numbers(200000000)) SETTINGS mutations_sync=2; +--SELECT count(), _part FROM 02581_trips WHERE id2 IN (SELECT (number*10 + 3)::UInt32 FROM numbers(10000000)) GROUP BY _part ORDER BY _part; +--EXPLAIN SELECT (), _part FROM 02581_trips WHERE id2 IN (SELECT (number*10 + 3)::UInt32 FROM numbers(10000000)); +ALTER TABLE 02581_trips UPDATE description='c' WHERE id2 IN (SELECT (number*10 + 3)::UInt32 FROM numbers(10000000)) SETTINGS mutations_sync=2; SELECT count() from 02581_trips WHERE description = ''; -- Run mutation with PK and non-PK IN big subquery ALTER TABLE 02581_trips UPDATE description='c' WHERE - (id IN (SELECT (number*10 + 4)::UInt32 FROM numbers(200000000))) OR - (id2 IN (SELECT (number*10 + 4)::UInt32 FROM numbers(200000000))) + (id IN (SELECT (number*10 + 4)::UInt32 FROM numbers(10000000))) OR + (id2 IN (SELECT (number*10 + 4)::UInt32 FROM numbers(10000000))) SETTINGS mutations_sync=2; SELECT count() from 02581_trips WHERE description = ''; -- Run mutation with PK and non-PK IN big subquery ALTER TABLE 02581_trips UPDATE description='c' WHERE - (id::UInt64 IN (SELECT (number*10 + 5)::UInt32 FROM numbers(200000000))) OR - (id2::UInt64 IN (SELECT (number*10 + 5)::UInt32 FROM numbers(200000000))) + (id::UInt64 IN (SELECT (number*10 + 5)::UInt32 FROM numbers(10000000))) OR + (id2::UInt64 IN (SELECT (number*10 + 5)::UInt32 FROM numbers(10000000))) SETTINGS mutations_sync=2; SELECT count() from 02581_trips WHERE description = ''; -- Run mutation with PK and non-PK IN big subquery ALTER TABLE 02581_trips UPDATE description='c' WHERE - (id::UInt32 IN (SELECT (number*10 + 6)::UInt32 FROM numbers(200000000))) OR - ((id2+1)::String IN (SELECT (number*10 + 6)::UInt32 FROM numbers(200000000))) + (id::UInt32 IN (SELECT (number*10 + 6)::UInt32 FROM numbers(10000000))) OR + ((id2+1)::String IN (SELECT (number*10 + 6)::UInt32 FROM numbers(10000000))) SETTINGS mutations_sync=2; SELECT count() from 02581_trips WHERE description = ''; +SYSTEM FLUSH LOGS; +-- Check that in every mutation there were parts that built sets (log messages like 'Created Set with 10000000 entries from 10000000 rows in 0.388989187 sec.' ) +-- and parts that shared sets (log messages like 'Got set from cache in 0.388930505 sec.' ) +WITH ( + SELECT uuid + FROM system.tables + WHERE (database = currentDatabase()) AND (name = '02581_trips') + ) AS table_uuid +SELECT + CAST(splitByChar('_', query_id)[5], 'UInt64') AS mutation_version, -- '5521485f-8a40-4aba-87a2-00342c369563::all_3_3_0_6' + sum(message LIKE 'Created Set with % entries%') >= 1 AS has_parts_for_which_set_was_built, + sum(message LIKE 'Got set from cache%') >= 1 AS has_parts_that_shared_set +FROM system.text_log +WHERE + query_id LIKE concat(CAST(table_uuid, 'String'), '::all\\_%') + AND (event_date >= yesterday()) + AND (message LIKE 'Created Set with % entries%' OR message LIKE 'Got set from cache%') +GROUP BY mutation_version ORDER BY mutation_version FORMAT TSVWithNames; + DROP TABLE 02581_trips; diff --git a/tests/queries/0_stateless/02582_async_reading_with_small_limit.sql b/tests/queries/0_stateless/02582_async_reading_with_small_limit.sql index cb6b1b6083e..406cab82183 100644 --- a/tests/queries/0_stateless/02582_async_reading_with_small_limit.sql +++ b/tests/queries/0_stateless/02582_async_reading_with_small_limit.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage +-- Tags: no-object-storage SET merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0; diff --git a/tests/queries/0_stateless/02661_read_from_archive.lib b/tests/queries/0_stateless/02661_read_from_archive.lib index 908b6bd38d2..56f1a0f163c 100644 --- a/tests/queries/0_stateless/02661_read_from_archive.lib +++ b/tests/queries/0_stateless/02661_read_from_archive.lib @@ -6,10 +6,10 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh function read_archive_file() { - $CLICKHOUSE_LOCAL --query "SELECT * FROM file('${user_files_path}/$1') ORDER BY 1, 2" - $CLICKHOUSE_CLIENT --query "SELECT * FROM file('${user_files_path}/$1') ORDER BY 1, 2" - $CLICKHOUSE_CLIENT --query "DESC file('${user_files_path}/$1')" - $CLICKHOUSE_CLIENT --query "CREATE TABLE 02661_archive_table Engine=File('CSV', '${user_files_path}/$1')" + $CLICKHOUSE_LOCAL --query "SELECT * FROM file('${USER_FILES_PATH}/$1') ORDER BY 1, 2" + $CLICKHOUSE_CLIENT --query "SELECT * FROM file('${USER_FILES_PATH}/$1') ORDER BY 1, 2" + $CLICKHOUSE_CLIENT --query "DESC file('${USER_FILES_PATH}/$1')" + $CLICKHOUSE_CLIENT --query "CREATE TABLE 02661_archive_table Engine=File('CSV', '${USER_FILES_PATH}/$1')" $CLICKHOUSE_CLIENT --query "SELECT * FROM 02661_archive_table ORDER BY 1, 2" $CLICKHOUSE_CLIENT --query "DROP TABLE 02661_archive_table" } @@ -20,16 +20,14 @@ function run_archive_test() { extension_without_dot=$(echo $1 | sed -e 's/\.//g') FILE_PREFIX="02661_read_from_archive_${CLICKHOUSE_DATABASE}_$extension_without_dot" - user_files_path=$(clickhouse-client --query "select _path, _file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep -o "/[^[:space:]]*nonexist.txt" | awk '{gsub("/nonexist.txt","",$1); print $1}') - touch ${FILE_PREFIX}_data0.csv echo -e "1,2\n3,4" > ${FILE_PREFIX}_data1.csv echo -e "5,6\n7,8" > ${FILE_PREFIX}_data2.csv echo -e "9,10\n11,12" > ${FILE_PREFIX}_data3.csv - eval "$2 ${user_files_path}/${FILE_PREFIX}_archive1.$1 ${FILE_PREFIX}_data0.csv ${FILE_PREFIX}_data1.csv ${FILE_PREFIX}_data2.csv > /dev/null" - eval "$2 ${user_files_path}/${FILE_PREFIX}_archive2.$1 ${FILE_PREFIX}_data1.csv ${FILE_PREFIX}_data3.csv > /dev/null" - eval "$2 ${user_files_path}/${FILE_PREFIX}_archive3.$1 ${FILE_PREFIX}_data2.csv ${FILE_PREFIX}_data3.csv > /dev/null" + eval "$2 ${USER_FILES_PATH}/${FILE_PREFIX}_archive1.$1 ${FILE_PREFIX}_data0.csv ${FILE_PREFIX}_data1.csv ${FILE_PREFIX}_data2.csv > /dev/null" + eval "$2 ${USER_FILES_PATH}/${FILE_PREFIX}_archive2.$1 ${FILE_PREFIX}_data1.csv ${FILE_PREFIX}_data3.csv > /dev/null" + eval "$2 ${USER_FILES_PATH}/${FILE_PREFIX}_archive3.$1 ${FILE_PREFIX}_data2.csv ${FILE_PREFIX}_data3.csv > /dev/null" echo "archive1 data1.csv" read_archive_file "${FILE_PREFIX}_archive1.$1 :: ${FILE_PREFIX}_data1.csv" @@ -44,10 +42,10 @@ function run_archive_test() { echo "archive* {2..3}.csv" read_archive_file "${FILE_PREFIX}_archive*.$1 :: ${FILE_PREFIX}_data{2..3}.csv" - $CLICKHOUSE_LOCAL --query "SELECT * FROM file('${user_files_path}/${FILE_PREFIX}_archive1.$1::nonexistent.csv')" 2>&1 | grep -q "CANNOT_EXTRACT_TABLE_STRUCTURE" && echo "OK" || echo "FAIL" - $CLICKHOUSE_LOCAL --query "SELECT * FROM file('${user_files_path}/${FILE_PREFIX}_archive3.$1::{2..3}.csv')" 2>&1 | grep -q "CANNOT_EXTRACT_TABLE_STRUCTURE" && echo "OK" || echo "FAIL" + $CLICKHOUSE_LOCAL --query "SELECT * FROM file('${USER_FILES_PATH}/${FILE_PREFIX}_archive1.$1::nonexistent.csv')" 2>&1 | grep -q "CANNOT_EXTRACT_TABLE_STRUCTURE" && echo "OK" || echo "FAIL" + $CLICKHOUSE_LOCAL --query "SELECT * FROM file('${USER_FILES_PATH}/${FILE_PREFIX}_archive3.$1::{2..3}.csv')" 2>&1 | grep -q "CANNOT_EXTRACT_TABLE_STRUCTURE" && echo "OK" || echo "FAIL" - rm ${user_files_path}/${FILE_PREFIX}_archive{1..3}.$1 + rm ${USER_FILES_PATH}/${FILE_PREFIX}_archive{1..3}.$1 rm ${FILE_PREFIX}_data{0..3}.csv } diff --git a/tests/queries/0_stateless/02680_mysql_ast_logical_err.sql b/tests/queries/0_stateless/02680_mysql_ast_logical_err.sql index bde91df83ca..78ce1b68b0d 100644 --- a/tests/queries/0_stateless/02680_mysql_ast_logical_err.sql +++ b/tests/queries/0_stateless/02680_mysql_ast_logical_err.sql @@ -1,4 +1,10 @@ CREATE TABLE foo (key UInt32, a String, b Int64, c String) ENGINE = TinyLog; -SELECT count() FROM mysql(mysql('127.0.0.1:9004', currentDatabase(), 'foo', 'default', ''), '127.0.0.1:9004', currentDatabase(), 'foo', '', ''); -- { serverError UNKNOWN_FUNCTION } -SELECT count() FROM mysql(mysql('127.0.0.1:9004', currentDatabase(), 'foo', 'default', '', SETTINGS connection_pool_size = 1), '127.0.0.1:9004', currentDatabase(), 'foo', '', ''); -- { serverError UNKNOWN_FUNCTION, UNSUPPORTED_METHOD } +SELECT count() FROM mysql( + mysql('127.0.0.1:9004', currentDatabase(), 'foo', 'default', ''), + '127.0.0.1:9004', currentDatabase(), 'foo', '', '', + SETTINGS connect_timeout = 100, connection_wait_timeout = 100, read_write_timeout = 300); -- { serverError UNKNOWN_FUNCTION } +SELECT count() FROM mysql( + mysql('127.0.0.1:9004', currentDatabase(), 'foo', 'default', '', SETTINGS connection_pool_size = 1), + '127.0.0.1:9004', currentDatabase(), 'foo', '', '', + SETTINGS connect_timeout = 100, connection_wait_timeout = 100, read_write_timeout = 300); -- { serverError UNKNOWN_FUNCTION, UNSUPPORTED_METHOD } diff --git a/tests/queries/0_stateless/02703_keeper_map_concurrent_create_drop.sh b/tests/queries/0_stateless/02703_keeper_map_concurrent_create_drop.sh index 17d1fa92377..3e629ece33f 100755 --- a/tests/queries/0_stateless/02703_keeper_map_concurrent_create_drop.sh +++ b/tests/queries/0_stateless/02703_keeper_map_concurrent_create_drop.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-ordinary-database, zookeeper, no-fasttest, no-parallel +# Tags: no-ordinary-database, zookeeper, no-fasttest CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -37,7 +37,7 @@ function create_drop_loop() export -f create_drop_loop; THREADS=10 -TIMEOUT=30 +TIMEOUT=20 for i in `seq $THREADS` do diff --git a/tests/queries/0_stateless/02703_max_local_read_bandwidth.sh b/tests/queries/0_stateless/02703_max_local_read_bandwidth.sh index c78cd202f1b..6f43c1ae869 100755 --- a/tests/queries/0_stateless/02703_max_local_read_bandwidth.sh +++ b/tests/queries/0_stateless/02703_max_local_read_bandwidth.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-s3-storage, no-random-settings, no-random-merge-tree-settings +# Tags: no-object-storage, no-random-settings, no-random-merge-tree-settings CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02703_max_local_write_bandwidth.sh b/tests/queries/0_stateless/02703_max_local_write_bandwidth.sh index 31cf6e9606e..4f6a300c5b3 100755 --- a/tests/queries/0_stateless/02703_max_local_write_bandwidth.sh +++ b/tests/queries/0_stateless/02703_max_local_write_bandwidth.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-s3-storage +# Tags: no-object-storage CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02704_max_backup_bandwidth.sh b/tests/queries/0_stateless/02704_max_backup_bandwidth.sh index 748bf856deb..8cb03a93a7a 100755 --- a/tests/queries/0_stateless/02704_max_backup_bandwidth.sh +++ b/tests/queries/0_stateless/02704_max_backup_bandwidth.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-s3-storage, no-random-settings, no-random-merge-tree-settings +# Tags: no-object-storage, no-random-settings, no-random-merge-tree-settings CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02722_database_filesystem.sh b/tests/queries/0_stateless/02722_database_filesystem.sh index 374dd246c96..2d0ff256c95 100755 --- a/tests/queries/0_stateless/02722_database_filesystem.sh +++ b/tests/queries/0_stateless/02722_database_filesystem.sh @@ -5,12 +5,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# see 01658_read_file_to_stringcolumn.sh -CLICKHOUSE_USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path, _file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - # Prepare data unique_name=${CLICKHOUSE_TEST_UNIQUE_NAME} -user_files_tmp_dir=${CLICKHOUSE_USER_FILES_PATH}/${unique_name} +user_files_tmp_dir=${USER_FILES_PATH}/${unique_name} mkdir -p ${user_files_tmp_dir}/tmp/ echo '"id","str","int","text"' > ${user_files_tmp_dir}/tmp.csv echo '1,"abc",123,"abacaba"' >> ${user_files_tmp_dir}/tmp.csv diff --git a/tests/queries/0_stateless/02724_decompress_filename_exception.sh b/tests/queries/0_stateless/02724_decompress_filename_exception.sh index e413910b934..8b5a2f23aa9 100755 --- a/tests/queries/0_stateless/02724_decompress_filename_exception.sh +++ b/tests/queries/0_stateless/02724_decompress_filename_exception.sh @@ -5,7 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') FILENAME="${USER_FILES_PATH}/corrupted_file.tsv.xx" echo 'corrupted file' > $FILENAME; diff --git a/tests/queries/0_stateless/02724_function_in_left_table_clause_asof_join.reference b/tests/queries/0_stateless/02724_function_in_left_table_clause_asof_join.reference index d00491fd7e5..6ed281c757a 100644 --- a/tests/queries/0_stateless/02724_function_in_left_table_clause_asof_join.reference +++ b/tests/queries/0_stateless/02724_function_in_left_table_clause_asof_join.reference @@ -1 +1,2 @@ 1 +1 diff --git a/tests/queries/0_stateless/02724_function_in_left_table_clause_asof_join.sql b/tests/queries/0_stateless/02724_function_in_left_table_clause_asof_join.sql index 13dfb5debe7..6aa70a379c1 100644 --- a/tests/queries/0_stateless/02724_function_in_left_table_clause_asof_join.sql +++ b/tests/queries/0_stateless/02724_function_in_left_table_clause_asof_join.sql @@ -6,3 +6,15 @@ ASOF LEFT JOIN ( select 1 as session_id, 4 as id ) as visitors ON visitors.session_id <= sessions.id AND arrayFirst(a -> a, arrayMap((a) -> a, sessions.arr)) = visitors.id +; + +select count(*) +from ( + select 1 as id, [1, 2, 3] as arr +) as sessions +ASOF LEFT JOIN ( + select 1 as session_id, 4 as id +) as visitors +ON visitors.session_id <= sessions.id AND arrayFirst(a -> a, arrayMap((a) -> a, sessions.arr)) = visitors.id +SETTINGS join_algorithm = 'full_sorting_merge' +; diff --git a/tests/queries/0_stateless/02724_limit_num_mutations.sh b/tests/queries/0_stateless/02724_limit_num_mutations.sh index a9d69b2ed48..60888db0e2e 100755 --- a/tests/queries/0_stateless/02724_limit_num_mutations.sh +++ b/tests/queries/0_stateless/02724_limit_num_mutations.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02724_show_indexes.reference b/tests/queries/0_stateless/02724_show_indexes.reference index e41f2521f5c..ac0461fc506 100644 --- a/tests/queries/0_stateless/02724_show_indexes.reference +++ b/tests/queries/0_stateless/02724_show_indexes.reference @@ -2,33 +2,33 @@ tbl 1 blf_idx 1 \N 0 \N \N \N BLOOM_FILTER YES d, b tbl 1 mm1_idx 1 \N 0 \N \N \N MINMAX YES a, c, d tbl 1 mm2_idx 1 \N 0 \N \N \N MINMAX YES c, d, e -tbl 1 PRIMARY 1 a A 0 \N \N \N PRIMARY YES -tbl 1 PRIMARY 2 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 1 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 2 a A 0 \N \N \N PRIMARY YES tbl 1 set_idx 1 \N 0 \N \N \N SET YES e tbl 1 blf_idx 1 \N 0 \N \N \N BLOOM_FILTER YES d, b tbl 1 mm1_idx 1 \N 0 \N \N \N MINMAX YES a, c, d tbl 1 mm2_idx 1 \N 0 \N \N \N MINMAX YES c, d, e -tbl 1 PRIMARY 1 a A 0 \N \N \N PRIMARY YES -tbl 1 PRIMARY 2 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 1 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 2 a A 0 \N \N \N PRIMARY YES tbl 1 set_idx 1 \N 0 \N \N \N SET YES e tbl 1 blf_idx 1 \N 0 \N \N \N BLOOM_FILTER YES d, b tbl 1 mm1_idx 1 \N 0 \N \N \N MINMAX YES a, c, d tbl 1 mm2_idx 1 \N 0 \N \N \N MINMAX YES c, d, e -tbl 1 PRIMARY 1 a A 0 \N \N \N PRIMARY YES -tbl 1 PRIMARY 2 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 1 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 2 a A 0 \N \N \N PRIMARY YES tbl 1 set_idx 1 \N 0 \N \N \N SET YES e tbl 1 blf_idx 1 \N 0 \N \N \N BLOOM_FILTER YES d, b tbl 1 mm1_idx 1 \N 0 \N \N \N MINMAX YES a, c, d tbl 1 mm2_idx 1 \N 0 \N \N \N MINMAX YES c, d, e -tbl 1 PRIMARY 1 a A 0 \N \N \N PRIMARY YES -tbl 1 PRIMARY 2 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 1 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 2 a A 0 \N \N \N PRIMARY YES tbl 1 set_idx 1 \N 0 \N \N \N SET YES e --- EXTENDED tbl 1 blf_idx 1 \N 0 \N \N \N BLOOM_FILTER YES d, b tbl 1 mm1_idx 1 \N 0 \N \N \N MINMAX YES a, c, d tbl 1 mm2_idx 1 \N 0 \N \N \N MINMAX YES c, d, e -tbl 1 PRIMARY 1 a A 0 \N \N \N PRIMARY YES -tbl 1 PRIMARY 2 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 1 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 2 a A 0 \N \N \N PRIMARY YES tbl 1 set_idx 1 \N 0 \N \N \N SET YES e --- WHERE --- Check with weird table names @@ -40,8 +40,8 @@ NULL 1 PRIMARY 1 c A 0 \N \N \N PRIMARY YES tbl 1 blf_idx 1 \N 0 \N \N \N BLOOM_FILTER YES d, b tbl 1 mm1_idx 1 \N 0 \N \N \N MINMAX YES a, c, d tbl 1 mm2_idx 1 \N 0 \N \N \N MINMAX YES c, d, e -tbl 1 PRIMARY 1 a A 0 \N \N \N PRIMARY YES -tbl 1 PRIMARY 2 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 1 c A 0 \N \N \N PRIMARY YES +tbl 1 PRIMARY 2 a A 0 \N \N \N PRIMARY YES tbl 1 set_idx 1 \N 0 \N \N \N SET YES e --- Equally named table in other database tbl 1 mmi_idx 1 \N 0 \N \N \N MINMAX YES b diff --git a/tests/queries/0_stateless/02725_agg_projection_resprect_PK.sql b/tests/queries/0_stateless/02725_agg_projection_resprect_PK.sql index a2355f78f4c..459ebc1bc22 100644 --- a/tests/queries/0_stateless/02725_agg_projection_resprect_PK.sql +++ b/tests/queries/0_stateless/02725_agg_projection_resprect_PK.sql @@ -20,6 +20,7 @@ CREATE TABLE t0 ) ENGINE = MergeTree ORDER BY (c1, c2) settings min_bytes_for_wide_part = 10485760, min_rows_for_wide_part = 0; +SET optimize_trivial_insert_select = 1; INSERT INTO t0 SELECT number, -number, diff --git a/tests/queries/0_stateless/02725_memory-for-merges.sql b/tests/queries/0_stateless/02725_memory-for-merges.sql index 1a8402dff4b..8e4d4f5b3e0 100644 --- a/tests/queries/0_stateless/02725_memory-for-merges.sql +++ b/tests/queries/0_stateless/02725_memory-for-merges.sql @@ -1,4 +1,4 @@ --- Tags: no-s3-storage, no-random-merge-tree-settings +-- Tags: no-object-storage, no-random-merge-tree-settings -- We allocate a lot of memory for buffers when reading or writing to S3 DROP TABLE IF EXISTS 02725_memory_for_merges SYNC; diff --git a/tests/queries/0_stateless/02726_async_insert_flush_queue.sql b/tests/queries/0_stateless/02726_async_insert_flush_queue.sql index 1ae24e4f3da..e393d96fc40 100644 --- a/tests/queries/0_stateless/02726_async_insert_flush_queue.sql +++ b/tests/queries/0_stateless/02726_async_insert_flush_queue.sql @@ -6,7 +6,8 @@ CREATE TABLE t_async_inserts_flush (a UInt64) ENGINE = Memory; SET async_insert = 1; SET wait_for_async_insert = 0; -SET async_insert_busy_timeout_min_ms = 1000000; +-- Disable adaptive timeout to prevent immediate push of the first message (if the queue last push was old) +SET async_insert_use_adaptive_busy_timeout=0; SET async_insert_busy_timeout_max_ms = 10000000; INSERT INTO t_async_inserts_flush VALUES (1) (2); diff --git a/tests/queries/0_stateless/02731_zero_objects_in_metadata.sh b/tests/queries/0_stateless/02731_zero_objects_in_metadata.sh index eef52002e36..78659b70129 100755 --- a/tests/queries/0_stateless/02731_zero_objects_in_metadata.sh +++ b/tests/queries/0_stateless/02731_zero_objects_in_metadata.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-s3-storage +# Tags: no-fasttest, no-object-storage CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02732_rename_after_processing.sh b/tests/queries/0_stateless/02732_rename_after_processing.sh index 9d44ff9fc34..c3f2274570e 100755 --- a/tests/queries/0_stateless/02732_rename_after_processing.sh +++ b/tests/queries/0_stateless/02732_rename_after_processing.sh @@ -4,12 +4,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# see 01658_read_file_to_stringcolumn.sh -CLICKHOUSE_USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path, _file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - # Prepare data unique_name=${CLICKHOUSE_TEST_UNIQUE_NAME} -tmp_dir=${CLICKHOUSE_USER_FILES_PATH}/${unique_name} +tmp_dir=${USER_FILES_PATH}/${unique_name} mkdir -p $tmp_dir rm -rf ${tmp_dir:?}/* diff --git a/tests/queries/0_stateless/02766_bitshift_with_const_arguments.sql b/tests/queries/0_stateless/02766_bitshift_with_const_arguments.sql index 6b2961f0555..91e8624057c 100644 --- a/tests/queries/0_stateless/02766_bitshift_with_const_arguments.sql +++ b/tests/queries/0_stateless/02766_bitshift_with_const_arguments.sql @@ -10,7 +10,7 @@ DROP TABLE IF EXISTS t1; CREATE TABLE t0 (vkey UInt32, pkey UInt32, c0 UInt32) engine = TinyLog; CREATE TABLE t1 (vkey UInt32) ENGINE = AggregatingMergeTree ORDER BY vkey; INSERT INTO t0 VALUES (15, 25000, 58); -SELECT ref_5.pkey AS c_2_c2392_6 FROM t0 AS ref_5 WHERE 'J[' < multiIf(ref_5.pkey IN ( SELECT 1 ), bitShiftLeft(multiIf(ref_5.c0 > NULL, '1', ')'), 40), NULL); +SELECT ref_5.pkey AS c_2_c2392_6 FROM t0 AS ref_5 WHERE 'J[' < multiIf(ref_5.pkey IN ( SELECT 1 ), bitShiftLeft(multiIf(ref_5.c0 > NULL, '1', ')'), 40), NULL); -- { serverError ARGUMENT_OUT_OF_BOUND } DROP TABLE t0; DROP TABLE t1; diff --git a/tests/queries/0_stateless/02771_multidirectory_globs_storage_file.sh b/tests/queries/0_stateless/02771_multidirectory_globs_storage_file.sh index 932837b83db..60df3ed2762 100755 --- a/tests/queries/0_stateless/02771_multidirectory_globs_storage_file.sh +++ b/tests/queries/0_stateless/02771_multidirectory_globs_storage_file.sh @@ -7,27 +7,22 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Data preparation. -# Now we can get the user_files_path by using the file function, we can also get it by this query: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') +rm -rf ${USER_FILES_PATH:?}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -rm -rf ${user_files_path:?}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ - -${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir{?/subdir?1/da,2/subdir2?/da}ta/non_existing.csv', CSV);" 2>&1 | grep -q "CANNOT_EXTRACT_TABLE_STRUCTURE" && echo 'OK' || echo 'FAIL' +${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir{?/subdir?1/da,2/subdir2?/da}ta/non_existing.csv', CSV);" 2>&1 | grep -q "CANNOT_EXTRACT_TABLE_STRUCTURE" && echo 'OK' || echo 'FAIL' # Create two files in different directories -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir1/subdir11/ -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir2/subdir22/ +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir1/subdir11/ +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir2/subdir22/ -echo 'This is file data1' > ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir1/subdir11/data1.csv -echo 'This is file data2' > ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir2/subdir22/data2.csv +echo 'This is file data1' > ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir1/subdir11/data1.csv +echo 'This is file data2' > ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir2/subdir22/data2.csv -${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir{?/subdir?1/da,2/subdir2?/da}ta1.csv', CSV);" -${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir{?/subdir?1/da,2/subdir2?/da}ta2.csv', CSV);" +${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir{?/subdir?1/da,2/subdir2?/da}ta1.csv', CSV);" +${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir{?/subdir?1/da,2/subdir2?/da}ta2.csv', CSV);" -${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir?/{subdir?1/data1,subdir2?/data2}.csv', CSV) WHERE _file == 'data1.csv';" -${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir?/{subdir?1/data1,subdir2?/data2}.csv', CSV) WHERE _file == 'data2.csv';" +${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir?/{subdir?1/data1,subdir2?/data2}.csv', CSV) WHERE _file == 'data1.csv';" +${CLICKHOUSE_CLIENT} --query "SELECT *, _file FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/dir?/{subdir?1/data1,subdir2?/data2}.csv', CSV) WHERE _file == 'data2.csv';" -rm -rf ${user_files_path:?}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} +rm -rf ${USER_FILES_PATH:?}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} diff --git a/tests/queries/0_stateless/02789_reading_from_s3_with_connection_pool.sh b/tests/queries/0_stateless/02789_reading_from_s3_with_connection_pool.sh index e4a1de9a2ec..751b2798243 100755 --- a/tests/queries/0_stateless/02789_reading_from_s3_with_connection_pool.sh +++ b/tests/queries/0_stateless/02789_reading_from_s3_with_connection_pool.sh @@ -18,10 +18,7 @@ query="SELECT a, b FROM test_s3" query_id=$(${CLICKHOUSE_CLIENT} --query "select queryID() from ($query) limit 1" 2>&1) ${CLICKHOUSE_CLIENT} --query "SYSTEM FLUSH LOGS" ${CLICKHOUSE_CLIENT} -nm --query " -WITH - ProfileEvents['DiskConnectionsReset'] AS reset, - ProfileEvents['DiskConnectionsPreserved'] AS preserved -SELECT preserved > reset +SELECT ProfileEvents['DiskConnectionsPreserved'] > 0 FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() diff --git a/tests/queries/0_stateless/02802_clickhouse_disks_s3_copy.sh b/tests/queries/0_stateless/02802_clickhouse_disks_s3_copy.sh index 2b9e5296a05..20b02bcba32 100755 --- a/tests/queries/0_stateless/02802_clickhouse_disks_s3_copy.sh +++ b/tests/queries/0_stateless/02802_clickhouse_disks_s3_copy.sh @@ -14,14 +14,11 @@ function run_test_for_disk() echo "$disk" - clickhouse-disks -C "$config" --disk "$disk" write --input "$config" $CLICKHOUSE_DATABASE/test - clickhouse-disks -C "$config" --log-level test --disk "$disk" copy $CLICKHOUSE_DATABASE/test $CLICKHOUSE_DATABASE/test.copy |& { + clickhouse-disks -C "$config" --disk "$disk" --query "write --path-from $config $CLICKHOUSE_DATABASE/test" + clickhouse-disks -C "$config" --log-level test --disk "$disk" --query "copy -r $CLICKHOUSE_DATABASE/test $CLICKHOUSE_DATABASE/test.copy" |& { grep -o -e "Single part upload has completed." -e "Single operation copy has completed." } - clickhouse-disks -C "$config" --disk "$disk" remove $CLICKHOUSE_DATABASE/test - # NOTE: this is due to "copy" does works like "cp -R from to/" instead of "cp from to" - clickhouse-disks -C "$config" --disk "$disk" remove $CLICKHOUSE_DATABASE/test.copy/test - clickhouse-disks -C "$config" --disk "$disk" remove $CLICKHOUSE_DATABASE/test.copy + clickhouse-disks -C "$config" --disk "$disk" --query "remove -r $CLICKHOUSE_DATABASE/test" } function run_test_copy_from_s3_to_s3(){ @@ -29,13 +26,12 @@ function run_test_copy_from_s3_to_s3(){ local disk_dest=$1 && shift echo "copy from $disk_src to $disk_dest" - clickhouse-disks -C "$config" --disk "$disk_src" write --input "$config" $CLICKHOUSE_DATABASE/test + clickhouse-disks -C "$config" --disk "$disk_src" --query "write --path-from $config $CLICKHOUSE_DATABASE/test" - clickhouse-disks -C "$config" --log-level test copy --disk-from "$disk_src" --disk-to "$disk_dest" $CLICKHOUSE_DATABASE/test $CLICKHOUSE_DATABASE/test.copy |& { + clickhouse-disks -C "$config" --log-level test --query "copy -r --disk-from $disk_src --disk-to $disk_dest $CLICKHOUSE_DATABASE/test $CLICKHOUSE_DATABASE/test.copy" |& { grep -o -e "Single part upload has completed." -e "Single operation copy has completed." } - clickhouse-disks -C "$config" --disk "$disk_dest" remove $CLICKHOUSE_DATABASE/test.copy/test - clickhouse-disks -C "$config" --disk "$disk_dest" remove $CLICKHOUSE_DATABASE/test.copy + clickhouse-disks -C "$config" --disk "$disk_dest" --query "remove -r $CLICKHOUSE_DATABASE/test.copy" } run_test_for_disk s3_plain_native_copy diff --git a/tests/queries/0_stateless/02805_distributed_queries_timeouts.sql b/tests/queries/0_stateless/02805_distributed_queries_timeouts.sql index 0b7337d1255..f6bccc99977 100644 --- a/tests/queries/0_stateless/02805_distributed_queries_timeouts.sql +++ b/tests/queries/0_stateless/02805_distributed_queries_timeouts.sql @@ -1,3 +1,4 @@ -select * from remote('127.2', view(select sleep(3) from system.one)) settings receive_timeout=1, async_socket_for_remote=0, use_hedged_requests=1 format Null; -select * from remote('127.2', view(select sleep(3) from system.one)) settings receive_timeout=1, async_socket_for_remote=1, use_hedged_requests=0 format Null; -select * from remote('127.2', view(select sleep(3) from system.one)) settings receive_timeout=1, async_socket_for_remote=0, use_hedged_requests=0 format Null; +create table dist as system.one engine=Distributed(test_shard_localhost, system, one); +select sleep(8) from dist settings function_sleep_max_microseconds_per_block=8e9, prefer_localhost_replica=0, receive_timeout=7, async_socket_for_remote=0, use_hedged_requests=1 format Null; +select sleep(8) from dist settings function_sleep_max_microseconds_per_block=8e9, prefer_localhost_replica=0, receive_timeout=7, async_socket_for_remote=1, use_hedged_requests=0 format Null; +select sleep(8) from dist settings function_sleep_max_microseconds_per_block=8e9, prefer_localhost_replica=0, receive_timeout=7, async_socket_for_remote=0, use_hedged_requests=0 format Null; diff --git a/tests/queries/0_stateless/02806_system_parts_columns_modification_time.sql.j2 b/tests/queries/0_stateless/02806_system_parts_columns_modification_time.sql.j2 index eee236ff681..1ca5cc0bb7e 100644 --- a/tests/queries/0_stateless/02806_system_parts_columns_modification_time.sql.j2 +++ b/tests/queries/0_stateless/02806_system_parts_columns_modification_time.sql.j2 @@ -1,5 +1,5 @@ --- Tags: no-s3-storage --- Tag: no-s3-storage because S3 updates metadata for the virtual link file on metadata disk (see CreateHardlinkOperation::execute() for details) +-- Tags: no-object-storage +-- Tag: no-object-storage because S3 updates metadata for the virtual link file on metadata disk (see CreateHardlinkOperation::execute() for details) set mutations_sync=1; diff --git a/tests/queries/0_stateless/02808_filesystem_cache_drop_query.sh b/tests/queries/0_stateless/02808_filesystem_cache_drop_query.sh index b44f9e50513..8a4a2e906b0 100755 --- a/tests/queries/0_stateless/02808_filesystem_cache_drop_query.sh +++ b/tests/queries/0_stateless/02808_filesystem_cache_drop_query.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel, no-s3-storage, no-random-settings +# Tags: no-fasttest, no-parallel, no-object-storage, no-random-settings # set -x diff --git a/tests/queries/0_stateless/02815_no_throw_in_simple_queries.sh b/tests/queries/0_stateless/02815_no_throw_in_simple_queries.sh index 50f5e565088..68c55f9b66a 100755 --- a/tests/queries/0_stateless/02815_no_throw_in_simple_queries.sh +++ b/tests/queries/0_stateless/02815_no_throw_in_simple_queries.sh @@ -45,9 +45,4 @@ expect eof EOF } -run "$CLICKHOUSE_LOCAL --disable_suggestion" -# Suggestions are off because the suggestion feature initializes itself by reading all available function -# names from "system.functions". Getting the value for field "is_obsolete" occasionally throws (e.g. for -# certain dictionary functions when dictionaries are not set up yet). Exceptions are properly handled, but -# they exist for a short time. This, in combination with CLICKHOUSE_TERMINATE_ON_ANY_EXCEPTION, terminates -# clickhouse-local and clickhouse-client when run in interactive mode *with* suggestions. +run "$CLICKHOUSE_LOCAL" diff --git a/tests/queries/0_stateless/02832_alter_max_sessions_for_user.sh b/tests/queries/0_stateless/02832_alter_max_sessions_for_user.sh index a3b0d17f1be..87fbffdb1e6 100755 --- a/tests/queries/0_stateless/02832_alter_max_sessions_for_user.sh +++ b/tests/queries/0_stateless/02832_alter_max_sessions_for_user.sh @@ -1,10 +1,12 @@ #!/usr/bin/env bash +# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh SESSION_ID_PREFIX="02832_alter_max_sessions_session_$$" +QUERY_ID_PREFIX="02832_alter_max_sessions_query_$$" PROFILE="02832_alter_max_sessions_profile_$$" USER="02832_alter_max_sessions_user_$$" USER2="02832_alter_max_sessions_user_two_$$" @@ -15,6 +17,26 @@ ${CLICKHOUSE_CLIENT} -q $"DROP PROFILE IF EXISTS ${PROFILE}" ${CLICKHOUSE_CLIENT} -q $"CREATE SETTINGS PROFILE ${PROFILE}" ${CLICKHOUSE_CLIENT} -q $"CREATE USER '${USER}' SETTINGS PROFILE '${PROFILE}'" +function run_sessions_set() +{ + local sessions_count="$1" + local session_check="$2" + for ((i = 1 ; i <= ${sessions_count} ; i++)); do + local session_id="${SESSION_ID_PREFIX}_${i}" + local query_id="${QUERY_ID_PREFIX}_${i}" + # Write only expected error text + # More than alter_sessions_count queries will not start. + ${CLICKHOUSE_CURL} -sS -X POST "${CLICKHOUSE_URL}&user=${USER}&query_id=${query_id}&session_id=${session_id}&session_check=${session_check}&session_timeout=600&function_sleep_max_microseconds_per_block=120000000" --data-binary "SELECT sleep(120)" | grep -o -m 1 'USER_SESSION_LIMIT_EXCEEDED' & + done + + for ((i = 1 ; i <= ${sessions_count} ; i++)); do + local query_id="${QUERY_ID_PREFIX}_${i}" + $CLICKHOUSE_CLIENT --query "KILL QUERY WHERE query_id='$query_id' SYNC" >/dev/null + done + + wait +} + function test_alter_profile() { local max_session_count="$1" @@ -24,23 +46,13 @@ function test_alter_profile() ${CLICKHOUSE_CLIENT} -q $"ALTER SETTINGS PROFILE ${PROFILE} SETTINGS max_sessions_for_user = ${max_session_count}" # Create sessions with $max_session_count restriction - for ((i = 1 ; i <= ${max_session_count} ; i++)); do - local session_id="${SESSION_ID_PREFIX}_${i}" - # Skip output from this query - ${CLICKHOUSE_CURL} -sS -X POST "${CLICKHOUSE_URL}&user=${USER}&session_id=${session_id}&session_check=0" --data-binary "SELECT 1" > /dev/null - done + run_sessions_set $max_session_count 0 # Update restriction to $alter_sessions_count ${CLICKHOUSE_CLIENT} -q $"ALTER SETTINGS PROFILE ${PROFILE} SETTINGS max_sessions_for_user = ${alter_sessions_count}" # Simultaneous sessions should use max settings from profile ($alter_sessions_count) - for ((i = 1 ; i <= ${max_session_count} ; i++)); do - local session_id="${SESSION_ID_PREFIX}_${i}" - # ignore select 1, we need only errors - ${CLICKHOUSE_CURL} -sS -X POST "${CLICKHOUSE_URL}&user=${USER}&session_id=${session_id}&session_check=1" --data-binary "select sleep(0.3)" | grep -o -m 1 'USER_SESSION_LIMIT_EXCEEDED' & - done - - wait + run_sessions_set $max_session_count 1 } test_alter_profile 1 1 diff --git a/tests/queries/0_stateless/02833_multiprewhere_extra_column.sql b/tests/queries/0_stateless/02833_multiprewhere_extra_column.sql index 3a751294cba..da2f050cf38 100644 --- a/tests/queries/0_stateless/02833_multiprewhere_extra_column.sql +++ b/tests/queries/0_stateless/02833_multiprewhere_extra_column.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel, no-random-settings, no-random-merge-tree-settings, no-s3-storage +-- Tags: no-parallel, no-random-settings, no-random-merge-tree-settings, no-object-storage drop table if exists t_multi_prewhere; drop row policy if exists policy_02834 on t_multi_prewhere; diff --git a/tests/queries/0_stateless/02834_apache_arrow_abort.sql b/tests/queries/0_stateless/02834_apache_arrow_abort.sql index 47db46f1e43..cb0eaebe554 100644 --- a/tests/queries/0_stateless/02834_apache_arrow_abort.sql +++ b/tests/queries/0_stateless/02834_apache_arrow_abort.sql @@ -1,4 +1,4 @@ --- Tags: no-fasttest +-- Tags: no-fasttest, no-tsan, no-asan, no-msan, no-ubsan -- This tests depends on internet access, but it does not matter, because it only has to check that there is no abort due to a bug in Apache Arrow library. - -INSERT INTO TABLE FUNCTION url('https://clickhouse-public-datasets.s3.amazonaws.com/hits_compatible/hits.parquet') SELECT * FROM url('https://clickhouse-public-datasets.s3.amazonaws.com/hits_compatible/hits.parquet'); -- { serverError CANNOT_WRITE_TO_OSTREAM, RECEIVED_ERROR_FROM_REMOTE_IO_SERVER, POCO_EXCEPTION } +SET optimize_trivial_insert_select=1; +INSERT INTO TABLE FUNCTION url('https://clickhouse-public-datasets.s3.amazonaws.com/hits_compatible/athena_partitioned/hits_9.parquet') SELECT * FROM url('https://clickhouse-public-datasets.s3.amazonaws.com/hits_compatible/athena_partitioned/hits_9.parquet'); -- { serverError CANNOT_WRITE_TO_OSTREAM, RECEIVED_ERROR_FROM_REMOTE_IO_SERVER, POCO_EXCEPTION } diff --git a/tests/queries/0_stateless/02841_not_ready_set_constraints.reference b/tests/queries/0_stateless/02841_not_ready_set_constraints.reference new file mode 100644 index 00000000000..daaac9e3030 --- /dev/null +++ b/tests/queries/0_stateless/02841_not_ready_set_constraints.reference @@ -0,0 +1,2 @@ +42 +42 diff --git a/tests/queries/0_stateless/02841_not_ready_set_constraints.sql b/tests/queries/0_stateless/02841_not_ready_set_constraints.sql new file mode 100644 index 00000000000..274940f50a3 --- /dev/null +++ b/tests/queries/0_stateless/02841_not_ready_set_constraints.sql @@ -0,0 +1,43 @@ +DROP TABLE IF EXISTS t1; +DROP TABLE IF EXISTS t2; + +CREATE TABLE t1 ( + `id` UInt64 +) +ENGINE = MergeTree ORDER BY id; + +INSERT INTO t1(id) VALUES (42); + +CREATE TABLE t2 ( + `conversation` UInt64, + CONSTRAINT constraint_conversation CHECK conversation IN (SELECT id FROM t1) +) +ENGINE = MergeTree ORDER BY conversation; + +INSERT INTO t2(conversation) VALUES (42); + +select * from t2; + +drop table t1; + +INSERT INTO t2(conversation) VALUES (42); -- { serverError UNKNOWN_TABLE } + +drop table t2; + +CREATE TABLE t2 ( + `conversation` UInt64, + CONSTRAINT constraint_conversation CHECK conversation IN (SELECT id FROM t1) +) +ENGINE = MergeTree ORDER BY conversation; + +INSERT INTO t2(conversation) VALUES (42); -- { serverError UNKNOWN_TABLE } + +CREATE TABLE t1 ( + `id` UInt64 +) +ENGINE = MergeTree ORDER BY id; + +INSERT INTO t1(id) VALUES (42); + +INSERT INTO t2(conversation) VALUES (42); +select * from t2; diff --git a/tests/queries/0_stateless/02842_truncate_database.reference b/tests/queries/0_stateless/02842_truncate_database.reference index 71f52bcd1da..bc8c0210d27 100644 --- a/tests/queries/0_stateless/02842_truncate_database.reference +++ b/tests/queries/0_stateless/02842_truncate_database.reference @@ -20,3 +20,5 @@ source_table_stripe_log source_table_tiny_log === DICTIONARIES IN test_truncate_database === dest_dictionary +new tables +new_table diff --git a/tests/queries/0_stateless/02842_truncate_database.sql b/tests/queries/0_stateless/02842_truncate_database.sql index 09ac844cfe2..bcd818f55ba 100644 --- a/tests/queries/0_stateless/02842_truncate_database.sql +++ b/tests/queries/0_stateless/02842_truncate_database.sql @@ -73,4 +73,8 @@ SELECT * FROM dest_dictionary; -- {serverError UNKNOWN_TABLE} SHOW TABLES FROM test_truncate_database; SHOW DICTIONARIES FROM test_truncate_database; +CREATE TABLE new_table (x UInt16) ENGINE = MergeTree ORDER BY x; +select 'new tables'; +SHOW TABLES FROM test_truncate_database; + DROP DATABASE test_truncate_database; diff --git a/tests/queries/0_stateless/02864_replace_partition_with_duplicated_parts_zookeeper.sh b/tests/queries/0_stateless/02864_replace_partition_with_duplicated_parts_zookeeper.sh index edfed206d87..07d2ee27d22 100755 --- a/tests/queries/0_stateless/02864_replace_partition_with_duplicated_parts_zookeeper.sh +++ b/tests/queries/0_stateless/02864_replace_partition_with_duplicated_parts_zookeeper.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: zookeeper, no-s3-storage +# Tags: zookeeper, no-object-storage # Because REPLACE PARTITION does not forces immediate removal of replaced data parts from local filesystem # (it tries to do it as quick as possible, but it still performed in separate thread asynchronously) diff --git a/tests/queries/0_stateless/02864_replace_regexp_string_fallback.reference b/tests/queries/0_stateless/02864_replace_regexp_string_fallback.reference new file mode 100644 index 00000000000..dd52d49eea3 --- /dev/null +++ b/tests/queries/0_stateless/02864_replace_regexp_string_fallback.reference @@ -0,0 +1 @@ +Hello l x Hexlo Hexxo diff --git a/tests/queries/0_stateless/02864_replace_regexp_string_fallback.sql b/tests/queries/0_stateless/02864_replace_regexp_string_fallback.sql new file mode 100644 index 00000000000..917c11fe8dd --- /dev/null +++ b/tests/queries/0_stateless/02864_replace_regexp_string_fallback.sql @@ -0,0 +1,11 @@ +-- Tests functions replaceRegexpAll and replaceRegexpOne with trivial patterns. These trigger internally a fallback to simple string replacement. + +-- _materialize_ because the shortcut is only implemented for non-const haystack + const needle + const replacement strings + +SELECT 'Hello' AS haystack, 'l' AS needle, 'x' AS replacement, replaceRegexpOne(materialize(haystack), needle, replacement), replaceRegexpAll(materialize(haystack), needle, replacement); + +-- negative tests + +-- Even if the fallback is used, invalid substitutions must throw an exception. +SELECT 'Hello' AS haystack, 'l' AS needle, '\\1' AS replacement, replaceRegexpOne(materialize(haystack), needle, replacement); -- { serverError BAD_ARGUMENTS } +SELECT 'Hello' AS haystack, 'l' AS needle, '\\1' AS replacement, replaceRegexpAll(materialize(haystack), needle, replacement); -- { serverError BAD_ARGUMENTS } diff --git a/tests/queries/0_stateless/02864_statistics_count_min_sketch.reference b/tests/queries/0_stateless/02864_statistics_count_min_sketch.reference new file mode 100644 index 00000000000..02c41656a36 --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_count_min_sketch.reference @@ -0,0 +1,14 @@ +CREATE TABLE default.tab\n(\n `a` String,\n `b` UInt64,\n `c` Int64,\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 +Test statistics count_min: + Prewhere info + Prewhere filter + Prewhere filter column: and(equals(a, \'0\'), equals(b, 0), equals(c, 0)) (removed) +Test statistics multi-types: + Prewhere info + Prewhere filter + Prewhere filter column: and(equals(a, \'0\'), less(c, -90), greater(b, 900)) (removed) + Prewhere info + Prewhere filter + Prewhere filter column: and(equals(a, \'10000\'), equals(b, 0), less(c, 0)) (removed) +Test LowCardinality and Nullable data type: +tab2 diff --git a/tests/queries/0_stateless/02864_statistics_count_min_sketch.sql b/tests/queries/0_stateless/02864_statistics_count_min_sketch.sql new file mode 100644 index 00000000000..c730aa7b4a7 --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_count_min_sketch.sql @@ -0,0 +1,70 @@ +-- Tags: no-fasttest + +DROP TABLE IF EXISTS tab SYNC; + +SET allow_experimental_statistics = 1; +SET allow_statistics_optimize = 1; +SET allow_suspicious_low_cardinality_types=1; +SET mutations_sync = 2; + +CREATE TABLE tab +( + a String, + b UInt64, + c Int64, + pk String, +) Engine = MergeTree() ORDER BY pk +SETTINGS min_bytes_for_wide_part = 0; + +SHOW CREATE TABLE tab; + +INSERT INTO tab select toString(number % 10000), number % 1000, -(number % 100), generateUUIDv4() FROM system.numbers LIMIT 10000; + +SELECT 'Test statistics count_min:'; + +ALTER TABLE tab ADD STATISTICS a TYPE count_min; +ALTER TABLE tab ADD STATISTICS b TYPE count_min; +ALTER TABLE tab ADD STATISTICS c TYPE count_min; +ALTER TABLE tab MATERIALIZE STATISTICS a, b, c; + +SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8|_UInt16|_String', '') +FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE c = 0/*100*/ and b = 0/*10*/ and a = '0'/*1*/) xx +WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; + +ALTER TABLE tab DROP STATISTICS a, b, c; + + +SELECT 'Test statistics multi-types:'; + +ALTER TABLE tab ADD STATISTICS a TYPE count_min; +ALTER TABLE tab ADD STATISTICS b TYPE count_min, uniq, tdigest; +ALTER TABLE tab ADD STATISTICS c TYPE count_min, uniq, tdigest; +ALTER TABLE tab MATERIALIZE STATISTICS a, b, c; + +SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8|_UInt16|_String', '') +FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE c < -90/*900*/ and b > 900/*990*/ and a = '0'/*1*/) +WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; + +SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8|_UInt16|_String', '') +FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE c < 0/*9900*/ and b = 0/*10*/ and a = '10000'/*0*/) +WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; + +ALTER TABLE tab DROP STATISTICS a, b, c; + +DROP TABLE IF EXISTS tab SYNC; + + +SELECT 'Test LowCardinality and Nullable data type:'; +DROP TABLE IF EXISTS tab2 SYNC; +SET allow_suspicious_low_cardinality_types=1; +CREATE TABLE tab2 +( + a LowCardinality(Int64) STATISTICS(count_min), + b Nullable(Int64) STATISTICS(count_min), + c LowCardinality(Nullable(Int64)) STATISTICS(count_min), + pk String, +) Engine = MergeTree() ORDER BY pk; + +select name from system.tables where name = 'tab2' and database = currentDatabase(); + +DROP TABLE IF EXISTS tab2 SYNC; diff --git a/tests/queries/0_stateless/02864_statistics_ddl.reference b/tests/queries/0_stateless/02864_statistics_ddl.reference new file mode 100644 index 00000000000..a7ff5caa0b0 --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_ddl.reference @@ -0,0 +1,31 @@ +CREATE TABLE default.tab\n(\n `a` Float64 STATISTICS(tdigest),\n `b` Int64 STATISTICS(tdigest),\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 +After insert + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10), less(b, 10)) (removed) +10 +0 +After drop statistic + Prewhere info + Prewhere filter + Prewhere filter column: and(less(b, 10), less(a, 10)) (removed) +10 +CREATE TABLE default.tab\n(\n `a` Float64,\n `b` Int64,\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 +After add statistic +CREATE TABLE default.tab\n(\n `a` Float64 STATISTICS(tdigest),\n `b` Int64 STATISTICS(tdigest),\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 +After materialize statistic + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10), less(b, 10)) (removed) +20 +After merge + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10), less(b, 10)) (removed) +20 +CREATE TABLE default.tab\n(\n `a` Float64 STATISTICS(tdigest),\n `c` Int64 STATISTICS(tdigest),\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 +After rename + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10), less(c, 10)) (removed) +20 diff --git a/tests/queries/0_stateless/02864_statistics_ddl.sql b/tests/queries/0_stateless/02864_statistics_ddl.sql new file mode 100644 index 00000000000..fe612efe2ac --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_ddl.sql @@ -0,0 +1,59 @@ +-- Tests that various DDL statements create/drop/materialize statistics + +DROP TABLE IF EXISTS tab; + +SET allow_experimental_statistics = 1; +SET allow_statistics_optimize = 1; + +CREATE TABLE tab +( + a Float64 STATISTICS(tdigest), + b Int64 STATISTICS(tdigest), + pk String, +) Engine = MergeTree() ORDER BY pk +SETTINGS min_bytes_for_wide_part = 0; + +SHOW CREATE TABLE tab; + +INSERT INTO tab select number, -number, generateUUIDv4() FROM system.numbers LIMIT 10000; + +SELECT 'After insert'; +SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; +SELECT count(*) FROM tab WHERE b < 10 and a < 10; +SELECT count(*) FROM tab WHERE b < NULL and a < '10'; + +ALTER TABLE tab DROP STATISTICS a, b; + +SELECT 'After drop statistic'; +SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; +SELECT count(*) FROM tab WHERE b < 10 and a < 10; + +SHOW CREATE TABLE tab; + +ALTER TABLE tab ADD STATISTICS a, b TYPE tdigest; + +SELECT 'After add statistic'; + +SHOW CREATE TABLE tab; + +ALTER TABLE tab MATERIALIZE STATISTICS a, b; +INSERT INTO tab select number, -number, generateUUIDv4() FROM system.numbers LIMIT 10000; + +SELECT 'After materialize statistic'; +SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; +SELECT count(*) FROM tab WHERE b < 10 and a < 10; + +OPTIMIZE TABLE tab FINAL; + +SELECT 'After merge'; +SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; +SELECT count(*) FROM tab WHERE b < 10 and a < 10; + +ALTER TABLE tab RENAME COLUMN b TO c; +SHOW CREATE TABLE tab; + +SELECT 'After rename'; +SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE c < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; +SELECT count(*) FROM tab WHERE c < 10 and a < 10; + +DROP TABLE IF EXISTS tab; diff --git a/tests/queries/0_stateless/02864_statistics_exception.sql b/tests/queries/0_stateless/02864_statistics_exception.sql index c531d39cd69..289ffee6600 100644 --- a/tests/queries/0_stateless/02864_statistics_exception.sql +++ b/tests/queries/0_stateless/02864_statistics_exception.sql @@ -1,57 +1,55 @@ -DROP TABLE IF EXISTS t1; +-- Tests creating/dropping/materializing statistics produces the right exceptions. -CREATE TABLE t1 +DROP TABLE IF EXISTS tab; + +-- Can't create statistics when allow_experimental_statistics = 0 +CREATE TABLE tab ( - a Float64 STATISTICS(tdigest), - b Int64 STATISTICS(tdigest), - pk String, -) Engine = MergeTree() ORDER BY pk; -- { serverError INCORRECT_QUERY } + a Float64 STATISTICS(tdigest) +) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } SET allow_experimental_statistics = 1; -CREATE TABLE t1 +-- The same type of statistics can't exist more than once on a column +CREATE TABLE tab ( - a Float64 STATISTICS(tdigest), - b Int64, - pk String STATISTICS(tdigest), -) Engine = MergeTree() ORDER BY pk; -- { serverError ILLEGAL_STATISTICS } + a Float64 STATISTICS(tdigest, tdigest) +) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } -CREATE TABLE t1 +-- Unknown statistics types are rejected +CREATE TABLE tab ( - a Float64 STATISTICS(tdigest, tdigest(10)), - b Int64, -) Engine = MergeTree() ORDER BY pk; -- { serverError INCORRECT_QUERY } + a Float64 STATISTICS(no_statistics_type) +) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } -CREATE TABLE t1 +-- tDigest statistics can only be created on numeric columns +CREATE TABLE tab ( - a Float64 STATISTICS(xyz), - b Int64, -) Engine = MergeTree() ORDER BY pk; -- { serverError INCORRECT_QUERY } + a String STATISTICS(tdigest), +) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } -CREATE TABLE t1 +CREATE TABLE tab ( a Float64, - b Int64, - pk String, -) Engine = MergeTree() ORDER BY pk; + b String +) Engine = MergeTree() ORDER BY tuple(); -ALTER TABLE t1 ADD STATISTICS a TYPE xyz; -- { serverError INCORRECT_QUERY } -ALTER TABLE t1 ADD STATISTICS a TYPE tdigest; -ALTER TABLE t1 ADD STATISTICS IF NOT EXISTS a TYPE tdigest; -ALTER TABLE t1 ADD STATISTICS a TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab ADD STATISTICS a TYPE no_statistics_type; -- { serverError INCORRECT_QUERY } +ALTER TABLE tab ADD STATISTICS a TYPE tdigest; +ALTER TABLE tab ADD STATISTICS IF NOT EXISTS a TYPE tdigest; +ALTER TABLE tab ADD STATISTICS a TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab MODIFY STATISTICS a TYPE tdigest; -- Statistics can be created only on integer columns -ALTER TABLE t1 MODIFY STATISTICS a TYPE tdigest; -ALTER TABLE t1 ADD STATISTICS pk TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } -ALTER TABLE t1 DROP STATISTICS b; -- { serverError ILLEGAL_STATISTICS } -ALTER TABLE t1 DROP STATISTICS a; -ALTER TABLE t1 DROP STATISTICS IF EXISTS a; -ALTER TABLE t1 CLEAR STATISTICS a; -- { serverError ILLEGAL_STATISTICS } -ALTER TABLE t1 CLEAR STATISTICS IF EXISTS a; -ALTER TABLE t1 MATERIALIZE STATISTICS b; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab ADD STATISTICS b TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab DROP STATISTICS b; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab DROP STATISTICS a; +ALTER TABLE tab DROP STATISTICS IF EXISTS a; +ALTER TABLE tab CLEAR STATISTICS a; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab CLEAR STATISTICS IF EXISTS a; +ALTER TABLE tab MATERIALIZE STATISTICS b; -- { serverError ILLEGAL_STATISTICS } -ALTER TABLE t1 ADD STATISTICS a TYPE tdigest; -ALTER TABLE t1 ADD STATISTICS b TYPE tdigest; -ALTER TABLE t1 MODIFY COLUMN a Float64 TTL toDateTime(b) + INTERVAL 1 MONTH; -ALTER TABLE t1 MODIFY COLUMN a Int64; -- { serverError ALTER_OF_COLUMN_IS_FORBIDDEN } +ALTER TABLE tab ADD STATISTICS a TYPE tdigest; +ALTER TABLE tab MODIFY COLUMN a Float64 TTL toDateTime(b) + INTERVAL 1 MONTH; +ALTER TABLE tab MODIFY COLUMN a Int64; -- { serverError ALTER_OF_COLUMN_IS_FORBIDDEN } -DROP TABLE t1; +DROP TABLE tab; diff --git a/tests/queries/0_stateless/03164_materialize_statistics.reference b/tests/queries/0_stateless/02864_statistics_materialize_in_merge.reference similarity index 100% rename from tests/queries/0_stateless/03164_materialize_statistics.reference rename to tests/queries/0_stateless/02864_statistics_materialize_in_merge.reference diff --git a/tests/queries/0_stateless/02864_statistics_materialize_in_merge.sql b/tests/queries/0_stateless/02864_statistics_materialize_in_merge.sql new file mode 100644 index 00000000000..3e15ec1148e --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_materialize_in_merge.sql @@ -0,0 +1,52 @@ +-- Tests delayed materialization of statistics in merge instead of during insert (setting 'materialize_statistics_on_insert = 0'). + +DROP TABLE IF EXISTS tab; + +SET allow_experimental_analyzer = 1; +SET allow_experimental_statistics = 1; +SET allow_statistics_optimize = 1; + +SET materialize_statistics_on_insert = 0; + +CREATE TABLE tab +( + a Int64 STATISTICS(tdigest), + b Int16 STATISTICS(tdigest), +) ENGINE = MergeTree() ORDER BY tuple() +SETTINGS min_bytes_for_wide_part = 0, enable_vertical_merge_algorithm = 0; -- TODO: there is a bug in vertical merge with statistics. + +INSERT INTO tab SELECT number, -number FROM system.numbers LIMIT 10000; + +SELECT count(*) FROM tab WHERE b < 10 and a < 10 SETTINGS log_comment = 'statistics not used'; + +OPTIMIZE TABLE tab FINAL; + +SELECT count(*) FROM tab WHERE b < 10 and a < 10 SETTINGS log_comment = 'statistics used after merge'; + +TRUNCATE TABLE tab; +SET mutations_sync = 2; + +INSERT INTO tab SELECT number, -number FROM system.numbers LIMIT 10000; +ALTER TABLE tab MATERIALIZE STATISTICS a, b; + +SELECT count(*) FROM tab WHERE b < 10 and a < 10 SETTINGS log_comment = 'statistics used after materialize'; + +DROP TABLE tab; + +SYSTEM FLUSH LOGS; + +SELECT log_comment, message FROM system.text_log JOIN +( + SELECT Settings['log_comment'] AS log_comment, query_id FROM system.query_log + WHERE current_database = currentDatabase() + AND query LIKE 'SELECT count(*) FROM tab%' + AND type = 'QueryFinish' +) AS query_log USING (query_id) +WHERE message LIKE '%moved to PREWHERE%' +ORDER BY event_time_microseconds; + +SELECT count(), sum(ProfileEvents['MergeTreeDataWriterStatisticsCalculationMicroseconds']) +FROM system.query_log +WHERE current_database = currentDatabase() + AND query LIKE 'INSERT INTO tab SELECT%' + AND type = 'QueryFinish'; diff --git a/tests/queries/0_stateless/02864_statistics_operate.reference b/tests/queries/0_stateless/02864_statistics_operate.reference deleted file mode 100644 index 6398a9bd000..00000000000 --- a/tests/queries/0_stateless/02864_statistics_operate.reference +++ /dev/null @@ -1,31 +0,0 @@ -CREATE TABLE default.t1\n(\n `a` Float64 STATISTICS(tdigest),\n `b` Int64 STATISTICS(tdigest),\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 -After insert - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), less(b, 10)) (removed) -10 -0 -After drop statistic - Prewhere info - Prewhere filter - Prewhere filter column: and(less(b, 10), less(a, 10)) (removed) -10 -CREATE TABLE default.t1\n(\n `a` Float64,\n `b` Int64,\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 -After add statistic -CREATE TABLE default.t1\n(\n `a` Float64 STATISTICS(tdigest),\n `b` Int64 STATISTICS(tdigest),\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 -After materialize statistic - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), less(b, 10)) (removed) -20 -After merge - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), less(b, 10)) (removed) -20 -CREATE TABLE default.t1\n(\n `a` Float64 STATISTICS(tdigest),\n `c` Int64 STATISTICS(tdigest),\n `pk` String\n)\nENGINE = MergeTree\nORDER BY pk\nSETTINGS min_bytes_for_wide_part = 0, index_granularity = 8192 -After rename - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), less(c, 10)) (removed) -20 diff --git a/tests/queries/0_stateless/02864_statistics_operate.sql b/tests/queries/0_stateless/02864_statistics_operate.sql deleted file mode 100644 index bf69c11bc91..00000000000 --- a/tests/queries/0_stateless/02864_statistics_operate.sql +++ /dev/null @@ -1,57 +0,0 @@ -DROP TABLE IF EXISTS t1; - -SET allow_experimental_statistics = 1; -SET allow_statistics_optimize = 1; - -CREATE TABLE t1 -( - a Float64 STATISTICS(tdigest), - b Int64 STATISTICS(tdigest), - pk String, -) Engine = MergeTree() ORDER BY pk -SETTINGS min_bytes_for_wide_part = 0; - -SHOW CREATE TABLE t1; - -INSERT INTO t1 select number, -number, generateUUIDv4() FROM system.numbers LIMIT 10000; - -SELECT 'After insert'; -SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT count(*) FROM t1 WHERE b < 10 and a < 10; -SELECT count(*) FROM t1 WHERE b < NULL and a < '10'; - -ALTER TABLE t1 DROP STATISTICS a, b; - -SELECT 'After drop statistic'; -SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT count(*) FROM t1 WHERE b < 10 and a < 10; - -SHOW CREATE TABLE t1; - -ALTER TABLE t1 ADD STATISTICS a, b TYPE tdigest; - -SELECT 'After add statistic'; - -SHOW CREATE TABLE t1; - -ALTER TABLE t1 MATERIALIZE STATISTICS a, b; -INSERT INTO t1 select number, -number, generateUUIDv4() FROM system.numbers LIMIT 10000; - -SELECT 'After materialize statistic'; -SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT count(*) FROM t1 WHERE b < 10 and a < 10; - -OPTIMIZE TABLE t1 FINAL; - -SELECT 'After merge'; -SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT count(*) FROM t1 WHERE b < 10 and a < 10; - -ALTER TABLE t1 RENAME COLUMN b TO c; -SHOW CREATE TABLE t1; - -SELECT 'After rename'; -SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE c < 10 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT count(*) FROM t1 WHERE c < 10 and a < 10; - -DROP TABLE IF EXISTS t1; diff --git a/tests/queries/0_stateless/02864_statistics_uniq.sql b/tests/queries/0_stateless/02864_statistics_uniq.sql index d496392668b..0f5f353c045 100644 --- a/tests/queries/0_stateless/02864_statistics_uniq.sql +++ b/tests/queries/0_stateless/02864_statistics_uniq.sql @@ -70,3 +70,4 @@ SETTINGS min_bytes_for_wide_part = 0; INSERT INTO t3 select number, -number, number/1000, generateUUIDv4() FROM system.numbers LIMIT 10000; DROP TABLE IF EXISTS t3; + diff --git a/tests/queries/0_stateless/02870_move_partition_to_volume_io_throttling.sql b/tests/queries/0_stateless/02870_move_partition_to_volume_io_throttling.sql index b03d9849a80..2d76ab0d8e3 100644 --- a/tests/queries/0_stateless/02870_move_partition_to_volume_io_throttling.sql +++ b/tests/queries/0_stateless/02870_move_partition_to_volume_io_throttling.sql @@ -2,6 +2,8 @@ -- Tag: no-fasttest -- requires S3 -- Tag: no-replicated-database -- ALTER MOVE PARTITION TO should not be replicated (will be fixed separatelly) +SET optimize_trivial_insert_select = 1; + CREATE TABLE test_move_partition_throttling (key UInt64 CODEC(NONE)) ENGINE = MergeTree ORDER BY tuple() SETTINGS storage_policy='local_remote'; INSERT INTO test_move_partition_throttling SELECT number FROM numbers(1e6); SELECT disk_name, partition, rows FROM system.parts WHERE database = currentDatabase() AND table = 'test_move_partition_throttling' and active; diff --git a/tests/queries/0_stateless/02874_parquet_multiple_batches_array_inconsistent_offsets.reference b/tests/queries/0_stateless/02874_parquet_multiple_batches_array_inconsistent_offsets.reference index ba63f2f7e9c..a9eae234dba 100644 --- a/tests/queries/0_stateless/02874_parquet_multiple_batches_array_inconsistent_offsets.reference +++ b/tests/queries/0_stateless/02874_parquet_multiple_batches_array_inconsistent_offsets.reference @@ -1,3 +1,3 @@ Parquet -e76a749f346078a6a43e0cbd25f0d18a - +3249508141921544766 400 diff --git a/tests/queries/0_stateless/02874_parquet_multiple_batches_array_inconsistent_offsets.sh b/tests/queries/0_stateless/02874_parquet_multiple_batches_array_inconsistent_offsets.sh index 83196458a84..c96531ffea9 100755 --- a/tests/queries/0_stateless/02874_parquet_multiple_batches_array_inconsistent_offsets.sh +++ b/tests/queries/0_stateless/02874_parquet_multiple_batches_array_inconsistent_offsets.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: no-ubsan, no-fasttest +# Tags: long, no-ubsan, no-fasttest, no-parallel, no-asan, no-msan, no-tsan +# This test requires around 10 GB of memory and it is just too much. CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -121,9 +122,12 @@ echo "Parquet" #} DATA_FILE=$CUR_DIR/data_parquet/string_int_list_inconsistent_offset_multiple_batches.parquet -${CLICKHOUSE_CLIENT} --query="DROP TABLE IF EXISTS parquet_load" -${CLICKHOUSE_CLIENT} --query="CREATE TABLE parquet_load (ints Array(Int64), strings Nullable(String)) ENGINE = Memory" -cat "$DATA_FILE" | ${CLICKHOUSE_CLIENT} -q "INSERT INTO parquet_load FORMAT Parquet" -${CLICKHOUSE_CLIENT} --query="SELECT * FROM parquet_load" | md5sum -${CLICKHOUSE_CLIENT} --query="SELECT count() FROM parquet_load" -${CLICKHOUSE_CLIENT} --query="drop table parquet_load" \ No newline at end of file + +${CLICKHOUSE_LOCAL} --multiquery " +DROP TABLE IF EXISTS parquet_load; +CREATE TABLE parquet_load (ints Array(Int64), strings Nullable(String)) ENGINE = Memory; +INSERT INTO parquet_load FROM INFILE '$DATA_FILE'; +SELECT sum(cityHash64(*)) FROM parquet_load; +SELECT count() FROM parquet_load; +DROP TABLE parquet_load; +" diff --git a/tests/queries/0_stateless/02884_async_insert_skip_settings.reference b/tests/queries/0_stateless/02884_async_insert_skip_settings.reference index 318966ce93b..957be52c59b 100644 --- a/tests/queries/0_stateless/02884_async_insert_skip_settings.reference +++ b/tests/queries/0_stateless/02884_async_insert_skip_settings.reference @@ -1,5 +1,5 @@ -4 +pending to flush 4 1 1 2 -1 +flush queries 1 diff --git a/tests/queries/0_stateless/02884_async_insert_skip_settings.sql b/tests/queries/0_stateless/02884_async_insert_skip_settings.sql index 9bc689fb4ec..df715af873a 100644 --- a/tests/queries/0_stateless/02884_async_insert_skip_settings.sql +++ b/tests/queries/0_stateless/02884_async_insert_skip_settings.sql @@ -9,7 +9,8 @@ ORDER BY id; SET async_insert = 1; SET async_insert_deduplicate = 1; SET wait_for_async_insert = 0; -SET async_insert_busy_timeout_min_ms = 100000; +-- Disable adaptive timeout to prevent immediate push of the first message (if the queue last push was old) +SET async_insert_use_adaptive_busy_timeout=0; SET async_insert_busy_timeout_max_ms = 1000000; SET insert_deduplication_token = '1'; @@ -30,7 +31,7 @@ INSERT INTO t_async_insert_skip_settings VALUES (2); SYSTEM FLUSH LOGS; -SELECT length(entries.bytes) FROM system.asynchronous_inserts +SELECT 'pending to flush', length(entries.bytes) FROM system.asynchronous_inserts WHERE database = currentDatabase() AND table = 't_async_insert_skip_settings' ORDER BY first_update; @@ -40,7 +41,7 @@ SELECT * FROM t_async_insert_skip_settings ORDER BY id; SYSTEM FLUSH LOGS; -SELECT uniqExact(flush_query_id) FROM system.asynchronous_insert_log +SELECT 'flush queries', uniqExact(flush_query_id) FROM system.asynchronous_insert_log WHERE database = currentDatabase() AND table = 't_async_insert_skip_settings'; DROP TABLE t_async_insert_skip_settings SYNC; diff --git a/tests/queries/0_stateless/02889_file_log_save_errors.sh b/tests/queries/0_stateless/02889_file_log_save_errors.sh index 8ef7816d57d..cf7ced0bd08 100755 --- a/tests/queries/0_stateless/02889_file_log_save_errors.sh +++ b/tests/queries/0_stateless/02889_file_log_save_errors.sh @@ -4,27 +4,25 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - ${CLICKHOUSE_CLIENT} --query "drop table if exists file_log;" ${CLICKHOUSE_CLIENT} --query "drop table if exists log_errors;" -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* for i in {0..9} do - echo "{\"key\" : $i, \"value\" : $i}" >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.jsonl - echo "Error $i" >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.jsonl + echo "{\"key\" : $i, \"value\" : $i}" >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.jsonl + echo "Error $i" >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/a.jsonl done for i in {10..19} do - echo "{\"key\" : $i, \"value\" : $i}" >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.jsonl - echo "Error $i" >> ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.jsonl + echo "{\"key\" : $i, \"value\" : $i}" >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.jsonl + echo "Error $i" >> ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/b.jsonl done -${CLICKHOUSE_CLIENT} --query "create table file_log(key UInt8, value UInt8) engine=FileLog('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'JSONEachRow') settings handle_error_mode='stream';" +${CLICKHOUSE_CLIENT} --query "create table file_log(key UInt8, value UInt8) engine=FileLog('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/', 'JSONEachRow') settings handle_error_mode='stream';" ${CLICKHOUSE_CLIENT} --query "create Materialized View log_errors engine=MergeTree order by tuple() as select _error as error, _raw_record as record, _filename as file from file_log where not isNull(_error);" function count() @@ -42,4 +40,4 @@ ${CLICKHOUSE_CLIENT} --query "select * from log_errors order by file, record;" ${CLICKHOUSE_CLIENT} --query "drop table file_log;" ${CLICKHOUSE_CLIENT} --query "drop table log_errors;" -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} diff --git a/tests/queries/0_stateless/02890_named_tuple_functions.reference b/tests/queries/0_stateless/02890_named_tuple_functions.reference new file mode 100644 index 00000000000..6b36ff2c54c --- /dev/null +++ b/tests/queries/0_stateless/02890_named_tuple_functions.reference @@ -0,0 +1,10 @@ +Tuple(\n i Int32,\n j Int32) +['i','j'] +Tuple(UInt8, Int32) +['1','2'] +Tuple(\n k UInt8,\n j Int32) +['k','j'] +Tuple(Int32, Int32, Int32, Int32) +['1','2','3','4'] +(1,2,3) +Tuple(Nullable(Nothing)) Tuple(Bool) Tuple(Bool) diff --git a/tests/queries/0_stateless/02890_named_tuple_functions.sql b/tests/queries/0_stateless/02890_named_tuple_functions.sql new file mode 100644 index 00000000000..0033102bd53 --- /dev/null +++ b/tests/queries/0_stateless/02890_named_tuple_functions.sql @@ -0,0 +1,34 @@ +set enable_named_columns_in_function_tuple = 1; +set allow_experimental_analyzer = 1; + +drop table if exists x; +create table x (i int, j int) engine MergeTree order by i; +insert into x values (1, 2); + +select toTypeName(tuple(i, j)) from x; +select tupleNames(tuple(i, j)) from x; + +select toTypeName(tuple(1, j)) from x; +select tupleNames(tuple(1, j)) from x; + +select toTypeName(tuple(1 as k, j)) from x; +select tupleNames(tuple(1 as k, j)) from x; + +select toTypeName(tuple(i, i, j, j)) from x; +select tupleNames(tuple(i, i, j, j)) from x; + +select tupleNames(1); -- { serverError 43 } + +drop table x; + +drop table if exists tbl; + +-- Make sure named tuple won't break Values insert +create table tbl (x Tuple(a Int32, b Int32, c Int32)) engine MergeTree order by (); +insert into tbl values (tuple(1, 2, 3)); -- without tuple it's interpreted differently inside values block. +select * from tbl; + +drop table tbl; + +-- Avoid generating named tuple for special keywords +select toTypeName(tuple(null)), toTypeName(tuple(true)), toTypeName(tuple(false)); diff --git a/tests/queries/0_stateless/02890_untuple_column_names.reference b/tests/queries/0_stateless/02890_untuple_column_names.reference index 388f974c45f..13a85c70138 100644 --- a/tests/queries/0_stateless/02890_untuple_column_names.reference +++ b/tests/queries/0_stateless/02890_untuple_column_names.reference @@ -57,6 +57,10 @@ t.1: 1 Row 1: ────── t.1: 1 +-- tuple() with enable_named_columns_in_function_tuple = 1 and allow_experimental_analyzer = 1 keeps the column names +Row 1: +────── +t.a: 1 -- thankfully JSONExtract() keeps them Row 1: ────── diff --git a/tests/queries/0_stateless/02890_untuple_column_names.sql b/tests/queries/0_stateless/02890_untuple_column_names.sql index ab6748cb54d..cd490ca3522 100644 --- a/tests/queries/0_stateless/02890_untuple_column_names.sql +++ b/tests/queries/0_stateless/02890_untuple_column_names.sql @@ -37,8 +37,11 @@ SELECT untuple(tuple(1)::Tuple(Int)), untuple(tuple(1)::Tuple(Int)) FORMAT Verti SELECT untuple(tuple(1)::Tuple(Int)), untuple(tuple(1)::Tuple(Int)) FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; -- Bug: doesn't throw an exception SELECT '-- tuple() loses the column names (would be good to fix, see #36773)'; -SELECT untuple(tuple(1 as a)) as t FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -SELECT untuple(tuple(1 as a)) as t FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; +SELECT untuple(tuple(1 as a)) as t FORMAT Vertical SETTINGS allow_experimental_analyzer = 0, enable_named_columns_in_function_tuple = 0; +SELECT untuple(tuple(1 as a)) as t FORMAT Vertical SETTINGS allow_experimental_analyzer = 1, enable_named_columns_in_function_tuple = 0; + +SELECT '-- tuple() with enable_named_columns_in_function_tuple = 1 and allow_experimental_analyzer = 1 keeps the column names'; +SELECT untuple(tuple(1 as a)) as t FORMAT Vertical SETTINGS allow_experimental_analyzer = 1, enable_named_columns_in_function_tuple = 1; SELECT '-- thankfully JSONExtract() keeps them'; SELECT untuple(JSONExtract('{"key": "value"}', 'Tuple(key String)')) x FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; diff --git a/tests/queries/0_stateless/02892_input_csv_cr_end_count_many_rows.sh b/tests/queries/0_stateless/02892_input_csv_cr_end_count_many_rows.sh index 42dde18de00..9f93396e368 100755 --- a/tests/queries/0_stateless/02892_input_csv_cr_end_count_many_rows.sh +++ b/tests/queries/0_stateless/02892_input_csv_cr_end_count_many_rows.sh @@ -6,11 +6,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep -E '^Code: 107.*FILE_DOESNT_EXIST' | head -1 | awk '{gsub("/nonexist.txt","",$9); print $9}') - cp "$CURDIR"/data_csv/1m_rows_cr_end_of_line.csv.xz $USER_FILES_PATH/ $CLICKHOUSE_CLIENT -q "SELECT count(1) from file('1m_rows_cr_end_of_line.csv.xz') settings input_format_csv_allow_cr_end_of_line=1, optimize_count_from_files=1" $CLICKHOUSE_CLIENT -q "SELECT count(1) from file('1m_rows_cr_end_of_line.csv.xz') settings input_format_csv_allow_cr_end_of_line=1, optimize_count_from_files=0" -rm $USER_FILES_PATH/1m_rows_cr_end_of_line.csv.xz \ No newline at end of file +rm $USER_FILES_PATH/1m_rows_cr_end_of_line.csv.xz diff --git a/tests/queries/0_stateless/02892_orc_filter_pushdown.reference b/tests/queries/0_stateless/02892_orc_filter_pushdown.reference index 903d42bf492..0be6d13adb3 100644 --- a/tests/queries/0_stateless/02892_orc_filter_pushdown.reference +++ b/tests/queries/0_stateless/02892_orc_filter_pushdown.reference @@ -205,7 +205,7 @@ select count(), sum(number) from file('02892.orc', ORC, 'number UInt64, negative 600 419700 select count(), min(negative_or_null), max(negative_or_null) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where (negative_or_null < -500); 596 -1099 -501 -select count(), sum(number) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where indexHint(negative_or_null is null); +select count(), sum(number) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where indexHint(negative_or_null is null) settings allow_experimental_analyzer=1; 0 0 select count(), min(negative_or_null), max(negative_or_null) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where (negative_or_null is null); 0 0 0 diff --git a/tests/queries/0_stateless/02892_orc_filter_pushdown.sql b/tests/queries/0_stateless/02892_orc_filter_pushdown.sql index e3736de6a17..29055eb911a 100644 --- a/tests/queries/0_stateless/02892_orc_filter_pushdown.sql +++ b/tests/queries/0_stateless/02892_orc_filter_pushdown.sql @@ -206,7 +206,7 @@ select count(), min(nEgAtIvE_oR_nUlL), max(nEgAtIvE_oR_nUlL) from file('02892.or select count(), sum(number) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where indexHint(negative_or_null < -500); select count(), min(negative_or_null), max(negative_or_null) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where (negative_or_null < -500); -select count(), sum(number) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where indexHint(negative_or_null is null); +select count(), sum(number) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where indexHint(negative_or_null is null) settings allow_experimental_analyzer=1; select count(), min(negative_or_null), max(negative_or_null) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where (negative_or_null is null); select count(), sum(number) from file('02892.orc', ORC, 'number UInt64, negative_or_null Int64') where indexHint(negative_or_null in (0, -1, -10, -100, -1000)); diff --git a/tests/queries/0_stateless/02895_npy_format.reference b/tests/queries/0_stateless/02895_npy_format.reference index f9e77644a35..52972f0acbd 100644 --- a/tests/queries/0_stateless/02895_npy_format.reference +++ b/tests/queries/0_stateless/02895_npy_format.reference @@ -85,12 +85,12 @@ c [4,5,6] [[1,2],[3,4]] [[5,6],[7,8]] -0 -0 -0 -0 -0 -0 +1 +1 +1 +1 +1 +1 1 [2.199219,1.099609,3.300781] [4.25,3.34961,6.628906] diff --git a/tests/queries/0_stateless/02895_npy_format.sh b/tests/queries/0_stateless/02895_npy_format.sh index 9d05303a091..194b2bc1fe4 100755 --- a/tests/queries/0_stateless/02895_npy_format.sh +++ b/tests/queries/0_stateless/02895_npy_format.sh @@ -52,14 +52,14 @@ $CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/two_dim.npy', Npy, 'v $CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/three_dim.npy', Npy, 'value Array(Array(Int8))')" $CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_float.npy', Npy, 'value Array(Float32)')" 2>&1 | grep -c "BAD_ARGUMENTS" -$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_float.npy', Npy, 'value UUID')" 2>&1 | grep -c "BAD_ARGUMENTS" -$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_float.npy', Npy, 'value Tuple(UInt8)')" 2>&1 | grep -c "BAD_ARGUMENTS" +$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_float.npy', Npy, 'value UUID')" 2>&1 | grep -c "UNKNOWN_TYPE" +$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_float.npy', Npy, 'value Tuple(UInt8)')" 2>&1 | grep -c "UNKNOWN_TYPE" -$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_float.npy', Npy, 'value Int8')" 2>&1 | grep -c "BAD_ARGUMENTS" -$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_str.npy', Npy, 'value Int8')" 2>&1 | grep -c "BAD_ARGUMENTS" -$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_unicode.npy', Npy, 'value Float32')" 2>&1 | grep -c "BAD_ARGUMENTS" +$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_float.npy', Npy, 'value Int8')" 2>&1 | grep -c "ILLEGAL_COLUMN" +$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_str.npy', Npy, 'value Int8')" 2>&1 | grep -c "ILLEGAL_COLUMN" +$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/one_dim_unicode.npy', Npy, 'value Float32')" 2>&1 | grep -c "ILLEGAL_COLUMN" -$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/complex.npy')" 2>&1 | grep -c "BAD_ARGUMENTS" +$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/complex.npy')" 2>&1 | grep -c "CANNOT_EXTRACT_TABLE_STRUCTURE" $CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_npy/float_16.npy')" diff --git a/tests/queries/0_stateless/02895_npy_output_format.sh b/tests/queries/0_stateless/02895_npy_output_format.sh index 934c80830c5..a364e447062 100755 --- a/tests/queries/0_stateless/02895_npy_output_format.sh +++ b/tests/queries/0_stateless/02895_npy_output_format.sh @@ -5,10 +5,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -user_files_path=$($CLICKHOUSE_CLIENT_BINARY -q "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') -mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* -chmod 777 ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/* +chmod 777 ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ ${CLICKHOUSE_CLIENT} -n -q --ignore-error " DROP DATABASE IF EXISTS npy_output_02895; @@ -33,43 +32,43 @@ ${CLICKHOUSE_CLIENT} -n -q --ignore-error " INSERT INTO npy_output_02895.data_types VALUES (1, 1, 1, 1, 1, 1, 1, 1, 0.1, 0.01, 'npy', 'npy'), (-1, -1, -1, -1, 0, 0, 0, 0, 0.2, 0.02, 'npy', 'npynpy'); - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int8.npy') SELECT i1 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int16.npy') SELECT i2 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int32.npy') SELECT i4 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int64.npy') SELECT i8 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint8.npy') SELECT u1 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint16.npy') SELECT u2 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint32.npy') SELECT u4 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint64.npy') SELECT u8 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float32.npy') SELECT f4 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float64.npy') SELECT f8 FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_fixedstring.npy') SELECT fs FROM npy_output_02895.data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_string.npy') SELECT s FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int8.npy') SELECT i1 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int16.npy') SELECT i2 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int32.npy') SELECT i4 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int64.npy') SELECT i8 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint8.npy') SELECT u1 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint16.npy') SELECT u2 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint32.npy') SELECT u4 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint64.npy') SELECT u8 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float32.npy') SELECT f4 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float64.npy') SELECT f8 FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_fixedstring.npy') SELECT fs FROM npy_output_02895.data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_string.npy') SELECT s FROM npy_output_02895.data_types; - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int8.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int16.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int32.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int64.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint8.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint16.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint32.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint64.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float32.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float64.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_fixedstring.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_string.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int8.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int16.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int32.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int64.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint8.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint16.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint32.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint64.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float32.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float64.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_fixedstring.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_string.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int8.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int16.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int32.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int64.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint8.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint16.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint32.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint64.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float32.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float64.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_fixedstring.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_string.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int8.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int16.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int32.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_int64.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint8.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint16.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint32.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_uint64.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float32.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_float64.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_fixedstring.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_dtype_string.npy'); SELECT '-- test nested data types --'; CREATE TABLE IF NOT EXISTS npy_output_02895.nested_data_types @@ -81,16 +80,16 @@ ${CLICKHOUSE_CLIENT} -n -q --ignore-error " INSERT INTO npy_output_02895.nested_data_types VALUES ([[[1], [2]], [[3], [4]]], [[0.1], [0.2]], ['a', 'bb']), ([[[1], [2]], [[3], [4]]], [[0.1], [0.2]], ['ccc', 'dddd']); - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_int32.npy') SELECT i4 FROM npy_output_02895.nested_data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_float64.npy') SELECT f8 FROM npy_output_02895.nested_data_types; - INSERT INTO TABLE FUNCTION file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_string.npy') SELECT s FROM npy_output_02895.nested_data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_int32.npy') SELECT i4 FROM npy_output_02895.nested_data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_float64.npy') SELECT f8 FROM npy_output_02895.nested_data_types; + INSERT INTO TABLE FUNCTION file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_string.npy') SELECT s FROM npy_output_02895.nested_data_types; - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_int32.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_float64.npy'); - SELECT * FROM file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_string.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_int32.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_float64.npy'); - DESC file('${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_string.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_int32.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_float64.npy'); + SELECT * FROM file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_string.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_int32.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_float64.npy'); + DESC file('${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/02895_nested_dtype_string.npy'); SELECT '-- test exceptions --'; CREATE TABLE IF NOT EXISTS npy_output_02895.exceptions @@ -115,4 +114,4 @@ ${CLICKHOUSE_CLIENT} -n -q --ignore-error " DROP DATABASE IF EXISTS npy_output_02895;" -rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} +rm -rf ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME:?} diff --git a/tests/queries/0_stateless/02903_parameterized_view_explain_ast.reference b/tests/queries/0_stateless/02903_parameterized_view_explain_ast.reference index 6ee8d0c3d23..27222968b1a 100644 --- a/tests/queries/0_stateless/02903_parameterized_view_explain_ast.reference +++ b/tests/queries/0_stateless/02903_parameterized_view_explain_ast.reference @@ -1,4 +1,4 @@ -CreateQuery numbers_pv (children 2) +CreateQuery numbers_pv (children 2) Identifier numbers_pv SelectWithUnionQuery (children 1) ExpressionList (children 1) diff --git a/tests/queries/0_stateless/02908_many_requests_to_system_replicas.reference b/tests/queries/0_stateless/02908_many_requests_to_system_replicas.reference index f376bb87044..fdefd2e3466 100644 --- a/tests/queries/0_stateless/02908_many_requests_to_system_replicas.reference +++ b/tests/queries/0_stateless/02908_many_requests_to_system_replicas.reference @@ -1,10 +1,10 @@ -Creating 300 tables -900 1 1 -900 1 1 -900 1 1 -900 1 1 -Making 200 requests to system.replicas +Creating 50 tables +150 1 1 +150 1 1 +150 1 1 +150 1 1 +Making 100 requests to system.replicas Query system.replicas while waiting for other concurrent requests to finish 0 -900 +150 1 diff --git a/tests/queries/0_stateless/02908_many_requests_to_system_replicas.sh b/tests/queries/0_stateless/02908_many_requests_to_system_replicas.sh index a247c99a818..81ba59fc591 100755 --- a/tests/queries/0_stateless/02908_many_requests_to_system_replicas.sh +++ b/tests/queries/0_stateless/02908_many_requests_to_system_replicas.sh @@ -7,8 +7,8 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) set -e -NUM_TABLES=300 -CONCURRENCY=200 +NUM_TABLES=50 +CONCURRENCY=100 echo "Creating $NUM_TABLES tables" @@ -46,10 +46,10 @@ wait; # Check results with different max_block_size -$CLICKHOUSE_CLIENT -q 'SELECT count(), sum(total_replicas) >= 2700, sum(active_replicas) >= 2700 FROM system.replicas WHERE database=currentDatabase()' -$CLICKHOUSE_CLIENT -q 'SELECT count(), sum(total_replicas) >= 2700, sum(active_replicas) >= 2700 FROM system.replicas WHERE database=currentDatabase() SETTINGS max_block_size=1' -$CLICKHOUSE_CLIENT -q 'SELECT count(), sum(total_replicas) >= 2700, sum(active_replicas) >= 2700 FROM system.replicas WHERE database=currentDatabase() SETTINGS max_block_size=77' -$CLICKHOUSE_CLIENT -q 'SELECT count(), sum(total_replicas) >= 2700, sum(active_replicas) >= 2700 FROM system.replicas WHERE database=currentDatabase() SETTINGS max_block_size=11111' +$CLICKHOUSE_CLIENT -q 'SELECT count() as c, sum(total_replicas) >= 3*c, sum(active_replicas) >= 3*c FROM system.replicas WHERE database=currentDatabase()' +$CLICKHOUSE_CLIENT -q 'SELECT count() as c, sum(total_replicas) >= 3*c, sum(active_replicas) >= 3*c FROM system.replicas WHERE database=currentDatabase() SETTINGS max_block_size=1' +$CLICKHOUSE_CLIENT -q 'SELECT count() as c, sum(total_replicas) >= 3*c, sum(active_replicas) >= 3*c FROM system.replicas WHERE database=currentDatabase() SETTINGS max_block_size=77' +$CLICKHOUSE_CLIENT -q 'SELECT count() as c, sum(total_replicas) >= 3*c, sum(active_replicas) >= 3*c FROM system.replicas WHERE database=currentDatabase() SETTINGS max_block_size=11111' echo "Making $CONCURRENCY requests to system.replicas" @@ -70,8 +70,8 @@ wait; $CLICKHOUSE_CLIENT -nq " SYSTEM FLUSH LOGS; --- without optimisation there are ~350K zk requests -SELECT sum(ProfileEvents['ZooKeeperTransactions']) < 30000 +-- Check that number of ZK request is less then a half of (total replicas * concurrency) +SELECT sum(ProfileEvents['ZooKeeperTransactions']) < (${NUM_TABLES} * 3 * ${CONCURRENCY} / 2) FROM system.query_log WHERE current_database=currentDatabase() AND log_comment='02908_many_requests'; " diff --git a/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.reference b/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.reference index 5b6c14ca24f..352e02941c3 100644 --- a/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.reference +++ b/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.reference @@ -3,7 +3,7 @@ SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR (t1.x IS NULL AND t2.x IS NULL)) O 2 2 2 2 3 3 3 33 \N \N \N \N -SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.x IS NULL AND t1.y <=> t2.y AND t2.x IS NULL) ORDER BY t1.x NULLS LAST; +SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.x IS NULL AND t2.x IS NULL) OR t1.y <=> t2.y ORDER BY t1.x NULLS LAST; 1 42 4 42 2 2 2 2 3 3 3 33 @@ -12,14 +12,14 @@ SELECT * FROM t1 JOIN t2 ON (t1.x = t2.x OR t1.x IS NULL AND t2.x IS NULL) ORDER 2 2 2 2 3 3 3 33 \N \N \N \N -SELECT * FROM t1 JOIN t2 ON t1.x <=> t2.x AND (t1.x = t1.y OR t1.x IS NULL AND t1.y IS NULL) ORDER BY t1.x; +SELECT * FROM t1 JOIN t2 ON t1.x <=> t2.x AND ((t1.x = t1.y) OR t1.x IS NULL AND t1.y IS NULL) ORDER BY t1.x; 2 2 2 2 3 3 3 33 \N \N \N \N SELECT * FROM t1 JOIN t2 ON (t1.x = t2.x OR t1.x IS NULL AND t2.x IS NULL) AND t1.y <=> t2.y ORDER BY t1.x NULLS LAST; 2 2 2 2 \N \N \N \N -SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.y <=> t2.y OR (t1.x IS NULL AND t1.y IS NULL AND t2.x IS NULL AND t2.y IS NULL)) ORDER BY t1.x NULLS LAST; +SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.y <=> t2.y OR (t1.x IS NULL AND t2.x IS NULL) OR (t1.y IS NULL AND t2.y IS NULL)) ORDER BY t1.x NULLS LAST; 1 42 4 42 2 2 2 2 3 3 3 33 @@ -31,3 +31,36 @@ SELECT x = y OR (x IS NULL AND y IS NULL) FROM t1 ORDER BY x NULLS LAST; 1 1 1 +SELECT * FROM t1 JOIN t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( (t2.x IS NULL) AND (t1.x IS NULL) ) ORDER BY t1.x NULLS LAST; +2 2 2 2 +3 3 3 33 +\N \N \N \N +SELECT * FROM t1 JOIN t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( t2.x <> t1.x AND (t2.x IS NULL) AND (t1.x IS NULL) ) ORDER BY t1.x NULLS LAST; +2 2 2 2 +3 3 3 33 +\N \N \N \N +SELECT * FROM t1 JOIN t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( t2.x <> t1.x AND t2.x <> t1.x ) ORDER BY t1.x NULLS LAST; -- { serverError INVALID_JOIN_ON_EXPRESSION } +SELECT * FROM t1 JOIN t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( t2.x <> t1.x AND (t2.x IS NULL) AND (t2.x IS NULL) ) ORDER BY t1.x NULLS LAST; -- { serverError INVALID_JOIN_ON_EXPRESSION } +-- aliases defined in the join condition are valid +SELECT *, e, e2 FROM t1 FULL JOIN t2 ON ( ( ((t1.x == t2.x) AS e) AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( (t2.x IS NULL) AND (t1.x IS NULL) ) AS e2 ) ORDER BY t1.x NULLS LAST, t2.x NULLS LAST; +1 42 \N \N \N 0 +2 2 2 2 1 1 +3 3 3 33 1 1 +\N \N 4 42 \N 0 +\N \N \N \N \N 1 +SELECT *, e, e2 FROM t1 FULL JOIN t2 ON ( ( ((t1.x == t2.x) AS e) AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) AS e2 ) ORDER BY t1.x NULLS LAST, t2.x NULLS LAST; +1 42 \N \N \N 0 +2 2 2 2 1 1 +3 3 3 33 1 1 +\N \N 4 42 \N 0 +\N \N \N \N \N 0 +\N \N \N \N \N 0 +-- check for non-nullable columns for which `is null` is replaced with constant +SELECT * FROM t1n as t1 JOIN t2n as t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( (t2.x IS NULL) AND (t1.x IS NULL) ) ORDER BY t1.x NULLS LAST; +2 2 2 2 +3 3 3 33 +-- +0 +0 +2 +2 diff --git a/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.sql b/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.sql index 5458370db8c..ee5e1582015 100644 --- a/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.sql +++ b/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.sql @@ -1,5 +1,7 @@ DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; +DROP TABLE IF EXISTS t1n; +DROP TABLE IF EXISTS t2n; CREATE TABLE t1 (x Nullable(Int64), y Nullable(UInt64)) ENGINE = TinyLog; CREATE TABLE t2 (x Nullable(Int64), y Nullable(UInt64)) ENGINE = TinyLog; @@ -7,24 +9,65 @@ CREATE TABLE t2 (x Nullable(Int64), y Nullable(UInt64)) ENGINE = TinyLog; INSERT INTO t1 VALUES (1,42), (2,2), (3,3), (NULL,NULL); INSERT INTO t2 VALUES (NULL,NULL), (2,2), (3,33), (4,42); +CREATE TABLE t1n (x Int64, y UInt64) ENGINE = TinyLog; +CREATE TABLE t2n (x Int64, y UInt64) ENGINE = TinyLog; + +INSERT INTO t1n VALUES (1,42), (2,2), (3,3); +INSERT INTO t2n VALUES (2,2), (3,33), (4,42); + SET allow_experimental_analyzer = 1; -- { echoOn } SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR (t1.x IS NULL AND t2.x IS NULL)) ORDER BY t1.x NULLS LAST; -SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.x IS NULL AND t1.y <=> t2.y AND t2.x IS NULL) ORDER BY t1.x NULLS LAST; +SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.x IS NULL AND t2.x IS NULL) OR t1.y <=> t2.y ORDER BY t1.x NULLS LAST; SELECT * FROM t1 JOIN t2 ON (t1.x = t2.x OR t1.x IS NULL AND t2.x IS NULL) ORDER BY t1.x; -SELECT * FROM t1 JOIN t2 ON t1.x <=> t2.x AND (t1.x = t1.y OR t1.x IS NULL AND t1.y IS NULL) ORDER BY t1.x; +SELECT * FROM t1 JOIN t2 ON t1.x <=> t2.x AND ((t1.x = t1.y) OR t1.x IS NULL AND t1.y IS NULL) ORDER BY t1.x; SELECT * FROM t1 JOIN t2 ON (t1.x = t2.x OR t1.x IS NULL AND t2.x IS NULL) AND t1.y <=> t2.y ORDER BY t1.x NULLS LAST; -SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.y <=> t2.y OR (t1.x IS NULL AND t1.y IS NULL AND t2.x IS NULL AND t2.y IS NULL)) ORDER BY t1.x NULLS LAST; +SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.y <=> t2.y OR (t1.x IS NULL AND t2.x IS NULL) OR (t1.y IS NULL AND t2.y IS NULL)) ORDER BY t1.x NULLS LAST; SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR (t1.x IS NULL AND t2.x IS NULL)) AND (t1.y == t2.y OR (t1.y IS NULL AND t2.y IS NULL)) AND COALESCE(t1.x, 0) != 2 ORDER BY t1.x NULLS LAST; SELECT x = y OR (x IS NULL AND y IS NULL) FROM t1 ORDER BY x NULLS LAST; + +SELECT * FROM t1 JOIN t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( (t2.x IS NULL) AND (t1.x IS NULL) ) ORDER BY t1.x NULLS LAST; +SELECT * FROM t1 JOIN t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( t2.x <> t1.x AND (t2.x IS NULL) AND (t1.x IS NULL) ) ORDER BY t1.x NULLS LAST; +SELECT * FROM t1 JOIN t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( t2.x <> t1.x AND t2.x <> t1.x ) ORDER BY t1.x NULLS LAST; -- { serverError INVALID_JOIN_ON_EXPRESSION } +SELECT * FROM t1 JOIN t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( t2.x <> t1.x AND (t2.x IS NULL) AND (t2.x IS NULL) ) ORDER BY t1.x NULLS LAST; -- { serverError INVALID_JOIN_ON_EXPRESSION } + +-- aliases defined in the join condition are valid +SELECT *, e, e2 FROM t1 FULL JOIN t2 ON ( ( ((t1.x == t2.x) AS e) AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( (t2.x IS NULL) AND (t1.x IS NULL) ) AS e2 ) ORDER BY t1.x NULLS LAST, t2.x NULLS LAST; +SELECT *, e, e2 FROM t1 FULL JOIN t2 ON ( ( ((t1.x == t2.x) AS e) AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) AS e2 ) ORDER BY t1.x NULLS LAST, t2.x NULLS LAST; + +-- check for non-nullable columns for which `is null` is replaced with constant +SELECT * FROM t1n as t1 JOIN t2n as t2 ON (t1.x == t2.x AND ((t2.x IS NOT NULL) AND (t1.x IS NOT NULL)) ) OR ( (t2.x IS NULL) AND (t1.x IS NULL) ) ORDER BY t1.x NULLS LAST; + -- { echoOff } +SELECT '--'; + +-- IS NOT NULL and constants are optimized out +SELECT count() FROM ( EXPLAIN QUERY TREE + SELECT * FROM t1 JOIN t2 ON ( (t1.x = t2.x) AND (t1.x IS NOT NULL) AND true AND (t2.x IS NOT NULL) ) +) WHERE explain like '%CONSTANT%' OR explain ilike '%is%null%'; + +SELECT count() FROM ( EXPLAIN QUERY TREE + SELECT * FROM t1 JOIN t2 ON ( (t1.x = t2.x) AND true ) +) WHERE explain like '%CONSTANT%' OR explain ilike '%is%null%'; + +-- this is not optimized out +SELECT count() FROM ( EXPLAIN QUERY TREE + SELECT * FROM t1 JOIN t2 ON (t1.x <=> t2.x OR t1.x IS NULL AND t1.y <=> t2.y AND t2.x IS NULL) +) WHERE explain like '%CONSTANT%' OR explain ilike '%is%null%'; + +SELECT count() FROM ( EXPLAIN QUERY TREE + SELECT * FROM t1 JOIN t2 ON t1.x <=> t2.x AND (t1.x = t1.y OR t1.x IS NULL AND t1.y IS NULL) +) WHERE explain like '%CONSTANT%' OR explain ilike '%is%null%'; + DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; +DROP TABLE IF EXISTS t1n; +DROP TABLE IF EXISTS t2n; diff --git a/tests/queries/0_stateless/02912_ingestion_mv_deduplication.reference b/tests/queries/0_stateless/02912_ingestion_mv_deduplication.reference index 335b55f05c8..07deb7c2565 100644 --- a/tests/queries/0_stateless/02912_ingestion_mv_deduplication.reference +++ b/tests/queries/0_stateless/02912_ingestion_mv_deduplication.reference @@ -10,13 +10,14 @@ 2022-09-01 12:23:34 42 2023-09-01 12:23:34 42 -- MV -2022-09-01 12:00:00 42 +2022-09-01 12:00:00 84 +2023-09-01 12:00:00 42 -- Original issue with deduplicate_blocks_in_dependent_materialized_views = 1 AND max_insert_delayed_streams_for_parallel_write > 1 -- Landing 2022-09-01 12:23:34 42 2023-09-01 12:23:34 42 -- MV -2022-09-01 12:00:00 42 +2022-09-01 12:00:00 84 2023-09-01 12:00:00 42 -- Regression introduced in https://github.com/ClickHouse/ClickHouse/pull/54184 -- Landing (Agg/Replacing)MergeTree diff --git a/tests/queries/0_stateless/02912_ingestion_mv_deduplication.sql b/tests/queries/0_stateless/02912_ingestion_mv_deduplication.sql index f206f0d7775..a2378fd8f67 100644 --- a/tests/queries/0_stateless/02912_ingestion_mv_deduplication.sql +++ b/tests/queries/0_stateless/02912_ingestion_mv_deduplication.sql @@ -54,8 +54,9 @@ SELECT '-- Original issue with deduplicate_blocks_in_dependent_materialized_view - 1st insert works for landing and mv tables - 2nd insert gets first block 20220901 deduplicated and second one inserted in landing table - - 2nd insert is not inserting anything in mv table due to a bug computing blocks to be discarded + - 2nd insert is not inserting anything in mv table due to a bug computing blocks to be discarded, now that block is inserted because deduplicate_blocks_in_dependent_materialized_views=0 + Now it is fixed. */ SET deduplicate_blocks_in_dependent_materialized_views = 0, max_insert_delayed_streams_for_parallel_write = 1000; @@ -97,7 +98,7 @@ SELECT '-- Original issue with deduplicate_blocks_in_dependent_materialized_view This is what happens now: - 1st insert works for landing and mv tables - - 2nd insert gets first block 20220901 deduplicated and second one inserted for landing and mv tables + - 2nd insert gets first block 20220901 deduplicated for landing and both rows are inserted for mv tables */ SET deduplicate_blocks_in_dependent_materialized_views = 1, max_insert_delayed_streams_for_parallel_write = 1000; diff --git a/tests/queries/0_stateless/02918_parallel_replicas_custom_key_unavailable_replica.sql b/tests/queries/0_stateless/02918_parallel_replicas_custom_key_unavailable_replica.sql index b9bc6974c47..c36b5bebd58 100644 --- a/tests/queries/0_stateless/02918_parallel_replicas_custom_key_unavailable_replica.sql +++ b/tests/queries/0_stateless/02918_parallel_replicas_custom_key_unavailable_replica.sql @@ -6,6 +6,9 @@ INSERT INTO 02918_parallel_replicas SELECT toString(number), number % 4 FROM num SET prefer_localhost_replica=0; +--- if we try to query unavaialble replica, connection will be retried +--- but a warning log message will be printed out +SET send_logs_level='error'; -- { echoOn } SELECT y, count() FROM cluster(test_cluster_1_shard_3_replicas_1_unavailable, currentDatabase(), 02918_parallel_replicas) @@ -26,5 +29,6 @@ GROUP BY y ORDER BY y SETTINGS max_parallel_replicas=3, parallel_replicas_custom_key='cityHash64(y)', parallel_replicas_custom_key_filter_type='default'; -- { echoOff } +SET send_logs_level='warning'; DROP TABLE 02918_parallel_replicas; diff --git a/tests/queries/0_stateless/02931_file_cluster.sh b/tests/queries/0_stateless/02931_file_cluster.sh index 8566e2ab08e..ebd3792e1dc 100755 --- a/tests/queries/0_stateless/02931_file_cluster.sh +++ b/tests/queries/0_stateless/02931_file_cluster.sh @@ -4,8 +4,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - mkdir -p "${USER_FILES_PATH}"/"${CLICKHOUSE_TEST_UNIQUE_NAME}"/ for i in {1..10} diff --git a/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.reference b/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.reference index 3124698d218..f9b72ba9c6a 100644 --- a/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.reference +++ b/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.reference @@ -245,21 +245,21 @@ EXPLAIN SYNTAX (SELECT 2 * count(uint64) - sum(uint64) From test_table); SELECT (2 * count(uint64)) - sum(uint64) FROM test_table SELECT sum(float64 + 2) From test_table; -26.5 +26.875 SELECT sum(2 + float64) From test_table; -26.5 +26.875 SELECT sum(float64 - 2) From test_table; -6.5 +6.875 SELECT sum(2 - float64) From test_table; --6.5 +-6.875 SELECT sum(float64) + 2 * count(float64) From test_table; -26.5 +26.875 SELECT 2 * count(float64) + sum(float64) From test_table; -26.5 +26.875 SELECT sum(float64) - 2 * count(float64) From test_table; -6.5 +6.875 SELECT 2 * count(float64) - sum(float64) From test_table; --6.5 +-6.875 EXPLAIN SYNTAX (SELECT sum(float64 + 2) From test_table); SELECT sum(float64) + (2 * count(float64)) FROM test_table @@ -375,25 +375,25 @@ EXPLAIN SYNTAX (SELECT (2 * count(uint64) - sum(uint64)) + (3 * count(uint64) - SELECT ((2 * count(uint64)) - sum(uint64)) + ((3 * count(uint64)) - sum(uint64)) FROM test_table SELECT sum(float64 + 2) + sum(float64 + 3) From test_table; -58 +58.75 SELECT sum(float64 + 2) - sum(float64 + 3) From test_table; -5 SELECT sum(float64 - 2) + sum(float64 - 3) From test_table; -8 +8.75 SELECT sum(float64 - 2) - sum(float64 - 3) From test_table; 5 SELECT sum(2 - float64) - sum(3 - float64) From test_table; -5 SELECT (sum(float64) + 2 * count(float64)) + (sum(float64) + 3 * count(float64)) From test_table; -58 +58.75 SELECT (sum(float64) + 2 * count(float64)) - (sum(float64) + 3 * count(float64)) From test_table; -5 SELECT (sum(float64) - 2 * count(float64)) + (sum(float64) - 3 * count(float64)) From test_table; -8 +8.75 SELECT (sum(float64) - 2 * count(float64)) - (sum(float64) - 3 * count(float64)) From test_table; 5 SELECT (2 * count(float64) - sum(float64)) + (3 * count(float64) - sum(float64)) From test_table; --8 +-8.75 EXPLAIN SYNTAX (SELECT sum(float64 + 2) + sum(float64 + 3) From test_table); SELECT (sum(float64) + (2 * count(float64))) + (sum(float64) + (3 * count(float64))) FROM test_table diff --git a/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.sql b/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.sql index c7b0ff82442..94baee6f1ba 100644 --- a/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.sql +++ b/tests/queries/0_stateless/02931_rewrite_sum_column_and_constant.sql @@ -23,11 +23,12 @@ CREATE TABLE test_table decimal32 Decimal32(5), ) ENGINE=MergeTree ORDER BY uint64; -INSERT INTO test_table VALUES (1, 1.1, 1.11); -INSERT INTO test_table VALUES (2, 2.2, 2.22); -INSERT INTO test_table VALUES (3, 3.3, 3.33); -INSERT INTO test_table VALUES (4, 4.4, 4.44); -INSERT INTO test_table VALUES (5, 5.5, 5.55); +-- Use Float64 numbers divisible by 1/16 (or some other small power of two), so that their sum doesn't depend on summation order. +INSERT INTO test_table VALUES (1, 1.125, 1.11); +INSERT INTO test_table VALUES (2, 2.250, 2.22); +INSERT INTO test_table VALUES (3, 3.375, 3.33); +INSERT INTO test_table VALUES (4, 4.500, 4.44); +INSERT INTO test_table VALUES (5, 5.625, 5.55); -- { echoOn } SELECT sum(uint64 + 1 AS i) from test_table where i > 0; diff --git a/tests/queries/0_stateless/02932_refreshable_materialized_views.sh b/tests/queries/0_stateless/02932_refreshable_materialized_views.sh index 89942e25b67..9081035579d 100755 --- a/tests/queries/0_stateless/02932_refreshable_materialized_views.sh +++ b/tests/queries/0_stateless/02932_refreshable_materialized_views.sh @@ -2,8 +2,6 @@ # Tags: atomic-database CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh @@ -134,7 +132,7 @@ while [ "`$CLICKHOUSE_CLIENT -nq "select status, next_refresh_time from refreshe do sleep 0.1 done -sleep 1 + $CLICKHOUSE_CLIENT -nq " select '<14: waiting for next cycle>', view, status, remaining_dependencies, next_refresh_time from refreshes; truncate src; @@ -172,13 +170,13 @@ $CLICKHOUSE_CLIENT -nq " drop table b; create materialized view c refresh every 1 second (x Int64) engine Memory empty as select * from src; drop table src;" -while [ "`$CLICKHOUSE_CLIENT -nq "select last_refresh_result from refreshes -- $LINENO" | xargs`" != 'Exception' ] +while [ "`$CLICKHOUSE_CLIENT -nq "select last_refresh_result from refreshes where view = 'c' -- $LINENO" | xargs`" != 'Exception' ] do sleep 0.1 done # Check exception, create src, expect successful refresh. $CLICKHOUSE_CLIENT -nq " - select '<19: exception>', exception ilike '%UNKNOWN_TABLE%' from refreshes; + select '<19: exception>', exception ilike '%UNKNOWN_TABLE%' ? '1' : exception from refreshes where view = 'c'; create table src (x Int64) engine Memory as select 1; system refresh view c;" while [ "`$CLICKHOUSE_CLIENT -nq "select last_refresh_result from refreshes -- $LINENO" | xargs`" != 'Finished' ] @@ -224,22 +222,27 @@ done $CLICKHOUSE_CLIENT -nq " rename table e to f; select '<24: rename during refresh>', * from f; - select '<25: rename during refresh>', view, status from refreshes; + select '<25: rename during refresh>', view, status from refreshes where view = 'f'; alter table f modify refresh after 10 year;" -sleep 2 # make it likely that at least one row was processed + # Cancel. $CLICKHOUSE_CLIENT -nq " system cancel view f;" -while [ "`$CLICKHOUSE_CLIENT -nq "select last_refresh_result from refreshes -- $LINENO" | xargs`" != 'Cancelled' ] +while [ "`$CLICKHOUSE_CLIENT -nq "select last_refresh_result from refreshes where view = 'f' -- $LINENO" | xargs`" != 'Cancelled' ] do sleep 0.1 done + +while [ "`$CLICKHOUSE_CLIENT -nq "select status from refreshes where view = 'f' -- $LINENO" | xargs`" = 'Running' ] +do + sleep 0.1 +done + # Check that another refresh doesn't immediately start after the cancelled one. -sleep 1 $CLICKHOUSE_CLIENT -nq " - select '<27: cancelled>', view, status from refreshes; + select '<27: cancelled>', view, status from refreshes where view = 'f'; system refresh view f;" -while [ "`$CLICKHOUSE_CLIENT -nq "select status from refreshes -- $LINENO" | xargs`" != 'Running' ] +while [ "`$CLICKHOUSE_CLIENT -nq "select status from refreshes where view = 'f' -- $LINENO" | xargs`" != 'Running' ] do sleep 0.1 done diff --git a/tests/queries/0_stateless/02933_change_cache_setting_without_restart.sh b/tests/queries/0_stateless/02933_change_cache_setting_without_restart.sh index ddad7a1904b..2e5a538007c 100755 --- a/tests/queries/0_stateless/02933_change_cache_setting_without_restart.sh +++ b/tests/queries/0_stateless/02933_change_cache_setting_without_restart.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel, no-s3-storage +# Tags: no-fasttest, no-parallel, no-object-storage CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -8,7 +8,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) disk_name="s3_cache_02933" $CLICKHOUSE_CLIENT --query "DESCRIBE FILESYSTEM CACHE '${disk_name}'" -config_path=/etc/clickhouse-server/config.d/storage_conf.xml +config_path=${CLICKHOUSE_CONFIG_DIR}/config.d/storage_conf.xml config_path_tmp=$config_path.tmp cat $config_path \ diff --git a/tests/queries/0_stateless/02933_replicated_database_forbid_create_as_select.reference b/tests/queries/0_stateless/02933_replicated_database_forbid_create_as_select.reference index d00491fd7e5..6ed281c757a 100644 --- a/tests/queries/0_stateless/02933_replicated_database_forbid_create_as_select.reference +++ b/tests/queries/0_stateless/02933_replicated_database_forbid_create_as_select.reference @@ -1 +1,2 @@ 1 +1 diff --git a/tests/queries/0_stateless/02933_replicated_database_forbid_create_as_select.sh b/tests/queries/0_stateless/02933_replicated_database_forbid_create_as_select.sh index 8a6904b6bd7..15f169d880f 100755 --- a/tests/queries/0_stateless/02933_replicated_database_forbid_create_as_select.sh +++ b/tests/queries/0_stateless/02933_replicated_database_forbid_create_as_select.sh @@ -9,8 +9,17 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh ${CLICKHOUSE_CLIENT} --query "CREATE DATABASE ${CLICKHOUSE_DATABASE}_db engine = Replicated('/clickhouse/databases/${CLICKHOUSE_TEST_ZOOKEEPER_PREFIX}/${CLICKHOUSE_DATABASE}_db', '{shard}', '{replica}')" + # Non-replicated engines are allowed ${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.test (id UInt64) ENGINE = MergeTree() ORDER BY id AS SELECT 1" +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --query "CREATE MATERIALIZED VIEW ${CLICKHOUSE_DATABASE}_db.test_mv (id UInt64) ENGINE = MergeTree() ORDER BY id POPULATE AS SELECT 1" + # Replicated storafes are forbidden -${CLICKHOUSE_CLIENT} --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.test2 (id UInt64) ENGINE = ReplicatedMergeTree('/clickhouse/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/test2', '1') ORDER BY id AS SELECT 1" |& grep -cm1 "SUPPORT_IS_DISABLED" +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.test2 (id UInt64) ENGINE = ReplicatedMergeTree ORDER BY id AS SELECT 1" |& grep -cm1 "SUPPORT_IS_DISABLED" +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --query "CREATE MATERIALIZED VIEW ${CLICKHOUSE_DATABASE}_db.test_mv2 (id UInt64) ENGINE = ReplicatedMergeTree ORDER BY id POPULATE AS SELECT 1" |& grep -cm1 "SUPPORT_IS_DISABLED" + +# But it is allowed with the special setting +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.test2 (id UInt64) ENGINE = ReplicatedMergeTree ORDER BY id AS SELECT 1" --database_replicated_allow_heavy_create=1 +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --query "CREATE MATERIALIZED VIEW ${CLICKHOUSE_DATABASE}_db.test_mv2 (id UInt64) ENGINE = ReplicatedMergeTree ORDER BY id POPULATE AS SELECT 1" --database_replicated_allow_heavy_create=1 + ${CLICKHOUSE_CLIENT} --query "DROP DATABASE ${CLICKHOUSE_DATABASE}_db" diff --git a/tests/queries/0_stateless/02941_variant_type_1.reference b/tests/queries/0_stateless/02941_variant_type_1.reference index 8a6e77d4f6d..53e5a556821 100644 --- a/tests/queries/0_stateless/02941_variant_type_1.reference +++ b/tests/queries/0_stateless/02941_variant_type_1.reference @@ -91,42 +91,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 0 1 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -145,21 +145,21 @@ lc_str_2 [0] [0,1] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 @@ -256,42 +256,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N 0 -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -310,23 +310,23 @@ lc_str_2 [0] [] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 ----------------------------------------------------------------------------------------------------------- test3 insert @@ -421,42 +421,42 @@ lc_str_15 (0,0) (16,17) (0,0) -\N -\N -\N -\N +0 +0 +0 +0 4 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 10 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 16 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 5 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 11 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 17 -\N +0 [] [] [] @@ -475,23 +475,23 @@ lc_str_15 [] [] [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17] -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 6 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 12 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 18 ----------------------------------------------------------------------------------------------------------- MergeTree compact @@ -587,42 +587,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 0 1 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -641,21 +641,21 @@ lc_str_2 [0] [0,1] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 @@ -751,42 +751,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 0 1 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -805,21 +805,21 @@ lc_str_2 [0] [0,1] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 @@ -916,42 +916,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N 0 -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -970,23 +970,23 @@ lc_str_2 [0] [] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 ----------------------------------------------------------------------------------------------------------- test2 select @@ -1080,42 +1080,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N 0 -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -1134,23 +1134,23 @@ lc_str_2 [0] [] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 ----------------------------------------------------------------------------------------------------------- test3 insert @@ -1245,42 +1245,42 @@ lc_str_15 (0,0) (16,17) (0,0) -\N -\N -\N -\N +0 +0 +0 +0 4 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 10 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 16 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 5 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 11 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 17 -\N +0 [] [] [] @@ -1299,23 +1299,23 @@ lc_str_15 [] [] [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17] -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 6 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 12 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 18 ----------------------------------------------------------------------------------------------------------- test3 select @@ -1409,42 +1409,42 @@ lc_str_15 (0,0) (16,17) (0,0) -\N -\N -\N -\N +0 +0 +0 +0 4 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 10 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 16 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 5 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 11 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 17 -\N +0 [] [] [] @@ -1463,23 +1463,23 @@ lc_str_15 [] [] [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17] -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 6 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 12 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 18 ----------------------------------------------------------------------------------------------------------- MergeTree wide @@ -1575,42 +1575,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 0 1 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -1629,21 +1629,21 @@ lc_str_2 [0] [0,1] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 @@ -1739,42 +1739,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 0 1 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -1793,21 +1793,21 @@ lc_str_2 [0] [0,1] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 2 3 @@ -1904,42 +1904,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N 0 -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -1958,23 +1958,23 @@ lc_str_2 [0] [] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 ----------------------------------------------------------------------------------------------------------- test2 select @@ -2068,42 +2068,42 @@ lc_str_2 (0,0) (0,0) (0,0) -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N 0 -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 2 -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 -\N -\N -\N +0 +0 +0 [] [] [] @@ -2122,23 +2122,23 @@ lc_str_2 [0] [] [0,1,2] -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 1 -\N +0 3 ----------------------------------------------------------------------------------------------------------- test3 insert @@ -2233,42 +2233,42 @@ lc_str_15 (0,0) (16,17) (0,0) -\N -\N -\N -\N +0 +0 +0 +0 4 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 10 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 16 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 5 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 11 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 17 -\N +0 [] [] [] @@ -2287,23 +2287,23 @@ lc_str_15 [] [] [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17] -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 6 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 12 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 18 ----------------------------------------------------------------------------------------------------------- test3 select @@ -2397,42 +2397,42 @@ lc_str_15 (0,0) (16,17) (0,0) -\N -\N -\N -\N +0 +0 +0 +0 4 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 10 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 16 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 5 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 11 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 17 -\N +0 [] [] [] @@ -2451,22 +2451,22 @@ lc_str_15 [] [] [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17] -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 6 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 12 -\N -\N -\N -\N -\N +0 +0 +0 +0 +0 18 ----------------------------------------------------------------------------------------------------------- diff --git a/tests/queries/0_stateless/02941_variant_type_2.reference b/tests/queries/0_stateless/02941_variant_type_2.reference index 20a5176cb5e..1d9126aa230 100644 --- a/tests/queries/0_stateless/02941_variant_type_2.reference +++ b/tests/queries/0_stateless/02941_variant_type_2.reference @@ -6,9 +6,6 @@ test4 select 100000 100000 100000 -100000 -100000 -100000 MergeTree compact test4 insert test4 select @@ -17,18 +14,12 @@ test4 select 100000 100000 100000 -100000 -100000 -100000 test4 select 500000 100000 100000 100000 100000 -100000 -100000 -100000 MergeTree wide test4 insert test4 select @@ -37,15 +28,9 @@ test4 select 100000 100000 100000 -100000 -100000 -100000 test4 select 500000 100000 100000 100000 100000 -100000 -100000 -100000 diff --git a/tests/queries/0_stateless/02941_variant_type_2.sh b/tests/queries/0_stateless/02941_variant_type_2.sh index f43cd2bb0d6..8453bce98dc 100755 --- a/tests/queries/0_stateless/02941_variant_type_2.sh +++ b/tests/queries/0_stateless/02941_variant_type_2.sh @@ -33,13 +33,10 @@ select v.\`LowCardinality(String)\` from test format Null; select count() from test where isNotNull(v.\`LowCardinality(String)\`); select v.\`Tuple(a UInt32, b UInt32)\` from test format Null; select v.\`Tuple(a UInt32, b UInt32)\`.a from test format Null; -select count() from test where isNotNull(v.\`Tuple(a UInt32, b UInt32)\`.a); select v.\`Tuple(a UInt32, b UInt32)\`.b from test format Null; -select count() from test where isNotNull(v.\`Tuple(a UInt32, b UInt32)\`.b); select v.\`Array(UInt64)\` from test format Null; select count() from test where not empty(v.\`Array(UInt64)\`); -select v.\`Array(UInt64)\`.size0 from test format Null; -select count() from test where isNotNull(v.\`Array(UInt64)\`.size0);" +select v.\`Array(UInt64)\`.size0 from test format Null;" } function run() diff --git a/tests/queries/0_stateless/02941_variant_type_3.reference b/tests/queries/0_stateless/02941_variant_type_3.reference index 1ccdb3acdff..d28aa7a594b 100644 --- a/tests/queries/0_stateless/02941_variant_type_3.reference +++ b/tests/queries/0_stateless/02941_variant_type_3.reference @@ -6,9 +6,6 @@ test5 select 100000 100000 100000 -100000 -100000 -100000 MergeTree compact test5 insert test5 select @@ -17,18 +14,12 @@ test5 select 100000 100000 100000 -100000 -100000 -100000 test5 select 500000 100000 100000 100000 100000 -100000 -100000 -100000 MergeTree wide test5 insert test5 select @@ -37,15 +28,9 @@ test5 select 100000 100000 100000 -100000 -100000 -100000 test5 select 500000 100000 100000 100000 100000 -100000 -100000 -100000 diff --git a/tests/queries/0_stateless/02941_variant_type_3.sh b/tests/queries/0_stateless/02941_variant_type_3.sh index f4b2b304f56..990eb25b5be 100755 --- a/tests/queries/0_stateless/02941_variant_type_3.sh +++ b/tests/queries/0_stateless/02941_variant_type_3.sh @@ -35,13 +35,10 @@ select v.\`LowCardinality(String)\` from test format Null; select count() from test where isNotNull(v.\`LowCardinality(String)\`); select v.\`Tuple(a UInt32, b UInt32)\` from test format Null; select v.\`Tuple(a UInt32, b UInt32)\`.a from test format Null; -select count() from test where isNotNull(v.\`Tuple(a UInt32, b UInt32)\`.a); select v.\`Tuple(a UInt32, b UInt32)\`.b from test format Null; -select count() from test where isNotNull(v.\`Tuple(a UInt32, b UInt32)\`.b); select v.\`Array(UInt64)\` from test format Null; select count() from test where not empty(v.\`Array(UInt64)\`); -select v.\`Array(UInt64)\`.size0 from test format Null; -select count() from test where isNotNull(v.\`Array(UInt64)\`.size0);" +select v.\`Array(UInt64)\`.size0 from test format Null;" } function run() diff --git a/tests/queries/0_stateless/02941_variant_type_4.reference b/tests/queries/0_stateless/02941_variant_type_4.reference index e13d5820343..d1630b04347 100644 --- a/tests/queries/0_stateless/02941_variant_type_4.reference +++ b/tests/queries/0_stateless/02941_variant_type_4.reference @@ -6,9 +6,6 @@ test6 select 200000 200000 200000 -200000 -200000 -200000 ----------------------------------------------------------------------------------------------------------- MergeTree compact test6 insert @@ -18,9 +15,6 @@ test6 select 200000 200000 200000 -200000 -200000 -200000 ----------------------------------------------------------------------------------------------------------- test6 select 1000000 @@ -28,9 +22,6 @@ test6 select 200000 200000 200000 -200000 -200000 -200000 ----------------------------------------------------------------------------------------------------------- MergeTree wide test6 insert @@ -40,9 +31,6 @@ test6 select 200000 200000 200000 -200000 -200000 -200000 ----------------------------------------------------------------------------------------------------------- test6 select 1000000 @@ -50,7 +38,4 @@ test6 select 200000 200000 200000 -200000 -200000 -200000 ----------------------------------------------------------------------------------------------------------- diff --git a/tests/queries/0_stateless/02941_variant_type_4.sh b/tests/queries/0_stateless/02941_variant_type_4.sh index f9a16847864..b8f619694b0 100755 --- a/tests/queries/0_stateless/02941_variant_type_4.sh +++ b/tests/queries/0_stateless/02941_variant_type_4.sh @@ -29,13 +29,10 @@ function test6_select() select count() from test where isNotNull(v.\`LowCardinality(String)\`); select v.\`Tuple(a UInt32, b UInt32)\` from test format Null; select v.\`Tuple(a UInt32, b UInt32)\`.a from test format Null; - select count() from test where isNotNull(v.\`Tuple(a UInt32, b UInt32)\`.a); select v.\`Tuple(a UInt32, b UInt32)\`.b from test format Null; - select count() from test where isNotNull(v.\`Tuple(a UInt32, b UInt32)\`.b); select v.\`Array(UInt64)\` from test format Null; select count() from test where not empty(v.\`Array(UInt64)\`); - select v.\`Array(UInt64)\`.size0 from test format Null; - select count() from test where isNotNull(v.\`Array(UInt64)\`.size0);" + select v.\`Array(UInt64)\`.size0 from test format Null;" echo "-----------------------------------------------------------------------------------------------------------" } diff --git a/tests/queries/0_stateless/02944_dynamically_change_filesystem_cache_size.sh b/tests/queries/0_stateless/02944_dynamically_change_filesystem_cache_size.sh index 2e344a6b6e5..cb099bb59ae 100755 --- a/tests/queries/0_stateless/02944_dynamically_change_filesystem_cache_size.sh +++ b/tests/queries/0_stateless/02944_dynamically_change_filesystem_cache_size.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel, no-s3-storage, no-random-settings +# Tags: no-fasttest, no-parallel, no-object-storage, no-random-settings CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -24,7 +24,7 @@ $CLICKHOUSE_CLIENT --query "SELECT * FROM test FORMAT Null" $CLICKHOUSE_CLIENT --query "SELECT count() FROM system.filesystem_cache WHERE state = 'DOWNLOADED'" $CLICKHOUSE_CLIENT --query "SELECT sum(size) FROM system.filesystem_cache WHERE state = 'DOWNLOADED'" -config_path=/etc/clickhouse-server/config.d/storage_conf_02944.xml +config_path=${CLICKHOUSE_CONFIG_DIR}/config.d/storage_conf_02944.xml config_path_tmp=$config_path.tmp echo 'set max_size from 100 to 10' diff --git a/tests/queries/0_stateless/02944_variant_as_common_type.sql b/tests/queries/0_stateless/02944_variant_as_common_type.sql index 49ea5f2769c..b3b86427b06 100644 --- a/tests/queries/0_stateless/02944_variant_as_common_type.sql +++ b/tests/queries/0_stateless/02944_variant_as_common_type.sql @@ -71,4 +71,3 @@ select toTypeName(res), array([1, 2, 3], [[1, 2, 3]]) as res; select toTypeName(res), map('a', 1, 'b', 'str_1') as res; select toTypeName(res), map('a', 1, 'b', map('c', 2, 'd', 'str_1')) as res; select toTypeName(res), map('a', 1, 'b', [1, 2, 3], 'c', [[4, 5, 6]]) as res; - diff --git a/tests/queries/0_stateless/02950_dictionary_ssd_cache_short_circuit.sh b/tests/queries/0_stateless/02950_dictionary_ssd_cache_short_circuit.sh index a02bdd0a1d2..3d2fe5d664d 100755 --- a/tests/queries/0_stateless/02950_dictionary_ssd_cache_short_circuit.sh +++ b/tests/queries/0_stateless/02950_dictionary_ssd_cache_short_circuit.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - $CLICKHOUSE_CLIENT -n --query=" DROP DATABASE IF EXISTS 02950_database_for_ssd_cache_dictionary; CREATE DATABASE 02950_database_for_ssd_cache_dictionary; @@ -32,7 +30,7 @@ $CLICKHOUSE_CLIENT -n --query=" PRIMARY KEY id SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'source_table')) LIFETIME(MIN 1 MAX 1000) - LAYOUT(SSD_CACHE(BLOCK_SIZE 4096 FILE_SIZE 8192 PATH '$USER_FILES_PATH/0d')); + LAYOUT(SSD_CACHE(BLOCK_SIZE 4096 FILE_SIZE 8192 PATH '$CLICKHOUSE_USER_FILES/0d')); SELECT dictGetOrDefault('02950_database_for_ssd_cache_dictionary.ssd_cache_dictionary', ('v1', 'v2'), 0, (intDiv(1, id), intDiv(1, id))) FROM 02950_database_for_ssd_cache_dictionary.source_table; SELECT dictGetOrDefault('02950_database_for_ssd_cache_dictionary.ssd_cache_dictionary', 'v2', id+1, intDiv(NULL, id)) FROM 02950_database_for_ssd_cache_dictionary.source_table; diff --git a/tests/queries/0_stateless/02950_part_log_bytes_uncompressed.sql b/tests/queries/0_stateless/02950_part_log_bytes_uncompressed.sql index 248475ab84b..24425062116 100644 --- a/tests/queries/0_stateless/02950_part_log_bytes_uncompressed.sql +++ b/tests/queries/0_stateless/02950_part_log_bytes_uncompressed.sql @@ -1,3 +1,6 @@ +-- Tags: no-random-merge-tree-settings, no-random-settings +-- Because we compare part sizes, and they could be affected by index granularity and index compression settings. + CREATE TABLE part_log_bytes_uncompressed ( key UInt8, value UInt8 @@ -17,7 +20,8 @@ ALTER TABLE part_log_bytes_uncompressed DROP PART 'all_4_4_0' SETTINGS mutations SYSTEM FLUSH LOGS; -SELECT event_type, table, part_name, bytes_uncompressed > 0, size_in_bytes < bytes_uncompressed FROM system.part_log +SELECT event_type, table, part_name, bytes_uncompressed > 0, (bytes_uncompressed > 0 ? (size_in_bytes < bytes_uncompressed ? '1' : toString((size_in_bytes, bytes_uncompressed))) : '0') +FROM system.part_log WHERE event_date >= yesterday() AND database = currentDatabase() AND table = 'part_log_bytes_uncompressed' AND (event_type != 'RemovePart' OR part_name = 'all_4_4_0') -- ignore removal of other parts ORDER BY part_name, event_type; diff --git a/tests/queries/0_stateless/02956_rocksdb_bulk_sink.reference b/tests/queries/0_stateless/02956_rocksdb_bulk_sink.reference index 74c71827e6e..1f140df1d6b 100644 --- a/tests/queries/0_stateless/02956_rocksdb_bulk_sink.reference +++ b/tests/queries/0_stateless/02956_rocksdb_bulk_sink.reference @@ -3,7 +3,7 @@ 1000 1 1000 -2 +1 1000000 1000 0 999001 diff --git a/tests/queries/0_stateless/02956_rocksdb_bulk_sink.sh b/tests/queries/0_stateless/02956_rocksdb_bulk_sink.sh index 45e65b18e07..95c136584f0 100755 --- a/tests/queries/0_stateless/02956_rocksdb_bulk_sink.sh +++ b/tests/queries/0_stateless/02956_rocksdb_bulk_sink.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-ordinary-database, use-rocksdb +# Tags: no-ordinary-database, use-rocksdb, no-random-settings CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # Normal importing, as we only insert 1000 rows, so it should be in memtable ${CLICKHOUSE_CLIENT} --query "CREATE TABLE IF NOT EXISTS rocksdb_worm (key UInt64, value UInt64) ENGINE = EmbeddedRocksDB() PRIMARY KEY key SETTINGS optimize_for_bulk_insert = 0;" -${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers(1000);" +${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers(1000) SETTINGS optimize_trivial_insert_select = 1;" ${CLICKHOUSE_CLIENT} --query "SELECT sum(value) FROM system.rocksdb WHERE database = currentDatabase() AND table = 'rocksdb_worm' AND name = 'no.file.opens';" # should be 0 because all data is still in memtable ${CLICKHOUSE_CLIENT} --query "SELECT count() FROM rocksdb_worm;" @@ -16,33 +16,32 @@ ${CLICKHOUSE_CLIENT} --query "ALTER TABLE rocksdb_worm MODIFY SETTING optimize_f # Testing that key serialization is identical w. and w/o bulk sink ${CLICKHOUSE_CLIENT} --query "TRUNCATE TABLE rocksdb_worm;" -${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+2 FROM numbers(1000);" # should override previous keys +${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+2 FROM numbers(1000) SETTINGS optimize_trivial_insert_select = 1;" # should override previous keys ${CLICKHOUSE_CLIENT} --query "SELECT count() FROM rocksdb_worm WHERE value = key + 2;" # With bulk insertion, there is no memtable, so a small insert should create a new file ${CLICKHOUSE_CLIENT} --query "TRUNCATE TABLE rocksdb_worm;" -${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers(1000);" +${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers(1000) SETTINGS optimize_trivial_insert_select = 1;" ${CLICKHOUSE_CLIENT} --query "SELECT sum(value) FROM system.rocksdb WHERE database = currentDatabase() AND table = 'rocksdb_worm' AND name = 'no.file.opens';" # should be 1 ${CLICKHOUSE_CLIENT} --query "SELECT count() FROM rocksdb_worm;" # Testing insert with multiple sinks and fixed block size ${CLICKHOUSE_CLIENT} --query "TRUNCATE TABLE rocksdb_worm;" # Must set both max_threads and max_insert_threads to 2 to make sure there is only two sinks -${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers_mt(1000000) SETTINGS max_threads = 2, max_insert_threads = 2, max_block_size = 10000, min_insert_block_size_rows = 0, min_insert_block_size_bytes = 0, insert_deduplication_token = '';" -${CLICKHOUSE_CLIENT} --query "SELECT sum(value) FROM system.rocksdb WHERE database = currentDatabase() AND table = 'rocksdb_worm' AND name = 'no.file.opens';" # should be 2 because default bulk sink size is ~1M rows / SST file +${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers_mt(1000000) SETTINGS max_threads = 2, max_insert_threads = 2, max_block_size = 10000, min_insert_block_size_rows = 0, min_insert_block_size_bytes = 0, insert_deduplication_token = '', optimize_trivial_insert_select = 1;" +${CLICKHOUSE_CLIENT} --query "SELECT sum(value) IN (1, 2) FROM system.rocksdb WHERE database = currentDatabase() AND table = 'rocksdb_worm' AND name = 'no.file.opens';" # should be not more than 2 because default bulk sink size is ~1M rows / SST file. ${CLICKHOUSE_CLIENT} --query "SELECT count() FROM rocksdb_worm;" # Testing insert with duplicated keys ${CLICKHOUSE_CLIENT} --query "TRUNCATE TABLE rocksdb_worm;" -${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number % 1000, number+1 FROM numbers_mt(1000000) SETTINGS max_block_size = 100000, max_insert_threads = 1;" +${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number % 1000, number+1 FROM numbers_mt(1000000) SETTINGS max_block_size = 100000, max_insert_threads = 1, optimize_trivial_insert_select = 1;" ${CLICKHOUSE_CLIENT} --query "SELECT count() FROM rocksdb_worm;" ${CLICKHOUSE_CLIENT} --query "SELECT * FROM rocksdb_worm WHERE key = 0;" # should be the latest value - 999001 # Testing insert with multiple threads ${CLICKHOUSE_CLIENT} --query "TRUNCATE TABLE rocksdb_worm;" -${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers_mt(1000000)" & -${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers_mt(1000000)" & +${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers_mt(1000000) SETTINGS optimize_trivial_insert_select = 1" & +${CLICKHOUSE_CLIENT} --query "INSERT INTO rocksdb_worm SELECT number, number+1 FROM numbers_mt(1000000) SETTINGS optimize_trivial_insert_select = 1" & wait ${CLICKHOUSE_CLIENT} --query "SELECT count() FROM rocksdb_worm;" - diff --git a/tests/queries/0_stateless/02956_rocksdb_with_ttl.reference b/tests/queries/0_stateless/02956_rocksdb_with_ttl.reference new file mode 100644 index 00000000000..508f73c8d78 --- /dev/null +++ b/tests/queries/0_stateless/02956_rocksdb_with_ttl.reference @@ -0,0 +1,2 @@ +0 foo +0 diff --git a/tests/queries/0_stateless/02956_rocksdb_with_ttl.sql b/tests/queries/0_stateless/02956_rocksdb_with_ttl.sql new file mode 100644 index 00000000000..01efe19cf45 --- /dev/null +++ b/tests/queries/0_stateless/02956_rocksdb_with_ttl.sql @@ -0,0 +1,13 @@ +-- Tags: no-ordinary-database, use-rocksdb + +-- TTL = 2s +CREATE TABLE dict_with_ttl (key UInt64, value String) ENGINE = EmbeddedRocksDB(2) PRIMARY KEY (key); +INSERT INTO dict_with_ttl VALUES (0, 'foo'); +-- Data inserted correctly +SELECT * FROM dict_with_ttl; +-- If possible, we should test that even we execute OPTIMIZE TABLE, the data is still there if TTL is not expired yet +-- Nevertheless, query time is unpredictable with different builds, so we can't test it. So we only test that after 3s +-- we execute OPTIMIZE and the data should be gone. +SELECT sleep(3); +OPTIMIZE TABLE dict_with_ttl; +SELECT * FROM dict_with_ttl; diff --git a/tests/queries/0_stateless/02961_storage_config_volume_priority.sh b/tests/queries/0_stateless/02961_storage_config_volume_priority.sh index 4e085541a8d..145b921a750 100755 --- a/tests/queries/0_stateless/02961_storage_config_volume_priority.sh +++ b/tests/queries/0_stateless/02961_storage_config_volume_priority.sh @@ -15,7 +15,7 @@ WHERE policy_name = 'policy_02961' ORDER BY volume_priority ASC; " -config_path=/etc/clickhouse-server/config.d/storage_conf_02961.xml +config_path=${CLICKHOUSE_CONFIG_DIR}/config.d/storage_conf_02961.xml config_path_tmp=$config_path.tmp echo 'check non-unique values dont work' diff --git a/tests/queries/0_stateless/02962_system_sync_replica_lightweight_from_modifier.sh b/tests/queries/0_stateless/02962_system_sync_replica_lightweight_from_modifier.sh index b61be87411d..9ef271632d0 100755 --- a/tests/queries/0_stateless/02962_system_sync_replica_lightweight_from_modifier.sh +++ b/tests/queries/0_stateless/02962_system_sync_replica_lightweight_from_modifier.sh @@ -62,7 +62,7 @@ export -f sync_and_drop_replicas export -f optimize_thread export -f mutations_thread -TIMEOUT=60 +TIMEOUT=30 timeout $TIMEOUT bash -c insert_thread 2> /dev/null & timeout $TIMEOUT bash -c sync_and_drop_replicas 2> /dev/null & diff --git a/tests/queries/0_stateless/02963_test_flexible_disk_configuration.sql b/tests/queries/0_stateless/02963_test_flexible_disk_configuration.sql index 8f67cd7e030..7ebef866360 100644 --- a/tests/queries/0_stateless/02963_test_flexible_disk_configuration.sql +++ b/tests/queries/0_stateless/02963_test_flexible_disk_configuration.sql @@ -22,7 +22,7 @@ create table test (a Int32) engine = MergeTree() order by tuple() settings disk=disk(name='test2', type = object_storage, object_storage_type = s3, - metadata_storage_type = local, + metadata_type = local, endpoint = 'http://localhost:11111/test/common/', access_key_id = clickhouse, secret_access_key = clickhouse); @@ -32,7 +32,7 @@ create table test (a Int32) engine = MergeTree() order by tuple() settings disk=disk(name='test3', type = object_storage, object_storage_type = s3, - metadata_storage_type = local, + metadata_type = local, metadata_keep_free_space_bytes = 1024, endpoint = 'http://localhost:11111/test/common/', access_key_id = clickhouse, @@ -43,7 +43,7 @@ create table test (a Int32) engine = MergeTree() order by tuple() settings disk=disk(name='test4', type = object_storage, object_storage_type = s3, - metadata_storage_type = local, + metadata_type = local, metadata_keep_free_space_bytes = 0, endpoint = 'http://localhost:11111/test/common/', access_key_id = clickhouse, diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.reference b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.reference new file mode 100644 index 00000000000..e1bf9c27a81 --- /dev/null +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.reference @@ -0,0 +1,30 @@ + +simple join with analyzer +4200000 4200000 4200000 -1400000 +4200006 4200006 4200006 -1400002 +4200012 4200012 4200012 -1400004 +4200018 4200018 4200018 -1400006 +4200024 4200024 4200024 -1400008 +4200030 4200030 4200030 -1400010 +4200036 4200036 4200036 -1400012 +4200042 4200042 4200042 -1400014 +4200048 4200048 4200048 -1400016 +4200054 4200054 4200054 -1400018 + +simple (global) join with analyzer and parallel replicas +4200000 4200000 4200000 -1400000 +4200006 4200006 4200006 -1400002 +4200012 4200012 4200012 -1400004 +4200018 4200018 4200018 -1400006 +4200024 4200024 4200024 -1400008 +4200030 4200030 4200030 -1400010 +4200036 4200036 4200036 -1400012 +4200042 4200042 4200042 -1400014 +4200048 4200048 4200048 -1400016 +4200054 4200054 4200054 -1400018 +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value` FROM `default`.`num_2` AS `__table1` (stage: WithMergeableState) +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value` FROM `default`.`num_2` AS `__table1` (stage: WithMergeableState) + DefaultCoordinator: Coordination done +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` GLOBAL ALL INNER JOIN `_data_` AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` GLOBAL ALL INNER JOIN `_data_` AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) + DefaultCoordinator: Coordination done diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh new file mode 100755 index 00000000000..1089eb4051f --- /dev/null +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh @@ -0,0 +1,51 @@ +#!/usr/bin/env bash +# Tags: long, no-random-settings, no-random-merge-tree-settings + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +$CLICKHOUSE_CLIENT -nm -q " +drop table if exists num_1; +drop table if exists num_2; + +create table num_1 (key UInt64, value String) engine = MergeTree order by key; +create table num_2 (key UInt64, value Int64) engine = MergeTree order by key; + +insert into num_1 select number * 2, toString(number * 2) from numbers(1e7); +insert into num_2 select number * 3, -number from numbers(1.5e6); +" + +############## +echo +echo "simple join with analyzer" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1" + +############## +echo +echo "simple (global) join with analyzer and parallel replicas" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, +max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, send_logs_level='trace', +max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0" 2>&1 | +grep "executeQuery\|.*Coordinator: Coordination done" | +grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | +sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.reference b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.reference new file mode 100644 index 00000000000..297ec311f3e --- /dev/null +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.reference @@ -0,0 +1,57 @@ + +simple (local) join with analyzer and parallel replicas +4200000 4200000 4200000 -1400000 +4200006 4200006 4200006 -1400002 +4200012 4200012 4200012 -1400004 +4200018 4200018 4200018 -1400006 +4200024 4200024 4200024 -1400008 +4200030 4200030 4200030 -1400010 +4200036 4200036 4200036 -1400012 +4200042 4200042 4200042 -1400014 +4200048 4200048 4200048 -1400016 +4200054 4200054 4200054 -1400018 +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) + DefaultCoordinator: Coordination done + +simple (local) join with analyzer and parallel replicas and full sorting merge join +4200000 4200000 4200000 -1400000 +4200006 4200006 4200006 -1400002 +4200012 4200012 4200012 -1400004 +4200018 4200018 4200018 -1400006 +4200024 4200024 4200024 -1400008 +4200030 4200030 4200030 -1400010 +4200036 4200036 4200036 -1400012 +4200042 4200042 4200042 -1400014 +4200048 4200048 4200048 -1400016 +4200054 4200054 4200054 -1400018 +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) + WithOrderCoordinator: Coordination done + +nested join with analyzer +420000 420000 420000 -140000 +420042 420042 420042 -140014 +420084 420084 420084 -140028 +420126 420126 420126 -140042 +420168 420168 420168 -140056 +420210 420210 420210 -140070 +420252 420252 420252 -140084 +420294 420294 420294 -140098 +420336 420336 420336 -140112 +420378 420378 420378 -140126 + +nested join with analyzer and parallel replicas, both local +420000 420000 420000 -140000 +420042 420042 420042 -140014 +420084 420084 420084 -140028 +420126 420126 420126 -140042 +420168 420168 420168 -140056 +420210 420210 420210 -140070 +420252 420252 420252 -140084 +420294 420294 420294 -140098 +420336 420336 420336 -140112 +420378 420378 420378 -140126 +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4` ALL INNER JOIN (SELECT `__table6`.`number` * 7 AS `key` FROM numbers(100000.) AS `__table6`) AS `__table5` ON `__table4`.`key` = `__table5`.`key` SETTINGS parallel_replicas_prefer_local_join = 1) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(10000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4` ALL INNER JOIN (SELECT `__table6`.`number` * 7 AS `key` FROM numbers(100000.) AS `__table6`) AS `__table5` ON `__table4`.`key` = `__table5`.`key` SETTINGS parallel_replicas_prefer_local_join = 1) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(10000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) + WithOrderCoordinator: Coordination done diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh new file mode 100755 index 00000000000..7a0e2d9bfdb --- /dev/null +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh @@ -0,0 +1,103 @@ +#!/usr/bin/env bash +# Tags: long, no-random-settings, no-random-merge-tree-settings + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +$CLICKHOUSE_CLIENT -nm -q " +drop table if exists num_1; +drop table if exists num_2; + +create table num_1 (key UInt64, value String) engine = MergeTree order by key; +create table num_2 (key UInt64, value Int64) engine = MergeTree order by key; + +insert into num_1 select number * 2, toString(number * 2) from numbers(1e7); +insert into num_2 select number * 3, -number from numbers(1.5e6); +" + +############## +echo +echo "simple (local) join with analyzer and parallel replicas" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1, +allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1, send_logs_level='trace', +allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | +grep "executeQuery\|.*Coordinator: Coordination done" | +grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | +sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' + + +############## +echo +echo "simple (local) join with analyzer and parallel replicas and full sorting merge join" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', +allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', send_logs_level='trace', +allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | +grep "executeQuery\|.*Coordinator: Coordination done" | +grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | +sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' + + +############## +echo +echo "nested join with analyzer" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2 inner join + (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r +on l.key = r.key order by l.key limit 10 offset 10000 +SETTINGS allow_experimental_analyzer=1" + + +############## +echo +echo "nested join with analyzer and parallel replicas, both local" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2 inner join + (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r +on l.key = r.key order by l.key limit 10 offset 10000 +SETTINGS allow_experimental_analyzer=1, +allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2 inner join + (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r +on l.key = r.key order by l.key limit 10 offset 10000 +SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', send_logs_level='trace', +allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | +grep "executeQuery\|.*Coordinator: Coordination done" | +grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | +sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer.reference b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.reference similarity index 55% rename from tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer.reference rename to tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.reference index d7fa419aeab..c0485b817c4 100644 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer.reference +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.reference @@ -1,91 +1,4 @@ -simple join with analyzer -4200000 4200000 4200000 -1400000 -4200006 4200006 4200006 -1400002 -4200012 4200012 4200012 -1400004 -4200018 4200018 4200018 -1400006 -4200024 4200024 4200024 -1400008 -4200030 4200030 4200030 -1400010 -4200036 4200036 4200036 -1400012 -4200042 4200042 4200042 -1400014 -4200048 4200048 4200048 -1400016 -4200054 4200054 4200054 -1400018 - -simple (global) join with analyzer and parallel replicas -4200000 4200000 4200000 -1400000 -4200006 4200006 4200006 -1400002 -4200012 4200012 4200012 -1400004 -4200018 4200018 4200018 -1400006 -4200024 4200024 4200024 -1400008 -4200030 4200030 4200030 -1400010 -4200036 4200036 4200036 -1400012 -4200042 4200042 4200042 -1400014 -4200048 4200048 4200048 -1400016 -4200054 4200054 4200054 -1400018 -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value` FROM `default`.`num_2` AS `__table1` (stage: WithMergeableState) -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value` FROM `default`.`num_2` AS `__table1` (stage: WithMergeableState) - DefaultCoordinator: Coordination done -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` GLOBAL ALL INNER JOIN `_data_` AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` GLOBAL ALL INNER JOIN `_data_` AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) - DefaultCoordinator: Coordination done - -simple (local) join with analyzer and parallel replicas -4200000 4200000 4200000 -1400000 -4200006 4200006 4200006 -1400002 -4200012 4200012 4200012 -1400004 -4200018 4200018 4200018 -1400006 -4200024 4200024 4200024 -1400008 -4200030 4200030 4200030 -1400010 -4200036 4200036 4200036 -1400012 -4200042 4200042 4200042 -1400014 -4200048 4200048 4200048 -1400016 -4200054 4200054 4200054 -1400018 -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) - DefaultCoordinator: Coordination done - -simple (local) join with analyzer and parallel replicas and full sorting merge join -4200000 4200000 4200000 -1400000 -4200006 4200006 4200006 -1400002 -4200012 4200012 4200012 -1400004 -4200018 4200018 4200018 -1400006 -4200024 4200024 4200024 -1400008 -4200030 4200030 4200030 -1400010 -4200036 4200036 4200036 -1400012 -4200042 4200042 4200042 -1400014 -4200048 4200048 4200048 -1400016 -4200054 4200054 4200054 -1400018 -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) - WithOrderCoordinator: Coordination done - -nested join with analyzer -420000 420000 420000 -140000 -420042 420042 420042 -140014 -420084 420084 420084 -140028 -420126 420126 420126 -140042 -420168 420168 420168 -140056 -420210 420210 420210 -140070 -420252 420252 420252 -140084 -420294 420294 420294 -140098 -420336 420336 420336 -140112 -420378 420378 420378 -140126 - -nested join with analyzer and parallel replicas, both local -420000 420000 420000 -140000 -420042 420042 420042 -140014 -420084 420084 420084 -140028 -420126 420126 420126 -140042 -420168 420168 420168 -140056 -420210 420210 420210 -140070 -420252 420252 420252 -140084 -420294 420294 420294 -140098 -420336 420336 420336 -140112 -420378 420378 420378 -140126 -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4` ALL INNER JOIN (SELECT `__table6`.`number` * 7 AS `key` FROM numbers(100000.) AS `__table6`) AS `__table5` ON `__table4`.`key` = `__table5`.`key` SETTINGS parallel_replicas_prefer_local_join = 1) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(10000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4` ALL INNER JOIN (SELECT `__table6`.`number` * 7 AS `key` FROM numbers(100000.) AS `__table6`) AS `__table5` ON `__table4`.`key` = `__table5`.`key` SETTINGS parallel_replicas_prefer_local_join = 1) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(10000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) - WithOrderCoordinator: Coordination done - nested join with analyzer and parallel replicas, both global 420000 420000 420000 -140000 420042 420042 420042 -140014 diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.sh similarity index 58% rename from tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer.sh rename to tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.sh index 2840482da6d..e49a340ab67 100755 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer.sh +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.sh @@ -17,125 +17,6 @@ insert into num_1 select number * 2, toString(number * 2) from numbers(1e7); insert into num_2 select number * 3, -number from numbers(1.5e6); " -############## -echo -echo "simple join with analyzer" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2) r on l.key = r.key -order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1" - -############## -echo -echo "simple (global) join with analyzer and parallel replicas" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2) r on l.key = r.key -order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, -max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2) r on l.key = r.key -order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, send_logs_level='trace', -max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0" 2>&1 | -grep "executeQuery\|.*Coordinator: Coordination done" | -grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | -sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' - -############## -echo -echo "simple (local) join with analyzer and parallel replicas" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2) r on l.key = r.key -order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2) r on l.key = r.key -order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, send_logs_level='trace', -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | -grep "executeQuery\|.*Coordinator: Coordination done" | -grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | -sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' - - -############## -echo -echo "simple (local) join with analyzer and parallel replicas and full sorting merge join" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2) r on l.key = r.key -order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2) r on l.key = r.key -order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', send_logs_level='trace', -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | -grep "executeQuery\|.*Coordinator: Coordination done" | -grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | -sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' - - -############## -echo -echo "nested join with analyzer" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2 inner join - (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r -on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1" - - -############## -echo -echo "nested join with analyzer and parallel replicas, both local" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2 inner join - (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r -on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" - -$CLICKHOUSE_CLIENT -q " -select * from (select key, value from num_1) l -inner join (select key, value from num_2 inner join - (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r -on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', send_logs_level='trace', -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | -grep "executeQuery\|.*Coordinator: Coordination done" | -grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | -sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' - - ############## echo echo "nested join with analyzer and parallel replicas, both global" diff --git a/tests/queries/0_stateless/02967_prewhere_no_columns.reference b/tests/queries/0_stateless/02967_prewhere_no_columns.reference new file mode 100644 index 00000000000..df105254618 --- /dev/null +++ b/tests/queries/0_stateless/02967_prewhere_no_columns.reference @@ -0,0 +1,2 @@ +105 +105 diff --git a/tests/queries/0_stateless/02967_prewhere_no_columns.sql b/tests/queries/0_stateless/02967_prewhere_no_columns.sql new file mode 100644 index 00000000000..efcc952caa2 --- /dev/null +++ b/tests/queries/0_stateless/02967_prewhere_no_columns.sql @@ -0,0 +1,51 @@ +CREATE TABLE t_02967 +( + `key` Date, + `value` UInt16 +) +ENGINE = MergeTree +ORDER BY key +SETTINGS + index_granularity_bytes = 0 --8192 --, min_index_granularity_bytes = 2 + , index_granularity = 100 + , min_rows_for_wide_part = 0, min_bytes_for_wide_part = 0 +-- +-- , min_bytes_for_wide_part = 2 +AS SELECT + number, + repeat(toString(number), 5) +FROM numbers(105.); + + + +-- Check with newly inserted data part. It's in-memory structured are filled at insert time. +SELECT + count(ignore(*)) +FROM t_02967 +PREWHERE CAST(ignore() + 1 as UInt8) +GROUP BY + ignore(65535, *), + ignore(255, 256, *) +SETTINGS + --send_logs_level='test', + max_threads=1; + + + +-- Reload part form disk to check that in-meory structures where properly serilaized-deserialized +DETACH TABLE t_02967; +ATTACH TABLE t_02967; + + +SELECT + count(ignore(*)) +FROM t_02967 +PREWHERE CAST(ignore() + 1 as UInt8) +GROUP BY + ignore(65535, *), + ignore(255, 256, *) +SETTINGS + --send_logs_level='test', + max_threads=1; + +DROP TABLE t_02967; diff --git a/tests/queries/0_stateless/02968_file_log_multiple_read.sh b/tests/queries/0_stateless/02968_file_log_multiple_read.sh index 199893a9428..d9bae05270a 100755 --- a/tests/queries/0_stateless/02968_file_log_multiple_read.sh +++ b/tests/queries/0_stateless/02968_file_log_multiple_read.sh @@ -4,12 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -# Data preparation. -# Now we can get the user_files_path by use the table file function for trick. also we can get it by query as: -# "insert into function file('exist.txt', 'CSV', 'val1 char') values ('aaaa'); select _path from file('exist.txt', 'CSV', 'val1 char')" -user_files_path=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') - -logs_dir=${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME} +logs_dir=${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME} rm -rf ${logs_dir} diff --git a/tests/queries/0_stateless/02970_generate_series.reference b/tests/queries/0_stateless/02970_generate_series.reference index 6e6f3c81587..f8a8e1891b7 100644 --- a/tests/queries/0_stateless/02970_generate_series.reference +++ b/tests/queries/0_stateless/02970_generate_series.reference @@ -5,6 +5,7 @@ 501 50 17928 +17928 0 10 13 diff --git a/tests/queries/0_stateless/02970_generate_series.sql b/tests/queries/0_stateless/02970_generate_series.sql index 0844714b3a6..edae884a561 100644 --- a/tests/queries/0_stateless/02970_generate_series.sql +++ b/tests/queries/0_stateless/02970_generate_series.sql @@ -5,6 +5,7 @@ SELECT count() FROM generate_series(7, 77, 10); SELECT count() FROM generate_series(0, 1000, 2); SELECT count() FROM generate_series(0, 999, 20); SELECT sum(generate_series) FROM generate_series(4, 1008, 4) WHERE generate_series % 7 = 1; +SELECT sum(generate_series) FROM generate_series(4, 1008, 4) WHERE generate_series % 7 = 1 SETTINGS max_block_size = 71; SELECT * FROM generate_series(5, 4); SELECT * FROM generate_series(0, 0); diff --git a/tests/queries/0_stateless/02971_analyzer_remote_id.sh b/tests/queries/0_stateless/02971_analyzer_remote_id.sh index 463e4cc1f0c..ab3c5292529 100755 --- a/tests/queries/0_stateless/02971_analyzer_remote_id.sh +++ b/tests/queries/0_stateless/02971_analyzer_remote_id.sh @@ -1,15 +1,9 @@ #!/usr/bin/env bash -# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} --query="DROP DATABASE IF EXISTS test_02971" -${CLICKHOUSE_CLIENT} --query="CREATE DATABASE test_02971" - -${CLICKHOUSE_CLIENT} --query="CREATE TABLE test_02971.x ENGINE = MergeTree() ORDER BY number AS SELECT * FROM numbers(2)" -${CLICKHOUSE_LOCAL} --query="SELECT count() FROM remote('127.0.0.{2,3}', 'test_02971.x') SETTINGS allow_experimental_analyzer = 1" 2>&1 \ +${CLICKHOUSE_CLIENT} --query="CREATE TABLE ${CLICKHOUSE_DATABASE}.x ENGINE = MergeTree() ORDER BY number AS SELECT * FROM numbers(2)" +${CLICKHOUSE_LOCAL} --query="SELECT count() FROM remote('127.0.0.{2,3}', '${CLICKHOUSE_DATABASE}.x') SETTINGS allow_experimental_analyzer = 1" 2>&1 \ | grep -av "ASan doesn't fully support makecontext/swapcontext functions" - -${CLICKHOUSE_CLIENT} --query="DROP DATABASE IF EXISTS test_02971" diff --git a/tests/queries/0_stateless/02971_functions_to_subcolumns_column_names.reference b/tests/queries/0_stateless/02971_functions_to_subcolumns_column_names.reference new file mode 100644 index 00000000000..3389ea44074 --- /dev/null +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_column_names.reference @@ -0,0 +1,5 @@ +SELECT + __table1.`arr.size0` AS `length(arr)`, + __table1.`n.null` AS `isNull(n)` +FROM default.t_column_names AS __table1 +{"length(arr)":"3","isNull(n)":0} diff --git a/tests/queries/0_stateless/02971_functions_to_subcolumns_column_names.sql b/tests/queries/0_stateless/02971_functions_to_subcolumns_column_names.sql new file mode 100644 index 00000000000..48e5232d18b --- /dev/null +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_column_names.sql @@ -0,0 +1,13 @@ +DROP TABLE IF EXISTS t_column_names; + +CREATE TABLE t_column_names (arr Array(UInt64), n Nullable(String)) ENGINE = Memory; + +INSERT INTO t_column_names VALUES ([1, 2, 3], 'foo'); + +SET optimize_functions_to_subcolumns = 1; +SET allow_experimental_analyzer = 1; + +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT length(arr), isNull(n) FROM t_column_names; +SELECT length(arr), isNull(n) FROM t_column_names FORMAT JSONEachRow; + +DROP TABLE t_column_names; diff --git a/tests/queries/0_stateless/02971_functions_to_subcolumns_map.reference b/tests/queries/0_stateless/02971_functions_to_subcolumns_map.reference new file mode 100644 index 00000000000..9488291c8ff --- /dev/null +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_map.reference @@ -0,0 +1,12 @@ +SELECT __table1.`m.size0` AS `length(m)` +FROM default.t_func_to_subcolumns_map AS __table1 +2 +1 +SELECT __table1.`m.size0` = 0 AS `empty(m)` +FROM default.t_func_to_subcolumns_map AS __table1 +0 +0 +SELECT __table1.`m.size0` != 0 AS `notEmpty(m)` +FROM default.t_func_to_subcolumns_map AS __table1 +1 +1 diff --git a/tests/queries/0_stateless/02971_functions_to_subcolumns_map.sql b/tests/queries/0_stateless/02971_functions_to_subcolumns_map.sql new file mode 100644 index 00000000000..e8a752a82d5 --- /dev/null +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_map.sql @@ -0,0 +1,19 @@ +DROP TABLE IF EXISTS t_func_to_subcolumns_map; + +CREATE TABLE t_func_to_subcolumns_map (id UInt64, m Map(String, UInt64)) ENGINE = MergeTree ORDER BY id; + +INSERT INTO t_func_to_subcolumns_map VALUES (1, map('aaa', 1, 'bbb', 2)) (2, map('ccc', 3)); + +SET optimize_functions_to_subcolumns = 1; +SET allow_experimental_analyzer = 1; + +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT length(m) FROM t_func_to_subcolumns_map; +SELECT length(m) FROM t_func_to_subcolumns_map; + +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT empty(m) FROM t_func_to_subcolumns_map; +SELECT empty(m) FROM t_func_to_subcolumns_map; + +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT notEmpty(m) FROM t_func_to_subcolumns_map; +SELECT notEmpty(m) FROM t_func_to_subcolumns_map; + +DROP TABLE t_func_to_subcolumns_map; diff --git a/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.reference b/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.reference new file mode 100644 index 00000000000..04616738a15 --- /dev/null +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.reference @@ -0,0 +1,4 @@ +SELECT __table1.`v.String` AS `variantElement(v, \'String\')` +FROM default.t_func_to_subcolumns_variant AS __table1 +foo +\N diff --git a/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.sql b/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.sql new file mode 100644 index 00000000000..511bcc44514 --- /dev/null +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.sql @@ -0,0 +1,15 @@ +DROP TABLE IF EXISTS t_func_to_subcolumns_variant; + +SET allow_experimental_variant_type = 1; + +CREATE TABLE t_func_to_subcolumns_variant (id UInt64, v Variant(String, UInt64)) ENGINE = MergeTree ORDER BY id; + +INSERT INTO t_func_to_subcolumns_variant VALUES (1, 'foo') (2, 111); + +SET optimize_functions_to_subcolumns = 1; +SET allow_experimental_analyzer = 1; + +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT variantElement(v, 'String') FROM t_func_to_subcolumns_variant; +SELECT variantElement(v, 'String') FROM t_func_to_subcolumns_variant; + +DROP TABLE t_func_to_subcolumns_variant; diff --git a/tests/queries/0_stateless/02973_parse_crlf_with_tsv_files.sh b/tests/queries/0_stateless/02973_parse_crlf_with_tsv_files.sh index 14f28f1ba4a..6edac86be5b 100755 --- a/tests/queries/0_stateless/02973_parse_crlf_with_tsv_files.sh +++ b/tests/queries/0_stateless/02973_parse_crlf_with_tsv_files.sh @@ -4,8 +4,6 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -# Data preparation step -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') UNIX_ENDINGS="${CLICKHOUSE_TEST_UNIQUE_NAME}_data_without_crlf.tsv" DOS_ENDINGS="${CLICKHOUSE_TEST_UNIQUE_NAME}_data_with_crlf.tsv" DATA_FILE_UNIX_ENDINGS="${USER_FILES_PATH:?}/${UNIX_ENDINGS}" diff --git a/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_MergeTree.reference b/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_MergeTree.reference index 531163e1d84..3135f2d01e1 100644 --- a/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_MergeTree.reference +++ b/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_MergeTree.reference @@ -3,28 +3,28 @@ data after ATTACH 1 Files before DETACH TABLE all_1_1_0 -backups/ordinary_default/data/ordinary_default/data/all_1_1_0: -primary.cidx -serialization.json -metadata_version.txt -default_compression_codec.txt +/backups/ordinary_default/data/ordinary_default/data/all_1_1_0: +checksums.txt +columns.txt +count.txt data.bin data.cmrk3 -count.txt -columns.txt -checksums.txt +default_compression_codec.txt +metadata_version.txt +primary.cidx +serialization.json Files after DETACH TABLE all_1_1_0 -backups/ordinary_default/data/ordinary_default/data/all_1_1_0: -primary.cidx -serialization.json -metadata_version.txt -default_compression_codec.txt +/backups/ordinary_default/data/ordinary_default/data/all_1_1_0: +checksums.txt +columns.txt +count.txt data.bin data.cmrk3 -count.txt -columns.txt -checksums.txt +default_compression_codec.txt +metadata_version.txt +primary.cidx +serialization.json diff --git a/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_MergeTree.sh b/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_MergeTree.sh index 12d08159012..d543f7195a9 100755 --- a/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_MergeTree.sh +++ b/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_MergeTree.sh @@ -49,11 +49,11 @@ path=$($CLICKHOUSE_CLIENT -q "SELECT replace(data_paths[1], 's3_plain', '') FROM path=${path%/} echo "Files before DETACH TABLE" -clickhouse-disks -C "$config" --disk s3_plain_disk list --recursive "${path:?}" | tail -n+2 +clickhouse-disks -C "$config" --disk s3_plain_disk --query "list --recursive $path" | tail -n+2 $CLICKHOUSE_CLIENT -q "detach table data" echo "Files after DETACH TABLE" -clickhouse-disks -C "$config" --disk s3_plain_disk list --recursive "$path" | tail -n+2 +clickhouse-disks -C "$config" --disk s3_plain_disk --query "list --recursive $path" | tail -n+2 # metadata file is left $CLICKHOUSE_CLIENT --force_remove_data_recursively_on_drop=1 -q "drop database if exists $CLICKHOUSE_DATABASE" diff --git a/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_ReplicatedMergeTree.reference b/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_ReplicatedMergeTree.reference index 1e191b719a5..a2dd196083e 100644 --- a/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_ReplicatedMergeTree.reference +++ b/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_ReplicatedMergeTree.reference @@ -3,28 +3,28 @@ data after ATTACH 1 Files before DETACH TABLE all_X_X_X -backups/ordinary_default/data/ordinary_default/data_read/all_X_X_X: -primary.cidx -serialization.json -metadata_version.txt -default_compression_codec.txt +/backups/ordinary_default/data/ordinary_default/data_read/all_X_X_X: +checksums.txt +columns.txt +count.txt data.bin data.cmrk3 -count.txt -columns.txt -checksums.txt +default_compression_codec.txt +metadata_version.txt +primary.cidx +serialization.json Files after DETACH TABLE all_X_X_X -backups/ordinary_default/data/ordinary_default/data_read/all_X_X_X: -primary.cidx -serialization.json -metadata_version.txt -default_compression_codec.txt +/backups/ordinary_default/data/ordinary_default/data_read/all_X_X_X: +checksums.txt +columns.txt +count.txt data.bin data.cmrk3 -count.txt -columns.txt -checksums.txt +default_compression_codec.txt +metadata_version.txt +primary.cidx +serialization.json diff --git a/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_ReplicatedMergeTree.sh b/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_ReplicatedMergeTree.sh index b079e67a000..eec05c81344 100755 --- a/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_ReplicatedMergeTree.sh +++ b/tests/queries/0_stateless/02980_s3_plain_DROP_TABLE_ReplicatedMergeTree.sh @@ -55,14 +55,14 @@ path=${path%/} echo "Files before DETACH TABLE" # sed to match any part, since in case of fault injection part name may not be all_0_0_0 but all_1_1_0 -clickhouse-disks -C "$config" --disk s3_plain_disk list --recursive "${path:?}" | tail -n+2 | sed 's/all_[^_]*_[^_]*_0/all_X_X_X/g' +clickhouse-disks -C "$config" --disk s3_plain_disk --query "list --recursive $path" | tail -n+2 | sed 's/all_[^_]*_[^_]*_0/all_X_X_X/g' $CLICKHOUSE_CLIENT -nm -q " detach table data_read; detach table data_write; " echo "Files after DETACH TABLE" -clickhouse-disks -C "$config" --disk s3_plain_disk list --recursive "$path" | tail -n+2 | sed 's/all_[^_]*_[^_]*_0/all_X_X_X/g' +clickhouse-disks -C "$config" --disk s3_plain_disk --query "list --recursive $path" | tail -n+2 | sed 's/all_[^_]*_[^_]*_0/all_X_X_X/g' # metadata file is left $CLICKHOUSE_CLIENT --force_remove_data_recursively_on_drop=1 -q "drop database if exists $CLICKHOUSE_DATABASE" diff --git a/tests/queries/0_stateless/02982_aggregation_states_destruction.reference b/tests/queries/0_stateless/02982_aggregation_states_destruction.reference index 72749c905a3..d00491fd7e5 100644 --- a/tests/queries/0_stateless/02982_aggregation_states_destruction.reference +++ b/tests/queries/0_stateless/02982_aggregation_states_destruction.reference @@ -1 +1 @@ -1 1 1 +1 diff --git a/tests/queries/0_stateless/02982_aggregation_states_destruction.sh b/tests/queries/0_stateless/02982_aggregation_states_destruction.sh index 263a4535c0e..84183606d48 100755 --- a/tests/queries/0_stateless/02982_aggregation_states_destruction.sh +++ b/tests/queries/0_stateless/02982_aggregation_states_destruction.sh @@ -11,4 +11,4 @@ $CLICKHOUSE_CLIENT --query_id $query_id --log_query_threads 1 --query="select nu $CLICKHOUSE_CLIENT -q "system flush logs;" -$CLICKHOUSE_CLIENT -q "select count() > 0, (countIf(thread_name = 'AggregDestruct') as aggs) > 0, aggs > 1 from system.query_thread_log where query_id = '$query_id' and current_database = currentDatabase();" +$CLICKHOUSE_CLIENT -q "select count() > 0 from system.query_thread_log where query_id = '$query_id' and current_database = currentDatabase() and thread_name = 'AggregDestruct';" diff --git a/tests/queries/0_stateless/02982_changeDate.reference b/tests/queries/0_stateless/02982_changeDate.reference new file mode 100644 index 00000000000..4a7f093ca2b --- /dev/null +++ b/tests/queries/0_stateless/02982_changeDate.reference @@ -0,0 +1,169 @@ +Negative tests +changeYear +-- Date +2001-01-01 +1970-01-01 +1970-01-01 +2149-06-06 +-- Date32 +2001-01-01 +1900-01-01 +1900-01-01 +2299-12-31 +-- DateTime +2001-01-01 11:22:33 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +2106-02-07 07:28:15 +-- DateTime64 +2001-01-01 11:22:33.4444 +1900-01-01 00:00:00.0000 +1900-01-01 00:00:00.0000 +2299-12-31 23:59:59.9999 +changeMonth +-- Date +2000-01-01 +2000-02-01 +2000-12-01 +1970-01-01 +1970-01-01 +1970-01-01 +-- Date32 +2000-01-01 +2000-02-01 +2000-12-01 +1900-01-01 +1900-01-01 +1900-01-01 +-- DateTime +2000-01-01 11:22:33 +2000-02-01 11:22:33 +2000-12-01 11:22:33 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +-- DateTime64 +2000-01-01 11:22:33.4444 +2000-02-01 11:22:33.4444 +2000-12-01 11:22:33.4444 +1900-01-01 00:00:00.0000 +1900-01-01 00:00:00.0000 +1900-01-01 00:00:00.0000 +changeDay +-- Date +2000-01-01 +2000-01-02 +2000-01-31 +1970-01-01 +1970-01-01 +1970-01-01 +-- Date32 +2000-01-01 +2000-01-02 +2000-01-31 +1900-01-01 +1900-01-01 +1900-01-01 +-- DateTime +2000-01-01 11:22:33 +2000-01-02 11:22:33 +2000-01-31 11:22:33 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +-- DateTime64 +2000-01-01 11:22:33.4444 +2000-01-02 11:22:33.4444 +2000-01-31 11:22:33.4444 +1900-01-01 00:00:00.0000 +1900-01-01 00:00:00.0000 +1900-01-01 00:00:00.0000 +-- Special case: change to 29 Feb in a leap year +2000-02-29 +2000-02-29 +2000-02-29 11:22:33 +2000-02-29 11:22:33.4444 +changeHour +-- Date +2000-01-01 00:00:00 +2000-01-01 02:00:00 +2000-01-01 23:00:00 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +-- Date32 +2000-01-01 00:00:00.000 +2000-01-01 02:00:00.000 +2000-01-01 23:00:00.000 +1900-01-01 00:00:00.000 +1900-01-01 00:00:00.000 +-- DateTime +2000-01-01 00:22:33 +2000-01-01 02:22:33 +2000-01-01 23:22:33 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +-- DateTime64 +2000-01-01 00:22:33.4444 +2000-01-01 02:22:33.4444 +2000-01-01 23:22:33.4444 +1900-01-01 00:00:00.0000 +1900-01-01 00:00:00.0000 +-- With different timezone +1970-01-01 07:00:00 +1970-01-01 07:00:00 +changeMinute +-- Date +2000-01-01 00:00:00 +2000-01-01 00:02:00 +2000-01-01 00:59:00 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +-- Date32 +2000-01-01 00:00:00.000 +2000-01-01 00:02:00.000 +2000-01-01 00:59:00.000 +1900-01-01 00:00:00.000 +1900-01-01 00:00:00.000 +-- DateTime +2000-01-01 11:00:33 +2000-01-01 11:02:33 +2000-01-01 11:59:33 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +-- DateTime64 +2000-01-01 11:00:33.4444 +2000-01-01 11:02:33.4444 +2000-01-01 11:59:33.4444 +1900-01-01 00:00:00.0000 +1900-01-01 00:00:00.0000 +-- With different timezone +1970-01-01 07:00:00 +1970-01-01 07:00:00 +changeSecond +-- Date +2000-01-01 00:00:00 +2000-01-01 00:00:02 +2000-01-01 00:00:59 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +-- Date32 +2000-01-01 00:00:00.000 +2000-01-01 00:00:02.000 +2000-01-01 00:00:59.000 +1900-01-01 00:00:00.000 +1900-01-01 00:00:00.000 +-- DateTime +2000-01-01 11:22:00 +2000-01-01 11:22:02 +2000-01-01 11:22:59 +1970-01-01 01:00:00 +1970-01-01 01:00:00 +-- DateTime64 +2000-01-01 11:22:00.4444 +2000-01-01 11:22:02.4444 +2000-01-01 11:22:59.4444 +1900-01-01 00:00:00.0000 +1900-01-01 00:00:00.0000 +-- With different timezone +1970-01-01 07:00:00 +1970-01-01 07:00:00 diff --git a/tests/queries/0_stateless/02982_changeDate.sql b/tests/queries/0_stateless/02982_changeDate.sql new file mode 100644 index 00000000000..2bc9aa95569 --- /dev/null +++ b/tests/queries/0_stateless/02982_changeDate.sql @@ -0,0 +1,185 @@ +SELECT 'Negative tests'; +-- as changeYear, changeMonth, changeDay, changeMinute, changeSecond share the same implementation, just testing one of them +SELECT changeYear(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT changeYear(toDate('2000-01-01')); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT changeYear(toDate('2000-01-01'), 2000, 1); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT changeYear(1999, 2000); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT changeYear(toDate('2000-01-01'), 'abc'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT changeYear(toDate('2000-01-01'), 1.5); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } + +-- Disable timezone randomization +SET session_timezone='CET'; + +SELECT 'changeYear'; +SELECT '-- Date'; +SELECT changeYear(toDate('2000-01-01'), 2001); +SELECT changeYear(toDate('2000-01-01'), 1800); -- out-of-bounds +SELECT changeYear(toDate('2000-01-01'), -5000); -- out-of-bounds +SELECT changeYear(toDate('2000-01-01'), 2500); -- out-of-bounds +SELECT '-- Date32'; +SELECT changeYear(toDate32('2000-01-01'), 2001); +SELECT changeYear(toDate32('2000-01-01'), 1800); -- out-of-bounds +SELECT changeYear(toDate32('2000-01-01'), -5000); -- out-of-bounds +SELECT changeYear(toDate32('2000-01-01'), 2500); -- out-of-bounds +SELECT '-- DateTime'; +SELECT changeYear(toDateTime('2000-01-01 11:22:33'), 2001); +SELECT changeYear(toDateTime('2000-01-01 11:22:33'), 1800); -- out-of-bounds +SELECT changeYear(toDateTime('2000-01-01 11:22:33'), -5000); -- out-of-bounds +SELECT changeYear(toDateTime('2000-01-01 11:22:33'), 2500); -- out-of-bounds +SELECT '-- DateTime64'; +SELECT changeYear(toDateTime64('2000-01-01 11:22:33.4444', 4), 2001); +SELECT changeYear(toDateTime64('2000-01-01 11:22:33.4444', 4), 1800); -- out-of-bounds +SELECT changeYear(toDateTime64('2000-01-01 11:22:33.4444', 4), -5000); -- out-of-bounds +SELECT changeYear(toDateTime64('2000-01-01 11:22:33.4444', 4), 2500); -- out-of-bounds + +SELECT 'changeMonth'; +SELECT '-- Date'; +SELECT changeMonth(toDate('2000-01-01'), 1); +SELECT changeMonth(toDate('2000-01-01'), 2); +SELECT changeMonth(toDate('2000-01-01'), 12); +SELECT changeMonth(toDate('2000-01-01'), 0); -- out-of-bounds +SELECT changeMonth(toDate('2000-01-01'), -1); -- out-of-bounds +SELECT changeMonth(toDate('2000-01-01'), 13); -- out-of-bounds +SELECT '-- Date32'; +SELECT changeMonth(toDate32('2000-01-01'), 1); +SELECT changeMonth(toDate32('2000-01-01'), 2); +SELECT changeMonth(toDate32('2000-01-01'), 12); +SELECT changeMonth(toDate32('2000-01-01'), 0); -- out-of-bounds +SELECT changeMonth(toDate32('2000-01-01'), -1); -- out-of-bounds +SELECT changeMonth(toDate32('2000-01-01'), 13); -- out-of-bounds +SELECT '-- DateTime'; +SELECT changeMonth(toDateTime('2000-01-01 11:22:33'), 1); +SELECT changeMonth(toDateTime('2000-01-01 11:22:33'), 2); +SELECT changeMonth(toDateTime('2000-01-01 11:22:33'), 12); +SELECT changeMonth(toDateTime('2000-01-01 11:22:33'), 0); -- out-of-bounds +SELECT changeMonth(toDateTime('2000-01-01 11:22:33'), -1); -- out-of-bounds +SELECT changeMonth(toDateTime('2000-01-01 11:22:33'), 13); -- out-of-bounds +SELECT '-- DateTime64'; +SELECT changeMonth(toDateTime64('2000-01-01 11:22:33.4444', 4), 1); +SELECT changeMonth(toDateTime64('2000-01-01 11:22:33.4444', 4), 2); +SELECT changeMonth(toDateTime64('2000-01-01 11:22:33.4444', 4), 12); +SELECT changeMonth(toDateTime64('2000-01-01 11:22:33.4444', 4), 0); -- out-of-bounds +SELECT changeMonth(toDateTime64('2000-01-01 11:22:33.4444', 4), -1); -- out-of-bounds +SELECT changeMonth(toDateTime64('2000-01-01 11:22:33.4444', 4), 13); -- out-of-bounds + +SELECT 'changeDay'; +SELECT '-- Date'; +SELECT changeDay(toDate('2000-01-01'), 1); +SELECT changeDay(toDate('2000-01-01'), 2); +SELECT changeDay(toDate('2000-01-01'), 31); +SELECT changeDay(toDate('2000-01-01'), 0); -- out-of-bounds +SELECT changeDay(toDate('2000-01-01'), -1); -- out-of-bounds +SELECT changeDay(toDate('2000-01-01'), 32); -- out-of-bounds +SELECT '-- Date32'; +SELECT changeDay(toDate32('2000-01-01'), 1); +SELECT changeDay(toDate32('2000-01-01'), 2); +SELECT changeDay(toDate32('2000-01-01'), 31); +SELECT changeDay(toDate32('2000-01-01'), 0); -- out-of-bounds +SELECT changeDay(toDate32('2000-01-01'), -1); -- out-of-bounds +SELECT changeDay(toDate32('2000-01-01'), 32); -- out-of-bounds +SELECT '-- DateTime'; +SELECT changeDay(toDateTime('2000-01-01 11:22:33'), 1); +SELECT changeDay(toDateTime('2000-01-01 11:22:33'), 2); +SELECT changeDay(toDateTime('2000-01-01 11:22:33'), 31); +SELECT changeDay(toDateTime('2000-01-01 11:22:33'), 0); -- out-of-bounds +SELECT changeDay(toDateTime('2000-01-01 11:22:33'), -1); -- out-of-bounds +SELECT changeDay(toDateTime('2000-01-01 11:22:33'), 32); -- out-of-bounds +SELECT '-- DateTime64'; +SELECT changeDay(toDateTime64('2000-01-01 11:22:33.4444', 4), 1); +SELECT changeDay(toDateTime64('2000-01-01 11:22:33.4444', 4), 2); +SELECT changeDay(toDateTime64('2000-01-01 11:22:33.4444', 4), 31); +SELECT changeDay(toDateTime64('2000-01-01 11:22:33.4444', 4), 0); -- out-of-bounds +SELECT changeDay(toDateTime64('2000-01-01 11:22:33.4444', 4), -1); -- out-of-bounds +SELECT changeDay(toDateTime64('2000-01-01 11:22:33.4444', 4), 32); -- out-of-bounds +SELECT '-- Special case: change to 29 Feb in a leap year'; +SELECT changeDay(toDate('2000-02-28'), 29); +SELECT changeDay(toDate32('2000-02-01'), 29); +SELECT changeDay(toDateTime('2000-02-01 11:22:33'), 29); +SELECT changeDay(toDateTime64('2000-02-01 11:22:33.4444', 4), 29); + +SELECT 'changeHour'; +SELECT '-- Date'; +SELECT changeHour(toDate('2000-01-01'), 0); +SELECT changeHour(toDate('2000-01-01'), 2); +SELECT changeHour(toDate('2000-01-01'), 23); +SELECT changeHour(toDate('2000-01-01'), -1); -- out-of-bounds +SELECT changeHour(toDate('2000-01-01'), 24); -- out-of-bounds +SELECT '-- Date32'; +SELECT changeHour(toDate32('2000-01-01'), 0); +SELECT changeHour(toDate32('2000-01-01'), 2); +SELECT changeHour(toDate32('2000-01-01'), 23); +SELECT changeHour(toDate32('2000-01-01'), -1); -- out-of-bounds +SELECT changeHour(toDate32('2000-01-01'), 24); -- out-of-bounds +SELECT '-- DateTime'; +SELECT changeHour(toDateTime('2000-01-01 11:22:33'), 0); +SELECT changeHour(toDateTime('2000-01-01 11:22:33'), 2); +SELECT changeHour(toDateTime('2000-01-01 11:22:33'), 23); +SELECT changeHour(toDateTime('2000-01-01 11:22:33'), -1); -- out-of-bounds +SELECT changeHour(toDateTime('2000-01-01 11:22:33'), 24); -- out-of-bounds +SELECT '-- DateTime64'; +SELECT changeHour(toDateTime64('2000-01-01 11:22:33.4444', 4), 0); +SELECT changeHour(toDateTime64('2000-01-01 11:22:33.4444', 4), 2); +SELECT changeHour(toDateTime64('2000-01-01 11:22:33.4444', 4), 23); +SELECT changeHour(toDateTime64('2000-01-01 11:22:33.4444', 4), -1); -- out-of-bounds +SELECT changeHour(toDateTime64('2000-01-01 11:22:33.4444', 4), 24); -- out-of-bounds +SELECT '-- With different timezone'; +SELECT changeHour(toDate('2000-01-01'), -1) SETTINGS session_timezone = 'Asia/Novosibirsk'; +SELECT changeHour(toDate('2000-01-01'), 24) SETTINGS session_timezone = 'Asia/Novosibirsk'; + +SELECT 'changeMinute'; +SELECT '-- Date'; +SELECT changeMinute(toDate('2000-01-01'), 0); +SELECT changeMinute(toDate('2000-01-01'), 2); +SELECT changeMinute(toDate('2000-01-01'), 59); +SELECT changeMinute(toDate('2000-01-01'), -1); -- out-of-bounds +SELECT changeMinute(toDate('2000-01-01'), 60); -- out-of-bounds +SELECT '-- Date32'; +SELECT changeMinute(toDate32('2000-01-01'), 0); +SELECT changeMinute(toDate32('2000-01-01'), 2); +SELECT changeMinute(toDate32('2000-01-01'), 59); +SELECT changeMinute(toDate32('2000-01-01'), -1); -- out-of-bounds +SELECT changeMinute(toDate32('2000-01-01'), 60); -- out-of-bounds +SELECT '-- DateTime'; +SELECT changeMinute(toDateTime('2000-01-01 11:22:33'), 0); +SELECT changeMinute(toDateTime('2000-01-01 11:22:33'), 2); +SELECT changeMinute(toDateTime('2000-01-01 11:22:33'), 59); +SELECT changeMinute(toDateTime('2000-01-01 11:22:33'), -1); -- out-of-bounds +SELECT changeMinute(toDateTime('2000-01-01 11:22:33'), 60); -- out-of-bounds +SELECT '-- DateTime64'; +SELECT changeMinute(toDateTime64('2000-01-01 11:22:33.4444', 4), 0); +SELECT changeMinute(toDateTime64('2000-01-01 11:22:33.4444', 4), 2); +SELECT changeMinute(toDateTime64('2000-01-01 11:22:33.4444', 4), 59); +SELECT changeMinute(toDateTime64('2000-01-01 11:22:33.4444', 4), -1); -- out-of-bounds +SELECT changeMinute(toDateTime64('2000-01-01 11:22:33.4444', 4), 60); -- out-of-bounds +SELECT '-- With different timezone'; +SELECT changeMinute(toDate('2000-01-01'), -1) SETTINGS session_timezone = 'Asia/Novosibirsk'; +SELECT changeMinute(toDate('2000-01-01'), 60) SETTINGS session_timezone = 'Asia/Novosibirsk'; + +SELECT 'changeSecond'; +SELECT '-- Date'; +SELECT changeSecond(toDate('2000-01-01'), 0); +SELECT changeSecond(toDate('2000-01-01'), 2); +SELECT changeSecond(toDate('2000-01-01'), 59); +SELECT changeSecond(toDate('2000-01-01'), -1); -- out-of-bounds +SELECT changeSecond(toDate('2000-01-01'), 60); -- out-of-bounds +SELECT '-- Date32'; +SELECT changeSecond(toDate32('2000-01-01'), 0); +SELECT changeSecond(toDate32('2000-01-01'), 2); +SELECT changeSecond(toDate32('2000-01-01'), 59); +SELECT changeSecond(toDate32('2000-01-01'), -1); -- out-of-bounds +SELECT changeSecond(toDate32('2000-01-01'), 60); -- out-of-bounds +SELECT '-- DateTime'; +SELECT changeSecond(toDateTime('2000-01-01 11:22:33'), 0); +SELECT changeSecond(toDateTime('2000-01-01 11:22:33'), 2); +SELECT changeSecond(toDateTime('2000-01-01 11:22:33'), 59); +SELECT changeSecond(toDateTime('2000-01-01 11:22:33'), -1); -- out-of-bounds +SELECT changeSecond(toDateTime('2000-01-01 11:22:33'), 60); -- out-of-bounds +SELECT '-- DateTime64'; +SELECT changeSecond(toDateTime64('2000-01-01 11:22:33.4444', 4), 0); +SELECT changeSecond(toDateTime64('2000-01-01 11:22:33.4444', 4), 2); +SELECT changeSecond(toDateTime64('2000-01-01 11:22:33.4444', 4), 59); +SELECT changeSecond(toDateTime64('2000-01-01 11:22:33.4444', 4), -1); -- out-of-bounds +SELECT changeSecond(toDateTime64('2000-01-01 11:22:33.4444', 4), 60); -- out-of-bounds +SELECT '-- With different timezone'; +SELECT changeSecond(toDate('2000-01-01'), -1) SETTINGS session_timezone = 'Asia/Novosibirsk'; +SELECT changeSecond(toDate('2000-01-01'), 60) SETTINGS session_timezone = 'Asia/Novosibirsk'; diff --git a/tests/queries/0_stateless/02984_form_format.sh b/tests/queries/0_stateless/02984_form_format.sh index ce5feb60130..471b48e0f68 100755 --- a/tests/queries/0_stateless/02984_form_format.sh +++ b/tests/queries/0_stateless/02984_form_format.sh @@ -5,8 +5,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -# Test setup -USER_FILES_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}') FILE_NAME="data.tmp" FORM_DATA="${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/${FILE_NAME}" mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ @@ -30,4 +28,4 @@ touch $FORM_DATA echo -ne "c.e=ls7xfkpm&c.tti.m=raf&rt.start=navigation&rt.bmr=390%2C11%2C10&rt.tstart=1707076768666&rt.bstart=1707076769091&rt.blstart=1707076769056&rt.end=1707076769078&t_resp=296&t_page=116&t_done=412&t_other=boomerang%7C6%2Cboomr_fb%7C425%2Cboomr_ld%7C390%2Cboomr_lat%7C35&rt.tt=2685&rt.obo=0&pt.fcp=407&nt_nav_st=1707076768666&nt_dns_st=1707076768683&nt_dns_end=1707076768684&nt_con_st=1707076768684&nt_con_end=1707076768850&nt_req_st=1707076768850&nt_res_st=1707076768962&nt_res_end=1707076768962&nt_domloading=1707076769040&nt_domint=1707076769066&nt_domcontloaded_st=1707076769067&nt_domcontloaded_end=1707076769068&nt_domcomp=1707076769069&nt_load_st=1707076769069&nt_load_end=1707076769078&nt_unload_st=1707076769040&nt_unload_end=1707076769041&nt_ssl_st=1707076768788&nt_enc_size=3209&nt_dec_size=10093&nt_trn_size=3940&nt_protocol=h2&nt_red_cnt=0&nt_nav_type=1&restiming=%7B%22https%3A%2F%2Fwww.basicrum.com%2F%22%3A%7B%22publications%2F%22%3A%226%2C88%2C88%2C54%2C54%2C3e%2Ci%2Ci%2Ch*12h5%2Ckb%2C5b8%22%2C%22assets%2Fjs%2F%22%3A%7B%22just-the-docs.js%22%3A%223am%2Ce%2Ce*12pc%2C_%2C8oj*20%22%2C%22boomerang-1.737.60.cutting-edge.min.js%22%3A%222au%2Cb%2Ca*1pu3%2C_%2C1m19*21*42%22%2C%22vendor%2Flunr.min.js%22%3A%223am%2Cd%2C8*16t2%2C_%2Cfym*20%22%7D%7D%7D&u=https%3A%2F%2Fwww.basicrum.com%2Fpublications%2F&r=https%3A%2F%2Fwww.basicrum.com%2Fcost-analyses%2F&v=1.737.60&sv=14&sm=p&rt.si=dd0c542f-7adf-4310-830a-6c0a3d157c90-s8cjr1&rt.ss=1707075325294&rt.sl=4&vis.st=visible&ua.plt=Linux%20x86_64&ua.vnd=&pid=8fftz949&n=1&c.t.fps=07*4*65*j*61&c.t.busy=2*4*0034&c.tti.vr=408&c.tti=408&c.b=2&c.f=60&c.f.d=2511&c.f.m=1&c.f.s=ls7xfl1h&dom.res=5&dom.doms=1&mem.lsln=0&mem.ssln=0&mem.lssz=2&mem.sssz=2&scr.xy=1920x1200&scr.bpp=24%2F24&scr.orn=0%2Flandscape-primary&cpu.cnc=16&dom.ln=114&dom.sz=10438&dom.ck=157&dom.img=0&dom.script=6&dom.script.ext=3&dom.iframe=0&dom.link=4&dom.link.css=1&sb=1" > $FORM_DATA $CLICKHOUSE_CLIENT -q "SELECT * FROM file('$FORM_DATA', Form) FORMAT Vertical" -rm $FORM_DATA \ No newline at end of file +rm $FORM_DATA diff --git a/tests/queries/0_stateless/02987_group_array_intersect.reference b/tests/queries/0_stateless/02987_group_array_intersect.reference index 7ec64a889f5..ec4d07742af 100644 --- a/tests/queries/0_stateless/02987_group_array_intersect.reference +++ b/tests/queries/0_stateless/02987_group_array_intersect.reference @@ -8,12 +8,12 @@ [1,4,5] [] [] -1000000 -999999 +100000 +99999 [9] ['a','c'] -1000000 -999999 +50000 +49999 ['1'] [] ['2023-01-01 00:00:00'] diff --git a/tests/queries/0_stateless/02987_group_array_intersect.sql b/tests/queries/0_stateless/02987_group_array_intersect.sql index 321e860b0a8..15acd0ca900 100644 --- a/tests/queries/0_stateless/02987_group_array_intersect.sql +++ b/tests/queries/0_stateless/02987_group_array_intersect.sql @@ -39,15 +39,15 @@ DROP TABLE test_numbers; DROP TABLE IF EXISTS test_big_numbers_sep; CREATE TABLE test_big_numbers_sep (a Array(Int64)) engine=MergeTree ORDER BY a; -INSERT INTO test_big_numbers_sep SELECT array(number) FROM numbers_mt(1000000); +INSERT INTO test_big_numbers_sep SELECT array(number) FROM numbers_mt(100000); SELECT groupArrayIntersect(*) FROM test_big_numbers_sep; DROP TABLE test_big_numbers_sep; DROP TABLE IF EXISTS test_big_numbers; CREATE TABLE test_big_numbers (a Array(Int64)) engine=MergeTree ORDER BY a; -INSERT INTO test_big_numbers SELECT range(1000000); +INSERT INTO test_big_numbers SELECT range(100000); SELECT length(groupArrayIntersect(*)) FROM test_big_numbers; -INSERT INTO test_big_numbers SELECT range(999999); +INSERT INTO test_big_numbers SELECT range(99999); SELECT length(groupArrayIntersect(*)) FROM test_big_numbers; INSERT INTO test_big_numbers VALUES ([9]); SELECT groupArrayIntersect(*) FROM test_big_numbers; @@ -63,9 +63,9 @@ DROP TABLE test_string; DROP TABLE IF EXISTS test_big_string; CREATE TABLE test_big_string (a Array(String)) engine=MergeTree ORDER BY a; -INSERT INTO test_big_string SELECT groupArray(toString(number)) FROM numbers_mt(1000000); +INSERT INTO test_big_string SELECT groupArray(toString(number)) FROM numbers_mt(50000); SELECT length(groupArrayIntersect(*)) FROM test_big_string; -INSERT INTO test_big_string SELECT groupArray(toString(number)) FROM numbers_mt(999999); +INSERT INTO test_big_string SELECT groupArray(toString(number)) FROM numbers_mt(49999); SELECT length(groupArrayIntersect(*)) FROM test_big_string; INSERT INTO test_big_string VALUES (['1']); SELECT groupArrayIntersect(*) FROM test_big_string; diff --git a/tests/queries/0_stateless/02992_all_columns_should_have_comment.sql b/tests/queries/0_stateless/02992_all_columns_should_have_comment.sql index 127c6fee07d..ad056384bfd 100644 --- a/tests/queries/0_stateless/02992_all_columns_should_have_comment.sql +++ b/tests/queries/0_stateless/02992_all_columns_should_have_comment.sql @@ -1,4 +1,8 @@ SYSTEM FLUSH LOGS; SELECT 'Column ' || name || ' from table ' || concat(database, '.', table) || ' should have a comment' FROM system.columns -WHERE (database = 'system') AND (comment = '') AND (table NOT ILIKE '%_log_%') AND (table NOT IN ('numbers', 'numbers_mt', 'one', 'generate_series', 'generateSeries', 'coverage_log')) AND (default_kind != 'ALIAS'); +WHERE (database = 'system') AND + (comment = '') AND + (table NOT ILIKE '%log%') AND + (table NOT IN ('numbers', 'numbers_mt', 'one', 'generate_series', 'generateSeries', 'coverage_log', 'filesystem_read_prefetches_log')) AND + (default_kind != 'ALIAS'); diff --git a/tests/queries/0_stateless/02992_analyzer_group_by_const.sql b/tests/queries/0_stateless/02992_analyzer_group_by_const.sql index ede6e0deed9..2a9e673d7bc 100644 --- a/tests/queries/0_stateless/02992_analyzer_group_by_const.sql +++ b/tests/queries/0_stateless/02992_analyzer_group_by_const.sql @@ -30,3 +30,46 @@ SELECT min(dummy) FROM remote('127.0.0.{2,3}', system.one) GROUP BY y; + +CREATE TABLE ttt (hr DateTime, ts DateTime) ENGINE=Memory +as select '2000-01-01' d, d; + +SELECT + count(), + now() AS c1 +FROM remote('127.0.0.{1,2}', currentDatabase(), ttt) +GROUP BY c1 FORMAT Null; + +SELECT + count(), + now() AS c1 +FROM remote('127.0.0.{3,2}', currentDatabase(), ttt) +GROUP BY c1 FORMAT Null; + +SELECT + count(), + now() AS c1 +FROM remote('127.0.0.{1,2}', currentDatabase(), ttt) +GROUP BY c1 + 1 FORMAT Null; + +SELECT + count(), + now() AS c1 +FROM remote('127.0.0.{3,2}', currentDatabase(), ttt) +GROUP BY c1 + 1 FORMAT Null; + +SELECT + count(), + tuple(nullIf(toDateTime(formatDateTime(hr, '%F %T', 'America/Los_Angeles'), 'America/Los_Angeles'), toDateTime(0))) as c1, + defaultValueOfArgumentType(toTimeZone(ts, 'America/Los_Angeles')) as c2, + formatDateTime(hr, '%F %T', 'America/Los_Angeles') as c3 +FROM remote('127.0.0.{1,2}', currentDatabase(), ttt) +GROUP BY c1, c2, c3 FORMAT Null; + +SELECT + count(), + tuple(nullIf(toDateTime(formatDateTime(hr, '%F %T', 'America/Los_Angeles'), 'America/Los_Angeles'), toDateTime(0))) as c1, + defaultValueOfArgumentType(toTimeZone(ts, 'America/Los_Angeles')) as c2, + formatDateTime(hr, '%F %T', 'America/Los_Angeles') as c3 +FROM remote('127.0.0.{3,2}', currentDatabase(), ttt) +GROUP BY c1, c2, c3 FORMAT Null; diff --git a/tests/queries/0_stateless/02993_lazy_index_loading.reference b/tests/queries/0_stateless/02993_lazy_index_loading.reference index 5bc329ae4eb..08f07a92815 100644 --- a/tests/queries/0_stateless/02993_lazy_index_loading.reference +++ b/tests/queries/0_stateless/02993_lazy_index_loading.reference @@ -1,4 +1,4 @@ -100000000 140000000 +100000000 100000000 0 0 1 100000000 100000000 diff --git a/tests/queries/0_stateless/02993_lazy_index_loading.sql b/tests/queries/0_stateless/02993_lazy_index_loading.sql index 7de4af9ef0e..ffb4b7547bf 100644 --- a/tests/queries/0_stateless/02993_lazy_index_loading.sql +++ b/tests/queries/0_stateless/02993_lazy_index_loading.sql @@ -1,6 +1,7 @@ DROP TABLE IF EXISTS test; CREATE TABLE test (s String) ENGINE = MergeTree ORDER BY s SETTINGS index_granularity = 1; +SET optimize_trivial_insert_select = 1; INSERT INTO test SELECT randomString(1000) FROM numbers(100000); SELECT round(primary_key_bytes_in_memory, -7), round(primary_key_bytes_in_memory_allocated, -7) FROM system.parts WHERE database = currentDatabase() AND table = 'test'; diff --git a/tests/queries/0_stateless/02995_index.reference b/tests/queries/0_stateless/02995_index.reference deleted file mode 100644 index 1e8639caa88..00000000000 --- a/tests/queries/0_stateless/02995_index.reference +++ /dev/null @@ -1,126 +0,0 @@ -12 4 21722 2209341 4 1415 2333 4 61 64 3 -21 1134 11363 58484 1106 1458 1592 136 26 62 32 -22 210 4504 5729 196 291 767 124 47 54 8 -26 196 1327684 5221 195 4140 5661 161 28 49 19 -28 5 2034378 7102 5 325 3255 2 53 60 4 -29 53 45041 45189 45 1580 211 31 55 84 18 -38 424 1600675 4653 424 562 5944 244 60 65 6 -45 17 62743 674873 17 6239 6494 17 65 76 8 -72 1862 1210073 6200 1677 2498 528 859 51 61 11 -79 2 2255228 2255293 2 5495 7057 2 65 65 1 -85 459 1051571 1829638 459 6402 7131 334 32 61 25 -86 10 1748130 1754217 10 4374 7003 10 56 59 4 -91 165 5718 5802 75 282 7113 112 41 63 22 -94 20 1231916 2050003 20 4802 4917 19 53 59 7 -99 2 3665 36667 2 497 697 2 70 71 2 -103 1 2446615 2446615 1 2498 2498 1 58 58 1 -106 72 6149 6699 67 527 826 40 61 61 1 -111 43 2273186 5272 43 492 4923 4 54 72 15 -120 3129 45117 6735 2868 1030 1625 561 59 64 6 -138 2 49243 49374 2 1428 1519 2 47 48 2 -143 100 23321 63639 100 1115 1624 88 51 51 1 -145 1 2447976 2447976 1 6173 6173 1 44 44 1 -153 16 13748 16881 16 1506 1636 16 54 68 9 -159 19952 1525336 7131 12957 1280 6163 2668 24 66 39 -171 5 15042 16698 5 1302 1608 5 65 65 1 -179 6264 1362341 2686 6244 2554 7132 2705 61 67 7 -192 1 1639623 1639623 1 3406 3406 1 32 32 1 -193 1 1429969 1429969 1 7131 7131 1 45 45 1 -207 12 23057 32500 12 1491 1726 12 32 46 7 -221 5081 1366870 6649 3432 4527 5226 687 24 69 39 -228 73 12281 17929 71 1328 2034 63 49 71 18 -229 2 1617478 1723791 2 4590 5578 2 41 42 2 -230 3916 1332729 6949 3668 1330 4703 845 62 65 4 -238 25 2624456 2625673 24 2535 6465 25 58 75 14 -241 154 2554929 2616444 154 2626 7131 148 34 57 17 -248 276 15529 30987 274 1040 1222 136 37 79 27 -254 3018 33966 6635 2837 1057 1622 539 24 60 33 -255 20 1581774 1811334 20 6068 6301 18 33 57 10 -256 5 5145 6841 5 367 376 5 58 58 1 -270 2 2195579 2262119 2 7102 7123 2 33 34 2 -281 32 2379460 616215 32 6042 6086 23 53 64 12 -282 7 1292651 24244 7 1607 2455 6 46 55 5 -286 123 1521935 5269 123 3793 3940 81 40 66 22 -291 21 2419080 3567 21 297 4731 21 54 55 2 -316 4 5221 5616 4 505 558 4 32 35 3 -319 232 56480 63033 230 1599 313 50 33 64 26 -327 15 51647 51894 14 1292 1585 14 47 57 7 -332 24 23484 54948 24 1609 1726 16 32 49 11 -333 1 14189 14189 1 1550 1550 1 63 63 1 -342 49 2579220 2622432 49 4626 6933 48 34 54 14 -344 1 6486 6486 1 509 509 1 24 24 1 -346 1987 53016 6735 1823 1334 174 294 26 62 32 -358 45 59058 60844 44 6746 722 40 57 84 15 -363 1198 1260033 2568811 1196 5710 5790 82 55 80 26 -384 150 2361175 476024 150 7008 7123 81 38 64 22 -387 277 5200 6553 252 243 521 130 65 65 1 -392 1877 1607428 2030850 1875 1416 7131 1379 54 66 13 -396 8181 1380803 6186 7920 545 798 1743 24 67 39 -398 3 5183 5213 2 291 352 3 53 59 3 -399 62 51494 59203 61 7073 754 42 55 78 18 -412 2141 1360120 2189792 2136 2491 5658 1371 71 75 5 -413 2 2036037 2064917 2 3963 4666 2 43 45 2 -431 33 2302331 2348449 33 4425 6516 32 69 69 1 -447 59 25125 33094 59 1176 1817 56 53 58 6 -456 1 53157 53157 1 1556 1556 1 26 26 1 -462 5 5456 6280 5 348 4337 5 28 40 5 -472 1 1443716 1443716 1 6122 6122 1 42 42 1 -491 34 1066102 1183673 34 6606 6822 32 46 67 15 -498 896 2230163 3054 895 537 7131 714 24 59 28 -504 108 12281 25180 108 1318 1784 94 55 66 12 -515 22 1588883 2640809 22 6554 6571 15 46 59 12 -518 1 37743 37743 1 1558 1558 1 72 72 1 -530 1 3033 3033 1 561 561 1 59 59 1 -532 26 5721 6355 25 549 665 14 44 50 7 -546 156 2577874 48517 156 1105 324 133 44 51 8 -554 12 1665194 2640066 12 1817 2951 12 57 57 1 -564 3865 2028049 2083433 3722 1115 985 2203 44 84 41 -566 4432 50605 57509 3217 1191 267 459 26 72 39 -567 8 5221 5893 7 333 558 8 27 35 4 -582 1172 1320619 2019743 1172 5819 7131 757 26 63 30 -584 43100 2500 5594 22561 134 4573 1660 48 84 37 -589 28 6046 6068 19 345 564 27 55 62 8 -595 139 1585165 1683606 138 2231 3598 132 54 84 28 -615 3 1056081 1116230 3 5794 5796 2 59 62 3 -619 7 1543114 5241 7 2442 3105 7 41 45 3 -634 2722 1221058 4999 2686 2426 7131 1735 54 60 7 -635 237 2119333 4667 237 561 5999 176 49 60 12 -644 5 1774169 2056171 5 5591 6091 4 33 39 3 -647 8 51632 64403 8 1457 1624 8 26 34 5 -651 1325 1620565 6281 1301 528 792 815 62 63 2 -665 13 4598 4789 13 511 558 11 39 46 7 -679 1560 1613200 25940 1552 1569 3118 781 49 84 35 -704 2 14226 15594 2 1086 1116 2 65 71 2 -715 25 1199352 3490 25 5036 5112 23 34 55 13 -716 1253 61989 6735 1050 1203 1625 397 52 65 14 -730 2584 5560 6170 634 2421 627 293 56 69 14 -736 8 1433153 4941 8 339 4594 8 28 36 5 -749 2 1326176 1339862 2 4339 6213 2 49 50 2 -753 1 53157 53157 1 1556 1556 1 26 26 1 -761 63 1443230 6881 63 3154 3204 26 56 73 14 -762 49 1449596 1968154 49 2437 3753 48 54 62 9 -775 35107 5330 769436 2471 447 6607 656 70 81 12 -789 1 1552458 1552458 1 2441 2441 1 62 62 1 -794 158 5585 6585 155 495 929 67 24 50 20 -839 9 29223 46530 9 1336 1465 9 52 52 1 -844 5 2377545 2377635 5 5129 6321 5 53 69 5 -846 50 2172273 2589295 50 1582 3053 48 64 68 5 -847 2577 56656 63658 1582 1444 838 474 26 63 33 -861 1333 5570 6909 839 457 489 37 33 70 34 -873 2360 1519811 50487 2248 1310 1784 316 60 68 9 -879 228 6704 6785 79 279 507 121 35 66 24 -889 5130 2070007 39692 5040 1151 6791 2606 44 66 23 -896 4 511246 859452 4 6554 6561 4 67 71 4 -912 146 1322641 2238040 146 1366 6354 143 59 59 1 -913 82 5495 6870 78 350 565 67 24 43 15 -921 763 1580790 416881 763 6191 7131 509 63 64 2 -925 318 2500952 5025 309 476 6114 182 32 56 21 -931 12 4277 4809 12 238 256 9 63 83 9 -942 954 1331 2228193 952 1121 5047 788 65 70 6 -948 14 1785593 2600431 14 6550 6598 13 34 49 9 -956 5 5755 6023 5 359 411 5 43 48 4 -963 4 3812 3835 4 444 537 4 47 53 4 -978 5 51632 58212 5 1127 1556 5 24 32 5 -980 53 47201 59744 53 1537 1625 36 41 49 9 -987 6033 2020131 763444 4306 256 792 1832 60 64 5 -993 4 1615159 1718339 4 1570 3093 4 62 63 2 diff --git a/tests/queries/0_stateless/02995_index_1.reference b/tests/queries/0_stateless/02995_index_1.reference new file mode 100644 index 00000000000..6c3b1230db6 --- /dev/null +++ b/tests/queries/0_stateless/02995_index_1.reference @@ -0,0 +1,15 @@ +12 4 21722 2209341 4 1415 2333 4 61 64 3 +21 1134 11363 58484 1106 1458 1592 136 26 62 32 +22 210 4504 5729 196 291 767 124 47 54 8 +26 196 1327684 5221 195 4140 5661 161 28 49 19 +28 5 2034378 7102 5 325 3255 2 53 60 4 +29 53 45041 45189 45 1580 211 31 55 84 18 +38 424 1600675 4653 424 562 5944 244 60 65 6 +45 17 62743 674873 17 6239 6494 17 65 76 8 +72 1862 1210073 6200 1677 2498 528 859 51 61 11 +79 2 2255228 2255293 2 5495 7057 2 65 65 1 +85 459 1051571 1829638 459 6402 7131 334 32 61 25 +86 10 1748130 1754217 10 4374 7003 10 56 59 4 +91 165 5718 5802 75 282 7113 112 41 63 22 +94 20 1231916 2050003 20 4802 4917 19 53 59 7 +99 2 3665 36667 2 497 697 2 70 71 2 diff --git a/tests/queries/0_stateless/02995_index.sh b/tests/queries/0_stateless/02995_index_1.sh similarity index 98% rename from tests/queries/0_stateless/02995_index.sh rename to tests/queries/0_stateless/02995_index_1.sh index 5125d03904e..a5f1b30c2e8 100755 --- a/tests/queries/0_stateless/02995_index.sh +++ b/tests/queries/0_stateless/02995_index_1.sh @@ -26,7 +26,7 @@ DETACH TABLE test; ATTACH TABLE test; " -for i in {1..1000} +for i in {1..100} do echo " WITH ${i} AS try diff --git a/tests/queries/0_stateless/02995_index_10.reference b/tests/queries/0_stateless/02995_index_10.reference new file mode 100644 index 00000000000..bfa38d03801 --- /dev/null +++ b/tests/queries/0_stateless/02995_index_10.reference @@ -0,0 +1,13 @@ +912 146 1322641 2238040 146 1366 6354 143 59 59 1 +913 82 5495 6870 78 350 565 67 24 43 15 +921 763 1580790 416881 763 6191 7131 509 63 64 2 +925 318 2500952 5025 309 476 6114 182 32 56 21 +931 12 4277 4809 12 238 256 9 63 83 9 +942 954 1331 2228193 952 1121 5047 788 65 70 6 +948 14 1785593 2600431 14 6550 6598 13 34 49 9 +956 5 5755 6023 5 359 411 5 43 48 4 +963 4 3812 3835 4 444 537 4 47 53 4 +978 5 51632 58212 5 1127 1556 5 24 32 5 +980 53 47201 59744 53 1537 1625 36 41 49 9 +987 6033 2020131 763444 4306 256 792 1832 60 64 5 +993 4 1615159 1718339 4 1570 3093 4 62 63 2 diff --git a/tests/queries/0_stateless/02995_index_10.sh b/tests/queries/0_stateless/02995_index_10.sh new file mode 100755 index 00000000000..d72c7c72705 --- /dev/null +++ b/tests/queries/0_stateless/02995_index_10.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {901..1000} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_2.reference b/tests/queries/0_stateless/02995_index_2.reference new file mode 100644 index 00000000000..098292a289f --- /dev/null +++ b/tests/queries/0_stateless/02995_index_2.reference @@ -0,0 +1,13 @@ +103 1 2446615 2446615 1 2498 2498 1 58 58 1 +106 72 6149 6699 67 527 826 40 61 61 1 +111 43 2273186 5272 43 492 4923 4 54 72 15 +120 3129 45117 6735 2868 1030 1625 561 59 64 6 +138 2 49243 49374 2 1428 1519 2 47 48 2 +143 100 23321 63639 100 1115 1624 88 51 51 1 +145 1 2447976 2447976 1 6173 6173 1 44 44 1 +153 16 13748 16881 16 1506 1636 16 54 68 9 +159 19952 1525336 7131 12957 1280 6163 2668 24 66 39 +171 5 15042 16698 5 1302 1608 5 65 65 1 +179 6264 1362341 2686 6244 2554 7132 2705 61 67 7 +192 1 1639623 1639623 1 3406 3406 1 32 32 1 +193 1 1429969 1429969 1 7131 7131 1 45 45 1 diff --git a/tests/queries/0_stateless/02995_index_2.sh b/tests/queries/0_stateless/02995_index_2.sh new file mode 100755 index 00000000000..e7451c7ee4b --- /dev/null +++ b/tests/queries/0_stateless/02995_index_2.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {101..200} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_3.reference b/tests/queries/0_stateless/02995_index_3.reference new file mode 100644 index 00000000000..9c2fca9fde6 --- /dev/null +++ b/tests/queries/0_stateless/02995_index_3.reference @@ -0,0 +1,16 @@ +207 12 23057 32500 12 1491 1726 12 32 46 7 +221 5081 1366870 6649 3432 4527 5226 687 24 69 39 +228 73 12281 17929 71 1328 2034 63 49 71 18 +229 2 1617478 1723791 2 4590 5578 2 41 42 2 +230 3916 1332729 6949 3668 1330 4703 845 62 65 4 +238 25 2624456 2625673 24 2535 6465 25 58 75 14 +241 154 2554929 2616444 154 2626 7131 148 34 57 17 +248 276 15529 30987 274 1040 1222 136 37 79 27 +254 3018 33966 6635 2837 1057 1622 539 24 60 33 +255 20 1581774 1811334 20 6068 6301 18 33 57 10 +256 5 5145 6841 5 367 376 5 58 58 1 +270 2 2195579 2262119 2 7102 7123 2 33 34 2 +281 32 2379460 616215 32 6042 6086 23 53 64 12 +282 7 1292651 24244 7 1607 2455 6 46 55 5 +286 123 1521935 5269 123 3793 3940 81 40 66 22 +291 21 2419080 3567 21 297 4731 21 54 55 2 diff --git a/tests/queries/0_stateless/02995_index_3.sh b/tests/queries/0_stateless/02995_index_3.sh new file mode 100755 index 00000000000..506429e2696 --- /dev/null +++ b/tests/queries/0_stateless/02995_index_3.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {201..300} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_4.reference b/tests/queries/0_stateless/02995_index_4.reference new file mode 100644 index 00000000000..deff7afaed3 --- /dev/null +++ b/tests/queries/0_stateless/02995_index_4.reference @@ -0,0 +1,16 @@ +316 4 5221 5616 4 505 558 4 32 35 3 +319 232 56480 63033 230 1599 313 50 33 64 26 +327 15 51647 51894 14 1292 1585 14 47 57 7 +332 24 23484 54948 24 1609 1726 16 32 49 11 +333 1 14189 14189 1 1550 1550 1 63 63 1 +342 49 2579220 2622432 49 4626 6933 48 34 54 14 +344 1 6486 6486 1 509 509 1 24 24 1 +346 1987 53016 6735 1823 1334 174 294 26 62 32 +358 45 59058 60844 44 6746 722 40 57 84 15 +363 1198 1260033 2568811 1196 5710 5790 82 55 80 26 +384 150 2361175 476024 150 7008 7123 81 38 64 22 +387 277 5200 6553 252 243 521 130 65 65 1 +392 1877 1607428 2030850 1875 1416 7131 1379 54 66 13 +396 8181 1380803 6186 7920 545 798 1743 24 67 39 +398 3 5183 5213 2 291 352 3 53 59 3 +399 62 51494 59203 61 7073 754 42 55 78 18 diff --git a/tests/queries/0_stateless/02995_index_4.sh b/tests/queries/0_stateless/02995_index_4.sh new file mode 100755 index 00000000000..1a0458728f9 --- /dev/null +++ b/tests/queries/0_stateless/02995_index_4.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {301..400} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_5.reference b/tests/queries/0_stateless/02995_index_5.reference new file mode 100644 index 00000000000..c5ab4d2417c --- /dev/null +++ b/tests/queries/0_stateless/02995_index_5.reference @@ -0,0 +1,9 @@ +412 2141 1360120 2189792 2136 2491 5658 1371 71 75 5 +413 2 2036037 2064917 2 3963 4666 2 43 45 2 +431 33 2302331 2348449 33 4425 6516 32 69 69 1 +447 59 25125 33094 59 1176 1817 56 53 58 6 +456 1 53157 53157 1 1556 1556 1 26 26 1 +462 5 5456 6280 5 348 4337 5 28 40 5 +472 1 1443716 1443716 1 6122 6122 1 42 42 1 +491 34 1066102 1183673 34 6606 6822 32 46 67 15 +498 896 2230163 3054 895 537 7131 714 24 59 28 diff --git a/tests/queries/0_stateless/02995_index_5.sh b/tests/queries/0_stateless/02995_index_5.sh new file mode 100755 index 00000000000..60c12a8146d --- /dev/null +++ b/tests/queries/0_stateless/02995_index_5.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {401..500} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_6.reference b/tests/queries/0_stateless/02995_index_6.reference new file mode 100644 index 00000000000..bac19179bb6 --- /dev/null +++ b/tests/queries/0_stateless/02995_index_6.reference @@ -0,0 +1,14 @@ +504 108 12281 25180 108 1318 1784 94 55 66 12 +515 22 1588883 2640809 22 6554 6571 15 46 59 12 +518 1 37743 37743 1 1558 1558 1 72 72 1 +530 1 3033 3033 1 561 561 1 59 59 1 +532 26 5721 6355 25 549 665 14 44 50 7 +546 156 2577874 48517 156 1105 324 133 44 51 8 +554 12 1665194 2640066 12 1817 2951 12 57 57 1 +564 3865 2028049 2083433 3722 1115 985 2203 44 84 41 +566 4432 50605 57509 3217 1191 267 459 26 72 39 +567 8 5221 5893 7 333 558 8 27 35 4 +582 1172 1320619 2019743 1172 5819 7131 757 26 63 30 +584 43100 2500 5594 22561 134 4573 1660 48 84 37 +589 28 6046 6068 19 345 564 27 55 62 8 +595 139 1585165 1683606 138 2231 3598 132 54 84 28 diff --git a/tests/queries/0_stateless/02995_index_6.sh b/tests/queries/0_stateless/02995_index_6.sh new file mode 100755 index 00000000000..4936f73f36b --- /dev/null +++ b/tests/queries/0_stateless/02995_index_6.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {501..600} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_7.reference b/tests/queries/0_stateless/02995_index_7.reference new file mode 100644 index 00000000000..4f1d28ab37e --- /dev/null +++ b/tests/queries/0_stateless/02995_index_7.reference @@ -0,0 +1,9 @@ +615 3 1056081 1116230 3 5794 5796 2 59 62 3 +619 7 1543114 5241 7 2442 3105 7 41 45 3 +634 2722 1221058 4999 2686 2426 7131 1735 54 60 7 +635 237 2119333 4667 237 561 5999 176 49 60 12 +644 5 1774169 2056171 5 5591 6091 4 33 39 3 +647 8 51632 64403 8 1457 1624 8 26 34 5 +651 1325 1620565 6281 1301 528 792 815 62 63 2 +665 13 4598 4789 13 511 558 11 39 46 7 +679 1560 1613200 25940 1552 1569 3118 781 49 84 35 diff --git a/tests/queries/0_stateless/02995_index_7.sh b/tests/queries/0_stateless/02995_index_7.sh new file mode 100755 index 00000000000..26be310abce --- /dev/null +++ b/tests/queries/0_stateless/02995_index_7.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {601..700} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_8.reference b/tests/queries/0_stateless/02995_index_8.reference new file mode 100644 index 00000000000..62fbfb2af9a --- /dev/null +++ b/tests/queries/0_stateless/02995_index_8.reference @@ -0,0 +1,12 @@ +704 2 14226 15594 2 1086 1116 2 65 71 2 +715 25 1199352 3490 25 5036 5112 23 34 55 13 +716 1253 61989 6735 1050 1203 1625 397 52 65 14 +730 2584 5560 6170 634 2421 627 293 56 69 14 +736 8 1433153 4941 8 339 4594 8 28 36 5 +749 2 1326176 1339862 2 4339 6213 2 49 50 2 +753 1 53157 53157 1 1556 1556 1 26 26 1 +761 63 1443230 6881 63 3154 3204 26 56 73 14 +762 49 1449596 1968154 49 2437 3753 48 54 62 9 +775 35107 5330 769436 2471 447 6607 656 70 81 12 +789 1 1552458 1552458 1 2441 2441 1 62 62 1 +794 158 5585 6585 155 495 929 67 24 50 20 diff --git a/tests/queries/0_stateless/02995_index_8.sh b/tests/queries/0_stateless/02995_index_8.sh new file mode 100755 index 00000000000..8c2620b59fd --- /dev/null +++ b/tests/queries/0_stateless/02995_index_8.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {701..800} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_9.reference b/tests/queries/0_stateless/02995_index_9.reference new file mode 100644 index 00000000000..86c7be173bd --- /dev/null +++ b/tests/queries/0_stateless/02995_index_9.reference @@ -0,0 +1,9 @@ +839 9 29223 46530 9 1336 1465 9 52 52 1 +844 5 2377545 2377635 5 5129 6321 5 53 69 5 +846 50 2172273 2589295 50 1582 3053 48 64 68 5 +847 2577 56656 63658 1582 1444 838 474 26 63 33 +861 1333 5570 6909 839 457 489 37 33 70 34 +873 2360 1519811 50487 2248 1310 1784 316 60 68 9 +879 228 6704 6785 79 279 507 121 35 66 24 +889 5130 2070007 39692 5040 1151 6791 2606 44 66 23 +896 4 511246 859452 4 6554 6561 4 67 71 4 diff --git a/tests/queries/0_stateless/02995_index_9.sh b/tests/queries/0_stateless/02995_index_9.sh new file mode 100755 index 00000000000..76160c62aaa --- /dev/null +++ b/tests/queries/0_stateless/02995_index_9.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a String, b String, c String) ENGINE = MergeTree ORDER BY (a, b, c) SETTINGS index_granularity = 11; + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/4)), round(pow(sipHash64(2, number), 1/6)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/3)), round(pow(sipHash64(2, number), 1/5)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +INSERT INTO test +SELECT round(pow(sipHash64(1, number), 1/5)), round(pow(sipHash64(2, number), 1/7)), round(pow(sipHash64(3, number), 1/10)) +FROM numbers(100000); + +DETACH TABLE test; +ATTACH TABLE test; +" + +for i in {801..900} +do + echo " +WITH ${i} AS try +SELECT try, count(), min(a), max(a), uniqExact(a), min(b), max(b), uniqExact(b), min(c), max(c), uniqExact(c) FROM test +WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1)::String + AND a <= (a1 + round(pow(sipHash64(3, try), 1 / (3 + sipHash64(4, try) % 8))))::String + AND b >= (round(pow(sipHash64(5, try), 1 / (3 + sipHash64(6, try) % 8))) AS b1)::String + AND b <= (b1 + round(pow(sipHash64(7, try), 1 / (3 + sipHash64(8, try) % 8))))::String + AND c >= (round(pow(sipHash64(9, try), 1 / (3 + sipHash64(10, try) % 8))) AS c1)::String + AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String +HAVING count() > 0; +" +done | ${CLICKHOUSE_CLIENT} --multiquery + +${CLICKHOUSE_CLIENT} --multiquery "DROP TABLE test" diff --git a/tests/queries/0_stateless/03000_virtual_columns_in_prewhere.sql b/tests/queries/0_stateless/03000_virtual_columns_in_prewhere.sql index d57db9151b9..c1e6eba6b6f 100644 --- a/tests/queries/0_stateless/03000_virtual_columns_in_prewhere.sql +++ b/tests/queries/0_stateless/03000_virtual_columns_in_prewhere.sql @@ -1,3 +1,5 @@ +SET optimize_trivial_insert_select = 1; + drop table if exists x; create table x (i int, j int, k int) engine MergeTree order by tuple() settings index_granularity=8192, index_granularity_bytes = '10Mi', min_bytes_for_wide_part=0, min_rows_for_wide_part=0, ratio_of_defaults_for_sparse_serialization=1; diff --git a/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql b/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql index 988d7058f21..2b10c72ae1b 100644 --- a/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql +++ b/tests/queries/0_stateless/03001_consider_lwd_when_merge.sql @@ -7,12 +7,14 @@ SETTINGS max_bytes_to_merge_at_max_space_in_pool = 80000, exclude_deleted_rows_f INSERT INTO lwd_merge SELECT number FROM numbers(10000); INSERT INTO lwd_merge SELECT number FROM numbers(10000, 10000); -OPTIMIZE TABLE lwd_merge; +SET optimize_throw_if_noop = 1; + +OPTIMIZE TABLE lwd_merge; -- { serverError CANNOT_ASSIGN_OPTIMIZE } SELECT count() FROM system.parts WHERE database = currentDatabase() AND table = 'lwd_merge' AND active = 1; DELETE FROM lwd_merge WHERE id % 10 > 0; -OPTIMIZE TABLE lwd_merge; +OPTIMIZE TABLE lwd_merge; -- { serverError CANNOT_ASSIGN_OPTIMIZE } SELECT count() FROM system.parts WHERE database = currentDatabase() AND table = 'lwd_merge' AND active = 1; ALTER TABLE lwd_merge MODIFY SETTING exclude_deleted_rows_for_part_size_in_merge = 1; diff --git a/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.reference b/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.reference deleted file mode 100644 index b19d389d8d0..00000000000 --- a/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.reference +++ /dev/null @@ -1,10 +0,0 @@ -before -rmt_master NewPart 0 1 -rmt_master MergeParts 0 1 -rmt_slave MergeParts 1 0 -rmt_slave DownloadPart 0 1 -after -rmt_master NewPart 0 1 -rmt_master MergeParts 0 1 -rmt_slave MergeParts 1 0 -rmt_slave DownloadPart 0 2 diff --git a/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.sql b/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.sql deleted file mode 100644 index 548a8e5570a..00000000000 --- a/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.sql +++ /dev/null @@ -1,35 +0,0 @@ --- Tags: no-replicated-database, no-parallel, no-shared-merge-tree --- SMT: The merge process is completely different from RMT - -drop table if exists rmt_master; -drop table if exists rmt_slave; - -create table rmt_master (key Int) engine=ReplicatedMergeTree('/clickhouse/{database}', 'master') order by key settings always_fetch_merged_part=0; --- always_fetch_merged_part=1, consider this table as a "slave" -create table rmt_slave (key Int) engine=ReplicatedMergeTree('/clickhouse/{database}', 'slave') order by key settings always_fetch_merged_part=1; - -insert into rmt_master values (1); - -system sync replica rmt_master; -system sync replica rmt_slave; -system stop replicated sends rmt_master; -optimize table rmt_master final settings alter_sync=1, optimize_throw_if_noop=1; - -select sleep(3) format Null; - -system flush logs; -select 'before'; -select table, event_type, error>0, countIf(error=0) from system.part_log where database = currentDatabase() group by 1, 2, 3 order by 1, 2, 3; - -system start replicated sends rmt_master; --- sleep few seconds to try rmt_slave to fetch the part and reflect this error --- in system.part_log -select sleep(3) format Null; -system sync replica rmt_slave; - -system flush logs; -select 'after'; -select table, event_type, error>0, countIf(error=0) from system.part_log where database = currentDatabase() group by 1, 2, 3 order by 1, 2, 3; - -drop table rmt_master; -drop table rmt_slave; diff --git a/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.reference b/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.reference deleted file mode 100644 index aac9e7527d1..00000000000 --- a/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.reference +++ /dev/null @@ -1,10 +0,0 @@ -before -rmt_master NewPart 0 1 -rmt_master MutatePart 0 1 -rmt_slave DownloadPart 0 1 -rmt_slave MutatePart 1 0 -after -rmt_master NewPart 0 1 -rmt_master MutatePart 0 1 -rmt_slave DownloadPart 0 2 -rmt_slave MutatePart 1 0 diff --git a/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.sql b/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.sql deleted file mode 100644 index d8b5ebb3148..00000000000 --- a/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.sql +++ /dev/null @@ -1,41 +0,0 @@ --- Tags: no-replicated-database, no-parallel, no-shared-merge-tree --- SMT: The merge process is completely different from RMT - -drop table if exists rmt_master; -drop table if exists rmt_slave; - -create table rmt_master (key Int) engine=ReplicatedMergeTree('/clickhouse/{database}', 'master') order by tuple() settings always_fetch_merged_part=0, old_parts_lifetime=600; --- prefer_fetch_merged_part_*_threshold=0, consider this table as a "slave" -create table rmt_slave (key Int) engine=ReplicatedMergeTree('/clickhouse/{database}', 'slave') order by tuple() settings prefer_fetch_merged_part_time_threshold=0, prefer_fetch_merged_part_size_threshold=0, old_parts_lifetime=600; - -insert into rmt_master values (1); - -system sync replica rmt_master; -system sync replica rmt_slave; -system stop replicated sends rmt_master; -system stop pulling replication log rmt_slave; -alter table rmt_master update key=key+100 where 1 settings alter_sync=1; - --- first we need to make the rmt_master execute mutation so that it will have --- the part, and rmt_slave will consider it instead of performing mutation on --- it's own, otherwise prefer_fetch_merged_part_*_threshold will be simply ignored -select sleep(3) format Null; -system start pulling replication log rmt_slave; --- and sleep few more seconds to try rmt_slave to fetch the part and reflect --- this error in system.part_log -select sleep(3) format Null; - -system flush logs; -select 'before'; -select table, event_type, error>0, countIf(error=0) from system.part_log where database = currentDatabase() group by 1, 2, 3 order by 1, 2, 3; - -system start replicated sends rmt_master; -select sleep(3) format Null; -system sync replica rmt_slave; - -system flush logs; -select 'after'; -select table, event_type, error>0, countIf(error=0) from system.part_log where database = currentDatabase() group by 1, 2, 3 order by 1, 2, 3; - -drop table rmt_master; -drop table rmt_slave; diff --git a/tests/queries/0_stateless/03003_functions_to_subcolumns_final.reference b/tests/queries/0_stateless/03003_functions_to_subcolumns_final.reference new file mode 100644 index 00000000000..3051c199363 --- /dev/null +++ b/tests/queries/0_stateless/03003_functions_to_subcolumns_final.reference @@ -0,0 +1,25 @@ +3 +2 +SELECT __table1.`arr.size0` AS `length(arr)` +FROM default.t_length_1 AS __table1 +WHERE __table1.`arr.size0` IN ( + SELECT __table1.arr_length AS arr_length + FROM default.t_length_2 AS __table1 +) +2 +SELECT __table1.`arr.size0` AS `length(arr)` +FROM default.t_length_1 AS __table1 +WHERE __table1.`arr.size0` IN ( + SELECT __table1.arr_length AS arr_length + FROM default.t_length_2 AS __table1 + FINAL +) +2 +SELECT length(__table1.arr) AS `length(arr)` +FROM default.t_length_1 AS __table1 +FINAL +WHERE length(__table1.arr) IN ( + SELECT __table1.arr_length AS arr_length + FROM default.t_length_2 AS __table1 + FINAL +) diff --git a/tests/queries/0_stateless/03003_functions_to_subcolumns_final.sql b/tests/queries/0_stateless/03003_functions_to_subcolumns_final.sql new file mode 100644 index 00000000000..3fe29139c5f --- /dev/null +++ b/tests/queries/0_stateless/03003_functions_to_subcolumns_final.sql @@ -0,0 +1,24 @@ +DROP TABLE IF EXISTS t_length_1; +DROP TABLE IF EXISTS t_length_2; + +SET optimize_functions_to_subcolumns = 1; +SET allow_experimental_analyzer = 1; +SET optimize_on_insert = 0; + +CREATE TABLE t_length_1 (id UInt64, arr Array(UInt64)) ENGINE = ReplacingMergeTree ORDER BY id; +CREATE TABLE t_length_2 (id UInt64, arr_length UInt64) ENGINE = ReplacingMergeTree ORDER BY id; + +INSERT INTO t_length_1 VALUES (1, [1, 2, 3]), (2, [4, 5]); +INSERT INTO t_length_2 VALUES (1, 3), (1, 2), (2, 2); + +SELECT length(arr) FROM t_length_1 WHERE length(arr) in (SELECT arr_length FROM t_length_2); +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT length(arr) FROM t_length_1 WHERE length(arr) in (SELECT arr_length FROM t_length_2); + +SELECT length(arr) FROM t_length_1 WHERE length(arr) in (SELECT arr_length FROM t_length_2 FINAL); +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT length(arr) FROM t_length_1 WHERE length(arr) in (SELECT arr_length FROM t_length_2 FINAL); + +SELECT length(arr) FROM t_length_1 FINAL WHERE length(arr) in (SELECT arr_length FROM t_length_2 FINAL); +EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT length(arr) FROM t_length_1 FINAL WHERE length(arr) in (SELECT arr_length FROM t_length_2 FINAL); + +DROP TABLE t_length_1; +DROP TABLE t_length_2; diff --git a/tests/queries/0_stateless/03008_deduplication.python b/tests/queries/0_stateless/03008_deduplication.python new file mode 100644 index 00000000000..bfaa496805f --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication.python @@ -0,0 +1,659 @@ +#!/usr/bin/env python3 + +import os +import sys +import argparse +import string + + +CURDIR = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, os.path.join(CURDIR, "helpers")) + + +def __format(template, **params): + field_names = [v[1] for v in string.Formatter().parse(template) if v[1] is not None] + kv_args = {} + for field in field_names: + if field in params: + kv_args[field] = params[field] + else: + kv_args[field] = "" + + return template.format(**kv_args) + + +def instance_create_statement( + table_name, + table_columns, + table_keys, + table_engine, + with_deduplication, + no_merges=True, +): + template = """ + CREATE TABLE {table_name} + {table_columns} + ENGINE = {table_engine} + ORDER BY {table_keys} + {table_settings}; + {table_no_merges} + """ + + params = dict() + params["table_name"] = table_name + params["table_columns"] = table_columns + params["table_keys"] = table_keys + params["table_no_merges"] = f"SYSTEM STOP MERGES {table_name};" if no_merges else "" + params["table_engine"] = ( + "MergeTree()" + if table_engine == "MergeTree" + else f"ReplicatedMergeTree('/clickhouse/tables/{{database}}/{table_name}', '1')" + ) + + deduplication_window_setting_name = ( + "non_replicated_deduplication_window" + if table_engine == "MergeTree" + else "replicated_deduplication_window" + ) + deduplication_window_setting_value = 1000 if with_deduplication else 0 + + settings = list() + settings += [ + f"{deduplication_window_setting_name}={deduplication_window_setting_value}" + ] + params["table_settings"] = "SETTINGS " + ",".join(settings) + + return __format(template, **params) + + +def instance_insert_statement( + table_name, count, insert_method, insert_unique_blocks, use_insert_token +): + insert_settings = ( + "" if not use_insert_token else "SETTINGS insert_deduplication_token='UDT'" + ) + + if insert_method == "InsertSelect": + template = """ + INSERT INTO {table_name} + SELECT {insert_columns} + FROM numbers({count}) {insert_settings}; + """ + return __format( + template, + table_name=table_name, + count=count, + insert_columns=( + "'src_4', 4" + if not insert_unique_blocks + else "'src_' || toString(number), number" + ), + insert_settings=insert_settings, + ) + + else: + template = """ + INSERT INTO {table_name} + {insert_settings} VALUES {insert_values}; + """ + + values = [] + for i in range(count): + values += ( + [f"('src_{i}', {i})"] if insert_unique_blocks else ["('src_4', 4)"] + ) + insert_values = ", ".join(values) + + return __format( + template, + table_name=table_name, + insert_settings=insert_settings, + insert_values=insert_values, + ) + + +def get_drop_tables_statements(tables): + return "".join( + [f"DROP TABLE IF EXISTS {table_name};\n" for table_name in tables[::-1]] + ) + + +def get_logs_statement(args): + if args.get_logs: + return "SET send_logs_level='test';" + return "" + + +def str2bool(v): + if isinstance(v, bool): + return v + if v.lower() in ("yes", "true", "t", "y", "1"): + return True + elif v.lower() in ("no", "false", "f", "n", "0"): + return False + else: + raise argparse.ArgumentTypeError("Boolean value expected.") + + +class ArgsFactory: + def __init__(self, parser): + self.__parser = parser + + def add_opt_engine(self): + self.__parser.add_argument( + "--table-engine", + choices=["ReplicatedMergeTree", "MergeTree"], + default="MergeTree", + ) + + def add_opt_user_token(self): + self.__parser.add_argument( + "--use-insert-token", type=str2bool, nargs="?", const=True, default=False + ) + + def add_opt_single_thread(self): + self.__parser.add_argument( + "--single-thread", type=str2bool, nargs="?", const=True, default=True + ) + + def add_opt_dedup_src(self): + self.__parser.add_argument( + "--deduplicate-src-table", + type=str2bool, + nargs="?", + const=True, + default=True, + ) + + def add_opt_dedup_dst(self): + self.__parser.add_argument( + "--deduplicate-dst-table", + type=str2bool, + nargs="?", + const=True, + default=True, + ) + + def add_opt_get_logs(self): + self.__parser.add_argument( + "--get-logs", type=str2bool, nargs="?", const=True, default=False + ) + + def add_opt_uniq_blocks(self): + self.__parser.add_argument( + "--insert-unique-blocks", type=str2bool, nargs="?", const=True, default=True + ) + + def add_opt_insert_method(self): + self.__parser.add_argument( + "--insert-method", + choices=["InsertSelect", "InsertValues"], + default="InsertSelect", + ) + + def add_all(self): + self.add_opt_engine() + self.add_opt_user_token() + self.add_opt_single_thread() + self.add_opt_dedup_src() + self.add_opt_dedup_dst() + self.add_opt_get_logs() + self.add_opt_insert_method() + self.add_opt_uniq_blocks() + + +def test_insert_several_blocks(parser): + ArgsFactory(parser).add_all() + + def calle(args): + create_table_a_b_statement = instance_create_statement( + table_name="table_a_b", + table_columns="(a String, b UInt64)", + table_keys="(a, b)", + table_engine=args.table_engine, + with_deduplication=args.deduplicate_src_table, + ) + + create_table_when_b_even_statement = instance_create_statement( + table_name="table_when_b_even", + table_columns="(a String, b UInt64)", + table_keys="(a, b)", + table_engine=args.table_engine, + with_deduplication=args.deduplicate_dst_table, + ) + + create_mv_statement = """ + CREATE MATERIALIZED VIEW mv_b_even + TO table_when_b_even + AS + SELECT a, b + FROM table_a_b + WHERE b % 2 = 0; + """ + + drop_tables_statements = get_drop_tables_statements( + ["table_a_b", "table_when_b_even", "mv_b_even"] + ) + + insert_statement = instance_insert_statement( + "table_a_b", + 10, + args.insert_method, + args.insert_unique_blocks, + args.use_insert_token, + ) + + print_details_statements = f""" + SELECT 'table_a_b'; + SELECT 'count', count() FROM table_a_b; + {"" if not args.get_logs else "SELECT _part, count() FROM table_a_b GROUP BY _part ORDER BY _part;"} + + SELECT 'table_when_b_even'; + SELECT 'count', count() FROM table_when_b_even; + {"" if not args.get_logs else "SELECT _part, count() FROM table_when_b_even GROUP BY _part ORDER BY _part;"} + """ + + if args.insert_unique_blocks: + assert_first_insert_statements = f""" + SELECT throwIf( count() != 10 ) + FROM table_a_b; + SELECT throwIf( count() != 5 ) + FROM table_when_b_even; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {10 if args.deduplicate_src_table else 20} ) + FROM table_a_b; + SELECT throwIf( count() != {5 if args.deduplicate_dst_table else 10} ) + FROM table_when_b_even; + """ + else: + if args.use_insert_token: + assert_first_insert_statements = """ + SELECT throwIf( count() != 10 ) + FROM table_a_b; + SELECT throwIf( count() != 10 ) + FROM table_when_b_even; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {10 if args.deduplicate_src_table else 20} ) + FROM table_a_b; + SELECT throwIf( count() != {10 if args.deduplicate_dst_table else 20} ) + FROM table_when_b_even; + """ + else: + assert_first_insert_statements = f""" + SELECT throwIf( count() != {1 if args.deduplicate_src_table else 10} ) + FROM table_a_b; + SELECT throwIf( count() != {1 if args.deduplicate_dst_table else 10} ) + FROM table_when_b_even; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {1 if args.deduplicate_src_table else 20} ) + FROM table_a_b; + SELECT throwIf( count() != {1 if args.deduplicate_dst_table else 20} ) + FROM table_when_b_even; + """ + + script = f""" + {get_logs_statement(args)} + + SET max_insert_threads={1 if args.single_thread else 10}; + SET update_insert_deduplication_token_in_dependent_materialized_views=1; + SET deduplicate_blocks_in_dependent_materialized_views=1; + + SET max_block_size=1; + SET min_insert_block_size_rows=0; + SET min_insert_block_size_bytes=0; + + {drop_tables_statements} + + {create_table_a_b_statement} + + {create_table_when_b_even_statement} + + {create_mv_statement} + + -- first insert + {insert_statement} + + {print_details_statements} + + {assert_first_insert_statements} + + -- second insert, it is retry + {insert_statement} + + {print_details_statements} + + {assert_second_insert_statements} + + {drop_tables_statements} + """ + + print(script) + + parser.set_defaults(func=calle) + + +def test_mv_generates_several_blocks(parser): + ArgsFactory(parser).add_all() + + def calle(args): + tables = [ + "table_for_join_with", + "table_a_b", + "table_when_b_even_and_joined", + "mv_b_even", + ] + drop_tables_statements = get_drop_tables_statements(tables) + + details_print_for_table_for_join_with = "" + if args.get_logs: + details_print_for_table_for_join_with = """ + SELECT 'table_for_join_with'; + SELECT a_join, b, _part FROM table_for_join_with ORDER BY _part, a_join, b; + """ + + create_table_a_b_statement = instance_create_statement( + table_name="table_a_b", + table_columns="(a_src String, b UInt64)", + table_keys="(a_src, b)", + table_engine=args.table_engine, + with_deduplication=args.deduplicate_src_table, + ) + + create_table_when_b_even_and_joined_statement = instance_create_statement( + table_name="table_when_b_even_and_joined", + table_columns="(a_src String, a_join String, b UInt64)", + table_keys="(a_src, a_join, b)", + table_engine=args.table_engine, + with_deduplication=args.deduplicate_dst_table, + ) + + insert_statement = instance_insert_statement( + "table_a_b", + 5, + args.insert_method, + args.insert_unique_blocks, + args.use_insert_token, + ) + + details_print_statements = f""" + SELECT 'table_a_b'; + SELECT 'count', count() FROM table_a_b; + + SELECT 'table_when_b_even_and_joined'; + SELECT 'count', count() FROM table_when_b_even_and_joined; + {"" if not args.get_logs else "SELECT _part, a_src, a_join, b FROM table_when_b_even_and_joined ORDER BY _part;"} + """ + + if args.insert_unique_blocks: + assert_first_insert_statements = f""" + SELECT throwIf( count() != 5 ) + FROM table_a_b; + + SELECT throwIf( count() != 9 ) + FROM table_when_b_even_and_joined; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {5 if args.deduplicate_src_table else 10} ) + FROM table_a_b; + + SELECT throwIf( count() != {9 if args.deduplicate_dst_table else 18} ) + FROM table_when_b_even_and_joined; + """ + else: + if args.use_insert_token: + assert_first_insert_statements = f""" + SELECT throwIf( count() != {5 if args.deduplicate_src_table else 5} ) + FROM table_a_b; + + SELECT throwIf( count() != {10 if args.deduplicate_dst_table else 10} ) + FROM table_when_b_even_and_joined; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {5 if args.deduplicate_src_table else 10} ) + FROM table_a_b; + + SELECT throwIf( count() != {10 if args.deduplicate_dst_table else 20} ) + FROM table_when_b_even_and_joined; + """ + else: + assert_first_insert_statements = f""" + SELECT throwIf( count() != {1 if args.deduplicate_src_table else 5} ) + FROM table_a_b; + + SELECT throwIf( count() != {2 if args.deduplicate_dst_table else 10} ) + FROM table_when_b_even_and_joined; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {1 if args.deduplicate_src_table else 10} ) + FROM table_a_b; + + SELECT throwIf( count() != {2 if args.deduplicate_dst_table else 20} ) + FROM table_when_b_even_and_joined; + """ + + script = f""" + {get_logs_statement(args)} + + SET max_insert_threads={1 if args.single_thread else 10}; + SET update_insert_deduplication_token_in_dependent_materialized_views=1; + SET deduplicate_blocks_in_dependent_materialized_views=1; + + SET max_block_size=1; + SET min_insert_block_size_rows=0; + SET min_insert_block_size_bytes=0; + + {drop_tables_statements} + + CREATE TABLE table_for_join_with + (a_join String, b UInt64) + ENGINE = MergeTree() + ORDER BY (a_join, b); + INSERT INTO table_for_join_with + SELECT 'joined_' || toString(number), number + FROM numbers(1); + {details_print_for_table_for_join_with} + + {create_table_a_b_statement} + SYSTEM STOP MERGES table_a_b; + + {create_table_when_b_even_and_joined_statement} + SYSTEM STOP MERGES table_when_b_even_and_joined; + + CREATE MATERIALIZED VIEW mv_b_even + TO table_when_b_even_and_joined + AS + SELECT a_src, a_join, table_for_join_with.b as b + FROM table_a_b + FULL OUTER JOIN table_for_join_with + ON table_a_b.b = table_for_join_with.b AND table_a_b.b % 2 = 0 + ORDER BY a_src, a_join, b; + + -- first insert + {insert_statement} + + {details_print_statements} + + -- first assertion + {assert_first_insert_statements} + + -- second insert + {insert_statement} + + {details_print_statements} + + -- second assertion + {assert_second_insert_statements} + + {drop_tables_statements} + """ + + print(script) + + parser.set_defaults(func=calle) + + +def test_several_mv_into_one_table(parser): + ArgsFactory(parser).add_all() + + def calle(args): + tables = ["table_src", "table_dst", "mv_b_even", "mv_b_even_even"] + drop_tables_statements = get_drop_tables_statements(tables) + + create_table_src_statement = instance_create_statement( + table_name="table_src", + table_columns="(a String, b UInt64)", + table_keys="(a, b)", + table_engine=args.table_engine, + with_deduplication=args.deduplicate_src_table, + ) + + create_table_dst_statement = instance_create_statement( + table_name="table_dst", + table_columns="(a String, b UInt64)", + table_keys="(a, b)", + table_engine=args.table_engine, + with_deduplication=args.deduplicate_dst_table, + ) + + insert_statement = instance_insert_statement( + "table_src", + 8, + args.insert_method, + args.insert_unique_blocks, + args.use_insert_token, + ) + + details_print_statements = f""" + SELECT 'table_src count', count() FROM table_src; + + SELECT 'table_dst count', count() FROM table_dst; + {"" if not args.get_logs else "SELECT _part, count() FROM table_dst GROUP BY _part ORDER BY _part;"} + """ + + if args.insert_unique_blocks: + assert_first_insert_statements = f""" + SELECT throwIf( count() != 8 ) + FROM table_src; + + SELECT throwIf( count() != 6 ) + FROM table_dst; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {8 if args.deduplicate_src_table else 16} ) + FROM table_src; + + SELECT throwIf( count() != {6 if args.deduplicate_dst_table else 12} ) + FROM table_dst; + """ + else: + if args.use_insert_token: + assert_first_insert_statements = f""" + SELECT throwIf( count() != {8 if args.deduplicate_src_table else 8} ) + FROM table_src; + + SELECT throwIf( count() != {16 if args.deduplicate_dst_table else 16} ) + FROM table_dst; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {8 if args.deduplicate_src_table else 16} ) + FROM table_src; + + SELECT throwIf( count() != {16 if args.deduplicate_dst_table else 32} ) + FROM table_dst; + """ + else: + assert_first_insert_statements = f""" + SELECT throwIf( count() != {1 if args.deduplicate_src_table else 8} ) + FROM table_src; + + SELECT throwIf( count() != {2 if args.deduplicate_dst_table else 16} ) + FROM table_dst; + """ + assert_second_insert_statements = f""" + SELECT throwIf( count() != {1 if args.deduplicate_src_table else 16} ) + FROM table_src; + + SELECT throwIf( count() != {2 if args.deduplicate_dst_table else 32} ) + FROM table_dst; + """ + + script = f""" + {get_logs_statement(args)} + + SET max_insert_threads={1 if args.single_thread else 10}; + SET update_insert_deduplication_token_in_dependent_materialized_views=1; + SET deduplicate_blocks_in_dependent_materialized_views=1; + + SET max_block_size=1; + SET min_insert_block_size_rows=0; + SET min_insert_block_size_bytes=0; + + {drop_tables_statements} + + {create_table_src_statement} + + {create_table_dst_statement} + + CREATE MATERIALIZED VIEW mv_b_even + TO table_dst + AS + SELECT a, b + FROM table_src + WHERE b % 2 = 0; + + CREATE MATERIALIZED VIEW mv_b_even_even + TO table_dst + AS + SELECT a, b + FROM table_src + WHERE b % 4 = 0; + + -- first insert + {insert_statement} + + {details_print_statements} + + {assert_first_insert_statements} + + -- second insert, retry + {insert_statement} + + {details_print_statements} + + {assert_second_insert_statements} + + {drop_tables_statements} + """ + + print(script) + + parser.set_defaults(func=calle) + + +def parse_args(): + parser = argparse.ArgumentParser() + subparsers = parser.add_subparsers(dest="test") + test_insert_several_blocks( + subparsers.add_parser("insert_several_blocks_into_table") + ) + test_mv_generates_several_blocks( + subparsers.add_parser("mv_generates_several_blocks") + ) + test_several_mv_into_one_table(subparsers.add_parser("several_mv_into_one_table")) + args = parser.parse_args() + if args.test is None: + parser.print_help() + return args + + +def main(): + args = parse_args() + if args.test is not None: + args.func(args) + + +if __name__ == "__main__": + main() diff --git a/tests/queries/0_stateless/03008_deduplication_cases_from_docs.reference b/tests/queries/0_stateless/03008_deduplication_cases_from_docs.reference new file mode 100644 index 00000000000..4893274c1cd --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_cases_from_docs.reference @@ -0,0 +1,41 @@ +Different materialized view insert into one underlayed table equal data. +first attempt +from dst 1 A all_1_1_0 +from mv_dst 0 A all_1_1_0 +from mv_dst 0 A all_2_2_0 +second attempt +from dst 1 A all_1_1_0 +from mv_dst 0 A all_1_1_0 +from mv_dst 0 A all_2_2_0 +Different insert operations generate the same data after transformation in underlied table of materialized view. +first attempt +from dst 1 A all_1_1_0 +from mv_dst 0 A all_1_1_0 +second attempt +from dst 1 A all_1_1_0 +from dst 2 A all_2_2_0 +from mv_dst 0 A all_1_1_0 +from mv_dst 0 A all_2_2_0 +Indentical blocks in insertion with `insert_deduplication_token` +first attempt +from dst 0 A all_1_1_0 +from dst 0 A all_2_2_0 +second attempt +from dst 0 A all_1_1_0 +from dst 0 A all_2_2_0 +third attempt +from dst 0 A all_1_1_0 +from dst 0 A all_2_2_0 +Indentical blocks in insertion +from dst 0 A all_1_1_0 +Indentical blocks after materialised view`s transformation +first attempt +from dst 1 B all_1_1_0 +from dst 2 B all_2_2_0 +from mv_dst 0 B all_1_1_0 +from mv_dst 0 B all_2_2_0 +second attempt +from dst 1 B all_1_1_0 +from dst 2 B all_2_2_0 +from mv_dst 0 B all_1_1_0 +from mv_dst 0 B all_2_2_0 diff --git a/tests/queries/0_stateless/03008_deduplication_cases_from_docs.sql b/tests/queries/0_stateless/03008_deduplication_cases_from_docs.sql new file mode 100644 index 00000000000..7927a6b1edf --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_cases_from_docs.sql @@ -0,0 +1,331 @@ +-- ######### +select 'Different materialized view insert into one underlayed table equal data.'; + +DROP TABLE IF EXISTS dst; +DROP TABLE IF EXISTS mv_dst; +DROP TABLE IF EXISTS mv_first; +DROP TABLE IF EXISTS mv_second; + +CREATE TABLE dst +( + `key` Int64, + `value` String +) +ENGINE = MergeTree +ORDER BY tuple() +SETTINGS non_replicated_deduplication_window=1000; + +CREATE TABLE mv_dst +( + `key` Int64, + `value` String +) +ENGINE = MergeTree +ORDER BY tuple() +SETTINGS non_replicated_deduplication_window=1000; + +CREATE MATERIALIZED VIEW mv_first +TO mv_dst +AS SELECT + 0 AS key, + value AS value +FROM dst; + +CREATE MATERIALIZED VIEW mv_second +TO mv_dst +AS SELECT + 0 AS key, + value AS value +FROM dst; + +SET deduplicate_blocks_in_dependent_materialized_views=1; + +select 'first attempt'; + +INSERT INTO dst VALUES (1, 'A'); + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +SELECT + 'from mv_dst', + *, + _part +FROM mv_dst +ORDER by all; + +select 'second attempt'; + +INSERT INTO dst VALUES (1, 'A'); + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +SELECT + 'from mv_dst', + *, + _part +FROM mv_dst +ORDER by all; + +DROP TABLE mv_second; +DROP TABLE mv_first; +DROP TABLE mv_dst; +DROP TABLE dst; + + +-- ######### +select 'Different insert operations generate the same data after transformation in underlied table of materialized view.'; + +DROP TABLE IF EXISTS dst; +DROP TABLE IF EXISTS mv_dst; + +CREATE TABLE dst +( + `key` Int64, + `value` String +) +ENGINE = MergeTree +ORDER BY tuple() +SETTINGS non_replicated_deduplication_window=1000; + +CREATE MATERIALIZED VIEW mv_dst +( + `key` Int64, + `value` String +) +ENGINE = MergeTree +ORDER BY tuple() +SETTINGS non_replicated_deduplication_window=1000 +AS SELECT + 0 AS key, + value AS value +FROM dst; + +SET deduplicate_blocks_in_dependent_materialized_views=1; + +select 'first attempt'; + +INSERT INTO dst VALUES (1, 'A'); + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +SELECT + 'from mv_dst', + *, + _part +FROM mv_dst +ORDER by all; + +select 'second attempt'; + +INSERT INTO dst VALUES (2, 'A'); + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +SELECT + 'from mv_dst', + *, + _part +FROM mv_dst +ORDER by all; + +DROP TABLE mv_dst; +DROP TABLE dst; + + +-- ######### +select 'Indentical blocks in insertion with `insert_deduplication_token`'; + +DROP TABLE IF EXISTS dst; + +CREATE TABLE dst +( + `key` Int64, + `value` String +) +ENGINE = MergeTree +ORDER BY tuple() +SETTINGS non_replicated_deduplication_window=1000; + +SET max_block_size=1; +SET min_insert_block_size_rows=0; +SET min_insert_block_size_bytes=0; + +select 'first attempt'; + +INSERT INTO dst SELECT + 0 AS key, + 'A' AS value +FROM numbers(2) +SETTINGS insert_deduplication_token='some_user_token'; + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +select 'second attempt'; + +INSERT INTO dst SELECT + 0 AS key, + 'A' AS value +FROM numbers(2) +SETTINGS insert_deduplication_token='some_user_token'; + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +select 'third attempt'; + +INSERT INTO dst SELECT + 1 AS key, + 'b' AS value +FROM numbers(2) +SETTINGS insert_deduplication_token='some_user_token'; + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +DROP TABLE dst; + + +-- ######### +select 'Indentical blocks in insertion'; + +DROP TABLE IF EXISTS dst; + +CREATE TABLE dst +( + `key` Int64, + `value` String +) +ENGINE = MergeTree +ORDER BY tuple() +SETTINGS non_replicated_deduplication_window=1000; + +SET max_block_size=1; +SET min_insert_block_size_rows=0; +SET min_insert_block_size_bytes=0; + +INSERT INTO dst SELECT + 0 AS key, + 'A' AS value +FROM numbers(2); + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +DROP TABLE dst; + + +-- ######### +select 'Indentical blocks after materialised view`s transformation'; + +DROP TABLE IF EXISTS dst; +DROP TABLE IF EXISTS mv_dst; + +CREATE TABLE dst +( + `key` Int64, + `value` String +) +ENGINE = MergeTree +ORDER BY tuple() +SETTINGS non_replicated_deduplication_window=1000; + +CREATE MATERIALIZED VIEW mv_dst +( + `key` Int64, + `value` String +) +ENGINE = MergeTree +ORDER BY tuple() +SETTINGS non_replicated_deduplication_window=1000 +AS SELECT + 0 AS key, + value AS value +FROM dst; + +SET max_block_size=1; +SET min_insert_block_size_rows=0; +SET min_insert_block_size_bytes=0; + +SET deduplicate_blocks_in_dependent_materialized_views=1; + +select 'first attempt'; + +INSERT INTO dst SELECT + number + 1 AS key, + IF(key = 0, 'A', 'B') AS value +FROM numbers(2); + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +SELECT + 'from mv_dst', + *, + _part +FROM mv_dst +ORDER by all; + +select 'second attempt'; + +INSERT INTO dst SELECT + number + 1 AS key, + IF(key = 0, 'A', 'B') AS value +FROM numbers(2); + +SELECT + 'from dst', + *, + _part +FROM dst +ORDER by all; + +SELECT + 'from mv_dst', + *, + _part +FROM mv_dst +ORDER by all; + +DROP TABLE mv_dst; +DROP TABLE dst; diff --git a/tests/queries/0_stateless/03008_deduplication_insert_into_partitioned_table.reference b/tests/queries/0_stateless/03008_deduplication_insert_into_partitioned_table.reference new file mode 100644 index 00000000000..c82a6eaa213 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_insert_into_partitioned_table.reference @@ -0,0 +1,35 @@ +no user deduplication token +partitioned_table is deduplicated bacause deduplication works in scope of one partiotion: +1 A +1 D +2 B +2 C +mv_table is not deduplicated because the inserted blocks was different: +1 A +1 A +1 D +2 B +2 B +2 C +with user deduplication token +partitioned_table is not deduplicated because different tokens: +1 A +1 A +1 D +2 B +2 B +2 C +mv_table is not deduplicated because different tokens: +1 A +1 A +1 D +2 B +2 B +2 C +with incorrect ussage of user deduplication token +partitioned_table is deduplicated because equal tokens: +1 A +2 B +mv_table is deduplicated because equal tokens: +1 A +2 B diff --git a/tests/queries/0_stateless/03008_deduplication_insert_into_partitioned_table.sql b/tests/queries/0_stateless/03008_deduplication_insert_into_partitioned_table.sql new file mode 100644 index 00000000000..2eb931f7f73 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_insert_into_partitioned_table.sql @@ -0,0 +1,83 @@ +DROP TABLE IF EXISTS partitioned_table; +DROP TABLE IF EXISTS mv_table; + + +SET deduplicate_blocks_in_dependent_materialized_views = 1; + + +SELECT 'no user deduplication token'; + +CREATE TABLE partitioned_table + (key Int64, value String) + ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/03008_deduplication_insert_into_partitioned_table', '{replica}') + partition by key % 10 + order by tuple(); + +CREATE MATERIALIZED VIEW mv_table (key Int64, value String) + ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/03008_deduplication_insert_into_partitioned_table_mv', '{replica}') + ORDER BY tuple() + AS SELECT key, value FROM partitioned_table; + +INSERT INTO partitioned_table VALUES (1, 'A'), (2, 'B'); +INSERT INTO partitioned_table VALUES (1, 'A'), (2, 'C'); +INSERT INTO partitioned_table VALUES (1, 'D'), (2, 'B'); + +SELECT 'partitioned_table is deduplicated bacause deduplication works in scope of one partiotion:'; +SELECT * FROM partitioned_table ORDER BY ALL; +SELECT 'mv_table is not deduplicated because the inserted blocks was different:'; +SELECT * FROM mv_table ORDER BY ALL; + +DROP TABLE partitioned_table; +DROP TABLE mv_table; + + +SELECT 'with user deduplication token'; + +CREATE TABLE partitioned_table + (key Int64, value String) + ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/03008_deduplication_insert_into_partitioned_table', '{replica}') + partition by key % 10 + order by tuple(); + +CREATE MATERIALIZED VIEW mv_table (key Int64, value String) + ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/03008_deduplication_insert_into_partitioned_table_mv', '{replica}') + ORDER BY tuple() + AS SELECT key, value FROM partitioned_table; + +INSERT INTO partitioned_table SETTINGS insert_deduplication_token='token_1' VALUES (1, 'A'), (2, 'B'); +INSERT INTO partitioned_table SETTINGS insert_deduplication_token='token_2' VALUES (1, 'A'), (2, 'C'); +INSERT INTO partitioned_table SETTINGS insert_deduplication_token='token_3' VALUES (1, 'D'), (2, 'B'); + +SELECT 'partitioned_table is not deduplicated because different tokens:'; +SELECT * FROM partitioned_table ORDER BY ALL; +SELECT 'mv_table is not deduplicated because different tokens:'; +SELECT * FROM mv_table ORDER BY ALL; + +DROP TABLE partitioned_table; +DROP TABLE mv_table; + + +SELECT 'with incorrect ussage of user deduplication token'; + +CREATE TABLE partitioned_table + (key Int64, value String) + ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/03008_deduplication_insert_into_partitioned_table', '{replica}') + partition by key % 10 + order by tuple(); + +CREATE MATERIALIZED VIEW mv_table (key Int64, value String) + ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/03008_deduplication_insert_into_partitioned_table_mv', '{replica}') + ORDER BY tuple() + AS SELECT key, value FROM partitioned_table; + +INSERT INTO partitioned_table SETTINGS insert_deduplication_token='token_0' VALUES (1, 'A'), (2, 'B'); +INSERT INTO partitioned_table SETTINGS insert_deduplication_token='token_0' VALUES (1, 'A'), (2, 'C'); +INSERT INTO partitioned_table SETTINGS insert_deduplication_token='token_0' VALUES (1, 'D'), (2, 'B'); + +SELECT 'partitioned_table is deduplicated because equal tokens:'; +SELECT * FROM partitioned_table ORDER BY ALL; +SELECT 'mv_table is deduplicated because equal tokens:'; +SELECT * FROM mv_table ORDER BY ALL; + +DROP TABLE partitioned_table; +DROP TABLE mv_table; diff --git a/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_nonreplicated.reference b/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_nonreplicated.reference new file mode 100644 index 00000000000..bf900aa84d2 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_nonreplicated.reference @@ -0,0 +1,962 @@ + +Test case 0: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 1: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 2: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 3: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 20 +0 +0 +OK + +Test case 4: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 5: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 6: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 7: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 8: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 9: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 10: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 11: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 20 +0 +0 +OK + +Test case 12: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 13: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 14: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 15: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 16: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 17: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +OK + +Test case 18: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 19: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even +count 20 +0 +0 +OK + +Test case 20: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 21: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 1 +0 +0 +table_a_b +count 20 +table_when_b_even +count 1 +0 +0 +OK + +Test case 22: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 23: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 24: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 25: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +OK + +Test case 26: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 27: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even +count 20 +0 +0 +OK + +Test case 28: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 29: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 1 +0 +0 +table_a_b +count 20 +table_when_b_even +count 1 +0 +0 +OK + +Test case 30: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 31: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 32: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 33: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 34: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 35: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 20 +0 +0 +OK + +Test case 36: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 37: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 38: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 39: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 40: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 41: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 42: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 43: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 20 +0 +0 +OK + +Test case 44: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 45: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 46: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 47: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 48: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 49: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +OK + +Test case 50: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 51: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even +count 20 +0 +0 +OK + +Test case 52: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 53: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 1 +0 +0 +table_a_b +count 20 +table_when_b_even +count 1 +0 +0 +OK + +Test case 54: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 55: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 56: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 57: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +OK + +Test case 58: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 59: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even +count 20 +0 +0 +OK + +Test case 60: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 61: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 1 +0 +0 +table_a_b +count 20 +table_when_b_even +count 1 +0 +0 +OK + +Test case 62: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 63: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +All cases executed diff --git a/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_nonreplicated.sh b/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_nonreplicated.sh new file mode 100755 index 00000000000..ef4866cb69e --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_nonreplicated.sh @@ -0,0 +1,59 @@ +#!/usr/bin/env bash +# Tags: long, no-fasttest, no-parallel, no-azure-blob-storage, no-s3-storage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +ENGINE="MergeTree" + +RUN_ONLY="" +#RUN_ONLY="Test case 52: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True" + +i=0 +for insert_method in "InsertSelect" "InsertValues"; do + for use_insert_token in "True" "False"; do + for single_thread in "True" "False"; do + for deduplicate_src_table in "True" "False"; do + for deduplicate_dst_table in "True" "False"; do + for insert_unique_blocks in "True" "False"; do + + THIS_RUN="Test case $i:" + THIS_RUN+=" insert_method=$insert_method" + THIS_RUN+=" engine=$ENGINE" + THIS_RUN+=" use_insert_token=$use_insert_token" + THIS_RUN+=" single_thread=$single_thread" + THIS_RUN+=" deduplicate_src_table=$deduplicate_src_table" + THIS_RUN+=" deduplicate_dst_table=$deduplicate_dst_table" + THIS_RUN+=" insert_unique_blocks=$insert_unique_blocks" + + i=$((i+1)) + + echo + if [ -n "$RUN_ONLY" ] && [ "$RUN_ONLY" != "$THIS_RUN" ]; then + echo "skip $THIS_RUN" + continue + fi + echo "$THIS_RUN" + + $CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python insert_several_blocks_into_table \ + --insert-method $insert_method \ + --table-engine $ENGINE \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) + " && echo OK || echo FAIL + done + done + done + done + done +done + +echo +echo "All cases executed" diff --git a/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_replicated.reference b/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_replicated.reference new file mode 100644 index 00000000000..c815324b455 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_replicated.reference @@ -0,0 +1,962 @@ + +Test case 0: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 1: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 2: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 3: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 20 +0 +0 +OK + +Test case 4: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 5: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 6: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 7: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 8: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 9: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 10: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 11: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 20 +0 +0 +OK + +Test case 12: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 13: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 14: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 15: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 16: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 17: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +OK + +Test case 18: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 19: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even +count 20 +0 +0 +OK + +Test case 20: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 21: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 1 +0 +0 +table_a_b +count 20 +table_when_b_even +count 1 +0 +0 +OK + +Test case 22: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 23: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 24: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 25: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +OK + +Test case 26: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 27: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even +count 20 +0 +0 +OK + +Test case 28: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 29: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 1 +0 +0 +table_a_b +count 20 +table_when_b_even +count 1 +0 +0 +OK + +Test case 30: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 31: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 32: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 33: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 34: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 35: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 20 +0 +0 +OK + +Test case 36: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 37: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 38: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 39: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 40: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 41: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 42: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 43: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even +count 20 +0 +0 +OK + +Test case 44: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 45: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 46: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 47: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 48: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 49: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +OK + +Test case 50: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 51: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even +count 20 +0 +0 +OK + +Test case 52: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 53: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 1 +0 +0 +table_a_b +count 20 +table_when_b_even +count 1 +0 +0 +OK + +Test case 54: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 55: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +Test case 56: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +OK + +Test case 57: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +table_a_b +count 1 +table_when_b_even +count 1 +0 +0 +OK + +Test case 58: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +OK + +Test case 59: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even +count 20 +0 +0 +OK + +Test case 60: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 5 +0 +0 +OK + +Test case 61: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 1 +0 +0 +table_a_b +count 20 +table_when_b_even +count 1 +0 +0 +OK + +Test case 62: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 10 +table_when_b_even +count 5 +0 +0 +table_a_b +count 20 +table_when_b_even +count 10 +0 +0 +OK + +Test case 63: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 10 +table_when_b_even +count 10 +0 +0 +table_a_b +count 20 +table_when_b_even +count 20 +0 +0 +OK + +All cases executed diff --git a/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_replicated.sh b/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_replicated.sh new file mode 100755 index 00000000000..dd4b1f7cec0 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_insert_several_blocks_replicated.sh @@ -0,0 +1,59 @@ +#!/usr/bin/env bash +# Tags: long, no-fasttest, no-parallel, no-azure-blob-storage, no-s3-storage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +ENGINE="ReplicatedMergeTree" + +RUN_ONLY="" +#RUN_ONLY="Test case 52: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True" + +i=0 +for insert_method in "InsertSelect" "InsertValues"; do + for use_insert_token in "True" "False"; do + for single_thread in "True" "False"; do + for deduplicate_src_table in "True" "False"; do + for deduplicate_dst_table in "True" "False"; do + for insert_unique_blocks in "True" "False"; do + + THIS_RUN="Test case $i:" + THIS_RUN+=" insert_method=$insert_method" + THIS_RUN+=" engine=$ENGINE" + THIS_RUN+=" use_insert_token=$use_insert_token" + THIS_RUN+=" single_thread=$single_thread" + THIS_RUN+=" deduplicate_src_table=$deduplicate_src_table" + THIS_RUN+=" deduplicate_dst_table=$deduplicate_dst_table" + THIS_RUN+=" insert_unique_blocks=$insert_unique_blocks" + + i=$((i+1)) + + echo + if [ -n "$RUN_ONLY" ] && [ "$RUN_ONLY" != "$THIS_RUN" ]; then + echo "skip $THIS_RUN" + continue + fi + echo "$THIS_RUN" + + $CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python insert_several_blocks_into_table \ + --insert-method $insert_method \ + --table-engine $ENGINE \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) + " && echo OK || echo FAIL + done + done + done + done + done +done + +echo +echo "All cases executed" diff --git a/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_nonreplicated.reference b/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_nonreplicated.reference new file mode 100644 index 00000000000..6e76ec46aa8 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_nonreplicated.reference @@ -0,0 +1,962 @@ + +Test case 0: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 1: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 2: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 3: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 4: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 5: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 6: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 7: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 8: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 9: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 10: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 11: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 12: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 13: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 14: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 15: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 16: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 17: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 18: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 19: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 20: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 21: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 22: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 23: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 24: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 25: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 26: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 27: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 28: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 29: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 30: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 31: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 32: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 33: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 34: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 35: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 36: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 37: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 38: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 39: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 40: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 41: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 42: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 43: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 44: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 45: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 46: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 47: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 48: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 49: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 50: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 51: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 52: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 53: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 54: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 55: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 56: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 57: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 58: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 59: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 60: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 61: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 62: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 63: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +All cases executed diff --git a/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_nonreplicated.sh b/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_nonreplicated.sh new file mode 100755 index 00000000000..33386c76edb --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_nonreplicated.sh @@ -0,0 +1,59 @@ +#!/usr/bin/env bash +# Tags: long, no-fasttest, no-parallel, no-azure-blob-storage, no-s3-storage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +ENGINE="MergeTree" + +RUN_ONLY="" +#RUN_ONLY="Test case 20: engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True" + +i=0 +for insert_method in "InsertSelect" "InsertValues"; do + for use_insert_token in "True" "False"; do + for single_thread in "True" "False"; do + for deduplicate_src_table in "True" "False"; do + for deduplicate_dst_table in "True" "False"; do + for insert_unique_blocks in "True" "False"; do + + THIS_RUN="Test case $i:" + THIS_RUN+=" insert_method=$insert_method" + THIS_RUN+=" engine=$ENGINE" + THIS_RUN+=" use_insert_token=$use_insert_token" + THIS_RUN+=" single_thread=$single_thread" + THIS_RUN+=" deduplicate_src_table=$deduplicate_src_table" + THIS_RUN+=" deduplicate_dst_table=$deduplicate_dst_table" + THIS_RUN+=" insert_unique_blocks=$insert_unique_blocks" + + i=$((i+1)) + + echo + if [ -n "$RUN_ONLY" ] && [ "$RUN_ONLY" != "$THIS_RUN" ]; then + echo "skip $THIS_RUN" + continue + fi + echo "$THIS_RUN" + + $CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python mv_generates_several_blocks \ + --insert-method $insert_method \ + --table-engine $ENGINE \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) + " && echo OK || echo FAIL + done + done + done + done + done +done + +echo +echo "All cases executed" diff --git a/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_replicated.reference b/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_replicated.reference new file mode 100644 index 00000000000..a25e8713c61 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_replicated.reference @@ -0,0 +1,962 @@ + +Test case 0: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 1: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 2: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 3: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 4: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 5: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 6: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 7: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 8: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 9: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 10: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 11: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 12: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 13: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 14: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 15: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 16: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 17: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 18: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 19: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 20: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 21: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 22: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 23: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 24: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 25: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 26: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 27: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 28: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 29: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 30: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 31: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 32: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 33: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 34: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 35: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 36: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 37: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 38: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 39: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 40: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 41: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 42: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 43: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 44: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 45: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 10 +0 +0 +OK + +Test case 46: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 47: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 48: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 49: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 50: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 51: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 52: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 53: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 54: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 55: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 56: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 57: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 58: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 5 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 59: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 1 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 1 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +Test case 60: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 9 +0 +0 +OK + +Test case 61: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 2 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 2 +0 +0 +OK + +Test case 62: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_a_b +count 5 +table_when_b_even_and_joined +count 9 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 18 +0 +0 +OK + +Test case 63: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_a_b +count 5 +table_when_b_even_and_joined +count 10 +0 +0 +table_a_b +count 10 +table_when_b_even_and_joined +count 20 +0 +0 +OK + +All cases executed diff --git a/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_replicated.sh b/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_replicated.sh new file mode 100755 index 00000000000..b66ef83abf2 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_mv_generates_several_blocks_replicated.sh @@ -0,0 +1,59 @@ +#!/usr/bin/env bash +# Tags: long, no-fasttest, no-parallel, no-azure-blob-storage, no-s3-storage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +ENGINE="ReplicatedMergeTree" + +RUN_ONLY="" +#RUN_ONLY="Test case 20: engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True" + +i=0 +for insert_method in "InsertSelect" "InsertValues"; do + for use_insert_token in "True" "False"; do + for single_thread in "True" "False"; do + for deduplicate_src_table in "True" "False"; do + for deduplicate_dst_table in "True" "False"; do + for insert_unique_blocks in "True" "False"; do + + THIS_RUN="Test case $i:" + THIS_RUN+=" insert_method=$insert_method" + THIS_RUN+=" engine=$ENGINE" + THIS_RUN+=" use_insert_token=$use_insert_token" + THIS_RUN+=" single_thread=$single_thread" + THIS_RUN+=" deduplicate_src_table=$deduplicate_src_table" + THIS_RUN+=" deduplicate_dst_table=$deduplicate_dst_table" + THIS_RUN+=" insert_unique_blocks=$insert_unique_blocks" + + i=$((i+1)) + + echo + if [ -n "$RUN_ONLY" ] && [ "$RUN_ONLY" != "$THIS_RUN" ]; then + echo "skip $THIS_RUN" + continue + fi + echo "$THIS_RUN" + + $CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python mv_generates_several_blocks \ + --insert-method $insert_method \ + --table-engine $ENGINE \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) + " && echo OK || echo FAIL + done + done + done + done + done +done + +echo +echo "All cases executed" diff --git a/tests/queries/0_stateless/03008_deduplication_random_setttings.reference b/tests/queries/0_stateless/03008_deduplication_random_setttings.reference new file mode 100644 index 00000000000..8e7e24a7dd2 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_random_setttings.reference @@ -0,0 +1,3 @@ +insert_several_blocks_into_table OK +mv_generates_several_blocks OK +several_mv_into_one_table OK diff --git a/tests/queries/0_stateless/03008_deduplication_random_setttings.sh b/tests/queries/0_stateless/03008_deduplication_random_setttings.sh new file mode 100755 index 00000000000..e9f59138177 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_random_setttings.sh @@ -0,0 +1,76 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +engine_options=("ReplicatedMergeTree" "MergeTree") +engine=${engine_options[ $RANDOM % ${#engine_options[@]} ]} + +insert_method_option=("InsertSelect" "InsertValues") +insert_method=${insert_method_option[ $RANDOM % ${#insert_method_option[@]} ]} + +use_insert_token_option=("True" "False") +use_insert_token=${use_insert_token_option[ $RANDOM % ${#use_insert_token_option[@]} ]} + +single_thread_options=("True" "False") +single_thread=${single_thread_options[ $RANDOM % ${#single_thread_options[@]} ]} + +deduplicate_src_table_options=("True" "False") +deduplicate_src_table=${deduplicate_src_table_options[ $RANDOM % ${#deduplicate_src_table_options[@]} ]} + +deduplicate_dst_table_options=("True" "False") +deduplicate_dst_table=${deduplicate_dst_table_options[ $RANDOM % ${#deduplicate_dst_table_options[@]} ]} + +insert_unique_blocks_options=("True" "False") +insert_unique_blocks=${insert_unique_blocks_options[ $RANDOM % ${#insert_unique_blocks_options[@]} ]} + + +THIS_RUN="Test case:" +THIS_RUN+=" insert_method=$insert_method" +THIS_RUN+=" engine=$engine" +THIS_RUN+=" use_insert_token=$use_insert_token" +THIS_RUN+=" single_thread=$single_thread" +THIS_RUN+=" deduplicate_src_table=$deduplicate_src_table" +THIS_RUN+=" deduplicate_dst_table=$deduplicate_dst_table" +THIS_RUN+=" insert_unique_blocks=$insert_unique_blocks" + +$CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python insert_several_blocks_into_table \ + --insert-method $insert_method \ + --table-engine $engine \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) +" 1>/dev/null 2>&1 && echo 'insert_several_blocks_into_table OK' || echo "FAIL: insert_several_blocks_into_table ${THIS_RUN}" + +$CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python mv_generates_several_blocks \ + --insert-method $insert_method \ + --table-engine $engine \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) +" 1>/dev/null 2>&1 && echo 'mv_generates_several_blocks OK' || echo "FAIL: mv_generates_several_blocks ${THIS_RUN}" + +$CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python several_mv_into_one_table \ + --insert-method $insert_method \ + --table-engine $engine \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) +" 1>/dev/null 2>&1 && echo 'several_mv_into_one_table OK' || echo "FAIL: several_mv_into_one_table ${THIS_RUN}" + diff --git a/tests/queries/0_stateless/03008_deduplication_remote_insert_select.reference b/tests/queries/0_stateless/03008_deduplication_remote_insert_select.reference new file mode 100644 index 00000000000..9dd45974e40 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_remote_insert_select.reference @@ -0,0 +1,36 @@ +-- { echoOn } +SELECT count() FROM src; +100 +SELECT a, sum(b), uniq(b), FROM src GROUP BY a ORDER BY a; +0 450 10 +1 460 10 +2 470 10 +3 480 10 +4 490 10 +5 500 10 +6 510 10 +7 520 10 +8 530 10 +9 540 10 +SELECT count() FROM remote('127.0.0.{1..2}', currentDatabase(), src); +200 +-- { echoOn } +INSERT INTO dst_null + SELECT a, b FROM src; +SELECT + a, + sumMerge(sum_b) AS sum_b, + uniqMerge(uniq_b) AS uniq_b +FROM mv_dst +GROUP BY a +ORDER BY a; +0 450 10 +1 460 10 +2 470 10 +3 480 10 +4 490 10 +5 500 10 +6 510 10 +7 520 10 +8 530 10 +9 540 10 diff --git a/tests/queries/0_stateless/03008_deduplication_remote_insert_select.sql b/tests/queries/0_stateless/03008_deduplication_remote_insert_select.sql new file mode 100644 index 00000000000..c8e092822da --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_remote_insert_select.sql @@ -0,0 +1,51 @@ +DROP TABLE IF EXISTS src; + +CREATE TABLE src (a UInt64, b UInt64) + ENGINE=ReplicatedMergeTree('/clickhouse/tables/{database}/03008_deduplication_remote_insert_select/src', '{replica}') + ORDER BY tuple(); + +INSERT INTO src SELECT number % 10 as a, number as b FROM numbers(100); + +SET allow_experimental_parallel_reading_from_replicas=1; +SET max_parallel_replicas=3; +SET parallel_replicas_for_non_replicated_merge_tree=1; +SET cluster_for_parallel_replicas='parallel_replicas'; + +-- { echoOn } +SELECT count() FROM src; +SELECT a, sum(b), uniq(b), FROM src GROUP BY a ORDER BY a; +SELECT count() FROM remote('127.0.0.{1..2}', currentDatabase(), src); +-- { echoOff } + +DROP TABLE IF EXISTS dst_null; +CREATE TABLE dst_null(a UInt64, b UInt64) + ENGINE = Null; + +DROP TABLE IF EXISTS mv_dst; +CREATE MATERIALIZED VIEW mv_dst + ENGINE = AggregatingMergeTree() + ORDER BY a + AS SELECT + a, + sumState(b) AS sum_b, + uniqState(b) AS uniq_b + FROM dst_null + GROUP BY a; + +-- { echoOn } +INSERT INTO dst_null + SELECT a, b FROM src; + +SELECT + a, + sumMerge(sum_b) AS sum_b, + uniqMerge(uniq_b) AS uniq_b +FROM mv_dst +GROUP BY a +ORDER BY a; +-- { echoOff } + +DROP TABLE src; +DROP TABLE mv_dst; +DROP TABLE dst_null; + diff --git a/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_nonreplicated.reference b/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_nonreplicated.reference new file mode 100644 index 00000000000..b6a3e0175a7 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_nonreplicated.reference @@ -0,0 +1,706 @@ + +Test case 0: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 1: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 16 +0 +0 +OK + +Test case 2: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 3: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 32 +0 +0 +OK + +Test case 4: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 5: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 16 +0 +0 +OK + +Test case 6: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 7: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 8: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 9: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 16 +0 +0 +OK + +Test case 10: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 11: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 32 +0 +0 +OK + +Test case 12: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 13: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 16 +0 +0 +OK + +Test case 14: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 15: insert_method=InsertSelect engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 16: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 17: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 1 +table_dst count 2 +0 +0 +table_src count 1 +table_dst count 2 +0 +0 +OK + +Test case 18: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 19: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 1 +table_dst count 16 +0 +0 +table_src count 1 +table_dst count 32 +0 +0 +OK + +Test case 20: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 21: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 2 +0 +0 +table_src count 16 +table_dst count 2 +0 +0 +OK + +Test case 22: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 23: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 24: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 25: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 1 +table_dst count 2 +0 +0 +table_src count 1 +table_dst count 2 +0 +0 +OK + +Test case 26: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 27: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 1 +table_dst count 16 +0 +0 +table_src count 1 +table_dst count 32 +0 +0 +OK + +Test case 28: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 29: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 2 +0 +0 +table_src count 16 +table_dst count 2 +0 +0 +OK + +Test case 30: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 31: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 32: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 33: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 16 +0 +0 +OK + +Test case 34: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 35: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 32 +0 +0 +OK + +Test case 36: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 37: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 16 +0 +0 +OK + +Test case 38: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 39: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 40: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 41: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 16 +0 +0 +OK + +Test case 42: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 43: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 32 +0 +0 +OK + +Test case 44: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 45: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 16 +0 +0 +OK + +Test case 46: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 47: insert_method=InsertValues engine=MergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 48: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 49: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 1 +table_dst count 2 +0 +0 +table_src count 1 +table_dst count 2 +0 +0 +OK + +Test case 50: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 51: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 1 +table_dst count 16 +0 +0 +table_src count 1 +table_dst count 32 +0 +0 +OK + +Test case 52: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 53: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 2 +0 +0 +table_src count 16 +table_dst count 2 +0 +0 +OK + +Test case 54: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 55: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 56: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 57: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 1 +table_dst count 2 +0 +0 +table_src count 1 +table_dst count 2 +0 +0 +OK + +Test case 58: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 59: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 1 +table_dst count 16 +0 +0 +table_src count 1 +table_dst count 32 +0 +0 +OK + +Test case 60: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 61: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 2 +0 +0 +table_src count 16 +table_dst count 2 +0 +0 +OK + +Test case 62: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 63: insert_method=InsertValues engine=MergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +All cases executed diff --git a/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_nonreplicated.sh b/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_nonreplicated.sh new file mode 100755 index 00000000000..f9e1838f491 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_nonreplicated.sh @@ -0,0 +1,59 @@ +#!/usr/bin/env bash +# Tags: long, no-fasttest, no-parallel, no-azure-blob-storage, no-s3-storage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +ENGINE="MergeTree" + +RUN_ONLY="" +#RUN_ONLY="Test case 17: insert_method=InsertSelect engine=MergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False" + +i=0 +for insert_method in "InsertSelect" "InsertValues"; do + for use_insert_token in "True" "False"; do + for single_thread in "True" "False"; do + for deduplicate_src_table in "True" "False"; do + for deduplicate_dst_table in "True" "False"; do + for insert_unique_blocks in "True" "False"; do + + THIS_RUN="Test case $i:" + THIS_RUN+=" insert_method=$insert_method" + THIS_RUN+=" engine=$ENGINE" + THIS_RUN+=" use_insert_token=$use_insert_token" + THIS_RUN+=" single_thread=$single_thread" + THIS_RUN+=" deduplicate_src_table=$deduplicate_src_table" + THIS_RUN+=" deduplicate_dst_table=$deduplicate_dst_table" + THIS_RUN+=" insert_unique_blocks=$insert_unique_blocks" + + i=$((i+1)) + + echo + if [ -n "$RUN_ONLY" ] && [ "$RUN_ONLY" != "$THIS_RUN" ]; then + echo "skip $THIS_RUN" + continue + fi + echo "$THIS_RUN" + + $CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python several_mv_into_one_table \ + --insert-method $insert_method \ + --table-engine $ENGINE \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) + " && echo OK || echo FAIL + done + done + done + done + done +done + +echo +echo "All cases executed" diff --git a/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_replicated.reference b/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_replicated.reference new file mode 100644 index 00000000000..1921103f49e --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_replicated.reference @@ -0,0 +1,706 @@ + +Test case 0: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 1: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 16 +0 +0 +OK + +Test case 2: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 3: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 32 +0 +0 +OK + +Test case 4: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 5: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 16 +0 +0 +OK + +Test case 6: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 7: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 8: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 9: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 16 +0 +0 +OK + +Test case 10: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 11: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 32 +0 +0 +OK + +Test case 12: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 13: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 16 +0 +0 +OK + +Test case 14: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 15: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 16: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 17: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 1 +table_dst count 2 +0 +0 +table_src count 1 +table_dst count 2 +0 +0 +OK + +Test case 18: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 19: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 1 +table_dst count 16 +0 +0 +table_src count 1 +table_dst count 32 +0 +0 +OK + +Test case 20: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 21: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 2 +0 +0 +table_src count 16 +table_dst count 2 +0 +0 +OK + +Test case 22: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 23: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 24: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 25: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 1 +table_dst count 2 +0 +0 +table_src count 1 +table_dst count 2 +0 +0 +OK + +Test case 26: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 27: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 1 +table_dst count 16 +0 +0 +table_src count 1 +table_dst count 32 +0 +0 +OK + +Test case 28: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 29: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 2 +0 +0 +table_src count 16 +table_dst count 2 +0 +0 +OK + +Test case 30: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 31: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 32: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 33: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 16 +0 +0 +OK + +Test case 34: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 35: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 32 +0 +0 +OK + +Test case 36: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 37: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 16 +0 +0 +OK + +Test case 38: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 39: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 40: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 41: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 16 +0 +0 +OK + +Test case 42: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 43: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 8 +table_dst count 32 +0 +0 +OK + +Test case 44: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 45: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 16 +0 +0 +OK + +Test case 46: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 47: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=True single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 48: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 49: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 1 +table_dst count 2 +0 +0 +table_src count 1 +table_dst count 2 +0 +0 +OK + +Test case 50: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 51: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 1 +table_dst count 16 +0 +0 +table_src count 1 +table_dst count 32 +0 +0 +OK + +Test case 52: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 53: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 2 +0 +0 +table_src count 16 +table_dst count 2 +0 +0 +OK + +Test case 54: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 55: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +Test case 56: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 6 +0 +0 +OK + +Test case 57: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False +table_src count 1 +table_dst count 2 +0 +0 +table_src count 1 +table_dst count 2 +0 +0 +OK + +Test case 58: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 8 +table_dst count 12 +0 +0 +OK + +Test case 59: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=True deduplicate_dst_table=False insert_unique_blocks=False +table_src count 1 +table_dst count 16 +0 +0 +table_src count 1 +table_dst count 32 +0 +0 +OK + +Test case 60: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 6 +0 +0 +OK + +Test case 61: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=True insert_unique_blocks=False +table_src count 8 +table_dst count 2 +0 +0 +table_src count 16 +table_dst count 2 +0 +0 +OK + +Test case 62: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=True +table_src count 8 +table_dst count 6 +0 +0 +table_src count 16 +table_dst count 12 +0 +0 +OK + +Test case 63: insert_method=InsertValues engine=ReplicatedMergeTree use_insert_token=False single_thread=False deduplicate_src_table=False deduplicate_dst_table=False insert_unique_blocks=False +table_src count 8 +table_dst count 16 +0 +0 +table_src count 16 +table_dst count 32 +0 +0 +OK + +All cases executed diff --git a/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_replicated.sh b/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_replicated.sh new file mode 100755 index 00000000000..698e70d4064 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_several_mv_into_one_table_replicated.sh @@ -0,0 +1,59 @@ +#!/usr/bin/env bash +# Tags: long, no-fasttest, no-parallel, no-azure-blob-storage, no-s3-storage + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +ENGINE="ReplicatedMergeTree" + +RUN_ONLY="" +#RUN_ONLY="Test case 17: insert_method=InsertSelect engine=ReplicatedMergeTree use_insert_token=False single_thread=True deduplicate_src_table=True deduplicate_dst_table=True insert_unique_blocks=False" + +i=0 +for insert_method in "InsertSelect" "InsertValues"; do + for use_insert_token in "True" "False"; do + for single_thread in "True" "False"; do + for deduplicate_src_table in "True" "False"; do + for deduplicate_dst_table in "True" "False"; do + for insert_unique_blocks in "True" "False"; do + + THIS_RUN="Test case $i:" + THIS_RUN+=" insert_method=$insert_method" + THIS_RUN+=" engine=$ENGINE" + THIS_RUN+=" use_insert_token=$use_insert_token" + THIS_RUN+=" single_thread=$single_thread" + THIS_RUN+=" deduplicate_src_table=$deduplicate_src_table" + THIS_RUN+=" deduplicate_dst_table=$deduplicate_dst_table" + THIS_RUN+=" insert_unique_blocks=$insert_unique_blocks" + + i=$((i+1)) + + echo + if [ -n "$RUN_ONLY" ] && [ "$RUN_ONLY" != "$THIS_RUN" ]; then + echo "skip $THIS_RUN" + continue + fi + echo "$THIS_RUN" + + $CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " + $(python3 $CURDIR/03008_deduplication.python several_mv_into_one_table \ + --insert-method $insert_method \ + --table-engine $ENGINE \ + --use-insert-token $use_insert_token \ + --single-thread $single_thread \ + --deduplicate-src-table $deduplicate_src_table \ + --deduplicate-dst-table $deduplicate_dst_table \ + --insert-unique-blocks $insert_unique_blocks \ + --get-logs false \ + ) + " && echo OK || echo FAIL + done + done + done + done + done +done + +echo +echo "All cases executed" diff --git a/tests/queries/0_stateless/03008_deduplication_wrong_mv.reference b/tests/queries/0_stateless/03008_deduplication_wrong_mv.reference new file mode 100644 index 00000000000..89b4f324d5e --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_wrong_mv.reference @@ -0,0 +1,15 @@ +-- { echo ON } +CREATE TABLE src (x UInt8) ENGINE = Memory; +CREATE TABLE dst (x UInt8) ENGINE = Memory; +CREATE MATERIALIZED VIEW mv1 TO dst AS SELECT * FROM src; +INSERT INTO src VALUES (0); +SELECT * from dst; +0 +TRUNCATE TABLE dst; +--DROP TABLE src SYNC; +--CREATE TABLE src (y String) ENGINE = MergeTree order by tuple(); +ALTER TABLE src ADD COLUMN y UInt8; +ALTER TABLE src DROP COLUMN x; +INSERT INTO src VALUES (0); +SELECT * from dst; +0 diff --git a/tests/queries/0_stateless/03008_deduplication_wrong_mv.sql b/tests/queries/0_stateless/03008_deduplication_wrong_mv.sql new file mode 100644 index 00000000000..dd74c38ab17 --- /dev/null +++ b/tests/queries/0_stateless/03008_deduplication_wrong_mv.sql @@ -0,0 +1,21 @@ +DROP TABLE IF EXISTS mv; +DROP TABLE IF EXISTS src; +DROP TABLE IF EXISTS dst; + +-- { echo ON } +CREATE TABLE src (x UInt8) ENGINE = Memory; +CREATE TABLE dst (x UInt8) ENGINE = Memory; +CREATE MATERIALIZED VIEW mv1 TO dst AS SELECT * FROM src; + +INSERT INTO src VALUES (0); +SELECT * from dst; + +TRUNCATE TABLE dst; + +--DROP TABLE src SYNC; +--CREATE TABLE src (y String) ENGINE = MergeTree order by tuple(); +ALTER TABLE src ADD COLUMN y UInt8; +ALTER TABLE src DROP COLUMN x; + +INSERT INTO src VALUES (0); +SELECT * from dst; diff --git a/tests/queries/0_stateless/03008_local_plain_rewritable.sh b/tests/queries/0_stateless/03008_local_plain_rewritable.sh index 5fac964a219..d51e180efc9 100755 --- a/tests/queries/0_stateless/03008_local_plain_rewritable.sh +++ b/tests/queries/0_stateless/03008_local_plain_rewritable.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-random-settings, no-s3-storage, no-replicated-database, no-shared-merge-tree +# Tags: no-random-settings, no-object-storage, no-replicated-database, no-shared-merge-tree # Tag no-random-settings: enable after root causing flakiness CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) diff --git a/tests/queries/0_stateless/03013_json_key_ignore_case.reference b/tests/queries/0_stateless/03013_json_key_ignore_case.reference new file mode 100644 index 00000000000..54683d8fbc5 --- /dev/null +++ b/tests/queries/0_stateless/03013_json_key_ignore_case.reference @@ -0,0 +1,3 @@ +1 77328912 Ben +2 77328913 Jim +3 77328914 Bill diff --git a/tests/queries/0_stateless/03013_json_key_ignore_case.sh b/tests/queries/0_stateless/03013_json_key_ignore_case.sh new file mode 100755 index 00000000000..7882a29ae4f --- /dev/null +++ b/tests/queries/0_stateless/03013_json_key_ignore_case.sh @@ -0,0 +1,18 @@ +#!/usr/bin/env bash + +# NOTE: this sh wrapper is required because of shell_config + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +USER_FILES_PATH=$($CLICKHOUSE_CLIENT --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep -E '^Code: 107.*FILE_DOESNT_EXIST' | head -1 | awk '{gsub("/nonexist.txt","",$9); print $9}') + +cp "$CURDIR"/data_json/key_ignore_case.json $USER_FILES_PATH/ + +$CLICKHOUSE_CLIENT -q "drop table if exists test_tbl" +$CLICKHOUSE_CLIENT -q "create table test_tbl (id UInt16, reqid UInt32, name String) engine=MergeTree order by id" +$CLICKHOUSE_CLIENT -q "INSERT INTO test_tbl SELECT * FROM file('key_ignore_case.json', 'JSONEachRow') SETTINGS input_format_json_case_insensitive_column_matching=true" +$CLICKHOUSE_CLIENT -q "select * from test_tbl" +$CLICKHOUSE_CLIENT -q "drop table test_tbl" diff --git a/tests/queries/0_stateless/03023_invalid_format_detection.reference b/tests/queries/0_stateless/03023_invalid_format_detection.reference new file mode 100644 index 00000000000..d00491fd7e5 --- /dev/null +++ b/tests/queries/0_stateless/03023_invalid_format_detection.reference @@ -0,0 +1 @@ +1 diff --git a/tests/queries/0_stateless/03023_invalid_format_detection.sh b/tests/queries/0_stateless/03023_invalid_format_detection.sh new file mode 100755 index 00000000000..984a42aa218 --- /dev/null +++ b/tests/queries/0_stateless/03023_invalid_format_detection.sh @@ -0,0 +1,10 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +touch $CLICKHOUSE_TEST_UNIQUE_NAME.xml +$CLICKHOUSE_LOCAL -q "select * from file('$CLICKHOUSE_TEST_UNIQUE_NAME.*')" 2>&1 | grep -c "CANNOT_DETECT_FORMAT" +rm $CLICKHOUSE_TEST_UNIQUE_NAME.xml + diff --git a/tests/queries/0_stateless/03030_system_flush_distributed_settings.reference b/tests/queries/0_stateless/03030_system_flush_distributed_settings.reference index 5caff40c4a0..3a05c8b3ee8 100644 --- a/tests/queries/0_stateless/03030_system_flush_distributed_settings.reference +++ b/tests/queries/0_stateless/03030_system_flush_distributed_settings.reference @@ -1 +1 @@ -10000 +30000 diff --git a/tests/queries/0_stateless/03030_system_flush_distributed_settings.sql b/tests/queries/0_stateless/03030_system_flush_distributed_settings.sql index da2a387e07c..fac673a4fe4 100644 --- a/tests/queries/0_stateless/03030_system_flush_distributed_settings.sql +++ b/tests/queries/0_stateless/03030_system_flush_distributed_settings.sql @@ -6,14 +6,17 @@ drop table if exists dist_out; create table ephemeral (key Int, value Int) engine=Null(); create table dist_in as ephemeral engine=Distributed(test_shard_localhost, currentDatabase(), ephemeral, key) settings background_insert_batch=1; -create table data (key Int, uniq_values Int) engine=Memory(); -create materialized view mv to data as select key, uniqExact(value) uniq_values from ephemeral group by key; +create table data (key Int, uniq_values Int) engine=TinyLog(); +create materialized view mv to data as select key, uniqExact(value::String) uniq_values from ephemeral group by key; system stop distributed sends dist_in; create table dist_out as data engine=Distributed(test_shard_localhost, currentDatabase(), data); set prefer_localhost_replica=0; +SET optimize_trivial_insert_select = 1; -insert into dist_in select number/100, number from system.numbers limit 1e6 settings max_memory_usage='20Mi'; +-- due to pushing to MV with aggregation the query needs ~300MiB +-- but it will be done in background via "system flush distributed" +insert into dist_in select number/100, number from system.numbers limit 3e6 settings max_block_size=3e6, max_memory_usage='100Mi'; system flush distributed dist_in; -- { serverError MEMORY_LIMIT_EXCEEDED } system flush distributed dist_in settings max_memory_usage=0; select count() from dist_out; diff --git a/tests/queries/0_stateless/03032_dynamically_resize_filesystem_cache_2.sh b/tests/queries/0_stateless/03032_dynamically_resize_filesystem_cache_2.sh index 526c4f84030..cba5317fcfa 100755 --- a/tests/queries/0_stateless/03032_dynamically_resize_filesystem_cache_2.sh +++ b/tests/queries/0_stateless/03032_dynamically_resize_filesystem_cache_2.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel, no-s3-storage, no-random-settings +# Tags: no-fasttest, no-parallel, no-object-storage, no-random-settings CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -18,7 +18,7 @@ $CLICKHOUSE_CLIENT --query "SELECT * FROM test FORMAT Null" prev_max_size=$($CLICKHOUSE_CLIENT --query "SELECT max_size FROM system.filesystem_cache_settings WHERE cache_name = '$disk_name'") $CLICKHOUSE_CLIENT --query "SELECT current_size > 0 FROM system.filesystem_cache_settings WHERE cache_name = '$disk_name' FORMAT TabSeparated" -config_path=/etc/clickhouse-server/config.d/storage_conf.xml +config_path=${CLICKHOUSE_CONFIG_DIR}/config.d/storage_conf.xml new_max_size=$($CLICKHOUSE_CLIENT --query "SELECT divide(max_size, 2) FROM system.filesystem_cache_settings WHERE cache_name = '$disk_name'") sed -i "s|$prev_max_size<\/max_size>|$new_max_size<\/max_size>|" $config_path diff --git a/tests/queries/0_stateless/03033_set_index_in.sql b/tests/queries/0_stateless/03033_set_index_in.sql index ad42a576444..bc0676fc5ef 100644 --- a/tests/queries/0_stateless/03033_set_index_in.sql +++ b/tests/queries/0_stateless/03033_set_index_in.sql @@ -1,3 +1,5 @@ +SET optimize_trivial_insert_select = 1; + create table a (k UInt64, v UInt64, index i (v) type set(100) granularity 2) engine MergeTree order by k settings index_granularity=8192, index_granularity_bytes=1000000000, min_index_granularity_bytes=0; insert into a select number, intDiv(number, 4096) from numbers(1000000); select sum(1+ignore(*)) from a where indexHint(v in (20, 40)); @@ -6,4 +8,4 @@ select sum(1+ignore(*)) from a where indexHint(v in (select 20 union all select SELECT 1 FROM a PREWHERE v IN (SELECT 1) WHERE v IN (SELECT 2); select 1 from a where indexHint(indexHint(materialize(0))); -select sum(1+ignore(*)) from a where indexHint(indexHint(v in (20, 40))); \ No newline at end of file +select sum(1+ignore(*)) from a where indexHint(indexHint(v in (20, 40))); diff --git a/tests/queries/0_stateless/03035_max_insert_threads_support.sh b/tests/queries/0_stateless/03035_max_insert_threads_support.sh index 1e6bfb414d8..cedb651a430 100755 --- a/tests/queries/0_stateless/03035_max_insert_threads_support.sh +++ b/tests/queries/0_stateless/03035_max_insert_threads_support.sh @@ -8,7 +8,7 @@ DATA_FILE="data_$CLICKHOUSE_TEST_UNIQUE_NAME.csv" $CLICKHOUSE_CLIENT --max_insert_threads=4 --query=" EXPLAIN PIPELINE INSERT INTO FUNCTION file('$DATA_FILE') SELECT * FROM numbers_mt(1000000) ORDER BY number DESC -" | grep -o MaterializingTransform | wc -l +" | grep -o StorageFileSink | wc -l DATA_FILE_PATH=$($CLICKHOUSE_CLIENT_BINARY --query "select _path from file('$DATA_FILE', 'One')") rm $DATA_FILE_PATH diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns.lib similarity index 79% rename from tests/queries/0_stateless/03036_dynamic_read_subcolumns.sh rename to tests/queries/0_stateless/03036_dynamic_read_subcolumns.lib index 65517061b99..4914051db82 100755 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns.sh +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns.lib @@ -1,14 +1,4 @@ #!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" - function test() { @@ -43,20 +33,3 @@ function test() $CH_CLIENT -q "select d.\`Array(Dynamic)\`.\`Variant(String, UInt64)\`.UInt64, d.\`Array(Dynamic)\`.size0, d.\`Array(Variant(String, UInt64))\`.UInt64 from test format Null" $CH_CLIENT -q "select d.\`Array(Array(Dynamic))\`.size1, d.\`Array(Array(Dynamic))\`.UInt64, d.\`Array(Array(Dynamic))\`.\`Map(String, Tuple(a UInt64))\`.values.a from test format Null" } - -$CH_CLIENT -q "drop table if exists test;" - -echo "Memory" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=Memory" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree compact" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" -test -$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.reference new file mode 100644 index 00000000000..0d51ecfac3b --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.reference @@ -0,0 +1,19 @@ +Memory +test +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +None +String +UInt64 +200000 +200000 +200000 +200000 +0 +0 +200000 +200000 +100000 +100000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh new file mode 100755 index 00000000000..aabba731816 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_1.sh @@ -0,0 +1,21 @@ +#!/usr/bin/env bash +# Tags: long + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# reset --log_comment +CLICKHOUSE_LOG_COMMENT= +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + + +# shellcheck source=./03036_dynamic_read_subcolumns.lib +. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib + +CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" + +$CH_CLIENT -q "drop table if exists test;" + +echo "Memory" +$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=Memory" +test +$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.reference new file mode 100644 index 00000000000..099b7574566 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.reference @@ -0,0 +1,19 @@ +MergeTree compact +test +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +None +String +UInt64 +200000 +200000 +200000 +200000 +0 +0 +200000 +200000 +100000 +100000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh new file mode 100755 index 00000000000..872f4c20a98 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_2.sh @@ -0,0 +1,21 @@ +#!/usr/bin/env bash +# Tags: long + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# reset --log_comment +CLICKHOUSE_LOG_COMMENT= +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + + +# shellcheck source=./03036_dynamic_read_subcolumns.lib +. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib + +CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" + +$CH_CLIENT -q "drop table if exists test;" + +echo "MergeTree compact" +$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000;" +test +$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.reference new file mode 100644 index 00000000000..35db4a22b4c --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.reference @@ -0,0 +1,19 @@ +MergeTree wide +test +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +None +String +UInt64 +200000 +200000 +200000 +200000 +0 +0 +200000 +200000 +100000 +100000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh new file mode 100755 index 00000000000..96276c96add --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_3.sh @@ -0,0 +1,21 @@ +#!/usr/bin/env bash +# Tags: long + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# reset --log_comment +CLICKHOUSE_LOG_COMMENT= +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + + +# shellcheck source=./03036_dynamic_read_subcolumns.lib +. "$CUR_DIR"/03036_dynamic_read_subcolumns.lib + +CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" + +$CH_CLIENT -q "drop table if exists test;" + +echo "MergeTree wide" +$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" +test +$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.reference new file mode 100644 index 00000000000..d75d75896f7 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.reference @@ -0,0 +1,17 @@ +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +None +String +UInt64 +200000 +200000 +200000 +200000 +0 +0 +200000 +200000 +100000 +100000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.sql b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.sql new file mode 100644 index 00000000000..66fbf006a8c --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_compact_merge_tree.sql @@ -0,0 +1,40 @@ +-- Tags: long + +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; + +drop table if exists test; +create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000; + +insert into test select number, number from numbers(100000) settings min_insert_block_size_rows=50000; +insert into test select number, 'str_' || toString(number) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1)) from numbers(200000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, NULL from numbers(300000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, multiIf(number % 4 == 3, 'str_' || toString(number), number % 4 == 2, NULL, number % 4 == 1, number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1))) from numbers(400000, 400000) settings min_insert_block_size_rows=50000; +insert into test select number, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; + +select distinct dynamicType(d) as type from test order by type; +select count() from test where dynamicType(d) == 'UInt64'; +select count() from test where d.UInt64 is not NULL; +select count() from test where dynamicType(d) == 'String'; +select count() from test where d.String is not NULL; +select count() from test where dynamicType(d) == 'Date'; +select count() from test where d.Date is not NULL; +select count() from test where dynamicType(d) == 'Array(Variant(String, UInt64))'; +select count() from test where not empty(d.`Array(Variant(String, UInt64))`); +select count() from test where dynamicType(d) == 'Array(Array(Dynamic))'; +select count() from test where not empty(d.`Array(Array(Dynamic))`); +select count() from test where d is NULL; +select count() from test where not empty(d.`Tuple(a Array(Dynamic))`.a.String); + +select d, d.UInt64, d.String, d.`Array(Variant(String, UInt64))` from test format Null; +select d.UInt64, d.String, d.`Array(Variant(String, UInt64))` from test format Null; +select d.Int8, d.Date, d.`Array(String)` from test format Null; +select d, d.UInt64, d.Date, d.`Array(Variant(String, UInt64))`, d.`Array(Variant(String, UInt64))`.size0, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.UInt64, d.Date, d.`Array(Variant(String, UInt64))`, d.`Array(Variant(String, UInt64))`.size0, d.`Array(Variant(String, UInt64))`.UInt64, d.`Array(Variant(String, UInt64))`.String from test format Null; +select d, d.`Tuple(a UInt64, b String)`.a, d.`Array(Dynamic)`.`Variant(String, UInt64)`.UInt64, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.`Array(Dynamic)`.`Variant(String, UInt64)`.UInt64, d.`Array(Dynamic)`.size0, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.`Array(Array(Dynamic))`.size1, d.`Array(Array(Dynamic))`.UInt64, d.`Array(Array(Dynamic))`.`Map(String, Tuple(a UInt64))`.values.a from test format Null; + +drop table test; diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_memory.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_memory.reference new file mode 100644 index 00000000000..d75d75896f7 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_memory.reference @@ -0,0 +1,17 @@ +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +None +String +UInt64 +200000 +200000 +200000 +200000 +0 +0 +200000 +200000 +100000 +100000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_memory.sql b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_memory.sql new file mode 100644 index 00000000000..bb03bdef704 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_memory.sql @@ -0,0 +1,40 @@ +-- Tags: long + +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; + +drop table if exists test; +create table test (id UInt64, d Dynamic) engine=Memory; + +insert into test select number, number from numbers(100000) settings min_insert_block_size_rows=50000; +insert into test select number, 'str_' || toString(number) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1)) from numbers(200000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, NULL from numbers(300000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, multiIf(number % 4 == 3, 'str_' || toString(number), number % 4 == 2, NULL, number % 4 == 1, number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1))) from numbers(400000, 400000) settings min_insert_block_size_rows=50000; +insert into test select number, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; + +select distinct dynamicType(d) as type from test order by type; +select count() from test where dynamicType(d) == 'UInt64'; +select count() from test where d.UInt64 is not NULL; +select count() from test where dynamicType(d) == 'String'; +select count() from test where d.String is not NULL; +select count() from test where dynamicType(d) == 'Date'; +select count() from test where d.Date is not NULL; +select count() from test where dynamicType(d) == 'Array(Variant(String, UInt64))'; +select count() from test where not empty(d.`Array(Variant(String, UInt64))`); +select count() from test where dynamicType(d) == 'Array(Array(Dynamic))'; +select count() from test where not empty(d.`Array(Array(Dynamic))`); +select count() from test where d is NULL; +select count() from test where not empty(d.`Tuple(a Array(Dynamic))`.a.String); + +select d, d.UInt64, d.String, d.`Array(Variant(String, UInt64))` from test format Null; +select d.UInt64, d.String, d.`Array(Variant(String, UInt64))` from test format Null; +select d.Int8, d.Date, d.`Array(String)` from test format Null; +select d, d.UInt64, d.Date, d.`Array(Variant(String, UInt64))`, d.`Array(Variant(String, UInt64))`.size0, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.UInt64, d.Date, d.`Array(Variant(String, UInt64))`, d.`Array(Variant(String, UInt64))`.size0, d.`Array(Variant(String, UInt64))`.UInt64, d.`Array(Variant(String, UInt64))`.String from test format Null; +select d, d.`Tuple(a UInt64, b String)`.a, d.`Array(Dynamic)`.`Variant(String, UInt64)`.UInt64, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.`Array(Dynamic)`.`Variant(String, UInt64)`.UInt64, d.`Array(Dynamic)`.size0, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.`Array(Array(Dynamic))`.size1, d.`Array(Array(Dynamic))`.UInt64, d.`Array(Array(Dynamic))`.`Map(String, Tuple(a UInt64))`.values.a from test format Null; + +drop table test; diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.reference b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.reference new file mode 100644 index 00000000000..d75d75896f7 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.reference @@ -0,0 +1,17 @@ +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +None +String +UInt64 +200000 +200000 +200000 +200000 +0 +0 +200000 +200000 +100000 +100000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.sql b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.sql new file mode 100644 index 00000000000..00aba3a57b6 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_wide_merge_tree.sql @@ -0,0 +1,40 @@ +-- Tags: long + +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; + +drop table if exists test; +create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1; + +insert into test select number, number from numbers(100000) settings min_insert_block_size_rows=50000; +insert into test select number, 'str_' || toString(number) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1)) from numbers(200000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, NULL from numbers(300000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, multiIf(number % 4 == 3, 'str_' || toString(number), number % 4 == 2, NULL, number % 4 == 1, number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1))) from numbers(400000, 400000) settings min_insert_block_size_rows=50000; +insert into test select number, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; + +select distinct dynamicType(d) as type from test order by type; +select count() from test where dynamicType(d) == 'UInt64'; +select count() from test where d.UInt64 is not NULL; +select count() from test where dynamicType(d) == 'String'; +select count() from test where d.String is not NULL; +select count() from test where dynamicType(d) == 'Date'; +select count() from test where d.Date is not NULL; +select count() from test where dynamicType(d) == 'Array(Variant(String, UInt64))'; +select count() from test where not empty(d.`Array(Variant(String, UInt64))`); +select count() from test where dynamicType(d) == 'Array(Array(Dynamic))'; +select count() from test where not empty(d.`Array(Array(Dynamic))`); +select count() from test where d is NULL; +select count() from test where not empty(d.`Tuple(a Array(Dynamic))`.a.String); + +select d, d.UInt64, d.String, d.`Array(Variant(String, UInt64))` from test format Null; +select d.UInt64, d.String, d.`Array(Variant(String, UInt64))` from test format Null; +select d.Int8, d.Date, d.`Array(String)` from test format Null; +select d, d.UInt64, d.Date, d.`Array(Variant(String, UInt64))`, d.`Array(Variant(String, UInt64))`.size0, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.UInt64, d.Date, d.`Array(Variant(String, UInt64))`, d.`Array(Variant(String, UInt64))`.size0, d.`Array(Variant(String, UInt64))`.UInt64, d.`Array(Variant(String, UInt64))`.String from test format Null; +select d, d.`Tuple(a UInt64, b String)`.a, d.`Array(Dynamic)`.`Variant(String, UInt64)`.UInt64, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.`Array(Dynamic)`.`Variant(String, UInt64)`.UInt64, d.`Array(Dynamic)`.size0, d.`Array(Variant(String, UInt64))`.UInt64 from test format Null; +select d.`Array(Array(Dynamic))`.size1, d.`Array(Array(Dynamic))`.UInt64, d.`Array(Array(Dynamic))`.`Map(String, Tuple(a UInt64))`.values.a from test format Null; + +drop table test; diff --git a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql index 3893ec41c0e..d94a68aa4d8 100644 --- a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql +++ b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql @@ -7,13 +7,13 @@ CREATE TABLE test_table_1 ( id UInt64, value String -) ENGINE=MergeTree ORDER BY id; +) ENGINE=MergeTree ORDER BY id SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; CREATE TABLE test_table_2 ( id UInt64, value String -) ENGINE=MergeTree ORDER BY id; +) ENGINE=MergeTree ORDER BY id SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; INSERT INTO test_table_1 SELECT number, number FROM numbers(10); INSERT INTO test_table_2 SELECT number, number FROM numbers(10); diff --git a/tests/queries/0_stateless/03036_reading_s3_archives.reference b/tests/queries/0_stateless/03036_reading_s3_archives.reference index 36ced212a1b..eacf16d0295 100644 --- a/tests/queries/0_stateless/03036_reading_s3_archives.reference +++ b/tests/queries/0_stateless/03036_reading_s3_archives.reference @@ -1,52 +1,52 @@ -1 Str1 example1.csv test/03036_archive1.zip::example1.csv -2 Str2 example1.csv test/03036_archive1.zip::example1.csv -3 Str3 example2.csv test/03036_archive2.zip::example2.csv -4 Str4 example2.csv test/03036_archive2.zip::example2.csv -5 Str5 example3.csv test/03036_archive2.zip::example3.csv -6 Str6 example3.csv test/03036_archive2.zip::example3.csv -3 Str3 example2.csv test/03036_archive1.zip::example2.csv -3 Str3 example2.csv test/03036_archive2.zip::example2.csv -4 Str4 example2.csv test/03036_archive1.zip::example2.csv -4 Str4 example2.csv test/03036_archive2.zip::example2.csv -1 Str1 example1.csv test/03036_archive1.zip::example1.csv -2 Str2 example1.csv test/03036_archive1.zip::example1.csv -3 Str3 example2.csv test/03036_archive1.zip::example2.csv -3 Str3 example2.csv test/03036_archive2.zip::example2.csv -4 Str4 example2.csv test/03036_archive1.zip::example2.csv -4 Str4 example2.csv test/03036_archive2.zip::example2.csv -5 Str5 example3.csv test/03036_archive2.zip::example3.csv -6 Str6 example3.csv test/03036_archive2.zip::example3.csv -1 Str1 example1.csv test/03036_archive1.tar::example1.csv -2 Str2 example1.csv test/03036_archive1.tar::example1.csv -7 Str7 example4.csv test/03036_archive1.tar::example4.csv -7 Str7 example4.csv test/03036_archive2.tar::example4.csv -8 Str8 example4.csv test/03036_archive1.tar::example4.csv -8 Str8 example4.csv test/03036_archive2.tar::example4.csv -5 Str5 example3.csv test/03036_archive2.tar::example3.csv -6 Str6 example3.csv test/03036_archive2.tar::example3.csv -7 Str7 example4.csv test/03036_archive2.tar::example4.csv -8 Str8 example4.csv test/03036_archive2.tar::example4.csv -9 Str9 example5.csv test/03036_archive2.tar::example5.csv -10 Str10 example5.csv test/03036_archive2.tar::example5.csv -3 Str3 example2.csv test/03036_archive3.tar.gz::example2.csv -4 Str4 example2.csv test/03036_archive3.tar.gz::example2.csv -11 Str11 example6.csv test/03036_archive3.tar.gz::example6.csv -12 Str12 example6.csv test/03036_archive3.tar.gz::example6.csv -3 Str3 example2.csv test/03036_archive3.tar.gz::example2.csv -4 Str4 example2.csv test/03036_archive3.tar.gz::example2.csv -5 Str5 example3.csv test/03036_archive2.tar::example3.csv -6 Str6 example3.csv test/03036_archive2.tar::example3.csv -3 Str3 example2.csv test/03036_archive2.zip::example2.csv -4 Str4 example2.csv test/03036_archive2.zip::example2.csv -5 Str5 example3.csv test/03036_archive2.tar::example3.csv -6 Str6 example3.csv test/03036_archive2.tar::example3.csv -7 Str7 example4.csv test/03036_archive2.tar::example4.csv -8 Str8 example4.csv test/03036_archive2.tar::example4.csv -9 Str9 example5.csv test/03036_archive2.tar::example5.csv -10 Str10 example5.csv test/03036_archive2.tar::example5.csv -3 Str3 example2.csv test/03036_archive3.tar.gz::example2.csv -4 Str4 example2.csv test/03036_archive3.tar.gz::example2.csv -5 Str5 example3.csv test/03036_archive2.tar::example3.csv -6 Str6 example3.csv test/03036_archive2.tar::example3.csv -13 Str13 example7.csv test/03036_compressed_file_archive.zip::example7.csv -14 Str14 example7.csv test/03036_compressed_file_archive.zip::example7.csv +1 Str1 25 example1.csv test/03036_archive1.zip::example1.csv +2 Str2 25 example1.csv test/03036_archive1.zip::example1.csv +3 Str3 25 example2.csv test/03036_archive2.zip::example2.csv +4 Str4 25 example2.csv test/03036_archive2.zip::example2.csv +5 Str5 25 example3.csv test/03036_archive2.zip::example3.csv +6 Str6 25 example3.csv test/03036_archive2.zip::example3.csv +3 Str3 25 example2.csv test/03036_archive1.zip::example2.csv +3 Str3 25 example2.csv test/03036_archive2.zip::example2.csv +4 Str4 25 example2.csv test/03036_archive1.zip::example2.csv +4 Str4 25 example2.csv test/03036_archive2.zip::example2.csv +1 Str1 25 example1.csv test/03036_archive1.zip::example1.csv +2 Str2 25 example1.csv test/03036_archive1.zip::example1.csv +3 Str3 25 example2.csv test/03036_archive1.zip::example2.csv +3 Str3 25 example2.csv test/03036_archive2.zip::example2.csv +4 Str4 25 example2.csv test/03036_archive1.zip::example2.csv +4 Str4 25 example2.csv test/03036_archive2.zip::example2.csv +5 Str5 25 example3.csv test/03036_archive2.zip::example3.csv +6 Str6 25 example3.csv test/03036_archive2.zip::example3.csv +1 Str1 25 example1.csv test/03036_archive1.tar::example1.csv +2 Str2 25 example1.csv test/03036_archive1.tar::example1.csv +7 Str7 25 example4.csv test/03036_archive1.tar::example4.csv +7 Str7 25 example4.csv test/03036_archive2.tar::example4.csv +8 Str8 25 example4.csv test/03036_archive1.tar::example4.csv +8 Str8 25 example4.csv test/03036_archive2.tar::example4.csv +5 Str5 25 example3.csv test/03036_archive2.tar::example3.csv +6 Str6 25 example3.csv test/03036_archive2.tar::example3.csv +7 Str7 25 example4.csv test/03036_archive2.tar::example4.csv +8 Str8 25 example4.csv test/03036_archive2.tar::example4.csv +9 Str9 27 example5.csv test/03036_archive2.tar::example5.csv +10 Str10 27 example5.csv test/03036_archive2.tar::example5.csv +3 Str3 25 example2.csv test/03036_archive3.tar.gz::example2.csv +4 Str4 25 example2.csv test/03036_archive3.tar.gz::example2.csv +11 Str11 29 example6.csv test/03036_archive3.tar.gz::example6.csv +12 Str12 29 example6.csv test/03036_archive3.tar.gz::example6.csv +3 Str3 25 example2.csv test/03036_archive3.tar.gz::example2.csv +4 Str4 25 example2.csv test/03036_archive3.tar.gz::example2.csv +5 Str5 25 example3.csv test/03036_archive2.tar::example3.csv +6 Str6 25 example3.csv test/03036_archive2.tar::example3.csv +3 Str3 25 example2.csv test/03036_archive2.zip::example2.csv +4 Str4 25 example2.csv test/03036_archive2.zip::example2.csv +5 Str5 25 example3.csv test/03036_archive2.tar::example3.csv +6 Str6 25 example3.csv test/03036_archive2.tar::example3.csv +7 Str7 25 example4.csv test/03036_archive2.tar::example4.csv +8 Str8 25 example4.csv test/03036_archive2.tar::example4.csv +9 Str9 27 example5.csv test/03036_archive2.tar::example5.csv +10 Str10 27 example5.csv test/03036_archive2.tar::example5.csv +3 Str3 25 example2.csv test/03036_archive3.tar.gz::example2.csv +4 Str4 25 example2.csv test/03036_archive3.tar.gz::example2.csv +5 Str5 25 example3.csv test/03036_archive2.tar::example3.csv +6 Str6 25 example3.csv test/03036_archive2.tar::example3.csv +13 Str13 57 example7.csv test/03036_compressed_file_archive.zip::example7.csv +14 Str14 57 example7.csv test/03036_compressed_file_archive.zip::example7.csv diff --git a/tests/queries/0_stateless/03036_reading_s3_archives.sql b/tests/queries/0_stateless/03036_reading_s3_archives.sql index 00d7cc25e1a..43bda4ee704 100644 --- a/tests/queries/0_stateless/03036_reading_s3_archives.sql +++ b/tests/queries/0_stateless/03036_reading_s3_archives.sql @@ -1,22 +1,22 @@ -- Tags: no-fasttest -- Tag no-fasttest: Depends on AWS -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive1.zip :: example1.csv') ORDER BY (id, _file, _path); -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive2.zip :: example*.csv') ORDER BY (id, _file, _path); -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive*.zip :: example2.csv') ORDER BY (id, _file, _path); -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive*.zip :: example*') ORDER BY (id, _file, _path); -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive1.tar :: example1.csv') ORDER BY (id, _file, _path); -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive*.tar :: example4.csv') ORDER BY (id, _file, _path); -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive2.tar :: example*.csv') ORDER BY (id, _file, _path); -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive*.tar.gz :: example*.csv') ORDER BY (id, _file, _path); -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_archive*.tar* :: example{2..3}.csv') ORDER BY (id, _file, _path); -select id, data, _file, _path from s3(s3_conn, filename='03036_archive2.zip :: nonexistent.csv'); -- { serverError CANNOT_EXTRACT_TABLE_STRUCTURE } -select id, data, _file, _path from s3(s3_conn, filename='03036_archive2.zip :: nonexistent{2..3}.csv'); -- { serverError CANNOT_EXTRACT_TABLE_STRUCTURE } +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive1.zip :: example1.csv') ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive2.zip :: example*.csv') ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive*.zip :: example2.csv') ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive*.zip :: example*') ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive1.tar :: example1.csv') ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive*.tar :: example4.csv') ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive2.tar :: example*.csv') ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive*.tar.gz :: example*.csv') ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_archive*.tar* :: example{2..3}.csv') ORDER BY (id, _file, _path); +select id, data, _size, _file, _path from s3(s3_conn, filename='03036_archive2.zip :: nonexistent.csv'); -- { serverError CANNOT_EXTRACT_TABLE_STRUCTURE } +select id, data, _size, _file, _path from s3(s3_conn, filename='03036_archive2.zip :: nonexistent{2..3}.csv'); -- { serverError CANNOT_EXTRACT_TABLE_STRUCTURE } CREATE TABLE table_zip22 Engine S3(s3_conn, filename='03036_archive2.zip :: example2.csv'); -select id, data, _file, _path from table_zip22 ORDER BY (id, _file, _path); +select id, data, _size, _file, _path from table_zip22 ORDER BY (id, _file, _path); CREATE table table_tar2star Engine S3(s3_conn, filename='03036_archive2.tar :: example*.csv'); -SELECT id, data, _file, _path FROM table_tar2star ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM table_tar2star ORDER BY (id, _file, _path); CREATE table table_tarstarglobs Engine S3(s3_conn, filename='03036_archive*.tar* :: example{2..3}.csv'); -SELECT id, data, _file, _path FROM table_tarstarglobs ORDER BY (id, _file, _path); +SELECT id, data, _size, _file, _path FROM table_tarstarglobs ORDER BY (id, _file, _path); CREATE table table_noexist Engine s3(s3_conn, filename='03036_archive2.zip :: nonexistent.csv'); -- { serverError UNKNOWN_STORAGE } -SELECT id, data, _file, _path FROM s3(s3_conn, filename='03036_compressed_file_archive.zip :: example7.csv', format='CSV', structure='auto', compression_method='gz') ORDER BY (id, _file, _path) +SELECT id, data, _size, _file, _path FROM s3(s3_conn, filename='03036_compressed_file_archive.zip :: example7.csv', format='CSV', structure='auto', compression_method='gz') ORDER BY (id, _file, _path) diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal.reference b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal.reference deleted file mode 100644 index 59297e46330..00000000000 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal.reference +++ /dev/null @@ -1,60 +0,0 @@ -MergeTree compact -test -50000 DateTime -60000 Date -70000 Array(UInt16) -80000 String -100000 None -100000 UInt64 -70000 Array(UInt16) -100000 None -100000 UInt64 -190000 String -70000 Array(UInt16) -100000 None -100000 UInt64 -190000 String -200000 Map(UInt64, UInt64) -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -260000 String -10000 Tuple(UInt64, UInt64) -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -260000 String -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -270000 String -MergeTree wide -test -50000 DateTime -60000 Date -70000 Array(UInt16) -80000 String -100000 None -100000 UInt64 -70000 Array(UInt16) -100000 None -100000 UInt64 -190000 String -70000 Array(UInt16) -100000 None -100000 UInt64 -190000 String -200000 Map(UInt64, UInt64) -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -260000 String -10000 Tuple(UInt64, UInt64) -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -260000 String -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -270000 String diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal.sh b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal.sh deleted file mode 100755 index 887b2ed94d7..00000000000 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal.sh +++ /dev/null @@ -1,52 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_dynamic_type=1" - -function test() -{ - echo "test" - $CH_CLIENT -q "system stop merges test" - $CH_CLIENT -q "insert into test select number, number from numbers(100000)" - $CH_CLIENT -q "insert into test select number, 'str_' || toString(number) from numbers(80000)" - $CH_CLIENT -q "insert into test select number, range(number % 10 + 1) from numbers(70000)" - $CH_CLIENT -q "insert into test select number, toDate(number) from numbers(60000)" - $CH_CLIENT -q "insert into test select number, toDateTime(number) from numbers(50000)" - $CH_CLIENT -q "insert into test select number, NULL from numbers(100000)" - - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - - $CH_CLIENT -q "system stop merges test" - $CH_CLIENT -q "insert into test select number, map(number, number) from numbers(200000)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - - $CH_CLIENT -q "system stop merges test" - $CH_CLIENT -q "insert into test select number, tuple(number, number) from numbers(10000)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" -} - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree compact" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_columns_to_activate=10, index_granularity_bytes=10485760, index_granularity=8192, merge_max_block_size=8192, merge_max_block_size_bytes=10485760;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_columns_to_activate=10, index_granularity_bytes=10485760, index_granularity=8192, merge_max_block_size=8192, merge_max_block_size_bytes=10485760;" -test -$CH_CLIENT -q "drop table test;" - diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.reference new file mode 100644 index 00000000000..d0d777a5a38 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.reference @@ -0,0 +1,28 @@ +50000 DateTime +60000 Date +70000 Array(UInt16) +80000 String +100000 None +100000 UInt64 +70000 Array(UInt16) +100000 None +100000 UInt64 +190000 String +70000 Array(UInt16) +100000 None +100000 UInt64 +190000 String +200000 Map(UInt64, UInt64) +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +260000 String +10000 Tuple(UInt64, UInt64) +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +260000 String +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +270000 String diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql new file mode 100644 index 00000000000..07371ee099b --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_merge_tree.sql @@ -0,0 +1,33 @@ +-- Tags: long +set allow_experimental_dynamic_type=1; + +drop table if exists test; +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_columns_to_activate=10, index_granularity_bytes=10485760, index_granularity=8192, merge_max_block_size=8192, merge_max_block_size_bytes=10485760, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(100000); +insert into test select number, 'str_' || toString(number) from numbers(80000); +insert into test select number, range(number % 10 + 1) from numbers(70000); +insert into test select number, toDate(number) from numbers(60000); +insert into test select number, toDateTime(number) from numbers(50000); +insert into test select number, NULL from numbers(100000); + +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; optimize table test final;; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, map(number, number) from numbers(200000); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, tuple(number, number) from numbers(10000); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +drop table test; \ No newline at end of file diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.reference new file mode 100644 index 00000000000..d0d777a5a38 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.reference @@ -0,0 +1,28 @@ +50000 DateTime +60000 Date +70000 Array(UInt16) +80000 String +100000 None +100000 UInt64 +70000 Array(UInt16) +100000 None +100000 UInt64 +190000 String +70000 Array(UInt16) +100000 None +100000 UInt64 +190000 String +200000 Map(UInt64, UInt64) +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +260000 String +10000 Tuple(UInt64, UInt64) +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +260000 String +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +270000 String diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql new file mode 100644 index 00000000000..2b55a31e937 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_horizontal_compact_wide_tree.sql @@ -0,0 +1,33 @@ +-- Tags: long +set allow_experimental_dynamic_type=1; + +drop table if exists test; +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_columns_to_activate=10, index_granularity_bytes=10485760, index_granularity=8192, merge_max_block_size=8192, merge_max_block_size_bytes=10485760, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(100000); +insert into test select number, 'str_' || toString(number) from numbers(80000); +insert into test select number, range(number % 10 + 1) from numbers(70000); +insert into test select number, toDate(number) from numbers(60000); +insert into test select number, toDateTime(number) from numbers(50000); +insert into test select number, NULL from numbers(100000); + +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; optimize table test final;; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, map(number, number) from numbers(200000); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, tuple(number, number) from numbers(10000); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +drop table test; \ No newline at end of file diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical.reference b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical.reference deleted file mode 100644 index 59297e46330..00000000000 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical.reference +++ /dev/null @@ -1,60 +0,0 @@ -MergeTree compact -test -50000 DateTime -60000 Date -70000 Array(UInt16) -80000 String -100000 None -100000 UInt64 -70000 Array(UInt16) -100000 None -100000 UInt64 -190000 String -70000 Array(UInt16) -100000 None -100000 UInt64 -190000 String -200000 Map(UInt64, UInt64) -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -260000 String -10000 Tuple(UInt64, UInt64) -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -260000 String -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -270000 String -MergeTree wide -test -50000 DateTime -60000 Date -70000 Array(UInt16) -80000 String -100000 None -100000 UInt64 -70000 Array(UInt16) -100000 None -100000 UInt64 -190000 String -70000 Array(UInt16) -100000 None -100000 UInt64 -190000 String -200000 Map(UInt64, UInt64) -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -260000 String -10000 Tuple(UInt64, UInt64) -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -260000 String -100000 None -100000 UInt64 -200000 Map(UInt64, UInt64) -270000 String diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical.sh b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical.sh deleted file mode 100755 index 371ae87c2ef..00000000000 --- a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical.sh +++ /dev/null @@ -1,51 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - - - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_dynamic_type=1" -function test() -{ - echo "test" - $CH_CLIENT -q "system stop merges test" - $CH_CLIENT -q "insert into test select number, number from numbers(100000)" - $CH_CLIENT -q "insert into test select number, 'str_' || toString(number) from numbers(80000)" - $CH_CLIENT -q "insert into test select number, range(number % 10 + 1) from numbers(70000)" - $CH_CLIENT -q "insert into test select number, toDate(number) from numbers(60000)" - $CH_CLIENT -q "insert into test select number, toDateTime(number) from numbers(50000)" - $CH_CLIENT -q "insert into test select number, NULL from numbers(100000)" - - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - - $CH_CLIENT -q "system stop merges test" - $CH_CLIENT -q "insert into test select number, map(number, number) from numbers(200000)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - - $CH_CLIENT -q "system stop merges test" - $CH_CLIENT -q "insert into test select number, tuple(number, number) from numbers(10000)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" -} - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree compact" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, index_granularity_bytes=10485760, index_granularity=8192, merge_max_block_size=8192, merge_max_block_size_bytes=10485760;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, index_granularity_bytes=10485760, index_granularity=8192, merge_max_block_size=8192, merge_max_block_size_bytes=10485760;" -test -$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.reference new file mode 100644 index 00000000000..d0d777a5a38 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.reference @@ -0,0 +1,28 @@ +50000 DateTime +60000 Date +70000 Array(UInt16) +80000 String +100000 None +100000 UInt64 +70000 Array(UInt16) +100000 None +100000 UInt64 +190000 String +70000 Array(UInt16) +100000 None +100000 UInt64 +190000 String +200000 Map(UInt64, UInt64) +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +260000 String +10000 Tuple(UInt64, UInt64) +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +260000 String +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +270000 String diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql new file mode 100644 index 00000000000..ea7295a9eab --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_compact_merge_tree.sql @@ -0,0 +1,34 @@ +-- Tags: long +set allow_experimental_dynamic_type=1; + +drop table if exists test; +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, index_granularity_bytes=10485760, index_granularity=8192, merge_max_block_size=8192, merge_max_block_size_bytes=10485760, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(100000); +insert into test select number, 'str_' || toString(number) from numbers(80000); +insert into test select number, range(number % 10 + 1) from numbers(70000); +insert into test select number, toDate(number) from numbers(60000); +insert into test select number, toDateTime(number) from numbers(50000); +insert into test select number, NULL from numbers(100000); + +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, map(number, number) from numbers(200000); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, tuple(number, number) from numbers(10000); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +drop table test; \ No newline at end of file diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.reference new file mode 100644 index 00000000000..d0d777a5a38 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.reference @@ -0,0 +1,28 @@ +50000 DateTime +60000 Date +70000 Array(UInt16) +80000 String +100000 None +100000 UInt64 +70000 Array(UInt16) +100000 None +100000 UInt64 +190000 String +70000 Array(UInt16) +100000 None +100000 UInt64 +190000 String +200000 Map(UInt64, UInt64) +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +260000 String +10000 Tuple(UInt64, UInt64) +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +260000 String +100000 None +100000 UInt64 +200000 Map(UInt64, UInt64) +270000 String diff --git a/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql new file mode 100644 index 00000000000..e888a14b323 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_1_vertical_wide_merge_tree.sql @@ -0,0 +1,33 @@ +-- Tags: long +set allow_experimental_dynamic_type=1; + +drop table if exists test; +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, index_granularity_bytes=10485760, index_granularity=8192, merge_max_block_size=8192, merge_max_block_size_bytes=10485760, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(100000); +insert into test select number, 'str_' || toString(number) from numbers(80000); +insert into test select number, range(number % 10 + 1) from numbers(70000); +insert into test select number, toDate(number) from numbers(60000); +insert into test select number, toDateTime(number) from numbers(50000); +insert into test select number, NULL from numbers(100000); + +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; optimize table test final;; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, map(number, number) from numbers(200000); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, tuple(number, number) from numbers(10000); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); + +drop table test; \ No newline at end of file diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2.reference b/tests/queries/0_stateless/03037_dynamic_merges_2.reference deleted file mode 100644 index 420b8185b16..00000000000 --- a/tests/queries/0_stateless/03037_dynamic_merges_2.reference +++ /dev/null @@ -1,20 +0,0 @@ -MergeTree compact + horizontal merge -test -1000000 Array(UInt16) -1000000 String -1000000 UInt64 -MergeTree wide + horizontal merge -test -1000000 Array(UInt16) -1000000 String -1000000 UInt64 -MergeTree compact + vertical merge -test -1000000 Array(UInt16) -1000000 String -1000000 UInt64 -MergeTree wide + vertical merge -test -1000000 Array(UInt16) -1000000 String -1000000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2.sh b/tests/queries/0_stateless/03037_dynamic_merges_2.sh deleted file mode 100755 index 40adbdd4262..00000000000 --- a/tests/queries/0_stateless/03037_dynamic_merges_2.sh +++ /dev/null @@ -1,45 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_dynamic_type=1" - - -function test() -{ - echo "test" - $CH_CLIENT -q "system stop merges test" - $CH_CLIENT -q "insert into test select number, number from numbers(1000000)" - $CH_CLIENT -q "insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000)" - $CH_CLIENT -q "insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000)" - - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" -} - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree compact + horizontal merge" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide + horizontal merge" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree compact + vertical merge" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide + vertical merge" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1;" -test -$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.reference new file mode 100644 index 00000000000..afd392002e5 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.reference @@ -0,0 +1,3 @@ +1000000 Array(UInt16) +1000000 String +1000000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql new file mode 100644 index 00000000000..e633b277ebd --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_compact_merge_tree.sql @@ -0,0 +1,14 @@ +-- Tags: long + +set allow_experimental_dynamic_type = 1; + +drop table if exists test; +create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, lock_acquire_timeout_for_background_operations=600; +system stop merges test; +insert into test select number, number from numbers(1000000); +insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000); +insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +drop table test; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.reference new file mode 100644 index 00000000000..afd392002e5 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.reference @@ -0,0 +1,3 @@ +1000000 Array(UInt16) +1000000 String +1000000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql new file mode 100644 index 00000000000..90dbc2d84f5 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_horizontal_wide_merge_tree.sql @@ -0,0 +1,14 @@ +-- Tags: long + +set allow_experimental_dynamic_type = 1; + +drop table if exists test; +create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, lock_acquire_timeout_for_background_operations=600; +system stop merges test; +insert into test select number, number from numbers(1000000); +insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000); +insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +drop table test; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.reference new file mode 100644 index 00000000000..afd392002e5 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.reference @@ -0,0 +1,3 @@ +1000000 Array(UInt16) +1000000 String +1000000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql new file mode 100644 index 00000000000..ffd2618ee51 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_compact_merge_tree.sql @@ -0,0 +1,14 @@ +-- Tags: long + +set allow_experimental_dynamic_type = 1; + +drop table if exists test; +create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, lock_acquire_timeout_for_background_operations=600; +system stop merges test; +insert into test select number, number from numbers(1000000); +insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000); +insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +drop table test; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.reference b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.reference new file mode 100644 index 00000000000..afd392002e5 --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.reference @@ -0,0 +1,3 @@ +1000000 Array(UInt16) +1000000 String +1000000 UInt64 diff --git a/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql new file mode 100644 index 00000000000..36dff88751b --- /dev/null +++ b/tests/queries/0_stateless/03037_dynamic_merges_2_vertical_wide_merge_tree.sql @@ -0,0 +1,14 @@ +-- Tags: long + +set allow_experimental_dynamic_type = 1; + +drop table if exists test; +create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, lock_acquire_timeout_for_background_operations=600; +system stop merges test; +insert into test select number, number from numbers(1000000); +insert into test select number, 'str_' || toString(number) from numbers(1000000, 1000000); +insert into test select number, range(number % 10 + 1) from numbers(2000000, 1000000); +system start merges test; +optimize table test final; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +drop table test; diff --git a/tests/queries/0_stateless/03038_move_partition_to_oneself_deadlock.sql b/tests/queries/0_stateless/03038_move_partition_to_oneself_deadlock.sql index 6eefa5270c5..f3072fb3539 100644 --- a/tests/queries/0_stateless/03038_move_partition_to_oneself_deadlock.sql +++ b/tests/queries/0_stateless/03038_move_partition_to_oneself_deadlock.sql @@ -1,3 +1,5 @@ +SET optimize_trivial_insert_select = 1; + DROP TABLE IF EXISTS move_partition_to_oneself; CREATE TABLE move_partition_to_oneself (key UInt64 CODEC(NONE)) ENGINE = MergeTree ORDER BY tuple(); INSERT INTO move_partition_to_oneself SELECT number FROM numbers(1e6); diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges.reference b/tests/queries/0_stateless/03038_nested_dynamic_merges.reference deleted file mode 100644 index 65034647775..00000000000 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges.reference +++ /dev/null @@ -1,92 +0,0 @@ -MergeTree compact + horizontal merge -test -16667 Tuple(a Dynamic(max_types=3)):Date -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):String -50000 Tuple(a Dynamic(max_types=3)):UInt64 -100000 UInt64:None -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):UInt64 -66667 Tuple(a Dynamic(max_types=3)):String -100000 UInt64:None -16667 Tuple(a Dynamic(max_types=3)):DateTime -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):UInt64 -66667 Tuple(a Dynamic(max_types=3)):String -100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -100000 UInt64:None -133333 Tuple(a Dynamic(max_types=3)):None -50000 Tuple(a Dynamic(max_types=3)):UInt64 -100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -100000 UInt64:None -116667 Tuple(a Dynamic(max_types=3)):String -133333 Tuple(a Dynamic(max_types=3)):None -MergeTree wide + horizontal merge -test -16667 Tuple(a Dynamic(max_types=3)):Date -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):String -50000 Tuple(a Dynamic(max_types=3)):UInt64 -100000 UInt64:None -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):UInt64 -66667 Tuple(a Dynamic(max_types=3)):String -100000 UInt64:None -16667 Tuple(a Dynamic(max_types=3)):DateTime -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):UInt64 -66667 Tuple(a Dynamic(max_types=3)):String -100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -100000 UInt64:None -133333 Tuple(a Dynamic(max_types=3)):None -50000 Tuple(a Dynamic(max_types=3)):UInt64 -100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -100000 UInt64:None -116667 Tuple(a Dynamic(max_types=3)):String -133333 Tuple(a Dynamic(max_types=3)):None -MergeTree compact + vertical merge -test -16667 Tuple(a Dynamic(max_types=3)):Date -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):String -50000 Tuple(a Dynamic(max_types=3)):UInt64 -100000 UInt64:None -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):UInt64 -66667 Tuple(a Dynamic(max_types=3)):String -100000 UInt64:None -16667 Tuple(a Dynamic(max_types=3)):DateTime -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):UInt64 -66667 Tuple(a Dynamic(max_types=3)):String -100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -100000 UInt64:None -133333 Tuple(a Dynamic(max_types=3)):None -50000 Tuple(a Dynamic(max_types=3)):UInt64 -100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -100000 UInt64:None -116667 Tuple(a Dynamic(max_types=3)):String -133333 Tuple(a Dynamic(max_types=3)):None -MergeTree wide + vertical merge -test -16667 Tuple(a Dynamic(max_types=3)):Date -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):String -50000 Tuple(a Dynamic(max_types=3)):UInt64 -100000 UInt64:None -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):UInt64 -66667 Tuple(a Dynamic(max_types=3)):String -100000 UInt64:None -16667 Tuple(a Dynamic(max_types=3)):DateTime -33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) -50000 Tuple(a Dynamic(max_types=3)):UInt64 -66667 Tuple(a Dynamic(max_types=3)):String -100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -100000 UInt64:None -133333 Tuple(a Dynamic(max_types=3)):None -50000 Tuple(a Dynamic(max_types=3)):UInt64 -100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -100000 UInt64:None -116667 Tuple(a Dynamic(max_types=3)):String -133333 Tuple(a Dynamic(max_types=3)):None diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges.sh b/tests/queries/0_stateless/03038_nested_dynamic_merges.sh deleted file mode 100755 index b82ddb3813e..00000000000 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges.sh +++ /dev/null @@ -1,53 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1" - - -function test() -{ - echo "test" - $CH_CLIENT -q "system stop merges test" - $CH_CLIENT -q "insert into test select number, number from numbers(100000)" - $CH_CLIENT -q "insert into test select number, tuple(if(number % 2 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=3)) from numbers(100000)" - $CH_CLIENT -q "insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=3)) from numbers(50000)" - - $CH_CLIENT -q "select count(), dynamicType(d) || ':' || dynamicType(d.\`Tuple(a Dynamic(max_types=3))\`.a) as type from test group by type order by count(), type" - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) || ':' || dynamicType(d.\`Tuple(a Dynamic(max_types=3))\`.a) as type from test group by type order by count(), type" - - $CH_CLIENT -q "insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(50000)" - $CH_CLIENT -q "insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(200000)" - - $CH_CLIENT -q "select count(), dynamicType(d) || ':' || dynamicType(d.\`Tuple(a Dynamic(max_types=3))\`.a) as type from test group by type order by count(), type" - $CH_CLIENT -nm -q "system start merges test; optimize table test final;" - $CH_CLIENT -q "select count(), dynamicType(d) || ':' || dynamicType(d.\`Tuple(a Dynamic(max_types=3))\`.a) as type from test group by type order by count(), type" -} - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree compact + horizontal merge" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide + horizontal merge" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree compact + vertical merge" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1;" -test -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide + vertical merge" -$CH_CLIENT -q "create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1;" -test -$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.reference b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.reference new file mode 100644 index 00000000000..4be740f6050 --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.reference @@ -0,0 +1,21 @@ +16667 Tuple(a Dynamic(max_types=3)):Date +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):String +50000 Tuple(a Dynamic(max_types=3)):UInt64 +100000 UInt64:None +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):UInt64 +66667 Tuple(a Dynamic(max_types=3)):String +100000 UInt64:None +16667 Tuple(a Dynamic(max_types=3)):DateTime +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):UInt64 +66667 Tuple(a Dynamic(max_types=3)):String +100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +100000 UInt64:None +133333 Tuple(a Dynamic(max_types=3)):None +50000 Tuple(a Dynamic(max_types=3)):UInt64 +100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +100000 UInt64:None +116667 Tuple(a Dynamic(max_types=3)):String +133333 Tuple(a Dynamic(max_types=3)):None diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql new file mode 100644 index 00000000000..1d5c63dcdf1 --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql @@ -0,0 +1,29 @@ +-- Tags: long + +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; +set enable_named_columns_in_function_tuple = 0; + +drop table if exists test;; +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(100000); +insert into test select number, tuple(if(number % 2 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=3)) from numbers(100000); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=3)) from numbers(50000); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(50000); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(200000); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +drop table test; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.reference b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.reference new file mode 100644 index 00000000000..4be740f6050 --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.reference @@ -0,0 +1,21 @@ +16667 Tuple(a Dynamic(max_types=3)):Date +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):String +50000 Tuple(a Dynamic(max_types=3)):UInt64 +100000 UInt64:None +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):UInt64 +66667 Tuple(a Dynamic(max_types=3)):String +100000 UInt64:None +16667 Tuple(a Dynamic(max_types=3)):DateTime +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):UInt64 +66667 Tuple(a Dynamic(max_types=3)):String +100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +100000 UInt64:None +133333 Tuple(a Dynamic(max_types=3)):None +50000 Tuple(a Dynamic(max_types=3)):UInt64 +100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +100000 UInt64:None +116667 Tuple(a Dynamic(max_types=3)):String +133333 Tuple(a Dynamic(max_types=3)):None diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql new file mode 100644 index 00000000000..2bffe35c577 --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql @@ -0,0 +1,29 @@ +-- Tags: long + +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; +set enable_named_columns_in_function_tuple = 0; + +drop table if exists test;; +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(100000); +insert into test select number, tuple(if(number % 2 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=3)) from numbers(100000); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=3)) from numbers(50000); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(50000); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(200000); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +drop table test; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.reference b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.reference new file mode 100644 index 00000000000..4be740f6050 --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.reference @@ -0,0 +1,21 @@ +16667 Tuple(a Dynamic(max_types=3)):Date +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):String +50000 Tuple(a Dynamic(max_types=3)):UInt64 +100000 UInt64:None +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):UInt64 +66667 Tuple(a Dynamic(max_types=3)):String +100000 UInt64:None +16667 Tuple(a Dynamic(max_types=3)):DateTime +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):UInt64 +66667 Tuple(a Dynamic(max_types=3)):String +100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +100000 UInt64:None +133333 Tuple(a Dynamic(max_types=3)):None +50000 Tuple(a Dynamic(max_types=3)):UInt64 +100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +100000 UInt64:None +116667 Tuple(a Dynamic(max_types=3)):String +133333 Tuple(a Dynamic(max_types=3)):None diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql new file mode 100644 index 00000000000..fb686091ebb --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql @@ -0,0 +1,29 @@ +-- Tags: long + +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; +set enable_named_columns_in_function_tuple = 0; + +drop table if exists test;; +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(100000); +insert into test select number, tuple(if(number % 2 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=3)) from numbers(100000); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=3)) from numbers(50000); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(50000); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(200000); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +drop table test; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.reference b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.reference new file mode 100644 index 00000000000..4be740f6050 --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.reference @@ -0,0 +1,21 @@ +16667 Tuple(a Dynamic(max_types=3)):Date +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):String +50000 Tuple(a Dynamic(max_types=3)):UInt64 +100000 UInt64:None +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):UInt64 +66667 Tuple(a Dynamic(max_types=3)):String +100000 UInt64:None +16667 Tuple(a Dynamic(max_types=3)):DateTime +33333 Tuple(a Dynamic(max_types=3)):Array(UInt8) +50000 Tuple(a Dynamic(max_types=3)):UInt64 +66667 Tuple(a Dynamic(max_types=3)):String +100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +100000 UInt64:None +133333 Tuple(a Dynamic(max_types=3)):None +50000 Tuple(a Dynamic(max_types=3)):UInt64 +100000 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) +100000 UInt64:None +116667 Tuple(a Dynamic(max_types=3)):String +133333 Tuple(a Dynamic(max_types=3)):None diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql new file mode 100644 index 00000000000..ed195452d56 --- /dev/null +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql @@ -0,0 +1,29 @@ +-- Tags: long + +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; +set enable_named_columns_in_function_tuple = 0; + +drop table if exists test;; +create table test (id UInt64, d Dynamic(max_types=3)) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1, vertical_merge_algorithm_min_rows_to_activate=1, vertical_merge_algorithm_min_columns_to_activate=1, lock_acquire_timeout_for_background_operations=600; + +system stop merges test; +insert into test select number, number from numbers(100000); +insert into test select number, tuple(if(number % 2 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=3)) from numbers(100000); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=3)) from numbers(50000); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(50000); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(200000); + +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +system start merges test; +optimize table test final; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; + +drop table test; diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_1.reference b/tests/queries/0_stateless/03040_dynamic_type_alters_1.reference deleted file mode 100644 index ca98ec0963c..00000000000 --- a/tests/queries/0_stateless/03040_dynamic_type_alters_1.reference +++ /dev/null @@ -1,526 +0,0 @@ -Memory -initial insert -alter add column 1 -3 None -0 0 \N \N \N \N -1 1 \N \N \N \N -2 2 \N \N \N \N -insert after alter add column 1 -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -alter modify column 1 -7 None -8 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -insert after alter modify column 1 -8 None -11 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -alter modify column 2 -4 UInt64 -7 String -8 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -insert after alter modify column 2 -1 Date -5 UInt64 -8 String -9 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -19 19 \N \N \N \N \N -20 20 20 \N 20 \N \N -21 21 str_21 str_21 \N \N \N -22 22 1970-01-23 \N \N 1970-01-23 \N -alter modify column 3 -1 Date -5 UInt64 -8 String -9 None -0 0 0 \N \N \N \N \N \N -1 1 1 \N \N \N \N \N \N -2 2 2 \N \N \N \N \N \N -3 3 3 \N \N \N 3 \N \N -4 4 4 \N \N \N 4 \N \N -5 5 5 \N \N \N 5 \N \N -6 6 6 \N \N str_6 \N \N \N -7 7 7 \N \N str_7 \N \N \N -8 8 8 \N \N str_8 \N \N \N -9 9 9 \N \N \N \N \N \N -10 10 10 \N \N \N \N \N \N -11 11 11 \N \N \N \N \N \N -12 12 12 \N \N \N 12 \N \N -13 13 13 \N \N str_13 \N \N \N -14 14 14 \N \N \N \N \N \N -15 15 15 \N \N \N \N \N \N -16 16 16 \N \N 16 \N \N \N -17 17 17 \N \N str_17 \N \N \N -18 18 18 \N \N 1970-01-19 \N \N \N -19 19 19 \N \N \N \N \N \N -20 20 20 \N \N \N 20 \N \N -21 21 21 \N \N str_21 \N \N \N -22 22 22 \N \N \N \N 1970-01-23 \N -insert after alter modify column 3 -1 Date -5 UInt64 -8 String -12 None -0 0 0 \N \N \N \N \N \N -1 1 1 \N \N \N \N \N \N -2 2 2 \N \N \N \N \N \N -3 3 3 \N \N \N 3 \N \N -4 4 4 \N \N \N 4 \N \N -5 5 5 \N \N \N 5 \N \N -6 6 6 \N \N str_6 \N \N \N -7 7 7 \N \N str_7 \N \N \N -8 8 8 \N \N str_8 \N \N \N -9 9 9 \N \N \N \N \N \N -10 10 10 \N \N \N \N \N \N -11 11 11 \N \N \N \N \N \N -12 12 12 \N \N \N 12 \N \N -13 13 13 \N \N str_13 \N \N \N -14 14 14 \N \N \N \N \N \N -15 15 15 \N \N \N \N \N \N -16 16 16 \N \N 16 \N \N \N -17 17 17 \N \N str_17 \N \N \N -18 18 18 \N \N 1970-01-19 \N \N \N -19 19 19 \N \N \N \N \N \N -20 20 20 \N \N \N 20 \N \N -21 21 21 \N \N str_21 \N \N \N -22 22 22 \N \N \N \N 1970-01-23 \N -23 \N \N \N \N \N \N \N \N -24 24 24 \N \N \N \N \N \N -25 str_25 \N str_25 \N \N \N \N \N -MergeTree compact -initial insert -alter add column 1 -3 None -0 0 \N \N \N \N -1 1 \N \N \N \N -2 2 \N \N \N \N -insert after alter add column 1 -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -alter modify column 1 -7 None -8 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -insert after alter modify column 1 -8 None -11 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -alter modify column 2 -8 None -11 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -insert after alter modify column 2 -1 Date -1 UInt64 -9 None -12 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -19 19 \N \N \N \N \N -20 20 20 \N 20 \N \N -21 21 str_21 str_21 \N \N \N -22 22 1970-01-23 \N \N 1970-01-23 \N -alter modify column 3 -1 Date -1 UInt64 -9 None -12 String -0 0 0 \N \N \N \N \N \N -1 1 1 \N \N \N \N \N \N -2 2 2 \N \N \N \N \N \N -3 3 3 \N \N 3 \N \N \N -4 4 4 \N \N 4 \N \N \N -5 5 5 \N \N 5 \N \N \N -6 6 6 \N \N str_6 \N \N \N -7 7 7 \N \N str_7 \N \N \N -8 8 8 \N \N str_8 \N \N \N -9 9 9 \N \N \N \N \N \N -10 10 10 \N \N \N \N \N \N -11 11 11 \N \N \N \N \N \N -12 12 12 \N \N 12 \N \N \N -13 13 13 \N \N str_13 \N \N \N -14 14 14 \N \N \N \N \N \N -15 15 15 \N \N \N \N \N \N -16 16 16 \N \N 16 \N \N \N -17 17 17 \N \N str_17 \N \N \N -18 18 18 \N \N 1970-01-19 \N \N \N -19 19 19 \N \N \N \N \N \N -20 20 20 \N \N \N 20 \N \N -21 21 21 \N \N str_21 \N \N \N -22 22 22 \N \N \N \N 1970-01-23 \N -insert after alter modify column 3 -1 Date -1 UInt64 -12 None -12 String -0 0 0 \N \N \N \N \N \N -1 1 1 \N \N \N \N \N \N -2 2 2 \N \N \N \N \N \N -3 3 3 \N \N 3 \N \N \N -4 4 4 \N \N 4 \N \N \N -5 5 5 \N \N 5 \N \N \N -6 6 6 \N \N str_6 \N \N \N -7 7 7 \N \N str_7 \N \N \N -8 8 8 \N \N str_8 \N \N \N -9 9 9 \N \N \N \N \N \N -10 10 10 \N \N \N \N \N \N -11 11 11 \N \N \N \N \N \N -12 12 12 \N \N 12 \N \N \N -13 13 13 \N \N str_13 \N \N \N -14 14 14 \N \N \N \N \N \N -15 15 15 \N \N \N \N \N \N -16 16 16 \N \N 16 \N \N \N -17 17 17 \N \N str_17 \N \N \N -18 18 18 \N \N 1970-01-19 \N \N \N -19 19 19 \N \N \N \N \N \N -20 20 20 \N \N \N 20 \N \N -21 21 21 \N \N str_21 \N \N \N -22 22 22 \N \N \N \N 1970-01-23 \N -23 \N \N \N \N \N \N \N \N -24 24 24 \N \N \N \N \N \N -25 str_25 \N str_25 \N \N \N \N \N -MergeTree wide -initial insert -alter add column 1 -3 None -0 0 \N \N \N \N -1 1 \N \N \N \N -2 2 \N \N \N \N -insert after alter add column 1 -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -alter modify column 1 -7 None -8 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -insert after alter modify column 1 -8 None -11 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -alter modify column 2 -8 None -11 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -insert after alter modify column 2 -1 Date -1 UInt64 -9 None -12 String -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 3 \N \N \N -4 4 4 4 \N \N \N -5 5 5 5 \N \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 12 \N \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -15 15 \N \N \N \N \N -16 16 16 16 \N \N \N -17 17 str_17 str_17 \N \N \N -18 18 1970-01-19 1970-01-19 \N \N \N -19 19 \N \N \N \N \N -20 20 20 \N 20 \N \N -21 21 str_21 str_21 \N \N \N -22 22 1970-01-23 \N \N 1970-01-23 \N -alter modify column 3 -1 Date -1 UInt64 -9 None -12 String -0 0 0 \N \N \N \N \N \N -1 1 1 \N \N \N \N \N \N -2 2 2 \N \N \N \N \N \N -3 3 3 \N \N 3 \N \N \N -4 4 4 \N \N 4 \N \N \N -5 5 5 \N \N 5 \N \N \N -6 6 6 \N \N str_6 \N \N \N -7 7 7 \N \N str_7 \N \N \N -8 8 8 \N \N str_8 \N \N \N -9 9 9 \N \N \N \N \N \N -10 10 10 \N \N \N \N \N \N -11 11 11 \N \N \N \N \N \N -12 12 12 \N \N 12 \N \N \N -13 13 13 \N \N str_13 \N \N \N -14 14 14 \N \N \N \N \N \N -15 15 15 \N \N \N \N \N \N -16 16 16 \N \N 16 \N \N \N -17 17 17 \N \N str_17 \N \N \N -18 18 18 \N \N 1970-01-19 \N \N \N -19 19 19 \N \N \N \N \N \N -20 20 20 \N \N \N 20 \N \N -21 21 21 \N \N str_21 \N \N \N -22 22 22 \N \N \N \N 1970-01-23 \N -insert after alter modify column 3 -1 Date -1 UInt64 -12 None -12 String -0 0 0 \N \N \N \N \N \N -1 1 1 \N \N \N \N \N \N -2 2 2 \N \N \N \N \N \N -3 3 3 \N \N 3 \N \N \N -4 4 4 \N \N 4 \N \N \N -5 5 5 \N \N 5 \N \N \N -6 6 6 \N \N str_6 \N \N \N -7 7 7 \N \N str_7 \N \N \N -8 8 8 \N \N str_8 \N \N \N -9 9 9 \N \N \N \N \N \N -10 10 10 \N \N \N \N \N \N -11 11 11 \N \N \N \N \N \N -12 12 12 \N \N 12 \N \N \N -13 13 13 \N \N str_13 \N \N \N -14 14 14 \N \N \N \N \N \N -15 15 15 \N \N \N \N \N \N -16 16 16 \N \N 16 \N \N \N -17 17 17 \N \N str_17 \N \N \N -18 18 18 \N \N 1970-01-19 \N \N \N -19 19 19 \N \N \N \N \N \N -20 20 20 \N \N \N 20 \N \N -21 21 21 \N \N str_21 \N \N \N -22 22 22 \N \N \N \N 1970-01-23 \N -23 \N \N \N \N \N \N \N \N -24 24 24 \N \N \N \N \N \N -25 str_25 \N str_25 \N \N \N \N \N diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_1.sh b/tests/queries/0_stateless/03040_dynamic_type_alters_1.sh deleted file mode 100755 index 7a73be20a4d..00000000000 --- a/tests/queries/0_stateless/03040_dynamic_type_alters_1.sh +++ /dev/null @@ -1,77 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_dynamic_type=1 --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_analyzer=1" - -function run() -{ - echo "initial insert" - $CH_CLIENT -q "insert into test select number, number from numbers(3)" - - echo "alter add column 1" - $CH_CLIENT -q "alter table test add column d Dynamic(max_types=3) settings mutations_sync=1" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, d, d.String, d.UInt64, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "insert after alter add column 1" - $CH_CLIENT -q "insert into test select number, number, number from numbers(3, 3)" - $CH_CLIENT -q "insert into test select number, number, 'str_' || toString(number) from numbers(6, 3)" - $CH_CLIENT -q "insert into test select number, number, NULL from numbers(9, 3)" - $CH_CLIENT -q "insert into test select number, number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL) from numbers(12, 3)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, d, d.String, d.UInt64, d.Date, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "alter modify column 1" - $CH_CLIENT -q "alter table test modify column d Dynamic(max_types=1) settings mutations_sync=1" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, d, d.String, d.UInt64, d.Date, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "insert after alter modify column 1" - $CH_CLIENT -q "insert into test select number, number, multiIf(number % 4 == 0, number, number % 4 == 1, 'str_' || toString(number), number % 4 == 2, toDate(number), NULL) from numbers(15, 4)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, d, d.String, d.UInt64, d.Date, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "alter modify column 2" - $CH_CLIENT -q "alter table test modify column d Dynamic(max_types=3) settings mutations_sync=1" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, d, d.String, d.UInt64, d.Date, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "insert after alter modify column 2" - $CH_CLIENT -q "insert into test select number, number, multiIf(number % 4 == 0, number, number % 4 == 1, 'str_' || toString(number), number % 4 == 2, toDate(number), NULL) from numbers(19, 4)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, d, d.String, d.UInt64, d.Date, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "alter modify column 3" - $CH_CLIENT -q "alter table test modify column y Dynamic settings mutations_sync=1" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, y.UInt64, y.String, y.\`Tuple(a UInt64)\`.a, d.String, d.UInt64, d.Date, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "insert after alter modify column 3" - $CH_CLIENT -q "insert into test select number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL), NULL from numbers(23, 3)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, y.UInt64, y.String, y.\`Tuple(a UInt64)\`.a, d.String, d.UInt64, d.Date, d.\`Tuple(a UInt64)\`.a from test order by x" -} - -$CH_CLIENT -q "drop table if exists test;" - -echo "Memory" -$CH_CLIENT -q "create table test (x UInt64, y UInt64) engine=Memory" -run -$CH_CLIENT -q "drop table test;" - -echo "MergeTree compact" -$CH_CLIENT -q "create table test (x UInt64, y UInt64) engine=MergeTree order by x settings min_rows_for_wide_part=100000000, min_bytes_for_wide_part=1000000000;" -run -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide" -$CH_CLIENT -q "create table test (x UInt64, y UInt64 ) engine=MergeTree order by x settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" -run -$CH_CLIENT -q "drop table test;" - diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_1_compact_merge_tree.reference b/tests/queries/0_stateless/03040_dynamic_type_alters_1_compact_merge_tree.reference new file mode 100644 index 00000000000..2ec301b747b --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_1_compact_merge_tree.reference @@ -0,0 +1,174 @@ +initial insert +alter add column 1 +3 None +0 0 \N \N \N 0 +1 1 \N \N \N 0 +2 2 \N \N \N 0 +insert after alter add column 1 +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +alter modify column 1 +7 None +8 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +insert after alter modify column 1 +8 None +11 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +alter modify column 2 +8 None +11 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +insert after alter modify column 2 +1 Date +1 UInt64 +9 None +12 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +19 19 \N \N \N \N 0 +20 20 20 \N 20 \N 0 +21 21 str_21 str_21 \N \N 0 +22 22 1970-01-23 \N \N 1970-01-23 0 +alter modify column 3 +1 Date +1 UInt64 +9 None +12 String +0 0 0 \N 0 \N \N \N 0 +1 1 1 \N 0 \N \N \N 0 +2 2 2 \N 0 \N \N \N 0 +3 3 3 \N 0 3 \N \N 0 +4 4 4 \N 0 4 \N \N 0 +5 5 5 \N 0 5 \N \N 0 +6 6 6 \N 0 str_6 \N \N 0 +7 7 7 \N 0 str_7 \N \N 0 +8 8 8 \N 0 str_8 \N \N 0 +9 9 9 \N 0 \N \N \N 0 +10 10 10 \N 0 \N \N \N 0 +11 11 11 \N 0 \N \N \N 0 +12 12 12 \N 0 12 \N \N 0 +13 13 13 \N 0 str_13 \N \N 0 +14 14 14 \N 0 \N \N \N 0 +15 15 15 \N 0 \N \N \N 0 +16 16 16 \N 0 16 \N \N 0 +17 17 17 \N 0 str_17 \N \N 0 +18 18 18 \N 0 1970-01-19 \N \N 0 +19 19 19 \N 0 \N \N \N 0 +20 20 20 \N 0 \N 20 \N 0 +21 21 21 \N 0 str_21 \N \N 0 +22 22 22 \N 0 \N \N 1970-01-23 0 +insert after alter modify column 3 +1 Date +1 UInt64 +12 None +12 String +0 0 0 \N 0 \N \N \N 0 +1 1 1 \N 0 \N \N \N 0 +2 2 2 \N 0 \N \N \N 0 +3 3 3 \N 0 3 \N \N 0 +4 4 4 \N 0 4 \N \N 0 +5 5 5 \N 0 5 \N \N 0 +6 6 6 \N 0 str_6 \N \N 0 +7 7 7 \N 0 str_7 \N \N 0 +8 8 8 \N 0 str_8 \N \N 0 +9 9 9 \N 0 \N \N \N 0 +10 10 10 \N 0 \N \N \N 0 +11 11 11 \N 0 \N \N \N 0 +12 12 12 \N 0 12 \N \N 0 +13 13 13 \N 0 str_13 \N \N 0 +14 14 14 \N 0 \N \N \N 0 +15 15 15 \N 0 \N \N \N 0 +16 16 16 \N 0 16 \N \N 0 +17 17 17 \N 0 str_17 \N \N 0 +18 18 18 \N 0 1970-01-19 \N \N 0 +19 19 19 \N 0 \N \N \N 0 +20 20 20 \N 0 \N 20 \N 0 +21 21 21 \N 0 str_21 \N \N 0 +22 22 22 \N 0 \N \N 1970-01-23 0 +23 \N \N \N 0 \N \N \N 0 +24 24 24 \N 0 \N \N \N 0 +25 str_25 \N str_25 0 \N \N \N 0 diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_1_compact_merge_tree.sql b/tests/queries/0_stateless/03040_dynamic_type_alters_1_compact_merge_tree.sql new file mode 100644 index 00000000000..4ab700306d4 --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_1_compact_merge_tree.sql @@ -0,0 +1,53 @@ +set allow_experimental_dynamic_type = 1; +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; + +drop table if exists test; +create table test (x UInt64, y UInt64) engine=MergeTree order by x settings min_rows_for_wide_part=100000000, min_bytes_for_wide_part=1000000000; +select 'initial insert'; +insert into test select number, number from numbers(3); + +select 'alter add column 1'; +alter table test add column d Dynamic(max_types=3) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter add column 1'; +insert into test select number, number, number from numbers(3, 3); +insert into test select number, number, 'str_' || toString(number) from numbers(6, 3); +insert into test select number, number, NULL from numbers(9, 3); +insert into test select number, number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL) from numbers(12, 3); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 1'; +alter table test modify column d Dynamic(max_types=1) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 1'; +insert into test select number, number, multiIf(number % 4 == 0, number, number % 4 == 1, 'str_' || toString(number), number % 4 == 2, toDate(number), NULL) from numbers(15, 4); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 2'; +alter table test modify column d Dynamic(max_types=3) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 2'; +insert into test select number, number, multiIf(number % 4 == 0, number, number % 4 == 1, 'str_' || toString(number), number % 4 == 2, toDate(number), NULL) from numbers(19, 4); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 3'; +alter table test modify column y Dynamic settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, y.UInt64, y.String, y.`Tuple(a UInt64)`.a, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 3'; +insert into test select number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL), NULL from numbers(23, 3); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, y.UInt64, y.String, y.`Tuple(a UInt64)`.a, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +drop table test; \ No newline at end of file diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_1_memory.reference b/tests/queries/0_stateless/03040_dynamic_type_alters_1_memory.reference new file mode 100644 index 00000000000..c592528c3cd --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_1_memory.reference @@ -0,0 +1,175 @@ +initial insert +alter add column 1 +3 None +0 0 \N \N \N 0 +1 1 \N \N \N 0 +2 2 \N \N \N 0 +insert after alter add column 1 +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +alter modify column 1 +7 None +8 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +insert after alter modify column 1 +8 None +11 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +alter modify column 2 +4 UInt64 +7 String +8 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +insert after alter modify column 2 +1 Date +5 UInt64 +8 String +9 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +19 19 \N \N \N \N 0 +20 20 20 \N 20 \N 0 +21 21 str_21 str_21 \N \N 0 +22 22 1970-01-23 \N \N 1970-01-23 0 +alter modify column 3 +1 Date +5 UInt64 +8 String +9 None +0 0 0 \N 0 \N \N \N 0 +1 1 1 \N 0 \N \N \N 0 +2 2 2 \N 0 \N \N \N 0 +3 3 3 \N 0 \N 3 \N 0 +4 4 4 \N 0 \N 4 \N 0 +5 5 5 \N 0 \N 5 \N 0 +6 6 6 \N 0 str_6 \N \N 0 +7 7 7 \N 0 str_7 \N \N 0 +8 8 8 \N 0 str_8 \N \N 0 +9 9 9 \N 0 \N \N \N 0 +10 10 10 \N 0 \N \N \N 0 +11 11 11 \N 0 \N \N \N 0 +12 12 12 \N 0 \N 12 \N 0 +13 13 13 \N 0 str_13 \N \N 0 +14 14 14 \N 0 \N \N \N 0 +15 15 15 \N 0 \N \N \N 0 +16 16 16 \N 0 16 \N \N 0 +17 17 17 \N 0 str_17 \N \N 0 +18 18 18 \N 0 1970-01-19 \N \N 0 +19 19 19 \N 0 \N \N \N 0 +20 20 20 \N 0 \N 20 \N 0 +21 21 21 \N 0 str_21 \N \N 0 +22 22 22 \N 0 \N \N 1970-01-23 0 +insert after alter modify column 3 +1 Date +5 UInt64 +8 String +12 None +0 0 0 \N 0 \N \N \N 0 +1 1 1 \N 0 \N \N \N 0 +2 2 2 \N 0 \N \N \N 0 +3 3 3 \N 0 \N 3 \N 0 +4 4 4 \N 0 \N 4 \N 0 +5 5 5 \N 0 \N 5 \N 0 +6 6 6 \N 0 str_6 \N \N 0 +7 7 7 \N 0 str_7 \N \N 0 +8 8 8 \N 0 str_8 \N \N 0 +9 9 9 \N 0 \N \N \N 0 +10 10 10 \N 0 \N \N \N 0 +11 11 11 \N 0 \N \N \N 0 +12 12 12 \N 0 \N 12 \N 0 +13 13 13 \N 0 str_13 \N \N 0 +14 14 14 \N 0 \N \N \N 0 +15 15 15 \N 0 \N \N \N 0 +16 16 16 \N 0 16 \N \N 0 +17 17 17 \N 0 str_17 \N \N 0 +18 18 18 \N 0 1970-01-19 \N \N 0 +19 19 19 \N 0 \N \N \N 0 +20 20 20 \N 0 \N 20 \N 0 +21 21 21 \N 0 str_21 \N \N 0 +22 22 22 \N 0 \N \N 1970-01-23 0 +23 \N \N \N 0 \N \N \N 0 +24 24 24 \N 0 \N \N \N 0 +25 str_25 \N str_25 0 \N \N \N 0 diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_1_memory.sql b/tests/queries/0_stateless/03040_dynamic_type_alters_1_memory.sql new file mode 100644 index 00000000000..e802fd034ce --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_1_memory.sql @@ -0,0 +1,53 @@ +set allow_experimental_dynamic_type = 1; +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; + +drop table if exists test; +create table test (x UInt64, y UInt64) engine=Memory; +select 'initial insert'; +insert into test select number, number from numbers(3); + +select 'alter add column 1'; +alter table test add column d Dynamic(max_types=3) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter add column 1'; +insert into test select number, number, number from numbers(3, 3); +insert into test select number, number, 'str_' || toString(number) from numbers(6, 3); +insert into test select number, number, NULL from numbers(9, 3); +insert into test select number, number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL) from numbers(12, 3); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 1'; +alter table test modify column d Dynamic(max_types=1) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 1'; +insert into test select number, number, multiIf(number % 4 == 0, number, number % 4 == 1, 'str_' || toString(number), number % 4 == 2, toDate(number), NULL) from numbers(15, 4); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 2'; +alter table test modify column d Dynamic(max_types=3) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 2'; +insert into test select number, number, multiIf(number % 4 == 0, number, number % 4 == 1, 'str_' || toString(number), number % 4 == 2, toDate(number), NULL) from numbers(19, 4); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 3'; +alter table test modify column y Dynamic settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, y.UInt64, y.String, y.`Tuple(a UInt64)`.a, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 3'; +insert into test select number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL), NULL from numbers(23, 3); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, y.UInt64, y.String, y.`Tuple(a UInt64)`.a, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +drop table test; \ No newline at end of file diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_1_wide_merge_tree.reference b/tests/queries/0_stateless/03040_dynamic_type_alters_1_wide_merge_tree.reference new file mode 100644 index 00000000000..2ec301b747b --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_1_wide_merge_tree.reference @@ -0,0 +1,174 @@ +initial insert +alter add column 1 +3 None +0 0 \N \N \N 0 +1 1 \N \N \N 0 +2 2 \N \N \N 0 +insert after alter add column 1 +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +alter modify column 1 +7 None +8 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +insert after alter modify column 1 +8 None +11 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +alter modify column 2 +8 None +11 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +insert after alter modify column 2 +1 Date +1 UInt64 +9 None +12 String +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 3 \N \N 0 +4 4 4 4 \N \N 0 +5 5 5 5 \N \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 12 \N \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +15 15 \N \N \N \N 0 +16 16 16 16 \N \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 1970-01-19 \N \N 0 +19 19 \N \N \N \N 0 +20 20 20 \N 20 \N 0 +21 21 str_21 str_21 \N \N 0 +22 22 1970-01-23 \N \N 1970-01-23 0 +alter modify column 3 +1 Date +1 UInt64 +9 None +12 String +0 0 0 \N 0 \N \N \N 0 +1 1 1 \N 0 \N \N \N 0 +2 2 2 \N 0 \N \N \N 0 +3 3 3 \N 0 3 \N \N 0 +4 4 4 \N 0 4 \N \N 0 +5 5 5 \N 0 5 \N \N 0 +6 6 6 \N 0 str_6 \N \N 0 +7 7 7 \N 0 str_7 \N \N 0 +8 8 8 \N 0 str_8 \N \N 0 +9 9 9 \N 0 \N \N \N 0 +10 10 10 \N 0 \N \N \N 0 +11 11 11 \N 0 \N \N \N 0 +12 12 12 \N 0 12 \N \N 0 +13 13 13 \N 0 str_13 \N \N 0 +14 14 14 \N 0 \N \N \N 0 +15 15 15 \N 0 \N \N \N 0 +16 16 16 \N 0 16 \N \N 0 +17 17 17 \N 0 str_17 \N \N 0 +18 18 18 \N 0 1970-01-19 \N \N 0 +19 19 19 \N 0 \N \N \N 0 +20 20 20 \N 0 \N 20 \N 0 +21 21 21 \N 0 str_21 \N \N 0 +22 22 22 \N 0 \N \N 1970-01-23 0 +insert after alter modify column 3 +1 Date +1 UInt64 +12 None +12 String +0 0 0 \N 0 \N \N \N 0 +1 1 1 \N 0 \N \N \N 0 +2 2 2 \N 0 \N \N \N 0 +3 3 3 \N 0 3 \N \N 0 +4 4 4 \N 0 4 \N \N 0 +5 5 5 \N 0 5 \N \N 0 +6 6 6 \N 0 str_6 \N \N 0 +7 7 7 \N 0 str_7 \N \N 0 +8 8 8 \N 0 str_8 \N \N 0 +9 9 9 \N 0 \N \N \N 0 +10 10 10 \N 0 \N \N \N 0 +11 11 11 \N 0 \N \N \N 0 +12 12 12 \N 0 12 \N \N 0 +13 13 13 \N 0 str_13 \N \N 0 +14 14 14 \N 0 \N \N \N 0 +15 15 15 \N 0 \N \N \N 0 +16 16 16 \N 0 16 \N \N 0 +17 17 17 \N 0 str_17 \N \N 0 +18 18 18 \N 0 1970-01-19 \N \N 0 +19 19 19 \N 0 \N \N \N 0 +20 20 20 \N 0 \N 20 \N 0 +21 21 21 \N 0 str_21 \N \N 0 +22 22 22 \N 0 \N \N 1970-01-23 0 +23 \N \N \N 0 \N \N \N 0 +24 24 24 \N 0 \N \N \N 0 +25 str_25 \N str_25 0 \N \N \N 0 diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_1_wide_merge_tree.sql b/tests/queries/0_stateless/03040_dynamic_type_alters_1_wide_merge_tree.sql new file mode 100644 index 00000000000..55c4f0b5f0c --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_1_wide_merge_tree.sql @@ -0,0 +1,53 @@ +set allow_experimental_dynamic_type = 1; +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; + +drop table if exists test; +create table test (x UInt64, y UInt64) engine=MergeTree order by x settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1; +select 'initial insert'; +insert into test select number, number from numbers(3); + +select 'alter add column 1'; +alter table test add column d Dynamic(max_types=3) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter add column 1'; +insert into test select number, number, number from numbers(3, 3); +insert into test select number, number, 'str_' || toString(number) from numbers(6, 3); +insert into test select number, number, NULL from numbers(9, 3); +insert into test select number, number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL) from numbers(12, 3); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 1'; +alter table test modify column d Dynamic(max_types=1) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 1'; +insert into test select number, number, multiIf(number % 4 == 0, number, number % 4 == 1, 'str_' || toString(number), number % 4 == 2, toDate(number), NULL) from numbers(15, 4); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 2'; +alter table test modify column d Dynamic(max_types=3) settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 2'; +insert into test select number, number, multiIf(number % 4 == 0, number, number % 4 == 1, 'str_' || toString(number), number % 4 == 2, toDate(number), NULL) from numbers(19, 4); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter modify column 3'; +alter table test modify column y Dynamic settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, y.UInt64, y.String, y.`Tuple(a UInt64)`.a, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter modify column 3'; +insert into test select number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL), NULL from numbers(23, 3); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, y.UInt64, y.String, y.`Tuple(a UInt64)`.a, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +drop table test; \ No newline at end of file diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_2.reference b/tests/queries/0_stateless/03040_dynamic_type_alters_2.reference deleted file mode 100644 index 18a181464e9..00000000000 --- a/tests/queries/0_stateless/03040_dynamic_type_alters_2.reference +++ /dev/null @@ -1,182 +0,0 @@ -MergeTree compact -initial insert -alter add column -3 None -0 0 \N \N \N \N -1 1 \N \N \N \N -2 2 \N \N \N \N -insert after alter add column 1 -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -alter rename column 1 -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -insert nested dynamic -3 Array(Dynamic) -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N [] [] [] -1 1 \N \N \N \N \N [] [] [] -2 2 \N \N \N \N \N [] [] [] -3 3 3 \N 3 \N \N [] [] [] -4 4 4 \N 4 \N \N [] [] [] -5 5 5 \N 5 \N \N [] [] [] -6 6 str_6 str_6 \N \N \N [] [] [] -7 7 str_7 str_7 \N \N \N [] [] [] -8 8 str_8 str_8 \N \N \N [] [] [] -9 9 \N \N \N \N \N [] [] [] -10 10 \N \N \N \N \N [] [] [] -11 11 \N \N \N \N \N [] [] [] -12 12 12 \N 12 \N \N [] [] [] -13 13 str_13 str_13 \N \N \N [] [] [] -14 14 \N \N \N \N \N [] [] [] -15 15 [15] \N \N \N \N [15] [NULL] [NULL] -16 16 ['str_16'] \N \N \N \N [NULL] ['str_16'] [NULL] -17 17 [17] \N \N \N \N [17] [NULL] [NULL] -alter rename column 2 -3 Array(Dynamic) -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N [] [] [] -1 1 \N \N \N \N \N [] [] [] -2 2 \N \N \N \N \N [] [] [] -3 3 3 \N 3 \N \N [] [] [] -4 4 4 \N 4 \N \N [] [] [] -5 5 5 \N 5 \N \N [] [] [] -6 6 str_6 str_6 \N \N \N [] [] [] -7 7 str_7 str_7 \N \N \N [] [] [] -8 8 str_8 str_8 \N \N \N [] [] [] -9 9 \N \N \N \N \N [] [] [] -10 10 \N \N \N \N \N [] [] [] -11 11 \N \N \N \N \N [] [] [] -12 12 12 \N 12 \N \N [] [] [] -13 13 str_13 str_13 \N \N \N [] [] [] -14 14 \N \N \N \N \N [] [] [] -15 15 [15] \N \N \N \N [15] [NULL] [NULL] -16 16 ['str_16'] \N \N \N \N [NULL] ['str_16'] [NULL] -17 17 [17] \N \N \N \N [17] [NULL] [NULL] -MergeTree wide -initial insert -alter add column -3 None -0 0 \N \N \N \N -1 1 \N \N \N \N -2 2 \N \N \N \N -insert after alter add column 1 -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -alter rename column 1 -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N -insert nested dynamic -3 Array(Dynamic) -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N [] [] [] -1 1 \N \N \N \N \N [] [] [] -2 2 \N \N \N \N \N [] [] [] -3 3 3 \N 3 \N \N [] [] [] -4 4 4 \N 4 \N \N [] [] [] -5 5 5 \N 5 \N \N [] [] [] -6 6 str_6 str_6 \N \N \N [] [] [] -7 7 str_7 str_7 \N \N \N [] [] [] -8 8 str_8 str_8 \N \N \N [] [] [] -9 9 \N \N \N \N \N [] [] [] -10 10 \N \N \N \N \N [] [] [] -11 11 \N \N \N \N \N [] [] [] -12 12 12 \N 12 \N \N [] [] [] -13 13 str_13 str_13 \N \N \N [] [] [] -14 14 \N \N \N \N \N [] [] [] -15 15 [15] \N \N \N \N [15] [NULL] [NULL] -16 16 ['str_16'] \N \N \N \N [NULL] ['str_16'] [NULL] -17 17 [17] \N \N \N \N [17] [NULL] [NULL] -alter rename column 2 -3 Array(Dynamic) -4 String -4 UInt64 -7 None -0 0 \N \N \N \N \N [] [] [] -1 1 \N \N \N \N \N [] [] [] -2 2 \N \N \N \N \N [] [] [] -3 3 3 \N 3 \N \N [] [] [] -4 4 4 \N 4 \N \N [] [] [] -5 5 5 \N 5 \N \N [] [] [] -6 6 str_6 str_6 \N \N \N [] [] [] -7 7 str_7 str_7 \N \N \N [] [] [] -8 8 str_8 str_8 \N \N \N [] [] [] -9 9 \N \N \N \N \N [] [] [] -10 10 \N \N \N \N \N [] [] [] -11 11 \N \N \N \N \N [] [] [] -12 12 12 \N 12 \N \N [] [] [] -13 13 str_13 str_13 \N \N \N [] [] [] -14 14 \N \N \N \N \N [] [] [] -15 15 [15] \N \N \N \N [15] [NULL] [NULL] -16 16 ['str_16'] \N \N \N \N [NULL] ['str_16'] [NULL] -17 17 [17] \N \N \N \N [17] [NULL] [NULL] diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_2.sh b/tests/queries/0_stateless/03040_dynamic_type_alters_2.sh deleted file mode 100755 index 6491e64372f..00000000000 --- a/tests/queries/0_stateless/03040_dynamic_type_alters_2.sh +++ /dev/null @@ -1,57 +0,0 @@ -#!/usr/bin/env bash -# Tags: long - -CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) -# reset --log_comment -CLICKHOUSE_LOG_COMMENT= -# shellcheck source=../shell_config.sh -. "$CUR_DIR"/../shell_config.sh - -CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_dynamic_type=1 --allow_experimental_variant_type=1 --use_variant_as_common_type=1" - -function run() -{ - echo "initial insert" - $CH_CLIENT -q "insert into test select number, number from numbers(3)" - - echo "alter add column" - $CH_CLIENT -q "alter table test add column d Dynamic settings mutations_sync=1" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, d, d.String, d.UInt64, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "insert after alter add column 1" - $CH_CLIENT -q "insert into test select number, number, number from numbers(3, 3)" - $CH_CLIENT -q "insert into test select number, number, 'str_' || toString(number) from numbers(6, 3)" - $CH_CLIENT -q "insert into test select number, number, NULL from numbers(9, 3)" - $CH_CLIENT -q "insert into test select number, number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL) from numbers(12, 3)" - $CH_CLIENT -q "select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d)" - $CH_CLIENT -q "select x, y, d, d.String, d.UInt64, d.Date, d.\`Tuple(a UInt64)\`.a from test order by x" - - echo "alter rename column 1" - $CH_CLIENT -q "alter table test rename column d to d1 settings mutations_sync=1" - $CH_CLIENT -q "select count(), dynamicType(d1) from test group by dynamicType(d1) order by count(), dynamicType(d1)" - $CH_CLIENT -q "select x, y, d1, d1.String, d1.UInt64, d1.Date, d1.\`Tuple(a UInt64)\`.a from test order by x" - - echo "insert nested dynamic" - $CH_CLIENT -q "insert into test select number, number, [number % 2 ? number : 'str_' || toString(number)]::Array(Dynamic) from numbers(15, 3)" - $CH_CLIENT -q "select count(), dynamicType(d1) from test group by dynamicType(d1) order by count(), dynamicType(d1)" - $CH_CLIENT -q "select x, y, d1, d1.String, d1.UInt64, d1.Date, d1.\`Tuple(a UInt64)\`.a, d1.\`Array(Dynamic)\`.UInt64, d1.\`Array(Dynamic)\`.String, d1.\`Array(Dynamic)\`.Date from test order by x" - - echo "alter rename column 2" - $CH_CLIENT -q "alter table test rename column d1 to d2 settings mutations_sync=1" - $CH_CLIENT -q "select count(), dynamicType(d2) from test group by dynamicType(d2) order by count(), dynamicType(d2)" - $CH_CLIENT -q "select x, y, d2, d2.String, d2.UInt64, d2.Date, d2.\`Tuple(a UInt64)\`.a, d2.\`Array(Dynamic)\`.UInt64, d2.\`Array(Dynamic)\`.String, d2.\`Array(Dynamic)\`.Date, from test order by x" -} - -$CH_CLIENT -q "drop table if exists test;" - -echo "MergeTree compact" -$CH_CLIENT -q "create table test (x UInt64, y UInt64) engine=MergeTree order by x settings min_rows_for_wide_part=100000000, min_bytes_for_wide_part=1000000000;" -run -$CH_CLIENT -q "drop table test;" - -echo "MergeTree wide" -$CH_CLIENT -q "create table test (x UInt64, y UInt64 ) engine=MergeTree order by x settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" -run -$CH_CLIENT -q "drop table test;" - diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_2_compact_merge_tree.reference b/tests/queries/0_stateless/03040_dynamic_type_alters_2_compact_merge_tree.reference new file mode 100644 index 00000000000..a2f2a19805d --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_2_compact_merge_tree.reference @@ -0,0 +1,90 @@ +initial insert +alter add column +3 None +0 0 \N \N \N 0 +1 1 \N \N \N 0 +2 2 \N \N \N 0 +insert after alter add column 1 +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +alter rename column 1 +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +insert nested dynamic +3 Array(Dynamic) +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 [] [] [] +1 1 \N \N \N \N 0 [] [] [] +2 2 \N \N \N \N 0 [] [] [] +3 3 3 \N 3 \N 0 [] [] [] +4 4 4 \N 4 \N 0 [] [] [] +5 5 5 \N 5 \N 0 [] [] [] +6 6 str_6 str_6 \N \N 0 [] [] [] +7 7 str_7 str_7 \N \N 0 [] [] [] +8 8 str_8 str_8 \N \N 0 [] [] [] +9 9 \N \N \N \N 0 [] [] [] +10 10 \N \N \N \N 0 [] [] [] +11 11 \N \N \N \N 0 [] [] [] +12 12 12 \N 12 \N 0 [] [] [] +13 13 str_13 str_13 \N \N 0 [] [] [] +14 14 \N \N \N \N 0 [] [] [] +15 15 [15] \N \N \N 0 [15] [NULL] [NULL] +16 16 ['str_16'] \N \N \N 0 [NULL] ['str_16'] [NULL] +17 17 [17] \N \N \N 0 [17] [NULL] [NULL] +alter rename column 2 +3 Array(Dynamic) +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 [] [] [] +1 1 \N \N \N \N 0 [] [] [] +2 2 \N \N \N \N 0 [] [] [] +3 3 3 \N 3 \N 0 [] [] [] +4 4 4 \N 4 \N 0 [] [] [] +5 5 5 \N 5 \N 0 [] [] [] +6 6 str_6 str_6 \N \N 0 [] [] [] +7 7 str_7 str_7 \N \N 0 [] [] [] +8 8 str_8 str_8 \N \N 0 [] [] [] +9 9 \N \N \N \N 0 [] [] [] +10 10 \N \N \N \N 0 [] [] [] +11 11 \N \N \N \N 0 [] [] [] +12 12 12 \N 12 \N 0 [] [] [] +13 13 str_13 str_13 \N \N 0 [] [] [] +14 14 \N \N \N \N 0 [] [] [] +15 15 [15] \N \N \N 0 [15] [NULL] [NULL] +16 16 ['str_16'] \N \N \N 0 [NULL] ['str_16'] [NULL] +17 17 [17] \N \N \N 0 [17] [NULL] [NULL] diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_2_compact_merge_tree.sql b/tests/queries/0_stateless/03040_dynamic_type_alters_2_compact_merge_tree.sql new file mode 100644 index 00000000000..cead110dd7d --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_2_compact_merge_tree.sql @@ -0,0 +1,39 @@ +set allow_experimental_dynamic_type = 1; +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; + +drop table if exists test; +create table test (x UInt64, y UInt64) engine=MergeTree order by x settings min_rows_for_wide_part=100000000, min_bytes_for_wide_part=1000000000; + +select 'initial insert'; +insert into test select number, number from numbers(3); + +select 'alter add column'; +alter table test add column d Dynamic settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter add column 1'; +insert into test select number, number, number from numbers(3, 3); +insert into test select number, number, 'str_' || toString(number) from numbers(6, 3); +insert into test select number, number, NULL from numbers(9, 3); +insert into test select number, number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL) from numbers(12, 3); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter rename column 1'; +alter table test rename column d to d1 settings mutations_sync=1; +select count(), dynamicType(d1) from test group by dynamicType(d1) order by count(), dynamicType(d1); +select x, y, d1, d1.String, d1.UInt64, d1.Date, d1.`Tuple(a UInt64)`.a from test order by x; + +select 'insert nested dynamic'; +insert into test select number, number, [number % 2 ? number : 'str_' || toString(number)]::Array(Dynamic) from numbers(15, 3); +select count(), dynamicType(d1) from test group by dynamicType(d1) order by count(), dynamicType(d1); +select x, y, d1, d1.String, d1.UInt64, d1.Date, d1.`Tuple(a UInt64)`.a, d1.`Array(Dynamic)`.UInt64, d1.`Array(Dynamic)`.String, d1.`Array(Dynamic)`.Date from test order by x; + +select 'alter rename column 2'; +alter table test rename column d1 to d2 settings mutations_sync=1; +select count(), dynamicType(d2) from test group by dynamicType(d2) order by count(), dynamicType(d2); +select x, y, d2, d2.String, d2.UInt64, d2.Date, d2.`Tuple(a UInt64)`.a, d2.`Array(Dynamic)`.UInt64, d2.`Array(Dynamic)`.String, d2.`Array(Dynamic)`.Date, from test order by x; + +drop table test; diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_2_wide_merge_tree.reference b/tests/queries/0_stateless/03040_dynamic_type_alters_2_wide_merge_tree.reference new file mode 100644 index 00000000000..a2f2a19805d --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_2_wide_merge_tree.reference @@ -0,0 +1,90 @@ +initial insert +alter add column +3 None +0 0 \N \N \N 0 +1 1 \N \N \N 0 +2 2 \N \N \N 0 +insert after alter add column 1 +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +alter rename column 1 +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 +insert nested dynamic +3 Array(Dynamic) +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 [] [] [] +1 1 \N \N \N \N 0 [] [] [] +2 2 \N \N \N \N 0 [] [] [] +3 3 3 \N 3 \N 0 [] [] [] +4 4 4 \N 4 \N 0 [] [] [] +5 5 5 \N 5 \N 0 [] [] [] +6 6 str_6 str_6 \N \N 0 [] [] [] +7 7 str_7 str_7 \N \N 0 [] [] [] +8 8 str_8 str_8 \N \N 0 [] [] [] +9 9 \N \N \N \N 0 [] [] [] +10 10 \N \N \N \N 0 [] [] [] +11 11 \N \N \N \N 0 [] [] [] +12 12 12 \N 12 \N 0 [] [] [] +13 13 str_13 str_13 \N \N 0 [] [] [] +14 14 \N \N \N \N 0 [] [] [] +15 15 [15] \N \N \N 0 [15] [NULL] [NULL] +16 16 ['str_16'] \N \N \N 0 [NULL] ['str_16'] [NULL] +17 17 [17] \N \N \N 0 [17] [NULL] [NULL] +alter rename column 2 +3 Array(Dynamic) +4 String +4 UInt64 +7 None +0 0 \N \N \N \N 0 [] [] [] +1 1 \N \N \N \N 0 [] [] [] +2 2 \N \N \N \N 0 [] [] [] +3 3 3 \N 3 \N 0 [] [] [] +4 4 4 \N 4 \N 0 [] [] [] +5 5 5 \N 5 \N 0 [] [] [] +6 6 str_6 str_6 \N \N 0 [] [] [] +7 7 str_7 str_7 \N \N 0 [] [] [] +8 8 str_8 str_8 \N \N 0 [] [] [] +9 9 \N \N \N \N 0 [] [] [] +10 10 \N \N \N \N 0 [] [] [] +11 11 \N \N \N \N 0 [] [] [] +12 12 12 \N 12 \N 0 [] [] [] +13 13 str_13 str_13 \N \N 0 [] [] [] +14 14 \N \N \N \N 0 [] [] [] +15 15 [15] \N \N \N 0 [15] [NULL] [NULL] +16 16 ['str_16'] \N \N \N 0 [NULL] ['str_16'] [NULL] +17 17 [17] \N \N \N 0 [17] [NULL] [NULL] diff --git a/tests/queries/0_stateless/03040_dynamic_type_alters_2_wide_merge_tree.sql b/tests/queries/0_stateless/03040_dynamic_type_alters_2_wide_merge_tree.sql new file mode 100644 index 00000000000..f58599b1d61 --- /dev/null +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_2_wide_merge_tree.sql @@ -0,0 +1,39 @@ +set allow_experimental_dynamic_type = 1; +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; + +drop table if exists test; +create table test (x UInt64, y UInt64) engine=MergeTree order by x settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1; + +select 'initial insert'; +insert into test select number, number from numbers(3); + +select 'alter add column'; +alter table test add column d Dynamic settings mutations_sync=1; +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.`Tuple(a UInt64)`.a from test order by x; + +select 'insert after alter add column 1'; +insert into test select number, number, number from numbers(3, 3); +insert into test select number, number, 'str_' || toString(number) from numbers(6, 3); +insert into test select number, number, NULL from numbers(9, 3); +insert into test select number, number, multiIf(number % 3 == 0, number, number % 3 == 1, 'str_' || toString(number), NULL) from numbers(12, 3); +select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select x, y, d, d.String, d.UInt64, d.Date, d.`Tuple(a UInt64)`.a from test order by x; + +select 'alter rename column 1'; +alter table test rename column d to d1 settings mutations_sync=1; +select count(), dynamicType(d1) from test group by dynamicType(d1) order by count(), dynamicType(d1); +select x, y, d1, d1.String, d1.UInt64, d1.Date, d1.`Tuple(a UInt64)`.a from test order by x; + +select 'insert nested dynamic'; +insert into test select number, number, [number % 2 ? number : 'str_' || toString(number)]::Array(Dynamic) from numbers(15, 3); +select count(), dynamicType(d1) from test group by dynamicType(d1) order by count(), dynamicType(d1); +select x, y, d1, d1.String, d1.UInt64, d1.Date, d1.`Tuple(a UInt64)`.a, d1.`Array(Dynamic)`.UInt64, d1.`Array(Dynamic)`.String, d1.`Array(Dynamic)`.Date from test order by x; + +select 'alter rename column 2'; +alter table test rename column d1 to d2 settings mutations_sync=1; +select count(), dynamicType(d2) from test group by dynamicType(d2) order by count(), dynamicType(d2); +select x, y, d2, d2.String, d2.UInt64, d2.Date, d2.`Tuple(a UInt64)`.a, d2.`Array(Dynamic)`.UInt64, d2.`Array(Dynamic)`.String, d2.`Array(Dynamic)`.Date, from test order by x; + +drop table test; diff --git a/tests/queries/0_stateless/03041_dynamic_type_check_table.reference b/tests/queries/0_stateless/03041_dynamic_type_check_table.reference index b1ea186a917..0dab4ea0d20 100644 --- a/tests/queries/0_stateless/03041_dynamic_type_check_table.reference +++ b/tests/queries/0_stateless/03041_dynamic_type_check_table.reference @@ -2,55 +2,55 @@ MergeTree compact initial insert alter add column 3 None -0 0 \N \N \N \N -1 1 \N \N \N \N -2 2 \N \N \N \N +0 0 \N \N \N 0 +1 1 \N \N \N 0 +2 2 \N \N \N 0 insert after alter add column 4 String 4 UInt64 7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 check table 1 MergeTree wide initial insert alter add column 3 None -0 0 \N \N \N \N -1 1 \N \N \N \N -2 2 \N \N \N \N +0 0 \N \N \N 0 +1 1 \N \N \N 0 +2 2 \N \N \N 0 insert after alter add column 4 String 4 UInt64 7 None -0 0 \N \N \N \N \N -1 1 \N \N \N \N \N -2 2 \N \N \N \N \N -3 3 3 \N 3 \N \N -4 4 4 \N 4 \N \N -5 5 5 \N 5 \N \N -6 6 str_6 str_6 \N \N \N -7 7 str_7 str_7 \N \N \N -8 8 str_8 str_8 \N \N \N -9 9 \N \N \N \N \N -10 10 \N \N \N \N \N -11 11 \N \N \N \N \N -12 12 12 \N 12 \N \N -13 13 str_13 str_13 \N \N \N -14 14 \N \N \N \N \N +0 0 \N \N \N \N 0 +1 1 \N \N \N \N 0 +2 2 \N \N \N \N 0 +3 3 3 \N 3 \N 0 +4 4 4 \N 4 \N 0 +5 5 5 \N 5 \N 0 +6 6 str_6 str_6 \N \N 0 +7 7 str_7 str_7 \N \N 0 +8 8 str_8 str_8 \N \N 0 +9 9 \N \N \N \N 0 +10 10 \N \N \N \N 0 +11 11 \N \N \N \N 0 +12 12 12 \N 12 \N 0 +13 13 str_13 str_13 \N \N 0 +14 14 \N \N \N \N 0 check table 1 diff --git a/tests/queries/0_stateless/03056_analyzer_double_subquery_alias.reference b/tests/queries/0_stateless/03056_analyzer_double_subquery_alias.reference deleted file mode 100644 index 72749c905a3..00000000000 --- a/tests/queries/0_stateless/03056_analyzer_double_subquery_alias.reference +++ /dev/null @@ -1 +0,0 @@ -1 1 1 diff --git a/tests/queries/0_stateless/03056_analyzer_double_subquery_alias.sql b/tests/queries/0_stateless/03056_analyzer_double_subquery_alias.sql deleted file mode 100644 index de471c1a091..00000000000 --- a/tests/queries/0_stateless/03056_analyzer_double_subquery_alias.sql +++ /dev/null @@ -1,17 +0,0 @@ --- https://github.com/ClickHouse/ClickHouse/issues/22627 -SET allow_experimental_analyzer=1; -WITH - x AS - ( - SELECT 1 AS a - ), - xx AS - ( - SELECT * - FROM x - , x AS x1 - , x AS x2 - ) -SELECT * -FROM xx -WHERE a = 1; diff --git a/tests/queries/0_stateless/03071_fix_short_circuit_logic.reference b/tests/queries/0_stateless/03071_fix_short_circuit_logic.reference new file mode 100644 index 00000000000..48aedfc3958 --- /dev/null +++ b/tests/queries/0_stateless/03071_fix_short_circuit_logic.reference @@ -0,0 +1 @@ +2024-01-02 16:54:59 diff --git a/tests/queries/0_stateless/03071_fix_short_circuit_logic.sql b/tests/queries/0_stateless/03071_fix_short_circuit_logic.sql new file mode 100644 index 00000000000..7745bceca0b --- /dev/null +++ b/tests/queries/0_stateless/03071_fix_short_circuit_logic.sql @@ -0,0 +1,62 @@ + + +CREATE FUNCTION IF NOT EXISTS unhexPrefixed AS value -> unhex(substring(value, 3)); +CREATE FUNCTION IF NOT EXISTS hex2bytes AS address -> CAST(unhexPrefixed(address), 'FixedString(20)'); +CREATE FUNCTION IF NOT EXISTS bytes2hex AS address -> concat('0x', lower(hex(address))); + +CREATE TABLE test +( + `transfer_id` String, + `address` FixedString(20), + `value` UInt256, + `block_timestamp` DateTime('UTC'), + `token_address` FixedString(20) +) +ENGINE = MergeTree +PARTITION BY toYYYYMM(block_timestamp) +PRIMARY KEY (address, block_timestamp) +ORDER BY (address, block_timestamp); + +INSERT INTO test SELECT 'token-transfer-0x758f1bbabb160683e1c80ed52dcd24a32b599d40edf1cec91b5f1199c0e392a2-56', hex2bytes('0xd387a6e4e84a6c86bd90c158c6028a58cc8ac459'), 3000000000000000000000, '2024-01-02 16:54:59', 'abc'; + +CREATE TABLE token_data +( + token_address_hex String, + chain String, + is_blacklisted Bool +) +ENGINE = TinyLog; + +INSERT INTO token_data SELECT bytes2hex('abc'), 'zksync', false; + +CREATE DICTIONARY token_data_map +( + token_address_hex String, + chain String, + is_blacklisted Bool +) +PRIMARY KEY token_address_hex, chain +SOURCE(Clickhouse(table token_data)) +LIFETIME(MIN 200 MAX 300) +LAYOUT(COMPLEX_KEY_HASHED_ARRAY()); + +SELECT block_timestamp +FROM +( + SELECT + block_timestamp, + bytes2hex(token_address) AS token_address_hex + FROM + ( + SELECT + transfer_id, + address, + value, + block_timestamp, + token_address, + 'zksync' AS chain + FROM test + ) + WHERE (address = hex2bytes('0xd387a6e4e84a6c86bd90c158c6028a58cc8ac459')) AND (transfer_id NOT LIKE 'gas%') AND (value > 0) AND (dictGetOrDefault(token_data_map, 'is_blacklisted', (token_address_hex, 'zksync'), true)) +) +SETTINGS max_threads = 1, short_circuit_function_evaluation = 'enable', allow_experimental_analyzer = 0; \ No newline at end of file diff --git a/tests/queries/0_stateless/03094_one_thousand_joins.sql b/tests/queries/0_stateless/03094_one_thousand_joins.sql index ea159f0e4c0..1f6bd99df7f 100644 --- a/tests/queries/0_stateless/03094_one_thousand_joins.sql +++ b/tests/queries/0_stateless/03094_one_thousand_joins.sql @@ -1,6 +1,7 @@ -- Tags: no-fasttest, no-tsan, long -- (no-tsan because it has a small maximum stack size and the test would fail with TOO_DEEP_RECURSION) +SET join_algorithm = 'default'; -- for 'full_sorting_merge' the query is 10x slower SET allow_experimental_analyzer = 1; -- old analyzer returns TOO_DEEP_SUBQUERIES -- Bug 33446, marked as 'long' because it still runs around 10 sec diff --git a/tests/queries/0_stateless/03127_system_unload_primary_key_table.reference b/tests/queries/0_stateless/03127_system_unload_primary_key_table.reference index 3ac6127fb21..2d33f7f6683 100644 --- a/tests/queries/0_stateless/03127_system_unload_primary_key_table.reference +++ b/tests/queries/0_stateless/03127_system_unload_primary_key_table.reference @@ -1,8 +1,8 @@ -100000000 140000000 -100000000 140000000 -100000000 140000000 +100000000 100000000 +100000000 100000000 +100000000 100000000 0 0 -100000000 140000000 +100000000 100000000 0 0 0 0 1 diff --git a/tests/queries/0_stateless/03128_system_unload_primary_key.reference b/tests/queries/0_stateless/03128_system_unload_primary_key.reference index c7b40ae5b06..2646dc7247f 100644 --- a/tests/queries/0_stateless/03128_system_unload_primary_key.reference +++ b/tests/queries/0_stateless/03128_system_unload_primary_key.reference @@ -1,4 +1,4 @@ -100000000 140000000 -100000000 140000000 +100000000 100000000 +100000000 100000000 0 0 0 0 diff --git a/tests/queries/0_stateless/03130_generateSnowflakeId.reference b/tests/queries/0_stateless/03130_generateSnowflakeId.reference index 39669d21bee..fd264f00d36 100644 --- a/tests/queries/0_stateless/03130_generateSnowflakeId.reference +++ b/tests/queries/0_stateless/03130_generateSnowflakeId.reference @@ -1,5 +1,11 @@ +Negative tests +The first bit must be zero 1 +Test disabling of common subexpression elimination via first parameter 0 0 1 +Test user-provided machine ID +1 +Generated Snowflake IDs are unique 100 diff --git a/tests/queries/0_stateless/03130_generateSnowflakeId.sql b/tests/queries/0_stateless/03130_generateSnowflakeId.sql index 0717c81aa0d..6c0a6cc0f2c 100644 --- a/tests/queries/0_stateless/03130_generateSnowflakeId.sql +++ b/tests/queries/0_stateless/03130_generateSnowflakeId.sql @@ -1,13 +1,22 @@ -- Test SQL function 'generateSnowflakeID' -SELECT bitAnd(bitShiftRight(toUInt64(generateSnowflakeID()), 63), 1) = 0; -- check first bit is zero +SELECT 'Negative tests'; +SELECT generateSnowflakeID(1, 2, 3); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT generateSnowflakeID(1, 'not_an_int'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT generateSnowflakeID(1, materialize(2)); -- { serverError ILLEGAL_COLUMN } + SELECT 'The first bit must be zero'; +SELECT bitAnd(bitShiftRight(generateSnowflakeID(), 63), 1) = 0; + +SELECT 'Test disabling of common subexpression elimination via first parameter'; SELECT generateSnowflakeID(1) = generateSnowflakeID(2); -- disabled common subexpression elimination --> lhs != rhs SELECT generateSnowflakeID() = generateSnowflakeID(1); -- same as ^^ -SELECT generateSnowflakeID(1) = generateSnowflakeID(1); -- enabled common subexpression elimination +SELECT generateSnowflakeID(1) = generateSnowflakeID(1); -- with common subexpression elimination -SELECT generateSnowflakeID(1, 2); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT 'Test user-provided machine ID'; +SELECT bitAnd(bitShiftRight(generateSnowflakeID(1, 123), 12), 1024 - 1) = 123; -- the machine id is actually set in the generated snowflake ID (1024 = 2^10) +SELECT 'Generated Snowflake IDs are unique'; SELECT count(*) FROM ( diff --git a/tests/queries/0_stateless/03131_hilbert_coding.sql b/tests/queries/0_stateless/03131_hilbert_coding.sql index ed293dc6910..b16a0efad5d 100644 --- a/tests/queries/0_stateless/03131_hilbert_coding.sql +++ b/tests/queries/0_stateless/03131_hilbert_coding.sql @@ -46,6 +46,7 @@ drop table if exists hilbert_numbers_1_03131; select '----- ERRORS -----'; select hilbertEncode(); -- { serverError TOO_FEW_ARGUMENTS_FOR_FUNCTION } +select hilbertEncode(1, 2, 3); -- { serverError TOO_MANY_ARGUMENTS_FOR_FUNCTION } select hilbertDecode(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } select hilbertEncode('text'); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } select hilbertDecode('text', 'text'); -- { serverError ILLEGAL_COLUMN } diff --git a/tests/queries/0_stateless/03143_asof_join_ddb_long.reference b/tests/queries/0_stateless/03143_asof_join_ddb_long.reference new file mode 100644 index 00000000000..2850a8aba98 --- /dev/null +++ b/tests/queries/0_stateless/03143_asof_join_ddb_long.reference @@ -0,0 +1,2 @@ +49999983751397 10000032 +49999983751397 10000032 diff --git a/tests/queries/0_stateless/03143_asof_join_ddb_long.sql b/tests/queries/0_stateless/03143_asof_join_ddb_long.sql new file mode 100644 index 00000000000..17a67511030 --- /dev/null +++ b/tests/queries/0_stateless/03143_asof_join_ddb_long.sql @@ -0,0 +1,50 @@ +-- Tags: long + +DROP TABLE IF EXISTS build; +DROP TABLE IF EXISTS skewed_probe; + +SET session_timezone = 'UTC'; + +CREATE TABLE build ENGINE = MergeTree ORDER BY (key, begin) +AS + SELECT + toDateTime('1990-03-21 13:00:00') + INTERVAL number MINUTE AS begin, + number % 4 AS key, + number AS value + FROM numbers(0, 10000000); + +CREATE TABLE skewed_probe ENGINE = MergeTree ORDER BY (key, begin) +AS + SELECT + toDateTime('1990-04-21 13:00:01') + INTERVAL number MINUTE AS begin, + 0 AS key + FROM numbers(0, 5) + UNION ALL + SELECT + toDateTime('1990-05-21 13:00:01') + INTERVAL number MINUTE AS begin, + 1 AS key + FROM numbers(0, 10) + UNION ALL + SELECT + toDateTime('1990-06-21 13:00:01') + INTERVAL number MINUTE AS begin, + 2 AS key + FROM numbers(0, 20) + UNION ALL + SELECT + toDateTime('1990-03-21 13:00:01') + INTERVAL number MINUTE AS begin, + 3 AS key + FROM numbers(0, 10000000); + + +SELECT SUM(value), COUNT(*) +FROM skewed_probe +ASOF JOIN build +USING (key, begin) +; + +SELECT SUM(value), COUNT(*) +FROM skewed_probe +ASOF JOIN build +USING (key, begin) +SETTINGS join_algorithm = 'full_sorting_merge' +; diff --git a/tests/queries/0_stateless/03144_asof_join_ddb_doubles.reference b/tests/queries/0_stateless/03144_asof_join_ddb_doubles.reference new file mode 100644 index 00000000000..f130f0a3f3b --- /dev/null +++ b/tests/queries/0_stateless/03144_asof_join_ddb_doubles.reference @@ -0,0 +1,58 @@ +1 0 +2 0 +3 1 +4 1 +5 1 +6 2 +7 2 +8 3 +9 3 +0 0 +1 0 +2 0 +3 1 +4 1 +5 1 +6 2 +7 2 +8 3 +9 3 +1 1 0 +1 2 0 +1 3 1 +1 4 1 +1 5 1 +1 6 2 +1 7 2 +1 8 3 +1 9 3 +2 0 10 +2 1 10 +2 2 10 +2 3 10 +2 4 10 +2 5 10 +2 6 10 +2 7 20 +2 8 20 +2 9 20 +1 0 0 +1 1 0 +1 2 0 +1 3 1 +1 4 1 +1 5 1 +1 6 2 +1 7 2 +1 8 3 +1 9 3 +2 0 10 +2 1 10 +2 2 10 +2 3 10 +2 4 10 +2 5 10 +2 6 10 +2 7 20 +2 8 20 +2 9 20 diff --git a/tests/queries/0_stateless/03144_asof_join_ddb_doubles.sql b/tests/queries/0_stateless/03144_asof_join_ddb_doubles.sql new file mode 100644 index 00000000000..87aece14628 --- /dev/null +++ b/tests/queries/0_stateless/03144_asof_join_ddb_doubles.sql @@ -0,0 +1,65 @@ +SET join_algorithm = 'full_sorting_merge'; +SET allow_experimental_analyzer = 1; + +DROP TABLE IF EXISTS events0; + +CREATE TABLE events0 ( + begin Float64, + value Int32 +) ENGINE = MergeTree ORDER BY begin; + +INSERT INTO events0 VALUES (1.0, 0), (3.0, 1), (6.0, 2), (8.0, 3); + +SELECT p.ts, e.value +FROM + (SELECT number :: Float64 AS ts FROM numbers(10)) p +ASOF JOIN events0 e +ON p.ts >= e.begin +ORDER BY p.ts ASC; + +SELECT p.ts, e.value +FROM + (SELECT number :: Float64 AS ts FROM numbers(10)) p +ASOF LEFT JOIN events0 e +ON p.ts >= e.begin +ORDER BY p.ts ASC +-- SETTINGS join_use_nulls = 1 +; + +DROP TABLE IF EXISTS events0; + +DROP TABLE IF EXISTS events; +DROP TABLE IF EXISTS probes; + +CREATE TABLE events ( + key Int32, + begin Float64, + value Int32 +) ENGINE = MergeTree ORDER BY (key, begin); + +INSERT INTO events VALUES (1, 1.0, 0), (1, 3.0, 1), (1, 6.0, 2), (1, 8.0, 3), (2, 0.0, 10), (2, 7.0, 20), (2, 11.0, 30); + +CREATE TABLE probes ( + key Int32, + ts Float64 +) ENGINE = MergeTree ORDER BY (key, ts) AS +SELECT + key.number, + ts.number +FROM + numbers(1, 2) as key, + numbers(10) as ts +SETTINGS join_algorithm = 'hash'; + +SELECT p.key, p.ts, e.value +FROM probes p +ASOF JOIN events e +ON p.key = e.key AND p.ts >= e.begin +ORDER BY p.key, p.ts ASC; + +SELECT p.key, p.ts, e.value +FROM probes p +ASOF LEFT JOIN events e +ON p.key = e.key AND p.ts >= e.begin +ORDER BY p.key, p.ts ASC NULLS FIRST; + diff --git a/tests/queries/0_stateless/03144_parallel_alter_add_drop_column_zookeeper_on_steroids.sh b/tests/queries/0_stateless/03144_parallel_alter_add_drop_column_zookeeper_on_steroids.sh index ea7bb8f7ad0..b3d0b08948b 100755 --- a/tests/queries/0_stateless/03144_parallel_alter_add_drop_column_zookeeper_on_steroids.sh +++ b/tests/queries/0_stateless/03144_parallel_alter_add_drop_column_zookeeper_on_steroids.sh @@ -30,9 +30,9 @@ function alter_thread() while true; do REPLICA=$(($RANDOM % 3 + 1)) ADD=$(($RANDOM % 5 + 1)) - $CLICKHOUSE_CLIENT --query "ALTER TABLE concurrent_alter_add_drop_steroids_$REPLICA ADD COLUMN value$ADD UInt32 DEFAULT 42 SETTINGS replication_alter_partitions_sync=0"; # additionaly we don't wait anything for more heavy concurrency + $CLICKHOUSE_CLIENT --query "ALTER TABLE concurrent_alter_add_drop_steroids_$REPLICA ADD COLUMN value$ADD UInt32 DEFAULT 42 SETTINGS replication_alter_partitions_sync=0"; # additionally we don't wait anything for more heavy concurrency DROP=$(($RANDOM % 5 + 1)) - $CLICKHOUSE_CLIENT --query "ALTER TABLE concurrent_alter_add_drop_steroids_$REPLICA DROP COLUMN value$DROP SETTINGS replication_alter_partitions_sync=0"; # additionaly we don't wait anything for more heavy concurrency + $CLICKHOUSE_CLIENT --query "ALTER TABLE concurrent_alter_add_drop_steroids_$REPLICA DROP COLUMN value$DROP SETTINGS replication_alter_partitions_sync=0"; # additionally we don't wait anything for more heavy concurrency sleep 0.$RANDOM done } @@ -85,7 +85,7 @@ export -f optimize_thread; export -f insert_thread; -TIMEOUT=30 +TIMEOUT=20 # Sometimes we detach and attach tables timeout $TIMEOUT bash -c alter_thread 2> /dev/null & diff --git a/tests/queries/0_stateless/03145_asof_join_ddb_inequalities.reference b/tests/queries/0_stateless/03145_asof_join_ddb_inequalities.reference new file mode 100644 index 00000000000..4aac918c98c --- /dev/null +++ b/tests/queries/0_stateless/03145_asof_join_ddb_inequalities.reference @@ -0,0 +1,73 @@ +- +2023-03-21 12:00:00 1970-01-01 00:00:00 -1 +2023-03-21 13:00:00 1970-01-01 00:00:00 -1 +2023-03-21 14:00:00 2023-03-21 13:00:00 0 +2023-03-21 15:00:00 2023-03-21 14:00:00 1 +2023-03-21 16:00:00 2023-03-21 15:00:00 2 +2023-03-21 17:00:00 2023-03-21 16:00:00 3 +2023-03-21 18:00:00 2023-03-21 16:00:00 3 +2023-03-21 19:00:00 2023-03-21 16:00:00 3 +2023-03-21 20:00:00 2023-03-21 16:00:00 3 +2023-03-21 21:00:00 2023-03-21 16:00:00 3 +2027-10-18 11:03:27 2023-03-21 16:00:00 3 +- +2023-03-21 12:00:00 1970-01-01 00:00:00 -1 +2023-03-21 13:00:00 1970-01-01 00:00:00 -1 +2023-03-21 14:00:00 2023-03-21 13:00:00 0 +2023-03-21 15:00:00 2023-03-21 14:00:00 1 +2023-03-21 16:00:00 2023-03-21 15:00:00 2 +2023-03-21 17:00:00 2023-03-21 16:00:00 3 +2023-03-21 18:00:00 2023-03-21 16:00:00 3 +2023-03-21 19:00:00 2023-03-21 16:00:00 3 +2023-03-21 20:00:00 2023-03-21 16:00:00 3 +2023-03-21 21:00:00 2023-03-21 16:00:00 3 +2027-10-18 11:03:27 2023-03-21 16:00:00 3 +\N \N \N +2023-03-21 12:00:00 2023-03-21 13:00:00 0 +2023-03-21 13:00:00 2023-03-21 13:00:00 0 +2023-03-21 14:00:00 2023-03-21 14:00:00 1 +2023-03-21 15:00:00 2023-03-21 15:00:00 2 +2023-03-21 16:00:00 2023-03-21 16:00:00 3 +2023-03-21 17:00:00 2027-10-18 11:03:27 9 +2023-03-21 18:00:00 2027-10-18 11:03:27 9 +2023-03-21 19:00:00 2027-10-18 11:03:27 9 +2023-03-21 20:00:00 2027-10-18 11:03:27 9 +2023-03-21 21:00:00 2027-10-18 11:03:27 9 +2027-10-18 11:03:27 2027-10-18 11:03:27 9 +- +2023-03-21 12:00:00 2023-03-21 13:00:00 0 +2023-03-21 13:00:00 2023-03-21 13:00:00 0 +2023-03-21 14:00:00 2023-03-21 14:00:00 1 +2023-03-21 15:00:00 2023-03-21 15:00:00 2 +2023-03-21 16:00:00 2023-03-21 16:00:00 3 +2023-03-21 17:00:00 2027-10-18 11:03:27 9 +2023-03-21 18:00:00 2027-10-18 11:03:27 9 +2023-03-21 19:00:00 2027-10-18 11:03:27 9 +2023-03-21 20:00:00 2027-10-18 11:03:27 9 +2023-03-21 21:00:00 2027-10-18 11:03:27 9 +2027-10-18 11:03:27 2027-10-18 11:03:27 9 +\N \N \N +- +2023-03-21 12:00:00 2023-03-21 13:00:00 0 +2023-03-21 13:00:00 2023-03-21 14:00:00 1 +2023-03-21 14:00:00 2023-03-21 15:00:00 2 +2023-03-21 15:00:00 2023-03-21 16:00:00 3 +2023-03-21 16:00:00 2027-10-18 11:03:27 9 +2023-03-21 17:00:00 2027-10-18 11:03:27 9 +2023-03-21 18:00:00 2027-10-18 11:03:27 9 +2023-03-21 19:00:00 2027-10-18 11:03:27 9 +2023-03-21 20:00:00 2027-10-18 11:03:27 9 +2023-03-21 21:00:00 2027-10-18 11:03:27 9 +- +2023-03-21 12:00:00 2023-03-21 13:00:00 0 +2023-03-21 13:00:00 2023-03-21 14:00:00 1 +2023-03-21 14:00:00 2023-03-21 15:00:00 2 +2023-03-21 15:00:00 2023-03-21 16:00:00 3 +2023-03-21 16:00:00 2027-10-18 11:03:27 9 +2023-03-21 17:00:00 2027-10-18 11:03:27 9 +2023-03-21 18:00:00 2027-10-18 11:03:27 9 +2023-03-21 19:00:00 2027-10-18 11:03:27 9 +2023-03-21 20:00:00 2027-10-18 11:03:27 9 +2023-03-21 21:00:00 2027-10-18 11:03:27 9 +2027-10-18 11:03:27 \N \N +\N \N \N diff --git a/tests/queries/0_stateless/03145_asof_join_ddb_inequalities.sql b/tests/queries/0_stateless/03145_asof_join_ddb_inequalities.sql new file mode 100644 index 00000000000..d67aa254bd6 --- /dev/null +++ b/tests/queries/0_stateless/03145_asof_join_ddb_inequalities.sql @@ -0,0 +1,66 @@ +DROP TABLE IF EXISTS events0; +DROP TABLE IF EXISTS probe0; + +SET allow_experimental_analyzer = 1; +SET join_algorithm = 'full_sorting_merge'; + +CREATE TABLE events0 ( + begin Nullable(DateTime('UTC')), + value Int32 +) ENGINE = MergeTree ORDER BY tuple(); + +INSERT INTO events0 SELECT toDateTime('2023-03-21 13:00:00', 'UTC') + INTERVAL number HOUR, number FROM numbers(4); +INSERT INTO events0 VALUES (NULL, -10),('0000-01-01 00:00:00', -1), ('9999-12-31 23:59:59', 9); + +CREATE TABLE probe0 ( + begin Nullable(DateTime('UTC')) +) ENGINE = MergeTree ORDER BY tuple(); + +INSERT INTO probe0 SELECT toDateTime('2023-03-21 12:00:00', 'UTC') + INTERVAl number HOUR FROM numbers(10); +INSERT INTO probe0 VALUES (NULL),('9999-12-31 23:59:59'); + +SET join_use_nulls = 1; + +SELECT '-'; +SELECT p.begin, e.begin, e.value +FROM probe0 p +ASOF JOIN events0 e +ON p.begin > e.begin +ORDER BY p.begin ASC; + +SELECT '-'; +SELECT p.begin, e.begin, e.value +FROM probe0 p +ASOF LEFT JOIN events0 e +ON p.begin > e.begin +ORDER BY p.begin ASC; + +SELECT p.begin, e.begin, e.value +FROM probe0 p +ASOF JOIN events0 e +ON p.begin <= e.begin +ORDER BY p.begin ASC; + +SELECT '-'; +SELECT p.begin, e.begin, e.value +FROM probe0 p +ASOF LEFT JOIN events0 e +ON p.begin <= e.begin +ORDER BY p.begin ASC; + +SELECT '-'; +SELECT p.begin, e.begin, e.value +FROM probe0 p +ASOF JOIN events0 e +ON p.begin < e.begin +ORDER BY p.begin ASC; + +SELECT '-'; +SELECT p.begin, e.begin, e.value +FROM probe0 p +ASOF LEFT JOIN events0 e +ON p.begin < e.begin +ORDER BY p.begin ASC; + +DROP TABLE IF EXISTS events0; +DROP TABLE IF EXISTS probe0; diff --git a/tests/queries/0_stateless/03145_non_loaded_projection_backup.sh b/tests/queries/0_stateless/03145_non_loaded_projection_backup.sh index 7df2118ad0c..95aef9bbc5b 100755 --- a/tests/queries/0_stateless/03145_non_loaded_projection_backup.sh +++ b/tests/queries/0_stateless/03145_non_loaded_projection_backup.sh @@ -6,8 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT -nm -q " drop table if exists tp_1; -create table tp_1 (x Int32, y Int32, projection p (select x, y order by x)) engine = MergeTree order by y partition by intDiv(y, 100); -system stop merges tp_1; +create table tp_1 (x Int32, y Int32, projection p (select x, y order by x)) engine = MergeTree order by y partition by intDiv(y, 100) settings max_parts_to_merge_at_once=1; insert into tp_1 select number, number from numbers(3); set mutations_sync = 2; @@ -39,7 +38,6 @@ $CLICKHOUSE_CLIENT -nm -q " set send_logs_level='fatal'; drop table tp_1; restore table tp_1 from Disk('backups', '$backup_id'); -system stop merges tp_1; " | grep -o "RESTORED" $CLICKHOUSE_CLIENT -q "select count() from tp_1;" diff --git a/tests/queries/0_stateless/03146_asof_join_ddb_merge_long.reference b/tests/queries/0_stateless/03146_asof_join_ddb_merge_long.reference new file mode 100644 index 00000000000..ca481c7fff0 --- /dev/null +++ b/tests/queries/0_stateless/03146_asof_join_ddb_merge_long.reference @@ -0,0 +1,2 @@ +26790 1488 +26790 1488 diff --git a/tests/queries/0_stateless/03146_asof_join_ddb_merge_long.sql.j2 b/tests/queries/0_stateless/03146_asof_join_ddb_merge_long.sql.j2 new file mode 100644 index 00000000000..49ba70c471e --- /dev/null +++ b/tests/queries/0_stateless/03146_asof_join_ddb_merge_long.sql.j2 @@ -0,0 +1,39 @@ +-- Tags: long + +SET allow_experimental_analyzer=1; + +SET session_timezone = 'UTC'; + +{% for join_algorithm in ['default', 'full_sorting_merge'] -%} + +SET join_algorithm = '{{ join_algorithm }}'; + +-- TODO: enable once USING and `join_use_nulls` is supported by `full_sorting_merge` +-- SET join_use_nulls = 1; + +WITH build AS ( + SELECT + tk.number AS k, + toDateTime('2021-01-01 00:00:00') + INTERVAL i.number SECONDS AS t, + i.number % 37 AS v + FROM numbers(3000000) AS i + CROSS JOIN numbers(2) AS tk + SETTINGS join_algorithm = 'hash', join_use_nulls = 0 +), +probe AS ( + SELECT + tk.number AS k, + toDateTime('2021-01-01 00:00:30') + INTERVAL tt.number HOUR AS t + FROM numbers(2) AS tk + CROSS JOIN numbers(toUInt32((toDateTime('2021-02-01 00:00:30') - toDateTime('2021-01-01 00:00:30')) / 3600)) AS tt + SETTINGS join_algorithm = 'hash', join_use_nulls = 0 +) +SELECT + SUM(v) AS v, + COUNT(*) AS n +FROM probe +ASOF LEFT JOIN build +USING (k, t) +; + +{% endfor -%} diff --git a/tests/queries/0_stateless/03147_asof_join_ddb_missing.reference b/tests/queries/0_stateless/03147_asof_join_ddb_missing.reference new file mode 100644 index 00000000000..11eb84463f4 --- /dev/null +++ b/tests/queries/0_stateless/03147_asof_join_ddb_missing.reference @@ -0,0 +1,10 @@ +108 +108 27 +513 +1218 +3528 +14553 +121275 +1495503 +12462525 +1249625025 diff --git a/tests/queries/0_stateless/03147_asof_join_ddb_missing.sql b/tests/queries/0_stateless/03147_asof_join_ddb_missing.sql new file mode 100644 index 00000000000..95a5f8ab3ff --- /dev/null +++ b/tests/queries/0_stateless/03147_asof_join_ddb_missing.sql @@ -0,0 +1,186 @@ +SET allow_experimental_analyzer=1; + +SET session_timezone = 'UTC'; +SET joined_subquery_requires_alias = 0; +SET allow_experimental_analyzer = 1; +SET join_algorithm = 'full_sorting_merge'; + +-- # 10 dates, 5 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(10), (SELECT number AS k FROM numbers(5)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); + +-- # Coverage: Missing right side bin +WITH build AS ( + SELECT + k * 2 AS k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(10), (SELECT number AS k FROM numbers(5)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + intDiv(k, 2) AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v), COUNT(*) +FROM probe ASOF JOIN build USING (k, t); + +-- # 20 dates, 5 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(20), (SELECT number AS k FROM numbers(5)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); + +-- # 30 dates, 5 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(30), (SELECT number AS k FROM numbers(5)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); + +-- # 50 dates, 5 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(50), (SELECT number AS k FROM numbers(5)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); + +-- # 100 dates, 5 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(100), (SELECT number AS k FROM numbers(5)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); + +-- # 100 dates, 50 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(100), (SELECT number AS k FROM numbers(50)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); + +-- # 1000 dates, 5 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(1000), (SELECT number AS k FROM numbers(5)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); + +-- # 1000 dates, 50 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(1000), (SELECT number AS k FROM numbers(50)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); + +-- # 10000 dates, 50 keys +WITH build AS ( + SELECT + k, + toDateTime('2001-01-01 00:00:00') + INTERVAL number MINUTE AS t, + number AS v + FROM numbers(10000), (SELECT number AS k FROM numbers(50)) + SETTINGS join_algorithm = 'default' +), +probe AS ( + SELECT + k * 2 AS k, + t - INTERVAL 30 SECOND AS t + FROM build +) +SELECT SUM(v) +FROM probe ASOF JOIN build USING (k, t); diff --git a/tests/queries/0_stateless/03148_asof_join_ddb_subquery.reference b/tests/queries/0_stateless/03148_asof_join_ddb_subquery.reference new file mode 100644 index 00000000000..387a4a8f249 --- /dev/null +++ b/tests/queries/0_stateless/03148_asof_join_ddb_subquery.reference @@ -0,0 +1,4 @@ +1 1 +3 1 +6 1 +8 1 diff --git a/tests/queries/0_stateless/03148_asof_join_ddb_subquery.sql b/tests/queries/0_stateless/03148_asof_join_ddb_subquery.sql new file mode 100644 index 00000000000..2ddf0f09b1e --- /dev/null +++ b/tests/queries/0_stateless/03148_asof_join_ddb_subquery.sql @@ -0,0 +1,29 @@ +DROP TABLE IF EXISTS events; +CREATE TABLE events (begin Float64, value Int32) ENGINE = MergeTree() ORDER BY begin; + +INSERT INTO events VALUES (1, 0), (3, 1), (6, 2), (8, 3); + +SET allow_experimental_analyzer = 1; +SET join_algorithm = 'full_sorting_merge'; +SET joined_subquery_requires_alias = 0; + +SELECT + begin, + value IN ( + SELECT e1.value + FROM ( + SELECT * + FROM events e1 + WHERE e1.value = events.value + ) AS e1 + ASOF JOIN ( + SELECT number :: Float64 AS begin + FROM numbers(10) + WHERE number >= 1 AND number < 10 + ) + USING (begin) + ) +FROM events +ORDER BY begin ASC; + +DROP TABLE IF EXISTS events; diff --git a/tests/queries/0_stateless/03148_async_queries_in_query_log_errors.sh b/tests/queries/0_stateless/03148_async_queries_in_query_log_errors.sh index 9c290133bf9..2b4b96a9cbf 100755 --- a/tests/queries/0_stateless/03148_async_queries_in_query_log_errors.sh +++ b/tests/queries/0_stateless/03148_async_queries_in_query_log_errors.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: no-parallel CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/03149_asof_join_ddb_timestamps.reference b/tests/queries/0_stateless/03149_asof_join_ddb_timestamps.reference new file mode 100644 index 00000000000..7cfc85d23a5 --- /dev/null +++ b/tests/queries/0_stateless/03149_asof_join_ddb_timestamps.reference @@ -0,0 +1,56 @@ +2023-03-21 13:00:00 0 +2023-03-21 14:00:00 1 +2023-03-21 15:00:00 2 +2023-03-21 16:00:00 3 +2023-03-21 17:00:00 3 +2023-03-21 18:00:00 3 +2023-03-21 19:00:00 3 +2023-03-21 20:00:00 3 +2023-03-21 21:00:00 3 +2106-02-07 06:28:15 9 +2023-03-21 13:00:00 0 +2023-03-21 14:00:00 1 +2023-03-21 15:00:00 2 +2023-03-21 16:00:00 3 +2023-03-21 17:00:00 3 +2023-03-21 18:00:00 3 +2023-03-21 19:00:00 3 +2023-03-21 20:00:00 3 +2023-03-21 21:00:00 3 +2106-02-07 06:28:15 9 +2023-03-21 12:00:00 \N +2023-03-21 13:00:00 0 +2023-03-21 14:00:00 1 +2023-03-21 15:00:00 2 +2023-03-21 16:00:00 3 +2023-03-21 17:00:00 3 +2023-03-21 18:00:00 3 +2023-03-21 19:00:00 3 +2023-03-21 20:00:00 3 +2023-03-21 21:00:00 3 +2106-02-07 06:28:15 9 +\N \N +2023-03-21 12:00:00 0 +2023-03-21 13:00:00 0 +2023-03-21 14:00:00 1 +2023-03-21 15:00:00 2 +2023-03-21 16:00:00 3 +2023-03-21 17:00:00 3 +2023-03-21 18:00:00 3 +2023-03-21 19:00:00 3 +2023-03-21 20:00:00 3 +2023-03-21 21:00:00 3 +2106-02-07 06:28:15 9 +\N 0 +2023-03-21 12:00:00 \N +2023-03-21 13:00:00 \N +2023-03-21 14:00:00 \N +2023-03-21 15:00:00 \N +2023-03-21 16:00:00 \N +2023-03-21 17:00:00 \N +2023-03-21 18:00:00 \N +2023-03-21 19:00:00 \N +2023-03-21 20:00:00 \N +2023-03-21 21:00:00 \N +2106-02-07 06:28:15 \N +\N \N diff --git a/tests/queries/0_stateless/03149_asof_join_ddb_timestamps.sql b/tests/queries/0_stateless/03149_asof_join_ddb_timestamps.sql new file mode 100644 index 00000000000..cd83d62dc70 --- /dev/null +++ b/tests/queries/0_stateless/03149_asof_join_ddb_timestamps.sql @@ -0,0 +1,95 @@ +DROP TABLE IF EXISTS events0; +DROP TABLE IF EXISTS probe0; + +SET session_timezone = 'UTC'; +SET allow_experimental_analyzer = 1; +SET join_algorithm = 'full_sorting_merge'; +SET join_use_nulls = 1; + +CREATE TABLE events0 +ENGINE = MergeTree() +ORDER BY COALESCE(begin, toDateTime('9999-12-31 23:59:59')) +AS +SELECT + toNullable(toDateTime('2023-03-21 13:00:00') + INTERVAL number HOUR) AS begin, + number AS value +FROM numbers(4); + +INSERT INTO events0 VALUES (NULL, -1), (toDateTime('9999-12-31 23:59:59'), 9); + +CREATE TABLE probe0 +ENGINE = MergeTree() +ORDER BY COALESCE(begin, toDateTime('9999-12-31 23:59:59')) +AS +SELECT + toNullable(toDateTime('2023-03-21 12:00:00') + INTERVAL number HOUR) AS begin +FROM numbers(10); + +INSERT INTO probe0 VALUES (NULL), (toDateTime('9999-12-31 23:59:59')); + +SELECT + p.begin, + e.value +FROM + probe0 p + ASOF JOIN events0 e ON p.begin >= e.begin +ORDER BY p.begin ASC; + +SELECT + p.begin, + e.value +FROM + probe0 p + ASOF JOIN events0 e USING (begin) +ORDER BY p.begin ASC +SETTINGS join_use_nulls = 0 +; + +SELECT + p.begin, + e.value +FROM + probe0 p + ASOF LEFT JOIN events0 e ON p.begin >= e.begin +ORDER BY p.begin ASC; + +SELECT + p.begin, + e.value +FROM + probe0 p + ASOF LEFT JOIN events0 e USING (begin) +ORDER BY p.begin ASC +SETTINGS join_use_nulls = 0 +; + +SELECT + p.begin, + e.value +FROM + probe0 p + ASOF RIGHT JOIN events0 e ON p.begin >= e.begin +ORDER BY e.begin ASC; -- { serverError NOT_IMPLEMENTED} + +SELECT + p.begin, + e.value +FROM + probe0 p + ASOF RIGHT JOIN events0 e USING (begin) +ORDER BY e.begin ASC; -- { serverError NOT_IMPLEMENTED} + + +SELECT + p.begin, + e.value +FROM + probe0 p + ASOF LEFT JOIN ( + SELECT * FROM events0 WHERE log(value + 5) > 10 + ) e ON p.begin + INTERVAL 2 HOUR >= e.begin + INTERVAL 1 HOUR +ORDER BY p.begin ASC; + + +DROP TABLE IF EXISTS events0; +DROP TABLE IF EXISTS probe0; diff --git a/tests/queries/0_stateless/03153_format_regexp_usability.sh b/tests/queries/0_stateless/03153_format_regexp_usability.sh index 03bed10dd17..561de3be893 100755 --- a/tests/queries/0_stateless/03153_format_regexp_usability.sh +++ b/tests/queries/0_stateless/03153_format_regexp_usability.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-parallel, no-ordinary-database, long +# Tags: no-fasttest, no-ordinary-database, long CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/03161_lightweight_delete_projection.reference b/tests/queries/0_stateless/03161_lightweight_delete_projection.reference index e69de29bb2d..c5a6cbab0bc 100644 --- a/tests/queries/0_stateless/03161_lightweight_delete_projection.reference +++ b/tests/queries/0_stateless/03161_lightweight_delete_projection.reference @@ -0,0 +1,2 @@ +1231 John 33 +8888 Alice 50 diff --git a/tests/queries/0_stateless/03161_lightweight_delete_projection.sql b/tests/queries/0_stateless/03161_lightweight_delete_projection.sql index cd29fae8fd7..b189388e356 100644 --- a/tests/queries/0_stateless/03161_lightweight_delete_projection.sql +++ b/tests/queries/0_stateless/03161_lightweight_delete_projection.sql @@ -5,7 +5,8 @@ CREATE TABLE users ( uid Int16, name String, age Int16, - projection p1 (select count(), age group by age) + projection p1 (select count(), age group by age), + projection p2 (select age, name group by age, name) ) ENGINE = MergeTree order by uid; INSERT INTO users VALUES (1231, 'John', 33); @@ -13,3 +14,18 @@ INSERT INTO users VALUES (6666, 'Ksenia', 48); INSERT INTO users VALUES (8888, 'Alice', 50); DELETE FROM users WHERE 1; -- { serverError NOT_IMPLEMENTED } + +DELETE FROM users WHERE uid = 8888 SETTINGS lightweight_mutation_projection_mode = 'throw'; -- { serverError NOT_IMPLEMENTED } + +DELETE FROM users WHERE uid = 6666 SETTINGS lightweight_mutation_projection_mode = 'drop'; + +-- expecting no projection +SELECT + name, + `table` +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'users'); + +SELECT * FROM users ORDER BY uid; + +DROP TABLE users; diff --git a/tests/queries/0_stateless/03164_materialize_statistics.sql b/tests/queries/0_stateless/03164_materialize_statistics.sql deleted file mode 100644 index 43c5724dd59..00000000000 --- a/tests/queries/0_stateless/03164_materialize_statistics.sql +++ /dev/null @@ -1,49 +0,0 @@ -DROP TABLE IF EXISTS t_statistics_materialize; - -SET allow_experimental_analyzer = 1; -SET allow_experimental_statistics = 1; -SET allow_statistics_optimize = 1; -SET materialize_statistics_on_insert = 0; - -CREATE TABLE t_statistics_materialize -( - a Int64 STATISTICS(tdigest), - b Int16 STATISTICS(tdigest), -) ENGINE = MergeTree() ORDER BY tuple() -SETTINGS min_bytes_for_wide_part = 0, enable_vertical_merge_algorithm = 0; -- TODO: there is a bug in vertical merge with statistics. - -INSERT INTO t_statistics_materialize SELECT number, -number FROM system.numbers LIMIT 10000; - -SELECT count(*) FROM t_statistics_materialize WHERE b < 10 and a < 10 SETTINGS log_comment = 'statistics not used'; - -OPTIMIZE TABLE t_statistics_materialize FINAL; - -SELECT count(*) FROM t_statistics_materialize WHERE b < 10 and a < 10 SETTINGS log_comment = 'statistics used after merge'; - -TRUNCATE TABLE t_statistics_materialize; -SET mutations_sync = 2; - -INSERT INTO t_statistics_materialize SELECT number, -number FROM system.numbers LIMIT 10000; -ALTER TABLE t_statistics_materialize MATERIALIZE STATISTICS a, b; - -SELECT count(*) FROM t_statistics_materialize WHERE b < 10 and a < 10 SETTINGS log_comment = 'statistics used after materialize'; - -DROP TABLE t_statistics_materialize; - -SYSTEM FLUSH LOGS; - -SELECT log_comment, message FROM system.text_log JOIN -( - SELECT Settings['log_comment'] AS log_comment, query_id FROM system.query_log - WHERE current_database = currentDatabase() - AND query LIKE 'SELECT count(*) FROM t_statistics_materialize%' - AND type = 'QueryFinish' -) AS query_log USING (query_id) -WHERE message LIKE '%moved to PREWHERE%' -ORDER BY event_time_microseconds; - -SELECT count(), sum(ProfileEvents['MergeTreeDataWriterStatisticsCalculationMicroseconds']) -FROM system.query_log -WHERE current_database = currentDatabase() - AND query LIKE 'INSERT INTO t_statistics_materialize SELECT%' - AND type = 'QueryFinish'; diff --git a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql index 652b27b8a67..ac2070fbd76 100644 --- a/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql +++ b/tests/queries/0_stateless/03164_s3_settings_for_queries_and_merges.sql @@ -1,4 +1,4 @@ --- Tags: no-random-settings, no-fasttest +-- Tags: no-random-settings, no-fasttest, no-parallel SET allow_prefetched_read_pool_for_remote_filesystem=0; SET allow_prefetched_read_pool_for_local_filesystem=0; diff --git a/tests/queries/0_stateless/03165_round_scale_as_column.reference b/tests/queries/0_stateless/03165_round_scale_as_column.reference index 9ad25ed466a..e0c9b6959ee 100644 --- a/tests/queries/0_stateless/03165_round_scale_as_column.reference +++ b/tests/queries/0_stateless/03165_round_scale_as_column.reference @@ -2162,4 +2162,17 @@ CHECKPOINT2 10 1.6275 1.6275 1.6275 1.6275 1 1 +3 +3.1 +3.14 +3.142 +3.1416 +3.14159 +3.141593 +3.1415927 +3.14159265 +3.141592654 +42 +42.4 +42.42 1 diff --git a/tests/queries/0_stateless/03165_round_scale_as_column.sql b/tests/queries/0_stateless/03165_round_scale_as_column.sql index 229f705808d..adae36564b8 100644 --- a/tests/queries/0_stateless/03165_round_scale_as_column.sql +++ b/tests/queries/0_stateless/03165_round_scale_as_column.sql @@ -118,6 +118,7 @@ DROP TABLE tab; SELECT round(1, 1); SELECT round(materialize(1), materialize(1)); -SELECT round(1, materialize(1)); --{serverError ILLEGAL_COLUMN} +SELECT round(pi(), number) FROM numbers(10); +SELECT round(toDecimal32(42.42, 2), number) from numbers(3); SELECT round(materialize(1), 1); SELECT materialize(10.1) AS x, ceil(x, toUInt256(123)); --{serverError ILLEGAL_TYPE_OF_ARGUMENT} diff --git a/tests/queries/0_stateless/03167_base64_url_functions.sql b/tests/queries/0_stateless/03167_base64_url_functions.sql index 674f1ae498b..6c394ba6c3a 100644 --- a/tests/queries/0_stateless/03167_base64_url_functions.sql +++ b/tests/queries/0_stateless/03167_base64_url_functions.sql @@ -2,35 +2,35 @@ -- no-fasttest because aklomp-base64 library is required -- incorrect number of arguments -SELECT base64UrlEncode(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -SELECT base64UrlDecode(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -SELECT tryBase64UrlDecode(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -SELECT base64UrlEncode('foo', 'excess argument'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -SELECT base64UrlDecode('foo', 'excess argument'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -SELECT tryBase64UrlDecode('foo', 'excess argument'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT base64URLEncode(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT base64URLDecode(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT tryBase64URLDecode(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT base64URLEncode('foo', 'excess argument'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT base64URLDecode('foo', 'excess argument'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +SELECT tryBase64URLDecode('foo', 'excess argument'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -- test with valid inputs -SELECT 'https://clickhouse.com' AS original, base64UrlEncode(original) AS encoded, base64UrlDecode(encoded), tryBase64UrlDecode(encoded); -SELECT '12?' AS original, base64UrlEncode(original) AS encoded, base64UrlDecode(encoded), tryBase64UrlDecode(encoded); -SELECT 'https://www.google.com/search?q=clickhouse+base64+decode&sca_esv=739f8bb380e4c7ed&ei=TfRiZqCDIrmnwPAP2KLRkA8&ved=0ahUKEwjg3ZHitsmGAxW5ExAIHVhRFPIQ4dUDCBA&uact=5&oq=clickhouse+base64+decode' AS original, base64UrlEncode(original) AS encoded, base64UrlDecode(encoded), tryBase64UrlDecode(encoded); +SELECT 'https://clickhouse.com' AS original, base64URLEncode(original) AS encoded, base64URLDecode(encoded), tryBase64URLDecode(encoded); +SELECT '12?' AS original, base64URLEncode(original) AS encoded, base64URLDecode(encoded), tryBase64URLDecode(encoded); +SELECT 'https://www.google.com/search?q=clickhouse+base64+decode&sca_esv=739f8bb380e4c7ed&ei=TfRiZqCDIrmnwPAP2KLRkA8&ved=0ahUKEwjg3ZHitsmGAxW5ExAIHVhRFPIQ4dUDCBA&uact=5&oq=clickhouse+base64+decode' AS original, base64URLEncode(original) AS encoded, base64URLDecode(encoded), tryBase64URLDecode(encoded); -- encoded value has no padding -SELECT 'aHR0cHM6Ly9jbGlj' AS encoded, base64UrlDecode(encoded), tryBase64UrlDecode(encoded); +SELECT 'aHR0cHM6Ly9jbGlj' AS encoded, base64URLDecode(encoded), tryBase64URLDecode(encoded); -- encoded value has one-byte padding -SELECT 'aHR0cHM6Ly9jbGlja2g' AS encoded, base64UrlDecode(encoded), tryBase64UrlDecode(encoded); +SELECT 'aHR0cHM6Ly9jbGlja2g' AS encoded, base64URLDecode(encoded), tryBase64URLDecode(encoded); -- encoded value has two-bytes padding -SELECT 'aHR0cHM6Ly9jbGljaw' AS encoded, base64UrlDecode(encoded), tryBase64UrlDecode(encoded); +SELECT 'aHR0cHM6Ly9jbGljaw' AS encoded, base64URLDecode(encoded), tryBase64URLDecode(encoded); -- test with invalid inputs -SELECT base64UrlDecode('https://clickhouse.com'); -- { serverError INCORRECT_DATA } -SELECT tryBase64UrlDecode('https://clickhouse.com'); -SELECT base64UrlDecode('12?'); -- { serverError INCORRECT_DATA } -SELECT tryBase64UrlDecode('12?'); -SELECT base64UrlDecode('aHR0cHM6Ly9jbGlja'); -- { serverError INCORRECT_DATA } -SELECT tryBase64UrlDecode('aHR0cHM6Ly9jbGlja'); +SELECT base64URLDecode('https://clickhouse.com'); -- { serverError INCORRECT_DATA } +SELECT tryBase64URLDecode('https://clickhouse.com'); +SELECT base64URLDecode('12?'); -- { serverError INCORRECT_DATA } +SELECT tryBase64URLDecode('12?'); +SELECT base64URLDecode('aHR0cHM6Ly9jbGlja'); -- { serverError INCORRECT_DATA } +SELECT tryBase64URLDecode('aHR0cHM6Ly9jbGlja'); -- test FixedString argument -SELECT toFixedString('https://clickhouse.com', 22) AS original, base64UrlEncode(original) AS encoded, base64UrlDecode(encoded), tryBase64UrlDecode(encoded); +SELECT toFixedString('https://clickhouse.com', 22) AS original, base64URLEncode(original) AS encoded, base64URLDecode(encoded), tryBase64URLDecode(encoded); diff --git a/tests/queries/0_stateless/03167_base64_url_functions_sh.reference b/tests/queries/0_stateless/03167_base64_url_functions_sh.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03167_base64_url_functions_sh.sh b/tests/queries/0_stateless/03167_base64_url_functions_sh.sh new file mode 100755 index 00000000000..57060b8c525 --- /dev/null +++ b/tests/queries/0_stateless/03167_base64_url_functions_sh.sh @@ -0,0 +1,166 @@ +#!/usr/bin/env bash +# Tags: no-fasttest +# shellcheck disable=SC2155 + +set -e + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +urls=( + "http://www.example.com" + "https://secure.example.com" + "http://example.com" + "https://www.example.org" + "https://subdomain.example.com" + "http://sub.sub.example.com" + "http://192.168.1.1" + "https://[2001:db8:85a3:8d3:1319:8a2e:370:7348]" + "http://example.com:8080" + "https://example.com:443" + "http://example.com/path/to/page.html" + "https://example.com/path/with/trailing/slash/" + "http://example.com/search?q=query&lang=en" + "https://example.com/path?param1=value1¶m2=value2" + "http://example.com/page.html#section1" + "https://example.com/document.pdf#page=10" + "http://user:password@example.com" + "https://user@example.com" + "https://user:pass@sub.example.com:8080/path/page.html?query=123#fragment" + "http://example.com/path%20with%20spaces" + "https://example.com/search?q=encode+this" + "http://例子.测试" + "https://mañana.com" + "http://example.com/%E2%82%AC" + "data:text/plain;base64,SGVsbG8sIFdvcmxkIQ==" + "file:///C:/path/to/file.txt" + "file:///home/user/document.pdf" + "ftp://ftp.example.com/pub/file.zip" + "ftps://secure-ftp.example.com/private/doc.pdf" + "mailto:user@example.com" + "mailto:user@example.com?subject=Hello&body=How%20are%20you" + "git://github.com/user/repo.git" + "ssh://user@host.xz:port/path/to/repo.git" + "https://example.com/path(1)/[2]/{3}" + "http://example.com/path;param?query,value" + "" + "http://" + "example.com" + "http:" + "//" + "?query=value" + "#fragment" + "http://?#" + "http://xn--bcher-kva.ch" + "https://xn--bcher-kva.xn--tckwe/xn--8ws00zhy3a/%E6%B8%AC%E8%A9%A6.php?xn--o39an51a5phao35a=xn--mgbh0fb&xn--fiq228c5hs=test" + "https://xn--3e0b707e.xn--79-8kcre8v3a/%ED%85%8C%EC%8A%A4%ED%8A%B8/%ED%8C%8C%EC%9D%BC.jsp?xn--i1b6b1a6a2e=xn--9t4b11yi5a&xn--3e0b707e=xn--80aaa1cbgbm" + "https://example.com/path?param=value&special=!@#$%^&*()" + + "http://example.com/path/with/~tilde" + "https://example.com/path/with/\`backtick\`" + + "https://example.com/path?param1=value1¶m2=value2¶m3=value3#section1#section2" + "http://example.com/page?q1=v1&q2=v2#frag1#frag2#frag3" + + "https://example.com/☃/snowman" + "http://example.com/path/⽇本語" + "https://example.com/ü/ñ/path?q=ç" + + "https://example.com/path/to/very/long/url/that/exceeds/two/hundred/and/fifty/five/characters/lorem/ipsum/dolor/sit/amet/consectetur/adipiscing/elit/sed/do/eiusmod/tempor/incididunt/ut/labore/et/dolore/magna/aliqua/ut/enim/ad/minim/veniam/quis/nostrud/exercitation/ullamco/laboris/nisi/ut/aliquip/ex/ea/commodo/consequat" + + "https://example.com//path///to//file" + "http://example.com/path?param1=value1&¶m2=value2&&¶m3=value3" + + "http://example.com/%70%61%74%68?%70%61%72%61%6d=%76%61%6c%75%65#%66%72%61%67%6d%65%6e%74" + + "HtTpS://ExAmPlE.cOm/PaTh" + "http://EXAMPLE.COM/PATH" + + "http://127.0.0.1:8080/path" + "https://[::1]/path" + "http://[2001:0db8:85a3:0000:0000:8a2e:0370:7334]:8080/path" + + "http://example.com:65535/path" + "https://example.com:0/path" + + "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAACklEQVR4nGMAAQAABQABDQottAAAAABJRU5ErkJggg==" + + "https://user:password@example.com:8080/path?query=value#fragment" + "ftp://anonymous:password@ftp.example.com/pub/" + + "http://example.com/path%20with%20spaces" + "https://example.com/search?q=query%20with%20spaces" + + "https://www.mañana.com/path" + "http://例子.测试/path" + "https://рм.рф/path" + + "https://user:pass@sub.example.com:8080/p/a/t/h?query=123&key=value#fragid1" + + "jdbc:mysql://localhost:3306/database" + "market://details?id=com.example.app" + "tel:+1-816-555-1212" + "sms:+18165551212" + + "http://[1080:0:0:0:8:800:200C:417A]/index.html" + "https://[2001:db8::1428:57ab]:8080/path" + + "http://.." + "http://../" + "http://??" + "http://??/" + "http:///a" + "http://example.com??" + "http://example.com??/" + "foo://example.com:8042/over/there?name=ferret#nose" + "//example.com/path" +) + + +base64URLEncode() { + echo -n "$1" | base64 -w0 | tr '+/' '-_' | tr -d '=' +} + +base64URLDecode() { + local len=$((${#1} % 4)) + local result="$1" + if [ $len -eq 2 ]; then result="$1"'==' + elif [ $len -eq 3 ]; then result="$1"'=' + fi + echo "$result" | tr '_-' '/+' | base64 -w0 -d +} + +test() { + local input="$1" + local encode_ch=$(${CLICKHOUSE_CLIENT} --query="SELECT base64URLEncode('$input')") + local encode_gold=$(base64URLEncode $input) + + local decode_ch=$(${CLICKHOUSE_CLIENT} --query="SELECT base64URLDecode('$encode_gold')") + local decode_gold=$(base64URLDecode $encode_gold) + + if [ "$encode_ch" != "$encode_gold" ]; then + echo "Input: $input" + echo "Expected: $encode_gold" + echo "Got: $encode_ch" + fi + + if [ "$decode_ch" != "$input" ] || [ "$decode_ch" != "$decode_gold" ]; then + echo "Input: $input" + echo "Decode gold: $decode_gold" + echo "Got: $decode_ch" + fi +} + + +for url in "${urls[@]}"; do + test "$url" +done + +# special case for ' +decode=$(${CLICKHOUSE_CLIENT} --query="SELECT base64URLDecode(base64URLEncode('http://example.com/!$&\'()*+,;=:@/path'))") +if [ "$decode" != "http://example.com/!$&\'()*+,;=:@/path" ]; then + echo "Special case fail" + echo "Got: $decode" +fi diff --git a/tests/queries/0_stateless/03168_inconsistent_ast_formatting.sql b/tests/queries/0_stateless/03168_inconsistent_ast_formatting.sql index d43d46d5b14..5333ea29ce7 100644 --- a/tests/queries/0_stateless/03168_inconsistent_ast_formatting.sql +++ b/tests/queries/0_stateless/03168_inconsistent_ast_formatting.sql @@ -1,4 +1,7 @@ create table a (x `Null`); -- { clientError SYNTAX_ERROR } create table a (x f(`Null`)); -- { clientError SYNTAX_ERROR } create table a (x Enum8(f(`Null`, 'World', 2))); -- { clientError SYNTAX_ERROR } -create table a (`value2` Enum8('Hello' = 1, equals(`Null`, 'World', 2), '!' = 3)); -- { clientError SYNTAX_ERROR } \ No newline at end of file +create table a (`value2` Enum8('Hello' = 1, equals(`Null`, 'World', 2), '!' = 3)); -- { clientError SYNTAX_ERROR } + +create table a (x Int8) engine Memory; +create table b empty as a; diff --git a/tests/queries/0_stateless/03168_read_in_order_buffering_1.reference b/tests/queries/0_stateless/03168_read_in_order_buffering_1.reference new file mode 100644 index 00000000000..306885a0974 --- /dev/null +++ b/tests/queries/0_stateless/03168_read_in_order_buffering_1.reference @@ -0,0 +1,6 @@ +1 +0 +1 +0 +0 +0 diff --git a/tests/queries/0_stateless/03168_read_in_order_buffering_1.sql b/tests/queries/0_stateless/03168_read_in_order_buffering_1.sql new file mode 100644 index 00000000000..75025dcadc8 --- /dev/null +++ b/tests/queries/0_stateless/03168_read_in_order_buffering_1.sql @@ -0,0 +1,45 @@ +DROP TABLE IF EXISTS t_read_in_order_1; + +CREATE TABLE t_read_in_order_1 (id UInt64, v UInt64) +ENGINE = MergeTree ORDER BY id +SETTINGS index_granularity = 1024, index_granularity_bytes = '10M'; + +INSERT INTO t_read_in_order_1 SELECT number, number FROM numbers(1000000); + +SET max_threads = 8; +SET optimize_read_in_order = 1; +SET read_in_order_use_buffering = 1; + +SELECT count() FROM +( + EXPLAIN PIPELINE SELECT * FROM t_read_in_order_1 ORDER BY id +) WHERE explain LIKE '%BufferChunks%'; + +SELECT count() FROM +( + EXPLAIN PIPELINE SELECT * FROM t_read_in_order_1 ORDER BY id LIMIT 10 +) WHERE explain LIKE '%BufferChunks%'; + +SELECT count() FROM +( + EXPLAIN PIPELINE SELECT * FROM t_read_in_order_1 WHERE v % 10 = 0 ORDER BY id LIMIT 10 +) WHERE explain LIKE '%BufferChunks%'; + +SET read_in_order_use_buffering = 0; + +SELECT count() FROM +( + EXPLAIN PIPELINE SELECT * FROM t_read_in_order_1 ORDER BY id +) WHERE explain LIKE '%BufferChunks%'; + +SELECT count() FROM +( + EXPLAIN PIPELINE SELECT * FROM t_read_in_order_1 ORDER BY id LIMIT 10 +) WHERE explain LIKE '%BufferChunks%'; + +SELECT count() FROM +( + EXPLAIN PIPELINE SELECT * FROM t_read_in_order_1 WHERE v % 10 = 0 ORDER BY id LIMIT 10 +) WHERE explain LIKE '%BufferChunks%'; + +DROP TABLE t_read_in_order_1; diff --git a/tests/queries/0_stateless/03168_read_in_order_buffering_2.reference b/tests/queries/0_stateless/03168_read_in_order_buffering_2.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03168_read_in_order_buffering_2.sql b/tests/queries/0_stateless/03168_read_in_order_buffering_2.sql new file mode 100644 index 00000000000..1d3a75412e0 --- /dev/null +++ b/tests/queries/0_stateless/03168_read_in_order_buffering_2.sql @@ -0,0 +1,17 @@ +-- Tags: long, no-random-settings, no-tsan, no-asan, no-msan, no-s3-storage + +DROP TABLE IF EXISTS t_read_in_order_2; + +CREATE TABLE t_read_in_order_2 (id UInt64, v UInt64) ENGINE = MergeTree ORDER BY id; + +INSERT INTO t_read_in_order_2 SELECT number, number FROM numbers(10000000); +OPTIMIZE TABLE t_read_in_order_2 FINAL; + +SET optimize_read_in_order = 1; +SET max_threads = 4; +SET read_in_order_use_buffering = 1; +SET max_memory_usage = '100M'; + +SELECT * FROM t_read_in_order_2 ORDER BY id FORMAT Null; + +DROP TABLE t_read_in_order_2; diff --git a/tests/queries/0_stateless/03169_cache_complex_dict_short_circuit_bug.sql b/tests/queries/0_stateless/03169_cache_complex_dict_short_circuit_bug.sql index 8463d13d251..f91aaf39081 100644 --- a/tests/queries/0_stateless/03169_cache_complex_dict_short_circuit_bug.sql +++ b/tests/queries/0_stateless/03169_cache_complex_dict_short_circuit_bug.sql @@ -27,5 +27,5 @@ LAYOUT(COMPLEX_KEY_CACHE(SIZE_IN_CELLS 10)); SELECT dictGetOrDefault('cache_dictionary_complex_key_simple_attributes_short_circuit', 'value_first', (number, concat(toString(number))), toString(materialize('default'))) AS value_first FROM system.numbers LIMIT 20 FORMAT Null; SELECT dictGetOrDefault('cache_dictionary_complex_key_simple_attributes_short_circuit', 'value_first', (number, concat(toString(number))), toString(materialize('default'))) AS value_first FROM system.numbers LIMIT 20 FORMAT Null; -DROP TABLE IF EXISTS complex_key_simple_attributes_source_short_circuit_table; DROP DICTIONARY IF EXISTS cache_dictionary_complex_key_simple_attributes_short_circuit; +DROP TABLE IF EXISTS complex_key_simple_attributes_source_short_circuit_table; diff --git a/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.reference b/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.reference new file mode 100644 index 00000000000..1fc6683620c --- /dev/null +++ b/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.reference @@ -0,0 +1,9 @@ +1 +2 1 +3 0 +0 450 +1 460 +2 470 +3 480 +4 490 +\N 4950 diff --git a/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.sql b/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.sql new file mode 100644 index 00000000000..f10019a78dd --- /dev/null +++ b/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.sql @@ -0,0 +1,49 @@ +SET optimize_functions_to_subcolumns = 1; +SET allow_experimental_analyzer = 1; + +DROP TABLE IF EXISTS t_func_to_subcolumns_map_2; + +CREATE TABLE t_func_to_subcolumns_map_2 (id UInt64, m Map(String, UInt64)) ENGINE = MergeTree ORDER BY id; + +INSERT INTO t_func_to_subcolumns_map_2 VALUES (1, map('aaa', 1, 'bbb', 2)) (2, map('ccc', 3)); + +SELECT sum(mapContains(m, toNullable('aaa'))) FROM t_func_to_subcolumns_map_2; + +DROP TABLE t_func_to_subcolumns_map_2; + +DROP TABLE IF EXISTS t_func_to_subcolumns_join; + +CREATE TABLE t_func_to_subcolumns_join (id UInt64, arr Array(UInt64), n Nullable(String), m Map(String, UInt64)) +ENGINE = MergeTree ORDER BY tuple(); + +INSERT INTO t_func_to_subcolumns_join VALUES (1, [1, 2, 3], 'abc', map('foo', 1, 'bar', 2)) (2, [], NULL, map()); + +SET join_use_nulls = 1; + +SELECT + id, + right.n IS NULL +FROM t_func_to_subcolumns_join AS left +FULL OUTER JOIN +( + SELECT + 1 AS id, + 'qqq' AS n + UNION ALL + SELECT + 3 AS id, + 'www' +) AS right USING (id) +WHERE empty(arr); + +DROP TABLE t_func_to_subcolumns_join; + +DROP TABLE IF EXISTS t_func_to_subcolumns_use_nulls; + +CREATE TABLE t_func_to_subcolumns_use_nulls (arr Array(UInt64), v UInt64) ENGINE = MergeTree ORDER BY tuple(); + +INSERT INTO t_func_to_subcolumns_use_nulls SELECT range(number % 10), number FROM numbers(100); + +SELECT length(arr) AS n, sum(v) FROM t_func_to_subcolumns_use_nulls GROUP BY n WITH ROLLUP HAVING n <= 4 OR isNull(n) ORDER BY n SETTINGS group_by_use_nulls = 1; + +DROP TABLE t_func_to_subcolumns_use_nulls; diff --git a/tests/queries/0_stateless/03172_dynamic_binary_serialization.reference b/tests/queries/0_stateless/03172_dynamic_binary_serialization.reference new file mode 100644 index 00000000000..f5668ed935b --- /dev/null +++ b/tests/queries/0_stateless/03172_dynamic_binary_serialization.reference @@ -0,0 +1,50 @@ +\N None +42 UInt8 +-42 Int8 +42 UInt16 +-42 Int16 +42 UInt32 +-42 Int32 +42 UInt64 +-42 Int64 +42 UInt128 +-42 Int128 +42 UInt256 +-42 Int256 +42.42 Float32 +42.42 Float64 +2020-01-01 Date +2020-01-01 Date32 +2020-01-01 00:00:00 DateTime(\'EST\') +2020-01-01 00:00:00 DateTime(\'CET\') +2020-01-01 00:00:00.000000 DateTime64(6, \'EST\') +2020-01-01 00:00:00.000000 DateTime64(6, \'CET\') +Hello, World! String +aaaaa FixedString(5) +a Enum8(\'c\' = -128, \'a\' = 1, \'b\' = 2) +a Enum16(\'c\' = -1280, \'a\' = 1, \'b\' = 2) +42.42 Decimal(9, 3) +42.42 Decimal(18, 3) +42.42 Decimal(38, 3) +42.42 Decimal(76, 3) +984ac60f-4d08-4ef1-9c62-d82f343fbc90 UUID +[1,2,3] Array(UInt64) +[[[1],[2]],[[3,4,5]]] Array(Array(Array(UInt64))) +(1,'str',42.42) Tuple(UInt32, String, Float32) +(1,'str',42.42) Tuple(a UInt32, b String, c Float32) +(1,('str',(42.42,-30))) Tuple(UInt32, Tuple(String, Tuple(Float32, Int8))) +(1,('str',(42.42,-30))) Tuple(a UInt32, b Tuple(c String, d Tuple(e Float32, f Int8))) +\0 \0\0\0\0\0\0\0\0\0\0\0\0\06364136223846793005 0 123459*\0\0\0\0\0\0\0 AggregateFunction(quantile(0.5), UInt64) +42 SimpleAggregateFunction(sum, UInt64) +Hello, World! LowCardinality(String) +{1:'str1',2:'str2'} Map(UInt64, String) +{1:{1:{1:'str1'}},2:{2:{2:'str2'}}} Map(UInt64, Map(UInt64, Map(UInt64, String))) +127.0.0.0 IPv4 +2001:db8:cafe:1::1 IPv6 +true Bool +[(1,2),(3,4)] Nested(a UInt32, b UInt32) +[(0,0),(10,0),(10,10),(0,10)] Ring +(0,0) Point +[[(20,20),(50,20),(50,50),(20,50)],[(30,30),(50,50),(50,30)]] Polygon +[[[(0,0),(10,0),(10,10),(0,10)]],[[(20,20),(50,20),(50,50),(20,50)],[(30,30),(50,50),(50,30)]]] MultiPolygon +[{42:(1,[(2,{1:2})])}] Array(Map(UInt8, Tuple(UInt8, Array(Tuple(UInt8, Map(UInt8, UInt8)))))) diff --git a/tests/queries/0_stateless/03172_dynamic_binary_serialization.sh b/tests/queries/0_stateless/03172_dynamic_binary_serialization.sh new file mode 100755 index 00000000000..9b57e5c8718 --- /dev/null +++ b/tests/queries/0_stateless/03172_dynamic_binary_serialization.sh @@ -0,0 +1,63 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT -q "drop table if exists test" +$CLICKHOUSE_CLIENT --allow_experimental_dynamic_type=1 -q "create table test (id UInt64, d Dynamic(max_types=255)) engine=Memory" + +$CLICKHOUSE_CLIENT -q "insert into test select 0, NULL" +$CLICKHOUSE_CLIENT -q "insert into test select 1, materialize(42)::UInt8" +$CLICKHOUSE_CLIENT -q "insert into test select 2, materialize(-42)::Int8" +$CLICKHOUSE_CLIENT -q "insert into test select 3, materialize(42)::UInt16" +$CLICKHOUSE_CLIENT -q "insert into test select 4, materialize(-42)::Int16" +$CLICKHOUSE_CLIENT -q "insert into test select 5, materialize(42)::UInt32" +$CLICKHOUSE_CLIENT -q "insert into test select 6, materialize(-42)::Int32" +$CLICKHOUSE_CLIENT -q "insert into test select 7, materialize(42)::UInt64" +$CLICKHOUSE_CLIENT -q "insert into test select 8, materialize(-42)::Int64" +$CLICKHOUSE_CLIENT -q "insert into test select 9, materialize(42)::UInt128" +$CLICKHOUSE_CLIENT -q "insert into test select 10, materialize(-42)::Int128" +$CLICKHOUSE_CLIENT -q "insert into test select 11, materialize(42)::UInt256" +$CLICKHOUSE_CLIENT -q "insert into test select 12, materialize(-42)::Int256" +$CLICKHOUSE_CLIENT -q "insert into test select 13, materialize(42.42)::Float32" +$CLICKHOUSE_CLIENT -q "insert into test select 14, materialize(42.42)::Float64" +$CLICKHOUSE_CLIENT -q "insert into test select 15, materialize('2020-01-01')::Date" +$CLICKHOUSE_CLIENT -q "insert into test select 16, materialize('2020-01-01')::Date32" +$CLICKHOUSE_CLIENT -q "insert into test select 17, materialize('2020-01-01 00:00:00')::DateTime('EST')" +$CLICKHOUSE_CLIENT -q "insert into test select 18, materialize('2020-01-01 00:00:00')::DateTime('CET')" +$CLICKHOUSE_CLIENT -q "insert into test select 19, materialize('2020-01-01 00:00:00.000000')::DateTime64(6, 'EST')" +$CLICKHOUSE_CLIENT -q "insert into test select 20, materialize('2020-01-01 00:00:00.000000')::DateTime64(6, 'CET')" +$CLICKHOUSE_CLIENT -q "insert into test select 21, materialize('Hello, World!')" +$CLICKHOUSE_CLIENT -q "insert into test select 22, materialize('aaaaa')::FixedString(5)" +$CLICKHOUSE_CLIENT -q "insert into test select 23, materialize('a')::Enum8('a' = 1, 'b' = 2, 'c' = -128)" +$CLICKHOUSE_CLIENT -q "insert into test select 24, materialize('a')::Enum16('a' = 1, 'b' = 2, 'c' = -1280)" +$CLICKHOUSE_CLIENT -q "insert into test select 25, materialize(42.42)::Decimal32(3)" +$CLICKHOUSE_CLIENT -q "insert into test select 26, materialize(42.42)::Decimal64(3)" +$CLICKHOUSE_CLIENT -q "insert into test select 27, materialize(42.42)::Decimal128(3)" +$CLICKHOUSE_CLIENT -q "insert into test select 28, materialize(42.42)::Decimal256(3)" +$CLICKHOUSE_CLIENT -q "insert into test select 29, materialize('984ac60f-4d08-4ef1-9c62-d82f343fbc90')::UUID" +$CLICKHOUSE_CLIENT -q "insert into test select 30, materialize([1, 2, 3])::Array(UInt64)" +$CLICKHOUSE_CLIENT -q "insert into test select 31, materialize([[[1], [2]], [[3, 4, 5]]])::Array(Array(Array(UInt64)))" +$CLICKHOUSE_CLIENT -q "insert into test select 32, materialize(tuple(1, 'str', 42.42))::Tuple(UInt32, String, Float32)" +$CLICKHOUSE_CLIENT -q "insert into test select 33, materialize(tuple(1, 'str', 42.42))::Tuple(a UInt32, b String, c Float32)" +$CLICKHOUSE_CLIENT -q "insert into test select 34, materialize(tuple(1, tuple('str', tuple(42.42, -30))))::Tuple(UInt32, Tuple(String, Tuple(Float32, Int8)))" +$CLICKHOUSE_CLIENT -q "insert into test select 35, materialize(tuple(1, tuple('str', tuple(42.42, -30))))::Tuple(a UInt32, b Tuple(c String, d Tuple(e Float32, f Int8)))" +$CLICKHOUSE_CLIENT -q "insert into test select 36, quantileState(0.5)(42::UInt64)" +$CLICKHOUSE_CLIENT -q "insert into test select 37, sumSimpleState(42::UInt64)" +$CLICKHOUSE_CLIENT -q "insert into test select 38, toLowCardinality('Hello, World!')" +$CLICKHOUSE_CLIENT -q "insert into test select 39, materialize(map(1, 'str1', 2, 'str2'))::Map(UInt64, String)" +$CLICKHOUSE_CLIENT -q "insert into test select 40, materialize(map(1, map(1, map(1, 'str1')), 2, map(2, map(2, 'str2'))))::Map(UInt64, Map(UInt64, Map(UInt64, String)))" +$CLICKHOUSE_CLIENT -q "insert into test select 41, materialize('127.0.0.0')::IPv4" +$CLICKHOUSE_CLIENT -q "insert into test select 42, materialize('2001:db8:cafe:1:0:0:0:1')::IPv6" +$CLICKHOUSE_CLIENT -q "insert into test select 43, materialize(true)::Bool" +$CLICKHOUSE_CLIENT -q "insert into test select 44, materialize([tuple(1, 2), tuple(3, 4)])::Nested(a UInt32, b UInt32)" +$CLICKHOUSE_CLIENT -q "insert into test select 45, materialize([(0, 0), (10, 0), (10, 10), (0, 10)])::Ring" +$CLICKHOUSE_CLIENT -q "insert into test select 46, materialize((0, 0))::Point" +$CLICKHOUSE_CLIENT -q "insert into test select 47, materialize([[(20, 20), (50, 20), (50, 50), (20, 50)], [(30, 30), (50, 50), (50, 30)]])::Polygon" +$CLICKHOUSE_CLIENT -q "insert into test select 48, materialize([[[(0, 0), (10, 0), (10, 10), (0, 10)]], [[(20, 20), (50, 20), (50, 50), (20, 50)],[(30, 30), (50, 50), (50, 30)]]])::MultiPolygon" +$CLICKHOUSE_CLIENT -q "insert into test select 49, materialize([map(42, tuple(1, [tuple(2, map(1, 2))]))])" + +$CLICKHOUSE_CLIENT -q "select * from test format RowBinary" | $CLICKHOUSE_LOCAL --allow_experimental_dynamic_type=1 --input-format RowBinary --structure 'id UInt64, d Dynamic(max_types=255)' -q "select d, dynamicType(d) from table order by id" +$CLICKHOUSE_CLIENT -q "drop table test" + diff --git a/tests/queries/0_stateless/03172_error_log_table_not_empty.sh b/tests/queries/0_stateless/03172_error_log_table_not_empty.sh index 8d74ebe1039..22a2fd82c64 100755 --- a/tests/queries/0_stateless/03172_error_log_table_not_empty.sh +++ b/tests/queries/0_stateless/03172_error_log_table_not_empty.sh @@ -1,9 +1,14 @@ #!/usr/bin/env bash +# Tags: no-fasttest +# Tag no-fasttest: this test relies on the timeouts, it always takes no less that 4 seconds to run CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh +# system.error_log is created lazy, flush logs query makes it sure that the table is created. +$CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS;" + # Get the previous number of errors for 111, 222 and 333 errors_111=$($CLICKHOUSE_CLIENT -q "SELECT sum(value) FROM system.error_log WHERE code = 111") errors_222=$($CLICKHOUSE_CLIENT -q "SELECT sum(value) FROM system.error_log WHERE code = 222") @@ -38,4 +43,4 @@ $CLICKHOUSE_CLIENT -mn -q " SELECT sum(value) > $(($errors_111+1)) FROM system.error_log WHERE code = 111; SELECT sum(value) > $(($errors_222+1)) FROM system.error_log WHERE code = 222; SELECT sum(value) > $(($errors_333+1)) FROM system.error_log WHERE code = 333; -" \ No newline at end of file +" diff --git a/tests/queries/0_stateless/03172_system_detached_tables.reference b/tests/queries/0_stateless/03172_system_detached_tables.reference new file mode 100644 index 00000000000..83d1ff13942 --- /dev/null +++ b/tests/queries/0_stateless/03172_system_detached_tables.reference @@ -0,0 +1,11 @@ +database atomic tests +test03172_system_detached_tables test_table 0 +test03172_system_detached_tables test_table_perm 1 +test03172_system_detached_tables test_table 0 +test03172_system_detached_tables test_table_perm 1 +test03172_system_detached_tables test_table 0 +----------------------- +database lazy tests +before attach test03172_system_detached_tables_lazy test_table 0 +before attach test03172_system_detached_tables_lazy test_table_perm 1 +DROP TABLE diff --git a/tests/queries/0_stateless/03172_system_detached_tables.sql b/tests/queries/0_stateless/03172_system_detached_tables.sql new file mode 100644 index 00000000000..1a3c2d7cc0f --- /dev/null +++ b/tests/queries/0_stateless/03172_system_detached_tables.sql @@ -0,0 +1,53 @@ +-- Tags: no-parallel + +SELECT 'database atomic tests'; +DROP DATABASE IF EXISTS test03172_system_detached_tables; +CREATE DATABASE IF NOT EXISTS test03172_system_detached_tables ENGINE=Atomic; + +CREATE TABLE test03172_system_detached_tables.test_table (n Int64) ENGINE=MergeTree ORDER BY n; +SELECT * FROM system.detached_tables WHERE database='test03172_system_detached_tables'; + +DETACH TABLE test03172_system_detached_tables.test_table; +SELECT database, table, is_permanently FROM system.detached_tables WHERE database='test03172_system_detached_tables'; + +ATTACH TABLE test03172_system_detached_tables.test_table; + +CREATE TABLE test03172_system_detached_tables.test_table_perm (n Int64) ENGINE=MergeTree ORDER BY n; +SELECT * FROM system.detached_tables WHERE database='test03172_system_detached_tables'; + +DETACH TABLE test03172_system_detached_tables.test_table_perm PERMANENTLY; +SELECT database, table, is_permanently FROM system.detached_tables WHERE database='test03172_system_detached_tables'; + +DETACH TABLE test03172_system_detached_tables.test_table SYNC; +SELECT database, table, is_permanently FROM system.detached_tables WHERE database='test03172_system_detached_tables'; + +SELECT database, table, is_permanently FROM system.detached_tables WHERE database='test03172_system_detached_tables' AND table='test_table'; + +DROP DATABASE test03172_system_detached_tables SYNC; + +SELECT '-----------------------'; +SELECT 'database lazy tests'; + +DROP DATABASE IF EXISTS test03172_system_detached_tables_lazy; +CREATE DATABASE test03172_system_detached_tables_lazy Engine=Lazy(10); + +CREATE TABLE test03172_system_detached_tables_lazy.test_table (number UInt64) engine=Log; +INSERT INTO test03172_system_detached_tables_lazy.test_table SELECT * FROM numbers(100); +DETACH TABLE test03172_system_detached_tables_lazy.test_table; + +CREATE TABLE test03172_system_detached_tables_lazy.test_table_perm (number UInt64) engine=Log; +INSERT INTO test03172_system_detached_tables_lazy.test_table_perm SELECT * FROM numbers(100); +DETACH table test03172_system_detached_tables_lazy.test_table_perm PERMANENTLY; + +SELECT 'before attach', database, table, is_permanently FROM system.detached_tables WHERE database='test03172_system_detached_tables_lazy'; + +ATTACH TABLE test03172_system_detached_tables_lazy.test_table; +ATTACH TABLE test03172_system_detached_tables_lazy.test_table_perm; + +SELECT 'after attach', database, table, is_permanently FROM system.detached_tables WHERE database='test03172_system_detached_tables_lazy'; + +SELECT 'DROP TABLE'; +DROP TABLE test03172_system_detached_tables_lazy.test_table SYNC; +DROP TABLE test03172_system_detached_tables_lazy.test_table_perm SYNC; + +DROP DATABASE test03172_system_detached_tables_lazy SYNC; diff --git a/tests/queries/0_stateless/03173_check_cyclic_dependencies_on_create_and_rename.reference b/tests/queries/0_stateless/03173_check_cyclic_dependencies_on_create_and_rename.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03173_check_cyclic_dependencies_on_create_and_rename.sql b/tests/queries/0_stateless/03173_check_cyclic_dependencies_on_create_and_rename.sql new file mode 100644 index 00000000000..0cadd4f5cee --- /dev/null +++ b/tests/queries/0_stateless/03173_check_cyclic_dependencies_on_create_and_rename.sql @@ -0,0 +1,77 @@ +-- Tags: atomic-database + +DROP TABLE IF EXISTS test; +CREATE TABLE test (id UInt64, value String) ENGINE=MergeTree ORDER BY id; +INSERT INTO test SELECT number, 'str_' || toString(number) FROM numbers(10); +DROP DICTIONARY IF EXISTS test_dict; +CREATE DICTIONARY test_dict +( + id UInt64, + value String +) +PRIMARY KEY id +SOURCE(CLICKHOUSE(TABLE test)) +LAYOUT(FLAT()) +LIFETIME(MIN 0 MAX 1000); +DROP TABLE IF EXISTS view_source; +CREATE TABLE view_source (id UInt64) ENGINE=MergeTree ORDER BY id; +INSERT INTO view_source SELECT * FROM numbers(5); +DROP VIEW IF EXISTS view; +CREATE VIEW view AS SELECT id, dictGet('test_dict', 'value', id) as value FROM view_source; + +CREATE OR REPLACE DICTIONARY test_dict +( + id UInt64, + value String +) +PRIMARY KEY id +SOURCE(CLICKHOUSE(TABLE view)) +LAYOUT(FLAT()) +LIFETIME(MIN 0 MAX 1000); -- {serverError INFINITE_LOOP} + +REPLACE DICTIONARY test_dict +( + id UInt64, + value String +) +PRIMARY KEY id +SOURCE(CLICKHOUSE(TABLE view)) +LAYOUT(FLAT()) +LIFETIME(MIN 0 MAX 1000); -- {serverError INFINITE_LOOP} + + +DROP DICTIONARY IF EXISTS test_dict_2; +CREATE DICTIONARY test_dict_2 +( + id UInt64, + value String +) +PRIMARY KEY id +SOURCE(CLICKHOUSE(TABLE view)) +LAYOUT(FLAT()) +LIFETIME(MIN 0 MAX 1000); + +EXCHANGE DICTIONARIES test_dict AND test_dict_2; -- {serverError INFINITE_LOOP} + +DROP DICTIONARY test_dict_2; + +CREATE OR REPLACE DICTIONARY test_dict_2 +( + id UInt64, + value String +) +PRIMARY KEY id +SOURCE(CLICKHOUSE(TABLE view)) +LAYOUT(FLAT()) +LIFETIME(MIN 0 MAX 1000); + +EXCHANGE DICTIONARIES test_dict AND test_dict_2; -- {serverError INFINITE_LOOP} + +DROP DICTIONARY test_dict; +RENAME DICTIONARY test_dict_2 to test_dict; -- {serverError INFINITE_LOOP} + +DROP DICTIONARY test_dict_2; +DROP VIEW view; +DROP TABLE test; +DROP TABLE view_source; + diff --git a/tests/queries/0_stateless/03173_parallel_replicas_join_bug.reference b/tests/queries/0_stateless/03173_parallel_replicas_join_bug.reference new file mode 100644 index 00000000000..93018551e1b --- /dev/null +++ b/tests/queries/0_stateless/03173_parallel_replicas_join_bug.reference @@ -0,0 +1,10 @@ +a1451105-722e-4fe7-bfaa-65ad2ae249c2 whatever +a1451105-722e-4fe7-bfaa-65ad2ae249c2 whatever +--------------------------- +a1451105-722e-4fe7-bfaa-65ad2ae249c2 +a1451105-722e-4fe7-bfaa-65ad2ae249c2 +--------------------------- +a1451105-722e-4fe7-bfaa-65ad2ae249c2 +--------------------------- +a1451105-722e-4fe7-bfaa-65ad2ae249c2 +a1451105-722e-4fe7-bfaa-65ad2ae249c2 diff --git a/tests/queries/0_stateless/03173_parallel_replicas_join_bug.sh b/tests/queries/0_stateless/03173_parallel_replicas_join_bug.sh new file mode 100755 index 00000000000..20a29e2734e --- /dev/null +++ b/tests/queries/0_stateless/03173_parallel_replicas_join_bug.sh @@ -0,0 +1,83 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +$CLICKHOUSE_CLIENT -nq " + CREATE TABLE ids (id UUID, whatever String) Engine=MergeTree ORDER BY tuple(); + INSERT INTO ids VALUES ('a1451105-722e-4fe7-bfaa-65ad2ae249c2', 'whatever'); + + CREATE TABLE data (id UUID, event_time DateTime, status String) Engine=MergeTree ORDER BY tuple(); + INSERT INTO data VALUES ('a1451105-722e-4fe7-bfaa-65ad2ae249c2', '2000-01-01', 'CREATED'); + + CREATE TABLE data2 (id UUID, event_time DateTime, status String) Engine=MergeTree ORDER BY tuple(); + INSERT INTO data2 VALUES ('a1451105-722e-4fe7-bfaa-65ad2ae249c2', '2000-01-02', 'CREATED'); +" + +$CLICKHOUSE_CLIENT -nq " +SET allow_experimental_analyzer = 1, cluster_for_parallel_replicas = 'parallel_replicas', max_parallel_replicas = 10, allow_experimental_parallel_reading_from_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, max_threads = 1; + +SELECT + id, + whatever +FROM ids AS l +INNER JOIN view( + SELECT * + FROM merge($CLICKHOUSE_DATABASE, 'data.*') +) AS s ON l.id = s.id +WHERE status IN ['CREATED', 'CREATING'] +ORDER BY event_time DESC; + +SELECT '---------------------------'; + +with +results1 as ( + SELECT id + FROM data t1 + inner join ids t2 + on t1.id = t2.id +), +results2 as ( + SELECT id + FROM ids t1 + inner join data t2 + on t1.id = t2.id +) +select * from results1 union all select * from results2; + +SELECT '---------------------------'; + +with +results1 as ( + SELECT id + FROM data t1 + inner join ids t2 + on t1.id = t2.id +), +results2 as ( + SELECT id + FROM ids t1 + inner join data t2 + on t1.id = t2.id +) +select * from results1 t1 inner join results2 t2 using (id); + +SELECT '---------------------------'; + +with +results1 as ( + SELECT t1.id + FROM data t1 + inner join ids t2 on t1.id = t2.id + left join data t3 on t2.id = t3.id +), +results2 as ( + SELECT id + FROM ids t1 + inner join data t2 + on t1.id = t2.id +) +select * from results1 union all select * from results2; +" diff --git a/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.reference b/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.reference new file mode 100644 index 00000000000..1ba147f9627 --- /dev/null +++ b/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.reference @@ -0,0 +1,114 @@ +42 UInt8 +42 UInt8 +\N Nullable(Nothing) +\N Nullable(Nothing) +42 UInt8 +42 UInt8 +-42 Int8 +-42 Int8 +42 UInt16 +42 UInt16 +-42 Int16 +-42 Int16 +42 UInt32 +42 UInt32 +-42 Int32 +-42 Int32 +42 UInt64 +42 UInt64 +-42 Int64 +-42 Int64 +42 UInt128 +42 UInt128 +-42 Int128 +-42 Int128 +42 UInt256 +42 UInt256 +-42 Int256 +-42 Int256 +42.42 Float32 +42.42 Float32 +42.42 Float64 +42.42 Float64 +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date32 +2020-01-01 Date32 +2020-01-01 00:00:00 DateTime +2020-01-01 00:00:00 DateTime +2020-01-01 00:00:00 DateTime(\'EST\') +2020-01-01 00:00:00 DateTime(\'EST\') +2020-01-01 00:00:00 DateTime(\'CET\') +2020-01-01 00:00:00 DateTime(\'CET\') +2020-01-01 00:00:00.000000 DateTime64(6) +2020-01-01 00:00:00.000000 DateTime64(6) +2020-01-01 00:00:00.000000 DateTime64(6, \'EST\') +2020-01-01 00:00:00.000000 DateTime64(6, \'EST\') +2020-01-01 00:00:00.000000 DateTime64(6, \'CET\') +2020-01-01 00:00:00.000000 DateTime64(6, \'CET\') +Hello, World! String +Hello, World! String +aaaaa FixedString(5) +aaaaa FixedString(5) +a Enum8(\'c\' = -128, \'a\' = 1, \'b\' = 2) +a Enum8(\'c\' = -128, \'a\' = 1, \'b\' = 2) +a Enum16(\'c\' = -1280, \'a\' = 1, \'b\' = 2) +a Enum16(\'c\' = -1280, \'a\' = 1, \'b\' = 2) +42.42 Decimal(9, 3) +42.42 Decimal(9, 3) +42.42 Decimal(18, 3) +42.42 Decimal(18, 3) +42.42 Decimal(38, 3) +42.42 Decimal(38, 3) +42.42 Decimal(76, 3) +42.42 Decimal(76, 3) +984ac60f-4d08-4ef1-9c62-d82f343fbc90 UUID +984ac60f-4d08-4ef1-9c62-d82f343fbc90 UUID +[1,2,3] Array(UInt64) +[1,2,3] Array(UInt64) +[[[1],[2]],[[3,4,5]]] Array(Array(Array(UInt64))) +[[[1],[2]],[[3,4,5]]] Array(Array(Array(UInt64))) +(1,'str',42.42) Tuple(UInt32, String, Float32) +(1,'str',42.42) Tuple(UInt32, String, Float32) +(1,'str',42.42) Tuple(\n a UInt32,\n b String,\n c Float32) +(1,'str',42.42) Tuple(\n a UInt32,\n b String,\n c Float32) +(1,('str',(42.42,-30))) Tuple(UInt32, Tuple(String, Tuple(Float32, Int8))) +(1,('str',(42.42,-30))) Tuple(UInt32, Tuple(String, Tuple(Float32, Int8))) +(1,('str',(42.42,-30))) Tuple(\n a UInt32,\n b Tuple(\n c String,\n d Tuple(\n e Float32,\n f Int8))) +(1,('str',(42.42,-30))) Tuple(\n a UInt32,\n b Tuple(\n c String,\n d Tuple(\n e Float32,\n f Int8))) +\0 \0\0\0\0\0\0\0\0\0\0\0\0\06364136223846793005 0 123459*\0\0\0\0\0\0\0 AggregateFunction(quantile(0.5), UInt64) +\0 \0\0\0\0\0\0\0\0\0\0\0\0\06364136223846793005 0 123459*\0\0\0\0\0\0\0 AggregateFunction(quantile(0.5), UInt64) +42 SimpleAggregateFunction(sum, UInt64) +42 SimpleAggregateFunction(sum, UInt64) +Hello, World! LowCardinality(String) +Hello, World! LowCardinality(String) +{1:'str1',2:'str2'} Map(UInt64, String) +{1:'str1',2:'str2'} Map(UInt64, String) +{1:{1:{1:'str1'}},2:{2:{2:'str2'}}} Map(UInt64, Map(UInt64, Map(UInt64, String))) +{1:{1:{1:'str1'}},2:{2:{2:'str2'}}} Map(UInt64, Map(UInt64, Map(UInt64, String))) +127.0.0.0 IPv4 +127.0.0.0 IPv4 +2001:db8:cafe:1::1 IPv6 +2001:db8:cafe:1::1 IPv6 +true Bool +true Bool +[(1,2),(3,4)] Nested(a UInt32, b UInt32) +[(1,2),(3,4)] Nested(a UInt32, b UInt32) +[(0,0),(10,0),(10,10),(0,10)] Ring +[(0,0),(10,0),(10,10),(0,10)] Ring +(0,0) Point +(0,0) Point +[[(20,20),(50,20),(50,50),(20,50)],[(30,30),(50,50),(50,30)]] Polygon +[[(20,20),(50,20),(50,50),(20,50)],[(30,30),(50,50),(50,30)]] Polygon +[[[(0,0),(10,0),(10,10),(0,10)]],[[(20,20),(50,20),(50,50),(20,50)],[(30,30),(50,50),(50,30)]]] MultiPolygon +[[[(0,0),(10,0),(10,10),(0,10)]],[[(20,20),(50,20),(50,50),(20,50)],[(30,30),(50,50),(50,30)]]] MultiPolygon +[{42:(1,[(2,{1:2})])}] Array(Map(UInt8, Tuple(UInt8, Array(Tuple(UInt8, Map(UInt8, UInt8)))))) +[{42:(1,[(2,{1:2})])}] Array(Map(UInt8, Tuple(UInt8, Array(Tuple(UInt8, Map(UInt8, UInt8)))))) +42 Variant(String, Tuple(\n a UInt32,\n b Array(Map(String, String))), UInt32) +42 Variant(String, Tuple(\n a UInt32,\n b Array(Map(String, String))), UInt32) +[{42:(1,[(2,{1:2})])}] Dynamic +[{42:(1,[(2,{1:2})])}] Dynamic +[{42:(1,[(2,{1:2})])}] Dynamic(max_types=10) +[{42:(1,[(2,{1:2})])}] Dynamic(max_types=10) +[{42:(1,[(2,{1:2})])}] Dynamic(max_types=255) +[{42:(1,[(2,{1:2})])}] Dynamic(max_types=255) diff --git a/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.sh b/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.sh new file mode 100755 index 00000000000..723b11ad620 --- /dev/null +++ b/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.sh @@ -0,0 +1,69 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +function test +{ + $CLICKHOUSE_LOCAL --allow_experimental_dynamic_type=1 --allow_experimental_variant_type=1 --output_format_binary_encode_types_in_binary_format=1 -q "select $1 as value format RowBinaryWithNamesAndTypes" | $CLICKHOUSE_LOCAL --input-format RowBinaryWithNamesAndTypes --input_format_binary_decode_types_in_binary_format=1 -q "select value, toTypeName(value) from table" + $CLICKHOUSE_LOCAL --allow_experimental_dynamic_type=1 --allow_experimental_variant_type=1 --output_format_native_encode_types_in_binary_format=1 -q "select $1 as value format Native" | $CLICKHOUSE_LOCAL --input-format Native --input_format_native_decode_types_in_binary_format=1 -q "select value, toTypeName(value) from table" +} + +test "materialize(42)::UInt8" +test "NULL" +test "materialize(42)::UInt8" +test "materialize(-42)::Int8" +test "materialize(42)::UInt16" +test "materialize(-42)::Int16" +test "materialize(42)::UInt32" +test "materialize(-42)::Int32" +test "materialize(42)::UInt64" +test "materialize(-42)::Int64" +test "materialize(42)::UInt128" +test "materialize(-42)::Int128" +test "materialize(42)::UInt256" +test "materialize(-42)::Int256" +test "materialize(42.42)::Float32" +test "materialize(42.42)::Float64" +test "materialize('2020-01-01')::Date" +test "materialize('2020-01-01')::Date32" +test "materialize('2020-01-01 00:00:00')::DateTime" +test "materialize('2020-01-01 00:00:00')::DateTime('EST')" +test "materialize('2020-01-01 00:00:00')::DateTime('CET')" +test "materialize('2020-01-01 00:00:00.000000')::DateTime64(6)" +test "materialize('2020-01-01 00:00:00.000000')::DateTime64(6, 'EST')" +test "materialize('2020-01-01 00:00:00.000000')::DateTime64(6, 'CET')" +test "materialize('Hello, World!')" +test "materialize('aaaaa')::FixedString(5)" +test "materialize('a')::Enum8('a' = 1, 'b' = 2, 'c' = -128)" +test "materialize('a')::Enum16('a' = 1, 'b' = 2, 'c' = -1280)" +test "materialize(42.42)::Decimal32(3)" +test "materialize(42.42)::Decimal64(3)" +test "materialize(42.42)::Decimal128(3)" +test "materialize(42.42)::Decimal256(3)" +test "materialize('984ac60f-4d08-4ef1-9c62-d82f343fbc90')::UUID" +test "materialize([1, 2, 3])::Array(UInt64)" +test "materialize([[[1], [2]], [[3, 4, 5]]])::Array(Array(Array(UInt64)))" +test "materialize(tuple(1, 'str', 42.42))::Tuple(UInt32, String, Float32)" +test "materialize(tuple(1, 'str', 42.42))::Tuple(a UInt32, b String, c Float32)" +test "materialize(tuple(1, tuple('str', tuple(42.42, -30))))::Tuple(UInt32, Tuple(String, Tuple(Float32, Int8)))" +test "materialize(tuple(1, tuple('str', tuple(42.42, -30))))::Tuple(a UInt32, b Tuple(c String, d Tuple(e Float32, f Int8)))" +test "quantileState(0.5)(42::UInt64)" +test "sumSimpleState(42::UInt64)" +test "toLowCardinality('Hello, World!')" +test "materialize(map(1, 'str1', 2, 'str2'))::Map(UInt64, String)" +test "materialize(map(1, map(1, map(1, 'str1')), 2, map(2, map(2, 'str2'))))::Map(UInt64, Map(UInt64, Map(UInt64, String)))" +test "materialize('127.0.0.0')::IPv4" +test "materialize('2001:db8:cafe:1:0:0:0:1')::IPv6" +test "materialize(true)::Bool" +test "materialize([tuple(1, 2), tuple(3, 4)])::Nested(a UInt32, b UInt32)" +test "materialize([(0, 0), (10, 0), (10, 10), (0, 10)])::Ring" +test "materialize((0, 0))::Point" +test "materialize([[(20, 20), (50, 20), (50, 50), (20, 50)], [(30, 30), (50, 50), (50, 30)]])::Polygon" +test "materialize([[[(0, 0), (10, 0), (10, 10), (0, 10)]], [[(20, 20), (50, 20), (50, 50), (20, 50)],[(30, 30), (50, 50), (50, 30)]]])::MultiPolygon" +test "materialize([map(42, tuple(1, [tuple(2, map(1, 2))]))])" +test "materialize(42::UInt32)::Variant(UInt32, String, Tuple(a UInt32, b Array(Map(String, String))))" +test "materialize([map(42, tuple(1, [tuple(2, map(1, 2))]))])::Dynamic" +test "materialize([map(42, tuple(1, [tuple(2, map(1, 2))]))])::Dynamic(max_types=10)" +test "materialize([map(42, tuple(1, [tuple(2, map(1, 2))]))])::Dynamic(max_types=255)" diff --git a/tests/queries/0_stateless/03173_set_transformed_partition_pruning.reference b/tests/queries/0_stateless/03173_set_transformed_partition_pruning.reference new file mode 100644 index 00000000000..3a6727b70e8 --- /dev/null +++ b/tests/queries/0_stateless/03173_set_transformed_partition_pruning.reference @@ -0,0 +1,50 @@ +-- Single partition by function +0 +2 +-- Nested partition by function +1 +2 +1 +1 +-- Nested partition by function, LowCardinality +1 +2 +1 +1 +-- Nested partition by function, Nullable +1 +2 +1 +1 +-- Nested partition by function, LowCardinality + Nullable +1 +2 +1 +1 +-- Non-safe cast +2 +2 +-- Multiple partition columns +1 +1 +1 +2 +-- LowCardinality set +1 +1 +-- Nullable set +1 +1 +-- LowCardinality + Nullable set +1 +1 +-- Not failing with date parsing functions +1 +0 +-- Pruning + not failing with nested date parsing functions +1 +2 +0 +-- Empty transform functions +2 +1 diff --git a/tests/queries/0_stateless/03173_set_transformed_partition_pruning.sql b/tests/queries/0_stateless/03173_set_transformed_partition_pruning.sql new file mode 100644 index 00000000000..8ffabacaa8c --- /dev/null +++ b/tests/queries/0_stateless/03173_set_transformed_partition_pruning.sql @@ -0,0 +1,258 @@ +SELECT '-- Single partition by function'; + +DROP TABLE IF EXISTS 03173_single_function; +CREATE TABLE 03173_single_function ( + dt Date, +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY toMonth(dt); + +INSERT INTO 03173_single_function +SELECT toDate('2000-01-01') + 10 * number FROM numbers(50) +UNION ALL +SELECT toDate('2100-01-01') + 10 * number FROM numbers(50); +OPTIMIZE TABLE 03173_single_function FINAL; + +SELECT count() FROM 03173_single_function WHERE dt IN ('2024-01-20', '2024-05-25') SETTINGS log_comment='03173_single_function'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_single_function'; + +DROP TABLE IF EXISTS 03173_single_function; + +SELECT '-- Nested partition by function'; + +DROP TABLE IF EXISTS 03173_nested_function; +CREATE TABLE 03173_nested_function( + id Int32, +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY xxHash32(id) % 3; + +INSERT INTO 03173_nested_function SELECT number FROM numbers(100); +OPTIMIZE TABLE 03173_nested_function FINAL; + +SELECT count() FROM 03173_nested_function WHERE id IN (10) SETTINGS log_comment='03173_nested_function'; +SELECT count() FROM 03173_nested_function WHERE xxHash32(id) IN (2158931063, 1449383981) SETTINGS log_comment='03173_nested_function_subexpr'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_function'; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_function_subexpr'; + +DROP TABLE IF EXISTS 03173_nested_function; + +SELECT '-- Nested partition by function, LowCardinality'; + +SET allow_suspicious_low_cardinality_types = 1; + +DROP TABLE IF EXISTS 03173_nested_function_lc; +CREATE TABLE 03173_nested_function_lc( + id LowCardinality(Int32), +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY xxHash32(id) % 3; + +INSERT INTO 03173_nested_function_lc SELECT number FROM numbers(100); +OPTIMIZE TABLE 03173_nested_function_lc FINAL; + +SELECT count() FROM 03173_nested_function_lc WHERE id IN (10) SETTINGS log_comment='03173_nested_function_lc'; +SELECT count() FROM 03173_nested_function_lc WHERE xxHash32(id) IN (2158931063, 1449383981) SETTINGS log_comment='03173_nested_function_subexpr_lc'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_function_lc'; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_function_subexpr_lc'; + +DROP TABLE IF EXISTS 03173_nested_function_lc; + +SELECT '-- Nested partition by function, Nullable'; + +DROP TABLE IF EXISTS 03173_nested_function_null; +CREATE TABLE 03173_nested_function_null( + id Nullable(Int32), +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY xxHash32(id) % 3 +SETTINGS allow_nullable_key=1; + +INSERT INTO 03173_nested_function_null SELECT number FROM numbers(100); +OPTIMIZE TABLE 03173_nested_function_null FINAL; + +SELECT count() FROM 03173_nested_function_null WHERE id IN (10) SETTINGS log_comment='03173_nested_function_null'; +SELECT count() FROM 03173_nested_function_null WHERE xxHash32(id) IN (2158931063, 1449383981) SETTINGS log_comment='03173_nested_function_subexpr_null'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_function_null'; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_function_subexpr_null'; + +DROP TABLE IF EXISTS 03173_nested_function_null; + +SELECT '-- Nested partition by function, LowCardinality + Nullable'; + +DROP TABLE IF EXISTS 03173_nested_function_lc_null; + +SET allow_suspicious_low_cardinality_types = 1; +CREATE TABLE 03173_nested_function_lc_null( + id LowCardinality(Nullable(Int32)), +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY xxHash32(id) % 3 +SETTINGS allow_nullable_key=1; + +INSERT INTO 03173_nested_function_lc_null SELECT number FROM numbers(100); +OPTIMIZE TABLE 03173_nested_function_lc_null FINAL; + +SELECT count() FROM 03173_nested_function_lc_null WHERE id IN (10) SETTINGS log_comment='03173_nested_function_lc_null'; +SELECT count() FROM 03173_nested_function_lc_null WHERE xxHash32(id) IN (2158931063, 1449383981) SETTINGS log_comment='03173_nested_function_subexpr_lc_null'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_function_lc_null'; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_function_subexpr_lc_null'; + +DROP TABLE IF EXISTS 03173_nested_function_lc_null; + +SELECT '-- Non-safe cast'; + +DROP TABLE IF EXISTS 03173_nonsafe_cast; +CREATE TABLE 03173_nonsafe_cast( + id Int64, +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY xxHash32(id) % 3; + +INSERT INTO 03173_nonsafe_cast SELECT number FROM numbers(100); +OPTIMIZE TABLE 03173_nonsafe_cast FINAL; + +SELECT count() FROM 03173_nonsafe_cast WHERE id IN (SELECT '50' UNION ALL SELECT '99') SETTINGS log_comment='03173_nonsafe_cast'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nonsafe_cast'; + +DROP TABLE IF EXISTS 03173_nonsafe_cast; + +SELECT '-- Multiple partition columns'; + +DROP TABLE IF EXISTS 03173_multiple_partition_cols; +CREATE TABLE 03173_multiple_partition_cols ( + key1 Int32, + key2 Int32 +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY (intDiv(key1, 50), xxHash32(key2) % 3); + +INSERT INTO 03173_multiple_partition_cols SELECT number, number FROM numbers(100); +OPTIMIZE TABLE 03173_multiple_partition_cols FINAL; + +SELECT count() FROM 03173_multiple_partition_cols WHERE key2 IN (4) SETTINGS log_comment='03173_multiple_columns'; +SELECT count() FROM 03173_multiple_partition_cols WHERE xxHash32(key2) IN (4251411170) SETTINGS log_comment='03173_multiple_columns_subexpr'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_multiple_columns'; +-- Due to xxHash32() in WHERE condition, MinMax is unable to eliminate any parts, +-- so partition pruning leave two parts (for key1 // 50 = 0 and key1 // 50 = 1) +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_multiple_columns_subexpr'; + +-- Preparing base table for filtering by LowCardinality/Nullable sets +DROP TABLE IF EXISTS 03173_base_data_source; +CREATE TABLE 03173_base_data_source( + id Int32, +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY xxHash32(id) % 3; + +INSERT INTO 03173_base_data_source SELECT number FROM numbers(100); +OPTIMIZE TABLE 03173_base_data_source FINAL; + +SELECT '-- LowCardinality set'; + +SET allow_suspicious_low_cardinality_types = 1; +DROP TABLE IF EXISTS 03173_low_cardinality_set; +CREATE TABLE 03173_low_cardinality_set (id LowCardinality(Int32)) ENGINE=Memory AS SELECT 10; + +SELECT count() FROM 03173_base_data_source WHERE id IN (SELECT id FROM 03173_low_cardinality_set) SETTINGS log_comment='03173_low_cardinality_set'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_low_cardinality_set'; + +DROP TABLE IF EXISTS 03173_low_cardinality_set; + +SELECT '-- Nullable set'; + +DROP TABLE IF EXISTS 03173_nullable_set; +CREATE TABLE 03173_nullable_set (id Nullable(Int32)) ENGINE=Memory AS SELECT 10; + +SELECT count() FROM 03173_base_data_source WHERE id IN (SELECT id FROM 03173_nullable_set) SETTINGS log_comment='03173_nullable_set'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nullable_set'; + +DROP TABLE IF EXISTS 03173_nullable_set; + +SELECT '-- LowCardinality + Nullable set'; + +DROP TABLE IF EXISTS 03173_lc_nullable_set; +CREATE TABLE 03173_lc_nullable_set (id LowCardinality(Nullable(Int32))) ENGINE=Memory AS SELECT 10 UNION ALL SELECT NULL; + +SELECT count() FROM 03173_base_data_source WHERE id IN (SELECT id FROM 03173_lc_nullable_set) SETTINGS log_comment='03173_lc_nullable_set'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_lc_nullable_set'; + +DROP TABLE IF EXISTS 03173_lc_nullable_set; + +SELECT '-- Not failing with date parsing functions'; + +DROP TABLE IF EXISTS 03173_date_parsing; +CREATE TABLE 03173_date_parsing ( + id String +) +ENGINE=MergeTree +ORDER BY tuple() +PARTITION BY toDate(id); + +INSERT INTO 03173_date_parsing +SELECT toString(toDate('2023-04-01') + number) +FROM numbers(20); + +SELECT count() FROM 03173_date_parsing WHERE id IN ('2023-04-02', '2023-05-02'); +SELECT count() FROM 03173_date_parsing WHERE id IN ('not a date'); + +DROP TABLE IF EXISTS 03173_date_parsing; + +SELECT '-- Pruning + not failing with nested date parsing functions'; + +DROP TABLE IF EXISTS 03173_nested_date_parsing; +CREATE TABLE 03173_nested_date_parsing ( + id String +) +ENGINE=MergeTree +ORDER BY tuple() +PARTITION BY toMonth(toDate(id)); + +INSERT INTO 03173_nested_date_parsing +SELECT toString(toDate('2000-01-01') + 10 * number) FROM numbers(50) +UNION ALL +SELECT toString(toDate('2100-01-01') + 10 * number) FROM numbers(50); + +SELECT count() FROM 03173_nested_date_parsing WHERE id IN ('2000-01-21', '2023-05-02') SETTINGS log_comment='03173_nested_date_parsing'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_nested_date_parsing'; +SELECT count() FROM 03173_nested_date_parsing WHERE id IN ('not a date'); + +DROP TABLE IF EXISTS 03173_nested_date_parsing; + +SELECT '-- Empty transform functions'; + +DROP TABLE IF EXISTS 03173_empty_transform; +CREATE TABLE 03173_empty_transform( + id Int32, +) +ENGINE = MergeTree +ORDER BY tuple() +PARTITION BY xxHash32(id) % 3; + +INSERT INTO 03173_empty_transform SELECT number FROM numbers(6); +OPTIMIZE TABLE 03173_empty_transform FINAL; + +SELECT id FROM 03173_empty_transform WHERE xxHash32(id) % 3 IN (xxHash32(2::Int32) % 3) SETTINGS log_comment='03173_empty_transform'; +SYSTEM FLUSH LOGS; +SELECT ProfileEvents['SelectedParts'] FROM system.query_log WHERE type = 'QueryFinish' AND current_database = currentDatabase() AND log_comment = '03173_empty_transform'; + +DROP TABLE IF EXISTS 03173_empty_transform; diff --git a/tests/queries/0_stateless/03174_merge_join_bug.reference b/tests/queries/0_stateless/03174_merge_join_bug.reference new file mode 100644 index 00000000000..af98bcd6397 --- /dev/null +++ b/tests/queries/0_stateless/03174_merge_join_bug.reference @@ -0,0 +1,10 @@ +0 0 +1 1 +2 2 +3 3 +4 4 +5 5 +6 6 +7 7 +8 8 +9 9 diff --git a/tests/queries/0_stateless/03174_merge_join_bug.sql b/tests/queries/0_stateless/03174_merge_join_bug.sql new file mode 100644 index 00000000000..ab4cb6cd4a9 --- /dev/null +++ b/tests/queries/0_stateless/03174_merge_join_bug.sql @@ -0,0 +1,10 @@ +-- Tags: no-random-settings + +SET allow_experimental_analyzer=1, join_algorithm = 'full_sorting_merge'; +CREATE TABLE xxxx_yyy (key UInt32, key_b ALIAS key) ENGINE=MergeTree() ORDER BY key SETTINGS ratio_of_defaults_for_sparse_serialization=0.0; +INSERT INTO xxxx_yyy SELECT number FROM numbers(10); + +SELECT * +FROM xxxx_yyy AS a +INNER JOIN xxxx_yyy AS b ON a.key = b.key_b +ORDER BY a.key; diff --git a/tests/queries/0_stateless/03174_projection_deduplicate.reference b/tests/queries/0_stateless/03174_projection_deduplicate.reference new file mode 100644 index 00000000000..1796b2f1dee --- /dev/null +++ b/tests/queries/0_stateless/03174_projection_deduplicate.reference @@ -0,0 +1,3 @@ +1 one +1 one +1 one diff --git a/tests/queries/0_stateless/03174_projection_deduplicate.sql b/tests/queries/0_stateless/03174_projection_deduplicate.sql new file mode 100644 index 00000000000..46222b69dc7 --- /dev/null +++ b/tests/queries/0_stateless/03174_projection_deduplicate.sql @@ -0,0 +1,30 @@ +-- https://github.com/ClickHouse/ClickHouse/issues/65548 +DROP TABLE IF EXISTS test_projection_deduplicate; + +CREATE TABLE test_projection_deduplicate +( + `id` Int32, + `string` String, + PROJECTION test_projection + ( + SELECT id + GROUP BY id + ) +) +ENGINE = MergeTree +PRIMARY KEY id; + +INSERT INTO test_projection_deduplicate VALUES (1, 'one'); +INSERT INTO test_projection_deduplicate VALUES (1, 'one'); + +OPTIMIZE TABLE test_projection_deduplicate DEDUPLICATE; -- { serverError NOT_IMPLEMENTED } + +SELECT * FROM test_projection_deduplicate; + +ALTER TABLE test_projection_deduplicate DROP PROJECTION test_projection; + +OPTIMIZE TABLE test_projection_deduplicate DEDUPLICATE; + +SELECT * FROM test_projection_deduplicate; + +DROP TABLE test_projection_deduplicate; diff --git a/tests/queries/0_stateless/03196_local_memory_limit.reference b/tests/queries/0_stateless/03196_local_memory_limit.reference new file mode 100644 index 00000000000..f2e22e8aa5b --- /dev/null +++ b/tests/queries/0_stateless/03196_local_memory_limit.reference @@ -0,0 +1 @@ +maximum: 95.37 MiB diff --git a/tests/queries/0_stateless/03196_local_memory_limit.sh b/tests/queries/0_stateless/03196_local_memory_limit.sh new file mode 100755 index 00000000000..346b37be006 --- /dev/null +++ b/tests/queries/0_stateless/03196_local_memory_limit.sh @@ -0,0 +1,7 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +${CLICKHOUSE_LOCAL} --config-file <(echo "100M") --query "SELECT number FROM system.numbers GROUP BY number HAVING count() > 1" 2>&1 | grep -o -P 'maximum: [\d\.]+ MiB' diff --git a/tests/queries/0_stateless/03197_fix_parse_mysql_iso_date.reference b/tests/queries/0_stateless/03197_fix_parse_mysql_iso_date.reference new file mode 100644 index 00000000000..bd9ab3be3fa --- /dev/null +++ b/tests/queries/0_stateless/03197_fix_parse_mysql_iso_date.reference @@ -0,0 +1,2 @@ +2024-06-20 00:00:00 +2024-06-20 00:00:00 diff --git a/tests/queries/0_stateless/03197_fix_parse_mysql_iso_date.sql b/tests/queries/0_stateless/03197_fix_parse_mysql_iso_date.sql new file mode 100644 index 00000000000..e83738f7214 --- /dev/null +++ b/tests/queries/0_stateless/03197_fix_parse_mysql_iso_date.sql @@ -0,0 +1,2 @@ +SELECT parseDateTime('2024-06-20', '%F', 'UTC') AS x; +SELECT parseDateTime('06/20/24', '%D', 'UTC') AS x; diff --git a/tests/queries/0_stateless/03197_storage_join_strictness_type_restriction.reference b/tests/queries/0_stateless/03197_storage_join_strictness_type_restriction.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03197_storage_join_strictness_type_restriction.sql b/tests/queries/0_stateless/03197_storage_join_strictness_type_restriction.sql new file mode 100644 index 00000000000..5aa3e4c2e0c --- /dev/null +++ b/tests/queries/0_stateless/03197_storage_join_strictness_type_restriction.sql @@ -0,0 +1,42 @@ +DROP TABLE IF EXISTS t1; +CREATE TABLE t1 +( + a Int64, + b Int64 +) Engine = Join(SEMI, ALL, a); -- { serverError BAD_ARGUMENTS } + +CREATE TABLE t1 +( + a Int64, + b Int64 +) Engine = Join(SEMI, INNER, a); -- { serverError BAD_ARGUMENTS } + +CREATE TABLE t1 +( + a Int64, + b Int64 +) Engine = Join(SEMI, FULL, a); -- { serverError BAD_ARGUMENTS } + +CREATE TABLE t1 +( + a Int64, + b Int64 +) Engine = Join(ANTI, ALL, a); -- { serverError BAD_ARGUMENTS } + +CREATE TABLE t1 +( + a Int64, + b Int64 +) Engine = Join(ANTI, INNER, a); -- { serverError BAD_ARGUMENTS } + +CREATE TABLE t1 +( + a Int64, + b Int64 +) Engine = Join(ANTI, FULL, a); -- { serverError BAD_ARGUMENTS } + +CREATE TABLE t1 +( + a Int64, + b Int64 +) Engine = Join(ANY, FULL, a); -- { serverError NOT_IMPLEMENTED } diff --git a/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.reference b/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.reference new file mode 100644 index 00000000000..33b8cd6ee26 --- /dev/null +++ b/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.reference @@ -0,0 +1,3 @@ +-- bitShiftRight +-- bitShiftLeft +OK diff --git a/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.sql b/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.sql new file mode 100644 index 00000000000..aec01753673 --- /dev/null +++ b/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.sql @@ -0,0 +1,17 @@ +SELECT '-- bitShiftRight'; +SELECT bitShiftRight(1, -1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftRight(toUInt8(1), 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftRight('hola', -1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftRight('hola', 4 * 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftRight(toFixedString('hola', 8), -1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftRight(toFixedString('hola', 8), 8 * 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } + +SELECT '-- bitShiftLeft'; +SELECT bitShiftLeft(1, -1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftLeft(toUInt8(1), 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftLeft('hola', -1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftLeft('hola', 4 * 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftLeft(toFixedString('hola', 8), -1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftLeft(toFixedString('hola', 8), 8 * 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } + +SELECT 'OK'; \ No newline at end of file diff --git a/tests/queries/0_stateless/03198_dictionary_validate_primary_key_type.reference b/tests/queries/0_stateless/03198_dictionary_validate_primary_key_type.reference new file mode 100644 index 00000000000..c7f2781aed9 --- /dev/null +++ b/tests/queries/0_stateless/03198_dictionary_validate_primary_key_type.reference @@ -0,0 +1,4 @@ +n1 UInt64 +n2 UInt32 +n1 UInt64 +n2 UInt32 diff --git a/tests/queries/0_stateless/03198_dictionary_validate_primary_key_type.sql b/tests/queries/0_stateless/03198_dictionary_validate_primary_key_type.sql new file mode 100644 index 00000000000..e7b8730abbd --- /dev/null +++ b/tests/queries/0_stateless/03198_dictionary_validate_primary_key_type.sql @@ -0,0 +1,41 @@ +CREATE DICTIONARY `test_dictionary0` ( + `n1` String, + `n2` UInt32 +) +PRIMARY KEY n1 +SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 DB 'test_db' TABLE 'table_01' USER 'default')) +LIFETIME(MIN 1 MAX 10) +LAYOUT(FLAT()); + +SET dictionary_validate_primary_key_type=1; + +CREATE DICTIONARY `test_dictionary1` ( + `n1` String, + `n2` UInt32 +) +PRIMARY KEY n1 +SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 DB 'test_db' TABLE 'table_01' USER 'default')) +LIFETIME(MIN 1 MAX 10) +LAYOUT(FLAT()); -- { serverError 36 } + +CREATE DICTIONARY `test_dictionary2` ( + `n1` UInt32, + `n2` UInt32 +) +PRIMARY KEY n1 +SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 DB 'test_db' TABLE 'table_01' USER 'default')) +LIFETIME(MIN 1 MAX 10) +LAYOUT(FLAT()); -- { serverError 36 } + +CREATE DICTIONARY `test_dictionary3` ( + `n1` UInt64, + `n2` UInt32 +) +PRIMARY KEY n1 +SOURCE(CLICKHOUSE(HOST 'localhost' PORT 9000 DB 'test_db' TABLE 'table_01' USER 'default')) +LIFETIME(MIN 1 MAX 10) +LAYOUT(FLAT()); + +DESCRIBE `test_dictionary0`; +DESCRIBE `test_dictionary3`; + diff --git a/tests/queries/0_stateless/03198_dynamic_read_subcolumns.reference b/tests/queries/0_stateless/03198_dynamic_read_subcolumns.reference new file mode 100644 index 00000000000..f9fe0c2ad77 --- /dev/null +++ b/tests/queries/0_stateless/03198_dynamic_read_subcolumns.reference @@ -0,0 +1,17 @@ +QUERY id: 0 + PROJECTION COLUMNS + d.String Nullable(String) + PROJECTION + LIST id: 1, nodes: 1 + COLUMN id: 2, column_name: d.String, result_type: Nullable(String), source_id: 3 + JOIN TREE + TABLE id: 3, alias: __table1, table_name: default.test_dynamic + SETTINGS allow_experimental_analyzer=1 +foo +\N +\N +foo +\N +\N +6 +6 diff --git a/tests/queries/0_stateless/03198_dynamic_read_subcolumns.sql b/tests/queries/0_stateless/03198_dynamic_read_subcolumns.sql new file mode 100644 index 00000000000..bbe701f022b --- /dev/null +++ b/tests/queries/0_stateless/03198_dynamic_read_subcolumns.sql @@ -0,0 +1,22 @@ +-- Tags: no-random-settings, no-object-storage, no-parallel +-- Tag no-object-storage: this test relies on the number of opened files in MergeTree that can differ in object storages + +SET allow_experimental_dynamic_type = 1; +DROP TABLE IF EXISTS test_dynamic; +CREATE TABLE test_dynamic (id UInt64, d Dynamic) ENGINE = MergeTree ORDER BY id SETTINGS min_bytes_for_wide_part = 0; +INSERT INTO test_dynamic VALUES (1, 'foo'), (2, 1111), (3, [1, 2, 3]); +EXPLAIN QUERY TREE SELECT d.String FROM test_dynamic SETTINGS allow_experimental_analyzer = 1; +SYSTEM DROP MARK CACHE; +SELECT d.String FROM test_dynamic SETTINGS allow_experimental_analyzer = 1; +SYSTEM DROP MARK CACHE; +SELECT d.String FROM test_dynamic SETTINGS allow_experimental_analyzer = 0; +SYSTEM FLUSH LOGS; +SELECT + ProfileEvents['FileOpen'] +FROM system.query_log +WHERE (type = 2) AND (query LIKE 'SELECT d.String %test_dynamic%') AND (current_database = currentDatabase()) +ORDER BY event_time_microseconds DESC +LIMIT 2; + +DROP TABLE test_dynamic; + diff --git a/tests/queries/0_stateless/03198_group_array_intersect.reference b/tests/queries/0_stateless/03198_group_array_intersect.reference new file mode 100644 index 00000000000..6a3ffdba388 --- /dev/null +++ b/tests/queries/0_stateless/03198_group_array_intersect.reference @@ -0,0 +1,4 @@ +['0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','3\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','4\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','5\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0'] +['1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','2\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','3\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','4\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','5\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','6\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0'] +['1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','2\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','4\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','5\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0'] +['1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','4\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0','5\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0'] diff --git a/tests/queries/0_stateless/03198_group_array_intersect.sql b/tests/queries/0_stateless/03198_group_array_intersect.sql new file mode 100644 index 00000000000..ab7ba0b0d70 --- /dev/null +++ b/tests/queries/0_stateless/03198_group_array_intersect.sql @@ -0,0 +1,11 @@ +DROP TABLE IF EXISTS test_numbers__fuzz_29; +SET max_threads=1, max_insert_threads=1; +CREATE TABLE test_numbers__fuzz_29 (`a` Array(Nullable(FixedString(19)))) ENGINE = MergeTree ORDER BY a SETTINGS allow_nullable_key=1; + +INSERT INTO test_numbers__fuzz_29 VALUES ([1,2,3,4,5,6]); +INSERT INTO test_numbers__fuzz_29 VALUES ([1,2,4,5]); +INSERT INTO test_numbers__fuzz_29 VALUES ([1,4,3,0,5,5,5]); + +SELECT arraySort(groupArrayIntersect(*)) FROM test_numbers__fuzz_29 GROUP BY a WITH ROLLUP ORDER BY ALL; + +DROP TABLE test_numbers__fuzz_29; diff --git a/tests/queries/0_stateless/03198_json_extract_more_types.reference b/tests/queries/0_stateless/03198_json_extract_more_types.reference new file mode 100644 index 00000000000..9a6580ff81b --- /dev/null +++ b/tests/queries/0_stateless/03198_json_extract_more_types.reference @@ -0,0 +1,21 @@ +2020-01-01 +2020-01-01 +2020-01-01 00:00:00 +2020-01-01 00:00:00.000000 +127.0.0.1 +2001:db8:85a3::8a2e:370:7334 +42 +42 +42 +42 +42 +42 +42 +42 +42 +42 +Hello +Hello +\0\0\0 +Hello\0\0\0\0\0 +5801c962-1182-458a-89f8-d077da5074f9 diff --git a/tests/queries/0_stateless/03198_json_extract_more_types.sql b/tests/queries/0_stateless/03198_json_extract_more_types.sql new file mode 100644 index 00000000000..28d24bbb271 --- /dev/null +++ b/tests/queries/0_stateless/03198_json_extract_more_types.sql @@ -0,0 +1,29 @@ +set allow_suspicious_low_cardinality_types=1; + +select JSONExtract('{"a" : "2020-01-01"}', 'a', 'Date'); +select JSONExtract('{"a" : "2020-01-01"}', 'a', 'Date32'); +select JSONExtract('{"a" : "2020-01-01 00:00:00"}', 'a', 'DateTime'); +select JSONExtract('{"a" : "2020-01-01 00:00:00.000000"}', 'a', 'DateTime64(6)'); +select JSONExtract('{"a" : "127.0.0.1"}', 'a', 'IPv4'); +select JSONExtract('{"a" : "2001:0db8:85a3:0000:0000:8a2e:0370:7334"}', 'a', 'IPv6'); + + +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(UInt8)'); +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(Int8)'); +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(UInt16)'); +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(Int16)'); +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(UInt32)'); +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(Int32)'); +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(UInt64)'); +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(Int64)'); + +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(Float32)'); +select JSONExtract('{"a" : 42}', 'a', 'LowCardinality(Float32)'); + +select JSONExtract('{"a" : "Hello"}', 'a', 'LowCardinality(String)'); +select JSONExtract('{"a" : "Hello"}', 'a', 'LowCardinality(FixedString(5))'); +select JSONExtract('{"a" : "Hello"}', 'a', 'LowCardinality(FixedString(3))'); +select JSONExtract('{"a" : "Hello"}', 'a', 'LowCardinality(FixedString(10))'); + +select JSONExtract('{"a" : "5801c962-1182-458a-89f8-d077da5074f9"}', 'a', 'LowCardinality(UUID)'); + diff --git a/tests/queries/0_stateless/03198_non_adaptive_granularity_no_errors.reference b/tests/queries/0_stateless/03198_non_adaptive_granularity_no_errors.reference new file mode 100644 index 00000000000..fcd78da1283 --- /dev/null +++ b/tests/queries/0_stateless/03198_non_adaptive_granularity_no_errors.reference @@ -0,0 +1,2 @@ +1000000 +1000000 diff --git a/tests/queries/0_stateless/03198_non_adaptive_granularity_no_errors.sql b/tests/queries/0_stateless/03198_non_adaptive_granularity_no_errors.sql new file mode 100644 index 00000000000..25798ef6d33 --- /dev/null +++ b/tests/queries/0_stateless/03198_non_adaptive_granularity_no_errors.sql @@ -0,0 +1,12 @@ +DROP TABLE IF EXISTS data_02051__fuzz_24; + +CREATE TABLE data_02051__fuzz_24 (`key` Int16, `value` String) ENGINE = MergeTree ORDER BY key SETTINGS index_granularity_bytes = 0, min_rows_for_wide_part = 0, min_bytes_for_wide_part=0 AS SELECT number, repeat(toString(number), 5) FROM numbers(1000000.); + +SELECT count(ignore(*)) FROM data_02051__fuzz_24 PREWHERE materialize(1) GROUP BY ignore(*); + +detach table data_02051__fuzz_24; +attach table data_02051__fuzz_24; + +SELECT count(ignore(*)) FROM data_02051__fuzz_24 PREWHERE materialize(1) GROUP BY ignore(*); + +DROP TABLE data_02051__fuzz_24; diff --git a/tests/queries/0_stateless/03198_orc_read_time_zone.reference b/tests/queries/0_stateless/03198_orc_read_time_zone.reference new file mode 100644 index 00000000000..809dba44400 --- /dev/null +++ b/tests/queries/0_stateless/03198_orc_read_time_zone.reference @@ -0,0 +1 @@ +1 2024-06-30 20:00:00.000 diff --git a/tests/queries/0_stateless/03198_orc_read_time_zone.sh b/tests/queries/0_stateless/03198_orc_read_time_zone.sh new file mode 100755 index 00000000000..27530c06237 --- /dev/null +++ b/tests/queries/0_stateless/03198_orc_read_time_zone.sh @@ -0,0 +1,12 @@ +#!/usr/bin/env bash +# Tags: no-fasttest + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT -q "drop table if exists test" +$CLICKHOUSE_CLIENT -q "create table test(id UInt64, t DateTime64) Engine=MergeTree order by id" +$CLICKHOUSE_CLIENT -q "insert into test from infile '$CURDIR/data_orc/test_reader_time_zone.snappy.orc' SETTINGS input_format_orc_read_use_writer_time_zone=true FORMAT ORC" +$CLICKHOUSE_CLIENT -q "select * from test SETTINGS session_timezone='Asia/Shanghai'" +$CLICKHOUSE_CLIENT -q "drop table test" \ No newline at end of file diff --git a/tests/queries/0_stateless/03198_settings_in_csv_tsv_schema_cache.reference b/tests/queries/0_stateless/03198_settings_in_csv_tsv_schema_cache.reference new file mode 100644 index 00000000000..aecacd10e00 --- /dev/null +++ b/tests/queries/0_stateless/03198_settings_in_csv_tsv_schema_cache.reference @@ -0,0 +1,28 @@ +c1 Nullable(Int64) +c2 Nullable(Int64) +c3 Nullable(Int64) +a Nullable(Int64) +b Nullable(Int64) +c Nullable(Int64) +2 +a Nullable(Int64) +b Nullable(Int64) +c Nullable(Int64) +a Nullable(String) +b Nullable(Int64) +c Nullable(Int64) +2 +a Nullable(String) +b Nullable(Int64) +c Nullable(Int64) +a Tuple(Nullable(Int64), Nullable(Int64), Nullable(Int64)) +b Nullable(Int64) +c Nullable(Int64) +2 +c1 Nullable(Int64) +c2 Nullable(Int64) +c3 Nullable(Int64) +a Nullable(Int64) +b Nullable(Int64) +c Nullable(Int64) +2 diff --git a/tests/queries/0_stateless/03198_settings_in_csv_tsv_schema_cache.sh b/tests/queries/0_stateless/03198_settings_in_csv_tsv_schema_cache.sh new file mode 100755 index 00000000000..ce53f467823 --- /dev/null +++ b/tests/queries/0_stateless/03198_settings_in_csv_tsv_schema_cache.sh @@ -0,0 +1,34 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +echo -e 'a,b,c\n1,2,3' > $CLICKHOUSE_TEST_UNIQUE_NAME.csv +$CLICKHOUSE_LOCAL -nm -q " +DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.csv') SETTINGS input_format_csv_skip_first_lines=1; +DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.csv') SETTINGS input_format_csv_skip_first_lines=0; +SELECT count() from system.schema_inference_cache where format = 'CSV' and additional_format_info like '%skip_first_lines%';" + +echo -e 'a,b,c\n"1",2,3' > $CLICKHOUSE_TEST_UNIQUE_NAME.csv +$CLICKHOUSE_LOCAL -nm -q " +DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.csv') SETTINGS input_format_csv_try_infer_numbers_from_strings=1; +DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.csv') SETTINGS input_format_csv_try_infer_numbers_from_strings=0; +SELECT count() from system.schema_inference_cache where format = 'CSV' and additional_format_info like '%try_infer_numbers_from_strings%';" + +echo -e 'a,b,c\n"(1,2,3)",2,3' > $CLICKHOUSE_TEST_UNIQUE_NAME.csv +$CLICKHOUSE_LOCAL -nm -q " +DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.csv') SETTINGS input_format_csv_try_infer_strings_from_quoted_tuples=1; +DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.csv') SETTINGS input_format_csv_try_infer_strings_from_quoted_tuples=0; +SELECT count() from system.schema_inference_cache where format = 'CSV' and additional_format_info like '%try_infer_strings_from_quoted_tuples%';" + +echo -e 'a\tb\tc\n1\t2\t3' > $CLICKHOUSE_TEST_UNIQUE_NAME.tsv +$CLICKHOUSE_LOCAL -nm -q " +DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.tsv') SETTINGS input_format_tsv_skip_first_lines=1; +DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.tsv') SETTINGS input_format_tsv_skip_first_lines=0; +SELECT count() from system.schema_inference_cache where format = 'TSV' and additional_format_info like '%skip_first_lines%';" + + +rm $CLICKHOUSE_TEST_UNIQUE_NAME.csv +rm $CLICKHOUSE_TEST_UNIQUE_NAME.tsv + diff --git a/tests/queries/0_stateless/03198_table_function_directory_path.reference b/tests/queries/0_stateless/03198_table_function_directory_path.reference new file mode 100644 index 00000000000..74cd8c6d31f --- /dev/null +++ b/tests/queries/0_stateless/03198_table_function_directory_path.reference @@ -0,0 +1,4 @@ +2 +2 +1 +1 diff --git a/tests/queries/0_stateless/03198_table_function_directory_path.sql b/tests/queries/0_stateless/03198_table_function_directory_path.sql new file mode 100644 index 00000000000..90f687ed6a3 --- /dev/null +++ b/tests/queries/0_stateless/03198_table_function_directory_path.sql @@ -0,0 +1,15 @@ +-- Tags: no-parallel + +INSERT INTO FUNCTION file('data_03198_table_function_directory_path.csv', 'csv') SELECT '1.csv' SETTINGS engine_file_truncate_on_insert=1; +INSERT INTO FUNCTION file('data_03198_table_function_directory_path/1.csv', 'csv') SELECT '1.csv' SETTINGS engine_file_truncate_on_insert=1; +INSERT INTO FUNCTION file('data_03198_table_function_directory_path/2.csv', 'csv') SELECT '2.csv' SETTINGS engine_file_truncate_on_insert=1; +INSERT INTO FUNCTION file('data_03198_table_function_directory_path/dir/3.csv', 'csv') SELECT '3.csv' SETTINGS engine_file_truncate_on_insert=1; +INSERT INTO FUNCTION file('data_03198_table_function_directory_path/dir1/dir/4.csv', 'csv') SELECT '4.csv' SETTINGS engine_file_truncate_on_insert=1; +INSERT INTO FUNCTION file('data_03198_table_function_directory_path/dir2/dir/5.csv', 'csv') SELECT '5.csv' SETTINGS engine_file_truncate_on_insert=1; + +SELECT COUNT(*) FROM file('data_03198_table_function_directory_path'); +SELECT COUNT(*) FROM file('data_03198_table_function_directory_path/'); +SELECT COUNT(*) FROM file('data_03198_table_function_directory_path/dir'); +SELECT COUNT(*) FROM file('data_03198_table_function_directory_path/*/dir', 'csv'); -- { serverError 74, 636 } +SELECT COUNT(*) FROM file('data_03198_table_function_directory_pat'); -- { serverError 400 } +SELECT COUNT(*) FROM file('data_03198_table_function_directory_path.csv'); diff --git a/tests/queries/0_stateless/03198_unload_primary_key_outdated.reference b/tests/queries/0_stateless/03198_unload_primary_key_outdated.reference new file mode 100644 index 00000000000..28655f938ba --- /dev/null +++ b/tests/queries/0_stateless/03198_unload_primary_key_outdated.reference @@ -0,0 +1,3 @@ +all_1_1_0 1 16 +all_1_1_0 0 0 +all_1_1_0_2 1 16 diff --git a/tests/queries/0_stateless/03198_unload_primary_key_outdated.sh b/tests/queries/0_stateless/03198_unload_primary_key_outdated.sh new file mode 100755 index 00000000000..4f217935123 --- /dev/null +++ b/tests/queries/0_stateless/03198_unload_primary_key_outdated.sh @@ -0,0 +1,32 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT -n " + DROP TABLE IF EXISTS t_unload_primary_key; + + CREATE TABLE t_unload_primary_key (a UInt64, b UInt64) + ENGINE = MergeTree ORDER BY a + SETTINGS old_parts_lifetime = 10000; + + INSERT INTO t_unload_primary_key VALUES (1, 1); + + SELECT name, active, primary_key_bytes_in_memory FROM system.parts WHERE database = '$CLICKHOUSE_DATABASE' AND table = 't_unload_primary_key' ORDER BY name; + + ALTER TABLE t_unload_primary_key UPDATE b = 100 WHERE 1 SETTINGS mutations_sync = 2; +" + +for _ in {1..100}; do + res=$($CLICKHOUSE_CLIENT -q "SELECT primary_key_bytes_in_memory FROM system.parts WHERE database = '$CLICKHOUSE_DATABASE' AND table = 't_unload_primary_key' AND name = 'all_1_1_0'") + if [[ $res -eq 0 ]]; then + break + fi + sleep 0.3 +done + +$CLICKHOUSE_CLIENT -n " + SELECT name, active, primary_key_bytes_in_memory FROM system.parts WHERE database = '$CLICKHOUSE_DATABASE' AND table = 't_unload_primary_key' ORDER BY name; + DROP TABLE IF EXISTS t_unload_primary_key; +" diff --git a/tests/queries/0_stateless/03199_dictionary_table_access.reference b/tests/queries/0_stateless/03199_dictionary_table_access.reference new file mode 100644 index 00000000000..4a703b3be84 --- /dev/null +++ b/tests/queries/0_stateless/03199_dictionary_table_access.reference @@ -0,0 +1,2 @@ +ACCESS_DENIED +ACCESS_DENIED diff --git a/tests/queries/0_stateless/03199_dictionary_table_access.sh b/tests/queries/0_stateless/03199_dictionary_table_access.sh new file mode 100755 index 00000000000..952b466b5da --- /dev/null +++ b/tests/queries/0_stateless/03199_dictionary_table_access.sh @@ -0,0 +1,41 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +username="user_${CLICKHOUSE_TEST_UNIQUE_NAME}" +dictname="dict_${CLICKHOUSE_TEST_UNIQUE_NAME}" +dicttablename="dict_table_${CLICKHOUSE_TEST_UNIQUE_NAME}" + +${CLICKHOUSE_CLIENT} -nm --query " + CREATE DICTIONARY IF NOT EXISTS ${dictname} + ( + id UInt64, + value UInt64 + ) + PRIMARY KEY id + SOURCE(NULL()) + LAYOUT(FLAT()) + LIFETIME(MIN 0 MAX 1000); + CREATE USER IF NOT EXISTS ${username} NOT IDENTIFIED; + GRANT SELECT, CREATE TEMPORARY TABLE ON *.* to ${username}; + SELECT * FROM ${dictname}; + CREATE TABLE ${dicttablename} (id UInt64, value UInt64) + ENGINE = Dictionary(${CLICKHOUSE_DATABASE}.${dictname}); + SELECT * FROM ${dicttablename}; +" + +$CLICKHOUSE_CLIENT -nm --user="${username}" --query " + SELECT * FROM ${dictname}; +" 2>&1 | grep -o ACCESS_DENIED | uniq + +$CLICKHOUSE_CLIENT -nm --user="${username}" --query " + SELECT * FROM ${dicttablename}; +" 2>&1 | grep -o ACCESS_DENIED | uniq + +${CLICKHOUSE_CLIENT} -nm --query " + DROP TABLE IF EXISTS ${dicttablename} SYNC; + DROP DICTIONARY IF EXISTS ${dictname}; + DROP USER IF EXISTS ${username}; +" diff --git a/tests/queries/0_stateless/03199_fix_auc_tie_handling.reference b/tests/queries/0_stateless/03199_fix_auc_tie_handling.reference new file mode 100644 index 00000000000..f35b39d5972 --- /dev/null +++ b/tests/queries/0_stateless/03199_fix_auc_tie_handling.reference @@ -0,0 +1,3 @@ +nan +0.58333 +0.58333 diff --git a/tests/queries/0_stateless/03199_fix_auc_tie_handling.sql b/tests/queries/0_stateless/03199_fix_auc_tie_handling.sql new file mode 100644 index 00000000000..5de0844f445 --- /dev/null +++ b/tests/queries/0_stateless/03199_fix_auc_tie_handling.sql @@ -0,0 +1,32 @@ +CREATE TABLE labels_unordered +( + idx Int64, + score Float64, + label Int64 +) +ENGINE = MergeTree +PRIMARY KEY idx +ORDER BY idx; + +SELECT floor(arrayAUC(array_concat_agg([score]), array_concat_agg([label])), 5) +FROM labels_unordered; + +INSERT INTO labels_unordered (idx,score,label) VALUES (1,0.1,0), (2,0.35,1), (3,0.4,0), (4,0.8,1), (5,0.8,0); + +SELECT floor(arrayAUC(array_concat_agg([score]), array_concat_agg([label])), 5) +FROM labels_unordered; + +CREATE TABLE labels_ordered +( + idx Int64, + score Float64, + label Int64 +) +ENGINE = MergeTree +PRIMARY KEY idx +ORDER BY idx; + +INSERT INTO labels_ordered (idx,score,label) VALUES (1,0.1,0), (2,0.35,1), (3,0.4,0), (4,0.8,0), (5,0.8,1); + +SELECT floor(arrayAUC(array_concat_agg([score]), array_concat_agg([label])), 5) +FROM labels_ordered; \ No newline at end of file diff --git a/tests/queries/0_stateless/01487_distributed_in_not_default_db.reference b/tests/queries/0_stateless/03199_has_lc_fixed_string.reference similarity index 100% rename from tests/queries/0_stateless/01487_distributed_in_not_default_db.reference rename to tests/queries/0_stateless/03199_has_lc_fixed_string.reference index 0d66ea1aee9..b261da18d51 100644 --- a/tests/queries/0_stateless/01487_distributed_in_not_default_db.reference +++ b/tests/queries/0_stateless/03199_has_lc_fixed_string.reference @@ -1,2 +1,2 @@ -0 1 +0 diff --git a/tests/queries/0_stateless/03199_has_lc_fixed_string.sql b/tests/queries/0_stateless/03199_has_lc_fixed_string.sql new file mode 100644 index 00000000000..3cb551804b7 --- /dev/null +++ b/tests/queries/0_stateless/03199_has_lc_fixed_string.sql @@ -0,0 +1,7 @@ +DROP TABLE IF EXISTS 03199_fixedstring_array; +CREATE TABLE 03199_fixedstring_array (arr Array(LowCardinality(FixedString(8)))) ENGINE = Memory; +INSERT INTO 03199_fixedstring_array VALUES (['a', 'b']), (['c', 'd']); + +SELECT has(arr, toFixedString(materialize('a'), 1)) FROM 03199_fixedstring_array; + +DROP TABLE 03199_fixedstring_array; diff --git a/tests/queries/0_stateless/03199_join_with_materialized_column.reference b/tests/queries/0_stateless/03199_join_with_materialized_column.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03199_join_with_materialized_column.sql b/tests/queries/0_stateless/03199_join_with_materialized_column.sql new file mode 100644 index 00000000000..8c53c5b3e66 --- /dev/null +++ b/tests/queries/0_stateless/03199_join_with_materialized_column.sql @@ -0,0 +1,6 @@ +SET allow_experimental_analyzer = 1; + +DROP TABLE IF EXISTS table_with_materialized; +CREATE TABLE table_with_materialized (col String MATERIALIZED 'A') ENGINE = Memory; +SELECT number FROM numbers(1) AS n, table_with_materialized; +DROP TABLE table_with_materialized; diff --git a/tests/queries/0_stateless/03199_json_extract_dynamic.reference b/tests/queries/0_stateless/03199_json_extract_dynamic.reference new file mode 100644 index 00000000000..759b7763cd1 --- /dev/null +++ b/tests/queries/0_stateless/03199_json_extract_dynamic.reference @@ -0,0 +1,30 @@ +true Bool +42 Int64 +-42 Int64 +18446744073709551615 UInt64 +42.42 Float64 +42 Int64 +-42 Int64 +18446744073709551615 UInt64 +Hello String +2020-01-01 Date +2020-01-01 00:00:00.000000000 DateTime64(9) +[1,2,3] Array(Nullable(Int64)) +['str1','str2','str3'] Array(Nullable(String)) +[[[1],[2,3,4]],[[5,6],[7]]] Array(Array(Array(Nullable(Int64)))) +['2020-01-01 00:00:00.000000000','2020-01-01 00:00:00.000000000'] Array(Nullable(DateTime64(9))) +['2020-01-01','2020-01-01 date'] Array(Nullable(String)) +['2020-01-01','2020-01-01 00:00:00','str'] Array(Nullable(String)) +['2020-01-01','2020-01-01 00:00:00','42'] Array(Nullable(String)) +['str','42'] Array(Nullable(String)) +[42,42.42] Array(Nullable(Float64)) +[42,18446744073709552000,42.42] Array(Nullable(Float64)) +[42,42.42] Array(Nullable(Float64)) +[NULL,NULL] Array(Nullable(String)) +[NULL,42] Array(Nullable(Int64)) +[[NULL],[],[42]] Array(Array(Nullable(Int64))) +[[],[NULL,NULL],[1,NULL,3],[NULL,2,NULL]] Array(Array(Nullable(Int64))) +[[],[NULL,NULL],['1',NULL,'3'],[NULL,'2',NULL],['2020-01-01']] Array(Array(Nullable(String))) +('str',42,[42]) Tuple(Nullable(String), Nullable(Int64), Array(Nullable(Int64))) +[42,18446744073709551615] Array(Nullable(UInt64)) +(-42,18446744073709551615) Tuple(Nullable(Int64), Nullable(UInt64)) diff --git a/tests/queries/0_stateless/03199_json_extract_dynamic.sql b/tests/queries/0_stateless/03199_json_extract_dynamic.sql new file mode 100644 index 00000000000..286949f4d3e --- /dev/null +++ b/tests/queries/0_stateless/03199_json_extract_dynamic.sql @@ -0,0 +1,37 @@ +set input_format_json_try_infer_numbers_from_strings=1; + +select JSONExtract(materialize('{"d" : true}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : 42}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : -42}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : 18446744073709551615}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : 42.42}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : "42"}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : "-42"}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : "18446744073709551615"}'), 'd', 'Dynamic') as d, dynamicType(d); + +select JSONExtract(materialize('{"d" : "Hello"}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : "2020-01-01"}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : "2020-01-01 00:00:00.000"}'), 'd', 'Dynamic') as d, dynamicType(d); + +select JSONExtract(materialize('{"d" : [1, 2, 3]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : ["str1", "str2", "str3"]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : [[[1], [2, 3, 4]], [[5, 6], [7]]]}'), 'd', 'Dynamic') as d, dynamicType(d); + +select JSONExtract(materialize('{"d" : ["2020-01-01", "2020-01-01 00:00:00"]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : ["2020-01-01", "2020-01-01 date"]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : ["2020-01-01", "2020-01-01 00:00:00", "str"]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : ["2020-01-01", "2020-01-01 00:00:00", "42"]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : ["str", "42"]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : [42, 42.42]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : [42, 18446744073709551615, 42.42]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : [42, 42.42]}'), 'd', 'Dynamic') as d, dynamicType(d); + +select JSONExtract(materialize('{"d" : [null, null]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : [null, 42]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : [[null], [], [42]]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"a" : [[], [null, null], ["1", null, "3"], [null, "2", null]]}'), 'a', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"a" : [[], [null, null], ["1", null, "3"], [null, "2", null], ["2020-01-01"]]}'), 'a', 'Dynamic') as d, dynamicType(d); + +select JSONExtract(materialize('{"d" : ["str", 42, [42]]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : [42, 18446744073709551615]}'), 'd', 'Dynamic') as d, dynamicType(d); +select JSONExtract(materialize('{"d" : [-42, 18446744073709551615]}'), 'd', 'Dynamic') as d, dynamicType(d); diff --git a/tests/queries/0_stateless/03199_merge_filters_bug.reference b/tests/queries/0_stateless/03199_merge_filters_bug.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03199_merge_filters_bug.sql b/tests/queries/0_stateless/03199_merge_filters_bug.sql new file mode 100644 index 00000000000..ed2ec2ea217 --- /dev/null +++ b/tests/queries/0_stateless/03199_merge_filters_bug.sql @@ -0,0 +1,70 @@ +drop table if exists t1; +drop table if exists t2; + +CREATE TABLE t1 +( + `s1` String, + `s2` String, + `s3` String +) +ENGINE = MergeTree +ORDER BY tuple(); + + +CREATE TABLE t2 +( + `fs1` FixedString(10), + `fs2` FixedString(10) +) +ENGINE = MergeTree +ORDER BY tuple(); + +INSERT INTO t1 SELECT + repeat('t', 15) s1, + 'test' s2, + 'test' s3; + +INSERT INTO t1 SELECT + substring(s1, 1, 10), + s2, + s3 +FROM generateRandom('s1 String, s2 String, s3 String') +LIMIT 10000; + +INSERT INTO t2 SELECT * +FROM generateRandom() +LIMIT 10000; + +WITH +tmp1 AS +( + SELECT + CAST(s1, 'FixedString(10)') AS fs1, + s2 AS sector, + s3 + FROM t1 + WHERE (s3 != 'test') +) + SELECT + fs1 + FROM t2 + LEFT JOIN tmp1 USING (fs1) + WHERE (fs1 IN ('test')) SETTINGS enable_multiple_prewhere_read_steps = 0; + +optimize table t1 final; + +WITH +tmp1 AS +( + SELECT + CAST(s1, 'FixedString(10)') AS fs1, + s2 AS sector, + s3 + FROM t1 + WHERE (s3 != 'test') +) + SELECT + fs1 + FROM t2 + LEFT JOIN tmp1 USING (fs1) + WHERE (fs1 IN ('test')); diff --git a/tests/queries/0_stateless/03199_queries_with_new_analyzer.reference b/tests/queries/0_stateless/03199_queries_with_new_analyzer.reference new file mode 100644 index 00000000000..10ce589000d --- /dev/null +++ b/tests/queries/0_stateless/03199_queries_with_new_analyzer.reference @@ -0,0 +1,27 @@ +5 (4230072075578472911,4230072075578472911) 71789584853496063 +2 (4401188181514187637,4401188181514187637) 878466845199253299 +4 (4940826638032106783,4940826638032106783) 3675164899122807807 +6 (10957420562507184961,10957420562507184961) 3732623117916254211 +0 (797076400500506358,797076400500506358) 3746094338409299772 +7 (10843611042193511775,10843611042193511775) 4607251742847087615 +3 (12588286986351526898,12588286986351526898) 13889114719560662796 +8 (452995860660674674,452995860660674674) 17365664920787500812 +9 (12206106972241516904,12206106972241516904) 17567684527097330880 +1 (14558425114501132193,14558425114501132193) 18445898820068822019 +3 255 255 +0 0 0 +0 0 0 +0 0 0 +0 0 0 +0 0 0 +0 0 0 +0 0 0 +0 0 0 +0 0 0 +0 +1 +2 +3 +4 +5 +6 diff --git a/tests/queries/0_stateless/03199_queries_with_new_analyzer.sql b/tests/queries/0_stateless/03199_queries_with_new_analyzer.sql new file mode 100644 index 00000000000..c32d7524492 --- /dev/null +++ b/tests/queries/0_stateless/03199_queries_with_new_analyzer.sql @@ -0,0 +1,41 @@ +SET allow_experimental_analyzer=1; + +SELECT *, ngramMinHash(*) AS minhash, mortonEncode(untuple(ngramMinHash(*))) AS z +FROM (SELECT toString(number) FROM numbers(10)) +ORDER BY z LIMIT 100; + +CREATE TABLE test ( + idx UInt64, + coverage Array(UInt64), + test_name String +) +ENGINE = MergeTree +ORDER BY tuple(); + +INSERT INTO test VALUES (10, [0,1,2,3], 'xx'), (20, [3,4,5,6], 'xxx'), (90, [3,4,5,6,9], 'xxxx'); + +WITH + 4096 AS w, 4096 AS h, w * h AS pixels, + arrayJoin(coverage) AS num, + num DIV (32768 * 32768 DIV pixels) AS idx, + mortonDecode(2, idx) AS coord, + 255 AS b, + least(255, uniq(test_name)) AS r, + 255 * uniq(test_name) / (max(uniq(test_name)) OVER ()) AS g +SELECT r::UInt8, g::UInt8, b::UInt8 +FROM test +GROUP BY coord +ORDER BY coord.2 * w + coord.1 +WITH FILL FROM 0 TO 10; + + +CREATE TABLE seq ( + number UInt64 +) +ENGINE = MergeTree +ORDER BY tuple(); + +INSERT INTO seq VALUES (0), (6), (7); + +WITH (Select min(number), max(number) from seq) as range Select * from numbers(range.1, range.2); + diff --git a/tests/queries/0_stateless/03199_unbin_buffer_overflow.reference b/tests/queries/0_stateless/03199_unbin_buffer_overflow.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03199_unbin_buffer_overflow.sh b/tests/queries/0_stateless/03199_unbin_buffer_overflow.sh new file mode 100755 index 00000000000..337debebb14 --- /dev/null +++ b/tests/queries/0_stateless/03199_unbin_buffer_overflow.sh @@ -0,0 +1,33 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +# check for buffer overflow in unbin (due to not enough memory preallocate for output buffer) +# we iterate over all remainders of input string length modulo word_size and check that no assertions are triggered + +word_size=8 +for i in $(seq 1 $((word_size+1))); do + str=$(printf "%${i}s" | tr ' ' 'x') + $CLICKHOUSE_CLIENT -q "SELECT count() FROM numbers(99) GROUP BY unbin(toFixedString(materialize('$str'), $i)) WITH ROLLUP WITH TOTALS FORMAT NULL" +done + +word_size=8 +for i in $(seq 1 $((word_size+1))); do + str=$(printf "%${i}s" | tr ' ' 'x') + $CLICKHOUSE_CLIENT -q "SELECT count() FROM numbers(99) GROUP BY unbin(materialize('$str')) WITH ROLLUP WITH TOTALS FORMAT NULL" +done + +word_size=2 +for i in $(seq 1 $((word_size+1))); do + str=$(printf "%${i}s" | tr ' ' 'x') + $CLICKHOUSE_CLIENT -q "SELECT count() FROM numbers(99) GROUP BY unhex(toFixedString(materialize('$str'), $i)) WITH ROLLUP WITH TOTALS FORMAT NULL" +done + +word_size=2 +for i in $(seq 1 $((word_size+1))); do + str=$(printf "%${i}s" | tr ' ' 'x') + $CLICKHOUSE_CLIENT -q "SELECT count() FROM numbers(99) GROUP BY unhex(materialize('$str')) WITH ROLLUP WITH TOTALS FORMAT NULL" +done diff --git a/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.reference b/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.reference new file mode 100644 index 00000000000..6d2c1334d6e --- /dev/null +++ b/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.reference @@ -0,0 +1,10 @@ +\N +\N +\N +\N +\N +\N +\N +\N +\N +\N diff --git a/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.sql b/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.sql new file mode 100644 index 00000000000..a01a595dbb5 --- /dev/null +++ b/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.sql @@ -0,0 +1,7 @@ +set allow_experimental_dynamic_type=1; +create table test (d Dynamic) engine=Memory; +insert into table test select * from numbers(5); +alter table test modify column d Dynamic(max_types=1); +select d.UInt64 from test settings allow_experimental_analyzer=1; +select d.UInt64 from test settings allow_experimental_analyzer=0; + diff --git a/tests/queries/0_stateless/00967_ubsan_bit_test.reference b/tests/queries/0_stateless/03200_subcolumns_join_use_nulls.reference similarity index 100% rename from tests/queries/0_stateless/00967_ubsan_bit_test.reference rename to tests/queries/0_stateless/03200_subcolumns_join_use_nulls.reference diff --git a/tests/queries/0_stateless/03200_subcolumns_join_use_nulls.sql b/tests/queries/0_stateless/03200_subcolumns_join_use_nulls.sql new file mode 100644 index 00000000000..2dd0a37657d --- /dev/null +++ b/tests/queries/0_stateless/03200_subcolumns_join_use_nulls.sql @@ -0,0 +1,13 @@ +DROP TABLE IF EXISTS t_subcolumns_join; + +CREATE TABLE t_subcolumns_join (id UInt64) ENGINE=MergeTree ORDER BY tuple(); + +INSERT INTO t_subcolumns_join SELECT number as number FROM numbers(10000); + +SELECT + count() +FROM (SELECT number FROM numbers(10)) as tbl LEFT JOIN t_subcolumns_join ON number = id +WHERE id is null +SETTINGS allow_experimental_analyzer = 1, optimize_functions_to_subcolumns = 1, join_use_nulls = 1; + +DROP TABLE t_subcolumns_join; diff --git a/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.reference b/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.reference new file mode 100644 index 00000000000..912bff45da5 --- /dev/null +++ b/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.reference @@ -0,0 +1,11 @@ +str_array Array(String) +1318 +5779 +1715 +6422 +5875 +1887 +3763 +4245 +4270 +758 diff --git a/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.sh b/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.sh new file mode 100755 index 00000000000..dcecd7b3bea --- /dev/null +++ b/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.sh @@ -0,0 +1,14 @@ +#!/usr/bin/env bash +# Tags: no-parallel, no-fasttest + +set -e + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +DATA_DIR=$CUR_DIR/data_avro + +# See https://github.com/ClickHouse/ClickHouse/issues/60438 +$CLICKHOUSE_LOCAL -q "DESC file('$DATA_DIR/negative_block_size_arrays.avro')" +$CLICKHOUSE_LOCAL -q "SELECT arraySum(arrayMap(x -> length(x), str_array)) AS res FROM file('$DATA_DIR/negative_block_size_arrays.avro')" diff --git a/tests/queries/0_stateless/03201_local_named_collections.reference b/tests/queries/0_stateless/03201_local_named_collections.reference new file mode 100644 index 00000000000..af5626b4a11 --- /dev/null +++ b/tests/queries/0_stateless/03201_local_named_collections.reference @@ -0,0 +1 @@ +Hello, world! diff --git a/tests/queries/0_stateless/03201_local_named_collections.sh b/tests/queries/0_stateless/03201_local_named_collections.sh new file mode 100755 index 00000000000..809b4b52f41 --- /dev/null +++ b/tests/queries/0_stateless/03201_local_named_collections.sh @@ -0,0 +1,20 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --multiquery " +DROP TABLE IF EXISTS test; +CREATE TABLE test (s String) ORDER BY (); +INSERT INTO test VALUES ('Hello, world!'); +" + +${CLICKHOUSE_LOCAL} --multiquery " +CREATE NAMED COLLECTION mydb AS host = '${CLICKHOUSE_HOST}', port = ${CLICKHOUSE_PORT_TCP}, user = 'default', password = '', db = '${CLICKHOUSE_DATABASE}'; +SELECT * FROM remote(mydb, table = 'test'); +" 2>&1 | grep --text -F -v "ASan doesn't fully support makecontext/swapcontext functions" + +${CLICKHOUSE_CLIENT} --multiquery " +DROP TABLE test; +" diff --git a/tests/queries/0_stateless/03201_sumIf_to_countIf_return_type.reference b/tests/queries/0_stateless/03201_sumIf_to_countIf_return_type.reference new file mode 100644 index 00000000000..62f5eb45106 --- /dev/null +++ b/tests/queries/0_stateless/03201_sumIf_to_countIf_return_type.reference @@ -0,0 +1,24 @@ +QUERY id: 0 + PROJECTION COLUMNS + (sumIf(toInt64(1), 1)) Tuple(Int64) + PROJECTION + LIST id: 1, nodes: 1 + FUNCTION id: 2, function_name: tuple, function_type: ordinary, result_type: Tuple(Int64) + ARGUMENTS + LIST id: 3, nodes: 1 + FUNCTION id: 4, function_name: sumIf, function_type: aggregate, result_type: Int64 + ARGUMENTS + LIST id: 5, nodes: 2 + CONSTANT id: 6, constant_value: Int64_1, constant_value_type: Int64 + EXPRESSION + FUNCTION id: 7, function_name: toInt64, function_type: ordinary, result_type: Int64 + ARGUMENTS + LIST id: 8, nodes: 1 + CONSTANT id: 9, constant_value: UInt64_1, constant_value_type: UInt8 + CONSTANT id: 10, constant_value: UInt64_1, constant_value_type: UInt8 + JOIN TREE + TABLE_FUNCTION id: 11, alias: __table1, table_function_name: numbers + ARGUMENTS + LIST id: 12, nodes: 1 + CONSTANT id: 13, constant_value: UInt64_100, constant_value_type: UInt8 + SETTINGS optimize_rewrite_sum_if_to_count_if=1 diff --git a/tests/queries/0_stateless/03201_sumIf_to_countIf_return_type.sql b/tests/queries/0_stateless/03201_sumIf_to_countIf_return_type.sql new file mode 100644 index 00000000000..24369fd6497 --- /dev/null +++ b/tests/queries/0_stateless/03201_sumIf_to_countIf_return_type.sql @@ -0,0 +1,2 @@ +SET allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT tuple(sumIf(toInt64(1), 1)) FROM numbers(100) settings optimize_rewrite_sum_if_to_count_if=1; diff --git a/tests/queries/0_stateless/03201_variant_null_map_subcolumn.reference b/tests/queries/0_stateless/03201_variant_null_map_subcolumn.reference new file mode 100644 index 00000000000..8565fe3d0fa --- /dev/null +++ b/tests/queries/0_stateless/03201_variant_null_map_subcolumn.reference @@ -0,0 +1,402 @@ +Memory +test +[] 1 0 0 [] +1 0 1 0 [] +\N 1 1 0 [] +['str_3','str_3','str_3'] 1 0 3 [1,1,1] +4 0 1 0 [] +\N 1 1 0 [] +[6,6,6,6,6,6] 1 0 6 [0,0,0,0,0,0] +7 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 9 [1,1,1,1,1,1,1,1,1] +10 0 1 0 [] +\N 1 1 0 [] +['str_12','str_12'] 1 0 2 [1,1] +13 0 1 0 [] +\N 1 1 0 [] +[15,15,15,15,15] 1 0 5 [0,0,0,0,0] +16 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 8 [1,1,1,1,1,1,1,1] +19 0 1 0 [] +\N 1 1 0 [] +['str_21'] 1 0 1 [1] +22 0 1 0 [] +\N 1 1 0 [] +[24,24,24,24] 1 0 4 [0,0,0,0] +25 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 7 [1,1,1,1,1,1,1] +28 0 1 0 [] +\N 1 1 0 [] +[] 1 0 0 [] +31 0 1 0 [] +\N 1 1 0 [] +[33,33,33] 1 0 3 [0,0,0] +34 0 1 0 [] +\N 1 1 0 [] +1 0 0 [] +0 1 0 [] +1 1 0 [] +1 0 3 [1,1,1] +0 1 0 [] +1 1 0 [] +1 0 6 [0,0,0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 9 [1,1,1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 2 [1,1] +0 1 0 [] +1 1 0 [] +1 0 5 [0,0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 8 [1,1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 1 [1] +0 1 0 [] +1 1 0 [] +1 0 4 [0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 7 [1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 0 [] +0 1 0 [] +1 1 0 [] +1 0 3 [0,0,0] +0 1 0 [] +1 1 0 [] +0 0 [] [] +1 0 [] [] +1 0 [] [] +0 3 [1,1,1] [0,0,0] +1 0 [] [] +1 0 [] [] +0 6 [0,0,0,0,0,0] [1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 9 [1,1,1,1,1,1,1,1,1] [1,1,1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 2 [1,1] [0,0] +1 0 [] [] +1 0 [] [] +0 5 [0,0,0,0,0] [1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 8 [1,1,1,1,1,1,1,1] [1,1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 1 [1] [0] +1 0 [] [] +1 0 [] [] +0 4 [0,0,0,0] [1,1,1,1] +1 0 [] [] +1 0 [] [] +0 7 [1,1,1,1,1,1,1] [1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 0 [] [] +1 0 [] [] +1 0 [] [] +0 3 [0,0,0] [1,1,1] +1 0 [] [] +1 0 [] [] +0 +2 +3 +5 +6 +8 +9 +11 +12 +14 +15 +17 +18 +20 +21 +23 +24 +26 +27 +29 +30 +32 +33 +35 +MergeTree compact +test +[] 1 0 0 [] +1 0 1 0 [] +\N 1 1 0 [] +['str_3','str_3','str_3'] 1 0 3 [1,1,1] +4 0 1 0 [] +\N 1 1 0 [] +[6,6,6,6,6,6] 1 0 6 [0,0,0,0,0,0] +7 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 9 [1,1,1,1,1,1,1,1,1] +10 0 1 0 [] +\N 1 1 0 [] +['str_12','str_12'] 1 0 2 [1,1] +13 0 1 0 [] +\N 1 1 0 [] +[15,15,15,15,15] 1 0 5 [0,0,0,0,0] +16 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 8 [1,1,1,1,1,1,1,1] +19 0 1 0 [] +\N 1 1 0 [] +['str_21'] 1 0 1 [1] +22 0 1 0 [] +\N 1 1 0 [] +[24,24,24,24] 1 0 4 [0,0,0,0] +25 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 7 [1,1,1,1,1,1,1] +28 0 1 0 [] +\N 1 1 0 [] +[] 1 0 0 [] +31 0 1 0 [] +\N 1 1 0 [] +[33,33,33] 1 0 3 [0,0,0] +34 0 1 0 [] +\N 1 1 0 [] +1 0 0 [] +0 1 0 [] +1 1 0 [] +1 0 3 [1,1,1] +0 1 0 [] +1 1 0 [] +1 0 6 [0,0,0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 9 [1,1,1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 2 [1,1] +0 1 0 [] +1 1 0 [] +1 0 5 [0,0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 8 [1,1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 1 [1] +0 1 0 [] +1 1 0 [] +1 0 4 [0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 7 [1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 0 [] +0 1 0 [] +1 1 0 [] +1 0 3 [0,0,0] +0 1 0 [] +1 1 0 [] +0 0 [] [] +1 0 [] [] +1 0 [] [] +0 3 [1,1,1] [0,0,0] +1 0 [] [] +1 0 [] [] +0 6 [0,0,0,0,0,0] [1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 9 [1,1,1,1,1,1,1,1,1] [1,1,1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 2 [1,1] [0,0] +1 0 [] [] +1 0 [] [] +0 5 [0,0,0,0,0] [1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 8 [1,1,1,1,1,1,1,1] [1,1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 1 [1] [0] +1 0 [] [] +1 0 [] [] +0 4 [0,0,0,0] [1,1,1,1] +1 0 [] [] +1 0 [] [] +0 7 [1,1,1,1,1,1,1] [1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 0 [] [] +1 0 [] [] +1 0 [] [] +0 3 [0,0,0] [1,1,1] +1 0 [] [] +1 0 [] [] +0 +2 +3 +5 +6 +8 +9 +11 +12 +14 +15 +17 +18 +20 +21 +23 +24 +26 +27 +29 +30 +32 +33 +35 +MergeTree wide +test +[] 1 0 0 [] +1 0 1 0 [] +\N 1 1 0 [] +['str_3','str_3','str_3'] 1 0 3 [1,1,1] +4 0 1 0 [] +\N 1 1 0 [] +[6,6,6,6,6,6] 1 0 6 [0,0,0,0,0,0] +7 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 9 [1,1,1,1,1,1,1,1,1] +10 0 1 0 [] +\N 1 1 0 [] +['str_12','str_12'] 1 0 2 [1,1] +13 0 1 0 [] +\N 1 1 0 [] +[15,15,15,15,15] 1 0 5 [0,0,0,0,0] +16 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 8 [1,1,1,1,1,1,1,1] +19 0 1 0 [] +\N 1 1 0 [] +['str_21'] 1 0 1 [1] +22 0 1 0 [] +\N 1 1 0 [] +[24,24,24,24] 1 0 4 [0,0,0,0] +25 0 1 0 [] +\N 1 1 0 [] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL] 1 0 7 [1,1,1,1,1,1,1] +28 0 1 0 [] +\N 1 1 0 [] +[] 1 0 0 [] +31 0 1 0 [] +\N 1 1 0 [] +[33,33,33] 1 0 3 [0,0,0] +34 0 1 0 [] +\N 1 1 0 [] +1 0 0 [] +0 1 0 [] +1 1 0 [] +1 0 3 [1,1,1] +0 1 0 [] +1 1 0 [] +1 0 6 [0,0,0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 9 [1,1,1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 2 [1,1] +0 1 0 [] +1 1 0 [] +1 0 5 [0,0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 8 [1,1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 1 [1] +0 1 0 [] +1 1 0 [] +1 0 4 [0,0,0,0] +0 1 0 [] +1 1 0 [] +1 0 7 [1,1,1,1,1,1,1] +0 1 0 [] +1 1 0 [] +1 0 0 [] +0 1 0 [] +1 1 0 [] +1 0 3 [0,0,0] +0 1 0 [] +1 1 0 [] +0 0 [] [] +1 0 [] [] +1 0 [] [] +0 3 [1,1,1] [0,0,0] +1 0 [] [] +1 0 [] [] +0 6 [0,0,0,0,0,0] [1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 9 [1,1,1,1,1,1,1,1,1] [1,1,1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 2 [1,1] [0,0] +1 0 [] [] +1 0 [] [] +0 5 [0,0,0,0,0] [1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 8 [1,1,1,1,1,1,1,1] [1,1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 1 [1] [0] +1 0 [] [] +1 0 [] [] +0 4 [0,0,0,0] [1,1,1,1] +1 0 [] [] +1 0 [] [] +0 7 [1,1,1,1,1,1,1] [1,1,1,1,1,1,1] +1 0 [] [] +1 0 [] [] +0 0 [] [] +1 0 [] [] +1 0 [] [] +0 3 [0,0,0] [1,1,1] +1 0 [] [] +1 0 [] [] +0 +2 +3 +5 +6 +8 +9 +11 +12 +14 +15 +17 +18 +20 +21 +23 +24 +26 +27 +29 +30 +32 +33 +35 diff --git a/tests/queries/0_stateless/03201_variant_null_map_subcolumn.sh b/tests/queries/0_stateless/03201_variant_null_map_subcolumn.sh new file mode 100755 index 00000000000..8231691e184 --- /dev/null +++ b/tests/queries/0_stateless/03201_variant_null_map_subcolumn.sh @@ -0,0 +1,44 @@ +#!/usr/bin/env bash +# Tags: long + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# reset --log_comment +CLICKHOUSE_LOG_COMMENT= +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_suspicious_variant_types=1" + +function test() +{ + echo "test" + $CH_CLIENT -q "insert into test select number, multiIf(number % 3 == 2, NULL, number % 3 == 1, number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10))) from numbers(36)" + $CH_CLIENT -q "select v, v.UInt64.null, v.\`Array(Variant(String, UInt64))\`.null, v.\`Array(Variant(String, UInt64))\`.size0, v.\`Array(Variant(String, UInt64))\`.UInt64.null from test order by id" + $CH_CLIENT -q "select v.UInt64.null, v.\`Array(Variant(String, UInt64))\`.null, v.\`Array(Variant(String, UInt64))\`.size0, v.\`Array(Variant(String, UInt64))\`.UInt64.null from test order by id" + $CH_CLIENT -q "select v.\`Array(Variant(String, UInt64))\`.null, v.\`Array(Variant(String, UInt64))\`.size0, v.\`Array(Variant(String, UInt64))\`.UInt64.null, v.\`Array(Variant(String, UInt64))\`.String.null from test order by id" + $CH_CLIENT -q "select id from test where v.UInt64 is null order by id" + + $CH_CLIENT -q "insert into test select number, multiIf(number % 3 == 2, NULL, number % 3 == 1, number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10))) from numbers(1000000) settings min_insert_block_size_rows=100000" + $CH_CLIENT -q "select v, v.UInt64.null, v.\`Array(Variant(String, UInt64))\`.null, v.\`Array(Variant(String, UInt64))\`.size0, v.\`Array(Variant(String, UInt64))\`.UInt64.null from test order by id format Null" + $CH_CLIENT -q "select v.UInt64.null, v.\`Array(Variant(String, UInt64))\`.null, v.\`Array(Variant(String, UInt64))\`.size0, v.\`Array(Variant(String, UInt64))\`.UInt64.null from test order by id format Null" + $CH_CLIENT -q "select v.\`Array(Variant(String, UInt64))\`.null, v.\`Array(Variant(String, UInt64))\`.size0, v.\`Array(Variant(String, UInt64))\`.UInt64.null, v.\`Array(Variant(String, UInt64))\`.String.null from test order by id format Null" + $CH_CLIENT -q "select id from test where v.UInt64 is null order by id format Null" +} + +$CH_CLIENT -q "drop table if exists test;" + +echo "Memory" +$CH_CLIENT -q "create table test (id UInt64, v Variant(UInt64, Array(Variant(String, UInt64)))) engine=Memory" +test +$CH_CLIENT -q "drop table test;" + +echo "MergeTree compact" +$CH_CLIENT -q "create table test (id UInt64, v Variant(UInt64, Array(Variant(String, UInt64)))) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000;" +test +$CH_CLIENT -q "drop table test;" + +echo "MergeTree wide" +$CH_CLIENT -q "create table test (id UInt64, v Variant(UInt64, Array(Variant(String, UInt64)))) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" +test +$CH_CLIENT -q "drop table test;" + diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns.reference b/tests/queries/0_stateless/03202_dynamic_null_map_subcolumn.reference similarity index 60% rename from tests/queries/0_stateless/03036_dynamic_read_subcolumns.reference rename to tests/queries/0_stateless/03202_dynamic_null_map_subcolumn.reference index 36984bc8b9b..8740726c7ef 100644 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns.reference +++ b/tests/queries/0_stateless/03202_dynamic_null_map_subcolumn.reference @@ -5,17 +5,17 @@ Array(Variant(String, UInt64)) None String UInt64 -200000 -200000 -200000 -200000 +20 +20 +20 +20 0 0 -200000 -200000 -100000 -100000 -200000 +20 +20 +10 +10 +20 0 MergeTree compact test @@ -24,17 +24,17 @@ Array(Variant(String, UInt64)) None String UInt64 -200000 -200000 -200000 -200000 +20 +20 +20 +20 0 0 -200000 -200000 -100000 -100000 -200000 +20 +20 +10 +10 +20 0 MergeTree wide test @@ -43,15 +43,15 @@ Array(Variant(String, UInt64)) None String UInt64 -200000 -200000 -200000 -200000 +20 +20 +20 +20 0 0 -200000 -200000 -100000 -100000 -200000 +20 +20 +10 +10 +20 0 diff --git a/tests/queries/0_stateless/03202_dynamic_null_map_subcolumn.sh b/tests/queries/0_stateless/03202_dynamic_null_map_subcolumn.sh new file mode 100755 index 00000000000..968c9e5271f --- /dev/null +++ b/tests/queries/0_stateless/03202_dynamic_null_map_subcolumn.sh @@ -0,0 +1,62 @@ +#!/usr/bin/env bash +# Tags: long + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# reset --log_comment +CLICKHOUSE_LOG_COMMENT= +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +CH_CLIENT="$CLICKHOUSE_CLIENT --allow_experimental_variant_type=1 --use_variant_as_common_type=1 --allow_experimental_dynamic_type=1 --optimize_functions_to_subcolumns=0" + + +function test() +{ + echo "test" + $CH_CLIENT -q "insert into test select number, number from numbers(10) settings min_insert_block_size_rows=50000" + $CH_CLIENT -q "insert into test select number, 'str_' || toString(number) from numbers(10, 10) settings min_insert_block_size_rows=50000" + $CH_CLIENT -q "insert into test select number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1)) from numbers(20, 10) settings min_insert_block_size_rows=50000" + $CH_CLIENT -q "insert into test select number, NULL from numbers(30, 10) settings min_insert_block_size_rows=50000" + $CH_CLIENT -q "insert into test select number, multiIf(number % 4 == 3, 'str_' || toString(number), number % 4 == 2, NULL, number % 4 == 1, number, arrayMap(x -> multiIf(number % 9 == 0, NULL, number % 9 == 3, 'str_' || toString(number), number), range(number % 10 + 1))) from numbers(40, 40) settings min_insert_block_size_rows=50000" + $CH_CLIENT -q "insert into test select number, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)) from numbers(10, 10) settings min_insert_block_size_rows=50000" + + $CH_CLIENT -q "select distinct dynamicType(d) as type from test order by type" + $CH_CLIENT -q "select count() from test where dynamicType(d) == 'UInt64'" + $CH_CLIENT -q "select count() from test where d.UInt64 is not NULL" + $CH_CLIENT -q "select count() from test where dynamicType(d) == 'String'" + $CH_CLIENT -q "select count() from test where d.String is not NULL" + $CH_CLIENT -q "select count() from test where dynamicType(d) == 'Date'" + $CH_CLIENT -q "select count() from test where d.Date is not NULL" + $CH_CLIENT -q "select count() from test where dynamicType(d) == 'Array(Variant(String, UInt64))'" + $CH_CLIENT -q "select count() from test where not empty(d.\`Array(Variant(String, UInt64))\`)" + $CH_CLIENT -q "select count() from test where dynamicType(d) == 'Array(Array(Dynamic))'" + $CH_CLIENT -q "select count() from test where not empty(d.\`Array(Array(Dynamic))\`)" + $CH_CLIENT -q "select count() from test where d is NULL" + $CH_CLIENT -q "select count() from test where not empty(d.\`Tuple(a Array(Dynamic))\`.a.String)" + + $CH_CLIENT -q "select d, d.UInt64.null, d.String.null, d.\`Array(Variant(String, UInt64))\`.null from test format Null" + $CH_CLIENT -q "select d.UInt64.null, d.String.null, d.\`Array(Variant(String, UInt64))\`.null from test format Null" + $CH_CLIENT -q "select d.Int8.null, d.Date.null, d.\`Array(String)\`.null from test format Null" + $CH_CLIENT -q "select d, d.UInt64.null, d.Date.null, d.\`Array(Variant(String, UInt64))\`.null, d.\`Array(Variant(String, UInt64))\`.size0, d.\`Array(Variant(String, UInt64))\`.UInt64.null from test format Null" + $CH_CLIENT -q "select d.UInt64.null, d.Date.null, d.\`Array(Variant(String, UInt64))\`.null, d.\`Array(Variant(String, UInt64))\`.size0, d.\`Array(Variant(String, UInt64))\`.UInt64.null, d.\`Array(Variant(String, UInt64))\`.String.null from test format Null" + $CH_CLIENT -q "select d, d.\`Tuple(a UInt64, b String)\`.a, d.\`Array(Dynamic)\`.\`Variant(String, UInt64)\`.UInt64.null, d.\`Array(Variant(String, UInt64))\`.UInt64.null from test format Null" + $CH_CLIENT -q "select d.\`Array(Dynamic)\`.\`Variant(String, UInt64)\`.UInt64.null, d.\`Array(Dynamic)\`.size0, d.\`Array(Variant(String, UInt64))\`.UInt64.null from test format Null" + $CH_CLIENT -q "select d.\`Array(Array(Dynamic))\`.size1, d.\`Array(Array(Dynamic))\`.UInt64.null, d.\`Array(Array(Dynamic))\`.\`Map(String, Tuple(a UInt64))\`.values.a from test format Null" +} + +$CH_CLIENT -q "drop table if exists test;" + +echo "Memory" +$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=Memory" +test +$CH_CLIENT -q "drop table test;" + +echo "MergeTree compact" +$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000;" +test +$CH_CLIENT -q "drop table test;" + +echo "MergeTree wide" +$CH_CLIENT -q "create table test (id UInt64, d Dynamic) engine=MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;" +test +$CH_CLIENT -q "drop table test;" diff --git a/tests/queries/0_stateless/03203_client_benchmark_options.reference b/tests/queries/0_stateless/03203_client_benchmark_options.reference new file mode 100644 index 00000000000..fd2996b1c78 --- /dev/null +++ b/tests/queries/0_stateless/03203_client_benchmark_options.reference @@ -0,0 +1,6 @@ +Ok +Ok +Ok +Ok +Ok +Ok diff --git a/tests/queries/0_stateless/03203_client_benchmark_options.sh b/tests/queries/0_stateless/03203_client_benchmark_options.sh new file mode 100755 index 00000000000..37a1f2cd3ac --- /dev/null +++ b/tests/queries/0_stateless/03203_client_benchmark_options.sh @@ -0,0 +1,26 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +output=$(${CLICKHOUSE_CLIENT} -t -q "SELECT sleepEachRow(2) FORMAT Null" 2>&1) +{ echo "$output" | grep -q "^2\." && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } + +output=$(${CLICKHOUSE_CLIENT} --time -q "SELECT sleepEachRow(2) FORMAT Null" 2>&1) +{ echo "$output" | grep -q "^2\." && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } + +output=$(${CLICKHOUSE_CLIENT} --memory-usage -q "SELECT sum(number) FROM numbers(10_000) FORMAT Null" 2>&1) +{ echo "$output" | grep -q "^[0-9]\+$" && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } + +output=$(${CLICKHOUSE_CLIENT} --memory-usage=none -q "SELECT sum(number) FROM numbers(10_000) FORMAT Null" 2>&1) +echo -n "$output" # expected no output + +output=$(${CLICKHOUSE_CLIENT} --memory-usage=default -q "SELECT sum(number) FROM numbers(10_000) FORMAT Null" 2>&1) +{ echo "$output" | grep -q "^[0-9]\+$" && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } + +output=$(${CLICKHOUSE_CLIENT} --memory-usage=readable -q "SELECT sum(number) FROM numbers(10_000) FORMAT Null" 2>&1) +{ echo "$output" | grep -q "^[0-9].*B$" && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } + +output=$(${CLICKHOUSE_CLIENT} --memory-usage=unknown -q "SELECT sum(number) FROM numbers(10_000) FORMAT Null" 2>&1) +{ echo "$output" | grep -q "BAD_ARGUMENTS" && echo "Ok"; } || { echo "Fail"; echo "'$output'"; } diff --git a/tests/queries/0_stateless/03203_count_with_non_deterministic_function.reference b/tests/queries/0_stateless/03203_count_with_non_deterministic_function.reference new file mode 100644 index 00000000000..6ed281c757a --- /dev/null +++ b/tests/queries/0_stateless/03203_count_with_non_deterministic_function.reference @@ -0,0 +1,2 @@ +1 +1 diff --git a/tests/queries/0_stateless/03203_count_with_non_deterministic_function.sql b/tests/queries/0_stateless/03203_count_with_non_deterministic_function.sql new file mode 100644 index 00000000000..bb3269da597 --- /dev/null +++ b/tests/queries/0_stateless/03203_count_with_non_deterministic_function.sql @@ -0,0 +1,4 @@ +CREATE TABLE t (p UInt8, x UInt64) Engine = MergeTree PARTITION BY p ORDER BY x; +INSERT INTO t SELECT 0, number FROM numbers(10) SETTINGS max_block_size = 100; +SELECT count() FROM t WHERE p = 0 AND rowNumberInAllBlocks() = 1 SETTINGS allow_experimental_analyzer = 0; +SELECT count() FROM t WHERE p = 0 AND rowNumberInAllBlocks() = 1 SETTINGS allow_experimental_analyzer = 1; diff --git a/tests/queries/0_stateless/03203_grpc_protocol.reference b/tests/queries/0_stateless/03203_grpc_protocol.reference new file mode 100644 index 00000000000..9766475a418 --- /dev/null +++ b/tests/queries/0_stateless/03203_grpc_protocol.reference @@ -0,0 +1 @@ +ok diff --git a/tests/queries/0_stateless/03203_grpc_protocol.sh b/tests/queries/0_stateless/03203_grpc_protocol.sh new file mode 100755 index 00000000000..d51d6382f67 --- /dev/null +++ b/tests/queries/0_stateless/03203_grpc_protocol.sh @@ -0,0 +1,14 @@ +#!/usr/bin/env bash +# Tags: no-fasttest +# Tag no-fasttest: In fasttest, ENABLE_LIBRARIES=0, so the grpc library is not built + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +if [[ -z "$CLICKHOUSE_GRPC_CLIENT" ]]; then + CLICKHOUSE_GRPC_CLIENT="$CURDIR/../../../utils/grpc-client/clickhouse-grpc-client.py" +fi + +# Simple test. +$CLICKHOUSE_GRPC_CLIENT --query "SELECT 'ok'" diff --git a/tests/queries/0_stateless/03203_multiif_and_where_2_conditions_old_analyzer_bug.reference b/tests/queries/0_stateless/03203_multiif_and_where_2_conditions_old_analyzer_bug.reference new file mode 100644 index 00000000000..5512e59cc73 --- /dev/null +++ b/tests/queries/0_stateless/03203_multiif_and_where_2_conditions_old_analyzer_bug.reference @@ -0,0 +1,6 @@ +this query used to be broken in old analyser: +c1 c2 yes true +this query worked: +yes true +experimental analyzer: +c1 c2 yes true diff --git a/tests/queries/0_stateless/03203_multiif_and_where_2_conditions_old_analyzer_bug.sql b/tests/queries/0_stateless/03203_multiif_and_where_2_conditions_old_analyzer_bug.sql new file mode 100644 index 00000000000..c97333f7df4 --- /dev/null +++ b/tests/queries/0_stateless/03203_multiif_and_where_2_conditions_old_analyzer_bug.sql @@ -0,0 +1,35 @@ +DROP TABLE IF EXISTS bugcheck1; + +CREATE TABLE bugcheck1 +ENGINE = MergeTree +ORDER BY tuple() +AS SELECT + 'c1' as column_a, + 'c2' as column_b; + +select 'this query used to be broken in old analyser:'; +SELECT *, + multiIf(column_b IN (SELECT 'c2' as someproduct), 'yes', 'no') AS condition_1, + multiIf(column_b = 'c2', 'true', 'false') AS condition_2 +FROM (SELECT column_a, column_b FROM bugcheck1) +WHERE (condition_1 IN ('yes')) AND (condition_2 in ('true')) +settings allow_experimental_analyzer=0; + +select 'this query worked:'; + +SELECT + multiIf(column_b IN (SELECT 'c2' as someproduct), 'yes', 'no') AS condition_1, + multiIf(column_b = 'c2', 'true', 'false') AS condition_2 +FROM (SELECT column_a, column_b FROM bugcheck1) +WHERE (condition_1 IN ('yes')) AND (condition_2 in ('true')) +settings allow_experimental_analyzer=0; + +select 'experimental analyzer:'; +SELECT *, + multiIf(column_b IN (SELECT 'c2' as someproduct), 'yes', 'no') AS condition_1, + multiIf(column_b = 'c2', 'true', 'false') AS condition_2 +FROM (SELECT column_a, column_b FROM bugcheck1) +WHERE (condition_1 IN ('yes')) AND (condition_2 in ('true')) +settings allow_experimental_analyzer=1; + +DROP TABLE bugcheck1; diff --git a/tests/queries/0_stateless/03203_optimize_disjunctions_chain_to_in.reference b/tests/queries/0_stateless/03203_optimize_disjunctions_chain_to_in.reference new file mode 100644 index 00000000000..353c8f98b03 --- /dev/null +++ b/tests/queries/0_stateless/03203_optimize_disjunctions_chain_to_in.reference @@ -0,0 +1,2 @@ +2020-01-01 +2020-01-02 diff --git a/tests/queries/0_stateless/03203_optimize_disjunctions_chain_to_in.sql b/tests/queries/0_stateless/03203_optimize_disjunctions_chain_to_in.sql new file mode 100644 index 00000000000..f9ba28bcd60 --- /dev/null +++ b/tests/queries/0_stateless/03203_optimize_disjunctions_chain_to_in.sql @@ -0,0 +1,12 @@ +SET allow_experimental_analyzer=1; +CREATE TABLE foo (i Date) ENGINE MergeTree ORDER BY i; +INSERT INTO foo VALUES ('2020-01-01'); +INSERT INTO foo VALUES ('2020-01-02'); + +SET optimize_min_equality_disjunction_chain_length = 3; +SELECT * +FROM foo +WHERE (foo.i = parseDateTimeBestEffort('2020-01-01')) + OR (foo.i = parseDateTimeBestEffort('2020-01-02')) + OR (foo.i = parseDateTimeBestEffort('2020-01-03')) +ORDER BY foo.i ASC diff --git a/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_complex.reference b/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_complex.reference new file mode 100644 index 00000000000..d38b21d2d05 --- /dev/null +++ b/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_complex.reference @@ -0,0 +1,14 @@ +18679 31 +0 +10 +20 +30 +40 +50 +60 +70 +80 +90 +100 +110 +4250 diff --git a/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_complex.sql b/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_complex.sql new file mode 100644 index 00000000000..756e08da27d --- /dev/null +++ b/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_complex.sql @@ -0,0 +1,38 @@ +--- The following query was buggy before, so let's use it as a test case +WITH + (num > 1) AND (arraySum(arrayMap(y -> ((y > 1) AND (y < num) AND ((num % y) = 0)), range(toUInt64(sqrt(num)) + 1))) = 0) AS is_prime_slow +SELECT + num, + ds, +FROM +( + WITH + arraySum(arrayMap(y -> toUInt8(y), splitByString('', toString(num)))) AS digits_sum + SELECT + 1 + (number * 2) AS num, + digits_sum AS ds + FROM numbers_mt(10000) + WHERE ds IN ( + WITH + (number > 1) AND (arraySum(arrayMap(y -> ((y > 1) AND (y < number) AND ((number % y) = 0)), range(toUInt64(sqrt(number)) + 1))) = 0) AS is_prime_slow + SELECT number + FROM numbers(180 + 1) + WHERE is_prime_slow + ) +) +WHERE is_prime_slow +ORDER BY num ASC +LIMIT 998, 1 +SETTINGS max_block_size = 64, max_threads=16; + +SELECT number +FROM numbers_mt(120) +WHERE (number % 10) = 0 +ORDER BY number ASC +SETTINGS max_block_size = 31, max_threads = 11; + +SELECT number +FROM numbers_mt(4242, 9) +WHERE (number % 10) = 0 +ORDER BY number ASC +SETTINGS max_block_size = 31, max_threads = 11; diff --git a/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_simple.reference b/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_simple.reference new file mode 100644 index 00000000000..6aad3ee0b4b --- /dev/null +++ b/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_simple.reference @@ -0,0 +1,6 @@ +case 1 +9900 +9910 +9920 +case 2 +9990 diff --git a/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_simple.sql b/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_simple.sql new file mode 100644 index 00000000000..df01ddf4312 --- /dev/null +++ b/tests/queries/0_stateless/03203_system_numbers_limit_and_offset_simple.sql @@ -0,0 +1,11 @@ +SELECT 'case 1'; +SELECT number FROM numbers_mt(10000) +WHERE (number % 10) = 0 +ORDER BY number ASC +LIMIT 990, 3; + +SELECT 'case 2'; +SELECT number FROM numbers_mt(10000) +WHERE (number % 10) = 0 +ORDER BY number ASC +LIMIT 999, 20 SETTINGS max_block_size = 31; diff --git a/tests/queries/0_stateless/03203_variant_convert_field_to_type_bug.reference b/tests/queries/0_stateless/03203_variant_convert_field_to_type_bug.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03203_variant_convert_field_to_type_bug.sql b/tests/queries/0_stateless/03203_variant_convert_field_to_type_bug.sql new file mode 100644 index 00000000000..b73bb8ffa6d --- /dev/null +++ b/tests/queries/0_stateless/03203_variant_convert_field_to_type_bug.sql @@ -0,0 +1,5 @@ +set allow_experimental_variant_type=1; +set use_variant_as_common_type=1; + +SELECT * FROM numbers([tuple(1, 2), NULL], 2); -- {serverError ILLEGAL_TYPE_OF_ARGUMENT} + diff --git a/tests/queries/0_stateless/03204_distributed_with_scalar_subquery.reference b/tests/queries/0_stateless/03204_distributed_with_scalar_subquery.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03204_distributed_with_scalar_subquery.sql b/tests/queries/0_stateless/03204_distributed_with_scalar_subquery.sql new file mode 100644 index 00000000000..0a07ce48268 --- /dev/null +++ b/tests/queries/0_stateless/03204_distributed_with_scalar_subquery.sql @@ -0,0 +1,10 @@ +DROP TABLE IF EXISTS t_c3oollc8r; +CREATE TABLE t_c3oollc8r (c_k37 Int32, c_y String, c_bou Int32, c_g1 Int32, c_lfntfzg Int32, c_kntw50q Int32) ENGINE = MergeTree ORDER BY (); + +SELECT ( + SELECT c_k37 + FROM t_c3oollc8r + ) > c_lfntfzg +FROM remote('127.0.0.{1,2}', currentDatabase(), t_c3oollc8r); + +DROP TABLE t_c3oollc8r; diff --git a/tests/queries/0_stateless/03204_format_join_on.reference b/tests/queries/0_stateless/03204_format_join_on.reference new file mode 100644 index 00000000000..846f36fcca4 --- /dev/null +++ b/tests/queries/0_stateless/03204_format_join_on.reference @@ -0,0 +1,4 @@ +SELECT * FROM t1 INNER JOIN t2 ON ((t1.x = t2.x) AND (t1.x IS NULL) AS e2) +SELECT * FROM t1 INNER JOIN t2 ON ((t1.x = t2.x) AND (t1.x IS NULL) AS e2) +SELECT * FROM t1 INNER JOIN t2 ON (t1.x = t2.x) AND ((t1.x IS NULL) AS e2) +SELECT * FROM t1 INNER JOIN t2 ON t1.x = t2.x diff --git a/tests/queries/0_stateless/03204_format_join_on.sh b/tests/queries/0_stateless/03204_format_join_on.sh new file mode 100755 index 00000000000..87b0afac042 --- /dev/null +++ b/tests/queries/0_stateless/03204_format_join_on.sh @@ -0,0 +1,15 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +# regression for the JOIN ON alias for the whole expression +phase1="$($CLICKHOUSE_FORMAT --oneline --query "SELECT * FROM t1 JOIN t2 ON ((t1.x = t2.x) AND (t1.x IS NULL) AS e2)")" +echo "$phase1" +# phase 2 +$CLICKHOUSE_FORMAT --oneline --query "$phase1" + +# other test cases +$CLICKHOUSE_FORMAT --oneline --query "SELECT * FROM t1 JOIN t2 ON (t1.x = t2.x) AND (t1.x IS NULL AS e2)" +$CLICKHOUSE_FORMAT --oneline --query "SELECT * FROM t1 JOIN t2 ON t1.x = t2.x" diff --git a/tests/queries/0_stateless/03204_index_hint_fuzzer.reference b/tests/queries/0_stateless/03204_index_hint_fuzzer.reference new file mode 100644 index 00000000000..f750ba35102 --- /dev/null +++ b/tests/queries/0_stateless/03204_index_hint_fuzzer.reference @@ -0,0 +1 @@ +(1,1) diff --git a/tests/queries/0_stateless/03204_index_hint_fuzzer.sql b/tests/queries/0_stateless/03204_index_hint_fuzzer.sql new file mode 100644 index 00000000000..5794f3eee53 --- /dev/null +++ b/tests/queries/0_stateless/03204_index_hint_fuzzer.sql @@ -0,0 +1 @@ +SELECT tuple(indexHint(toLowCardinality('aaa')), 1); diff --git a/tests/queries/0_stateless/03205_column_type_check.reference b/tests/queries/0_stateless/03205_column_type_check.reference new file mode 100644 index 00000000000..466cabedd2e --- /dev/null +++ b/tests/queries/0_stateless/03205_column_type_check.reference @@ -0,0 +1,5 @@ +1 nan 1048575 2 +1 1 1 1 1 +() 1 nan 1048575 2 + +() 1 nan 1048575 2 diff --git a/tests/queries/0_stateless/03205_column_type_check.sql b/tests/queries/0_stateless/03205_column_type_check.sql new file mode 100644 index 00000000000..332061d6be8 --- /dev/null +++ b/tests/queries/0_stateless/03205_column_type_check.sql @@ -0,0 +1,9 @@ +SELECT * FROM (SELECT toUInt256(1)) AS t, (SELECT greatCircleAngle(toLowCardinality(toNullable(toUInt256(1048575))), 257, -9223372036854775808, 1048576), 1048575, materialize(2)) AS u; + + +SET join_algorithm='hash'; +SET allow_experimental_join_condition=1; +SELECT * FROM ( SELECT 1 AS a, toLowCardinality(1), 1) AS t1 CROSS JOIN (SELECT toLowCardinality(1 AS a), 1 AS b) AS t2; + + +SELECT * FROM (SELECT tuple(), 1 GROUP BY greatCircleAngle(toNullable(1048575), 257, toInt128(-9223372036854775808), materialize(1048576)) WITH TOTALS) AS t, (SELECT greatCircleAngle(toUInt256(1048575), 257, toNullable(-9223372036854775808), 1048576), 1048575, 2) AS u diff --git a/tests/queries/0_stateless/03205_hashing_empty_tuples.reference b/tests/queries/0_stateless/03205_hashing_empty_tuples.reference new file mode 100644 index 00000000000..e24b5809aee --- /dev/null +++ b/tests/queries/0_stateless/03205_hashing_empty_tuples.reference @@ -0,0 +1,22 @@ +16324913028386710556 +16324913028386710556 +5049034479224883533 +7385293435322750976 +12248912094175844631 +5049034479224883533 +5887129541803688833 +5887129541803688833 +13747979201178469747 +5887129541803688833 +15520217392480966957 +16324913028386710556 +16324913028386710556 +5049034479224883533 +7385293435322750976 +12248912094175844631 +5049034479224883533 +5887129541803688833 +5887129541803688833 +13747979201178469747 +5887129541803688833 +15520217392480966957 diff --git a/tests/queries/0_stateless/03205_hashing_empty_tuples.sql b/tests/queries/0_stateless/03205_hashing_empty_tuples.sql new file mode 100644 index 00000000000..4a97f30ced3 --- /dev/null +++ b/tests/queries/0_stateless/03205_hashing_empty_tuples.sql @@ -0,0 +1,23 @@ +SELECT sipHash64(()); +SELECT sipHash64((), ()); +SELECT sipHash64((), 1); +SELECT sipHash64(1, ()); +SELECT sipHash64(1, (), 1); +SELECT sipHash64((), 1, ()); +SELECT sipHash64((), (1, 2)); +SELECT sipHash64((), (1, 2)); +SELECT sipHash64((1, 2), ()); +SELECT sipHash64((), (1, 2), ()); +SELECT sipHash64((1, 2), (), (3, 4)); + +SELECT sipHash64(materialize(())); +SELECT sipHash64(materialize(()), materialize(())); +SELECT sipHash64(materialize(()), 1); +SELECT sipHash64(1, materialize(())); +SELECT sipHash64(1, materialize(()), 1); +SELECT sipHash64((), 1, materialize(())); +SELECT sipHash64(materialize(()), (1, 2)); +SELECT sipHash64(materialize(()), (1, 2)); +SELECT sipHash64((1, 2), materialize(())); +SELECT sipHash64(materialize(()), (1, 2), ()); +SELECT sipHash64((1, 2), materialize(()), (3, 4)); diff --git a/tests/queries/0_stateless/03205_parallel_replicas_alter_select_ubsan.reference b/tests/queries/0_stateless/03205_parallel_replicas_alter_select_ubsan.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03205_parallel_replicas_alter_select_ubsan.sql b/tests/queries/0_stateless/03205_parallel_replicas_alter_select_ubsan.sql new file mode 100644 index 00000000000..2ec9368327d --- /dev/null +++ b/tests/queries/0_stateless/03205_parallel_replicas_alter_select_ubsan.sql @@ -0,0 +1,35 @@ +SET alter_sync = 2; +SET max_parallel_replicas = 3, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_for_non_replicated_merge_tree = true; + +DROP TABLE IF EXISTS t1; +DROP TABLE IF EXISTS t1__fuzz_26; + +CREATE TABLE t1__fuzz_26 (`a` Nullable(Float64), `b` Nullable(Float32), `pk` Int64) ENGINE = MergeTree ORDER BY pk; +CREATE TABLE t1 ( a Float64, b Int64, pk String) Engine = MergeTree() ORDER BY pk; + +ALTER TABLE t1 + (MODIFY COLUMN `a` Float64 TTL toDateTime(b) + toIntervalMonth(viewExplain('EXPLAIN', 'actions = 1', ( + SELECT + toIntervalMonth(1), + 2 + FROM t1__fuzz_26 + GROUP BY + toFixedString('%Prewhere%', 10), + toNullable(12) + WITH ROLLUP + )), 1)) settings allow_experimental_parallel_reading_from_replicas = 1; -- { serverError INCORRECT_RESULT_OF_SCALAR_SUBQUERY } + +ALTER TABLE t1 + (MODIFY COLUMN `a` Float64 TTL toDateTime(b) + toIntervalMonth(viewExplain('EXPLAIN', 'actions = 1', ( + SELECT + toIntervalMonth(1), + 2 + FROM t1__fuzz_26 + GROUP BY + toFixedString('%Prewhere%', 10), + toNullable(12) + WITH ROLLUP + )), 1)) settings allow_experimental_parallel_reading_from_replicas = 0; -- { serverError INCORRECT_RESULT_OF_SCALAR_SUBQUERY } + +DROP TABLE t1; +DROP TABLE t1__fuzz_26; diff --git a/tests/queries/0_stateless/03205_parallel_window_finctions_and_column_sparse_bug.reference b/tests/queries/0_stateless/03205_parallel_window_finctions_and_column_sparse_bug.reference new file mode 100644 index 00000000000..13e229432ae --- /dev/null +++ b/tests/queries/0_stateless/03205_parallel_window_finctions_and_column_sparse_bug.reference @@ -0,0 +1,10 @@ +false 1 1 +true 1 1 +--- +false 1 1 +false 1 2 +false 1 3 +true 1 1 +true 1 2 +--- +-755809149 0 diff --git a/tests/queries/0_stateless/03205_parallel_window_finctions_and_column_sparse_bug.sql b/tests/queries/0_stateless/03205_parallel_window_finctions_and_column_sparse_bug.sql new file mode 100644 index 00000000000..4cc54bb5ac2 --- /dev/null +++ b/tests/queries/0_stateless/03205_parallel_window_finctions_and_column_sparse_bug.sql @@ -0,0 +1,33 @@ +create table t(c Int32, d Bool) Engine=MergeTree order by c; +system stop merges t; + +insert into t values (1, 0); +insert into t values (1, 0); +insert into t values (1, 1); +insert into t values (1, 0)(1, 1); + +SELECT d, c, row_number() over (partition by d order by c) as c8 FROM t qualify c8=1 order by d settings max_threads=2, allow_experimental_analyzer = 1; +SELECT '---'; +SELECT d, c, row_number() over (partition by d order by c) as c8 FROM t order by d, c8 settings max_threads=2; +SELECT '---'; + +drop table t; + +create table t ( + c Int32 primary key , + s Bool , + w Float64 + ); + +system stop merges t; + +insert into t values(439499072,true,0),(1393290072,true,0); +insert into t values(-1317193174,false,0),(1929066636,false,0); +insert into t values(-2,false,0),(1962246186,true,0),(2054878592,false,0); +insert into t values(-1893563136,true,41.55); +insert into t values(-1338380855,true,-0.7),(-991301833,true,0),(-755809149,false,43.18),(-41,true,0),(3,false,0),(255,false,0),(255,false,0),(189195893,false,0),(195550885,false,9223372036854776000); + +SELECT * FROM ( +SELECT c, min(w) OVER (PARTITION BY s ORDER BY c ASC, s ASC, w ASC) +FROM t limit toUInt64(-1)) +WHERE c = -755809149; diff --git a/tests/queries/0_stateless/03206_is_null_constant_result_old_analyzer_bug.reference b/tests/queries/0_stateless/03206_is_null_constant_result_old_analyzer_bug.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03206_is_null_constant_result_old_analyzer_bug.sql b/tests/queries/0_stateless/03206_is_null_constant_result_old_analyzer_bug.sql new file mode 100644 index 00000000000..39e9ebfcd77 --- /dev/null +++ b/tests/queries/0_stateless/03206_is_null_constant_result_old_analyzer_bug.sql @@ -0,0 +1,23 @@ +CREATE TABLE left (x UUID) ORDER BY tuple(); + +CREATE TABLE right (x UUID) ORDER BY tuple(); + +set allow_experimental_analyzer=0; + +SELECT left.x, (right.x IS NULL)::Boolean FROM left LEFT OUTER JOIN right ON left.x = right.x GROUP BY ALL; + +SELECT isNullable(number)::Boolean, now() FROM numbers(2) GROUP BY isNullable(number)::Boolean, now() FORMAT Null; + +SELECT isNull(number)::Boolean, now() FROM numbers(2) GROUP BY isNull(number)::Boolean, now() FORMAT Null; + +SELECT (number IS NULL)::Boolean, now() FROM numbers(2) GROUP BY (number IS NULL)::Boolean, now() FORMAT Null; + +set allow_experimental_analyzer=1; + +SELECT left.x, (right.x IS NULL)::Boolean FROM left LEFT OUTER JOIN right ON left.x = right.x GROUP BY ALL; + +SELECT isNullable(number)::Boolean, now() FROM numbers(2) GROUP BY isNullable(number)::Boolean, now() FORMAT Null; + +SELECT isNull(number)::Boolean, now() FROM numbers(2) GROUP BY isNull(number)::Boolean, now() FORMAT Null; + +SELECT (number IS NULL)::Boolean, now() FROM numbers(2) GROUP BY (number IS NULL)::Boolean, now() FORMAT Null; diff --git a/tests/queries/0_stateless/03207_composite_expressions_lambda_consistent_formatting.reference b/tests/queries/0_stateless/03207_composite_expressions_lambda_consistent_formatting.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03207_composite_expressions_lambda_consistent_formatting.sql b/tests/queries/0_stateless/03207_composite_expressions_lambda_consistent_formatting.sql new file mode 100644 index 00000000000..2e2f5510876 --- /dev/null +++ b/tests/queries/0_stateless/03207_composite_expressions_lambda_consistent_formatting.sql @@ -0,0 +1,7 @@ +SET allow_experimental_analyzer = 1; +SELECT [1, (x -> 1)]; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT (1, (x -> 1)); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT map(1, (x -> 1)); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT [1, lambda(x, 1)]; -- { serverError UNKNOWN_IDENTIFIER } +SELECT (1, lambda(x, 1)); -- { serverError UNKNOWN_IDENTIFIER } +SELECT map(1, lambda(x, 1)); -- { serverError UNKNOWN_IDENTIFIER } diff --git a/tests/queries/0_stateless/03208_groupArrayIntersect_serialization.reference b/tests/queries/0_stateless/03208_groupArrayIntersect_serialization.reference new file mode 100644 index 00000000000..e84856c90fd --- /dev/null +++ b/tests/queries/0_stateless/03208_groupArrayIntersect_serialization.reference @@ -0,0 +1,13 @@ +010101 AggregateFunction(groupArrayIntersect, Array(UInt8)) +[1] +1 [2,4,6,8,10] +2 [2,4,6,8,10] +3 [2,4,6,8,10] +5 [2,6,10] +6 [10] +7 [] +a [(['2','4','6','8','10'])] +b [(['2','4','6','8','10'])] +c [(['2','4','6','8','10'])] +d [] +e [] diff --git a/tests/queries/0_stateless/03208_groupArrayIntersect_serialization.sql b/tests/queries/0_stateless/03208_groupArrayIntersect_serialization.sql new file mode 100644 index 00000000000..1b3d48ce0c3 --- /dev/null +++ b/tests/queries/0_stateless/03208_groupArrayIntersect_serialization.sql @@ -0,0 +1,43 @@ +SELECT hex(groupArrayIntersectState([1]) AS a), toTypeName(a); +SELECT finalizeAggregation(CAST(unhex('010101'), 'AggregateFunction(groupArrayIntersect, Array(UInt8))')); + +DROP TABLE IF EXISTS grouparray; +CREATE TABLE grouparray +( + `v` AggregateFunction(groupArrayIntersect, Array(UInt8)) +) +ENGINE = Log; + +INSERT INTO grouparray Select groupArrayIntersectState([2, 4, 6, 8, 10]::Array(UInt8)); +SELECT '1', arraySort(groupArrayIntersectMerge(v)) FROM grouparray; +INSERT INTO grouparray Select groupArrayIntersectState([2, 4, 6, 8, 10]::Array(UInt8)); +SELECT '2', arraySort(groupArrayIntersectMerge(v)) FROM grouparray; +INSERT INTO grouparray Select groupArrayIntersectState([1, 2, 3, 4, 5, 6, 7, 8, 9, 10]::Array(UInt8)); +SELECT '3', arraySort(groupArrayIntersectMerge(v)) FROM grouparray; +INSERT INTO grouparray Select groupArrayIntersectState([2, 6, 10]::Array(UInt8)); +SELECT '5', arraySort(groupArrayIntersectMerge(v)) FROM grouparray; +INSERT INTO grouparray Select groupArrayIntersectState([10]::Array(UInt8)); +SELECT '6', arraySort(groupArrayIntersectMerge(v)) FROM grouparray; +INSERT INTO grouparray Select groupArrayIntersectState([]::Array(UInt8)); +SELECT '7', arraySort(groupArrayIntersectMerge(v)) FROM grouparray; + +DROP TABLE IF EXISTS grouparray; + + +DROP TABLE IF EXISTS grouparray_string; +CREATE TABLE grouparray_string +( + `v` AggregateFunction(groupArrayIntersect, Array(Tuple(Array(String)))) +) +ENGINE = Log; + +INSERT INTO grouparray_string Select groupArrayIntersectState([tuple(['2', '4', '6', '8', '10'])]); +SELECT 'a', arraySort(groupArrayIntersectMerge(v)) FROM grouparray_string; +INSERT INTO grouparray_string Select groupArrayIntersectState([tuple(['2', '4', '6', '8', '10']), tuple(['2', '4', '6', '8', '10'])]); +SELECT 'b', arraySort(groupArrayIntersectMerge(v)) FROM grouparray_string; +INSERT INTO grouparray_string Select groupArrayIntersectState([tuple(['2', '4', '6', '8', '10']), tuple(['2', '4', '6', '8', '10', '14'])]); +SELECT 'c', arraySort(groupArrayIntersectMerge(v)) FROM grouparray_string; +INSERT INTO grouparray_string Select groupArrayIntersectState([tuple(['2', '4', '6', '8', '10', '20']), tuple(['2', '4', '6', '8', '10', '14'])]); +SELECT 'd', arraySort(groupArrayIntersectMerge(v)) FROM grouparray_string; +INSERT INTO grouparray_string Select groupArrayIntersectState([]::Array(Tuple(Array(String)))); +SELECT 'e', arraySort(groupArrayIntersectMerge(v)) FROM grouparray_string; diff --git a/tests/queries/0_stateless/03208_inconsistent_formatting_of_not_subquery.reference b/tests/queries/0_stateless/03208_inconsistent_formatting_of_not_subquery.reference new file mode 100644 index 00000000000..a1afeb1ab82 --- /dev/null +++ b/tests/queries/0_stateless/03208_inconsistent_formatting_of_not_subquery.reference @@ -0,0 +1 @@ +SELECT NOT ((SELECT 1)) diff --git a/tests/queries/0_stateless/03208_inconsistent_formatting_of_not_subquery.sh b/tests/queries/0_stateless/03208_inconsistent_formatting_of_not_subquery.sh new file mode 100755 index 00000000000..594d316b621 --- /dev/null +++ b/tests/queries/0_stateless/03208_inconsistent_formatting_of_not_subquery.sh @@ -0,0 +1,7 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +$CLICKHOUSE_FORMAT --oneline --query "SELECT NOT((SELECT 1))" diff --git a/tests/queries/0_stateless/03208_numbers_total_rows_approx.reference b/tests/queries/0_stateless/03208_numbers_total_rows_approx.reference new file mode 100644 index 00000000000..0cfbf08886f --- /dev/null +++ b/tests/queries/0_stateless/03208_numbers_total_rows_approx.reference @@ -0,0 +1 @@ +2 diff --git a/tests/queries/0_stateless/03208_numbers_total_rows_approx.sql b/tests/queries/0_stateless/03208_numbers_total_rows_approx.sql new file mode 100644 index 00000000000..7855dfb6207 --- /dev/null +++ b/tests/queries/0_stateless/03208_numbers_total_rows_approx.sql @@ -0,0 +1 @@ +SELECT number FROM numbers(2, 1) WHERE number % 2 = 0 SETTINGS max_rows_to_read = 10; diff --git a/tests/queries/0_stateless/03209_functions_json_msan_fuzzer_issue.reference b/tests/queries/0_stateless/03209_functions_json_msan_fuzzer_issue.reference new file mode 100644 index 00000000000..e02f3666d40 --- /dev/null +++ b/tests/queries/0_stateless/03209_functions_json_msan_fuzzer_issue.reference @@ -0,0 +1 @@ +0 0 0 1.1 diff --git a/tests/queries/0_stateless/03209_functions_json_msan_fuzzer_issue.sql b/tests/queries/0_stateless/03209_functions_json_msan_fuzzer_issue.sql new file mode 100644 index 00000000000..a05b07d5971 --- /dev/null +++ b/tests/queries/0_stateless/03209_functions_json_msan_fuzzer_issue.sql @@ -0,0 +1,2 @@ +WITH '{ "v":1.1}' AS raw SELECT JSONExtract(raw, 'float') AS float32_1, JSONExtract(concat(tuple('1970-01-05', 10, materialize(10), 10, 10, 10, toUInt256(10), 10, toNullable(10), 10, 10), materialize(toUInt256(0)), ', ', 2, 2, toLowCardinality(toLowCardinality(2))), raw, toLowCardinality('v'), 'Float32') AS float32_2, JSONExtractFloat(raw) AS float64_1, JSONExtract(raw, 'v', 'double') AS float64_2; + diff --git a/tests/queries/0_stateless/03210_convert_outer_join_to_inner_join_any_join.reference b/tests/queries/0_stateless/03210_convert_outer_join_to_inner_join_any_join.reference new file mode 100644 index 00000000000..3d6a23045fb --- /dev/null +++ b/tests/queries/0_stateless/03210_convert_outer_join_to_inner_join_any_join.reference @@ -0,0 +1,3 @@ +1 tx1 US +1 tx2 US +1 tx3 US diff --git a/tests/queries/0_stateless/03210_convert_outer_join_to_inner_join_any_join.sql b/tests/queries/0_stateless/03210_convert_outer_join_to_inner_join_any_join.sql new file mode 100644 index 00000000000..599875e90cf --- /dev/null +++ b/tests/queries/0_stateless/03210_convert_outer_join_to_inner_join_any_join.sql @@ -0,0 +1,33 @@ +DROP TABLE IF EXISTS user_country; +DROP TABLE IF EXISTS user_transactions; + +CREATE TABLE user_country ( + user_id UInt64, + country String +) +ENGINE = ReplacingMergeTree +ORDER BY user_id; + +CREATE TABLE user_transactions ( + user_id UInt64, + transaction_id String +) +ENGINE = MergeTree +ORDER BY user_id; + +INSERT INTO user_country (user_id, country) VALUES (1, 'US'); +INSERT INTO user_transactions (user_id, transaction_id) VALUES (1, 'tx1'), (1, 'tx2'), (1, 'tx3'), (2, 'tx1'); + +-- Expected 3 rows, got only 1. Removing 'ANY' and adding 'FINAL' fixes +-- the issue (but it is not always possible). Moving filter by 'country' to +-- an outer query doesn't help. Query without filter by 'country' works +-- as expected (returns 3 rows). +SELECT * FROM user_transactions +ANY LEFT JOIN user_country USING (user_id) +WHERE + user_id = 1 + AND country = 'US' +ORDER BY ALL; + +DROP TABLE user_country; +DROP TABLE user_transactions; diff --git a/tests/queries/0_stateless/03210_fix_single_value_data_assertion.reference b/tests/queries/0_stateless/03210_fix_single_value_data_assertion.reference new file mode 100644 index 00000000000..d8f7e13db55 --- /dev/null +++ b/tests/queries/0_stateless/03210_fix_single_value_data_assertion.reference @@ -0,0 +1,12 @@ +0 1 1 1 0 0 0 +1 3 3 3 2 2 2 +2 5 5 5 4 4 4 +3 7 7 7 6 6 6 +4 9 9 9 8 8 8 +5 11 11 11 10 10 10 +6 13 13 13 12 12 12 +7 15 15 15 14 14 14 +8 17 17 17 16 16 16 +9 19 19 19 18 18 18 + +0 107351244 107351244 107351244 107354520 107354520 107354520 diff --git a/tests/queries/0_stateless/03210_fix_single_value_data_assertion.sql b/tests/queries/0_stateless/03210_fix_single_value_data_assertion.sql new file mode 100644 index 00000000000..a1243ef0b25 --- /dev/null +++ b/tests/queries/0_stateless/03210_fix_single_value_data_assertion.sql @@ -0,0 +1,19 @@ +SELECT + intDiv(number, 2) AS k, + sumArgMax(number, number % 20), + sumArgMax(number, leftPad(toString(number % 20), 5, '0')), -- Pad with 0 to preserve number ordering + sumArgMax(number, [number % 20, number % 20]), + sumArgMin(number, number % 20), + sumArgMin(number, leftPad(toString(number % 20), 5, '0')), + sumArgMin(number, [number % 20, number % 20]), +FROM +( + SELECT number + FROM system.numbers + LIMIT 65537 +) +GROUP BY k + WITH TOTALS +ORDER BY k ASC + LIMIT 10 +SETTINGS group_by_overflow_mode = 'any', totals_mode = 'before_having', max_rows_to_group_by = 100000; diff --git a/tests/queries/0_stateless/03211_convert_outer_join_to_inner_join_anti_join.reference b/tests/queries/0_stateless/03211_convert_outer_join_to_inner_join_anti_join.reference new file mode 100644 index 00000000000..d717a29ab23 --- /dev/null +++ b/tests/queries/0_stateless/03211_convert_outer_join_to_inner_join_anti_join.reference @@ -0,0 +1,19 @@ +DATA + ┏━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━┓ + ┃ c0 ┃ c1 ┃ c2 ┃ + ┡━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━┩ +1. │ 826636805 │ 0 │ │ + ├───────────┼───────────┼────┤ +2. │ 0 │ 150808457 │ │ + └───────────┴───────────┴────┘ +NUMBER OF ROWS IN FIRST SHOULD BE EQUAL TO SECOND +FISRT + +SECOND +1 +TO DEBUG I TOOK JUST A SUBQUERY AND IT HAS 1 ROW +THIRD +1 +AND I ADDED SINGLE CONDITION THAT CONDITION <>0 THAT IS 1 IN THIRD QUERY AND IT HAS NO RESULT!!! +FOURTH +1 diff --git a/tests/queries/0_stateless/03211_convert_outer_join_to_inner_join_anti_join.sql b/tests/queries/0_stateless/03211_convert_outer_join_to_inner_join_anti_join.sql new file mode 100644 index 00000000000..77b1d52dd18 --- /dev/null +++ b/tests/queries/0_stateless/03211_convert_outer_join_to_inner_join_anti_join.sql @@ -0,0 +1,45 @@ +DROP TABLE IF EXISTS t0; + +CREATE TABLE t0 (c0 Int32, c1 Int32, c2 String) ENGINE = Log() ; +INSERT INTO t0(c0, c1, c2) VALUES (826636805,0, ''), (0, 150808457, ''); + +SELECT 'DATA'; +SELECT * FROM t0 FORMAT PrettyMonoBlock; + +SELECT 'NUMBER OF ROWS IN FIRST SHOULD BE EQUAL TO SECOND'; + + +SELECT 'FISRT'; +SELECT left.c2 FROM t0 AS left +LEFT ANTI JOIN t0 AS right_0 ON ((left.c0)=(right_0.c1)) +WHERE (abs ((- ((sign (right_0.c1)))))); + +SELECT 'SECOND'; +SELECT SUM(check <> 0) +FROM +( + SELECT (abs ((- ((sign (right_0.c1)))))) AS `check` + FROM t0 AS left + LEFT ANTI JOIN t0 AS right_0 ON ((left.c0)=(right_0.c1)) +); + + +SELECT 'TO DEBUG I TOOK JUST A SUBQUERY AND IT HAS 1 ROW'; + +SELECT 'THIRD'; + +SELECT (abs ((- ((sign (right_0.c1)))))) AS `check` +FROM t0 AS left +LEFT ANTI JOIN t0 AS right_0 ON ((left.c0)=(right_0.c1)); + + +SELECT 'AND I ADDED SINGLE CONDITION THAT CONDITION <>0 THAT IS 1 IN THIRD QUERY AND IT HAS NO RESULT!!!'; + + +SELECT 'FOURTH'; +SELECT (abs ((- ((sign (right_0.c1)))))) AS `check` +FROM t0 AS left +LEFT ANTI JOIN t0 AS right_0 ON ((left.c0)=(right_0.c1)) +WHERE check <> 0; + +DROP TABLE t0; diff --git a/tests/queries/0_stateless/data_avro/negative_block_size_arrays.avro b/tests/queries/0_stateless/data_avro/negative_block_size_arrays.avro new file mode 100644 index 00000000000..ec785a885dc Binary files /dev/null and b/tests/queries/0_stateless/data_avro/negative_block_size_arrays.avro differ diff --git a/tests/queries/0_stateless/data_json/key_ignore_case.json b/tests/queries/0_stateless/data_json/key_ignore_case.json new file mode 100644 index 00000000000..ad8f7cb4507 Binary files /dev/null and b/tests/queries/0_stateless/data_json/key_ignore_case.json differ diff --git a/tests/queries/0_stateless/data_orc/test_reader_time_zone.snappy.orc b/tests/queries/0_stateless/data_orc/test_reader_time_zone.snappy.orc new file mode 100644 index 00000000000..ab1b785dbbf Binary files /dev/null and b/tests/queries/0_stateless/data_orc/test_reader_time_zone.snappy.orc differ diff --git a/tests/queries/0_stateless/replication.lib b/tests/queries/0_stateless/replication.lib index 2e21f351d2a..05651531fba 100755 --- a/tests/queries/0_stateless/replication.lib +++ b/tests/queries/0_stateless/replication.lib @@ -51,6 +51,14 @@ function check_replication_consistency() table_name_prefix=$1 check_query_part=$2 + # Try to kill some mutations because sometimes tests run too much (it's not guarenteed to kill all mutations, see below) + # Try multiple replicas, because queries are not finished yet, and "global" KILL MUTATION may fail due to another query (like DROP TABLE) + readarray -t tables_arr < <(${CLICKHOUSE_CLIENT} -q "SELECT name FROM system.tables WHERE database=currentDatabase() AND name like '$table_name_prefix%'") + for t in "${tables_arr[@]}" + do + ${CLICKHOUSE_CLIENT} -q "KILL MUTATION WHERE database=currentDatabase() AND table='$t'" > /dev/null 2>/dev/null ||: + done + # Wait for all queries to finish (query may still be running if thread is killed by timeout) num_tries=0 while [[ $($CLICKHOUSE_CLIENT -q "SELECT count() FROM system.processes WHERE current_database=currentDatabase() AND query LIKE '%$table_name_prefix%'") -ne 1 ]]; do @@ -79,7 +87,9 @@ function check_replication_consistency() fi # Touch all data to check that it's readable (and trigger PartCheckThread if needed) - while ! $CLICKHOUSE_CLIENT -q "SELECT * FROM merge(currentDatabase(), '$table_name_prefix') FORMAT Null" 2>/dev/null; do + # it's important to disable prefer warmed unmerged parts because + # otherwise it can read non-syncrhonized state of replicas + while ! $CLICKHOUSE_CLIENT --prefer_warmed_unmerged_parts_seconds=0 -q "SELECT * FROM merge(currentDatabase(), '$table_name_prefix') FORMAT Null" 2>/dev/null; do sleep 1; num_tries=$((num_tries+1)) if [ $num_tries -eq 250 ]; then @@ -94,7 +104,7 @@ function check_replication_consistency() some_table=$($CLICKHOUSE_CLIENT -q "SELECT name FROM system.tables WHERE database=currentDatabase() AND name like '$table_name_prefix%' ORDER BY rand() LIMIT 1") $CLICKHOUSE_CLIENT -q "SYSTEM SYNC REPLICA $some_table PULL" 1>/dev/null 2>/dev/null ||: - # Forcefully cancel mutations to avoid waiting for them to finish + # Forcefully cancel mutations to avoid waiting for them to finish. Kills the remaining mutations ${CLICKHOUSE_CLIENT} -q "KILL MUTATION WHERE database=currentDatabase() AND table like '$table_name_prefix%'" > /dev/null # SYNC REPLICA is not enough if some MUTATE_PARTs are not assigned yet @@ -102,7 +112,9 @@ function check_replication_consistency() try_sync_replicas "$table_name_prefix" "$time_left" || exit 1 - res=$($CLICKHOUSE_CLIENT -q \ + # it's important to disable prefer warmed unmerged parts because + # otherwise it can read non-syncrhonized state of replicas + res=$($CLICKHOUSE_CLIENT --prefer_warmed_unmerged_parts_seconds=0 -q \ "SELECT if((countDistinct(data) as c) == 0, 1, c) FROM diff --git a/tests/queries/1_stateful/00162_mmap_compression_none.sql b/tests/queries/1_stateful/00162_mmap_compression_none.sql index d2cbcea8aaa..48d6ada821e 100644 --- a/tests/queries/1_stateful/00162_mmap_compression_none.sql +++ b/tests/queries/1_stateful/00162_mmap_compression_none.sql @@ -1,6 +1,6 @@ DROP TABLE IF EXISTS hits_none; CREATE TABLE hits_none (Title String CODEC(NONE)) ENGINE = MergeTree ORDER BY tuple() SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; -INSERT INTO hits_none SELECT Title FROM test.hits; +INSERT INTO hits_none SELECT Title FROM test.hits SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16; SET min_bytes_to_use_mmap_io = 1; SELECT sum(length(Title)) FROM hits_none; diff --git a/tests/queries/1_stateful/00166_explain_estimate.reference b/tests/queries/1_stateful/00166_explain_estimate.reference index 71ddd681581..85ecd0b9a71 100644 --- a/tests/queries/1_stateful/00166_explain_estimate.reference +++ b/tests/queries/1_stateful/00166_explain_estimate.reference @@ -1,5 +1,5 @@ test hits 1 57344 7 -test hits 1 8839168 1079 -test hits 1 835584 102 +test hits 1 8832938 1079 +test hits 1 829354 102 test hits 1 8003584 977 test hits 2 581632 71 diff --git a/tests/queries/1_stateful/00180_no_seek_avoiding_when_reading_from_cache.sh b/tests/queries/1_stateful/00180_no_seek_avoiding_when_reading_from_cache.sh index 2e1b807c496..f9fea2c1dad 100755 --- a/tests/queries/1_stateful/00180_no_seek_avoiding_when_reading_from_cache.sh +++ b/tests/queries/1_stateful/00180_no_seek_avoiding_when_reading_from_cache.sh @@ -17,11 +17,11 @@ $CLICKHOUSE_CLIENT -q "OPTIMIZE TABLE hits_s3_sampled FINAL" $CLICKHOUSE_CLIENT -q "SYSTEM DROP FILESYSTEM CACHE" # Warm up the cache -$CLICKHOUSE_CLIENT -q "SELECT * FROM hits_s3_sampled WHERE URL LIKE '%google%' ORDER BY EventTime LIMIT 10 FORMAT Null" -$CLICKHOUSE_CLIENT -q "SELECT * FROM hits_s3_sampled WHERE URL LIKE '%google%' ORDER BY EventTime LIMIT 10 FORMAT Null" +$CLICKHOUSE_CLIENT -q "SELECT * FROM hits_s3_sampled WHERE URL LIKE '%google%' ORDER BY EventTime LIMIT 10 FORMAT Null SETTINGS filesystem_cache_reserve_space_wait_lock_timeout_milliseconds=2000" +$CLICKHOUSE_CLIENT -q "SELECT * FROM hits_s3_sampled WHERE URL LIKE '%google%' ORDER BY EventTime LIMIT 10 FORMAT Null SETTINGS filesystem_cache_reserve_space_wait_lock_timeout_milliseconds=2000" query_id=02906_read_from_cache_$RANDOM -$CLICKHOUSE_CLIENT --query_id ${query_id} -q "SELECT * FROM hits_s3_sampled WHERE URL LIKE '%google%' ORDER BY EventTime LIMIT 10 FORMAT Null" +$CLICKHOUSE_CLIENT --query_id ${query_id} -q "SELECT * FROM hits_s3_sampled WHERE URL LIKE '%google%' ORDER BY EventTime LIMIT 10 FORMAT Null SETTINGS filesystem_cache_reserve_space_wait_lock_timeout_milliseconds=2000" $CLICKHOUSE_CLIENT -nq " SYSTEM FLUSH LOGS; diff --git a/tests/queries/shell_config.sh b/tests/queries/shell_config.sh index 614bfcece8f..f7017958635 100644 --- a/tests/queries/shell_config.sh +++ b/tests/queries/shell_config.sh @@ -54,9 +54,17 @@ export CLICKHOUSE_OBFUSCATOR=${CLICKHOUSE_OBFUSCATOR:="${CLICKHOUSE_BINARY}-obfu export CLICKHOUSE_COMPRESSOR=${CLICKHOUSE_COMPRESSOR:="${CLICKHOUSE_BINARY}-compressor"} export CLICKHOUSE_GIT_IMPORT=${CLICKHOUSE_GIT_IMPORT="${CLICKHOUSE_BINARY}-git-import"} +export CLICKHOUSE_CONFIG_DIR=${CLICKHOUSE_CONFIG_DIR:="/etc/clickhouse-server"} export CLICKHOUSE_CONFIG=${CLICKHOUSE_CONFIG:="/etc/clickhouse-server/config.xml"} export CLICKHOUSE_CONFIG_CLIENT=${CLICKHOUSE_CONFIG_CLIENT:="/etc/clickhouse-client/config.xml"} +export CLICKHOUSE_USER_FILES=${CLICKHOUSE_USER_FILES:="/var/lib/clickhouse/user_files"} +export CLICKHOUSE_USER_FILES_UNIQUE=${CLICKHOUSE_USER_FILES_UNIQUE:="${CLICKHOUSE_USER_FILES}/${CLICKHOUSE_TEST_UNIQUE_NAME}"} +# synonym +export USER_FILES_PATH=$CLICKHOUSE_USER_FILES + +export CLICKHOUSE_SCHEMA_FILES=${CLICKHOUSE_SCHEMA_FILES:="/var/lib/clickhouse/format_schemas"} + [ -x "${CLICKHOUSE_BINARY}-extract-from-config" ] && CLICKHOUSE_EXTRACT_CONFIG=${CLICKHOUSE_EXTRACT_CONFIG:="$CLICKHOUSE_BINARY-extract-from-config --config=$CLICKHOUSE_CONFIG"} [ -x "${CLICKHOUSE_BINARY}" ] && CLICKHOUSE_EXTRACT_CONFIG=${CLICKHOUSE_EXTRACT_CONFIG:="$CLICKHOUSE_BINARY extract-from-config --config=$CLICKHOUSE_CONFIG"} export CLICKHOUSE_EXTRACT_CONFIG=${CLICKHOUSE_EXTRACT_CONFIG:="$CLICKHOUSE_BINARY-extract-from-config --config=$CLICKHOUSE_CONFIG"} @@ -124,7 +132,7 @@ export CLICKHOUSE_URL_INTERSERVER=${CLICKHOUSE_URL_INTERSERVER:="${CLICKHOUSE_PO export CLICKHOUSE_CURL_COMMAND=${CLICKHOUSE_CURL_COMMAND:="curl"} # The queries in CI are prone to sudden delays, and we often don't check for curl # errors, so it makes sense to set a relatively generous timeout. -export CLICKHOUSE_CURL_TIMEOUT=${CLICKHOUSE_CURL_TIMEOUT:="60"} +export CLICKHOUSE_CURL_TIMEOUT=${CLICKHOUSE_CURL_TIMEOUT:="120"} export CLICKHOUSE_CURL=${CLICKHOUSE_CURL:="${CLICKHOUSE_CURL_COMMAND} -q -s --max-time ${CLICKHOUSE_CURL_TIMEOUT}"} export CLICKHOUSE_TMP=${CLICKHOUSE_TMP:="."} mkdir -p ${CLICKHOUSE_TMP} diff --git a/tests/result b/tests/result new file mode 100644 index 00000000000..b76f44f1e6a --- /dev/null +++ b/tests/result @@ -0,0 +1,12 @@ +Using queries from 'queries' directory +Connecting to ClickHouse server...... OK +Connected to server 24.7.1.1 @ 246f421f2402799fd11b22a608b4d0d497cb8438 chesema-processor-onCancel + +Running 1 stateless tests (MainProcess). + +00993_system_parts_race_condition_drop_zookeeper: [ OK ] + +1 tests passed. 0 tests skipped. 124.59 s elapsed (MainProcess). + +0 tests passed. 0 tests skipped. 0.00 s elapsed (MainProcess). +All tests have finished. diff --git a/utils/backup/backup b/utils/backup/backup new file mode 100755 index 00000000000..6aa9c179033 --- /dev/null +++ b/utils/backup/backup @@ -0,0 +1,47 @@ +#!/bin/bash + +user="default" +path="." + +usage() { + echo + echo "A trivial script to upload your files into ClickHouse." + echo "You might want to use something like Dropbox instead, but..." + echo + echo "Usage: $0 --host [--user ] --password " + exit 1 +} + +while [[ "$#" -gt 0 ]]; do + case "$1" in + --host) + host="$2" + shift 2 + ;; + --user) + user="$2" + shift 2 + ;; + --password) + password="$2" + shift 2 + ;; + --help) + usage + ;; + *) + path="$1" + shift 1 + ;; + esac +done + +if [ -z "$host" ] || [ -z "$password" ]; then + echo "Error: --host and --password are mandatory." + usage +fi + +clickhouse-client --host "$host" --user "$user" --password "$password" --secure --query "CREATE TABLE IF NOT EXISTS default.files (time DEFAULT now(), path String, content String CODEC(ZSTD(6))) ENGINE = MergeTree ORDER BY (path, time)" && +find "$path" -type f | clickhouse-local --input-format LineAsString \ + --max-block-size 1 --min-insert-block-size-rows 0 --min-insert-block-size-bytes '100M' --max-insert-threads 1 \ + --query "INSERT INTO FUNCTION remoteSecure('$host', default.files, '$user', '$password') (path, content) SELECT line, file(line) FROM table" --progress diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt index b4ed577cb63..4f71d952559 100644 --- a/utils/check-style/aspell-ignore/en/aspell-dict.txt +++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt @@ -1,4 +1,4 @@ -personal_ws-1.1 en 2758 +personal_ws-1.1 en 2942 AArch ACLs ALTERs @@ -48,7 +48,6 @@ AutoML Autocompletion AvroConfluent BIGINT -bigrams BIGSERIAL BORO BSON @@ -223,7 +222,6 @@ DatabaseOrdinaryThreadsActive DateTime DateTimes DbCL -deallocated Decrypted Deduplicate Deduplication @@ -295,7 +293,6 @@ FilesystemMainPathUsedBytes FilesystemMainPathUsedINodes FixedString FlameGraph -flameGraph Flink ForEach FreeBSD @@ -452,7 +449,7 @@ Kahan Kaser KeeperAliveConnections KeeperMap -KeeperOutstandingRequets +KeeperOutstandingRequests Kerberos Khanna KittenHouse @@ -470,7 +467,7 @@ LOCALTIME LOCALTIMESTAMP LONGLONG LOONGARCH -Lemir +Lemire Levenshtein Liao LibFuzzer @@ -577,6 +574,7 @@ MySQLDump MySQLThreads NATS NCHAR +NDJSON NEKUDOTAYIM NEWDATE NEWDECIMAL @@ -719,6 +717,8 @@ PlantUML PointDistKm PointDistM PointDistRads +PostHistory +PostLink PostgreSQLConnection PostgreSQLThreads Postgres @@ -747,6 +747,7 @@ PromQL Promql Promtail Protobuf +ProtobufList ProtobufSingle ProxySQL Punycode @@ -763,6 +764,7 @@ QueryCacheMisses QueryPreempted QueryThread QuickAssist +QuickSight QuoteMeta RBAC RClickHouse @@ -994,6 +996,8 @@ UPDATEs URIs URL URL's +URLDecode +URLEncode URLHash URLHierarchy URLPathHierarchy @@ -1011,13 +1015,10 @@ UncompressedCacheBytes UncompressedCacheCells UnidirectionalEdgeIsValid UniqThetaSketch -unigrams Updatable Uppercased Uptime Uptrace -UrlDecode -UrlEncode UserID Util VARCHAR @@ -1101,6 +1102,8 @@ aggregatefunction aggregatingmergetree aggregatio aggretate +aggthrow +aggThrow aiochclient allocator alphaTokens @@ -1223,6 +1226,7 @@ basename bcrypt benchmarking bfloat +bigrams binlog bitAnd bitCount @@ -1319,6 +1323,12 @@ cfg cgroup cgroups chadmin +changeDay +changeHour +changeMinute +changeMonth +changeSecond +changeYear changelog changelogs charset @@ -1472,6 +1482,7 @@ dbeaver dbgen dbms ddl +deallocated deallocation deallocations debian @@ -1511,11 +1522,11 @@ deserializing destructor destructors detectCharset -detectTonality detectLanguage detectLanguageMixed detectLanguageUnknown detectProgrammingLanguage +detectTonality determinator deterministically dictGet @@ -1531,8 +1542,8 @@ dictIsIn disableProtocols disjunction disjunctions -displaySecretsInShowAndSelect displayName +displaySecretsInShowAndSelect distro divideDecimal dmesg @@ -1582,11 +1593,11 @@ evalMLMethod exFAT expiryMsec exponentialMovingAverage -exponentialmovingaverage exponentialTimeDecayedAvg exponentialTimeDecayedCount exponentialTimeDecayedMax exponentialTimeDecayedSum +exponentialmovingaverage expr exprN extendedVerification @@ -1613,6 +1624,7 @@ filesystem filesystemAvailable filesystemCapacity filesystemFree +filesystemUnreserved filesystems finalizeAggregation fips @@ -1623,6 +1635,7 @@ firstSignificantSubdomainCustom firstSignificantSubdomainCustomRFC firstSignificantSubdomainRFC fixedstring +flameGraph flamegraph flatbuffers flattenTuple @@ -1644,6 +1657,7 @@ formated formatschema formatter formatters +fqdn frac freezed fromDaysSinceYearZero @@ -1656,6 +1670,7 @@ fsync func fuzzBits fuzzJSON +fuzzQuery fuzzer fuzzers gRPC @@ -1805,8 +1820,8 @@ incrementing indexHint indexOf infi -infty inflight +infty initcap initcapUTF initialQueryID @@ -1853,6 +1868,7 @@ jdbc jemalloc jeprof joinGet +joinGetOrNull json jsonMergePatch jsonasstring @@ -1897,11 +1913,13 @@ kurtosis kurtpop kurtsamp laion +lagInFrame lang laravel largestTriangleThreeBuckets latencies ldap +leadInFrame leftPad leftPadUTF leftUTF @@ -1954,9 +1972,9 @@ loghouse london lookups loongarch -lowcardinality lowCardinalityIndices lowCardinalityKeys +lowcardinality lowerUTF lowercased lttb @@ -1979,6 +1997,8 @@ mapExtractKeyLike mapFilter mapFromArrays mapKeys +mapPartialReverseSort +mapPartialSort mapPopulateSeries mapReverseSort mapSort @@ -1995,6 +2015,7 @@ maxMap maxintersections maxintersectionsposition maxmap +minMappedArrays maxmind mdadm meanZTest @@ -2012,6 +2033,7 @@ metrica metroHash mfedotov minMap +minMappedArrays minSampleSizeContinuous minSampleSizeConversion mindsdb @@ -2119,8 +2141,10 @@ noaa nonNegativeDerivative noop normalizeQuery +normalizeQueryKeepNames normalizeUTF normalizedQueryHash +normalizedQueryHashKeepNames notEmpty notEquals notILike @@ -2183,6 +2207,8 @@ parseReadableSizeOrZero parseTimeDelta parseable parsers +partitionId +partitionID pathFull pclmulqdq pcre @@ -2264,9 +2290,10 @@ proleptic prometheus proportionsZTest proto -protocol protobuf +protobuflist protobufsingle +protocol proxied pseudorandom pseudorandomize @@ -2518,6 +2545,8 @@ sqlite sqrt src srcReplicas +sshkey +stackoverflow stacktrace stacktraces startsWith @@ -2757,6 +2786,9 @@ topLevelDomain topLevelDomainRFC topk topkweighted +transactionID +transactionLatestSnapshot +transactionOldestSnapshot transactional transactionally translateUTF @@ -2790,6 +2822,7 @@ tupleModulo tupleModuloByNumber tupleMultiply tupleMultiplyByNumber +tupleNames tupleNegate tuplePlus tupleToNameValuePairs @@ -2810,6 +2843,7 @@ unescaping unhex unicode unidimensional +unigrams unintuitive uniq uniqCombined @@ -2856,6 +2890,7 @@ userver utils uuid uuidv +vCPU varPop varPopStable varSamp diff --git a/utils/check-style/check-pylint b/utils/check-style/check-pylint index 8cfbc68ac96..919a1661e45 100755 --- a/utils/check-style/check-pylint +++ b/utils/check-style/check-pylint @@ -12,5 +12,4 @@ function xargs-pylint { # exclude ci unittest scripts from check: test_* find "$ROOT_PATH/tests" -maxdepth 2 -type f -exec file -F' ' --mime-type {} + | grep -v "/test_" | xargs-pylint 50 -# Beware, there lambdas are checked. All of them contain `app`, and it causes brain-cucumber-zalgo -find "$ROOT_PATH/tests/ci" -mindepth 2 -type f -exec file -F' ' --mime-type {} + | xargs-pylint 1 + diff --git a/utils/check-style/check-style b/utils/check-style/check-style index 722dfbcad16..380656cd1ca 100755 --- a/utils/check-style/check-style +++ b/utils/check-style/check-style @@ -322,10 +322,14 @@ std_cerr_cout_excludes=( src/Client/LineReader.cpp src/Client/QueryFuzzer.cpp src/Client/Suggest.cpp + src/Client/ClientBase.h + src/Client/LineReader.h + src/Client/ReplxxLineReader.h src/Bridge/IBridge.cpp src/Daemon/BaseDaemon.cpp src/Loggers/Loggers.cpp src/Common/GWPAsan.cpp + src/Common/ProgressIndication.h ) sources_with_std_cerr_cout=( $( find $ROOT_PATH/{src,base} -name '*.h' -or -name '*.cpp' | \ diff --git a/utils/corrector_utf8/CMakeLists.txt b/utils/corrector_utf8/CMakeLists.txt index 4744dd5e0a5..17f047a8cff 100644 --- a/utils/corrector_utf8/CMakeLists.txt +++ b/utils/corrector_utf8/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable(corrector_utf8 corrector_utf8.cpp) -target_link_libraries(corrector_utf8 PRIVATE clickhouse_common_io) +target_link_libraries(corrector_utf8 PRIVATE clickhouse_common_io clickhouse_common_config) diff --git a/utils/grpc-client/generate_pb2.py b/utils/grpc-client/generate_pb2.py new file mode 100755 index 00000000000..95a39023ed7 --- /dev/null +++ b/utils/grpc-client/generate_pb2.py @@ -0,0 +1,52 @@ +#!/usr/bin/env python3 + +# This is a helper utility. +# It generates files in the "pb2" folder using the protocol buffer compiler. +# This script must be called manually after any change pf "clickhouse_grpc.proto" + +import grpc_tools # pip3 install grpcio-tools + +import os, shutil, subprocess + + +# Settings. +script_path = os.path.realpath(__file__) +script_name = os.path.basename(script_path) +script_dir = os.path.dirname(script_path) +root_dir = os.path.abspath(os.path.join(script_dir, "../..")) + +grpc_proto_dir = os.path.abspath(os.path.join(root_dir, "src/Server/grpc_protos")) +grpc_proto_filename = "clickhouse_grpc.proto" + +# Files in the "pb2" folder which will be generated by this script. +pb2_filenames = ["clickhouse_grpc_pb2.py", "clickhouse_grpc_pb2_grpc.py"] +pb2_dir = os.path.join(script_dir, "pb2") + + +# Processes the protobuf schema with the protocol buffer compiler and generates the "pb2" folder. +def generate_pb2(): + print(f"Generating files:") + for pb2_filename in pb2_filenames: + print(os.path.join(pb2_dir, pb2_filename)) + + os.makedirs(pb2_dir, exist_ok=True) + + cmd = [ + "python3", + "-m", + "grpc_tools.protoc", + "-I" + grpc_proto_dir, + "--python_out=" + pb2_dir, + "--grpc_python_out=" + pb2_dir, + os.path.join(grpc_proto_dir, grpc_proto_filename), + ] + subprocess.run(cmd) + + for pb2_filename in pb2_filenames: + assert os.path.exists(os.path.join(pb2_dir, pb2_filename)) + print("Done! (generate_pb2)") + + +# MAIN +if __name__ == "__main__": + generate_pb2() diff --git a/utils/grpc-client/pb2/generate.py b/utils/grpc-client/pb2/generate.py deleted file mode 100755 index 2f4b3bf5af7..00000000000 --- a/utils/grpc-client/pb2/generate.py +++ /dev/null @@ -1,29 +0,0 @@ -#!/usr/bin/env python3 - -import grpc_tools # pip3 install grpcio-tools - -import os -import subprocess - - -script_dir = os.path.dirname(os.path.realpath(__file__)) -dest_dir = script_dir -src_dir = os.path.abspath(os.path.join(script_dir, "../../../src/Server/grpc_protos")) -src_filename = "clickhouse_grpc.proto" - - -def generate(): - cmd = [ - "python3", - "-m", - "grpc_tools.protoc", - "-I" + src_dir, - "--python_out=" + dest_dir, - "--grpc_python_out=" + dest_dir, - os.path.join(src_dir, src_filename), - ] - subprocess.run(cmd) - - -if __name__ == "__main__": - generate() diff --git a/utils/keeper-bench/Runner.cpp b/utils/keeper-bench/Runner.cpp index ed7e09685f0..59761d827e1 100644 --- a/utils/keeper-bench/Runner.cpp +++ b/utils/keeper-bench/Runner.cpp @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -785,7 +786,7 @@ struct SetupNodeCollector if (snapshot_result.storage == nullptr) { std::cerr << "No initial snapshot found" << std::endl; - initial_storage = std::make_unique( + initial_storage = std::make_unique( /* tick_time_ms */ 500, /* superdigest */ "", keeper_context, /* initialize_system_nodes */ false); initial_storage->initializeSystemNodes(); } @@ -931,7 +932,7 @@ struct SetupNodeCollector std::cerr << "Generating snapshot with starting data" << std::endl; DB::SnapshotMetadataPtr snapshot_meta = std::make_shared(initial_storage->getZXID(), 1, std::make_shared()); - DB::KeeperStorageSnapshot snapshot(initial_storage.get(), snapshot_meta); + DB::KeeperStorageSnapshot snapshot(initial_storage.get(), snapshot_meta); snapshot_manager->serializeSnapshotToDisk(snapshot); new_nodes = false; @@ -939,9 +940,9 @@ struct SetupNodeCollector std::mutex nodes_mutex; DB::KeeperContextPtr keeper_context; - Coordination::KeeperStoragePtr initial_storage; + std::shared_ptr initial_storage; std::unordered_set nodes_created_during_replay; - std::optional snapshot_manager; + std::optional> snapshot_manager; bool new_nodes = false; }; @@ -1238,9 +1239,13 @@ void Runner::createConnections() std::shared_ptr Runner::getConnection(const ConnectionInfo & connection_info, size_t connection_info_idx) { - Coordination::ZooKeeper::Node node{Poco::Net::SocketAddress{connection_info.host}, static_cast(connection_info_idx), connection_info.secure}; - std::vector nodes; - nodes.push_back(node); + zkutil::ShuffleHost host; + host.host = connection_info.host; + host.secure = connection_info.secure; + host.original_index = static_cast(connection_info_idx); + host.address = Poco::Net::SocketAddress{connection_info.host}; + + zkutil::ShuffleHosts nodes{host}; zkutil::ZooKeeperArgs args; args.session_timeout_ms = connection_info.session_timeout_ms; args.connection_timeout_ms = connection_info.connection_timeout_ms; @@ -1307,9 +1312,9 @@ void removeRecursive(Coordination::ZooKeeper & zookeeper, const std::string & pa while (!children_span.empty()) { Coordination::Requests ops; - for (size_t i = 0; i < 1000 && !children.empty(); ++i) + for (size_t i = 0; i < 1000 && !children_span.empty(); ++i) { - removeRecursive(zookeeper, fs::path(path) / children.back()); + removeRecursive(zookeeper, fs::path(path) / children_span.back()); ops.emplace_back(zkutil::makeRemoveRequest(fs::path(path) / children_span.back(), -1)); children_span = children_span.subspan(0, children_span.size() - 1); } diff --git a/utils/keeper-bench/example.yaml b/utils/keeper-bench/example.yaml index e800e923482..c3a62a01eac 100644 --- a/utils/keeper-bench/example.yaml +++ b/utils/keeper-bench/example.yaml @@ -18,45 +18,46 @@ connections: host: "localhost:9181" -generator: - setup: +setup: + node: + name: "test3" + node: + name: "test_create" + node: + name: "test4" + node: + name: "test" + data: "somedata" node: - name: "test3" + repeat: 4 + name: + random_string: + size: 15 + data: + random_string: + size: + min_value: 10 + max_value: 20 node: - name: "test_create" - node: - name: "test4" - node: - name: "test" - data: "somedata" - node: - repeat: 4 - name: - random_string: - size: 15 - data: - random_string: - size: - min_value: 10 - max_value: 20 + repeat: 2 node: repeat: 2 - node: - repeat: 2 - name: - random_string: - size: 12 name: random_string: - size: 15 - data: - random_string: - size: - min_value: 10 - max_value: 20 - node: - name: "test2" - data: "somedata" + size: 12 + name: + random_string: + size: 15 + data: + random_string: + size: + min_value: 10 + max_value: 20 + node: + name: "test2" + data: "somedata" + +generator: requests: create: path: "/test_create" diff --git a/utils/keeper-data-dumper/main.cpp b/utils/keeper-data-dumper/main.cpp index 25be8549dd6..60170d67a2e 100644 --- a/utils/keeper-data-dumper/main.cpp +++ b/utils/keeper-data-dumper/main.cpp @@ -2,6 +2,7 @@ #include #include +#include #include #include #include @@ -15,7 +16,7 @@ using namespace Coordination; using namespace DB; -void dumpMachine(std::shared_ptr machine) +void dumpMachine(std::shared_ptr> machine) { auto & storage = machine->getStorageUnsafe(); std::queue keys; @@ -69,7 +70,7 @@ int main(int argc, char *argv[]) keeper_context->setLogDisk(std::make_shared("LogDisk", argv[2])); keeper_context->setSnapshotDisk(std::make_shared("SnapshotDisk", argv[1])); - auto state_machine = std::make_shared(queue, snapshots_queue, keeper_context, nullptr); + auto state_machine = std::make_shared>(queue, snapshots_queue, keeper_context, nullptr); state_machine->init(); size_t last_commited_index = state_machine->last_commit_index(); diff --git a/docker/test/libfuzzer/run_libfuzzer.py b/utils/libfuzzer/runner.py old mode 100755 new mode 100644 similarity index 72% rename from docker/test/libfuzzer/run_libfuzzer.py rename to utils/libfuzzer/runner.py index 5ed019490d5..bbe648dbbc2 --- a/docker/test/libfuzzer/run_libfuzzer.py +++ b/utils/libfuzzer/runner.py @@ -20,6 +20,7 @@ def run_fuzzer(fuzzer: str): options_file = f"{fuzzer}.options" custom_libfuzzer_options = "" + fuzzer_arguments = "" with Path(options_file) as path: if path.exists() and path.is_file(): @@ -27,19 +28,19 @@ def run_fuzzer(fuzzer: str): parser.read(path) if parser.has_section("asan"): - os.environ[ - "ASAN_OPTIONS" - ] = f"{os.environ['ASAN_OPTIONS']}:{':'.join('%s=%s' % (key, value) for key, value in parser['asan'].items())}" + os.environ["ASAN_OPTIONS"] = ( + f"{os.environ['ASAN_OPTIONS']}:{':'.join('%s=%s' % (key, value) for key, value in parser['asan'].items())}" + ) if parser.has_section("msan"): - os.environ[ - "MSAN_OPTIONS" - ] = f"{os.environ['MSAN_OPTIONS']}:{':'.join('%s=%s' % (key, value) for key, value in parser['msan'].items())}" + os.environ["MSAN_OPTIONS"] = ( + f"{os.environ['MSAN_OPTIONS']}:{':'.join('%s=%s' % (key, value) for key, value in parser['msan'].items())}" + ) if parser.has_section("ubsan"): - os.environ[ - "UBSAN_OPTIONS" - ] = f"{os.environ['UBSAN_OPTIONS']}:{':'.join('%s=%s' % (key, value) for key, value in parser['ubsan'].items())}" + os.environ["UBSAN_OPTIONS"] = ( + f"{os.environ['UBSAN_OPTIONS']}:{':'.join('%s=%s' % (key, value) for key, value in parser['ubsan'].items())}" + ) if parser.has_section("libfuzzer"): custom_libfuzzer_options = " ".join( @@ -47,9 +48,17 @@ def run_fuzzer(fuzzer: str): for key, value in parser["libfuzzer"].items() ) + if parser.has_section("fuzzer_arguments"): + fuzzer_arguments = " ".join( + ("%s" % key) if value == "" else ("%s=%s" % (key, value)) + for key, value in parser["fuzzer_arguments"].items() + ) + cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {corpus_dir}" if custom_libfuzzer_options: cmd_line += f" {custom_libfuzzer_options}" + if fuzzer_arguments: + cmd_line += f" {fuzzer_arguments}" if not "-dict=" in cmd_line and Path(f"{fuzzer}.dict").exists(): cmd_line += f" -dict={fuzzer}.dict" @@ -70,8 +79,6 @@ def main(): if (current / fuzzer).is_file() and os.access(current / fuzzer, os.X_OK): run_fuzzer(fuzzer) - exit(0) - if __name__ == "__main__": main() diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 1bff4cb0b09..271065a78fb 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -1,8 +1,13 @@ +v24.6.2.17-stable 2024-07-05 +v24.6.1.4423-stable 2024-07-01 +v24.5.4.49-stable 2024-07-01 v24.5.3.5-stable 2024-06-13 v24.5.2.34-stable 2024-06-13 v24.5.1.1763-stable 2024-06-01 +v24.4.3.25-stable 2024-06-14 v24.4.2.141-stable 2024-06-07 v24.4.1.2088-stable 2024-05-01 +v24.3.5.46-lts 2024-07-03 v24.3.4.147-lts 2024-06-13 v24.3.3.102-lts 2024-05-01 v24.3.2.23-lts 2024-04-03 diff --git a/utils/zookeeper-cli/CMakeLists.txt b/utils/zookeeper-cli/CMakeLists.txt index cad7164b775..fd2fa669f40 100644 --- a/utils/zookeeper-cli/CMakeLists.txt +++ b/utils/zookeeper-cli/CMakeLists.txt @@ -3,4 +3,6 @@ clickhouse_add_executable(clickhouse-zookeeper-cli ${ClickHouse_SOURCE_DIR}/src/Client/LineReader.cpp) target_link_libraries(clickhouse-zookeeper-cli PRIVATE clickhouse_common_zookeeper_no_log - dbms) + dbms + clickhouse_functions +) diff --git a/utils/zookeeper-dump-tree/CMakeLists.txt b/utils/zookeeper-dump-tree/CMakeLists.txt index 85e4d18c19f..3f3df65776a 100644 --- a/utils/zookeeper-dump-tree/CMakeLists.txt +++ b/utils/zookeeper-dump-tree/CMakeLists.txt @@ -3,4 +3,5 @@ target_link_libraries(zookeeper-dump-tree PRIVATE clickhouse_common_zookeeper_no_log clickhouse_common_io dbms + clickhouse_functions boost::program_options) diff --git a/utils/zookeeper-remove-by-list/CMakeLists.txt b/utils/zookeeper-remove-by-list/CMakeLists.txt index 50aaed76110..a4d7dccef65 100644 --- a/utils/zookeeper-remove-by-list/CMakeLists.txt +++ b/utils/zookeeper-remove-by-list/CMakeLists.txt @@ -2,4 +2,5 @@ clickhouse_add_executable (zookeeper-remove-by-list main.cpp ${SRCS}) target_link_libraries(zookeeper-remove-by-list PRIVATE clickhouse_common_zookeeper_no_log dbms + clickhouse_functions boost::program_options)